35999 world development report 2007 Development and the Next Generation world development report 2007 Development and the Next Generation THE WORLD BANK Washington, DC ©2006 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved 1 2 3 4 09 08 07 06 This volume is a product of the staff of the International Bank for Reconstruction and Develop- ment / The World Bank. The findings, interpretations, and conclusions expressed herein do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any ter- ritory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Cover design by Chris Lester of Rock Creek Creative, Inc. Typesetting by Precision Graphics. Cover images by Paul Olaja. Paul is a 19-year-old artist studying at Kyambogo University in Kampala, Uganda. In his paintings, Paul represents the many facets of youth: youth performing as a way to share their rich culture, working hard to survive, and growing into their adult roles. He depicts the central role of women, the tremendous diversity in the world and how young people strive to become a part of it. In his own words "youth need to do something for the world to hold it up." As a teenager, Paul lost his parents to HIV/AIDS. Art is Paul's passion, and his goal is to use art to share the stories of his homeland and to help other young people in need in Africa. Softcover Hardcover ISBN-10: 0-8213-6541-X ISBN-10: 0-8213-6549-5 ISBN-13: 978-0-8213-6541-0 ISBN-13: 978-0-8213-6549-6 ISSN: 0163-5085 ISSN: 0163-5085 eISBN-10: 0-8213-6542-8 eISBN-10: 0-8213-6550-9 eISBN-13: 978-0-8213-6542-7 eISBN-13: 978-0-8213-6550-2 DOI: 10.1596/978-0-8213-6541-0 DOI: 10.1596/978-0-8213-6549-6 Contents Foreword xi Acknowledgments xiii Methodological Note xv Abbreviations and Data Notes xvi Overview 1 Invest in young people--now 2 Investments during youth's five life transitions 5 Policies should focus not only on youth's opportunities but also their capabilities and second chances 10 Policies to broaden opportunities 11 Policies that enhance capabilities: Youth as decision makers 15 Policies to offer second chances 19 Moving forward 21 Part I Why now, and how? 25 1 Youth, poverty reduction, and growth 26 Young people are critical to further progress with poverty reduction and growth 27 How the challenges confronting young people have changed 31 Do numbers matter? How demographic changes affect opportunities for youth 33 How prepared are youth for today's challenges? A glass half empty 36 What should policy makers focus on? The five transitions 39 Spotlight: Differing demographics 43 2 Opportunities, capabilities, second chances: A framework for policy 45 Broadening opportunities 46 Developing the capabilities of young people as decision-making agents 53 Offering second chances 59 Spotlight: A gender filter on the youth lens 65 v vi CONTENTS Part II Transitions 67 3 Learning for work and life 68 Educational preparation of youth for work and life is low 68 A solid foundation: Improving the readiness for postprimary education 71 Enhancing postprimary education opportunities 72 Better education choices by young people 81 Offering second chances 89 Spotlight: Vietnamese youth: Managing prosperity 94 4 Going to work 96 Youth challenges in the labor market 96 What makes youth vulnerable in the labor market? 103 Broadening labor market opportunities 106 Choosing to work and developing the skills to do so 111 Providing a springboard to reintegrate the most vulnerable 114 Spotlight: Do baby booms lead to employment busts? Not in OECD countries 120 5 Growing up healthy 122 Promoting the health of young people stimulates growth and reduces poverty and health care expenditures 123 Public intervention is needed to promote youth health 124 Strengthening young people's capability to practice healthy behavior 129 Enhancing opportunities to make healthy choices 135 What if prevention fails? Helping young people deal with the adverse consequences of poor health behavior or misfortune 137 Spotlight: Addressing disparities among Brazil's youth 142 6 Forming families 144 Preparing for family formation is good for growth and poverty reduction 144 Preparation for family formation is poor 147 Providing opportunities for youth to prepare for parenthood 151 Strengthening young people's decision-making capabilities to prepare for parenthood 153 Supporting those who become mothers at an early age 156 7 Exercising citizenship 160 Youth participation: Rising, declining, or both? 161 What youth citizenship means for adult citizenship and development 164 Contents vii Opportunities for political participation and active citizenship 167 Acquiring an identity and a sense of belonging 172 Young people need legally recognized second chances 178 Spotlight: Rebuilding lives and institutions in Sierra Leone 185 Part III Across transitions and next steps 187 8 Moving and communicating across borders 188 Youth and international migration 188 Youth and the global flow of information and ideas 201 Spotlight: What donors can do 209 9 Youth policy: Doing it and getting it right 211 Youth policy priorities vary by country context 212 Youth policy often fails young people 214 Getting it right--by developing a coherent framework and integrating it with national policy 215 Getting it right--by listening to young people 219 Getting it right--through monitoring and evaluation 221 Spotlight: It's up to you(th)--taking action for development 225 Bibliographical note 227 Endnotes 229 References 243 Selected indicators 271 Technical notes 283 Selected world development indicators 285 Data sources and methodology 285 Classification of economies and summary measures 285 Terminology and country coverage 286 Technical notes 286 Index 305 viii CONTENTS Boxes 1 Investing in young people pays off big time: Estimating 3.9 Reaching out-of-school youth in Bangladesh 90 the long-term and interactive effects of human capital 4.1 Measuring youth activity in the transition to work 98 investments 5 4.2 Some youths are neither in the labor force nor studying 100 2 The poor quality of basic education severely limits 4.3 Early unemployment persists in Bosnia and opportunities for young people 6 Herzegovina 101 3 International migration offers opportunities and risks 4.4 Reforming part of the labor market has been no substitute for youth 14 for comprehensive reform in Spain and France 108 4 Knowing what's good for you: Telling young people about 4.5 Off-farm opportunities for youth in Palanpur, India 110 the benefits of school can affect outcomes 17 4.6 Employing youth with disabilities 115 5 All dressed up and somewhere to go in Bungoma and Butere-Mumias 23 4.7 Joven programs increased employment and earnings for some disadvantaged youths 117 1.1 What is youth? 27 5.1 Street children abusing drugs 127 1.2 "The Millennium Development Goals are not about youth, right?" No, wrong! 29 5.2 The role of education in behavior change 130 1.3 Losing a decade--what HIV/AIDS is doing to human 5.3 Reducing HIV risk in Namibia 131 capital accumulation and growth in Kenya 30 5.4 Social marketing can change behavior--Horizon Jeunes 1.4 A youth perspective on equity and development 38 in Cameroon 135 1.5 Work and marriage for men in the Middle East and 5.5 Russia limited the sale of alcohol, and deaths and North Africa 40 illnesses fell 137 2.1 Applying the human capital model to young people 47 5.6 Technology can help change young people's behavior: Using text messages in New Zealand to reduce smoking 137 2.2 Making services work for poor young people--World Development Report 2004 redux, with a youth lens 50 6.1 The sequencing of marriage and childbearing 145 2.3 A tale of two exports: How electronics in Malaysia and 6.2 Education shapes family formation 147 garments in Bangladesh promoted work for young women 6.3 Voices of Bangladeshi youth: Searching for the in traditional societies 51 ideal spouse 148 2.4 Young people's voice in budget setting improved outcomes 6.4 Anemia: The outcome of multiple deficiencies 149 in Ceará, Brazil 53 6.5 Grameen Bank's "Sixteen Decisions"--convincing men 2.5 Seen but not heard: Who decides when to drop out of to have fewer children 155 school? Or when to marry? Or do anything else? 54 6.6 Cash transfers conditional on delaying marriage to promote 2.6 Is microcredit an answer to relieving the young's school attendance for girls in Bangladesh 157 resource constraints? 57 7.1 What is citizenship? 161 2.7 Are youths rational (at least according to 7.2 Do large youth cohorts cause violence? Maybe, if economic economists)? 58 growth rates are low 166 2.8 Even those still in school need second chances 60 7.3 The Otpor youth movement in the former Yugoslavia 168 2.9 Brain development among youth: Neuroscience meets 7.4 The citizenship of Big George--from youth social science 61 to adulthood 169 3.1 The ignored side of skills development: Building behavioral 7.5 Promoting voluntary and independent opportunities 173 skills for school, work, and life 72 7.6 Last chance in Texas 180 3.2 Life skills programs and nonformal teaching methods in schools 75 7.7 Private sector interventions to deter youth crime 181 3.3 Violence and harassment in schools 77 7.8 War-affected youth in Uganda 182 3.4 Georgia: Fighting corruption in higher education 78 8.1 Small islands, large migrations 190 3.5 Chile's higher education: Diversifying the sources 8.2 Poor job prospects fuel migration in Morocco 199 of funding 79 8.3 Moving in fits and starts with technology--the African 3.6 Korea's secondary education: Expansion without Virtual University 205 sacrificing quality 79 8.4 Staying alive: HIV prevention using ICTs 206 3.7 Income-contingent loans 88 9.1 What do ministries of youth do? 215 3.8 Integrated approaches address the many constraints on 9.2 Does addressing gender issues hold lessons for youth policy? young people 89 The view from East Asia 216 Contents ix 9.3 Neither seen, nor heard--youth in the poverty reduction 9.7 Vozes Jovens: Opening national youth policy to youth voice strategy process 217 and participation in Brazil 221 9.4 How do developed countries handle youth issues? 9.8 A youth scorecard? The many indicators of youth Consider Sweden 218 development 222 9.5 Where departments of youth are headed: Evidence from 9.9 Credible proof of a program's success can ensure continuity: Latin America 219 The case of Oportunidades 223 9.6 Successful policy coordination and implementation: How health and education policies can work together to combat the AIDS shock 220 Figures 1 High enrollment rates in primary school are followed by 2.3 Private funds finance a big part of tertiary education costs significantly lower rates at secondary levels in Indonesia and in diverse countries 49 Zambia 3 2.4 Who has the final say? Young people's influence on their 2 Opening and closing demographic windows of own key transitions varies greatly across societies and for opportunity 5 males and females 54 3 Youth enrollment rates decline with age 6 3.1 The transition to secondary school is a stumbling block for 4 Entry into the workforce increases with age 7 many young people 69 5 The unemployment rate everywhere is higher for young 3.3 The ability to read a simple sentence increases with people than older people--with gaps much larger in some schooling, but few acquire even that most basic skill 70 countries 8 3.2 Learning achievement in poor countries can be very low 70 6 Risky behaviors peak during youth 8 3.4 Influencing effort directly can be a cost-effective way to 7 Family formation increases with age 9 improve learning in Kenya 82 8 Civic engagement rises with age 10 3.5 More rich children than poor children completing primary school make it to higher levels 85 9 Transitions seen through three lenses focus policies and magnify impact 11 3.6 Adolescents in many low-income and lower-middle-income countries combine school and work 86 10 Early childhood interventions (at ages 1­5) can have long- lasting effects on young people (at ages 13­18) 12 4.1 Child labor is highest in Africa 97 11 Who has the final say? The percentage of young women 4.2 Unemployment is higher for youths than for adults 99 (ages 15­24) who feel they themselves have had the most 4.3 Where young people cannot afford to be unemployed, youth influence on key life transitions varies greatly across unemployment is low and employment is high 100 societies 15 4.4 Returns to experience are highest for the young 102 12 It pays to go to school 18 4.5 In Chile employment is more responsive to demand 1.1 The number of children completing primary school has fluctuations for young males than for adults 103 risen in most regions 31 4.6 Returns to education are higher for workers who are 1.3 People today participate more in political processes and more mobile 105 have greater civil rights and liberties 32 4.7 Productivity and earnings increased faster for youth 1.4 While access to the Internet varies greatly, young people relative to older workers in Slovenia during the transition, dominate usage 32 (1992­2001) 107 1.5 Trends in the developing world's population of young 5.1 In the United States, mortality from lung cancer among people vary significantly across regions 33 men increased dramatically for nearly 40 years after 1.6 The relative size of youth cohorts is declining in most of the smoking peaked 123 developing world, as is the dependency ratio 35 5.2 Sexually active youth are unlikely to use condoms, even 1.7 Many young people from poor households do not have the where HIV prevalence is high 125 most basic skill of literacy 37 5.3 Alcohol consumption is common among 15- to 1.8 The transitions for Haitians ages 12­24 41 19-year-olds in some countries 127 1.9 The transitions for Chileans ages 12­24 42 5.4 Many young people, especially young men, consume 2.1 Transitions seen through three lenses focus policies and tobacco 127 magnify impact 45 5.5 Many young women in developing countries are 2.2 Attending is not always learning 48 overweight 128 x CONTENTS 5.6 Being overweight as a youth increases the chances of being 7.2 Young people's confidence in the civil service is increasing overweight as an adult 129 in low-income countries 163 5.7 The proportion of 15- to 24-year-olds infected with HIV is 7.3 Young people are more willing than older people to give up high, especially among girls 129 income to prevent environmental pollution 167 5.8 Knowledge of the ABCs (Abstain--Be faithful--use 7.4 Fewer youth are registering to vote in urban Chile 169 Condoms) to avoid HIV prevention is low among young 7.5 Countries incarcerate young people at very different people 132 rates 179 5.9 Knowing that condoms prevent HIV does not always lead to 8.1 Young people add disproportionately to the flow of condom use 133 developing country migrants 189 5.10 Knowing that condoms prevent HIV increases with 8.2 Migrants from developing countries tend to return home at education, but so does the gap between knowledge and a relatively young age 192 behavior 133 8.3 Patterns of migration and youth transitions vary according 5.11 Reported STI incidence varies among youths in to age, gender, and circumstance 193 Sub-Saharan Africa 138 8.4 Youth are more likely than older people to migrate 6.1 Teenage motherhood is common in some regions 146 illegally 196 6.2 The share of youth fertility is high in countries with 8.5 Leave a light on for them--most young people wish to low fertility 146 migrate temporarily 197 6.3 Young women of short stature risk developing obstetric 8.6 In the Kyrgyz Republic, young people use the Internet complications 149 more than older people and account for much of its 6.4 Anemia is highly prevalent among young women 149 growth 202 6.5 Young women are less likely to use contraceptives than 8.7 Public Internet access points are important for young older women 150 people 203 6.6 Young mothers do not get full care during antenatal 8.8 Youth are more likely than older people to visits 150 communicate with people abroad, especially using new 6.7 In India, acquiring life skills can stimulate young women's technologies 203 demand for health services 156 7.1 Interest in politics is on the rise among youth ages 18­24 in low-income countries 162 Tables 1 Policy goals, directions, actions, and programs 22 7.3 Women (ages 18­29) are less likely to discuss politics 3.1 Summary of youth education policies 92 with friends 165 4.1 Employment transitions for youth (ages 16­30) 113 7.4 Youth opinions on military and national service in 2005 (percentage agreeing or strongly agreeing) 171 4.2 Summary of youth employment policy directions and examples of programs 118 7.5 Summary of citizenship policy directions and examples of programs 184 5.1 Proportion of countries that include specific health topics in the school curriculum (percent) 134 8.1 Youth make up a large share of developing country migrants 191 5.2 Summary of policies to improve young people's health and health behavior 141 8.2 Countries that restrict the right of women to emigrate 200 6.1 Programs and interventions that prepare youth for transition to family formation 158 8.3 Catching up fast: The rise of new technologies 201 7.1 Do young people care less about politics than older groups? 8.4 The digital divide among Indonesian youth 202 Not everywhere 163 8.5 What do youth do online? 204 7.2 Should private ownership of business be increased? 164 8.6 Policies for youth in a global world 208 Foreword The time has never been better to invest in young people living in developing countries--that is the message of this year's World Development Report, the twenty-ninth in the series. The num- ber of people worldwide aged 12­24 years has reached 1.3 billion, the largest in history. It is also the healthiest and best educated--a strong base to build on in a world that demands more than basic skills. Today's youth are tomorrow's workers, entrepreneurs, parents, active citizens, and, indeed, leaders. And, because of falling fertility, they will have fewer children than their parents as they move through adulthood. This in turn may boost growth--by raising the share of the popula- tion that is working and by boosting household savings. Rich and poor countries alike need to seize this opportunity before the aging of societies closes it. Doing so will enable them to grow faster and reduce poverty even further. This Report examines five pivotal phases of life that can help unleash the development of young people's potential with the right government policies: learning, working, staying healthy, forming families,and exercising citizenship.Within each of these transitions,governments need not only to increase investments directly but also to cultivate an environment for young people and their families to invest in themselves. The Report identifies three policy directions for help- ing youth develop themselves and contribute to society: expanding opportunities, enhancing capabilities, and providing second chances. Investing in young people strongly contributes to the Bank's overarching mission of fight- ing poverty. At the same time, investing in young people is a challenge for governments in all countries, rich and poor. It is my hope that this Report contributes to addressing this challenge by sharing the experiences of countries where young people, supported by good policies and institutions, have been able not only to cope but to flourish--and in the process, contribute to a future of hope and opportunity for all generations. Paul Wolfowitz President World Bank Group xi Acknowledgments This Report has been prepared by a core team led by Emmanuel Y. Jimenez and compris- ing Jean Fares, Varun Gauri, Mattias K. A. Lundberg, David McKenzie, Mamta Murthi, Cristobal Ridao-Cano, and Nistha Sinha. The team was assisted by Amer Hasan, Sarojini Hirshleifer, Natsuko Kiso, and Annette Richter, all of whom also contributed to drafting parts of the Report, as well as Mehmet Ziya Gorpe, Claudio E. Montenegro, and Victor Sulla. Additional contributions were made by Deon Filmer, Paul Gertler, Elizabeth King, and Peter Orazem. The work was conducted under the general guidance of François Bourguignon and Jean-Louis Sarbib. Extensive and excellent advice (including help in preparing background papers) was received from Jere Behrman, Robert Blum, David Lam, and Cynthia Lloyd, to whom the team is grateful without implication. Many others inside and outside the World Bank provided inputs and helpful comments (see the bibliographic note). The Development Data Group contributed to the data appen- dix and was responsible for the Selected World Development Indicators. The team would also like to acknowledge the generous support of a multidonor programmatic trust fund, the Knowledge for Change Program. Additional support was also provided by the Hewlett Foundation and the governments of Denmark, Germany, Japan, Sweden, and the United Kingdom. The team benefited greatly from a wide range of consultations managed by Maya Brahmam, Stephen Commins, Viviana Mangiaterra, Juan Felipe Sanchez, Gerold Thilo Vollmer, and Kavita Watsa. Consultations included events and workshops held in 26 devel- oping countries and a few developed countries in Europe, North America, and Asia, as well as online discussions of the draft. The team wishes to thank participants in these workshops, videoconferences, and discussions, which included staff in country offices, researchers, gov- ernment officials, staff of nongovernmental and private sector organizations and, of course, the young people themselves. Particularly noteworthy were the efforts in the three countries "spotlighted" in this Report: Brazil, Sierra Leone, and Vietnam. Rebecca Sugui served as senior executive assistant to the team, Ofelia Valladolid as pro- gram assistant, and Jason Victor as team assistant. Evangeline Santo Domingo served as resource management assistant. Bruce Ross-Larson was the principal editor. Book design, editing, and production were coordinated by the World Bank's Office of the Publisher under the supervision of Dana Vorisek, Susan Graham, Andrés Meneses, and Randi Park. xiii Methodological Note Writing about young people hours, others three days. In a few, members of the core team One of the biggest challenges in writing this Report was that for this Report engaged directly; in others, they were sent the evidence base was uneven. Data to carry out diagnostic reports, which are available on our Web site, www.worldbank analysis for some topics, such as youth citizenship and migra- .org/wdr2007. The team also engaged with representatives of tion, were limited. More importantly, there were very few global youth organizations and youth leaders in Europe,Japan, rigorous evaluations of youth programs and policies for any and North America, including members of the Bank's Youth, of the transitions and issues covered in the Report. To help Development and Peace Network and the Francophonie and address these gaps, the team used consultations and surveys, the Organización Ibero-Americana de Juventud (OIJ). New as described below. It also supported several impact evalua- media were used for e-discussions and videoconferences with tions that were either ongoing or could be concluded within youth leaders from around the globe, as well as for an open the Report's timeframe. youth commentary on the World Bank's Web site. The quotations in the margins of this Report were taken Listening to young people from the WDR youth consultations (see Mangiaterra and In one of the most elaborate and wide-reaching consulta- Vollmer [2006] and www.worldbank.org/consultations) and tions for a WDR, over 3,000 young people participated in the e-discussion connected to the Youth Social Technopreneur- focus group discussions in 26 developing countries: Argen- ship Conference in October 2005. For quotations taken from tina, Bangladesh, Brazil, Burkina Faso, China, the Domini- sources other than these, the source is given in an endnote or can Republic, the Arab Republic of Egypt, Georgia, Ghana, below the quotation. Honduras, India, Kenya, the former Yugoslav Republic of Macedonia, Mexico, Mozambique, Nepal, Nigeria, Papua Surveying young people New Guinea, Peru, the Russian Federation, Sierra Leone, Several of the issues covered in this Report are not covered by Thailand, Timor Leste, Turkey, Vietnam, and the Republic of existing developing country surveys. To provide insights on Yemen. We consulted youth workers and experts from gov- these issues, the Report team added questions to nationally ernments, nongovernmental organizations, partner organi- representative audience surveys by InterMedia in late 2005 and zations, and the World Bank to ensure that participants were early 2006 for Albania, Bangladesh, Ethiopia, Iraq, Malaysia, representative of the youth in their country. From Novem- Romania, and Tajikistan. In addition, rich data on information ber 2005 to May 2006, these young women and men debated and communication technology use and political attitudes were and discussed the Report's five life transitions. The process obtained from InterMedia's survey databank. was undertaken by the country offices, supported strongly A large new database was constructed from existing coun- by a joint team of the Children and Youth Unit of the Bank's try household surveys in 97 developing countries, covering all Human Development Network, the External Relations Vice- Bank regions and 21 developed countries. For all developed Presidency (EXT), and the WDR. A special effort was made countries and more than half of the developing countries, at to reach out to young people whose voice is often not heard, least two survey points per country, covering the 1990s and such as young women, rural youth, and those living with a early 2000s, were used for the cross-country comparisons. disability. The team found the discussions immensely help- These data complemented those from the Demographic and ful in forming hypotheses, validating quantitative findings, Health Surveys (ORC Macro) and Living Standards Measure- and providing a rich context. Some meetings lasted a few ment Study surveys. xv Abbreviations and Data Notes Abbreviations ABC Abstain-Be faithful-use Condoms MENA Middle East and North Africa region AGETIP Agence d'Exécution des Travaux d'Intérêt Public MTV Music Television AIDS Acquired immune deficiency syndrome NEPAD New Partnership for Africa's Development ART Antiretroviral therapy NER Net enrollment rate ASER Annual Survey of Education Report (India) NFHS National Family Health Survey AVU African Virtual University NGO Nongovernmental organization BMI Body mass index OECD Organisation for Economic Co-operation BRAC Bangladesh Rural Advancement Committee and Development CDC U.S. Centers for Disease Control and Prevention ORC Opinion Research Corporation CDCA Centro de Defesa da Criança e do Adolescente ORT Oral rehydration therapy CEDECA Center of Defense of Children and Adolescents PETI Program to Eradicate Child Labor (Brazil) CEDPA Center for Development and Population Activities PIRLS Progress in International Reading and CORFO Corporación de Fomento de la Producción (Chile) Literacy Study DDR Disarmament, demobilization, and rehabilitation PISA Program for International Student Assessment DHS Demographic and Health Surveys PRS Poverty Reduction Strategy DPT Diptheria, pertussis, tetanus PRSP Poverty Reduction Strategy Paper ECD Early childhood development SACMEQ Southern and Eastern Africa Consortium for EFA Education for All Monitoring Education EPL Employment Protection Legislation Sida Swedish International Development Agency EU European Union SMS Short Messaging Service FAO Food and Agriculture Organization of the SPW Student Partnerships Worldwide United Nations STD Sexually transmitted disease GDP Gross domestic product STI Sexually transmitted infection GNI Gross national income TIMSS Trends in International Mathematics and GNP Gross national product Science Study HIV Human Immunodeficiency virus UCEP Underprivileged Children's Education Program ICL Income contingent loan UNAIDS Joint United Nations Program on HIV/AIDS ICT Information and communication technology UNDP United Nations Development Programme IEC Information education and communication UNESCO United Nations Education, Scientific, and IFPRI International Food Policy Research Institute Cultural Organization ILA Individual learning accounts UNHCR United Nations High Commissioner for Refugees ILO International Labour Organization UNICEF United Nations Children's Fund IMF International Monetary Fund UNODCCP United Nations Office for Drug Control and INCAP Institute of Nutrition of Central America Crime Prevention and Panama USAID U.S. Agency for International Development IUD Intrauterine device WDR World Development Report LRA Lord's Resistance Army WHO World Health Organization xvi Abbreviations and Data Notes xvii Data notes The countries included in regional and income groupings in middle-income economies and thus may include economies in this Report are listed in the Classification of Economies table at transition from central planning, as a matter of convenience. the end of the Selected World Development Indicators.Income The term advanced countries may be used as a matter of conve- classifications are based on gross national income (GNP) per nience to denote high-income economies. capita; thresholds for income classifications in this edition may Dollar figures are current U.S.dollars,unless otherwise spec- be found in the Introduction to Selected World Development ified. Billion means 1,000 million; trillion means 1,000 billion. Indicators. Group averages reported in the figures and tables Serbia and Montenegro is used in this Report either because are unweighted averages of the countries in the group, unless the event being discussed occurred prior to the independence noted to the contrary. of the Republic of Montenegro in June 2006 or because sepa- The use of the word countries to refer to economies implies rate data for the Republic of Serbia and the Republic of Mon- no judgment by the World Bank about the legal or other status tenegro are not available. of a territory. The term developing countries includes low- and Overview Her performance is riveting. Geórgia, Across the ocean in Freetown, Sierra Half a world away, Van, 21, a third-year 15, an impoverished street child, failing Leone, Simeon, 23, is wondering what to student at one of Hanoi's most presti- student, and aspiring actress, moistens do next. For the past 15 years his life had gious tertiary institutions, is at a friend's the eyes of the international aid officials been disrupted by the long civil war. He house rehearsing Celine Dion songs visiting a halfway house for young girls and his family, living in Koidu, a hotspot with her band. A conscientious stu- in Recife, Brazil. She is playing the part of the conflict, had to flee for their lives dent, she passed the rigorous entrance of an abused young girl on a makeshift several times. They were once captured examination with the unconditional stage, where she dreams about coping and forced to serve the rebel forces for support of her parents, both profession- with the stresses of her life: the indiffer- two years. The impact of such unrelent- als. She earns extra money by trans- ence of her family, the difficulty of stay- ing exposure to violence is clear when he lating newswires from English to ing in school, the pressure to sniff glue, says that he feels as if he has already died Vietnamese on her home computer-- the unwelcome advances from men, the three times: when his father was killed for experience she hopes would help her enter part-time work as a housekeeper. Perhaps failing to provide a fighter with enough the journalism field. Her enthusiasm for she finds it easy to play the role because it food, when his mother was raped and playing computer games and surfing the mirrors her life so closely . . . but she also later died, and when his sister was forced Internet gives her uncommon self-con- has plenty of talent. to return to Koidu as a sex worker. fidence in technology. It also helps that After the play, as she speaks to the vis- Still, he wants to restart his life. Work- she is in almost constant contact with her iting foreigners,the precocious performer ing as a volunteer for a nonprofit organi- friends, thanks to Internet telephony and reverts to the shy, awkward teenager that zation that serves and rehabilitates youths instant messaging. she still is. She is thankful for the oppor- through counseling and education, he Her most immediate concern is that tunity to develop her craft in a safe space feels good to be part of a group and learn- her parents may not let her join her boy- while also improving her reading, writ- ing again, including how to operate com- friend on his newly acquired scooter to ing, and knowledge about life's practi- puters. He also wants to bring his sister to cruise Hanoi's streets on a busy Saturday calities. She is anxious about her future, Freetown and away from her nightmarish night. She knows that they rightly fear for especially how to get motivated for the life in Koidu. If only he had a paying job. her safety, having heard of several friends boring classes in the public school that who recently had serious motorcycle she occasionally attends. But for the first accidents. time in her young life, she is hopeful. There are many young people like Geór- quality of the next generation of workers, gia, Simeon, andVan--indeed, more of them parents, and leaders. Decisions about devel- than at any time in world history. Each is oping their skills, about starting on the road entering an age fraught with risks and laden to financial independence, and about engag- with opportunities, not just for them but for ing with the broader civic community will their families,their societies,their economies. have long-lasting effects that have repercus- Together, their experience will determine the sions far beyond them and their families. 1 2 WORLD DEVELOPMENT REPORT 2007 Most policy makers know that young they want and a chance to participate in people will greatly influence the future delivering it. of their nations. Trying to help, they face · Capabilities. Develop young people's dilemmas. When primary school comple- capabilities to choose well among these tion has gone up so dramatically, thanks to opportunities by recognizing them as public investment, why does illiteracy seem decision-making agents and by help- so persistent? Why do large numbers of ing ensure that their decisions are well university graduates go jobless for months informed, adequately resourced, and or even years, while businesses complain of judicious. the lack of skilled workers? Why do young · Second chances. Provide an effective sys- people start smoking, when there are very tem of second chances through targeted "Our parents never got the visible global campaigns to control it? What programs that give young people the is to be done with demobilized combatants, opportunities we have; it is up hope and the incentive to catch up from still in their late teens, who can barely read to us to make the most of them bad luck--or bad choices. but are too old to go to primary school? and achieve [something], and to Tough questions, these, and there are Invest in young people--now make sure we can look after our many more. The answers are important for parents in their old age." Thesituationof youngpeopletodaypresents growth and poverty reduction. This World the world with an unprecedented opportu- Young male,Development Report offers a framework and nity to accelerate growth and reduce poverty Dhaka, Bangladesh provides examples of policies and programs (chapter 1 of the report). First, thanks to the January 2006to address the issues. development achievements of past decades, Decisions during five youth transi- more young people are completing primary tions have the biggest long-term impacts school and surviving childhood diseases. on how human capital is kept safe, devel- However, to succeed in today's competi- oped, and deployed: continuing to learn, tive global economy, they must be equipped starting to work, developing a healthful with advanced skills beyond literacy; to lifestyle, beginning a family, and exercis- stay healthy, they must confront new dis- ing citizenship. The report's focus on these ease burdens, such as sexually transmitted transitions defines our choice of whom to diseases and obesity. Second, lower fertility include as "the next generation." Because rates in many countries mean that today's they take place at different times in differ- youths will enter the workforce with fewer ent societies, the report does not adhere to nonworking dependents, and thus fewer one defined age range, but it takes 12­24 to support. If they remain unemployed for years as the relevant range to cover the long periods, though, they could be a drain transitions from puberty to economic independence.1 on the economy. Young people and their families make Building on a stronger base the decisions--but policies and institutions of human capital also affect the risks, the opportunities, and ultimately the outcomes. Putting a "youth Because labor is the main asset of the poor, lens" on these policies, the report presents making it more productive is the best way to three strategic directions for reform: reduce poverty. This requires enhancing the opportunities to earn money and develop- · Opportunities. Broaden the opportuni- ing the human capital to take advantage of ties for developing human capital by those opportunities. Broad-based economic expanding access to and improving the growth is important.2 So is providing basic quality of education and health ser- education and health care, especially for vices; by facilitating the start to a work- children--to provide the foundation of ing life; and by giving young people a basic skills and well-being. Doing both has voice to articulate the kind of assistance brought significant progress.Primary school Overview 3 enrollment rates in low-income countries of skilled and educated workers in a more outside China and India rose from 50 per- integrated global economy may not neces- cent in 1970 to 88 percent in 2000. Average sarily lead to falling returns to skills. It may life expectancy at birth worldwide rose from actually boost the demand for skills even 51 years to 65 in less than 40 years.3 further by inducing faster skill-intensive With these advances come new challenges. technological change.5 Investment climate Further progress requires young people who surveys show that more than a fifth of all ". . . even the most low skill jobs are more capable and involved. But higher firms in developing countries as diverse as . . . require secondary school completion rates at primary levels strain the Algeria, Bangladesh, Brazil, China, Esto- completion, sometimes even capacity for places in secondary school (fig- nia, and Zambia rate inadequate skills and ure 1). Almost all Indonesian children attend education of workers as a major or severe university studies--even though six years of schooling, and 80 percent of even obstacle to their operations.6 The private there is no need for it. It leaves the poorest complete primary levels. Then, returns to secondary and higher education behind those kids who, for some however, enrollments drop dramatically, have been rising, especially in countries that reason or other, could not finish especially for the poor. Fewer poor Zambian have close to universal primary education. secondary school." children enroll to begin with, but they, too, There are also new challenges in health. Young person, fall off at secondary levels. Girls particularly Having survived the scourges of childhood, Buenos Aires, Argentina are left behind, just as they were in the expan- young people confront health threats at a December 2005 sion of primary education, except in South very vulnerable time, initiating sexual activ- America, Eastern Europe, and the former ity and entering the age of identity-seeking Soviet Union.4 Even more disturbing, the vast and risk-taking. In 2005, more than half the numbers spilling out of primary schools have estimated 5 million people who contracted not learned what they should. Standardized HIV worldwide were young people between tests--not just for science and technology 15 and 24, the majority of them young but for the command of basic skills--show women and girls (discussed in chapters 1 that students in developing countries lag and 5 of the Report). The economic effect of far behind those in the Organisation for such devastating diseases can be enormous. Economic Co-operation and Development In South Africa, HIV/AIDS can reduce GDP (OECD) countries (chapter 3). growth by as much as a fifth. It is by far the Concerns about the quality and rel- leading cause of death among young people evance of basic training come just when ages 15­29 in Sub-Saharan Africa. In other the demand for advanced skills, such as regions, noncommunicable diseases are now problem-solving abilities critical for many the leading cause of death for young women. industries, is increasing. Contrary to what Injuries caused by accidents and violence are might be expected, the greater availability the leading cause for young men. Figure 1 High enrollment rates in primary school are followed by significantly lower rates at secondary levels in Indonesia and Zambia Zambia (2001) Indonesia (2002) Percentage completing each grade Percentage completing each grade 100 100 80 80 60 60 40 40 Richest 20% males Richest 20% females Poorest 40% males 20 20 Poorest 40% females 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Grade Grade Source: Authors' calculations from Demographic and Health Surveys. Note: Quintiles are based on an index of assets and housing characteristics. 4 WORLD DEVELOPMENT REPORT 2007 Addressing these challenges will affect or more. They have more classically shaped poverty reduction far into the future for population pyramids with broad bases for at least two reasons. First, the capacity to the youngest ages, tapering up gradually learn is much greater for the young than with age. for older people, so missed opportunities These numbers can be a fiscal and eco- to acquire skills, good health habits, and nomic risk. A recent study estimates the the desire to engage in the community and yearly cost per secondary school student in society can be extremely costly to remedy. Sub-Saharan Africa to be almost three times Second, human capital outcomes of young that of public cost per pupil in the primary people affect those of their children. Bet- level.7 Add to that the cost of address- ter educated parents have fewer, healthier, ing AIDS and noncommunicable diseases, and better educated children. In all devel- and financing the fiscal burden, difficult oping countries, but especially in the low- to manage in the best of times, can be a income regions of South Asia and Sub- constraint on growth. Moreover, if youth Saharan Africa, immunization rates are remain unemployed for long periods, as higher among families whose mothers have happened when the baby boom occurred in some secondary education. These intergen- Europe and the United States, this not only erational effects lift families out of poverty wastes human resources--it also risks mis- over the long term. aligned expectations and social unrest that could dampen the investment climate and Seizing the opportunities from a growth.8 "youth bulge" in the population These large numbers can also be an The need to address youth issues now is opportunity. The fertility transition means also rooted in demographics--because of that many developing countries are in, the fiscal demands of the sheer number of or will soon enter, a phase when they can today's young and their share in the future expect to see a larger share of people of labor force. working age. This expansion of a workforce Today, 1.5 billion people are ages 12­24 that has fewer children and elderly to sup- worldwide, 1.3 billion of them in develop- port provides a window of opportunity ing countries, the most ever in history. This to spend on other things, such as building number will rise but not by much more, human capital. because it is fast approaching a plateau as The window of falling dependency rates fertility rates decline, producing a "bulge" can stay open for up to 40 years, depending in the world's population structure. Perhaps on the rate of fertility decline. Then aging as important as this bulge is the diversity in closes it. The good news is that almost all age structures across the world's countries, developing countries are still in this window due to differences in the timing of the fall in (figure 2). Of those that entered the window fertility rates. For developed countries, this early, some have taken full advantage, and fertility transition occurred so long ago that others have not. One study attributes more the bulge is composed of the middle-aged, than 40 percent of the higher growth in East the baby boomers. Their immediate chal- Asia over Latin America in 1965­90 to the lenge is how to ensure adequate and sus- faster growth of its working-age population tainable old age income support. and better policies for trade and human A few developing countries, especially capital development.9 If countries fail to those in transition in Europe and Central invest in human capital--which is most Asia,mirror developed country age patterns. profitable for the young--they cannot hope However, in most developing countries, the to reap this demographic dividend. number of young people is peaking or will Other poor countries in Sub-Saharan peak in the next 10 years. Others, includ- Africa, South Asia, the Middle East and ing all of Sub-Saharan Africa, Afghanistan, North Africa are about to see the window Iraq, West Bank and Gaza, and the Republic of opportunity open (figure 2). If they are of Yemen, will not hit the peak for 20 years to follow the Asian economies' growth Overview 5 path, they need policies and institutions Figure 2 Opening and closing demographic windows of opportunity that broaden the opportunities for young Window closed people to develop their human capital and Japan use it productively in work. Indeed, the Italy overall skills of the labor force, built largely in childhood and youth, strongly affect the Window will close < 10 years climate for investment in firms. And, where China enrollment in postprimary education is Year of Chile peak youth high, skill shortages, a feature of all devel- population oping countries, are lower. Window will close > 10 years India Getting it right today can have huge pay- offs for the future because young people, as Bolivia the next generation of household heads and Window has yet to open parents, will have profound impacts on their Afghanistan children.10 To illustrate, consider Kenya, Uganda where AIDS is projected to have very dam- aging effects on human capital investment 1955 1965 1975 1985 1995 2005 2015 2025 2035 2045 because premature parental death weakens Source: United Nations (2005b), medium variant. the mechanisms for forming it. Reinforcing Note: Bars show the range of years for which the dependency ratio--the number of dependents relative to people of these effects is the fact that higher expected working age--is falling. mortality among young adults in the next generation reduces the family's expected returns to educational investment. Public B O X 1 Investing in young people pays off big time: investment in young people, costly as it may Estimating the long-term and interactive effects be, is well worth it (box 1). of human capital investments Adapting an overlapping-generations would lead to income per capita that is 7 Investments during youth's five model that was used to estimate the macro- percent higher than without the interven- life transitions economic impact of AIDS, researchers have tion, with gains continuing far beyond 2040. recently applied it to a broader range of The net present value of the benefits, at The decisions that will affect young people's investments in human capital in Africa:"By plausible discount rates, would be between well-being and society's are those that shape killing mostly young adults, AIDS does more 2.0 and 3.5 times that of costs--a worth- the foundational human capital to be pro- than destroy the human capital embodied while investment indeed. in them; it deprives their children of the Because of the long-run synergy ductive workers, family heads, citizens, and very things they need to become economi- between postprimary education and the community leaders. That is why this report cally productive adults--their parents'lov- health of young adults, combining this focuses on the transitions that the young ing care, knowledge, and capacity to finance subsidy with direct measures to combat the undergo in learning, work, health, fam- education."11 AIDS epidemic and treat its victims would In a recent paper that explicitly models do better still.A program that combines a ily, and citizenship. If made well, decisions the effects of secondary education, the lower educational subsidy with measures to about these transitions will develop, safe- AIDS epidemic that shocked Kenya in 1990 combat the epidemic and treat its victims guard, and properly deploy human capital. is estimated to lower human capital and would, for the same amount of money, pro- per capita income so much that it does not duce even more dramatic gains.Those gains If made badly, the consequences will be very recover its 1990 levels until 2030.An educa- come not only from saving lives but also costly to correct because dropping out of tion investment--in the form of a 30-year from increasing the incentive to invest in school, prolonged periods of unemploy- program to subsidize secondary education education, a result of the reduced mortality. ment, or risky health behaviors can leave costing about 0.9 percent of GDP starting in 2000 and rising to 1.8 percent in 2020-- Source: Bell, Bruhns, and Gersbach (2006). permanent scars. Public policy can do much in determin- ing which way things go. As youth undergo each transition, the development of their Learning after primary school age human capital is hampered, not only by At the age of 12, more than 85 percent of poverty, but also by policy failures that all children in developing countries are in affect their options, especially the failure to school, a proportion that declines as they provide or finance adequate services when grow older (figure 3). (Figures 3, 4, 6, 7, and markets do not work. 8 are stylized representations of figures from 6 WORLD DEVELOPMENT REPORT 2007 Figure 3 Youth enrollment rates decline with age port because some of the gains to schooling accrue to society rather than individuals. Percent of age cohort 100 Governments also try to level the playing In school field for rich and poor, so that those with the greatest aptitude attain the most schooling. Some countries are more successful at 50 managing this transition than others. Sev- eral East Asian governments have done it so well that it has been called a miracle.12 Much as one would want to believe in mir- 0 acles, many other countries are much less 12 18 24 likely to see them. Why? Age · Despite dramatic recent progress in the Source: Authors. numbers completing primary school, a Millennium Development Goal, children chapter 1 that use actual data.) Almost all are are not learning as much as they should out of school by age 24.What they learn early (box 2). in life lasts a lifetime and is much harder to · Many completing primary school can- master if they try to do so as adults. not further their education because of a Skills are nurtured very early in life by lack of school places, a lack of resources, parents, who then turn their children over or a pregnancy--or all three. to preschools and primary schools. The · The global wave of economic and techno- expected path for young people is to con- logical change is demanding more from tinue on to secondary school, when decisions workers than basic skills. For example, in are made about whether to enter the labor many Latin American countries, supply force, marry, and go on to higher education. has not kept up with the rising demand These decisions may need government sup- for skills.13 The problem is not only one B O X 2 The poor quality of basic education severely limits opportunities for young people The dramatic recent progress in the numbers than the rates of knowledge would imply. Such reaching age 40 is projected to be 36 percent of children completing primary schools, a Mil- ignorance can be devastating. In Kenya, the in 2010--without AIDS it would have been 8 lennium Development Goal, does not fully probability that a 20-year-old will die before percent.14 address country needs because the children are not learning as much as they should. Many, There are big gaps in the book knowledge and life knowledge of young people even those who reach lower secondary levels, can hardly read or write and are unprepared to Percentage of young women ages 15­24 who are Percentage of young women ages 15­24 who know cope with the practicalities of daily life. In sev- able to read a simple sentence condom use prevents HIV/AIDS eral African countries, half or fewer of all young 60 60 women ages 15­24 can read a simple sentence after three years of primary school (chapter Zambia 50 50 3)--and in Ghana and Zambia after even grade Zambia 6 (left panel of figure). Even among those who 40 40 go on to lower secondary (typically grades 7­9), preparation is low. 30 30 The gap is not limited to book learning. Ghana 20 20 Many young people do not know basic facts Ghana that could save their lives, such as what causes 10 10 HIV/AIDS, at a time when many begin sexual activity. Knowledge about condom use is very 0 0 0 1 2 3 4 5 6 0 1 2 3 4 5 6 low regardless of grade attained in both high- Grade completed Grade completed and low-HIV-prevalence countries (far right figure).The risks are even higher Source: Authors' calculations from Demographic and Health Surveys. because the use of condoms is typically lower Note: Zambia has a high prevalence of HIV/AIDS. Overview 7 of insufficient numbers, but also of irrel- Figure 4 Entry into the workforce increases evant training. with age · Many young people are not motivated to Percent of age cohort exert effort because of poor teaching or 100 In school poor school environments. Working · Others are completing school late or not at all. In Mali, only about 20 percent of 50 15- to 29-year-olds have completed pri- mary school. In Malawi, more than half of all 19-year-olds in school are still at the primary level (chapters 2 and 3). 0 The skills that these young people need 12 18 24 and the ways they can be developed dif- Age fer from those of younger children pro- Source: Authors. gressing through the system on time. Starting a productive working life when they start full-time work prema- Once built, skills must be well deployed. turely, which is why many countries In almost all countries this process begins have agreed to international conven- sometime during the ages 12 to 24 (figure tions banning the worst forms of child 4). The start toward a working life is con- labor. Starting too early also prevents sidered by many social scientists as the most them from acquiring enough basic important marker of independence. But the skills in school--skills that would make transition is not always easy. Long spells of them more marketable to more types of inactivity during youth are costly--forgoing employers. the accumulation of skills on the job and an · Breaking into the job market. Across all employment history that could send posi- societies, starting an independent liveli- tive signals to future employers. Research hood is not easy, especially because the from OECD countries indicates that, while key to opening opportunities is one's many young people make up for such epi- track record--in work habits, job-related sodes, the scars are felt most by the less- skills, and repaying loans. No won- educated and disadvantaged (chapter 4). der that unemployment rates for youth "Graduates are a thousand One role for government policy is to cor- are systematically higher than for older times more numerous than rect for information failures that can inhibit generations (figure 5). In some middle- existing posts." opportunities--such as inaccurate assess- income countries with rigid labor mar- ments by a potential employer of an inex- Young male, ket institutions, the gaps between youth perienced young person's potential produc- Foum Zaouia, Morocco and adult rates are wide and can have tivity, skills, and work habits. A second is May 2005 enormous costs in forgone skill building. to ensure that poor families are not forced In Guatemala, for example, experience to have their children start work too early, in skilled jobs increases adult reading before they acquire basic skills. A third is comprehension and nonverbal cognitive to correct for the consequences of broader skills.15 policies, such as minimum wages that are · Moving to new jobs and up the skill lad- set so high as to diminish, unintentionally, der. Especially in poor countries, young the incentives to create jobs for relatively people are not idle--they are working inexperienced young people. very hard but earning very little. Chang- The main issues for countries in manag- ing jobs to earn higher wages or get- ing this transition for young people include ting into the formal sector is one way to the following: move up the skill ladder. For too many, · Starting work too early. Young chil- however, where they begin in work is dren and adolescents can be exploited where they end up. 8 WORLD DEVELOPMENT REPORT 2007 Figure 5 The unemployment rate everywhere is higher for young people than older people--with gaps much larger in some countries Unemployment rate (percent) 60 50 40 Regional average (youth) 30 Regional average (adult) 20 10 0 Middle Europe & East Latin America South Sub-Saharan East & Central Asia & & the Caribbean Asia Africa North Asia Pacific Africa Source: Authors' calculations based on Fares, Montenegro, and Orazem (2006a). Country names available on request. Note: The full length for each bar shows the unemployment rate of youth in a single country; the light segment shows the unemployment rate of adults in the same country. Adopting a healthful lifestyle countries young people begin to have sex If death rates are the benchmark, young before age 15, and fewer than half of sexu- people are a healthy group: the average 10- ally active youth use condoms (chapter 5). year-old has a 97 percent chance to reach The full effects of some of these youth- the age of 25. Mortality is a misleading ful behaviors on health will be felt only in measure of youth health, however, because adulthood. The negative consequences of it does not reflect the behavior that puts these early decisions can have far-reaching their health at risk later on. Youth is when consequences--depleting the economy of people begin smoking, consuming alco- productive human capital and increasing hol and drugs, engaging in sex, and having public health costs. more control over their diet and physical Because the (sometimes catastrophic) activity--behaviors that persist and affect healthconsequencesof thesebehaviorsshow their future health. In Nepal and Indone- up only later in life, they are much more dif- sia, almost 60 percent of all young males ficult and expensive to treat than to prevent. ages 15­24 are currently smoking. In many But for many young people, the search for a stable identity, combined with short time horizons and limited information, encour- Figure 6 Risky behaviors peak during youth ages them to experiment with activities that put their health at risk. For example, they Percent of age cohort 100 tend to discount too heavily the long-term In school negative consequences of smoking or sexual Working encounters. As young people experiment, they take more health risks; as they get Taking health risks 50 older, the tendency falls (figure 6). Reducing risk-taking among youth requires that they have the information and the capacity to make and act on decisions. Policies can do 0 much to help young people manage these 12 18 24 risks, especially if they make young people Age more aware of the long-term consequences Source: Authors. of their actions today. Overview 9 Forming a family amongtheyoungasitoncewas,micronutrient In most countries, first births, the entry into deficiencies are. And while the use of family parenthood,takeplaceduringyouth.Because planning, maternal, and child health services most births occur among women ages 15­ has increased in many countries, it remains 24--a result of the fertility decline in most low in others. Even where the use of services countries over the last two decades--the has increased, young women and first-time young are driving countries' demographic mothers often do not receive all services. transition. By the time girls in developing countries reach 25, nearly 60 percent of them Exercising citizenship have become mothers (figure 7). Boys make Youth is the period when people begin to this transition later, with most becoming be heard and recognized outside their fami- fathers between the ages of 25 and 29. lies (figure 8). They establish their identities The ability and willingness of young par- as individuals while they begin to interact "Young people will always ents to invest in their children is the single independently with the broader commu- have dreams; we hardly ever most important factor determining the out- nity. Simply by being community members, lose hope. For this, the role of comes for future generations. Young people they receive rights (to a fair trial) as well as tend to under invest in family planning or obligations (paying taxes). Citizenship also the State must be as a guide to maternal health services. That is why gov- emphasizes how individuals should hold shore up democratic practices, ernments may have to be involved in what public officials accountable for their actions, and to grant young people more would otherwise seem to be intimately pri- demand justice, and tolerate people from areas of political and business vate decisions.A further justification for gov- different ethnic or religious groups. The participation." ernment investment in supporting the tran- right to vote is commonly granted to those Flor de Maria, 24 sition to parenthood is to ensure equity. The who turn 18. The willingness and ability Peru percentage of women who gave birth before to exercise such a right and other duties of age 15 is significantly higher in the lowest citizenship are formed early in life and, once wealth quintile than in others for 15 coun- formed, tend to be durable (chapter 7). tries with data. Limited economic oppor- Without opportunities for productive tunities, poor access to services, and tradi- civic engagement, young people's frustra- tional norms surrounding sexual behavior, tions may boil over into violent behavior marriage, and parenthood may encourage and lead to economic and social instability, marriage at very young ages--even at age 12 sparks that can ignite long-simmering dis- or younger among some girls (chapter 6). putes. One of the initial causes of the ethnic Nutrition and reproductive health ser- conflict between Sinhalese and Tamils in vices are among the most important human Sri Lanka was the frustration felt by Tamil capital investments that prepare young people students, who had been shut out of places to become the next generation of parents. in universities and other avenues for civic Although undernutrition is not as widespread involvement.16 Formal political participation and Figure 7 Family formation increases with age involvement in social organizations, essen- tial for good governance, are, in turn, nec- Percent of age cohort 100 essary for private investment and growth. In school They broaden access to economic opportu- Working nities, especially among previously excluded groups, most obviously women. They can Taking health risks also enhance collective action to increase 50 the pressure on governments to provide good public services. Forming families 0 12 18 24 The transitions overlap (figure 8). Some Age young people have uncomplicated lives and Source: Authors. undergo only one or two of these transitions 10 WORLD DEVELOPMENT REPORT 2007 Figure 8 Civic engagement rises with age do with the "supply side"--a shortage of opportunities to access services and work Percent of age cohort 100 experiences that build human capital. Oth- In school ers have to do with the "demand side" Working because of behaviors that reflect a lack of information, resources, or experienced deci- Taking health risks 50 sion making. Both can be addressed by the right policies. All nations already have policies and pro- Forming families grams that affect the lives of young people. Exercising citizenship 0 They have schools, universities, labor market 12 18 24 regulations, hospitals, and laws that allow Age youth to vote. The strategies that frame Source: Authors. them are set mostly in well-established sec- tor departments. This report applies a youth at a time. Others may already be budding lens to evaluate them. Are those strategies multitaskers:they are in school,working part- appropriate to youth's life transitions and time, married, driving fast, and participating environments? Are they adequate to serve in their local council. This is why policies that youth's needs? change a decision in one transition can easily The starting point for this assessment is a affect the investment climate for human capi- solid foundation of human capital acquired tal formation in other transitions. before adolescence--an application of the Because basic skills in schools are learned principles of investment models long rec- early, the failure to invest in education can ognized in economics.17 Beyond support- greatly increase the costs of pursuing healthy ing the preparation for those under 12 years lifestyles and of working. By the same token, old, the report uses a framework that splits risky behaviors leading to a young man's the youth lens into three mutually support- premature death or a girl's unexpected ive lenses that focus policies and magnify early pregnancy can significantly lower the their impact (figure 9). The framework returns to schooling. Prolonged unemploy- extends the human capital model by treat- ment can lead to disinterest in investing in ing not only governments and families, but further schooling, possible mental distress, also young people, as potential investors. delayed family formation, and negative · The first lens focuses on the gaps in manifestations of citizenship. opportunities for building human capital These transitions can also have very dif- and on policies that help young people ferent trajectories across gender. At puberty acquire, improve, and deploy their skills. the transition for adolescent girls diverges · The second lens focuses on the capa- dramatically from that of boys. This event bilities of young people as they choose signals the potential of motherhood--asso- among the opportunities available to ciated with some societies' concerns about them and on policies that dispense the protecting girls, sometimes to the point of information and incentives to help them overly restricting them. It also marks when make good decisions. boys are expected to work for pay (a spotlight · The third lens focuses on remedying on gender follows chapter 2 in the Report). undesirable outcomes and on policies offering second chances that put young Policies should focus not only people back on the path to build their on youth's opportunities but human capital for the future. also their capabilities and Just as the three lenses have to be aligned second chances for an image to be in focus, so policies must Some of the challenges to forming human be well coordinated to have maximum capital during youth's transitions have to impact. Opportunities can be missed if Overview 11 Figure 9 Transitions seen through three lenses focus policies and magnify impact YOUTH LENSES Low Policies affecting literacy human capital and knowledge formation during of basic skills youth transitions: despite higher Inexperienced as · Economywide primary completion decision makers Consequences policies and of poor outcomes institutions: Insufficient access Still forming longer lasting macro stability, to secondary and identity and than for adults "Youth friendly" investment climate, tertiary preferences policies governance, labor Remediation market regulations Few job Lack of resources more costly for · Education opportunities adults · Training Myopia and · Health services No platform for risk taking · Welfare and civic family services engagement · Infrastructure Opportunities Capabilities Second chances the capabilities to grasp them are blunted numbers surviving childhood diseases and or misdirected. Having better decision- completing primary school grow. Not making capabilities (agency) can lead to addressing these challenges passes poverty frustration if the opportunities are far below to succeeding generations, because the poor aspirations. Not having second chances can outcomes of young people today are trans- lead to a free fall in outcomes. Some of the mitted to their children. Countries that have lenses loom larger in some transitions than broken out of this spiral have improved the in others. In the transitions toward sustain- basic skills of adolescents and young adults, ing a healthy lifestyle and forming families, met demands for even higher-order skills, for example, outcomes are influenced most and smoothed the start of young people's by young people's behavior, so the emphasis work and civic lives. would be on capabilities. Viewing economywide and sectoral poli- Improving basic skills-- cies through these lenses make them "youth intervene earlier in the friendly" by identifying gaps and setting life cycle and focus on quality priorities. The need to narrow gaps does The lesson from the massive education not necessarily mean that the benevolent expansion in the 1980s and 1990s is clear-- hand of government should do all the heavy expanding places rapidly can come at the lifting--even if well intentioned, many gov- cost of quality, reflected in high enrollment ernments lack the resources and capacity rates but low achievement. In Morocco and to provide all the necessary investments. Namibia, more than 80 percent of school Instead, public policy needs to improve the children stay until the last grade of primary climate for young people, with the support education, but fewer than 20 percent have of their families, to invest in themselves-- minimum mastery of the material (chap- by addressing the costs, risks, and perceived ter 3). Young people are already paying the returns of investing in people, just as they price; many of the large numbers of ado- should do for firms. The next three sections lescents completing primary education do fill the right side of figure 9 with examples not know enough to be literate and numer- of specific policies and programs. ate members of society. A youth lens thus reinforces the point--well documented Policies to broaden in worldwide monitoring reports18--to opportunities improve the balance between expanding Developing and deploying youth's human primary enrollments and ensuring a mini- capital become special challenges as the mum quality standard. 12 WORLD DEVELOPMENT REPORT 2007 "There are practical things in What should countries do? First, measure students, and local communities are only life that secondary school doesn't quality well.Quantity may have been stressed some of the measures (chapter 3). Because a even touch upon, for example, because it is easier to measure enrollment and focus on quality is not costless, what can be how to confront and resolve completion rates than learning outcomes. deferred until upper secondary and beyond problems." This is slowly changing with the introduc- is the selection and specialization that some tion of standardized tests comparable across countries have from early grades. In Chile's Young person, schools within and across countries.19 secondary education reform, for example, Honduras Second, consider the system of learning all vocational specialization moved to upper January 2006 over the entire life cycle rather than as stand- secondary school, better establishing a solid alone pre-primary, primary, secondary, and academic base. tertiary education. For many countries this means improving the foundations before Meeting the demand for higher the children reach adolescence through order skills--improve the early investments in nutrition, health, and psycho-social development. In countries as relevance of upper secondary diverse as Jamaica, the Philippines, Turkey, and tertiary education and the United States,20 enriched child care Even as countries struggle with basic needs, and preschool programs have led to higher the global economy demands more techni- achievement test scores, higher gradua- cal and behavioral skills, especially those tion rates from high school, and even lower formed during the ages 15­24. Competition crime rates for participants well into their has driven up the demand for skill-intensive twenties (figure 10). technological innovation in Asia and Latin Establishing the basic skills for a well- America, much of it in export industries, functioning society may require making which tend to use disproportionately more lower secondary school universal, as many young people (chapter 4). countries are doing. Again, this should This pressure can be eased if upper sec- not come at the cost of quality. Setting ondary schools and universities turn out standards, developing accreditation and more students. But simply increasing quan- evaluation systems, training and motivat- tity is not enough because content and how ing teachers, and increasing the account- it is delivered matters most. If quality is ability of school administrators to parents, low or if what is learned is not relevant in the job market, unemployment rates can be Figure 10 Early childhood interventions (at ages high even for some of the most highly edu- 1­5) can have long-lasting effects on young people cated. Countries like South Africa are try- (at ages 13­18) ing to respond to employers' demand for Percentage still in school quality and relevance by revamping upper 100 Turkey Jamaica secondary curricula to emphasize practical 80 thinking and behavioral skills and offering more of a blend of academic and vocational 60 subjects (chapter 3). Policies that link educa- tional institutions with prospective employ- 40 ers from the private sector through regular 20 consultations and joint university-industry research projects help, as in China. 0 Such reforms can be costly because of the Parenting No Psychosocial No skills parenting stimulation psychosocial high unit cost of educating students beyond training skills stimulation the basics.Teacher shortages in math and the training sciences are especially acute in Sub-Saharan Sources: Walker and others (2005) and Kagitcibasi, Sunar, and Africa. Efficiency gains may be possible Bekman (2001). Note: In Turkey, the intervention was for four years between ages 3 through better incentives for administrators and 9, and involved both parenting skills training and daycare. Only the parenting skills had an effect at the followup at ages 13­15. and teachers, such as performance-based In Jamaica, children ages 1­2 received two years of professional pay schemes now being tried in some Latin psychosocial stimulation, and follow-up was at ages 17­18. Both are controlled impact evaluations. American countries (chapter 3). Overview 13 Many educational systems can also more than twice the national average--truly expand and improve by diversifying their "youth intensive" sectors (chapter 4). They funding sources. Families already contrib- have had a particularly stimulating effect on ute significantly to the cost of tertiary edu- previously excluded groups, such as young cation in some countries--up to 80 percent women in Penang, Malaysia, whose entry of the cost in high-performing economies into the labor force 20­30 years ago fueled with relatively high enrollment rates such the growth in a fledgling electronics industry as Chile and the Republic of Korea--when and altered social stereotypes about women. they feel they are getting value for money. So policies that open the economy to free Countries such as the Slovak Republic, trade tend to be youth friendly. A youth lens Turkey, and Uruguay, where private funds would not necessarily change these poli- contribute 20 percent or less of the total cies--it would merely strengthen the argu- cost of tertiary levels,21 could mobilize ment for pursuing them in the first place. more resources through fees, public-private A youth lens may also mean that some partnerships, income-generating activities general policies need to be changed once (consultancies, leasing unused property), the implications for youth are taken into and donor support. account. Such is the case for labor market Programs that enable rich and poor to regulations that affect new entrants dispro- compete more evenly are only beginning to portionately. Policies that limit flexibility be tried. What is clear is that "free tuition" and mobility across sectors tend to lengthen for universities is neither financially sus- the transition to work and constrain young tainable nor directing benefits to the poor, people more than others. Employment pro- because so few poor students seek entry to tection laws in Latin American and OECD tertiary institutions. In Uruguay, more than countries can increase the unemploy- 60 percent of those enjoying free tuition in ment rate for young people (chapter 4). If public tertiary institutions come from the minimum wages are set too high, they may two richest quintiles.22 Loans or targeted discourage employment of the unskilled, scholarships to students from disadvantaged mostly young workers who are only begin- families could be both efficient and sustain- ning their working lives. These are not argu- able. For example, some countries subsidize ments for scrapping all such laws and regu- private secondary and tertiary institutions lations. Instead, they are a call to develop according to the number of low-income policies that provide adequate protection students they enroll. The institutions apply without stifling opportunities for already for the subsidies and are then screened to disadvantaged groups. meet quality standards (chapter 3). In poorer countries such as Burkina Faso, The Gambia, Nicaragua, Paraguay, Rwanda, and Sierra Leone, many of the young are Accumulating skills on the job-- more likely to begin work in the informal ease barriers to start work sector (chapter 4). Although this sector will and facilitate mobility not solve all issues of youth employment-- Broadening opportunities for young peo- even selling on the street requires some sales ple's employment works best when premised skills and language skills and conditions can on economywide growth that stimulates be harsh--evidence suggests that it can be a demand: a rising tide lifts young people's remarkably resilient and productive stepping boats, as well as everyone else's.23 In many stone, sometimes to formal employment. economies, an export orientation and for- If these jobs are to be the first rungs rather eign direct investment expanded the demand than the last stop up the skill ladder, youth for young workers. Such policies have been have to be able to move freely to seize the cited, along with sound basic education, opportunities that arise. Practical training as a source of growth to explain the "East that combines occupational and behavioral Asian miracle."24 In Indonesia, in heavily skills can make young people more mobile. export-oriented sectors such as electronics But the track record of schools and even and textiles, youth employment shares are large public national training institutions 14 WORLD DEVELOPMENT REPORT 2007 in providing such skills has, at best, been do not have the same access to social services mixed. Are there alternatives? Advanced as those with an urban residence permit. countries' experiences with formal appren- Nor do they enjoy the full range of social ticeships and internships, which provide a security and safety net benefits, contributing "structured work experience," hold lessons to social pressures. Policies that smooth the for middle-income countries that are rap- transition to cities and across borders, such idly developing a modern wage sector. In as recognizing the rights of migrants, would other countries, traditional apprenticeships be efficient and equitable as are now being in informal sector firms are more common, introduced in China. and incentives can be used to improve qual- Young people represent a large propor- ity and encourage innovation, as in Kenya's tion of the world's international migrants. Jua Kali program (chapter 4). Migration broadens the opportunities to Another option for the young is self- work (chapter 8). It also expands options employment. Some are entrepreneurs of for education, especially in higher educa- necessity, others by opportunity. Both types tion. Opportunities for secondary educa- face constraints made more binding by their tion back home also can be broadened age, such as access to capital and to business thanks to remittances from migrant parents networks. Programs to provide seed capital or other relatives. Measures in both send- to build financing and contacts have started ing and recipient countries can ensure that in Latin America. migrants take these opportunities (box 3). Geographic mobility also broadens "We only have observer status opportunities, and the young are a dispro- when things are decided in this Participating in civic life-- portionately large share of all migrants, both community." to urban areas and to other countries. In enhance youth voice in policy and service delivery Young person, China,118millionruraldwellershavesought Bonthe District, Sierra Leone employment outside their home villages, Opportunities to be recognized and heard as February 2006 and because of the rising productivity in citizens,and to be included in community ini- rural areas, this has not significantly reduced tiatives, are important for the delivery of ser- agricultural production. But the migrants vices that affect young people directly. World B O X 3 International migration offers opportunities and risks for youth The probability of migrating internationally Easing immigration restrictions for temporary Mexicans migrate young and return to peaks in the late teens and early twenties (see workers would be one way developed coun- Mexico young figure), so youth make up a disproportionate tries could help.To maximize the development share of the world's migrants--and often also a impacts,developing countries need policies that Percent large share of the return migrants.About a third 4.5 · increase the benefits from their existing of the migrant flow from all developing coun- youth migrants, such as lowering remittance 4.0 tries are ages 12 to 24, and half are ages 12 to costs, providing access to finance, and for 3.5 29.The youth bulge in developing countries and returning migrants, recognizing the skills 3.0 the aging populations in most developed coun- Returning gained abroad; 2.5 tries will further increase the demand for migra- · expand the opportunities for other youth tion over the coming years.Even so, much of the 2.0 to migrate by reducing high passport costs, youth demand for legal migration is unmet, with 1.5 removing legal restrictions on emigration,and 50 to 90 percent in some countries reporting 1.0 developing active bilateral work arrangements; that they would migrate, usually temporarily, 0.5 Migrating if they had the opportunity.So the young are · mitigate the risks of migration through infor- 0 more likely to resort to illegal migration and to mation campaigns to reduce trafficking and 0 10 20 30 40 50 60 70 80 90 fall victim to trafficking than older age groups. the spread of infectious diseases,and through Age Youth migration has large direct effects on broadening the opportunities for work at home; growth and poverty reduction through remit- · facilitate the return of migrants who have Source: Authors. Note: Shaded portion represents 12­24 age range. tances and the return of migrants with skills gained useful overseas experience by improv- acquired abroad.It can thus be a very important ing the investment climate at home. way for youth to use and develop their human cap- ital to help reduce poverty in their own countries. Source: Authors. Overview 15 Development Report 200425 referred to this comes. It is also a good way to develop skills "[The] majority of youth in engagement as "client power," arguing that it in decision making. Georgia now realize that the key can make providers more responsive to those factor . . . in finding proper jobs who matter--the beneficiaries. Voice also Policies that enhance capabilities: lies in themselves." matters because young people's disposition toward citizenship tends to be durable, and Youth as decision makers Young person, Tbilisi, Georgia participating early in life is a good predictor of The second youth lens focuses on the need December 2005 ability and willingness to engage in the future. to help young people decide more capably How can governments give that voice more among life's opportunities. When young prominence and not be threatened by it? people enter adolescence, the most impor- For most young people, direct consulta- tant decisions are made for them by their tion and recognition through their participa- parents and the elders in their families. As tion in policy decisions and implementation they get older, decision making shifts from may be a more important platform than, say, parents and families to youth themselves. voting. Such opportunities include not only The speed of the shift varies greatly for the state-sponsored channels but also social and different transitions. In some societies, the civil organizations, such as rural associations transition comes early. For many others, it in West Africa (kafoolu), samba schools and comes only for some decisions--and at an sports clubs in Brazil, and 4-H and Scouts older age. For some, such as young women everywhere. National service, whether mili- in traditional societies where decision mak- tary or civil, is another channel countries have ing simply shifts from parents to husbands, explored for active participation. The success- independence never comes. ful youth service programs emphasize flexibil- These differences are well illustrated in ity, youth input, accountability, and organiza- the responses to an international survey of tional autonomy in their design (chapter 7). 15- to 24-year-olds that queries who has the Often the young are more likely to have most influence on decisions about marriage, opportunities to be heard and have a more education, and occupation (figure 11). Very direct impact in local settings. In Ceará, Bra- few young Bangladeshi women think they zil, youth had a chance to review the state have the most influence on their schooling budget and identify initiatives previously or marriage choices. In contrast, Albanian, not on the agenda. The effect of broadening Malaysian,and Romanian youth feel remark- such opportunities is not simply to ensure ably empowered to decide for themselves. stability--it is often a way to get better out- Ethiopians, Iraqis, and Tajiks are mixed. The Figure 11 Who has the final say? The percentage of young women (ages 15­24) who feel they themselves have had the most influence on key life transitions varies greatly across societies Albania Bangladesh Ethiopia Iraq Malaysia Romania Tajikistan 53% Work 60% 50% 79% 43% 89% 84% 32% School 39% 82% 18% 52% 82% 65% 25% Marriage 77% 4% 55% 65% 82% 96% Source: WDR 2007 InterMedia surveys. Note: The figure shows the percentage of people ages 15­24 who answered "myself" (rather than parents, government, or other) to the question: "Thinking of [each transition: your current or most recent occupation, your years of schooling, and your marriage partner], who has had the most influence?" 16 WORLD DEVELOPMENT REPORT 2007 results for males mirror these proportions where traffic accidents are the leading cause with a few exceptions (chapter 2). Iraqi males of death and serious injury for those ages feel less in control about work and school 15­19 and where motorbike use by urban than females; Bangladeshi males feel signifi- young people exceeds 70 percent,only about cantly more in control than females about a quarter use a helmet--many are simply work and marriage, but not about school. not convinced of its protective value. Even if there is no outright independence, What can be done to better inform youth young people everywhere make important about the benefits and costs of investing in decisions that can affect their futures, even their human capital? Successful interven- in what are seemingly tradition-bound com- tions use schools, use the broader media, munities. Young primary school completers improve the content of dissemination cam- may dutifully enroll in secondary school to paigns, and harness new technology. please their parents but their own efforts will be important to their success. Young couples Improve the curricula in and convey the may be prohibited by laws from marrying too value of schools. Inculcating life skills in youngbutcanstillhavesexualencountersthat schools is the surest way to enhance the could lead to unwanted pregnancies. That is, capabilities of young people. This goes they are exercising their "agency," defined by beyond skills needed for further school- social scientists as the ability of young people ing and work. School-based reproductive to define their goals and to act on them.26 health education programs can increase Agency needs to be recognized. This knowledge and the adoption of safe sexual might be as simple as ensuring that young behavior.28 A school-based sex education people have legal identity, including the intervention in Kenya--providing young basic documentation often crucial for girls with specific information, such as the access to basic services. But recognition prevalence of HIV infections among older alone is not sufficient--agency must be men--reduced pregnancies (chapter 5). informed, resourced, and responsible. That And there is no evidence that sex education is, if it is to help young people take advan- increases sexual activity among youth. tage of existing opportunities, it must be However, it is not enough just to intervene "capable." What determines this capabil- in schools,because so many youth in develop- ity? Access to information, command over ing countries drop out. This is often because real resources, and the ability to process of poverty, but it may also be that young peo- and act on the information. Policies can ple are not well informed about the benefits help in all three. of continuing their education. In the Domin- ican Republic, simply telling young boys Informing youth about the "real" earnings premium to edu- Young people know a lot. Their literacy cation, a fairly cheap intervention, increased rates are much higher than before. They are secondary completion (box 4). also much more frequent users of one of the most ubiquitous sources of information Examine options outside schools. How best now available--the Internet (chapter 8). to target those who have dropped out or However, there is much more to know and never sought education? Success in contain- young people's knowledge base to inform ing the spread of HIV/AIDS in Cambodia key decisions about human capital invest- and Thailand is associated with structured ment and risk-taking behavior is often information campaigns that worked through deficient. Recall the low rates of knowledge media and information providers.29 of condom use among young females in While it is difficult to establish that a cam- Africa, even those who are educated (see paign causes the desired effects because of box 2). A survey of otherwise well-informed many other influences, a few rigorous stud- and educated Vietnamese youth ages 14­25 ies have attempted to solve the attribution conducted in 2003 indicates that fewer than problem. The Better Life Options program 60 percent of rural youth had ever heard provides a combination of various services of syphilis or gonorrhea.27 In a country to young women (ages 12­20) in periurban Overview 17 slums and rural areas in India.It disseminates information on reproductive health and B O X 4 Knowing what's good for you: Telling young people services, provides vocational training, and about the benefits of school can affect outcomes promotes women's empowerment through recreational events and dissemination of Do young boys know the value of school- were 12 percent more likely to be attending ing? Not always. In the Dominican Repub- school in the following school year relative information material. A multivariate analy- lic a survey of boys in 2001 enrolled in the to those who did not. sis indicates that those in the program were final year of primary school compared the The measured actual earnings gains for significantly more involved in key life deci- returns they perceived to continuing their completing secondary education in the sions--such as spending in the household, education with the actual returns in terms Dominican Republic are much higher of differences in earnings from age-earn- when to marry, and whether to continue than the perceived gains ings profiles. It found that they accurately education--than those who were not.30 estimated the returns to completing pri- mary school (but not completing second- Incomplete secondary Harvest worldwide knowledge through ary school) consistently with estimated Complete secondary earnings profiles. However, they severely new technologies, such as the Internet, to Weekly earnings (Dominican pesos) underestimated the returns to completing 5,000 inform youth. A better climate for private secondary school.The measured actual investment in technology is important for average earnings gains (from surveys) 4,000 young people. They are especially likely to between secondary and primary comple- tion, at about 1,300 Dominican pesos use communal modes of access, such as 3,000 (about $200), was 10 times the perceived Internet cafes, so government regulations to gain of 140 Dominican pesos (about 2,000 allow easy entry of firms into these sectors $21) (figure).The differences were most would help. A reform of the licensing pro- pronounced for the youth in the poorest households. 1,000 cess in Algeria that made it more affordable Some students at randomly selected to obtain authorization to provide Internet schools were then given information about 0 Measured Perceived service led to an explosion of Internet cafes the estimated actual earnings profiles. earnings earnings between 1998 and 2000 (chapter 8). How- Follow-up surveys in 2005 indicate that those who were given the information Source: Jensen (2006). ever, because much of the information on the Internet would be in a foreign language, efforts to kick start local content are needed. many societies, especially as the young seek Many youth also need guidance on how to their identity, peers could have at least as avoid the risks of using the Internet and much influence on decisions as families or to learn how to find reliable information schools. So, informing a young person can among the mass of content available. have spillover effects on others. Programs have begun to include them as part-time Improve the delivery and management of service providers, such as Jamaica's Health information to ensure that what should Ministry, which encourages peer-to-peer be taught is taught well. Some who pur- learning to combat HIV/AIDS. port to be trainers are often poorly trained themselves. Addressing this in developing Helping young people countries requires training the trainers bet- command resources ter and improving their incentives. School- Because young people are only beginning to based career guidance services show some be financially independent, they naturally promise in Chile, the Philippines, Poland, confront more constraints on their con- Romania, the Russian Federation, South sumption and investment decisions. Indeed, Africa, and Turkey. One consistent finding one of the reasons that the age of leaving is that success depends on the information home is getting later even in rich countries available to the counselors (chapter 3). is that young adults rely on their families to The consequences of ill-informed con- get on a firmer economic footing. For those clusions, because of the "noise" in the infor- from poor families, or for those who for mation flow, can be profound. Many studies one reason or another (orphanhood, bro- show that young people tend to overestimate ken families) can no longer rely on family the amount of sexual activity and high-risk resources, the result could be a rocky start behaviors in the population, putting more on the road to a sustainable livelihood-- pressure on them to conform (chapter 5). In and for young females, a weak negotiating 18 WORLD DEVELOPMENT REPORT 2007 "My situation is difficult; I can't position within families, especially regard- schemes, which are worth monitoring and go to school because I am the ing marriage and childbearing. evaluating. For countries with poorly devel- main breadwinner of my family. Choosing to invest in skills presents sub- oped income tax systems, alternatives such In rural areas, you do not have stantial costs to young people. Out-of-pocket as targeted school vouchers and individual opportunities to learn after costs tend to vary--for the half of all univer- learning accounts, which encourage savings sity students in private universities in Argen- for education, may be better (chapter 3). school hours . . . I did not pass tina, Brazil, Chile, and Colombia, they range The income constraint is binding in the university exam." from 30 percent to 100 percent of GDP per poorer countries even for secondary educa- Dang, 25, capita.31 Even for students in free public uni- tion. Because parents are the main means of Bac Can, Vietnam versities the opportunity costs are substantial. support for young people at this age, some May 2006 Because of the big personal payoffs to higher subsidies to encourage enrollment target education, such costs would not be a bind- the household, but the transfers are condi- ing constraint if liquidity were not an issue. tional on achieving youth-related outcomes. But it is. A recent study from Mexico showed Mexico's Oportunidades provides such an that households are less likely to send their incentive by giving larger transfers to house- offspring to university if their income falls holds if young females (versus males) stay "In my area, if you can pass the temporarily, even if their permanent long- in school. In rural areas, girls' enrollment university entrance exam you term income remains unchanged.32 in all three years of middle school increased continue to study. If you fail, The obvious way to lift this constraint is much more than that of boys (figure 12). you join the army or stay at to provide credit. Because commercial loans Some innovative programs have chan- home to breed buffalo." arenotavailabletothepooreststudents,who neled subsidies directly to students, particu- do not have the collateral or parental guar- larly young girls--partly as an inducement Hoang (Dang's friend), antees to back them up, such credit schemes for them to perform well in school but also Bac Can, Vietnam could not function effectively for students to ensure that they "own" the decision to May 2006 without government support. Moreover, the attend, circumventing age-old biases against pressures on the young to begin a livelihood girls' schooling. The Bangladesh Female Sec- are high enough even without having to pay ondary Stipend Assistance Program (FSSAP) back a debt that is many multiples of ini- targets girls ages 11­14, transferring a tial earnings. And many public institutions monthly payment to bank accounts in the have found it difficult to administer such girls' names, contingent on them staying schemes because of low repayment rates, unmarried and performing well enough to especially given the many episodes of youth pass in school (chapter 6). This program unemployment. Australia has pioneered a has yet to be rigorously evaluated, but it has system that makes repayment contingent been associated with the enormous increase on graduates' incomes, as tracked in tax in girls' enrollment there.33 Such incentive- systems. Middle-income countries such as based schemes may not work too well if they Thailand are only now starting to try such force the provider to only increase quantity and not quality. Concerns about learning Figure 12 It pays to go to school outcomes in the first round of FSSAP are being addressed in subsequent programs. Percentage increase in middle school enrollment due Such programs do more than help young to Mexico's Opportunidades 1997­2001 40 girls go to school. They also highlight the Boys Girls role of resources in enhancing the capability 30 of young women as decision-making agents within the family. Some early marriages are 20 arranged more for the convenience of fami- lies and parents than for the young couple. While many are successful, there still are out- 10 rages, illegal in almost all countries, such as the exchange of very young brides, some less 0 1 2 3 than 12 years old, to much older men for a Middle school grade debt cancellation. Poor young women are Source: Parker (2003). also more likely to succumb to pressure to Overview 19 accept money or goods in exchange for sex, Sometimes changing attitudes requires increasing their risk of sexually transmitted changing the environment for learning--to infections. Helping them earn a living or pro- thwart entrenched practices, some not even viding credit and savings may empower them recognized, that channel behavior. Research to withstand the pressure to put their health from secondary schools in Thailand indi- and future at risk (chapter 6). cates that girls who study in single-sex envi- Young people, unable to earn a living by ronments do better in math and develop working for others, often work for them- more leadership skills than girls in mixed- selves. Some seize opportunities voluntarily; sex environments, a finding consistent with others do so out of necessity--but the bar- research in developed countries.34 Another rier of finance is common to all. It takes example includes U.S. residential job-train- money to make money. The young, even if ing programs, which allow participants to they have the brightest of prospects, have avoid interactions in their own disadvan- no credit ratings, are inexperienced, and can taged neighborhoods and thus develop their offer no collateral to borrow money to start self-image. The programs are more success- a livelihood. More programs now try to help ful than those that try to save money by not such young entrepreneurs, but they still need offering the expensive housing component.35 to be evaluated rigorously. Early lessons from Coming to the correct decision can also Latin America's Endeavor programs, pro- be influenced by incentives, especially if the viding financing and technical assistance to young do not take into account the effects on young people, are promising (chapter 4). others (or on themselves in the long run)-- even if they had the information and knew Enhancing the capacity to decide well how to decide for themselves.In Indonesia,an Young people, once resourced and informed, increase in the price of cigarettes is estimated still have to filter and assess the informa- to affect consumption for the young more tion--so much more of it, given the Inter- than it does for adults, a finding consistent net--and consider the consequences of their with those in richer countries.36 By contrast, actions. The process of evaluating informa- banning cigarette and alcohol advertisements tion and acting on that evaluation is difficult and setting minimum ages for drinking enough even for the most confident. For attempt to increase the"price"of such behav- many young people, it is more complicated iors, but their impact tends to be weak. because they are still seeking their identities. The process of developing a person's sense Policies to offer second chances of self has been used by some economists Broadening opportunities available to young to explain seemingly irrational acts that people and helping them choose wisely put human capital at risk, including youth's among them are the priorities, especially tendency to engage in risky behavior, such becauseremediationiscostly.Yetmanycannot as joining violence-prone gangs, despite take advantage of these opportunities, even if "[Many] of my acquaintances the low expected economic payoffs to such offered. What can be done for the 19-year-old [whom] I met during some activities (chapter 2). whose poor parents took her out of school at [previous] phase of my life . . . The ability to process information starts the age of 9? Or her twin brother who is still in are dead, in jail or crippled. to develop early in school. But many educa- primary school? What about the unemployed Marcos [a youth activist] tion systems fail because they emphasize rote 23-year-old who has just started a family but introduced me to the [second learning of facts.Few emphasize thinking and never learned any on-the-job skills because he chance] movement. . . . I've behavioral skills--motivation, persistence, could not find work in the formal sector? Or participated in meetings, met cooperation, team-building, the ability to the petty pickpocket incarcerated in an adult the people. . . We've started to be manage risk and conflict--that help individ- jail along with hardened criminals? What are integrated as a community, [to] uals process information and come to sensi- the options for those who have been dealt life's ble,informed decisions.These programs have proverbial bad hand? develop a conscience . . . to get been well tested in developed country settings Some young people have undesirable out of the drug trade." such as the Netherlands and the United States outcomes because they have had their Bruno, 21, and are now beginning to be tried in develop- opportunities restricted--others, because Ceará, Brazil ing countries as well (chapter 3). they chose the wrong path. More than half May 2006 20 WORLD DEVELOPMENT REPORT 2007 of all HIV/AIDS infections occur among increased the uptake of STI services among young people under 25, most instances due sexually experienced students and reduced to voluntary sexual behavior or intravenous the incidence of STIs (chapter 5). drug use (chapter 5).Half of all murders and violent crimes in Jamaica are committed by Integrating second chances with young males 18­25, who make up 10 per- mainstream programs cent of the population.37 This is not totally Countries have programs that try to miti- unexpected, because behavioral research gate the effects of undesirable human devel- shows young people to be less risk-averse opment outcomes for youth: youth rehabil- than older people38 and criminal activity is itation programs, treatment programs for a risky business. those infected with communicable diseases, Policies that help youth recover from bad and retraining programs for dropouts. outcomes can provide a safety net and benefit Many of them are small and disconnected society well into the future. Referred to in this from each other, risking very costly paral- Report as second chances, they must be well lel programs. Worse, they may not allow for designed, well targeted, and well coordinated, reentry into mainstream programs. and give the right incentives to beneficiaries. Coordination is the key. Graduate equiva- lency, for example, allows dropouts to take Targeting programs finely classes that will eventually get them the equiv- Because of the high cost of second-chance alent of a primary or secondary diploma. programs, it is important to direct them Even without the paper certificate, getting the to the neediest youth, such as orphans and equivalent skills would help. The Underprivi- those from families too poor to provide a leged Children Education Program (UCEP) safety net. Interventions that use means in Bangladesh helps 10- to 16-year-olds who testing, geographic targeting, and self-selec- have dropped out of primary school--the aim tion are all associated with getting a larger is to educate them for three years and direct share of the benefits to the lower two quin- them into UCEP-run vocational programs. tiles of the population. Studies show that UCEP, which served 36,000 Aside from avoiding errors of inclusion students in 2002,has costs per student roughly (giving subsidies to those who do not need the equivalent of the regular school system them) it is important to avoid errors of (around $20 per year). At the tertiary level exclusion--leaving out young people who the community college system in the United need to be reached. This is especially true States, designed originally to provide second regarding health risks, because some risky chances for adults, is now being used increas- behaviors may not have an immediate and ingly as a second-chance program by young discernible impact on health. More than 100 high school graduates--three-quarters of all million young people are afflicted by sexu- remedial students are in community colleges. ally transmitted infections (STIs), including Mainstream programs have to be flexible HIV, every year. Some infections are easily so that early mistakes do not turn into per- treated if identified early. Many, however, manent liabilities. Some countries stream go unnoticed (especially when initial symp- students as young as 10 into differing ability toms are mild) unless diagnosed by trained schools, while others keep the schools com- health workers. In South Africa, many prehensive. A recent study across 18 coun- reproductive health services are not easily tries, comparing the performance of these accessible by young people; when they do students on standardized international go, they feel that facility staff are judgmen- secondary-level tests, finds that early track- tal and hostile. It might not be surprising, ing not only increases education inequality then, that those who contract STIs would (there is no catching up despite the segrega- rather go to traditional healers than to low- tion) but may also lessen performance.39 quality and high-cost formal services. In Nigeria, providing school students with STI Rehabilitation with accountability education and training pharmacists and Rehabilitation is very costly, but the payoffs private doctors to treat STIs in adolescents are highest for young people who still have Overview 21 a lifetime of potential productivity ahead engage in more risk-taking than warranted. of them. For those who commit crimes at Some express fear that the availability of a young age, they must be made to face the antiretroviral therapy could cause young consequences of their actions without being people to take fewer precautions. Indeed, in made to lose hope. Many of these young Kenya, condom use fell after the government people--some with relatively minor misde- announced reported "cures" for AIDS.41 The meanors, some simply homeless--are often solution is not to deny second chances like incarcerated along with hardened crimi- treatment--that would be unethical as well nals. In the United States, where more than as wasteful. Instead, it is to build in incen- 10,000 juveniles are housed in adult facili- tives that encourage the care-taking behavior ties, harsher prison conditions are associ- to persist even as people undergo treatment. ated with higher recidivism rates (chapter Programs that enhance both capability and 7). Consequences should be commensurate second chances are more likely to succeed. with the gravity of the crime, and programs This concept is well illustrated in voca- should facilitate the reintegration of these tional training programs for out-of-school young people into healthy and productive youth. In a variety of country settings the roles in society. programs tend not to pass cost-benefit tests. Some interventions (chapter 7) have However, when training is provided as part overcome the limited capacity of court sys- of a comprehensive package that gives recip- tems, such as the Justice on Wheels program ients the incentives and information to find in the Philippines, where judges travel to jobs--such as employment services, coun- correctional facilities around the country to seling, and life-skills training--they have speed up the trial process. As an alternative better outcomes. Jovenes programs in Latin to traditional retributive justice, more than America, targeted to training disadvantaged 80 countries have restorative justice pro- youth,ages16­29,canhavesignificanteffects grams to promote rehabilitation. The pro- on employability and earnings. The training grams provide opportunities for victims and may also be costly, but the costs compare offenders to meet, and to agree on a plan of well to other human capital development restitution. The most famous of these is the programs for young people (chapter 4). Truth and Reconciliation program in South Africa instituted after apartheid. Moving forward Young people have been involved in con- The broad policy directions recommended flicts in every part of the globe: estimates in this report--divided according to the suggest that 300,000 under the age of 18 youth lenses of opportunities, capabilities, "When it comes to `youth are now or have recently been involved in and second chances--are summarized in armed conflict, and another 500,000 have table 1. Some of the actions and programs making a difference in been recruited into military or paramili- require a reallocation of resources. These communities' I think the tary forces.40 Experience with disarmament, include the recommendations to attend to value of youth has been demobilization, and rehabilitation programs quality in the development of basic skills underestimated everywhere. shows that it is possible for young combat- for adolescents and young adults, as govern- Youth are excellent in delivering ants to reconstruct their lives in peacetime. ments press ahead with meeting quantitative grass roots levels development Ex-combatants clearly require skills training targets for children. In countries that have projects at minimal budgets and to preparethemforpost-conflictlife,butthey already met their quantity and quality tar- very effectively. Due to the fact also need medical and psychosocial support gets, the priority is to expand access to upperthat they are involved at the to overcome the traumas they have suffered. secondary and tertiary education, especially grass roots level, they can easily Young female ex-combatants may have a dis- by stimulating the demand for education. implement a project without the tinct set of needs that may not be addressed Public spending alone will not do the bureaucracy of organizations . . . by programs designed for male soldiers. trick. Policies must stimulate young people, Any remediation program confronts their parents, and their communities to they often have a lower cost what economists call moral hazard. If some- invest in themselves. The Report describes base too." one knows that the consequences of risk-tak- the failures in markets, institutions, and Shasheen, 20, ing behavior are mitigated by a government policies that contribute to an unfavorable Australia program or by insurance, that person might climate for human capital investments in June 2006 22 WORLD DEVELOPMENT REPORT 2007 Table 1 Policy goals, directions, actions, and programs Broaden opportunities for young people Develop capabilities of young people Offer second chances to manage consequences of to develop their human capital as decision-making agents bad outcomes that occur early in life Policy goals Policy actions and programs Policy goals Policy actions and programs Policy goals Policy actions and programs Children enter Improve quality at primary Young people Targeted scholarships based Allow young people Demand-driven programs adolescence with and lower secondary have appropriate on merit and need, conditional to regain access that help youth reenter basic skills for command over on outcomes (Bangladesh to services that mainstream education Universalize lower further learning and resources that Female Secondary Stipend safeguard and develop systems (graduate secondary practical living affect human Program) human capital equivalency) Redesign inflexible capital decisions Micro credit for youth Treatment for HIV/AIDS for educational systems to young people be more diverse and to Income-contingent loans (e.g., integrate academic with Australia and Thailand) Retraining programs linked life skills (Chile Education well to labor demand (e.g., Reform) Young have School-based information, Latin America's Jovenes) sufficient education, and communication Motivate teachers with and accurate campaigns (Kenya) Give hope to those Restorative justice and incentives information who have committed rehabilitation programs Better Life Options Curriculum Address demand-side about human crimes or who were that are cost-effective (e.g., (India) constraints among capital needs combatants in armed Romania, South Africa Truth girls through women and constraints, Job counseling programs (e.g., conflict and Reconciliation) teachers, improved school and programs to Philippines Overseas Workers environments address them Program) Curriculum reform to stress Young people enter Align minimum wage with noncognitive skills training labor force at the market realities right time and are Break down barriers to Support Include students in school mobile to be able to mobility (e.g., relax overly decision-making decision making accumulate higher- rigid employment protection, by recognizing order skills Cash transfers conditional residence regulations) identity, and on outcomes (e.g., Mexico's giving incentives Oportunidades) All young people Recognize youth as to shift behavior are given a voice in significant stakeholders in Cigarette taxes civic life public institutions and as legal entities (e.g., policy consultations in Ceará, Brazil) young people.The good news is that reforms · Better coordination and integration with to correct for these failures may not be as national policy. Youth issues by nature costly to the public purse as direct invest- cut across sectors, while most policies ment. The bad news is that they may require that influence them are set within sec- more difficult political trade-offs. For tors (box 5). So the challenge of coordi- example, the returns to investing in young nation looms large. Countries that have people would be substantially enhanced by experienced success are those that have trade and labor market reforms that deploy drawn up a coherent national frame- human capital more efficiently through work for youth, supported by all min- more open competition--but this may istries. This framework needs to be well threaten older workers who would like to integrated into national policy planning maintain their entitlements. Measures that and budgeting (like the poverty reduc- enhance capability and offer second chances tion strategy processes), rather than may be controversial too. Some societies see be seen as standalone programs run by decision making in the hands of the young underfunded and overmandated youth as a threat or consider it too costly to miti- ministries, which are more effective as gate the effects of bad decisions, even if the coordinating bodies. young were not responsible for them. · Stronger voice. Young people's lack of If countries are to mobilize the eco- voice means they are a weak constitu- nomic and political resources to stimu- ency for reform. Parents do not repre- late reform, they will have to resolve three sent the views and aspirations of young issues (chapter 9): adults like they do for younger children. Overview 23 Yet youth may lack the opportunities or self-confidence to represent themselves B O X 5 All dressed up and somewhere to go in in public forums. Young people need to Bungoma and Butere-Mumias be encouraged to participate more fully in public life. Governments and other Providing school uniforms in Bungoma A recently concluded randomized trial in and Butere-Mumias Districts in Kenya Kenya evaluated a multisector program agencies need to learn to communicate not only lowered drop out rates--it also that gave incentives to young people by with them, make their programs attrac- delayed childbearing among teens providing them with school uniforms, a sig- tive to them, and tap their immense tal- nificant expense for young people and their Percent ents as partners in service delivery. 16 families.It showed that lowering the cost of education by providing these uniforms had · More evaluation. The dearth of rigor- 14 a strong effect not just on male and female ously evaluated youth-oriented pro- dropout rates--it also lowered the percent- 12 grams and policies can undermine the age of teenage females who have started 10 childbearing (figure).This effect was greater credibility of them, even if most are than that of a program aimed specifically at promising. Preparing this Report was 8 early childbearing--a teacher training pro- a challenge because of this gap--stud- 6 gram for Kenya's HIV/AIDS curriculum.Such ies like the one in box 5 were fairly rare. spillover effects might not figure in if one 4 were to take only a sectoral perspective. Addressing this gap requires capacity 2 building in government and incentives to use evidenced-based criteria in deciding 0 Male Female Females among programs. Because such knowl- dropout dropout who have Control edge is a public good, it also requires rate rate started Students receiving school uniforms childbearing international funding. The policies and Source: Duflo and others (2006). programs mentioned in table 1 include Note: The differences between treatment and control groups are statistically significant. not only those that have been rigorously evaluated, but those that appear to be promising based on professional judg- ments. Otherwise it would have been a patchy table indeed. identifies other "actionable" concerns about the development of youth's human capi- Some issues raised in this Report may tal--concerns that, if unaddressed, can put never be resolved.After all,parents have been all development at risk. Fortunately, exam- complaining about their teenagers (and vice ples abound of young people, supported by versa) for a long time. Such issues are part of good policies and institutions, who have not human maturation and outside the realm of only coped but flourished, thereby contrib- development economics. But the Report also uting to the future of all generations. Launched in 1997 by Chilean university stu- dents as a summer project to provide mini- mum living conditions for the poor, a pro- gram mobilized 4,000 volunteers to build 2,156 homes in the poorest parts of south- ern Chile in the first two years. By 2004, Un Techo Para Chile had built more than 24,000 basic housing units and recruited more than 18,000 volunteers each year. The group has also developed initiatives on technical training, education, microcredit, health, judicial services, and community service. The initiative has also spawned Un Techo Para mi País to share its experiences with other countries across Latin America. (http://www.untechoparachile.cl/) Why now, and how? DEVELOPING COUNTRIES' POPULATIONS HAVE more young people than ever; indeed, in many countries, there are more young people now than there likely will ever be, because of falling fertility. This presents P A R T I challenges but many more rewards--if policies and institutions for the next generation of workers, household heads, citizens, and leaders are well designed and implemented. Hard to do, but, as some coun- tries have shown, not impossible. Opportunity. The capability to choose. Another chance when choices go wrong. These are the lenses through which policies must be viewed to examine whether they create the right climate for invest- ment in the human capital of the young. Youth, poverty reduction, and growth 1 The developing world's 1.3 billion young opportunity to invest more in their human people ages 12­24 are its next generation capital because of the decline in the ratio of of economic and social actors. Making dependents (children and elderly) to work- c h a p t e r sure that they are well prepared for their ers. This could even lead to a demographic futures--as workers,entrepreneurs,parents, dividend--an acceleration in the rate of citizens, and community leaders--is thus growth, as witnessed by some countries in enormously important to the course of pov- East Asia--if the right policies are in place erty reduction and growth. Because human to employ the growing labor force. development is cumulative, missed oppor- In countries where the demographic tunities to invest in and prepare this genera- transition is yet to get under way (a hand- tion will be extremely costly to reverse, both ful of countries in Sub-Saharan Africa), a for young people and for society. case can still be made for paying attention That it is hard for children to recover from to young people because of the changing early setbacks in human development is well health landscape and the need to engage recognized. New circumstances, however, young people effectively as workers and citi- mean that many developing country govern- zens. At the same time service delivery will ments now have to deal with the needs of those need to focus on basic education and health a little older--with next-generation issues of services for children and mothers, essential human capital development among youth--if for lowering infant mortality and fertility. they wish to consolidate and build on the gains In countries where the demographic transi- so far. Rising primary school completion rates tion is far advanced (as in some countries in have put enormous pressure on higher levels East Asia and some former transition econ- of education, even in the poorest countries. omies), there is a strong case for investing in Even as primary education becomes more the young before the rising fiscal burden of widespread throughout the developing world, a growing elderly population heightens the changesintechnologymeanthatyoungpeople trade-offs between the young and the old. needmorethanbasicskillstocompetesuccess- How well are countries preparing this fully in the job market. As a result of the epi- next wave of workers, entrepreneurs, par- demiological transition from communicable ents,and citizens? The answer varies.Despite to noncommunicable diseases, along with the enormous progress with primary schooling emergence of new diseases such as HIV/AIDS, in the poorest countries, many young peo- young people are exposed to a different range ple cannot read. In other developing coun- of health risks than before. Finally, changes tries many of the young, especially from the in the political landscape and the growth poorest households, stumble on the way of civil society have altered the meaning of from primary to secondary school. In still citizenship--and with it, what young people others, the main obstacle is continuing from need to learn to engage effectively in commu- secondary to higher education. Consider- nity and society. ing other dimensions important to reduc- Even as the development needs of the ing poverty and sustaining growth--such as young come into sharper focus, the demo- improving the quality of formal schooling to graphic transition under way in most devel- provide skills relevant to the changing needs oping countries is creating an enormous of the labor market, or the knowledge and 26 Youth, poverty reduction, and growth 27 ability to avoid risky health behaviors--it is deepest reductions in East Asia (with and clear that much remains to be done. without China) and South Asia (with and As developing countries focus on their without India). As a result, the developing young, at least five dimensions--or transi- world is likely to meet the Millennium Devel- tions--are relevant from a development opment Goal of reducing poverty. However, perspective. Already mentioned are learning, all regions other than East Asia and South going to work, and avoiding risks to health. Asia will fall short.1 As a fourth dimension, young people also What can be done to reduce poverty need to learn how to become good parents. more widely? Past World Development Finally, as democratization brings greater Reports have articulated strategies based on representation and civil liberties, they need the building blocks of economic growth, to learn how to engage as citizens. For coun- human capital, empowerment, and social tries that address the challenges of these five protection.2 The blocks are mutually depen- transitions--learning, going to work, staying dent, so they are like the wheels of an "all- healthy, forming families, and exercising citi- wheel drive" vehicle, complementing each zenship--the payoffs can be enormous. other to navigate difficult terrain. Young people and poverty reduction Young people are critical to "Early marriage spells doom for Where do young people fit into this? This further progress with poverty children because they end up being Report considers young people as, broadly reduction and growth like their parents." speaking, those in the age group 12­24 Poverty in the developing world has declined years (box 1.1). Youth is a period of intense Young person, Abuja, Nigeria significantly in the last 20 years, with the learning, when people can acquire the December 2005 B O X 1 . 1 What is youth? Youth is a transitional phase from childhood status.While much research is based in devel- National policies on youth typically estab- to adulthood when young people,through a oped countries,it suggests that young people lish an age range for beneficiaries.The lower process of intense physiological,psychological, in their late teens and early twenties often see bound ranges from around 12 years (Jordan) to social,and economic change,gradually come to themselves as not yet adult.Some argue that around 18 years (Bangladesh).In some cases it be recognized--and to recognize themselves-- this prolonged period of semi-autonomy can is not strictly defined,as in Hungary,where the as adults.So it is more a stage in life than an age. be viewed as a new life stage in which young youth secretariat deals with both 0- to 14-year- It can also be a period of great energy,enthusi- people experiment with adult roles but do not olds and 15- to 26-year-olds.The upper bound asm,and creativity giving rise to the expression fully commit to them. ranges from around 24 (Jamaica) to even 35 or that you are"as young as you feel,"which is Laws in most countries designate ages 40 (Kenya,Pakistan). especially popular among those who are well when people can be treated as adults and The UN's World Program of Action for Youth past their youth! are thus no longer offered the protections of defines"youth"as people ages 15­24,while the For research and policy it is useful to pin childhood.One can thus change from being a World Health Organization (WHO) and UNICEF down the period of youth more precisely.Per- child to being an adult overnight.But the age use the terms"adolescent"for those 10­19, spectives on the most relevant age range vary at which school attendance is no longer com- "youth"for those 15­24,and"young people" across disciplines.In the health field youth is pulsory,and employment is legally permitted for those 10­24.The wider band of 10­24 years associated with the ages of physical maturation typically ranges between 11 and 16 years.Legal used by these agencies recognizes that many that begins with menarche for girls and more responsibility for crime can begin earlier,but policies directed at youth often need to influ- gradually for boys,typically between the ages of individuals are generally not charged as adults ence outcomes before the age of 15. 10 and 16.In the social sciences youth is defined until around 16.Political participation through Recognizing the diversity in perspective, by the acquisition of various adult statuses, voting is postponed,typically to around 18 or this Report uses different ranges depending on marked by events such as menarche,leaving later.Likewise,service in the military,whether the context. However, the focus, by and large, school,employment,marriage,and voting,with compulsory or voluntary,is often restricted until is on the age range 12­24, when important the recognition that becoming an adult is a age 18.The purchase of cigarettes,in countries foundations are laid for learning and skills. lengthy,self-reinforcing process,often extend- where there are restrictions on sales to minors,is The Report uses the terms"youth"and"young ing into the twenties.Social psychologists argue allowed from around 15 to 18.Consumption of people"interchangeably. that the subjective experience of feeling adult alcohol,where it is legally prohibited for minors, matters at least as much as the objective mark- is allowed from around 18 to 21,though in some ers of adulthood,such as age or a particular cases it is prohibited until the age of 25. Source: Fussell (2006). 28 WORLD DEVELOPMENT REPORT 2007 human capital they need to move them- child health and birth weight. Throughout D E F I N I T I O N 1 . 1 selves and their families out of poverty the developing world, but especially in the Human capital (definition 1.1). Not confined to the skills low-income regions of South Asia and Sub- needed to become an economically produc- Saharan Africa, the percentage of children This Report uses the term tive adult, learning extends to other aspects immunized is higher when mothers have "human capital"to refer to a of life such as navigating health risks and some secondary education.5 broad range of knowledge,skills, and capabilities that people becoming a responsible spouse or parent Parental schooling is also important for need for life and work.Tradition- or citizen.3 It can happen in several ways, the cognitive development of children. Evi- ally,human capital refers to the often through the formal school system, but dence from Ecuador shows that the more education and health levels of also through learning from parents, peers, educated the parents, the greater the posi- people as they affect economic productivity.In addition to this family, community, and work experience. tive impacts on the cognitive development traditional notion,this Report Because the capacity for learning is so great of children as young as three years--impacts highlights skills and capabilities relative to older ages, missed opportuni- that become more pronounced as children required for successful living. These fall under three main ties--to acquire skills in school or on the grow older.6 These findings hold even when categories:jobs,family,and com- job, or good health habits, or the desire to controlling for the better health of children munity.Under jobs are a range engage in community and society--can be of more educated parents. of skills and capabilities required extremely difficult to reverse. Given the importance of building human to obtain and retain a job above and beyond the technical com- The high cost of remedying missed capital in youth, it comes as no surprise petence to do the job,such as opportunities is easiest to appreciate in that this stage of life is given some promi- self-discipline and teamwork. formal schooling. For many, youth marks nence in the international commitment to Under family are health and the transition from primary to secondary development as reflected in the Millennium such skills as good parenting and managing or resolving con- school. Dropout rates can be very sharp at Development Goals. Seven of the eight goals flict.Under community are the this stage, especially in countries that have relate to outcomes for young people either skills and capabilities involved made primary education universal (see directly or indirectly (box 1.2). in belonging to a community, figure 1 in the overview). Those who drop Many aspects of human capital develop- enjoying its privileges and protections,and living up to its out at this stage typically never return--in ment among the young are, however, not obligations. part because few countries have programs covered by the Millennium Development of remedial education to ease the transition Goals. Preparing young people to be active back into school, or provide formal equiva- citizens is notably absent. Within education, lency with school degrees. The competing the quality of education is not sufficiently "One of my neighbors is an demands of work and, for young women, emphasized, while the health goals do not intelligent girl, but she could not of family and children are further obstacles take into account health risks beyond HIV/ register at school because her mom to returning to school. Discouragement and AIDS and those arising from motherhood. did not have the 250 lempiras stigma can also play a role, and there is a [about $13] she needed for the question of whether cognitive development, Young people and equity matriculation." once arrested, can be resumed, especially if Poverty reduction is more than a reduction the gaps are long (chapter 3). Similar argu- in absolute deprivation. Relative deprivation Young person, Honduras ments apply to other missed opportunities, or inequity is also enormously important.7 A January 2006 such as learning from continuing employ- concern for equity leads to an emphasis on ment (chapter 4) or civic engagement early intervention to build human capacity "Education is anything but (chapter 7). because inequity--or"inequality of opportu- inclusive." Building human capacity early is impor- nity"intheparlanceof lastyear'sWorldDevel- Disabled young person, tant not just for the future opportunities opment Report--quickly becomes deeply Recife, Brazilopen to young people but also to mitigate entrenched. In developing countries where January 2006 the intergenerational transmission of pov- basic education has become widespread, erty. More educated youth are more will- many inequalities of opportunity--at least ing to control family size and invest in the as they relate to schooling--appear in youth health and well-being of their offspring. The as poor young people drop out of school, or impacts are particularly strong for women.4 receive poorer quality education than the rich In all developing regions the average num- (chapter 3). For many young women from ber of children born is significantly lower poor households, youth marks the entry into for women with at least secondary educa- early marriage or early childbearing, effec- tion.Maternal education strongly influences tively sealing off further opportunities for Youth, poverty reduction, and growth 29 B O X 1 . 2 "The Millennium Development Goals are not about youth, right?" No, wrong! Although the widespread impression is that well-recognized links between female school- young people have the opportunity to build the Millennium Development Goals are about ing,fertility,and child health,the education of and use human capital--whether through bet- providing basic services to children,seven of the young women and their ongoing reproductive ter schooling,better health,or more productive eight goals have outcomes that relate to young health needs are thus critical to meeting goal employment--will take the world a long way people. 4 on child mortality and goal 5 on improving toward meeting goal 1,eradicating extreme Education outcomes for the young are maternal health.All in all,making sure that poverty and hunger. explicitly targeted both as a part of achiev- Seven out of the eight Millennium Development Goals target youth outcomes ing universal primary education (goal 2),and promoting gender equality and empowering Millennium Development Goal Direct or indirect youth-specific target women (goal 3).The high risk of HIV/AIDS faced Goal 1. Eradicate extreme poverty and hunger Indirect by young people,especially in Sub-Saharan Africa,is targeted through three indicators Goal 2. Achieve universal primary education Target 8. Literacy rate of 15- to 24-year-olds relating to how well the young are informed of Goal 3. Promote gender equality and Target 9. Ratio of girls to boys in primary, secondary, and risks,to infection rates among young pregnant empower women tertiary education women,and to risks facing orphans--all criti- Target 10. Ratio of literate women to men, ages 15­24 cal to combating the spread of HIV/AIDS (goal Goal 4. Reduce child mortality Indirect 6).The role of young people as stakeholders in the future of international development is rec- Goal 5. Improve maternal health Indirect ognized through an emphasis on employment Goal 6. Combat HIV/AIDS, malaria, Target 18. HIV prevalence among pregnant women ages opportunities for them as a part of building a and other diseases 15­24 years global partnership for international develop- Target 19. Percentage of population ages 15­24 years with ment (goal 8). comprehensive and correct knowledge of HIV/AIDS Even where young people are not the Target 20. Ratio of school attendance of orphans to school attendance of nonorphans ages 10­14 years explicit focus,their involvement can be impor- tant to achieving goals.Young women con- Goal 8. Develop a global partnership Target 45. Unemployment rate of young people ages 15­24, tribute 20­30 percent of total fertility in high- for development by sex fertility countries and upward of 50 percent Source: Authors.For Millennium Development Goals and targets see http://unstats.un.org/unsd/mi/ in low-fertility regimes (chapter 6).Given the mi_goals.asp. schooling or employment (chapters 5 and higher skilled workers. The evidence extends 6). Solidarity with other human beings, and beyond the wage sector--educated farmers attitudes toward those who are ethnically or are more likely to adopt new technologies, religiously different, are also formed in youth and almost all studies on agricultural pro- and can be important to young people's ductivity show that better educated farm- sense of identity and how they relate to oth- ers get higher returns on their land.8 Many ers in society (chapter 7). Youth can thus be studies have documented the importance an important period for promoting equality of a large pool of educated workers--par- of opportunity and allowing individuals to ticularly if they are educated to the second- attain their full potential. ary level--for knowledge spillovers and for foreign direct investment.9 Lower computer Young people and growth penetration and productivity in Latin Amer- Youth is an important stage of life for building ica and the Caribbean compared with the the human capital that allows young people East Asian economies has been attributed to to escape poverty and lead better and more the lower proportion of the workforce with fulfilling lives. The human capital formed in secondary schooling.10 youth--whether in skill levels, or health, or Macroeconomic models also suggest "When we work, [...] there remains civic and societal engagement--is also an large potential impacts on growth and liv- no time or energy for depression, important determinant of long-term growth. ing standards.The strongest evidence relates juvenile delinquency, aggression, The most compelling evidence is on the to the impact of HIV/AIDS on growth and and the like." microeconomic side. Schooling is persis- productivity. HIV/AIDS disproportionately Young person, Nepal tently found to increase productivity, as affects young people ages 15­24 in high January 2006 reflected in earnings. As discussed below, prevalence countries, where they account average earnings for those with secondary for more than half of new infections. In andparticularlytertiaryeducationhaverisen South Africa, where prevalence is over 20 over time because of the growing demand for percent, the unchecked spread of the epi- 30 WORLD DEVELOPMENT REPORT 2007 B O X 1 . 3 Losing a decade--what HIV/AIDS is doing to human capital accumulation and growth in Kenya Young people in Kenya,as in most parts of the affected family's lifetime income shrinks,and per capita income is 15 percent lower than with- world,confront the real choice between con- with it the means to finance children's educa- out AIDS.Note that with higher population in tinuing their educations and starting to work. tion,and because children lose the parental the no-AIDS scenario,per capita income is actu- They also become sexually active,with all the knowledge and guidance that complement ally lower until 2010 than in the AIDS scenario. gratifications and risks that entails,including formal education. The reductions in growth are driven in large contracting such sexually transmitted diseases The model suggests that as a result of HIV/ part by the setback in the spread of secondary as HIV/AIDS.Somewhat later,as young adults, AIDS,national income in Kenya will be roughly education,which both lowers the productivity they choose partners,have children,and take on halved by 2040 (box figure).Per capita income of school leavers and weakens the transmis- the responsibilities of family life.To examine the growth,faltering since the 1980s in Kenya,takes sion of human capital to their children.By 2040, impact of these decisions on human capital and half a decade longer to recover to 1990 levels in HIV/AIDS delays human capital attainment on long-term growth,Bell,Bruhns,and Gersbach the AIDS scenario than without AIDS.By 2040, average by about a decade. (2006) set up an overlapping generations model in which the wellspring of growth is human HIV/AIDS can have a big impact on the growth of national and per capita income in Kenya capital accumulation. In the model,the formation of human Simulated trends in national income Simulated trends in per capita income capital involves the intergenerational transmis- 500 140 sion of knowledge and abilities through both 450 130 child rearing and formal education.The growth 400 National income 120 process can be derailed by the failure to expand 350 without AIDS Per capita income human capital as a result of poor quality school- 300 110 without AIDS ing or the limited expansion in schooling,or pre- 250 100 mature adult mortality from HIV/AIDS.In Kenya, Per capita 200 90 the victims of AIDS are overwhelmingly young income 150 National income with AIDS adults or those in the early prime years,most 80 100 with AIDS with children to raise and care for. 50 70 A parent's death does more than destroy 0 60 the victim's human capital.It also weakens the 1990 2000 2010 2020 2030 2040 1990 2000 2010 2020 2030 2040 mechanism to form human capital in the next generation and beyond.Why? Because the Source: Bell, Bruhns, and Gersbach (2006). demic is estimated to reduce the growth of growth and the quality of human capital GDP in the range 0.8 to 1.5 percentage points embodied in the labor force.14 a year.11 In Kenya, where the epidemic is hit- Young people are also an enormous ting a peak currently, one estimate suggests resource for growth in the short run. In it will take more than 40 years for per capita many developing countries, they form the income to recover to 1990 levels (box 1.3). largest group of job seekers. Their share in In Russia, with its smaller but more rapidly the unemployed is upward of 50 percent growing epidemic, GDP may decline by 10 in most countries in the Middle East and percent in 2020 if no attempt is made to limit North Africa, and upward of 40 percent in the spread of the disease.12 the Caribbean. Having young people sit idle More conventional approaches to esti- is costly in forgone output. Estimates indi- mating macroeconomic effects, through cate that lowering youth unemployment cross-country regression analysis, provide could raise GDP by anywhere from 0.3 to less clear-cut evidence.13 In part it is diffi- 2.7 percent in a range of Caribbean coun- cult to measure human capital consistently tries15 based on forgone earnings alone. across countries, let alone that which is Failing to direct young people into pro- formed in youth, other than through fairly ductive pursuits can prove costly in other crude measures such as completed years of ways.In many countries crime often peaks in schooling. Plus weaknesses of institutions this age group and can hurt the investment or demand mean than human capital does climate. In Jamaica, youth 17­29 are respon- not consistently add to growth everywhere. sible for more than half of all prosecuted However, one attempt to get beyond quan- crime except arson.16 A 1 percent reduction tity finds a robust relationship between in youth crime could increase tourist flows, Youth, poverty reduction, and growth 31 and raise tourism revenues by J$40 million of primary schooling. In countries where the Figure 1.2a Returns to higher (about $1 million), or 4 percent.17 gains to primary completion are recent, the education have risen over time main challenge is providing access to second- Latin America (1990=100) How the challenges confronting ary education, especially lower secondary. In 130 young people have changed other countries that have been successful in 120 Higher Not only is human capital development in providing access to secondary (especially 110 youth important to poverty reduction and lower secondary) education, the balanced Primary growth, changing circumstances mean that expansion of access to upper secondary and 100 many developing countries have to focus tertiary education is more of an issue. But 90 Secondary more on this stage than ever before. throughout the developing world, as school- ing becomes more widespread, the education 80 1990 1992 1994 1996 1998 Expansion in access to system also needs to recognize that students basic education are becoming more diverse and may require Source: Behrman, Birdsall, and Szekely (2003). Note: Estimates are based on log wage regres- more options for learning in keeping with sions for 30- to 55-year-old urban males that In many of the poorest countries, progress their different interests and abilities. control for various factors affecting earnings, with providing basic education to children based on 71 sample surveys from 18 countries with an average sample per year of over 7,000 means that governments are now confronted individuals. Rising demand for workers with the learning needs of those who are somewhat older--youth. Despite outstand- with higher education Figure 1.2b Returns to higher ing challenges, primary completion rates The global context is changing in other ways. education have also risen relative to have increased substantially, particularly in Unlike 20 years ago,when earnings of workers primary schooling South Asia, Sub-Saharan Africa, and Latin with secondary and tertiary education were Ghana (1987 primary=100) America and the Caribbean. Combined with low relative to those with primary education 400 the larger number of children of graduating in many developing countries, earnings have age, the pressure on education systems at lev- now risen substantially, particularly for work- 300 University els above primary is enormous (figure 1.1). ers with tertiary education but also, in some The only region in the world experiencing countries, for those with secondary educa- 200 little pressure on this account is Europe and tion.18 In Latin America and the Caribbean, Upper secondary 100 Central Asia,where falling fertility is reducing labor market returns to those who completed Lower secondary the number of primary school graduates. primaryorsecondaryeducationhavedeclined Primary 0 Building on the gains to primary school- sharply, while the returns to those with ter- 1987 1998 ing requires balancing opportunities for tiary education have increased (figure 1.2a). Vietnam (1998 primary=100) postprimary schooling with the expansion In Ghana there has been a marked increase in 800 the return to university education over time, University and as a result, those returns are significantly 600 Figure 1.1 The number of children completing primary school has risen in most regions higher than returns to primary than they were Technical/professional in the past. In Vietnam, returns to both ter- 400 Number of primary school graduates in 2003 (1990=100) High school 180 tiary and upper secondary (high school) have 200 risen relative to primary (figure 1.2b). 160 Secondary Primary Due to rise in primary completion rate The rising premium on higher educa- 0 140 tion reflects the growing demand for skills 1998 2002 120 driven in large part by the spread of new Source: Gian and others (2006); and Schultz (2003). 100 technologies.19 Twenty years ago the expan- Note: For Ghana, estimates are based on log sion of trade was expected to increase the wage regressions for 35- to 54-year-old males 80 demand for unskilled workers in developing with controls for post-schooling experience and other factors. For Vietnam, estimates are based 60 countries. However, the spread of new tech- on log wage regressions for workers ages 18­65 whose primary employment was in the wage 40 nologies to developing countries has sharply sector. Due to change in cohort size 20 increased the demand for skilled workers, substantially increasing returns even as the 0 East Europe Latin Middle South Sub- supply of skilled workers has increased. Asia & & America East & Asia Saharan Recent research suggests that the increase Pacific Central & the North Africa Asia Caribbean Africa in the number of skilled workers may have in fact boosted skill-biased technological Source: World Bank, UNESCO Institute for Statistics (UIS), and OECD (2006) for primary completion rates; World Bank (2006h). change and raised the demand for skills.20 32 WORLD DEVELOPMENT REPORT 2007 Figure 1.3 People today participate The growing demand for skills has in developing countries, and illicit drugs are more in political processes and have increased the value of further education and available more readily.22 greater civil rights and liberties made it more important for growth. Indeed, Percentage of countries in many countries building a workforce with The changing nature of politics 100 higher order skills is an important part of Not and the growth of civil society free improvingtheclimateforinvestment,acquir- The number of countries in which people 75 ing a competitive edge, and generally main- cannot participate freely in the political pro- Partly taining the engine of growth. Because much cess or where there is limited freedom of 50 free of the return to higher education--especially expression or belief has declined significantly tertiary education--is private, rising returns from 4 in 10 countries in 1975 to 1 in 4 in 25 are not an argument for public funding, or at 2005 (figure 1.3). And the number of elec- Free least for public funding that goes beyond the toral democracies has vastly increased. Many 0 need for equity of access. It does draw atten- 1975 2005 of the gains have come since the fall of the tion, however, to important changes in how Berlin Wall and the end of communism in Source: Freedom House (2006). the global market is rewarding skills. Note: The freedom ratings are a composite of Eastern Europe and the former Soviet Union. two measures that relate to political rights and civil liberties. Each is ranked on a scale from 1 With greater democratization in the develop- New health risks (least free) to 7 (most free). See http://www ing world there has been a tendency toward .freedomhouse.org for details. The health environment has also changed. decentralizing public decision making to As infant and child health has gradually lower levels of government23 and toward improved, new diseases have emerged. As a more civic participation through civil society result, sexual initiation and sexual experi- organizations, community interest groups, mentation in youth carries far greater and other nongovernmental organizations.24 risks than before, especially in some Sub- This has increased the opportunities for Saharan African countries with very high people to participate in political activity and HIV-prevalence rates. Outside Sub-Saharan exercise their voice through a wide range of Africa, ongoing progress with tackling forums. Preparing young people for their communicable diseases, both new and old, rights and responsibilities as citizens--build- means that noncommunicable diseases and ing their social capital for this kind of engage- injuries have risen in prominence for young ment--has thus become more important. people. The greater coverage of roads and vehicular traffic is contributing to road Globalization and new technologies fatalities. In Vietnam, road traffic accidents Young people are growing up in a more are the most important cause of death global world. Information flows have among young men.21 Tobacco is now mar- increased substantially because of the keted far more aggressively to young people greater reach of global media, movies, music, and other cultural exports, though access varies significantly. Freer trade has expanded the goods and services that peo- Figure 1.4 While access to the Internet varies greatly, young people dominate usage ple are exposed to. And greater mobility-- Egypt, Arab Kyrgyz seen in rapid urbanization and the flows Ethiopia Indonesia Ghana Rep. of Armenia Republic China of people across borders--has increased 2005 2005 2005 2005 2005 2005 2005 awareness of consumption possibilities. Share of all New technologies, such as the Internet and 15- to 24-year- 29% 30% 53% mobile telephony, have a strong following olds with 0.05% 12% 13% 15% access to among youth. New data collected for this the Internet Report show that young people are among the primary users of the Internet, account- Share of all Internet ing for 40 percent or more of Internet users 67% 70% 52% 60% 50% 61% 43% users ages in a range of developing countries. Access, 15­24 however, varies from less than 1 percent of youth in Ethiopia to more than 50 percent Source: WDR 2007 InterMedia surveys. Note: "Access to the Internet" is defined as having used the Internet in the four weeks before the survey. in China (figure 1.4). Youth, poverty reduction, and growth 33 Theimpactof newinformationandcom- Largest numbers ever munications technologies is likely to vary The current cohort of young people in both across and within countries because developing countries is the largest the of differences in access. At one end of the world has ever seen--around 1.3 billion.26 spectrum, young people in many middle- On current projections, the number of 12- income countries, especially if urban or to 24-year-olds will reach a maximum of middle class, have easy access to informa- 1.5 billion in 2035 and decline only gradu- tion through radio or television or the ally thereafter. Numbers are at a plateau Internet. At the other end, large numbers because birth rates are falling. Combined of young people, especially in low-income with a slowdown in the increase in the countries, continue to have very limited number of women of childbearing age, access not only to new sources of informa- cohorts are at or near a maximum in many tion, but even to traditional sources such as countries and in the developing world as radio and television. a whole (see spotlight on differing demo- The effects of exposure to more infor- graphics following this chapter). mation can be both positive and nega- The plateau hides distinct regional tive. The Internet is an important source trends (figure 1.5). At one end, the num- of information for many young people on bers of youth in East Asia (dominated by matters related to sexual and reproductive China) have peaked and are set to decline. health, especially where little is provided The same is true of Europe and Central in school or through the family. Both old Asia. At the other end, the youth popula- and new media, however, can expose the tion in Sub-Saharan Africa, already more young to unfiltered, beguiling, or confus- than four times its 1950 level, is projected ing images of sex and violence. Both types to continue growing rapidly into the fore- of media can also promote exaggerated seeable future.27 South Asia, Latin America images of Western consumption and lead and the Caribbean, and the Middle East and to frustration if opportunities are not com- North Africa are between the two extremes. mensurate with the expectations young Numbers in Latin America and the Carib- people have formed. bean have reached a peak, or a long plateau; The impact of greater exposure to media those in South Asia and the Middle East and also depends on the local culture and its North Africa are expected to grow slowly response to imports. In many parts of the until they peak in the next 25 years or so. developing world, the resurgence of reli- gious movements has acted as a counter- vailing influence to the more permissive attitudes sometimes purveyed by the media. Figure 1.5 Trends in the developing world's population of young people vary significantly These movements have been associated across regions with changes in the behavior and attitudes Population of youth, millions of young people toward sex and marriage 500 (promoting virginity before marriage) and South Asia health (leading to a rejection of the use of alcohol and drugs).25 400 East Asia & Pacific Do numbers matter? 300 How demographic changes Sub-Saharan Africa affect opportunities for youth 200 Even as changing circumstances increase Latin America & the Caribbean the need to focus on human capital devel- 100 opment for youth, many developing coun- Europe & Central Asia Middle East & North Africa tries are helped by favorable demographic 0 changes that are creating a unique opportu- 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 nity to invest in youth. Source: United Nations (2005b), medium variant. 34 WORLD DEVELOPMENT REPORT 2007 Falling relative shares-- a small group of countries (Sierra Leone is with a few exceptions one of them), relative sizes are still increas- ing. Others in this group are Afghanistan, Absolute numbers are interesting--but Chad, Democratic Republic of Congo, Equa- only to a point. It is not clear that this is a torial Guinea, Ethiopia, Guinea-Bissau, Libe- critical variable unless inputs to economic ria, Niger, Republic of Congo, and Somalia. and social well-being (factors of produc- Except for Afghanistan, the countries in this tion) are in fixed supply. Some of these fac- group are in Sub-Saharan Africa.30 tors, such as land, may indeed be relatively The fact that the relative size of youth fixed in supply, though the supply of land cohorts is declining in most developing has been relatively unimportant in explain- countries means that the negative effect of ing economic growth, wages, or poverty being part of a large cohort, strong though it during the last 50 years of rapid population growth.28 Other inputs, such as physical may be, is declining. However, for the small group of countries where relative sizes are capital (machinery and infrastructure) can set to grow, the rising relative size of cohorts be expanded. could well aggravate difficult employment Evidence from developed countries conditions for young people--if the right suggests that the size of youth cohorts rela- macroeconomic and labor market policies tive to older workers is more important for are lacking.31 long-term outcomes than absolute num- bers. The postwar baby boom in Organi- sation for Economic Co-operation and A window of opportunity Development (OECD) countries resulted from falling dependency in a relatively large number of young peo- The declining relative size of youth cohorts ple entering the labor market in the 1970s signals the decline in the ratio of depen- (see the spotlight on baby booms follow- dents (children and elderly) to working ing chapter 4). This cohort experienced population that occurs during the demo- lower wages and higher unemployment graphic transition. (The convention is to than preceding generations. define children as those below 15 and the Studies that examine whether there were elderly as those above 65.) The trajectory long-term effects--or scarring--from these of the relative size of youth cohorts mirrors negative early experiences are varied. Some the dependency ratio,diverging significantly conclude that the impact of being in a rela- only when the demographic transition is so tively large youth cohort tends to diminish far advanced that the rising burden of the over time, in some cases disappearing. Oth- elderly on the working population pushes ers find more persistent effects, especially up the dependency ratio (figure 1.6). among those with lower education (chap- This decline in dependency (the rise ter 4). Caution is needed in deriving con- in the ratio of working population to the clusions for developing countries, but note nonworking population) during the demo- that the baby boom generation is broadly graphic transition can boost economic comparable in relative size to the large growth.32 The strongest evidence comes youth cohorts entering the labor market in from East Asia where between 25 and 40 developing countries today.29 percent of the rapid growth between 1965 The difference is that in most developing and 1990 in Japan, Hong Kong (China), countries, the relative size of youth cohorts the Republic of Korea, and Singapore has is shrinking. Figure 1.6 shows countries in been attributed to the higher growth of the different stages of the demographic transi- working age population.33 tion--China is far along, while Sierra Leone The potential for enhanced growth is lagging. Other than in Sierra Leone, the through a demographic dividend arises highest peak of the youth population relative for two reasons. First, the rise in labor sup- to the population of older workers occurred ply per capita, reinforced by the increase in the 1970s or 1980s. The declines in this in female labor supply that often accom- ratio since then have been considerable, panies fertility decline, increases potential ranging from 25 percent to 50 percent. For output per capita. Of course, this greater Youth, poverty reduction, and growth 35 Figure 1.6 The relative size of youth cohorts is declining in most of the developing world, as is the dependency ratio China Brazil Percent Percent 100 100 80 80 60 60 40 40 20 20 0 0 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Vietnam Egypt, Arab Rep. of Percent Percent 100 100 80 80 60 60 40 40 20 20 0 0 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 India Sierra Leone Percent Percent 100 100 80 80 60 60 40 40 20 20 0 0 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Relative size of population ages 12­24 to population 25­60 Dependency ratio of population ages 0­14 and 65+ to population 15­64 Source: United Nations (2005b), medium variant. Note: Dotted line represents 2005. labor supply would need to be produc- working age population, because the micro tively employed. Second, higher savings evidence is not consistent with a life-cycle and investment per capita associated with hypothesis of saving.34 Other factors--such the rising share of the working age popula- as rising life expectancy--may have con- tion (more likely to be in the saving phase tributed to rising savings.35 of the life cycle than the rest of the popu- This literature on the pathways for the lation) could also boost growth. Countries demographic dividend is not conclusive, in East Asia were particularly successful in but the idea of a demographic dividend absorbing their growing cohorts of new from changes in the dependency ratio has labor market entrants. Savings also grew, practical significance for public expendi- though it remains unclear whether this tures. Funded by taxes on income or con- can be attributed to the rising share of the sumption, those expenditures are likely to 36 WORLD DEVELOPMENT REPORT 2007 be affected by the ratio of the tax-paying population. In China, which underwent population to the beneficiary population. an early and fairly rapid fertility transi- Dependents are typically net beneficiaries tion, dependency ratios will start increas- rather than net taxpayers, depending on ing as early as 2010 because of the rapid governments to finance primary and sec- aging of the population. Other countries ondary schooling, postsecondary education expected to see an increase in dependency (in many cases), training programs, health in the next decade include Armenia, Geor- programs, and pensions. Early in the demo- gia, and Thailand. The pressure to increase graphic transition, when the population of spending on the elderly will likely generate elderly is small, the main net beneficiaries pressures to raise taxes, cut expenditures are children and youth. With dependents (including those on children and youth), "We young women are not prepared and youth cohorts falling in size relative to or both. So there is no time like the present to become mothers. I would like the tax-paying population in many devel- for investing in the young. to continue my studies, but since I oping countries, the economic circum- have had my daughter my options stances for investing public resources in have changed, because I have many How prepared are youth children and youth are likely to improve. more obligations now. I hope that for today's challenges? In a country with a 25 percent decline in this will not be a barrier for me to the dependency ratio, every real or baht or A glass half empty succeed in life." lira or rupee collected in taxes from each How well are countries building the human Eylin, 19, Honduras working age person could pay for a 33 per- capital of their young? The overall pic- January 2006 cent increase in spending per dependent, ture is one of steady progress in averages compared with the year of the peak ratio. but uneven improvement across different This increased investment can help build dimensions of human capital and differ- the human capital of children and youth, ent groups. Thus, much more remains to be which can have subsequent positive effects done. on growth. Education levels have been rising world- Not every developing country has wide, and today's youth cohorts have on declining dependency ratios. In the average more years of schooling than their countries such as Sierra Leone, men- predecessors. On conventional measures of tioned earlier, the dependency ratio is health, such as mortality in childhood or still rising because the demographic height-for-age, they are also the healthiest transition is not yet under way.36 There, generation by far. Outside countries with a rising dependency ratios are steadily very high prevalence of HIV/AIDS and some reducing the resources available per child parts of the former Soviet Union afflicted and young person, possibly leading to a with premature male mortality, their chances vicious cycle of underinvestment and low of surviving to old age are higher than ever growth. Without serious efforts to lower before.37 These are achievements to be proud child mortality and fertility, this situa- of, but the averages hide enormous differ- tion is not going to change. For the other ences across and within countries. developing countries, circumstances have At one end of the spectrum are young never been better. Just as a changing world people in almost every developing country is increasing the need for policy to focus who have had the benefit of good second- on the young, demographic changes are ary education. They can be expected to go making it easier to do so. This surely is on to university. Some may even pursue good news. an advanced degree in a developed coun- The window of opportunity presented try. In a country such as Niger, many more by declining dependency ratios will also young people go on to study in France than close, however, earlier in some countries in tertiary education institutions at home than in others. Many developing coun- (chapter 8). While well off youth in devel- tries are set to become more like devel- oping countries may have experimented oped countries, which already face the with alcohol and drugs or engaged in sex consequences of rising dependency caused while at school, their good prospects, their by the rising share of the elderly in the means to deal with poor outcomes, and Youth, poverty reduction, and growth 37 Figure 1.7 Many young people from poor households do not have the most basic skill of literacy Literacy rate of population ages 15­24 (percent) 100 80 60 40 20 Top quintile Bottom quintile Average 0 of BeninFaso Peru Rica Chile Ethiopia Nigeria Bhutan Jordan TurkeyBrazil Mexico Nicaragua ParaguaySwaziland Colombia Pakistan States BurkinaBangladesh MauritaniaCameroonCambodia Honduras Vietnam Indonesia Armenia MoldovaGuatemala Costa Fed. Micronesia, Source: Authors' calculations based on Fares, Montenegro, and Orazem (2006a). Note: Countries are ordered from lowest to highest GDP per capita (PPP). possibly their greater awareness mean that as children. The International Labor Orga- this experimentation is less likely to have nization (ILO) estimates there were 84 mil- adverse consequences. Because they have lion 12- to 14-year-olds working in 2000.38 access to better quality education, many will As child laborers grow older they face very have developed "softer" life skills, such as limited options for employment and earn- working in teams or handling difficult situ- ings. For many poor young women, their ations with confidence. Finally, being the limited (or nonexistent) education is asso- children of more educated and wealthier ciated with early marriage, which is still parents, many may be developing a sense of high in a number of countries (chapter their place in community and society. 6). Many studies from Sub-Saharan Africa At the other end of the spectrum are a find that early marriage increases the risk vast number of young people for whom the of HIV/AIDS among women because of opportunities look very different. Many the higher frequency of sexual contact either did not attend school as children within marriage. Young women are more or dropped out too early to acquire even likely to bear children at an early age if they the most basic skills, leaving more than are from poor households (box 1.4). This 130 million 15- to 24-year-olds illiterate. combination of poverty and low education Most of them are in South Asia and Sub- means that they are not well-placed to take Saharan Africa, more of them women than care of themselves or their offspring, per- men (figure 1.7). For those who progress petuating poverty. through school, average grade completed Between the two ends of the spectrum shows a similar pattern, with rich young- are the vast swath of today's young people. sters completing more years of school than Many of them are still in primary school, poor youngsters (box 1.4). having started late and frequently repeat- For poor young people with little or ing grades. Late starts are more common no education, life's opportunities are in postconflict environments: in Cambo- restricted. Young men with little or no dia, 15 percent of 15- to 19-year-olds were education face enormous disadvantages in enrolled in primary school in 2001. Others the labor market. Many may have worked will have made it through primary school 38 WORLD DEVELOPMENT REPORT 2007 B O X 1 . 4 A youth perspective on equity and development World Development Report 2006 made a case for But even among those who go to school,oppor- likely to find paid employment outside the focusing on inequalities in key dimensions of tunities are vastly different (figure 1). home than those from rich households, in opportunity--such as education,health,and As might be expected,per capita income some cases by a substantial margin.The con- the capacity to participate in society--because and schooling attainment among 15- to 24- trast is particularly striking in countries such as these inequalities tend to perpetuate them- year-olds are broadly correlated.In many low- Bolivia, Panama, and Paraguay. selves,both across groups in societies and income countries,though,young people from For young women,puberty and adolescence over time.This can result in inequality traps, the poorest 20 percent of households cannot often mark the divergence in opportunity with which some groups or people are unable to aspire to complete the primary cycle (six years). their male counterparts (see spotlight on gen- escape.This is detrimental on both intrinsic On the other hand,those from the richest 20 der following chapter 2),differences heightened grounds,because people cannot realize their percent of households can expect to do so by poverty.Poor young women are not only full potential,and on instrumental grounds, almost everywhere.Among those from poor less likely than their richer sisters to complete because inequality traps can curtail growth and households,girls are far less likely to complete primary schooling--they are also far more likely dynamism. primary school than boys.These differences,in to bear children before the age of 15 (figure 3), Many inequalities become entrenched in addition to perpetuating poverty,are an enor- which can be detrimental to their health and youth because of the vastly different opportuni- mous cost to society--with the poor unable to well-being and that of their children.The life tra- ties confronting youngsters from different eco- make the most of their talents. jectories of 15- to 24-year-old women from poor nomic backgrounds. Young people's access to productive jobs, households can thus be very circumscribed. We have already discussed how differences which allow them to build skills and experi- Addressing these inequalities is an impor- in literacy--the most basic of skills--affect the ence, also varies enormously between those tant part of addressing inequality of opportu- future income-earning potential of youngsters from rich and poor households (figure 2). nity by allowing young people to participate and the economic prospects of their families. Youngsters from poor households are far less more fully in their societies. Opportunities for young people at school and work vary enormously between rich and poor Figure 1 Completed years of schooling, 15- to 24-year-olds Years 14 Top quintile Bottom quintile Average 12 10 8 6 4 2 0 of Faso Haiti Peru Brazil Rica Chile TanzaniaEthiopiaNigeriaBenin d'Ivoire Lanka Pakistan Mexico Paraguay Swaziland Colombia States Thailand Bangladesh Mauritania Cameroon CambodiaVietnamBolivia Bhutan JordanAlbania Honduras Indonesia Nicaragua Guatemala Sri FederationCosta BurkinaCôte Fed. Russian Figure 2 Incidence of paid work, 15- to 24-year-olds Micronesia, Percent 100 Top quintile Bottom quintile Average 80 60 40 20 0 Haiti Peru Brazil Rica Chile TanzaniaEthiopiaZambiaNigeria Bolivia TurkeyTonga Pakistan Mexico Cambodia Indonesia Thailand Venezuela SwazilandPanama Colombia Argentina Bangladesh NicaraguaEcuadorJordan ParaguaySalvador El FederationCosta Mozambique de B. R. Russian Youth, poverty reduction, and growth 39 B O X 1 . 4 continued Opportunities for young women from poor households can be restricted by early childbearing Figure 3 Incidence of early childbearing (before age 15), 15- to 24-year-olds Percent 30 Top quintile Bottom quintile Average 20 10 0 MaliBenin Faso Chad Haiti India MalawiEthiopia Zambia Kenya RwandaUgandaRepublic Guinea Ghana Bolivia Gabon Cameroon Zimbabwe NicaraguaArmeniaMorocco NamibiaColombia Guatemala Philippines Mozambique Burkina Kazakhstan Kyrgyz Source: Figures 1 and 2: Fares, Montenegro, and Orazem (2006a). Figure 3: authors' calculations based on recent Demographic and Health Surveys and various household income and expenditure surveys. Note: For educational attainment and paid work, quintiles are based on per capita consumption estimates; for early childbearing, on a household durables possession index. Countries are ordered in terms of GDP per capita (PPP) on time, but they may have faltered in the in Armenia (44 percent), Indonesia (58 transition to secondary school--because percent), Mexico (29 percent), and Nepal of poor access to schools, a curriculum (55 percent).39 Many developing countries that fails to engage and instruct, and high provide civic instruction at school, but it is costs (including opportunity costs), espe- only as effective as the teaching methods, cially for the poor. Still others will be pro- which can leave much to be desired (chap- gressing through secondary school, for the ter 7). Opportunities to participate in the most part, acquiring fewer skills for work community--whether in school councils, and life than young people in developed community organizations, or local govern- countries. Those who are from poor fami- ments--vary substantially both across and lies will learn even less than those from within countries (chapters 3 and 7). richer backgrounds (chapter 3). Finally, many young people cannot find The challenges young people confront work. Unemployment rates for young peo- go beyond acquiring skills relevant to the ple, which are higher than those for adults labor market and extend to skills to navi- all over the world, are five to seven times the gate health risks and engage constructively adult unemployment rates in some develop- in community and society. Many young ing countries (chapter 4). Unemployment is people remain far from well-informed not the only problem, as many young peo- about the consequences of excessive con- ple are stuck in low productivity jobs or are sumption of alcohol or drugs or engag- neither in work or school. ing in unprotected sex. Knowledge of how All these problems set back the human to prevent HIV/AIDS, which is nowhere capital development of young people. near universal, is in many cases confined to a very small proportion of young peo- What should policy makers ple (chapter 5). Regular consumption of tobacco is very common in many develop- focus on? The five transitions ing countries: upward of a quarter of 15- to Much of what governments need to focus 24-year-old men currently smoke regularly on to make young people fulfill their 40 WORLD DEVELOPMENT REPORT 2007 "Regarding the young gang potential and contribute to their own well- the transition to adulthood requires the members, it is not a matter that being and that of society has already been development of human capital, the capa- they cannot study. The problem is touched on. Young people need to con- bility to make competent choices, and the that . . . the teachers are very strict, tinue learning to build skills and acquire development of a sense of well-being.40 authoritarian. During lessons, only human capital. Skill-building needs to When exactly these transitions occur the teachers talk, they do not discuss cover not only the skills for work, because varies by age, and one can think of three [things] with the students. Young young people need to learn to manage a distinct phases when different transitions people can study, but they are not range of health risks. They also need to be might be more evident. In the early phase, motivated." adequately prepared to become parents to roughly ages 12­14, the focus is more likely reduce the intergenerational transmission to be on learning. In the middle phase, Elvis, 21, Peru of poverty that occurs because of the fail- ages 15­18 or 15­20, learning contin- January 2006 ure to plan and space births, and nurture ues but work begins to come into greater children appropriately. They need also to prominence. Many behaviors that endanger learn to become actively engaged as citi- health increase with age and so can be more zens in the communities and societies in important at this stage. At the upper end of which they live. this range, young people may begin to for- The five dimensions--learning, going mally exercise citizenship through voting, to work, staying healthy, forming families, and for many young women this may also and exercising citizenship--are referred be a phase of childbearing. In the late phase, to in the social science literature as "tran- ages 18­24, work and childbearing assume sitions." The term "transition" is a little greater importance. misleading because some of the dimen- How these transitions play out naturally sions, such as going to work or forming a varies by country and individual. In Haiti, family, have more of an element of transi- a poor and largely rural economy, the early tion than others. All, however, are critical phase of ages 12­14 is dominated by pri- for poverty reduction and growth because mary school (figure 1.8). By age 14 more they relate to building, maintaining, than a quarter of 12- to 24-year-olds con- using, and reproducing human capital. tinue to be enrolled in primary school-- Recent research also highlights these five which formally ends at 12 years--owing transitions, emphasizing that success in to repetition and delayed or interrupted primary schooling. A smaller proportion of 12- to 24-year-olds are enrolled in sec- ondary school. As young people progress B O X 1 . 5 Work and marriage for men in the Middle through the teens, enrollment in second- East and North Africa ary school declines, falling quite sharply in the late teens. Girls have uniformly Countries in the Middle East and North of incumbents,making it difficult for new lower enrollment than boys. Labor market Africa (MENA) have increased school- entrants--typically youth. attachment rises with age but picks up par- ing among both young men and young Difficulties in transitioning to work have ticularly after the ages of 15­16, more for women.In 1960,women over age 15 had devalued education credentials and fostered less than one year of schooling on average, cynicism among successive cohorts of youth. boys than for girls. but in 2000 more than 4.5 years.Men's aver- Another consequence has been a delay in Experimentation with tobacco in Haiti age years of schooling rose from around 1.5 marriage,particularly for young men.The also rises with age but is low overall, a com- years to more than 6.As a result,the MENA MENA region stands out for having one of the mon finding in many low-income environ- region now has schooling levels around the highest declines in the proportion of ever- developing country average. married men in the 20­24 year age group ments where access to tobacco products is MENA's unemployment rates,how- since the 1970s.Qualitative studies point to constrained by income. Engaging in sex is ever,are among the highest in the world, a sense of frustration among young men at more common, with 30 percent of females leaving over one in four young men and their inability to start forming families.How- and 50 percent of males reporting having women in search of employment.In part, ever,more research is needed to establish high unemployment reflects growth rates the consequences of these trends on family had sex by the age of 16. Reported differ- lower than the developing country aver- and society. ences between the sexes can be large, with age.The failure to find employment is also women understood to under report and men the result of schooling systems that do not Source: National Research Council and Institute to over report sexual activity. Marriage is not impart market-relevant learning and skills. of Medicine (2005); Mensch,Singh,and Caster- In addition,labor markets protect the rights line (2006). common for young men before age 24, but it rises quite sharply for women after age 18. Youth, poverty reduction, and growth 41 Figure 1.8 The transitions for Haitians ages 12­24 Going to school Going to work Cumulative percentage in school Cumulative percentage employed 50 30 Boys in primary 40 Boys in secondary Males Girls in primary 20 30 Girls in secondary 20 10 Females 10 0 0 12 13 14 15 16 17 18 19 20 21 22 23 24 12 13 14 15 16 17 18 19 20 21 22 23 24 Age Age Smoking Engaging in sex Cumulative percentage who have smoked Cumulative percentage who have had sex 100 80 Males 80 60 60 Females 40 40 20 20 Males Females 0 0 12 13 14 15 16 17 18 19 20 21 22 23 24 12 13 14 15 16 17 18 19 20 21 22 23 24 Age Age Forming a family Exercising citizenship Cumulative percentage married or pregnant Cumulative percentage in civic or social associations 30 40 Females 30 20 Females Males 20 10 Males 10 Girls ever pregnant 0 0 12 13 14 15 16 17 18 19 20 21 22 23 24 12 13 14 15 16 17 18 19 20 21 22 23 24 Age Age Source: Authors' calculations based on Enquête sur les Conditions de Vie en Haiti (2001). By 24, nearly a quarter of young women are proportion in secondary schools is higher. married and nearly 10 percent have borne Very few young people remain in primary children. Participation in civic or social school after age 15 and in secondary after associations outside the home rises almost age 18. Labor market engagement rises with uniformly from age 12 onward, slowing as age but sharply increases after age 18, as young people approach their mid-twenties. opposed to ages 15­16 in Haiti. Engagement In Chile, an upper-middle-income in civic associations rises over this age range country, the five transitions are both similar and looks very similar to that in Haiti. to and different from those in Haiti (figure The big difference from Haiti is in health 1.9).A good proportion of young people are behavior and family formation. Experimen- still in primary school until age 14, but the tation with alcohol, tobacco, and drugs is 42 WORLD DEVELOPMENT REPORT 2007 Figure 1.9 The transitions for Chileans ages 12­24 Going to school Going to work Cumulative percentage in school Cumulative percentage employed 50 30 Boys in primary 40 Boys in secondary Males Girls in primary 20 30 Girls in secondary 20 10 Females 10 0 0 12 13 14 15 16 17 18 19 20 21 22 23 24 12 13 14 15 16 17 18 19 20 21 22 23 24 Age Age Smoking Engaging in sex Cumulative percentage who have smoked Cumulative percentage who have had sex 100 80 Males Males 80 60 Females Females 60 40 40 20 20 0 0 15 16 17 18 19 20 21 22 23 24 12 13 14 15 16 17 18 19 20 21 22 23 24 Age Age Forming a family Exercising citizenship Cumulative percentage married or pregnant Cumulative percentage in civic or social associations 30 40 Males 30 20 20 Females Females 10 10 Males 0 0 12 13 14 15 16 17 18 19 20 21 22 23 24 12 13 14 15 16 17 18 19 20 21 22 23 24 Age Age Source: Authors' calculations based on Encuesta de Caracterización Socioeconómica Nacional (2003) and Encuesta Nacional de Juventud (2000). much higher among Chilean youth, with The five transitions, though discussed more than two-thirds of young people hav- separately, interact. Some of the inter- ing tried one of these substances by age 16. relationships are obvious, such as longer Many young people also report that they schooling delaying labor force entry, or engage in sex at a fairly young age--more childbearing reducing labor market attach- than 20 percent of females and 40 percent ment (at least temporarily). Some are less of males report having had sex by age 16. obvious, such as the relationship between These figures are comparable to those work and marriage for men (box 1.5). For reported by youngsters in Haiti. However, governments that create a policy environ- Chile's incidence of marriage among 12- to ment conducive to all the transitions, the 24-year-olds is lower than Haiti's. payoffs can be enormous. spotlight Differing demographics Differing demographics 43 Most young people today are born into smaller families than their parents, but there are many more of them than in their parents' generation. The developing world as a whole and many countries are seeing a peak in the numbers of youth because population momentum--the inertia in population growth related to the large size of the childbearing population--is being gradually overtaken by falling fertility. Now at 1.3 billion, the population of young people is expected to grow slowly into the foreseeable future because the continuing growth in youth populations in Sub-Saharan Africa, Middle East and North Africa, and South Asia will counter the slow declines in East Asia and Europe and Central Asia. Why youth populations are and 1960s, generating population growth reflecting the rapid decline in fertility rates approaching a peak rates in excess of 4 percent a year in some that began in the 1960s. Cohort size then Why are youth populations so large? Recall countries. The timing of fertility decline has grew rapidly again in the late 1970s, reach- the main elements of the demographic varied, but it occurred in many developing ing a peak in 1982, driven by what demog- transition. Before the demographic tran- countries in the 1960s, when world popula- raphers call population momentum--the sition, death rates and birth rates are high tion growth hit a peak of around 2 percent increase in the size of the childbearing and in balance, implying low rates of popu- a year. The rapid population growth of the population as the birth cohorts of the 1950s lation growth. The demographic transition 1960s--the "population explosion"--is reached childbearing age. begins with a decline in death rates. With the origin of today's large youth cohorts. The experience for other developing death rates falling, birth rates typically Today's youth are the children of the popu- countries is similar to Brazil's, with differ- remain high for some period, accelerating lation explosion generation. ences only in the timing of the largest sur- population growth. Eventually birth rates Consider Brazil (figure 1).1 The demo- viving birth cohort. With the large declines also fall, slowing population growth. The graphic transition was already well under in fertility starting around the 1960s, transition ends when birth rates and death way by 1950, with the death rate having the developing world as a whole is now rates have both stabilized at a new low level, fallen to 15 per 1,000, while the birth rate approaching a plateau in youth numbers. implying a return to low (or zero) popula- was almost 45 per 1,000. Population growth There are now 1.3 billion young people ages tion growth. was about 2.8 percent a year, higher than 12­24 in the developing world, a number High-income countries went through a ever experienced by high-income countries expected to grow to 1.5 billion in 2035 and demographic transition in the 1800s and when they went through the demographic begin declining thereafter. early 1900s and had a long and slow decline transition. Although the birth rate was fall- Country patterns are distinct in mortality. The gap between birth rates ing in the 1950s, death rates were falling and death rates was never very large, and faster, causing a peak population growth Depending on the timing and speed of population growth rates rarely exceeded 1 rate of 3.0 percent in 1960­65. This was fertility decline, countries can expect to percent a year. also when world population growth rates see different trajectories of number of The demographic transition in develop- reached their historic peak. youth (figure 3). Four distinct patterns ing countries is quantitatively very different. Birth cohorts grew rapidly in the 1950s are the result of a complex interaction Death rates declined very fast in the 1950s in response to the rapid decline in death between fertility, mortality, and population rates (figure 2), driven largely by declines in momentum. infant and child mortality. Cohort size lev- Group 1. Countries in this group typi- eled off in the late 1960s and early 1970s, cally experienced an early transition to low Figure 1 The demographic transition in Brazil led to a peak in population growth during 1960­65 Figure 2 The largest surviving birth cohort in Brazil was born in 1982 Rate per 1,000 population Size of birth cohort reported in census (millions) 50 4 Crude birth rate 40 3 30 2 1960 census 20 Rate of natural increase 1970 census 1980 census 1 10 1991 census Crude death rate 2000 census 0 0 1955 1965 1975 1985 1995 2005 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 Source: United Nations (2005b), medium variant. Source: Lam (2006) 44 WORLD DEVELOPMENT REPORT 2007 Figure 3 Country trends in youth population vary significantly the fertility transition is halted, proceed- ing slowly, or yet to get under way. Other Population ages 12­24 (1950=100) countries in this group include Afghani- 800 stan, Cambodia, Chad, Republic of Congo, Democratic Republic of Congo, Equato- Sierra Leone rial Guinea, Eritrea, Ethiopia, Guatemala, Pakistan Guinea-Bissau, Kenya, the Lao People's 600 Democratic Republic, Liberia, Mozam- bique, Niger, Nigeria, Pakistan, Rwanda, Senegal, Somalia, Uganda, and the Repub- lic of Yemen. Egypt, Arab. Rep. of Within this group there are marked dif- 400 ferences in countries. Some with very late India fertility declines, such as the Democratic Republic of Congo and Sierra Leone, are Brazil Vietnam projected to have continuing rapid growth 200 of the youth population for the next several China decades. Dependency ratios have yet to fall, so resources available per youth are falling, Russian Federation and youth cohorts are growing relative to 0 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 older workers, intensifying the pressure on the labor market from new entrants. Others Source: United Nations (2005b), medium variant. such as Pakistan and Senegal are projected to have slower growth. Dependency ratios are falling, steadily improving the economic fertility and have seen a peak in their youth countries in this group include Argentina, circumstances for investing in youth. Also numbers (China, Russia). Other countries Chile, Costa Rica, the Islamic Republic of falling is the relative size of youth cohorts, in this group include Albania, Armenia, Iran, Indonesia, South Africa, Sri Lanka, easing labor market pressures. Cuba, Georgia, Mauritius, Poland, and and Turkey. These distinct country patterns under- Thailand. Group 3. Countries in this group will pin the approaching plateau in the absolute Group 2. Countries in this group expe- experience a peak between 2010 and 2030, number of young people in the develop- rienced the fertility transition somewhat for some relatively sharp (India) and for ing world. While numbers are declining later than the first group and are seeing a others long drawn (the Arab Republic of in several countries, they are countered by peak about now (2000­10). In some cases Egypt).3 Other countries in this group increases elsewhere. At the regional level, the peak is relatively sharp (Vietnam). In include Bangladesh, Malaysia, Nicaragua, East Asia and Europe and Central Asia are others there is a long plateau, with coun- Peru, and the Philippines. already experiencing contraction, while tries projected to experience 20 to 30 years Group 4. Countries in this group will Middle East and North Africa, South Asia, of relatively constant youth populations not experience a peak in the foreseeable and Sub-Saharan Africa are set to grow, the after they reach their peak (Brazil).2 Other future (Pakistan, Sierra Leone). For most latter the fastest. Opportunities, capabilities, second chances: A framework for policy 2 Navigating through the transitions of the a youth lens on these policies would focus next generation of workers, leaders, and them on three broad directions (figure 2.1): entrepreneurs offers great potential (and c h a p t e r · Broadening the opportunities for young poses great risks) for growth and poverty people to develop skills and use them reduction in developing countries. The productively. human and social capital of the young · Helping them acquire the capabilities to will determine national incomes. Their make good decisions in pursuing those decisions, as the next generation heading opportunities. households, communities, work forces, and nations, will affect the welfare of everyone · Offering them second chances, to recover else in society. How can policies help? from bad decisions, either by them or Governments already do many things by others, such as their governments or that affect the lives of young people--from families. setting the broad economic context in which Governments provide education and they live and work, to providing education, health services for youth and set the broad to setting laws about early marriage, vot- economic policies and regulations for labor ing, and child labor. Are these activities suf- markets where they work. A youth lens on ficient or even appropriate? How can they policies, however, reveals important gaps. be improved? This chapter suggests that In many countries the quick and massive Figure 2.1 Transitions seen through three lenses focus policies and magnify impact YOUTH LENSES Low Policies affecting literacy human capital and knowledge formation during of basic skills youth transitions: despite higher Inexperienced as · Economywide primary completion decision makers Consequences policies and of poor outcomes institutions: Insufficient access Still forming longer lasting macro stability, to secondary and identity and than for adults "Youth friendly" investment climate, tertiary preferences policies governance, labor Remediation market regulations Few job Lack of resources more costly for · Education opportunities adults · Training Myopia and · Health services No platform for risk taking · Welfare and civic family services engagement · Infrastructure Opportunities Capabilities Second chances Source: Authors. 45 46 WORLD DEVELOPMENT REPORT 2007 expansion of primary education has resulted In sum, the three broad policy directions in unprecedented pressure for places at sec- that emanate from a youth lens imply that ondary schools but not always transmitted governments need not only reorient spend- literacy and basic life skills. Labor market ing and publicly provided services but also regulations may have very different effects improve the climate for young people and on younger workers than on older workers. their families to invest in themselves. These Youth in many developing countries lack a are the "youth friendly policies" (figure platform for civic engagement. Therefore, 2.1) that the rest of the Report elaborates. policy makers need to improve access to ser- Just as the three lenses have to be aligned vices that help the young develop and pre- for the image to be in focus, so must poli- serve their human capital. They also need cies be well-coordinated to have maximum to give them ways of beginning to use that impact. Opportunities can be missed if the human capital to sustain their livelihoods capabilities to grasp them are blunted or and participate fully in community life. misdirected. Having better decision-making At some point (the exact age varies across capabilities (agency) can lead to frustration cultures) young people make their own deci- if the opportunities are too far below aspi- sions to seize (and in some cases create) rations. Not having second chances can lead these opportunities. Choosing well from an to a free fall in outcomes, regardless of the expanded set of opportunities depends on opportunities or capabilities. developing the capability to define one's goals This framework derives directly from the and act on them--one's agency. Youth must human capital model familiar to economists make these choices at a time when they are (box 2.1). It is also important to add a gen- acquiring an identity and are inexperienced der filter to the youth lens. The youth tran- as decision makers.Governments can help by sitions vary for males and females. Societies providing information and tools to manage that try to protect the well-being of young the risks that many youth confront for the girls often wind up restricting their oppor- first time. In some cases, government poli- tunities and capabilities for schooling and cies also need to enhance the role of young meeting health needs. (See the spotlight on people as decision-making agents, particu- a gender filter on the youth lens following larly adolescent girls in societies where they this chapter.) are further behind in skills or where they are unable to choose for themselves. Because many young people are deprived Broadening opportunities of opportunities to develop their human Economywide and sectoral policies affect the capital or participate responsibly in the opportunities for human capital investment. decisions that affect them, they need second Public financing, provision, and regulation chances. The consequences of not offering of education and health systems have largely those chances will reverberate for decades determined not only the number and quality afterward, not only as youth become the of skilled workers available to the economy, next generation of workers and leaders, but but also of skilled parents and active citi- also as they become parents who influence zens available to families and communities. the succeeding generations. These pro- Economywide policies for aggregate public grams for second chances can be costly, but spending, the labor market, external trade, not as costly as remediation for adults. To and the money supply affect growth and lower costs, the programs should be well investment--and thus influence how many integrated in the overall delivery system for workers of different skills will be demanded. developing human capital. They also need To what extent are countries' human capi- to target those who need them most, resolve tal policies and institutions adequately pro- moral hazards if they encourage risk-taking, viding opportunities for young people to and supplement rather than replace family- develop their future skills--broadly defined based safety nets. to include not just work, but also social Opportunities, capabilities, second chances: A framework for policy 47 B O X 2 . 1 Applying the human capital model to young people [P]eoples of the world differ enormously in productiv- dictators the models would have them be, duti- Are investors in human capital well ity; ...these differences are in turn largely related to ... fully correcting for externalities and equalizing informed decision makers? the accumulation of knowledge and the maintenance opportunities for all.Instead, their behavior A third set of assumptions concerns the motives of health.The concept of investment in human capital simply organizes and stresses these basic truths ... may depend on their accountability.Is it to the and preferences of the young person as a deci- I would venture the judgment that human capital is ultimate beneficiaries? Is it to their supervisors? sion maker.The model assumes that the human going to be an important part of the thinking about Or is it to special interest groups, to voters at capital investor is well-informed about the ben- development, income distribution, labor turnover, and large, or even to their own families or pockets? efits and costs, discounts the future appropri- many other problems for a long time to come." Getting that right will determine outcomes, as ately at the prevailing economic discount rate, --Gary Becker (1964) in Human Capital examined in World Development Report 2004.1 If accounts for the riskiness of the investment by If the true test of the value of an economic theory these policy makers do not feel accountable to comparing it rationally with other risky assets, is longevity,the human capital model passes with young people, or if young people do not have and has well-formed views, not just about their flying colors.Its basics are simple and empirically a platform for holding providers accountable, a present preferences, but about future desired testable (and generally validated).An individual gap in youth's opportunities for human capital consumption bundles as well. will invest in his or her human capital--an investment may result (box 2.2). For many young people, the reality is that additional year of schooling,on-the-job training, they are inexperienced decision makers who Who makes investment decisions? or acquiring a healthy lifestyle--as long as the are only selectively informed about the risks, A second set of assumptions has to do with marginal gain from that investment exceeds its costs, and benefits of most human capital who is making the decisions about investing in added cost.The gains extend over a lifetime and investments.Some tend to be more myopic and human capital.The model assumes that a young are discounted to the present.If some of these impatient than adults, which may lead them to person makes his or her own decisions.The gains accrue to others,governments need to discount the value of long-term investments like age range 12­24 is precisely when the locus of stimulate individuals to take them into account human capital.They are still forming their own decision making shifts from parents or house- in making decisions.Public action may also be identities, too, so that their own preferences holds--and in most cases, both parties have a needed if poor individuals cannot mobilize the for consumption are still evolving and may be say in the final outcome. But how much say does resources to finance the investment now,despite easily influenced by peers.Extensions of the each party have? How do they resolve conflicts a promise of big gains in the future. basic model, discussed more in boxes 2.7 and if they have different views--such as whether Like all basic models, the elegance comes at 2.9, include the explicit modeling of information the young person should work and contribute the price of simplifying assumptions--assump- asymmetries, identity formation, the dynamic to the household income or go to school? Eco- tions that have led to extensions that this effects of cumulative learning, and the synthesis nomic models have tended to treat children as Report uses to analyze the hard realities facing of behavioral science with economic thinking.3 passive receivers of parental decisions.2 In some youth in developing countries. The broad policy lenses of opportunities, societies this assumption is a stretch of reality, capabilities, and second chances come directly To whom are policy makers accountable? a point especially obvious to parents with teen- from the human capital model and its exten- One set of these assumptions has to do with agers; in others it may very well describe how sions as applied to young people.The rest of this the ability and willingness of government to youth's opportunities can be limited by social chapter shows how. correct for the failure of the market so as to conventions.The model's extensions include optimize the welfare of all.Government policy bargaining among household members to makers do not always behave as the benevolent determine human capital investments. Source: Authors. skills? What would this "youth lens" imply · For countries that have long been able for the priorities for public action? to provide access to primary and lower The answers are far from easy because secondary education of acceptable qual- some constraints on young people vary ity, address barriers to expanding oppor- greatly across countries while others are tunities for building higher level skills in more common. This section focuses on the upper secondary and tertiary institutions. following policy priorities: · For all countries, help people build skills · Improve the quality and relevance of at work while they are young and most services that enhance basic skills, such apt to learn from work experiences--and as literacy and knowledge about health start toward an independent livelihood. outcomes, especially for low- and some · For all countries, provide opportunities middle-income countries that have raced for young people to be heard and to par- to meet quantitative targets for educa- ticipate in civic life outside the family--to tion for all. build their skills for exercising citizenship. 48 WORLD DEVELOPMENT REPORT 2007 Addressing these priorities poses diffi- Figure 2.2 Attending is not always learning cult (but not insurmountable) challenges Grade 6 Grade 4 Grade 5 for government. Services beyond the basic reading reading French mean higher unit costs, both for provid- test, 1995 test, 2001 test, 1990 ers and for beneficiaries, who face greater Percent opportunity costs. The case for public inter- 100 Net enrollment rate vention is not as strong for upper secondary Minimum mastery 80 and higher education as it is for primary. 60 Improving the quality and relevance 40 of services that enhance basic skills 20 A youth lens points to improving the qual- ity of basic services for children as well as 0 for young adults, especially in countries that Faso Togo Malawi Namibia d'Ivoire GuineaSenegal have recently improved quantitative indica- Mauritius Tanzania Colombia Morocco Cameroon Burkina Côte Madagascar tors of primary school enrollment. Despite Source: UNESCO (2004b). the higher primary completion rates, many Note: Net enrollment ratio is the ratio of the number of children of children enter adolescence ill-prepared for official school age (as defined by the national education system) who are enrolled in school to the population of the corresponding work, further schooling, and the practicali- official school age. Minimum mastery is calculated by multiplying ties of coping with life in a more connected the percentage of children in the study who have achieved the mini- mum standard by the percentage of children who made it to grade and complex world. Literacy and numeracy, 5. The grade 5 reading test is SACMEQ (Southern and Eastern Africa Consortium for Monitoring Education). The grade 4 reading test is the backbones of the arguments to justify PIRLS (Progress in International Literacy Study), which determines public subsidies to primary education, are the percent of students with reading scores above the 25thpercen- often not in place by grade 6.4 The most tile of the international benchmark. The grade 5 French test is PASEC (Programme d'Analyse des Systèmes Educatifs de la Confemen). recent UNESCO report to monitor the Edu- cation for All (EFA) campaign puts it starkly: "Simply focusing on quantitative goals such Basic services have lasting impacts well into as universal primary education will not young adulthood in both developed and deliver EFA. . . . [I]n many parts of the world, developing countries. Aside from primary an enormous gap prevails between the num- schooling and basic health care of those bers who are graduating from schools and age 6 and older, early childhood develop- those among them who have managed to ment programs, such as enriched child care, master a minimum set of cognitive skills."5 increase the probability of graduating from Fewer than a third of children in the coun- high school and attending college--and tries named in figure 2.2 achieved minimum reduce the likelihood of being charged with mastery of the relevant topic by grades 4­6, a crime.8 Studies in the Philippines find even though the average net enrollment rate that nutritional status at a young age leads was about 65 percent. to academic success, as measured by greater Improper health practices mean that chances of school enrollment and lower rep- many children enter adolescence stunted etition rates, even by the time students are and malnourished. More than 40 percent of 11 years old.9 In Turkey, a mother-child edu- all 10- to 14-year-olds in Cameroon, Gua- cation program providing early enrichment temala, Nepal, and the Philippines are ane- for young children and training and support mic.6 Ill health puts them at a disadvantage for mothers increased educational attain- when they go to school or work. For men ment, reduced delinquency, and improved and women who work in urban Brazil, a 1 the status of mothers in the family.10 percent increase in height leads to a 2.0 to One could go further and extend the basic 2.4 percent increase in wages or earnings.7 agenda to include grades 6­9 (or roughly Because many children enter adoles- middle school). These are grades necessary cence on the proverbial wrong foot, the first to consolidate the gains, and in some cases priority is still their adequate preparation. remediate the missed opportunities, from Opportunities, capabilities, second chances: A framework for policy 49 earlier grades. Many students in these grades for these services, which are more expen- "School is ... like an island, are adolescents 12­15 years old. Many coun- sive per learner, for all who want them? where there are all kinds of people tries are already expanding their definition They cannot, under present circumstances. and you learn a lot from them. of basic education to include this age.11 There are two ways out: one is to have more However, I know that real life is As countries embark on these extensions, options to finance these services, the other, not going to be like this, it will be however, the lesson from the attempt to to make delivery even more efficient. difficult, we will have to face that, universalize primary is clear--ignore qual- but at school you start learning ity at your peril. A balanced approach that Financing and targeting. Students and how . . . to be prepared." improves quality along with quantity may their families are already paying for a Young person, Cuzco, Peru mean higher costs, but there will be savings big part of these costs, especially at ter- January 2006 as well, because higher quality education tiary levels, in countries as diverse as Côte implies less repetition.12 d'Ivoire, Indonesia, the Republic of Korea, and Paraguay (figure 2.3). This would not Addressing barriers to expanding be especially worrisome if all promising opportunities for building higher young people could afford to finance such level skills beyond literacy, numeracy, training, because the returns accrue mostly and basic health knowledge to the individual student. More often than Countries need to broaden opportunities not, however, large numbers of poor (and in postbasic services--in upper-second- some not so poor) are unable to finance it. ary, technical, and higher education, and With private spending on higher education in training centers and informal settings, exceeding 60 percent of GDP per capita in including the workplace. On this, there is an Benin and Côte d'Ivoire, it is not surpris- emerging consensus, especially for middle- ing that tertiary enrollment rates are below income countries (chapter 1). But how can 5 percent--and very few of those attending already strapped governments afford to pay are poor. Guaranteed free universal access to higher education would hardly be a solu- tion--it is neither fiscally feasible nor equi- table because poor young people are severely underrepresented among the pool of sec- Figure 2.3 Private funds finance a big part of tertiary education costs in diverse countries ondary completers and could probably not afford to forgo work. Best would be provid- Public Private ing a financial environment for institutions Greece Korea, Rep. of to offer student loans. Because this takes Portugal time and scarce management resources to Hungary develop, scholarships--tightly targeted to Slovak Republic Argentina the poor and merit-based--would be war- Poland ranted to encourage promising youth from Chile low-income families. Malaysia Mexico Uruguay Spending cost-effectively. For govern- Turkey ments already struggling financially with Peru Paraguay providing primary education and basic Indonesia health care, it is even more important to Jamaica pay attention to cost effectiveness. Because India Côte d'Ivoire basic services are failing poor young people, Benin the lessons of World Development Report 0 20 40 60 80 100 200413 are relevant here, especially those Percentage of tertiary education on strengthening the accountability of ser- expenditure vice providers to clients and citizens (box Source: Calculated from UNESCO (2005). 2.2). The lessons apply perhaps even more Note: Countries are ordered by (PPP adjusted) GDP per capita from top to bottom. strongly, because many of the services used 50 WORLD DEVELOPMENT REPORT 2007 B O X 2 . 2 Making services work for poor young people-- World Development Report 2004 redux, with a youth lens World Development Report 2004: Making Services enhancing the power of clients to hold provid- · Citizen voice--improves services by increas- Work for Poor People developed a framework for ers accountable for service delivery; strengthen- ing the ability of young people to articulate improving delivery in such services as educa- ing the voice of citizens to influence, and pos- their needs and wants, and ensuring that pol- tion and health.The framework unbundles the sibly sanction, policy makers for decisions about icy makers listen (see box 2.3 for an example delivery chain into the relationships between service delivery; and ensuring that the compact and chapter 7). three sets of actors. As clients--patients in between policy makers and providers gives the · Provider compact--improves services by clinics or students in schools and their par- right incentives for the delivery of good services. ensuring that providers have the right train- ents--they are the ultimate beneficiaries of Making services work for poor youth is fully ing, motivation, and incentives to deliver high services.They have a relationship with the consistent with the 2004 Report's framework. quality.Examples include contracting with frontline providers, who are schoolteachers and Services that work well for the poor will, of private providers to deliver youth-oriented doctors, for example. In a competitive-market course, typically work for poor youth as well. services, with payment conditional on suc- transaction the consumer holds the service But a youth lens sheds additional light on the cessful implementation, and by training pro- provider accountable through the power of framework. As young people age they become viders to create more youth-friendly services the purse--by paying for satisfactory service or actors in their own right.They become an (chapters 3, 5, and 6). taking his or her business elsewhere. However, important and independent group of clients. for services such as health and education there They also become a distinct part of the citi- An overarching recommendation from the is often no direct financial accountability of the zenry, with their own needs, priorities, and goals 2004 Report rings even more true in the con- provider to the client, as when service provid- for policy.Viewing the 2004 Report's framework text of youth-oriented services--and echoes ers are financed by the government.There is through a youth lens produces the following through this year's Report. Although a variety then a"long route"of accountability--with implications. of interventions and programs have been tried, clients as citizens influencing policy makers, and little reliable evaluation of their impacts has policy makers influencing providers.When the · Client power--improves services by empow- been undertaken. So to guide youth-oriented relationships along this long route break down, ering youth to actively engage in the provi- policy development, it is urgent to build the service delivery fails. sion of services, either by giving them the knowledge base of what works under what The recommendations from World Devel- means to choose among providers or to par- conditions. opment Report 2004 center on strengthening ticipate in the management of decentralized the relationships between these three actors: delivery (see chapters 3 and 5). Source: Authors. by young people (rather than those used young people do not start full-time employ- by children) tend to be more expensive per ment too early and that they have free entry person. The one dimension not emphasized and mobility when they are ready. in that agenda, which does not differentiate among generations, is the voice of young Start at the right time. Poverty can force people in improving service availability, dis- children to work at home or enter the labor cussed in box 2.2 and in the next section. market prematurely, encouraging adoles- cents to drop out of school early. Because Enhancing opportunities to begin dropouts often never return, the short- a sustainable livelihood term income gain to the parents and fam- "Most pupils don't leave school Human capital, once developed, needs to ily comes at the cost of forgone education [on] their own. It is due to lack of be used productively to sustain a liveli- for the child. Their chances of being able to financial support. It is difficult for hood. This has long been recognized to be read, for example, may fall because literacy our parents especially in Africa to a main path to poverty reduction, because rates increase significantly as individuals go send their children to school." labor is what the poor have in abundance. from no schooling to 1­3 years of schooling How the young start off in their working and then again to 4­6 years.15 Kwami, 20, Togo lives has an enormous effect on their later Cash incentives to remain in school may September 2005 prospects. Because young people are on a enable youth to stay long enough to acquire very steep part of their learning curves, they a threshold level of basic skills that are diffi- can acquire skills quickly when working, cult to accumulate outside the school setting. an attribute that diminishes with age.14 For Literacy and numeracy would also provide a most of them, this means working in the strong base for more advanced skills, which private sector, which in most countries is the help in insulating one during downturns. largest employer. Policies should ensure that The East Asian financial crisis of 1997­98 Opportunities, capabilities, second chances: A framework for policy 51 hit the least-skilled and other marginal first few years of their working lives, become workers--women, young workers, the less basically unemployable thereafter, at least in educated, recent school dropouts, and first- the formal sector. time job seekers--in Korea the hardest. What policies could ease entry of young Young workers ages 15­29 accounted for people into the labor market? Broadening the lion's share of job destruction, espe- opportunities for young people's employ- cially young female workers.16 Job recov- ment is best premised on economywide ery also took longer for the less-educated growth that stimulates demand: a rising tide young worker. Similarly in Argentina, lifts the boats of young people and everyone workers with less education took longer to else. Some of these policies unintention- be reabsorbed into the labor force after its ally have a disproportionately large effect recent financial crisis.17 on young people. In many economies, such as those in East Asia, exports and foreign Free entry into the labor market. The direct investment expanded the demand for young continue to gain skills after they enter young workers and were a principal source the labor market, but because they are inex- for growth explaining the East Asian Mira- perienced, their unemployment rates tend cle.18 They have had a particularly stimulat- to be higher than those of adults. For indus- ing effect on increasing the labor force par- trial countries, the unemployment rate of ticipation of previously excluded groups, young people ages 15­24 is two to three such as young women (box 2.3). times that of adults, and for some develop- At the same time, some broadly based ing countries up to five to seven times (chap- policies, especially those regulating the labor ter 4). These differences may not matter so market, can hurt new entrants dispropor- much if they are due to the natural search for tionately. For example, minimum wages are jobs. However, they are an enormous prob- sometimes set too high, and some studies in lem if young people, unemployed for the the United States and Latin America have B O X 2 . 3 A tale of two exports: How electronics in Malaysia and garments in Bangladesh promoted work for young women in traditional societies Malaysia in the 1970s and 1980s and Bangla- Bangladesh reveal that young women preferred also delayed marriage, expect greater influence desh in the 1980s and 1990s experienced rapid working in the garment industry to agriculture in the choice of their marriage partner, as had increases in the labor force participation of or domestic service.Work in the garment sector already occurred in Malaysia. young women, especially in some export indus- meant higher wages and social status.Though Some long-term economic and social tries.19 In Malaysia the electronics assembly sec- factory work was strenuous, those who did it effects are unclear. Once married, young tor attracted a steady stream of young women had higher self esteem and autonomy.In a sur- women tended to leave wage employment, from rural areas to the state of Penang.Similarly, vey of Bangladeshi garment workers, 90 percent and most young women expect to work only the garment sector in Bangladesh drew many of young female respondents had a high opin- three or four years. Aside from the short tenure, young women from rural areas to such cities as ion of themselves, compared with 57 percent of many of the jobs were tedious, low-skilled, Chittagong and Dhaka. female workers in nonexport industries. had limited upward mobility, and were man- Women constituted 90 percent of the The salaries of these young women aged in traditional patriarchal ways.Young garment workforce in Bangladesh and the enhanced their status within their families as women experienced greater autonomy as they semiconductor assembly workforce in Malaysia. well as benefited their families.Their wages migrated to urban settings, but were some- These sectors provided many young women could contribute up to 43 percent of household times stigmatized for having a more Western, with their first opportunities to enter the labor incomes in Bangladesh.A 1995 study indicated individualistic lifestyle. Nevertheless, what is force: 93 percent of female workers in Bangla- that 80 percent of families of female garment clear is that these young women in Bangladesh desh's garment sector and two-thirds of women workers would fall below the poverty line with- and Malaysia have broken some new ground in Malaysia's electronics sector had no previous out their daughters'salaries.While these contri- and in the process contributed economically to work experience. butions did not always improve their bargaining their families and societies. The previous generation of Malaysian power in the household, they seem to have had women had neither the educational attainment more inputs into some key decisions, such as Sources: Ackerman (1996); Amin and others (1998); nor the employment opportunities of their the timing of marriage. In Malaysia work before Chaudhuri and Paul-Majumder (1995); Kibria (1995); daughters--nor the modern lifestyle associ- marriage has become the norm.Young women Ong (1987); Paul-Majumder and Begum (2000); and ated with wage employment.Case studies from working in Bangladesh's garment sector, having Rahman (1995). 52 WORLD DEVELOPMENT REPORT 2007 found that they can reduce the employment be entrepreneurs end up working for some- of youth.20 Employment protection laws one else at one time or another. provide stability to those already employed, For the bulk of the young workforce, but could inhibit employers from taking a the surest way to a better job that pro- risk in hiring promising but inexperienced vides adequate training and skill-building workers. A study of 15 Latin American and opportunities is to move eventually into Caribbean and 28 Organisation for Eco- the formal sector. Here, again, policies that nomic Co-operation and Development provide an incentive for the private sector (OECD) countries found the impact of such to hire young people will be important. In regulations on young people's employment Côte d'Ivoire, Rwanda, and Senegal in the rates to be more than twice that on prime- late 1990s and early 2000s, because of the age male workers'.21 Overly generous social small size of the formal sector and overly protection schemes have the same effect. high expectations, the unemployment rate Social attitudes, including discrimina- for those with postsecondary education was tion by employers and the reluctance of seven to eight times that for those with just families to let young people work outside primary education.23 Pursuing investment the home, can also suppress opportunities, climate improvements for firms can create especially for females.Young women already more and better jobs, as discussed in World contribute much to the economy through Development Report 2005.24 unpaid work at home, work notoriously The lack of domestic opportunities has missing from GDP estimates.22 The entry pushed young people to look outside their of such a trained and motivated pool into local environment, including outside the the formal labor force can produce higher country. Most of the world's migrants are growth because they are a large share of the young. Because they are some of the best population relative to those not working, and the brightest, many observers are con- even without a dramatic fertility transition. cerned about the brain drain (chapter 8). Discriminatory practices can change, too, Policies to get the most from international as countries develop, though some social migration include improving the infra- norms are remarkably resilient (see exam- structure for migrants to send remittances, ples in the spotlight on gender following enabling migrants to stay connected with this chapter). their home countries, and encouraging the successful return of young migrants. They Mobility. Once employed, young people also include better access to information on tend to be highly mobile, more so than jobs, as in the Philippines. older workers. Changing jobs is one way of getting better jobs and in the process accu- Opening opportunities mulating human capital. Policies and insti- to be heard outside the family tutions that hinder mobility thus especially The skills and the desire to interact with the affect the young. broader community and engage in civic life Except for the lucky (or the well-con- are also formed early (chapter 7). More- nected), many young people start work in over, having a voice expands opportunities the informal sector. Some thrive there, espe- for better service. ("Voice" is shorthand cially those who manage to set up successful to indicate not just expression but recog- businesses. This is a fairly small proportion nition and inclusion--a chance to con- of the labor force (chapter 4) and a fairly tribute to society and be acknowledged.) limited and risky route. For many young World Development Report 2004 argued people who find it difficult to get credit, self- that the voice of the beneficiary, the poor employment is a survival strategy, discussed person, can be routed directly through later in this chapter. Moreover, because suc- contact with the service provider or indi- cess in business can be elusive, many would- rectly through influence on policy makers, Opportunities, capabilities, second chances: A framework for policy 53 who then direct providers to deliver better services (see box 2.2).25 B O X 2 . 4 Young people's voice in budget setting This is true for all in society, but for improved outcomes in Ceará, Brazil young people--only beginning to make CEDECA (Centro de Defesa da Criança e do the budget was marked by the participa- their voices heard outside the home, less Adolescente)-Ceará started"Children and tion of young people who contributed to patient with incremental change, and Young People in Action--Participating in the discussions in a competent manner.To searching for their identity--the oppor- Budget Work,"in Fortaleza, a city with more add value to their proposed amendments, than 2 million people, in 2002. (CEDECA- they gathered in front of the Town Hall on tunities for expressing voice through vot- Ceará educates citizens on the importance the day of the budget vote.There was a ing or consultation are perhaps even more of Brazil's budget process.) mixture of surprise and incredulity among relevant. Opportunities to speak out are After identifying 50 young people from the councilors.They could not believe that particularly important because their par- different areas of Fortaleza, CEDECA-Ceará a group of young people could understand trained them on budget content and the budget, present proposals, make their ents may not fully represent their interests, process.The training covered public admin- voices heard, and above all protect the right as discussed more fully below. The absence istration, political organizations, budget to participation in the drafting of policies of such opportunities can make alternative legislation, revenues, expenditures, the bud- aimed at them. get cycle, budget execution, ways for civil forms of expression, such as violent behav- Of the 33 suggested amendments society to engage with the budget process, to the 2004 budget, three passed.The ior, more attractive. and the roles of executive, legislative, and project has since scaled up.The young Indirect routes of accountability--from judicial authorities. people in the project are now training oth- citizens to policy makers to service provid- The 50 young people researched and ers.CEDECA-Ceará says that their efforts documented the needs of young people in resulted in an additional allocation of ers--may become more important as people various parts of Fortaleza by making visits $400,000 to children and young people in vote or otherwise make their views known and taking photographs.Having learned to 2004 and $760,000 in 2005. after age 18, officially, but in practice, much analyze municipal budgets and interpret the The government of Sweden is sup- later. For most young people, such indirect results,they engaged with the Fortaleza city porting an evaluation of the effects of the government's formal budget process.In 2003 project on social and private outcomes, and routes are not available. Direct consultation they submitted 33 amendments to the bud- its results are expected to be ready in the and recognition through participation in get proposal for 2004,related to programs near future. implementing development projects or in for education,drugs,and sexual abuse. setting budgets are also key (box 2.4). The According to respondents, for the first Source: Swedish International Development time in the city of Fortaleza, the debate on Agency (SIDA) (2005). young are more likely to have opportuni- ties to be heard and have an impact in local (rather than national) settings. The effect of broadening such opportu- nities is not simply to ensure stability--it well-being, and that of the nation, if policy is often a way to get better outcomes if the makers recognize young people as deci- young can use their creative energies pro- sion-making agents who define their own ductively. It is a marvelous way to develop goals and act on them.26 skills in decision making, the topic of the Recognizing agency is not enough--it next section. must be resourced, informed, and respon- sible. That is, it must be "capable." What Developing the capabilities determines this capability? The section of young people as highlights the importance of addressing the decision-making agents most important issues youth confront as they participate more in decision making: With an expanded array of opportuni- ties, how should one choose among them? · Having access to sound information-- "Many parents fear for their And who should choose? The extent to necessary because young people have children and would `ask' them to which young people participate in deci- had less opportunity to learn from life follow their way of doing things, sions that affect how they acquire skills, than adults. without allowing youth to try new begin to work, and express a voice varies · Commanding real resources--necessary things, for fear of the danger." across countries, societies, and cultures. because young people may not be able This section explores how young people to afford to invest in themselves, even if Young person, Thailand can contribute enormously to their own they have promising prospects. December 2005 54 WORLD DEVELOPMENT REPORT 2007 Figure 2.4 Who has the final say? Young people's influence on their own key transitions varies · Deciding judiciously--necessary in the greatly across societies and for males and females light of their evolving preferences in Romania their search for identity. Malaysia Young people make more-- and more important-- Albania decisions as they age Parents, or adult guardians, make most deci- Iraq sions affecting the schooling, health, social life, and general well-being of children. As Ethiopia children age they gradually begin to exert more control over many decisions affecting Tajikistan their lives. Those who are 24 make differ- ent decisions from those who are 12--in a Bangladesh process that some have called a progression from adolescence to emerging adulthood.27 100 80 60 40 20 0 20 40 60 80 100 The pace of this shift (and the decisions) Percent Females Males varies widely across countries and cultures Work School Marriage and may have nothing to do with level of development. A 2005­06 survey in a set of Source: WDR 2007 InterMedia surveys. Note: The figure shows the percentage of people ages 15­24 who answered "myself" (rather than parents, govern- diverse countries asked whether young peo- ment, or other) to the question: "Thinking of [each transition: your current or most recent occupation; your years of ple ages 15­24 felt that they themselves had schooling; and your marriage partner], who has had the most influence?" the most influence regarding decisions across the important transitions of education, mar- riage, and work (figure 2.4). Despite the vari- B O X 2 . 5 Seen but not heard: Who decides when to ations, at least half the youth surveyed felt in drop out of school? Or when to marry? most cases that they had the most influence. Or do anything else? Notable exceptions are Bangladeshis and Tajiks regarding marriage and school.28 Who participates in the decision to drop For example, psychologists have conducted Understanding more about the nature out? A survey in Pakistan found big differ- experiments to show that"pro-social" and consequences of choosing among life's ences. More than three-quarters of male behavior develops with age.According to youths (ages 20­24) report a role in the Kohlberg (1973) it is around the age of 12 opportunities is critical to effective gov- decision to leave school, but fewer than a that individuals start to go beyond behav- ernment policy. Where parents, older rela- quarter of the responsible adults felt that ing according to rules and regulations and tives, or community leaders do most of the young males should be involved. Similarly, more as a matter of fulfilling obligations to deciding, their incentives, resources, and nearly 50 percent of female youths report others. Most of the experimental work sup- a role in this decision, while only 11 per- porting these hypotheses has been done in constraints matter most (see box 2.5). How- cent of adult respondents felt that young developed countries. ever, even when young people do not have females should play a role. The findings point to the need to outright decision-making responsibilities A third of respondents (ages 15­24) understand better the role of young peo- until they are young adults above 18 or even reported a role in the decision to drop ple in the decision making of households. out of school but that they were not able Models of household behavior should older, they can still exercise considerable to express an opinion during household treat children not as"dormant agents,"but, agency early on. discussions on this issue.While the contrast- like spousal pairs, as negotiating partners This independence is perhaps most ing percentages may represent only the in a bargaining game (see box 2.7). Some obvious for schooling and health behav- difference between a desired norm and recent historical analysis uses expenditure actual practice, they call into question both data from the U.S. in 1917­19 to show that iors, which are most difficult for adults to the actual agency exercised by adolescents youths were also able to amass bargaining monitor and control. Young primary school and youth and the reliability and veracity of power in the household as a result of their completers may dutifully enroll in second- adult responses. work. ary education in accord with their parents' Generational differences in views go Sources: Moehling (2005) and Sathar and oth- beyond education--and beyond Pakistan. ers (2002). wishes, but they will not necessarily go to class or study hard enough. The state may Opportunities, capabilities, second chances: A framework for policy 55 impose minimum ages at which young from Albania, Bulgaria, and China indicate "I would support young people people may marry, and parents may try to that simply having a youth in a household [having] freedom in decision impose rules about whom young people increases the likelihood of adopting new making mechanisms. And also, will see, but sexual activity may (and does) technologies.32 in Turkey, I would work for take place surreptitiously. Adolescents can eliminating `family dominance' express their agency even in seeking work What more is there to know? Plenty.Young in young people's choices and opportunities. A study of 21 villages in 1998 people's general knowledge may not extend decisions." in the Indian state of Karnataka, all just over to human capital investment and risk-taking Sezin, 22, Eskisehir, Turkey 100 kilometers from the technology hub of behavior. A 2003 survey of otherwise well- December 2005 Bangalore, indicates that 12 percent of all informed and educated Vietnamese youths boys ages 10­14 were current or returned ages 14­25 indicates that fewer than 60 per- migrants. Of these, almost a third were con- cent of rural youths had ever heard of syphi- sidered strongly "autonomous migrants"-- lis or gonorrhea. Only about a third had they left without any parental pressure or heard of the menstrual cycle. In a country even involvement in decision making or in where traffic accidents are the leading cause facilitating work or living arrangements.29 of death and serious injury for those ages So, applying a youth lens means that 15­19 and where motorbike use by urban policy, in addition to addressing parents, young people exceeds 70 percent, only about needs to account directly for the behavior a quarter use a helmet--many were simply of young people, even when they are very not convinced of its protective value.33 young or living in what are seemingly tra- The availability of some information is ditional settings. How? By enhancing their doubtless rooted in social norms for topics access to information, their command over appropriate for discussion. A recent survey resources, and their skills to decide. in Russia quotes a 20-year-old woman from the North Caucasus as saying: "I am inter- Supporting the search ested in diseases and infections transferred for information sexually. The problem is that I cannot ask Young people know a lot. And because they my parents, because my breeding does not are on average better educated than previous allow me to discuss such issues with [them]." generations, their rates of literacy are much Instead, the most common source of health higher, especially in low-income countries, information cited by young people is televi- where the literacy rate of those 15­24 is 75 sion.34 Peers, some equally ill informed, are percent, compared with 59 percent for those often an alternative to families or schools as a 35­44.30 Indeed, the effect of this literacy is source of information. greater if one of these young people hap- The consequence is ill-informed conclu- pens to live in a household where all others sions because the noise in the information are illiterate--because this would enable a flow can be profound. Many studies show transfer of information to those who can- that young people tend to overestimate sex- not read, especially important in remote ual activity and other high-risk behaviors in areas where using fertilizer and other tech- the population, putting more pressure on nologies or following a regimen for the them to conform.35 treatment of tuberculosis entails under- standing directions. Researchers argue that What can be done? Three main avenues such proxy literacy, the exposure of house- can be followed to address the problem. One holds to people who are literate, would add is to improve the curriculum already being another third to India's literacy rate.31Young taught in existing institutions. School-based people are also much more frequent users reproductive health education programs can of the most ubiquitous source of informa- be effective in increasing knowledge and the tion now available worldwide--the Inter- adoption of safe sexual behavior (chapters net (chapters 1 and 8). Household surveys 5 and 6). Moreover, evaluations have found 56 WORLD DEVELOPMENT REPORT 2007 no evidence that sex education increases ing school year relative to those who were sexual activity among youth.36 Such pro- not given the information.38 grams are clearly not enough, however. A third way is that purveyors of infor- There are concerns, too, that sex education mation must themselves be well trained and programs reach only a select group--those motivated with the right incentives. Recent still in school. research on absenteeism among service pro- A second avenue is to develop options viders in health and education, discussed in outside the traditional institutions of skill World Development Report 2004, shows that formation, including social marketing of this is easier said than done. In Latin Amer- reproductive health services targeted to ica, teaching quality responds to the level youth, peer counseling programs, mass and structure of compensation. However, media programs, and workplace and com- a review of seven country initiatives indi- munity outreach services targeted to youth. cates that for an incentive scheme to work, These programs can promote good repro- it must be very selective, of sufficient size, ductive health if targeted to the right audi- and very closely linked to monitorable per- ence and linked to services already in place. formance.39 (Otherwise they may simply lead to unmet Using peers is also a promising mecha- demand.) Although not subjected to rig- nism for transmitting knowledge, but orous evaluation, they have been tried in peers are no panacea, for they too must Cambodia, Thailand, and Uganda--coun- be informed--and they must be a positive tries that have contained the spread of HIV/ influence. For example, U.S. residential job- AIDS. training programs, which allow participants These programs need to address the to avoid interactions in their dysfunctional source of the information gap.In the absence neighborhoods and develop their own self- of information,people will act based on their image, are more successful than programs own perceptions. Teenage girls in Kenya, that do not have the expensive housing when asked about the age patterns of HIV, component.40 massively underestimated the prevalence of the disease among males in their late twen- Helping the young command resources ties, making them more complacent about Of the observable markers of transition-- "sugar daddies." A randomized intervention such as completing school, leaving home, provided HIV profiles by gender and led to a beginning one's work life, marrying and closer matching among like-age groups.37 becoming a parent--financial indepen- An information gap also occurs when dence seems to be the best predictor of youngsters observe trends from a very self-perceived adulthood, in rich and poor select sample. In the Dominican Republic countries alike.41 In this Report's Bangla- in 2001 a survey of boys enrolled in the desh consultations, the recurring phrase was final year of primary school accurately esti- "nijer paye darano"--that is, "standing on mated the returns to completing primary one's feet." It refers to economic indepen- school consistently with estimated earn- dence--not having to depend on parents ings profiles but severely underestimated or other family members for livelihoods.42 the returns to having a high school degree. As young people begin to be financially This was because they based their estimates independent, they naturally confront more on observing the wages only of youth who severe restraints on their own consumption remained in the neighborhood after com- and investment decisions. pleting secondary schools (the high earners For the nonpoor who can still rely on moved out). Students at randomly selected resources from families, this is not as much schools were then given information of the of an issue. Indeed, one of the reasons that estimated actual earnings profiles. Follow- the age of leaving home is getting later even up surveys in 2005 indicate that those given in richer countries is that young adults rely the information were 12 percent more on their families to get a firmer economic likely to be attending school in the follow- footing.In France and Germany,the share of Opportunities, capabilities, second chances: A framework for policy 57 young men ages 20­24 living in the paren- that is many multiples of initial earnings. tal home increased from around 56 percent Some promising schemes to mitigate these in 1986 to 62 percent in 1994; in Italy the issues are being offered, such as loans that increase was steeper, from 50 percent to make repayments contingent on income 66 percent, even among men 25­29.43 For actually earned (see box 3.6 in chapter 3). those from poor families, however, or for those who for one reason or another (such Resources for starting a livelihood. Aside as orphanhood or family breakup) can no from being a barrier to education, inexperi- longer rely on family resources, the result ence and lack of collateral can also hinder could lead to underinvestment in human youths who have the motivation and skill capital, to a bad start on the road to a sus- to start new businesses. Evidence indicates tainable livelihood, and, for females, to a that liquidity is more of a constraint on weak negotiating position in marriage. young entrepreneurs than on more estab- lished ones. Their lack of a credit history Resources for investing in human and their inexperience puts them at a dis- capital. Young people who have to make advantage to adults. This is a tricky area choices about investing in skill formation are because it would not be prudent to direct confronted with substantial costs (see figure government funds to subsidize risky com- 2.3). Out-of-pocket costs tend to vary--for mercial ventures. Some recent experiments the half of university students in private to expand microcredit to young people may universities in Argentina, Brazil, Chile, and point to a promising avenue (box 2.6). Colombia, they range from 30 percent of GDP per capita to 100 percent.44 Even for Strengthening negotiating positions in fam- those at free public universities, the oppor- ilies and among peers. Emerging research tunity costs are substantial. Because of the shows that command over resources is high payoffs to tertiary education in those an important way to enhance the agency countries, such costs would not be an issue of young women in marriages, especially unless liquidity were a binding constraint-- ones arranged more for the convenience of and it is. A recent study for Mexico found families and parents than the young couple. that households with the same permanent income are less likely to send their offspring to university if they had a bad year eco- nomically.45 In many countries, parents lack B O X 2 . 6 Is microcredit an answer to relieving the the wherewithal to pay for such schooling young's resource constraints? directly or simply expect that young people will contribute significantly as they get older. Microcredit has enhanced opportunities for people (ages 12­24) have taken advantage of But will young people do so, since they face poor people in countries like Bangladesh. It microcredit loans.Ten percent of youth have more binding resource constraints? supplies credit in areas too poor or remote gotten a loan,compared with 33 percent of all to be served by traditional banks.Through adults.Of all loans to youth,the majority come The obvious way to ease this constraint group lending, microcredit provides poor from microcredit institutions (73 percent).Of is to provide credit to the student--she people, especially women, with a viable all who were eligible to receive loans in 1999, who benefits pays, even if it's later on. In alternative to the collateral required for 17 percent received their first loan before the traditional loans. Group membership serves age of 25.There is no indication that young advanced countries where students bear a as a monitoring device to ensure that each people's loans ran into disproportionately significant burden of the cost, education is member makes her loan payment, thus more problems than those to adults.They financed by students who take out loans. absorbing part of the risk. Research shows tended to pay similar interest rates and to use Such credit schemes could not work with- that it has improved the income-earning them for similar activities. opportunities of women. Because young Initiatives have been undertaken out government support, because commer- people also lack collateral, can microcredit recently to broaden the use of microcredit cial loans are not available to the poorest open doors for them? Would such a system schemes as a way to channel resources and students, who lack the collateral or parental work for a group that is, on average, more empower young people.It is important to guarantees to back them up. Moreover, the mobile and perhaps less risk-averse? evaluate these experiences and to share The final answers are not known,but analy- their lessons. pressures to begin earning a livelihood are sis of data from Bangladesh may be indica- high enough even without the enormous tive of microcredit's promise.There,young Source: Data calculated from Khandker (2005). responsibility of having to pay back a debt 58 WORLD DEVELOPMENT REPORT 2007 Indeed, in some societies, young people, women--and that credit to women had the especially girls, are never empowered to greatest impact on variables associated with make decisions themselves. That power women's power and independence.47 simply shifts from the parent to other older people, such as the husband or mother-in- Helping young people to decide well law.46 Poor youths are also more likely to Information, once received, has to be filtered succumb to pressures to accept money or and assessed before it can be acted on. This is goods in exchange for sex, putting them at difficult even for the most confident of people, increased risk of HIV and other sexually and it can be paralyzing for those unaccus- transmitted infections. Many young people tomed to the analytic processes of decision take up smoking because of peer pressure. making.Tosomeextentthiscouldbeaddressed Programs that help young people earn inschools.Educationalsystemsthatemphasize a living or that provide credit may also memorization and rote learning may teach empower them to withstand social or peer facts but fail to inculcate the analytical skills to pressure to engage in risky behavior. An come to well-informed decisions. evaluation of the justly celebrated micro- Such life skills can be taught in schools credit programs in Bangladesh showed that with changes to the curricula. Sometimes they had an effect on education, health, and this requires changing the environment labor market outcomes of both men and for learning to counteract well-entrenched B O X 2 . 7 Are youths rational (at least according to economists)? There have been three ways of modeling what weights.Halpern-Felsher and Cauffman (2001) overappreciate their current preferences and separates youths from adults, based on some found adults to be more competent decision project them onto their future preferences recent literature reviews.They are based on makers than youths when asked about the without duly taking into account the effects of traditional economic analysis, developmental short- and long-run costs and benefits of dif- changes in their situations in the interim.For psychology, and the intersection of the two, ferent interventions such as cosmetic surgery example, while increased education can bring a behavioral economics. or whether to participate in an experimental young man better job prospects as an adult, he Traditional economic analysis.In this frame- medical study.Competence was measured by may underestimate the importance of a good work, dictated by expected utility maximization, consideration of all options, risks, and long-term job to him in his adulthood because his current as long as benefits exceed costs in present consequences.The differences were particularly preferences do not put great weight on job value, an activity will be pursued, and the same striking between adults and younger adoles- opportunities. goes for risky activity. Even addictive activi- cents (those in the 6th and 8th grades). Third, risk-taking is affected by past engage- ties, like smoking, are considered rationally There is evidence that youth and adults ment in risky behavior. Suppose that one has by forward-looking agents who choose their use different considerations in decisions.When engaged in a risky activity in the past, the out- own optimal consumption, fully cognizant of asked about the perceived consequences of come of which--good or bad--is still uncertain, long-run negative consequences. Disparities such risky activities as smoking, drinking, and and is now choosing whether to repeat the between youths and adults thus do not rely on drug use, youths considered consequences same activity. If the activity brings short-run variant psychologies or levels of rationality, but involving social reactions more than adults. gratification, the higher the probability of a bad can result from income differences or sensitivity Steinberg and Cauffman (1996) show an outcome, the more likely one will engage in it to price. Indeed, preliminary evidence suggests inverted-U relationship between age and again, because the marginal risk is decreased. that, in the decision to pursue risky activities, susceptibility of peer influences.Susceptibility The implications are frightening.The higher the youths are quite sensitive to economic factors, increases between childhood and early ado- risk of contracting AIDS from unprotected sex, spawning incentives to curb teen smoking. lescence, peaking around age 14, and declines the more likely someone who has had unsafe Gruber (2001) cites a U.S. government study during the high school years. sex in the past will do so again.The worse that concludes that increasing the price of ciga- Behavioral economics. According to behav- the long-term effects of drugs, the higher the rettes is the most reliable method for reducing ioral economics, incongruous behavior and probability that someone who has so much as teen smoking. decision making by youths and adults may be experimented with them will become a regular Developmental psychology. Unlike traditional the upshot of different perspectives regarding user. economic analysis, developmental psychol- the future.First, younger people are more likely So economists would generally say that ogy treats the decision making of youths and to heavily discount the future over the present. youths are probably quite rational, given their adults as separate processes.In some studies, Activities with high short-run benefits or high preferences, the resources they command, and the factors youths take into account when long-run costs (or both), and underinvestment the perceived costs of their actions.Whether making decisions or formulating opinions are in activities with both high short-run costs and youths would say the same of economists is a subset of those used by adults, while in other high long-run benefits would be the result. another matter. cases, youths and adults consider different fac- Second, future utility can also be affected tors altogether, or with significantly different by projection bias--the tendency of people to Source: Basu, Ku, and Zarghamee (2006). Opportunities, capabilities, second chances: A framework for policy 59 practices that channel behavior, some of as an inducement for them to perform which may not even be recognized. Research well in school but also to ensure that they from secondary schools in Thailand indi- "own" the decision to attend, circumvent- cates that girls who study in single-sex envi- ing age-old biases against girls' schooling. ronments do better in math and develop The Bangladesh Female Secondary School more leadership skills than girls in mixed- Stipend program, targeted to girls ages sex environments, a finding consistent with 11­14, transfers money to a bank account research in developed countries.48 for girls directly, contingent on their per- For many young people, however, it is forming well and staying unmarried. While more than a matter of learning decision- rigorous impact evaluation has yet to be making skills. Because many youths are still done, indications are that this was a major seeking their identity, their decisions are factor in the large increase in girls' second- more complicated. The process of develop- ary school enrollment in the country (see ing a person's sense of self has been used chapters 3 and 6). by some economists to explain seemingly irrational acts, including young people's Offering second chances tendency to engage in risky behavior, such Broadening the opportunities available to as joining violence-prone gangs, despite young people for services and the start of the low expected economic payoffs to such a sustainable livelihood, and helping them activities.49 The search for identity, and the choose wisely among them, are the pri- influence of peer groups on a young per- orities. Many young people cannot take son, go some way toward explaining how advantage of these opportunities, how- much risk a person is willing to take and ever, because they were ill prepared dur- how much to discount the future relative ing younger ages. The worldwide median to the present. Experimental results show, primary school completion rate is now for example, that young people tend to take approaching 85 percent, but many coun- greater risks in driving automobiles when tries and regions still lag behind (Sub-Saha- there are peers present than do older adults ran Africa is at 55 percent) either because in a similar situation (box 2.7). children have dropped out or never went Coming to the correct decision for some- to school in the first place (box 2.8).52 In one who discounts future costs too heavily,or countries ravaged by civil war, the propor- who adopts risk-taking behavior to belong to tion could amount to an entire school-age a group, requires making that person aware generation (see the spotlight on Sierra of future consequences and forcing him or Leone following chapter 7). her to confront them. Incentives can some- Bad nutrition at an early age can also times influence such decisions, especially if diminish life chances for many young the young would not take the effects on oth- people. Height may reflect previous health ers (or on themselves in the long run) into investments and is accepted as a proxy for account, even if they had the information earning ability later in life.53 Many people and knew how to decide for themselves. In are disabled and cannot take advantage of the United States, higher cigarette prices sig- opportunities--estimates vary widely, but nificantly reduce smoking by young people.50 they range around 75 million to 150 million In Indonesia, a 1 percent increase in the price of those ages 10­24.54 of cigarettes reduces consumption by about Inevitably, some young people will a third of a percentage point.51 By contrast, choose or be led down the wrong path. banning cigarette and alcohol advertisements Early childbearing is still a problem in many and setting minimum ages for drinking are developing countries. More than half of all attempts to increase the price of such behav- infections of HIV/AIDS occur among young iors, but evidence suggests that their impact people under 25. In the United States, those tends to be weak (chapter 5). ages 15­19, 7 percent of the population, Innovative schemes that have channeled account for more than a fifth of the arrests subsidies directly to young girls--partly for violent offenses and roughly a third of 60 WORLD DEVELOPMENT REPORT 2007 strictures. Early marriage, deep-rooted in B O X 2 . 8 Even those still in school need second chances some societies, can close off opportunities prematurely. A sound primary education (six years) is the figure).Malawi has an enrollment rate Bad outcomes are also, however, partly relevant for youth not only as a foundation; of almost 75 percent for all its 15-year-olds, for many young adults, it is also the place compared with 70 percent for Vietnam, but the result of the misjudgments of young where they are still learning. Some have almost all the Vietnamese kids are in sec- people inexperienced in decision making repeated many years.Others started school ondary level, compared with only about 5 and less averse to risk. Even though adults very late.Still others are trying to get back percent for Malawi.The situation persists for into the educational system after many older children.Malawi has a higher enroll- and youths are not that different in their years away--in postconflict countries, per- ment rate for 19-year-olds than Vietnam, ability to carry out decisions about risky haps a generation of school-age kids. but half of them are still in primary school. situations, experimental results show that The age-enrollment profiles for differ- What programs would be needed to make youths differ in how they value the con- ent educational levels show that, although sure that they are educated to the same some poorer countries such as Vietnam levels as those who are much younger? sequences of choices. This could be due have managed to get their young people Vietnam has no such issue with age-mix- to myopia, or to a preference for immedi- through primary levels more or less on ing: all 19-year-olds in school are already at ate gratification. It could also be linked to time, others such as Malawi have not (see secondary levels. the search for identity--a subject of end- Large numbers of older youth are still in primary school in some countries less fascination to academics (see box 2.6). The differences in valuing the consequences Vietnam Malawi of choices could have tremendous effects, Age Age especially when the costs of actions are 15 15 borne now--such as going to school or 16 16 using contraception--but the payoffs occur 17 17 farther in the future.58 Recent research also shows that these 18 18 Primary school attitudes may also have physiological roots. 19 19 Secondary school Brain imaging indicates that the part of 0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80 the brain governing the ability to assess Enrollment (percent) Enrollment (percent) risk matures last, around the early twenties Source: Authors' calculations based on Demographic and Health Surveys (2002 in Vietnam and 2000 in (box 2.9). Malawi). How such biological development affects behavior is not clear, but it offers yet another possible explanation for the observed will- ingness of the young to take more risks than "The system in my country ... those for all property crimes.55 Figures are adults. In some countries the insurance does not track down those [who] difficult to get in developing countries, but market takes account of this difference. The `drop out' in time to ensure that there is evidence for some Caribbean coun- car insurance industry is well aware of the they do not end up becoming a tries, such as Jamaica, where nearly half the penchant for the young to indulge in risk liability to the world. That is what murders and more than half the crimes are taking. In Ireland a 19-year-old pays almost is happening here; thousands of committed by young people, mostly males, three times more than a 29-year-old for car young people walking aimlessly in between the ages of 17 and 25.56 insurance.59 Companies have long offered the streets [with no] jobs, no social incentives to reward responsible past behav- services, and therefore nothing." Why second chances? Mitigating the ior, such as a good track record. Some have effects of past policies and behaviors even reportedly offered to reward future Chernor, 21, Sierra Leone These undesirable outcomes are partly a leg- behavior, such as good grades, presumed to September 2005 acy of past policies and practices that failed be associated with prudence. However, such to deliver the basic services so important to insurance markets are thin in developing the foundations for better youth outcomes. countries and fail to protect the vast major- Apart from not spending enough for these ity. For example, the one insurance market services, the failure to spend properly has that young people could use is motor insur- had enormous implications that were docu- ance, given that the leading cause of death mented well in World Development Report among males outside Africa is injury, most 2004 (see box 2.2).57 Undesirable outcomes caused by vehicular accidents. The spread are also a legacy of family and community of this industry is confined to only a few Opportunities, capabilities, second chances: A framework for policy 61 B O X 2 . 9 Brain development among youth: Neuroscience meets social science A decade ago,the prevailing notion was that starting at the back of the brain and ending otherwise. brain growth ended at about the age of 2 years. at the prefrontal cortex.The prefrontal cortex · Learning and teaching strategies should Since then,we have learned that brain growth regulates impulses, risk taking, planning, deci- be timed to increase neurodevelopmental continues well into adolescence (between ages sion making, empathy, and insight.Research also capacities.Because neurodevelopmental 10 and 19) and into young adulthood (see the shows that the cerebellum, recently discovered maturation occurs at different chronological figure below).During this period the brain under- to be important for mathematics, music, deci- ages for different people, their inability to goes a series of changes,and parts of the brain sion making, social skills, and understanding grasp a concept at one age does not mean associated with social skills,problem solving,and humor, continues to grow through adolescence that they are unable to learn the material.This identifying emotions mature only by the early and well into emerging adulthood.The last speaks to the risk of educational"tracking" twenties.However,this process of brain develop- structure of the brain to stop growing, it devel- based on comprehension or performance ment cannot entirely explain adolescent decision ops until the mid-twenties. examinations at a young age. making and behavior.Nor does it override the Implications for social policies · Without a fully mature prefrontal cortex, effect of the environment--parents,schools, What does this new brain research mean for adolescents may be more impulsive than communities--in which young people live. understanding adolescent decision making adults and perhaps more susceptible to peer Brain development: and behavior? Although much more research is influences.This impulsiveness--especially in arborization and pruning needed before definitive policies can be recom- reactive decision making, as when faced with The brain is made up of nerve cells--about 10 mended based on the new brain research, it a situation or threatened to make an immedi- billion of them--connected by branches or den- suggests some interesting policy considerations: ate decision--suggests the value of second- drites.These branches move information from chance programs. · The loss of neuronal excitation in adolescence one cell to another, but these connections are is associated with a rise in depression, espe- It is, however, too early in the research to not soldered together; rather, there are spaces cially among adolescent females, suggesting draw definitive conclusions about brain devel- between the branch of one cell and the body of a biological basis for the epidemiological opment and behavior.Also, physical develop- another.These spaces are called synapses, and finding that gender differences in depres- ment interacts with the social environments to information moves from cell to cell across these sion start around the time of puberty.These determine behaviors and outcomes. So parental spaces by releasing tiny packets of chemicals. biological changes combine with external behaviors and expectations, effective schools, When there are abnormalities in the chemicals sources of stress to increase the risk of suicide communities that are youth oriented and sup- in the synapses, a variety of clinical conditions for youth in many countries of the world. portive, all make a difference in determining result, such as depression and attention-deficit · As the brain matures during adolescence, young people's behavior and how well they and hyperactivity disorders. alternations in the synaptic chemicals may learn complex decision-making skills. Different parts of the brain handle differ- influence learning (drugs for attention-deficit ent activities--that much is well-known.What disorders improve information transfer at the Source: Blum (2006). is new is the finding that during adolescence level of the neuronal synapse).For example, certain areas of the brain grow in size and other antidepressive drugs may allow for certain regions become more efficient.For example, excitatory neurotransmitters to stay in the the area of the brain responsible for language space between two brain cells longer than more than doubles in size between ages 8 and 14.Consequently, language acquisition Timeline of brain development is optimal at those ages. So, too, connections grow and strengthen between the brain stem and the spinal cord, increasing the connections between the emotions and what the body feels. Throughout childhood and adolescence, more and more nerve cells grow sheaths around them called white matter or myelin.This is like build- ing a superhighway, allowing information to be 10 cm 10 cm interpreted and recalled much faster than was ever possible as a young child. Prenatal Postnatal These structural changes are only some Week: 0 6 12 18 24 30 36 Month: 0 6 12 18 24 30 36 Year: 4 8 12 16 20 24 of the brain's alterations during adolescence. Another major change is called"pruning." Cell birth Majority of neurons Throughout early childhood, the number of con- Migration Fewer neurons, primarily nections between cells increase, and because in cortex the process is much like the growth of branches Axonal/dendritic outgrowth on a tree, it is called arborization.It allows a Programmed cell death child's brain to be very excitable--which is why Synaptic production children seem to be perpetual motion machines. In adolescence, many of those branches die-- Myelination through pruning.The brain is less excitable but Synaptic elimination/pruning also more efficient in carrying information. The pruning follows a consistent pattern throughout adolescence and young adulthood Source: Andersen (2003). 62 WORLD DEVELOPMENT REPORT 2007 developing countries and its penetration is In richer countries, the incidence of return- shallow in middle- and low-income econo- ing to the parental home, at least temporar- mies, where average premiums per capita ily, has increased after a failed marriage or are only a tenth of those in high-income relationship, job loss, or devastating illness. economies.60 Households able to cope with this are less at risk and therefore need less subsidy than What's to be done? young people who have frayed family con- Given the legacy of past policies and behav- nections, especially in developing countries. iors, and the many market failures, what's There is evidence, for example, that young to be done? Remediation tends to be rela- people return home in smaller numbers if tively costly for many of the transitions, they are from families that have experienced which is why early attention to basic needs the death of a parent or divorce.62 Orphans for younger children, as well as broadening are thus more vulnerable and a priority for opportunities for young people and help- social programs. ing them decide wisely, are essential. It is Another reason for targeting is that difficult to get precise estimates because some programs that mean to benefit mostly remedial programs are so different from the youths but do not identify them as the main "I used to steal cellular phones, but original ones. A rough order of magnitude beneficiary may fail to reach the neediest, that is dangerous. They can put you comparing adult basic literacy programs who may not seek the service if they are to prison and even kill you. I did with primary education programs in Ban- less aware of where it is. This is why cam- that because I wanted money and gladesh and Senegal yields a range of dif- paigns to address reproductive health, such under the influence of my gang. I ferences that make the former one to three as postnatal care services for very young quit that life and now I can walk the times as costly, even without taking into mothers (chapter 5), are so aggressive at streets again . . ." account their opportunity costs.61 So, the social marketing. first lesson: try to get it right the first time. Freddy, 16, Honduras Beyond prevention, countries rightly Coordinating with the mainstream. Sec- January 2006 have many programs that try to mitigate the ond-chance programs tend to be discon- effects of undesirable but sadly inevitable nected, run as separate initiatives by NGOs youth outcomes. These include reinvesting or agencies (chapter 9). The lesson from the in human capital (adult literacy programs), few analyses of these programs is that inte- treating those infected with communi- gration and coordination make them more cable diseases, providing drug rehabilita- effective. To avoid the danger of devel- tion, integrating the long-term unemployed oping very costly parallel programs, it is into the labor force (retraining programs important to ensure reentry. One example for dropouts and public works schemes for is graduate equivalency programs, which the young), and reintegrating young people allow dropouts from secondary systems to into the social fabric (demobilization pro- take classes that will eventually get them the grams after civil wars, the juvenile justice equivalent of a secondary diploma (chapter system). These programs cover all the tran- 3). Another is the U.S. community college sitions that the Report deals with. system. Designed originally to provide sec- Three other lessons seem to matter most: ond chances for adults, it is now being used target programs closely, coordinate them increasingly by young high school graduates with mainstream "first-chance" programs, as a second-chance program--three-quar- and mitigate without encouraging risky ters of all remedial students are in commu- behavior (or encouraging moral hazard). nity colleges.63 By the same token, it is important for Targeting. Because of the cost of second- mainstream programs to be flexible so chance programs, it is important to direct that early mistakes do not turn into per- them to the neediest, as with other costly manent liabilities. Some countries track schemes. Many young people have access to students as young as age 10 into differing family resources that can act as a safety net. ability schools--while others keep schools Opportunities, capabilities, second chances: A framework for policy 63 comprehensive. A recent study of coun- such as treatment. That would be unethi- tries' achievements over time shows that cal as well as wasteful. Instead, it is to build early tracking not only increases educa- incentives that encourage the care-taking tion inequality (there is no catching up behavior to persist even in treatment. despite segregation) but may also lower performance.64 The three pathways implied by the youth Remediation with accountability. No lenses in the framework discussed here-- one doubts that all people, including opportunities, capabilities, and second youths, should face the consequences of chances--need to be applied to each of their risky behaviors--for their own good the transitions introduced in chapter 1. and for society's. In the case of criminal Sustained greater spending on the basics, behavior, moral justice and deterrence especially quality, is a top priority in the demand it. After one has strayed, though, poorest countries. In others, more attention what is best for society? Rehabilitation is to upper secondary and tertiary education very costly, but the payoffs are highest for is key. However, it is not all about spending. young people who still have a lifetime of Equally important is to ensure that young potential productivity ahead of them. For people, with the support of their families, those who commit crimes at a young age, are stimulated to invest in themselves. As what is most critical is that they face the with any investment, improving the climate consequences of their action without being for investing in human capital can raise the made to lose hope. In the Philippines, 3,700 returns and lower the risks. The returns in children are reported to be in adult jails, the form of a good job and an active civic "in conditions detrimental to their health life can be enhanced by policies that level and well-being."65 Many of them--some the opportunities for young and old alike. with relatively minor misdemeanors, some Prudence in taking health risks can be homeless--are incarcerated with hard- encouraged by the right incentives. ened criminals. The consequences should The youth lenses must be aligned to be commensurate with the gravity of the get the most impact. An illustration of the crime, and programs should allow young magnitude of this joint effect can be seen people to rehabilitate themselves where that by considering the returns to a hypotheti- is possible and appropriate--"restorative" cal secondary education project using data rather than "retributive" justice, in the par- from the Dominican Republic, Indonesia, lance of chapter 7. More demobilization and Mexico.66 Take a project that enhances programs are being designed to rehabili- opportunities by building enough schools tate young combatants (see the spotlight on to achieve a gain in upper secondary enroll- Sierra Leone following chapter 7). ment to 57 percent from 52 percent of the Any remediation program confronts relevant age cohort. Earnings and cost data what economists call a moral hazard. If for Mexico indicate that such a project someone knows that the consequences of would have a benefit-to-cost ratio of 1.03 in his or her risk-taking behavior are miti- present value terms--an acceptable return gated by a government program or insur- on investment that broadens opportunities. ance, that person may engage in more risk- Recent research shows that secondary taking behavior than warranted. There students severely underestimate the real has been some fear that the availability of returns to completing secondary school.67 antiretroviral therapy for HIV may cause A project that not only provided another young people to take fewer precautions. secondary school seat but also enhanced the In Kenya, condom use fell after the gov- capability of young people simply by pro- ernment announced reported "cures" for viding information to correct their misper- AIDS (chapter 5). The solution to such ceptions of the returns to education, would problems is not to deny second chances, raise the enrollment rate to 62 percent. The 64 WORLD DEVELOPMENT REPORT 2007 benefit-to-cost ratio for this project would The framework introduced here must be 1.68--a much better investment. be applied in a way that takes into account Finally, a project that makes the invest- different economic, demographic, and ment less risky for youth would increase social environments. Some countries deal enrollment even more. The risk could be well with the next generation of develop- in the form of an economic shock that ment challenges--others, less so. Countries would make secondary education unafford- could also be characterized by the speed at able, forcing them to drop out--a waste of which young people start to make decisions resources. If a student or her family had for themselves: in traditional societies the the insurance of a scholarship scheme in young make fewer decisions about major the event of an income shock (a second life transitions, and in others they exercise chance), enrollments would rise to 69 per- independence faster than their predeces- cent. Such a program, combined with the sors. How countries can apply the principles first two, would have a benefit-to-cost ratio in these different settings across each of the of 2.15--better, yet again. transitions is the subject of part 2. spotlight A gender filter on the youth lens Gender differences emerge sharply with the onset of puberty, affecting the life trajectories of girls and boys in profoundly different ways. Governments have a range of policy levers to counter the inequalities in opportunities, capabilities, and second chances of young women and men. Some policies are not targeted--they equalize opportunities indirectly by addressing constraints. But disadvantaged from the start, many young women also need directly targeted programs to help them catch up. Life trajectories diverge early-- undergone significant change. Across coun- Programs that equalize and permanently--for tries, school attendance and labor force opportunities can reduce adolescent girls and boys participation rates have risen faster for gender disparities young women than young men, in both Initiation rites in puberty, which vary from Broadening opportunities so that young the 15­19 and 20­24 age groups.5 However, circumcision ceremonies in some African men and women have better access to ser- the schooling experience can still be gen- societies to debutante balls in the Ameri- vices will reduce gender disparities.Govern- der-sensitive. For example, in Kenya where cas, signal the break with the homogene- ments have a range of policy levers--pric- adolescents attend mixed schools and are ity of childhood and the emergence of ing policies, legal and regulatory reform, fairly free to associate with the opposite sex, gender-based expectations for girls and better designed service delivery, selected teachers view girls negatively, often saying boys. With these rites of passage, the lives investments in infrastructure--that could that they are too lazy or less capable than of young men and women begin to follow have equalizing effects even without gen- boys are to learn.6 social trajectories defined by gender. Gen- der-based targeting. How? Through alle- Time-use studies indicate other differ- der-defined roles and responsibilities tend viating the constraints that disadvantage ences in the lives of adolescent girls and to curtail girls' opportunities and decision- girls. Some could reduce the relative costs boys. The amount of time devoted to work, making capabilities but broaden those of associated with investing in girls' produc- paid or unpaid, rises with age for boys and boys. In some societies girls are married tive skills. Others could increase the infor- girls, but girls tend to work more hours very young. In others, families severely mation to counter labor market biases than boys, spending long hours fetching restrict girls' freedom to protect their vir- against girls, or increase infrastructure ginity before marriage.1 In still others, water and firewood, cleaning and cooking, investments that ease the burden of house- and minding younger siblings.7 Time-use adolescent girls often become pregnant hold work for girls. and have children outside marriage.2 These studies in Kenya, India, Nicaragua, Paki- Consider the gender-specific impact of stan, and South Africa show that, on aver- gendered trajectories for young women an ostensibly gender-neutral program, such age, girls ages 15­29 work about one hour and men are revealed in gender differences as building more schools in rural areas. more a day than boys.8 in youth outcomes. More schools should benefit both boys and For young women in many regions, The interplay of differences in sex girls, but studies show different effects for the transition from school to work does (biology) and gender (society) also shapes boys and girls. In countries as diverse as not take place or is interrupted because of the different disease burdens for adoles- Pakistan and Ghana, reducing the distance marriage and childbearing or because it is cent boys and girls. In addition to health to a secondary school elicits greater respon- socially unacceptable for them to work for risks associated with early and frequent siveness in schooling for girls than for boys, pay. Compared to young women, men are childbearing, young women ages 15­29 partly because girls' enrollment is lower to more likely to work for pay--especially suffer disproportionately from HIV/AIDS, begin with and because distance is a big- if they are not enrolled in school--and to unipolar depressive disorders, panic dis- ger constraint for girls, for whom personal work less on domestic chores and unpaid orders, and fires. Rape and domestic vio- security is a concern for parents.12 economic activities.9 The pressures on lence account for 5­16 percent of healthy Trade policies also can have dispropor- boys to earn income may explain the rise years of life lost by women of reproduc- tionately larger effects on young women's tive age.3 In contrast, young men suffer in school dropout rates for boys.10 Girls employment opportunities. Take Ban- who are not in school are more likely to more from disabilities related to violence, gladesh, where young women joined the alcohol use, and accidents,4 patterns simi- participate in unpaid domestic work and export-driven textile sector in large num- household surveys may not capture such lar to those in high-income countries, bers, increasing their wages dramatically. work. This is evident, for example, in the suggesting common risk-taking behaviors The increase in women's employment and percentage of "idle" girls in many develop- and underscoring the global nature of the earnings enhanced their bargaining power ing countries: by age 14 the proportion of problems of violence and drugs in adoles- within their households.13 Gender-informed girls who are absent from both school and cent males. health safety in the workplace and other reg- paid employment in six developing coun- In education, the trend has been toward ulations to ensure good employment condi- tries ranged from 6 percent in Nepal to 44 gender convergence, marking one area tions for female workers should increase the percent in the Republic of Yemen.11 where girls' transition to adulthood has gains for young women even more--and 66 WORLD DEVELOPMENT REPORT 2007 Percentage increase in middle school health care, and nutrition, girls and young Facilitating young women's school-to- enrollment due to Oportunidades in rural areas women tend to bear much of the costs. As work transition requires targeted training in Mexico by sex and grade, 1997­2001 household incomes rise, spending on these programs and labor intermediation ser- items also rises,with girls and young women vices. To attract young women, Chile Joven, Percentage increase 40 often benefiting proportionately. For exam- an employment training and traineeship Boys Girls ple, giving larger cash transfers to families program for low-income, high-risk youth, to send and keep girls in schools, as with included awareness campaigns for the busi- 30 Oportunidades in Mexico, has increased ness sector to recruit women, offered child enrollment for girls (see the figure). care services, provided gender training for 20 A pilot education project in the prov- trainers, and expanded the traineeships in ince of Balochistan, Pakistan, subsidized firms for young women. Sixty-five percent 10 the establishment of private neighborhood of participants found a job on finishing the schools. Parents were given the financial program, and 70 percent of them found a 0 resources and technical assistance to con- job in their area of study, proportions much 1 2 3 tract a school operator to open a neigh- higher than for the control group, and the Middle school grade borhood private school, with the financial returns to labor market performance were resources being tied to the number of girls Source: Parker (2003). better for women than for men.15 Similarly, enrolled in the new school. Opening a the youth job training program ProJoven counter the consequences of their inexperi- new neighborhood school increased girls' in Peru--created in 1997 to help economi- ence in the workforce. enrollments by 33 percentage points, a far cally disadvantaged youths ages 16­24 to In addition to general policies and pro- larger effect than on boys' enrollments. This enter the formal labor market--increased grams, gender-targeted programs must project was for primary education. Because real incomes and reduced the gender wage be implemented especially because gen- distance to school is an even greater worry gap. Fifty-four percent of the participants der-defined transitions to adulthood often for secondary school girls in Pakistan, the were young women. Before the program, severely limit girls' voice or exposure to relative gains for girls of greater school young men were earning 45 percent more information. When low-income families availability at secondary level is likely to be than women--after, only 2.7 percent are forced to ration spending on education, even higher.14 more.16 The young founders of Grupo Cultural Afro Reggae, after losing many friends to the vio- lence and drug trade in the favelas of Rio de Janeiro, decided that a teenage death should no longer be accepted or expected. They created a program of music, dance, and cul- tural workshops to steer children and youth away from the drug trade and violence of the favelas--and to break down stereotypes and communicate with broader society. (http://www.afroreggae.org.br/) Transitions EVERY GENERATION GOES THROUGH TRANSITIONS. For infants, it is the process of weaning. For those about to enter the "golden years," it is the process of retiring. For what this Report has been calling the next P A R T II generation of workers, households heads, and leaders, the following five transitions can be anticipated: · Learning as adolescents and young adults · Beginning to work · Taking risks that impact health · Forming families · Exercising active citizenship When and how these transitions occur vary enormously across countries. But the patterns can be represented in a highly stylized (but empirically based) way in the figure. At the age of 12, most chil- dren are in school. They start to leave school shortly thereafter and Percent of age cohort almost all are out by age 24. Children begin to work at an early age 100 In school in developing countries but most do not work full time until they Working are at least teenagers. Young people also begin to engage in risk- taking behaviors, such as having sex, smoking, trying drugs, which Taking health risks have possible health implications. Young people start puberty early 50 but form families later. Finally, young people gradually begin to make themselves heard outside the family and to exercise citizenship. The chapters in this part of the Report discuss how countries can Forming families Exercising citizenship address the challenges posed by these overlapping transitions, by pass- 0 12 18 24 ing country policies through the three lenses of opportunity, capabil- Age ity, and second chances. Learning for work and life 3 Young people need to acquire the right sions about their own education, but face knowledge and skills to become productive constraints in doing so. Enabling them to workers, good parents, and responsible citi- make better education choices requires c h a p t e r zens. Learning takes place in many environ- developing their behavioral skills and ments--home, school, the workplace--but involving them as stakeholders in their own most investments in learning take place in education, as well as providing better infor- schools. Those investments need to happen mation about learning options and the job during childhood and adolescence, and the market and providing the financial incen- investments in adolescence are needed to tives to make better choices. make earlier investments pay off. Learning opportunities need to be pro- Despite great progress in primary school- vided for all, including young people who ing in developing countries, the preparation failed to acquire basic skills the first time of youth for work and life is very low, just as around. Society cannot afford to neglect demand for skills and knowledge is rising. them--without second chances, these Past education policies focused on increas- young people and their families would ing the number of people who go through be condemned to poverty. Governments the education system, rather than learning should therefore develop a system of reme- that takes place in schools. This chapter dial education, equivalency programs, liter- asserts that, to improve the skills of young acy programs, and skills training that takes people for work and life, education oppor- into account the diversity of these young tunities must be made more relevant to the people. needs of all young people as learners and future workers, parents, and citizens, and Educational preparation of young people need to be provided with the tools to develop their capabilities so they youth for work and life is low can make the most of opportunities. The demand for workers with postprimary This involves improving educational education, particularly tertiary education, is preparation for adolescence by providing increasing as a result of skill-based techno- quality basic education (including lower sec- logical change and the growing importance of ondary) for all. It also involves meeting the knowledge (chapter 1). Rising primary com- growing demand for postbasic skills, by pro- pletion rates are substantially increasing the viding diverse and flexible learning options number of potential secondary-school goers. in upper secondary and higher education; At the same time, many children in South by implementing a relevant curriculum that Asia and Sub-Saharan Africa drop out before teaches practical subjects, thinking skills, and completing primary school (or never start). behavioral skills; and by connecting school The transition to secondary school and work. To be successful, these reforms remains a barrier around the world, even must be abetted by teachers who are well- in countries with high primary completion prepared and motivated as well as schools rates. Why? Because of demand factors-- that are accountable for student learning. low levels of preparedness, the perceived Education opportunities are not enough irrelevance of secondary schooling, and the when young people cannot benefit from high direct and indirect costs--and because them. Young people make important deci- of low physical access to secondary schools. 68 Learning for work and life 69 Figure 3.1 The transition to secondary school is a stumbling block for many young people Indonesia (2002) Nicaragua (2001) Completion of each grade (percent) Completion of each grade (percent) 100 100 80 80 60 60 40 40 20 20 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Grade Grade Senegal (2000) Thailand (2002) Completion of each grade (percent) Completion of each grade (percent) 100 100 80 80 60 60 40 40 20 20 0 0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Grade Grade Richest 20% male Richest 20% female Poorest 40% male Poorest 40% female Source: http://econ.worldbank.org/projects/edattain. Kaplan-Meier estimates based on nationally representative samples of 10- to 19-year- olds. Note: The richest 20 percent and poorest 40 percent are derived from an index of assets and housing characteristics. Secondary school typically starts in grade 7. The transition to secondary school is espe- tests--which evaluate 15-year-olds' abili- cially a problem for the poor (figure 3.1). ties to apply basic skills--is approximately In middle-income countries such as Indo- 20 percent lower than the Organisation for nesia and Thailand, the poorest 40 percent Economic Co-operation and Development experience a sharp drop in the proportion (OECD) average of 500 (figure 3.2), and completing the transition from primary worse for the poorest quintile of the stu- to secondary; in poorer countries such as dents in these countries. In some countries, Nicaragua and Senegal, the drop in comple- such as Mexico, performance is substantially tion rates begins earlier, before plunging lower than in others at similar incomes. even faster during the transition. In many These numbers are all the more worri- countries, particularly in South Asia and some because the students in these assess- Sub-Saharan Africa, gender differences tend ments are a select group. Those who have to get accentuated in the transition to sec- already dropped out of school--likely the ondary school and remain large. poorest and lowest performing--are not The educational preparation of youth included. In Brazil, only slightly more than for work and life is very low in develop- 60 percent of 15-year-olds are in school ing countries, particularly among the poor. in grades 7 and above. These low levels of The average performance of adolescents in learning achievement in secondary schools the poorest countries on Programme for reflect in part the failure of primary and International Student Assessment (PISA) secondary schools to provide the basic skills 70 WORLD DEVELOPMENT REPORT 2007 Figure 3.2 Learning achievement in poor countries can be very low Proportion of 15-year-olds in school Mathematics (grade 7 and above) Score Percent 600 100 80 500 60 400 40 300 Richest quintile of students 20 Average score Poorest quintile of students 200 0 on on BrazilTurkey RepublicPolandMexico Republic KoreaStates BrazilTurkey Mexico KoreaStates Indonesia ThailandTunisia Uruguay Hungary of ThailandTunisia Uruguay RepublicPoland Hungary of Republic Federati Indonesia Federati Rep. United Rep. United Slovak Czech Slovak Czech Russian Russian Source: Authors' analysis of PISA data. Note: Countries ordered by gross national income (GNI) per capita in 2003, from lowest to highest. needed for work and life. They also reflect The many children who stop in primary inequalities in access to good schools. Most school or drop out before completing it of the difference in test scores between acquire only minimal skills along the way. the poor and the rich is explained by their In Mali, only about 20 percent of 15- to 19- attending different schools. year-olds had completed the primary cycle of six years of schooling--of those, only 60 Figure 3.3 The ability to read a simple sentence increases with schooling, but few acquire percent could read a simple sentence (fig- even that most basic skill ure 3.3). In Nepal 15- to 19-year-olds who had completed the five years of the primary 15- to 19-year-olds who can read a simple 15- to 19-year-olds who have completed sentence, by highest grade completed at least each grade cycle could typically read a sentence--but Percent Percent among the close to 30 percent who make it 100 100 no further than grade 3, fewer than 60 per- cent could read a sentence. Rising demand for postprimary educa- 80 80 tion and poor preparation for work and life pose three policy challenges: 60 60 · Expanding postprimary learning oppor- tunities to meet increasing demand, while increasing the readiness for, and quality of, 40 40 postprimaryeducation.Althoughthemea- sured expansion of secondary education 20 20 opportunities (that maintain a minimum level of quality) is most relevant for coun- tries with increasing demand for second- 0 0 0 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 ary education (Brazil, Mexico), improving Grade Grade learning achievement is most relevant for countries with high secondary education Dominican Republic (2002) Kenya (2003) Nepal (2001) coverage (Hungary, South Africa). Mali (2001) Philippines (2003) · Enhancing young people's ability to make Source: http://econ.worldbank.org/projects/edattain and additional authors' analysis of Demographic and Health the most of these opportunities by alleviat- Surveys data. Learning for work and life 71 ing the decision-making, information, and hood investments pay off. For example, the financial constraints they face. This chal- effect of preschool interventions on learn- D E F I N I T I O N 3 . 1 lenge is most relevant in countries where ing achievement may not persist without Skills and knowledge the opportunities exist (secondary schools) appropriate further schooling investments. but many young people do not take advan- In the U.S. Head Start preschool program, The types of skills discussed in this Report include thinking skills tage of them (Morocco, Thailand). the initial improvements in test scores often (critical and creative thinking), · Providing second-chance learning oppor- vanish among children who attend poor behavioral skills (perseverance, tunities for out-of-school youth who lack quality schools.4 self-discipline, teamwork, the ability to negotiate conflict and the basic skills for work and life, which Investments during early childhood reap manage risks), specific knowl- is most relevant for countries with sig- significant dividends, and can reduce the edge (including numeracy and nificant numbers of young people who intergenerational transmission of inequal- literacy), and vocational skills (a never went to school or dropped out ity. Many young children in developing mix of specific knowledge and skills to perform jobs that rely on before completing primary school (Paki- countries suffer profound deficits in nutri- clearly defined tasks). stan, Senegal). tion, health, and cognitive and socio-emo- Basic skills denote the set tional development very early in their lives, of minimal abilities needed for with lifelong consequences for educational further learning,work,and life, A solid foundation: achievement, employment, and earnings.5 including numeracy and literacy Improving the readiness and basic levels of behavioral A study in Ecuador demonstrates that large skills such as perseverance,self- for postprimary education differences in language acquisition among discipline,and self-confidence. Learning is a life-cycle process, in which the children from different socioeconomic Postbasic skills include thinking skills,higher order behavioral skills timing and continuity of investments mat- backgrounds become larger as they mature, (decision-making skills,teamwork, ter.1 Investments in learning need to happen partly because of differences in child health the ability to negotiate conflict during childhood and adolescence; failures and parenting skills.6 Programs that com- and manage risks),specific knowl- to invest at this stage are very costly to rem- bine health, parenting skills, and preschool edge applied to real-life situations, and vocational skills. edy later. Learning is most intense in child- education are likely to be most success- hood and adolescence, when physical and ful, and rigorous evaluations of programs intellectual capabilities are growing rapidly. in Argentina, Jamaica, and the Philippines From a neurological perspective, childhood demonstrate their effectiveness in devel- and adolescence are critical periods for the oping countries.7 Such programs must normal development of most skills--there be combined with efforts to improve later is no good substitute for investments during education for those investments to have the these years (see definition 3.1 of skills and greatest effect. knowledge). For example, language skills Early entry, adequate progress, and are much more easily acquired by children learning basic literacy and numeracy than adults. Different abilities are formed at in primary school are keys to success in different stages of the life cycle--although postprimary education. In many countries, language skills are hard to develop after however, youths are still attending primary adolescence, behavioral skills (such as moti- school, particularly in Africa and Latin vation,persistence,self-confidence,and self- America. The proportions of primary stu- discipline) can still be developed through a dents beyond the primary school age are 34 person's early twenties (box 3.1).2 percent in Brazil, 28 percent in Kenya, and Because learning is cumulative, invest- 27 percent in the Lao People's Democratic ments in learning during childhood and Republic.8 Overage enrollment is mainly adolescence have larger returns than later due to repetition in Brazil, late school investments because they increase the pro- entry in Kenya, and a combination of the ductivity of those later investments. For two in Lao PDR. In rural Bangladesh, late example, early childhood interventions can school entry and grade repetition in pri- increase subsequent learning achievement. mary school reduce the chances of going to Also, the impact of job training programs secondary school and completing it. Mak- on earnings is typically larger for those who ing school entry at age 6 compulsory and have more schooling.3 Investments during secondary school more available increase adolescence are also needed to make child- the chances of starting secondary school.9 72 WORLD DEVELOPMENT REPORT 2007 B O X 3 . 1 The ignored side of skills development: Building behavioral skills for school, work, and life Some high IQ people can fail in life because they leads to higher wages.They affect employment country contexts.The Lion's Quest program in lack self-discipline, while low IQ individuals can and the choice of occupation.They also lead to the United States is being tried in Japan, and do well thanks to motivation and persistence. significant declines in smoking, marijuana use, nonformal teaching methods are being used by Several studies in countries as diverse participation in illegal activities, and teenage many programs in many developing countries, as India and the United States show that job pregnancy and marriage. although none been well evaluated yet (see stability and dependability are the traits most box 3.2). valued by employers.Yet academic and policy School-based programs Programs outside school discussions focus almost exclusively on mastery School-based mentoring programs provide The Make a Connection program strengthens of specific knowledge. Although it is some- evidence of the malleability and importance of the connections of young people to their com- times believed that behavioral skills cannot be behavior in adolescence. A randomized impact munities, to their families and peers, and to measured, psychologists have developed tests evaluation of the Big Brother­Big Sister mentor- themselves by developing such behavioral skills to measure these skills--and they are used by ing program in the United States shows that it as self-confidence, motivation, teamwork, and companies to screen workers.10 reduced the likelihood to have initiated drug or conflict management, as well as critical and Behavioral skills, developed from a very early alcohol use, to hit someone, to miss school, or creative thinking skills; together they are often age through late youth, have long-lasting effects to lie to parents. It also led to higher grades and referred as"life skills."It operates in 25 countries, on schooling, work, and social outcomes.They greater competency in school and work.The including Brazil, China, the Russian Federation, incorporate many traits, including motivation, Quantum Opportunity Program, which offered and South Africa, adapted to local needs. In the persistence, self-discipline, self-confidence, and disadvantaged minority students long-term Philippines, it targets indigenous out-of-school the ability to weigh options and come to a deci- mentoring and financial incentives for activities youth.The program lasts for about 13 months sion (decision-making skills).They also include aimed at improving social and labor market and includes training in identity, cultural appre- social skills, such as teamwork and the ability to skills, led to higher high school graduation rates ciation, and leadership, after which participants negotiate conflict and to resist peer pressure. and lower arrest rates. Such programs have not identify livelihood projects and are eligible for Significant differences in behavioral skills yet been tried in developing countries. loans to finance them. Self-reported retrospec- appear across income levels at early ages and Several classroom-based programs to teach tive information reveals significant effects on persist over time.The long-term impacts of early behavioral skills directly have also produced employment, school reentry, and community childhood interventions come mainly from persistent effects in controlled studies in the service. the social skills and motivation they impart to United States and the Netherlands.Their sucess- children and the better home environment they ful role in reducing risky health behavior has Sources: Bowles and Gintis (1976); Carneiro and Heckman (2003); Cunha and others (2005); Hahn produce. In the United States, behavioral skills been most frequently evaluated, but they also (1999); Hahn, Lanspery, and Leavitt (2005); Heck- measured at ages 14­24 reduce the probability improve academic and other behavioral out- man, Stixrud, and Urzua (2006); Schweinhart, Barnes, of dropping out of high school and increase the comes. Such programs often have curricula and and Weikart (1993); Sternberg (1985); and Zins and probability of attending college, which in turn teaching methods that can be adapted to other others (2004). Improving the quality of primary educa- a diverse and flexible menu of learning tion requires a renewed emphasis on basic options, so that young people can fulfill literacy and numeracy combined with some their potential. of the policies mentioned later in the chap- ·Make school curricula more relevant ter, such as student-centered teaching meth- by teaching practical subjects, think- ods, teachers who are well-prepared and ing skills, and behavioral skills as well motivated, and schools that are accountable as blending the academic and voca- for student learning. Supplementary reme- tional curricula in upper secondary. In dial education can also play an important addition, strengthening the connection role (see second chances section). between school and work facilitates the school-to-work transition. Enhancing postprimary Hire motivated and well-prepared teach- education opportunities · ers and make schools accountable for The postprimary education system should student learning to ensure success of serve the diverse needs of young people as education reform. learners and future workers, parents, and ·Implement cost-sharing strategies, public- citizens. This section shows what is needed private partnerships, and efficiency- to do this: enhancing mechanisms to finance the · Provide quality basic education (primary expansion and improvement of postpri- plus lower secondary) for all, as well as mary education. Learning for work and life 73 Expanding options and between school participation and distance improving the organization to secondary school, and little evidence of postprimary education that increasing school availability reduces inequality in enrollment by household Providing basic education for all. Countries wealth or gender. can provide all youth with basic skills for work and life by deferring selection and spe- Diversifying education options while ensur- cialization until after lower secondary and by ing quality. Upper secondary and higher making lower secondary school part of the education have to accommodate diverse basic and compulsory education cycle. Inter- student needs, interests, and capabili- national test scores show that early tracking ties--diversity that increases as they expand significantly increases inequality in learning and become mass systems. The vocational achievement--and reduces learning.11 The education sector in developing countries secondary education reform in Chile moved is small (22 percent of student enrollment) all vocational specialization to the upper sec- relative to that in OECD countries. Upper ondary school. This allows time for building secondary is still heavily directed to more a solid academic base for later occupational academic university degrees, with programs specialization. Deferring selection and spe- oriented to the labor market playing a mar- cialization can be combined with compul- ginal role. However, graduation rates are sory schooling laws, which have been shown higher in systems geared to work (Malaysia) to have a positive effect on educational than in those geared to university (Argen- attainment and other social outcomes.12 tina and Chile).16 In Tanzania and Tunisia, using entrance Higher education is also heavily con- exams to strictly ration the number of centrated on more academic university students progressing to secondary led to degrees, though this is slowly changing with overcrowding in primary school and lower the appearance of new institutions, includ- student performance overall.13 Early track- ing technical institutes offering short-term ing or selection may also create or perpetu- degrees, community colleges, polytechnics, ate social exclusion, as in some Caribbean distance education centers, and open uni- countries that sort students into better and versities. Shorter and more occupationally worse schools according to test scores, each oriented programs have half or more of the with a different uniform.14 Where there students in China, Jamaica, Malaysia, and are fiscal constraints on expanding lower Zimbabwe.17 secondary, more places can be created by a The expansion and diversification of combination of sharing costs (with a com- postprimary education systems can be pensating demand-side mechanism for the greatly facilitated by reaching out to the poor) and working with the private sector private sector through public-private part- (see finance section below). For example, nerships. Governments should encourage Tanzania discouraged private schools, but private participation while ensuring quality Kenya allowed them and even provided standards. The share of the private sector is some subsidies--and during the 1960s and larger in higher education (33 percent) than 1970s secondary school enrollment in Kenya in upper secondary (25 percent).18 In upper expanded much faster than in Tanzania. secondary, the share has remained stable Building schools can meet increases in or declined in most countries, but it has demand for secondary education, but these increased in tertiary, especially in Brazil and investments will not necessarily increase Peru. Despite some positive trends, there is school participation or reduce inequal- still more room for private sector involve- ity in many developing countries. A study ment in postprimary education. using data from 21 developing countries Public-private partnerships allow sys- shows that most rural residents live within tems to expand in a fiscally constrained fairly easy reach of a formal primary school environment, particularly in the more but not a secondary school.15 It finds there expensive tertiary sector, and improve is typically a small negative relationship learning outcomes and efficiency overall by 74 WORLD DEVELOPMENT REPORT 2007 increasing choices and injecting competi- For the noncompulsory and more tion. For that competition to work, public expensive upper secondary sector, using institutions need sufficient autonomy and competitive entrance examinations will resources to manage for results (see finance help ease fiscal pressures, but financing section below) and private institutions need mechanisms similar to those for lower sec- to be accountable for meeting well-defined ondary (described above) will also create quality standards. more spaces. Those who do not pass the Several studies conclude that govern- examination will need to be offered alterna- ments should provide information and qual- tive learning options (vocational) and the ity assurance while promoting diversity.19 possibility of going on to higher education. Few developing countries have accreditation Well-designed higher education admission and evaluation systems for higher education, tests are likely to be beneficial, because edu- resulting in a proliferation of low-quality cating the most capable students can foster private providers (as in Cambodia). Chile innovation, driving the economy. Geor- and the Republic of Korea set quality stan- gia recently reformed its tertiary entrance dards lower at the entry point (licensing) exam, which limited access and improved to give new institutions the chance to grow, the quality of students, but a remaining and later make the standards more stringent challenge is to provide learning alternatives (accreditation) for both public and private to those who failed to make the cut.22 institutions to allow fair competition. Improving the relevance and quality "School prepares spectators instead A flexible education system. Diversified of postprimary education of creators." postbasic education systems need enough The low learning achievement in develop- flexibility to allow students to experiment and ing countries shows that schools are failing Young person, develop their full potential. Open systems can to prepare young people for work and life. A Buenos Aires, Argentina facilitate student mobility by recognizing rel- high-quality education system must improve December 2005 evant prior experience, degree equivalencies, the relevance of school curricula by teaching and credits earned elsewhere. Recent reforms students the practical knowledge, thinking, in secondary education have upgraded pre- and behavioral skills demanded by the labor viously terminal vocational tracks, allowing market; using teaching methods that lead vocational education graduates entry to ter- to high learning achievement; and blend- tiary education after taking school-leaving ing the academic and vocational curricula. examinations. For example, vocational grad- It must, in addition, strengthen the connec- uates in South Africa and Tunisia can now tion between school and the local economy qualify for any higher education institution. to facilitate the school-to-work transition For tertiary education, many institutions and boost economic development. The suc- throughout the world have adopted credit- cess of these reforms relies on motivated and based courses, including single institutions well-prepared teachers, who ensure a safe (the University of Niger), networks of insti- school environment, as well as schools that tutions (such as the Indian Institutes of are accountable for student learning. Technology), and entire national university systems (as in Thailand).20 In Colombia, Improving the relevance of school people already in the labor market can get curricula. In many developing countries, university-equivalent certification through the secondary curriculum is not relevant to any accredited training institution. Many the social and economic needs of students. young people need to combine school and Norisittaughtinawaythatmaximizeslearn- work, and they can be accommodated-- ing attainment or keeps students engaged in through, for example, part-time educa- school.23 Sometimes, it is relevant only to the tion. Today, part-time tertiary education needs of a few privileged students. In addi- accounts for only 5 percent of enrollment in tion, few countries routinely assess their university in developing countries, and 13 curricula for relevance and effectiveness. percent in non-university programs.21 Comparative analysis of national secondary Learning for work and life 75 curricula between 1985 and 2000 shows very well evaluated, in both developed and devel- little change.24 The tertiary education curri- oping countries.29 The structured teaching cula also have many of the same problems.25 model consists of presenting the material In some countries, however, both secondary in a progression from simple to complex, and tertiary institutions have begun to teach pausing to check for student understand- more practical subjects and become more ing, and eliciting active participation from responsive to the labor market. all students. To improve teaching methods Although curricula and teaching meth- or develop a life-skills curriculum, teachers ods have remained largely unchanged, should have adequate materials and train- labor markets are demanding workers who ing, with routine assessments of student have strong thinking and interpersonal progress and of the teaching methods and skills. Job tasks requiring problem-solv- life-skills curriculum.30 Information and ing and communication skills have grown communication technologies can also facil- steadily since the 1970s in the United States itate teaching and learning (chapter 8). while manual and routine cognitive tasks To provide the practical skills demanded have declined. Surveys of employers and in the labor market, new subject areas need workers in several developing countries, to be added to the secondary curriculum such as India and Malaysia, also indicate without overloading it,31 including sci- an increasing demand for communication ence, technology, economics, and foreign skills, which is unmet by current education languages--accompanied by renewed systems.26 Entrepreneurship also requires emphasis on basic mathematics and read- thinking skills to solve problems and such ing. Computer literacy is becoming a baseline behavioral skills as self confidence and lead- requirement for many jobs, and the demand ership.27 So, in today's complex and chang- for high-skilled information communica- ing environment, the challenge is to build tions technology workers has increased.32 skills that allow young people to think criti- The command of international languages, cally and creatively, to process information, particularly English, is becoming an asset for to make decisions, to manage conflict, and adopting technology and for communicating to work in teams. Teaching such life skills can be integrated into every aspect of the curriculum through discovery-oriented teaching methods that include interactive learning, applying B O X 3 . 2 Life skills programs and nonformal knowledge to real-life problems, integrating teaching methods in schools teamwork and peer tutoring into the learn- ing process, and inviting student input into Young people need problem-solving skills, education methods. Some very preliminary the structure and subject matter of lessons. not only to succeed in the labor market but evidence suggests that such programs may also to process information for a healthy affect health knowledge (chapter 5), peer It can be difficult for teachers to use such life, to participate as citizens, and to care educators'future outcomes, and academic methods while ensuring students learn the for their families. In fact, many of what are outcomes. Such programs merit careful core material, and some developing coun- called "life skills" programs combine teach- evaluation, and the U.S. National Institutes tries have found it particularly difficult to ing of behavioral and thinking skills with of Health is doing a clinical trial of an SPW implement this type of reform,28 so teaching practical information about health, citizen- program in Zimbabwe, which will provide ship, or financial literacy. Many governments clear evidence on health outcomes, to be life skills as a separate subject may be bet- have worked with nongovernmental orga- finished in 2007. Some governments (India ter (box 3.2). For example, Japan and South nizations (NGOs) to provide nonformal life and Zambia) have also begun to cooperate Africa have recently included life skills as a skills programs both to students in school with NGOs to train regular teachers in non- and to youth out of school. formal education. Life-skills programs that subject in their secondary school curricula. The programs sometimes use peer emphasize civic and peace education are The teaching of such skills should be educators. For example, Student Partner- operating in several countries, including the accompanied by efforts to reform the tra- ships Worldwide (SPW), working in several Arab Republic of Egypt, Georgia, and Liberia; ditional methods used to teach other sub- countries in Africa and South Asia, trains they seem promising, although they have young people (those just finished with sec- not been formally evaluated (chapter 7). jects. Structured, student-centered teaching ondary school) to teach life skills and health Source: UZ-UCSF (http://www.uz-ucsf.co.zw/ methods, which even inexperienced teach- education in schools using nonformal research/researchprojects/current/rds.html).. ers can use successfully, are effective and 76 WORLD DEVELOPMENT REPORT 2007 with other countries, although basic literacy and sustainable employment thereafter.37 is best acquired in local languages.33 Teach- In Japan, full-time schooling is followed by ing financial literacy narrows the knowl- full-time employment in enterprises closely edge gap between the rich and the poor and connected with the school. (For more on enables young people to make wiser finan- apprenticeships, see chapter 4.) cial choices.34 Effective feedback from the labor market The general and vocational curricula and regular consultations with employers should be more integrated. The skills and alumni are indispensable for adjust- demanded by young people and labor mar- ing curricula to meet changing needs, as in kets go beyond and cut across the traditional Chile, where vocational training institutes division between general and vocational are governed by representatives of employ- curricula. Postponing vocational educa- ers, workers, and the government (see chap- tion until upper secondary and connecting ter 4). Formal university-industry part- it to higher education are two ways to inte- nerships are rare in developing countries, grate vocational and general education. Also though there are interesting experiences. needed is greater blending of content, bring- Universities and research institutes have ing more vocational content into the general contributed much to the growth of the Chi- curriculum (Botswana, Ghana, and Kenya) nese economy. In Beijing, such institutions and more vocationally relevant academic collaborate with local industry through subjects (science, mathematics, language) joint projects and technology transfers and into the vocational curriculum. The cur- establish firms (spin-offs) to commercial- riculum reform in Chile moved vocational ize their inventions. Some of those firms education to high school and increased the (Lenovo, Tongfang) are among the largest academic content of vocational education Chinese high-technology firms.38 to a third of the total teaching time. Special- ized courses were added to the general cur- Increasing the preparedness and motiva- riculum for students in the academic strand tion of teachers. Without teachers who are to choose from, while vocational specialties motivated and well-prepared, reforms to were streamlined,their content and practices improve the quality and relevance of cur- redesigned to give students flexible skills.35 ricula are unlikely to be successful. Teach- ers are often not well-prepared, however-- "Education should be more Strengthening the connection between either in their knowledge of the material or interactive. . . . during the lessons, school and work. Strengthening the con- in their use of effective teaching methods, only the teachers talk, they do not nection is the best way to make school more particularly when it comes to the learning discuss with the students, the young relevant to work and to facilitate the school- needs of youth.39 Teacher absenteeism is sig- people can study, but they are not to-work transition. Career academies in the nificant (24 percent in India), and teacher motivated." United States combine academic and tech- shortages block the expansion of secondary nical curricula around a career theme and education, especially in Africa.40 Teachers Young person, Lima, Peru establish partnerships with local employers also sometimes perpetuate or exacerbate January 2006 to provide work-based learning opportuni- violence and harassment in schools, which ties. The academies reduce dropout rates affects education outcomes (box 3.3). and improve school engagement among Despite the imperatives, few countries have students least likely to do well in a regular effective training, incentive, and account- school environment. They also improve the ability systems. labor market prospects of young people, but To ensure that teachers are well-prepared, sometimes they benefit young men more education and training are important. Sec- than young women.36 The successful Ger- ondary school teachers need to know a large man "dual system" combining part-time amount of material and be able to interact schooling with work has been tried in several successfully with independent-minded stu- countries in Asia, Africa, and Latin America. dents. There is little evidence of the impact Success has been limited, however, because of teacher training on student learning,41 of the inability to create jobs for apprentices but research suggests some lessons. Ideally, Learning for work and life 77 B O X 3 . 3 Violence and harassment in schools The teacher beat me once . . . so I waited for school to increases dropouts. In some countries, students violence and that women who were raped end and when the teacher left the school building I are discriminated against and physically abused "asked for it."About half of teachers surveyed in beat him up worse. I haven't gone back to school since. on the basis of caste (Dalit children in Uttar Kenya say that having sex with a student should Basti boy, Bangladesh46 Pradesh, India), social status, or disabilities. Stu- not result in dismissal or other serious action. Violence and harassment in schools is common dents in Latin American countries, who were Integrating training, incentives, and other in many countries. Despite its negative impact consulted for this report, said that fear of vio- accountability mechanisms is likely to be most on the school outcomes of all students, data lence in schools leads to dropping out. effective. For example, a pilot in Ghana and are not routinely collected.The pervasiveness South Africa, where violence in schools is Malawi is trying to address sexual violence in of sexual abuse, particularly of girls, by teach- widespread, has recently implemented some schools by improving codes of conduct for ers and other students has been documented corrective policies including dismissing teachers teachers; addressing the attitudes of students, in some countries in Sub-Saharan Africa--6 who commit serious offenses, such as having sex community members, and teachers; and devel- percent of 10- to 24-year-olds in Kenya report with students, and prohibiting corporal punish- oping referral and support systems for students having been sexually abused by a teacher. Cor- ment.There is still no system, however, for stu- who lodge complaints. poral punishment of students, particularly boys, dents to file complaints.Implementing such reg- by teachers is common in many countries--25 ulations may not be enough to change pervasive Sources: Dréze and Gazdar (1997); Human Rights percent of 10- to 24-year-olds in Kenya report cultural attitudes.For example, nearly a third of Watch (2001b); Interagency Gender Working Group being physically abused by a teacher, and a young men in Johannesburg schools say that (2005); Lloyd, Mensch, and Clark (2000); Lloyd (2003); study in Egypt demonstrated that teacher abuse forcing sex on someone you know is not sexual World Bank (2005s); and World Bank (2003a). solid preservice should be combined with parents) and design flaws (small incen- regular in-service training that is designed tives, little link to actual performance).44 to improve teaching practice as well as fos- Also, while in some cases program-induced ter the sharing of experience among teach- improvements in student outcomes come ers and allow teachers to provide feedback from better quality teaching, in other cases on the effectiveness of curriculum reform. these improvements are the result of teach- If there are fiscal constraints, research sug- ers misreporting scores, teaching to the test, gests that in-service training is especially or excluding low-achieving students. There effective and is sometimes less costly.42 is some evidence in support of performance Training should also take into account spe- awards based on student progress rather cific needs assessments--and be designed than levels of student performance, and to help foster positive interactions with stu- awards based on school-level average prog- dents, which increases the likelihood that ress--which promote team work--rather students will exert effort and stay in school than progress at the class level.45 To ensure "The teacher pretends to teach, the (discussed below). teaching quality, incentives need to be com- student pretends to learn, and the Incentives,if well designed and well imple- bined with quality training, good working state pretends that it fulfills its role." mented, can motivate teachers and make conditions, some teacher autonomy, oppor- them accountable for performance. Chile tunities for professional development, and Young person, Recife, Brazil more than doubled average teacher salaries in school autonomy in hiring teachers. January 2006 the 1990s, and the quality of students enter- ing teaching programs increased, suggesting Increasing system accountability for that the level of teacher salaries matters.43 performance. The World Development Performance-based incentives are, in princi- Report 2004 shows that additional public ple, superior to across-the-board increases in spending on education will not improve teacher salaries--on both fiscal and efficiency learning unless motivated providers can grounds. But there has been little experience take the required actions (see box 2.2).47 So with them--few teachers are accountable for frontline providers--school principals and what they do in the classroom. teachers--should be given enough auton- In practice, the impact of incentive pro- omy and resources to manage for results grams in countries as diverse as Chile, India, and be made accountable for those results. and Mexico has been reduced by imple- Two building blocks of accountability are mentation constraints (resistance from providing information on performance to teachers, indifference of headmasters and students, their families, and other stake- 78 WORLD DEVELOPMENT REPORT 2007 prehensive strategy that improves account- B O X 3 . 4 Georgia: Fighting corruption in ability systems but also addresses a number higher education of other constraints, including the institu- tions that underpin the delivery of services. For years, university entrance examinations testing center was equipped with surveil- The reform measures introduced in Georgia were an opportunity for bribery. Corrupt lance cameras and TV monitors that let to fight corruption in higher education illus- officials favored candidates who either had students'relatives observe the examination personal connections or had bribed the process. trate the effectiveness of combining a unified members of the examination committee. "Young people are no longer afraid, examination system, control mechanisms, Well-off students who paid the right profes- when they take university exams, that other and improved transparency (box 3.4). sors could get advance tips on examina- people will pay money and get in instead of tion topics.This meant students were not them,"says Maka, a third-year law student in selected on the basis of merit. Tbilisi. Faculty have also reported significant Financing the expansion In 2005 a new law on higher education improvements in the competence and com- made the unified national examinations mitment of students, allowing students to and improvement of mandatory for all potential students seek- finish their education sooner. postprimary education ing to enroll in a higher education institu- The new exam is part of the broad slate tion in Georgia. A high level of security of reforms to increase the quality of educa- In selecting the right set of education surrounded the testing process in July 2005. tion, including a new national curriculum, reforms, countries must consider the state Candidates were identified by a barcode on building up teacher training capacity, and of their education systems (how well they each exam paper to ensure confidential- introducing per capita financing to schools. ity in grading. Examination booklets were prepare youth for work and life), the needs Source: National Assessment and Examination printed in a secure facility overseas. Each Center (2005). of their young people, and their overall development priorities. This choice is con- strained by a country's available resources and its ability to use innovative ways to holders; and engaging their participation finance large-scale reforms. Some countries in school management. Youth participation will choose reforms that require relatively in school management, covered later in this few resources and still have a fairly large chapter, has been very limited. impact. Others will take on comprehen- Although many developing countries sive reforms that require a large amount of now collect national data on student achieve- additional resources. ment, the information is rarely made pub- Financing the types of reforms recom- lic and used to hold teachers and schools mended in this chapter is possible, even for accountable for performance (through social poor countries, but choosing the right type accountability, accreditation, funding alloca- of reform is critical. Investments in qual- tion, or performance-based pay). A large civil ity and relevance can lead to substantial society initiative in India, the Annual Survey improvements in enrollment and learning. of Education Report (ASER), collects regular One-time investments can have long-term data on literacy and numeracy skills of school impacts. For example, a recent project tar- children in rural India. The program also has geting poor lower secondary schools in a strategy for dissemination and community Guyana included curriculum reform, stu- mobilization, which includes the preparation dent assessment, and training and materi- of annual regional reports and brief sum- als for teachers and principals, which only maries at the district level. The 2005 survey required a one-time 6 percent increase in includes key findings for policy:48 31 percent school expenditures. It was associated with of 11- to 14-year-olds need remedial language large increases in test scores and in the num- instruction (they cannot read a story text ber of students completing lower secondary with some long sentences), 47 percent need school.49 The success led the government remedial numeracy instruction (they can- to adopt the changes in all lower secondary not do division), 23 percent of teachers were schools and to reform the upper secondary absent (on average, per school), and 8 percent curriculum. of schools had no teacher present at all. While some fast-growing developing Improving governance in education countries can rely on economic growth to requires more than just strengthening finance improvements in the education accountability for performance. For exam- system, most other countries will need ple, curtailing corruption requires a com- additional resources from a combination Learning for work and life 79 B O X 3 . 5 Chile's higher education: Diversifying the sources of funding Chile financed a big expansion in higher edu- the research and teaching efficiency of the insti- undergraduate degree structure, and increasing cation in recent years by charging students tution they attend.The system thus gives insti- the production of PhDs. tuition, encouraging diversified funding sources, tutions an incentive to improve their efficiency A remaining flaw in the financing system and allocating of public subsidies innovatively. while enhancing access for less-privileged is the lack of accountability for results.The Thanks to sizable private contributions, it groups. system allocates recurrent funds per student to increased tertiary enrollment to about 42 per- Public support incentives are designed traditional universities based on historical lev- cent of the 18­24 age group while being among to encourage universities to be responsive to els--but in 2006 pilot performance agreements the Latin American countries that allocate the both students needs and national priorities. are being negotiated between each institution least public funding to tertiary education rela- About 7 percent of public support is allocated and the ministry of education.The objectives tive to GDP. based on the ability of institutions to attract the are to link campus missions to national and Significant student contributions have students who receive the highest scores in the regional priorities, university autonomy to pub- provided the fiscal space to reorient state sub- university admission exam. Universities receive lic accountability, and institutional performance sidies to core public sector responsibilities. A funds by proposing projects to a competitive to government funding.The agreements, to run priority has been to increase access to income- investment fund for improving the quality of for three years, will contain funding commit- contingent student loans for students unable technical, undergraduate, and graduate tertiary ments, agreed targets, and indicators to monitor to finance their studies. Financial support is education.This approach has generated reforms progress. determined by the student's socioeconomic closely linked to national priorities: improving Sources: Bernasconi and Rojas (2004) and Thorn, profile, while tuition fees are set according to the quality of teacher education, reforming the Holm-Nielsen, and Jeppesen (2004). of cost-sharing, public-private partnerships, cation policy, strong demand for quality, and and efficiency enhancements--and some partnership with the private sector allowed will need help from donors. Chile's higher the secondary education sector to expand education system shows how blending fund- without compromising quality (box 3.6). ing sources can expand and improve tertiary Burkina Faso successfully financed improve- education (box 3.5). Korea's responsive edu- ments in access to, and the relevance of, B O X 3 . 6 Korea's secondary education: Expansion without sacrificing quality Korea's secondary schools do well on many and 1960s, when public funds mainly targeted and with declining quality;students preparing for fronts.Access is easy and equitable: gross enroll- primary education, secondary schools financed entrance exams relied on private tutoring even ment at the secondary level is at 90 percent, for almost half their expenses through parent more.In response,the government adopted an both boys and girls.Korean students score at the teacher associations. However, the rapid expan- equalization program for upper secondary in top in such international evaluations as PISA and sion of primary education put enormous pres- 1974,opening high school entry.The program Trends in International Mathematics and Science sure on secondary schools, and student compe- also aimed to narrow the quality gap between Study (TIMSS).Schools are adequately funded-- tition for good secondary schools increased. In urban and rural high schools and increase enroll- more than 2.4 percent of Korea's GDP is spent on preparation for entrance exams--"examination ment in vocational schools to meet demand from secondary education, a third of that privately. hell"--students often repeated grades, and the fast-growing manufacturing sector.Without Where did such achievements come families paid up to a quarter of their income for any curriculum change,however,vocational from? First, building a strong education sector private tutoring. education remained largely a terminal track,and was part of Korea's economic development In the face of criticism, the government enrollment did not change much. strategies as early as the 1950s. Dynamic and implemented a national equalization program Although the combination of private fund- motivated institutions promptly implemented in 1968, eliminating entrance examinations ing and public control worked well in the earlier policies to expand education. Second, from the and instituting a lottery for schools in high years of expansion, concerns about quality grew. earliest days the focus was on access and quality demand. Secondary school enrollment soared, In response, the government carried out a series for all, motivated by the desire to bring edu- and private providers stepped up to provide of reforms in secondary education starting in cated workers into the workforce.Third, parents the needed capacity.The equalization program 1999 and increased public funding to the sector contributed to the expansion's costs because of guaranteed any deficit in operating cost (but by 7 percent annually until 2003. It relaxed con- the high value they placed on quality education. not in capital cost) of all private schools. By trols over school management, instituted school In Korea, compulsory basic education 1971, most private schools were receiving direct councils to facilitate parental involvement, and includes primary and lower secondary. Primary financial assistance, subsidies, and tax exemp- legalized teachers unions.The curriculum reform education is free, but parents must pay tuition tions. In return, they gave up control over key introduced foreign languages and information for secondary schools. One in five lower second- decisions (curriculum, tuition rates, and teacher technology at earlier ages and emphasized stu- ary students and more than half of high-school salaries). dent-centered learning.There are also efforts to students attend private schools. Forty percent The equalization program for lower second- improve vocational curricula and link vocational of high school students are enrolled in technical ary, while improving enrollment, removed the high schools with technical colleges. and vocational schools. competition among the elite schools, and qual- Korea focused on one education cycle at a ity at the top declined. Examination hell resur- time, starting with basic education. In the 1950s faced at the end of the lower secondary cycle, Sources: Gill and Chon-Sun (2000) and Kim (2002). 80 WORLD DEVELOPMENT REPORT 2007 postprimary education by reducing subsidies Public-private partnerships. Public-pri- and charging fees for tertiary education. vate partnerships can expand and improve postprimary education. In addition to Improving efficiency. Efficiency gains can alleviating fiscal constraints, they improve be achieved by using formula funding to learning outcomes and efficiency by education institutions, particularly when increasing choice and competition. Analy- they have greater autonomy. More develop- sis of PISA data shows that private compe- ing countries are moving from traditional tition is associated with higher test scores, line-item budgeting to direct formula fund- and systems combining private operation ing, which channels funds to schools for with public funding do best.54 Partnerships operating expenditures according to some vary depending on the services procured known rule, such as enrollment.50 With and, though still uncommon in developing block grants, as opposed to earmarked countries, a few lessons are emerging.55 grants, schools have discretion over the Contracting with schools to enroll pub- use of funds (as with autonomous schools licly funded students (used extensively) in Nicaragua). School autonomy over per- has rapidly expanded access to education sonnel management and process decisions while avoiding large public capital costs. (hiring teachers, choosing textbooks, allo- Voucher-type programs have been imple- cating budgets within schools) is related to mented in a few developing countries, superior student performance.51 including Chile, Colombia, Côte d'Ivoire, Formula funding, particularly block and the Czech Republic. Their positive grants, can bring several efficiency bene- impact for beneficiaries has been estab- fits--increasing transparency and account- lished (Colombia)56 but the overall effects ability, reducing corruption, making fund- are still inconclusive. Vouchers are also ing more predictable (which allows better found to improve performance in public planning), and increasing flexibility (for school through increased competition (the block grants).52 It can also be combined Czech Republic).57 with other efficiency-enhancing mecha- Contracting a private actor to operate nisms, such as competitive funding and per- a public school can a have positive impact formance-based funding--but it requires on enrollment. Colombia's Colegios en good management capacity. Concesión turns over the management of some public schools to private institutions Cost sharing. Appropriate cost sharing through a competitive bidding process. and demand-side financing can generate Concession schools are paid less per student the needed resources equitably. Govern- than regular public schools, must accept ments should finance more of the compul- all students, and must meet outcome tar- sory phase of secondary education, because gets for test scores and dropout rates. They of higher social benefits and lower unit costs are carefully monitored and evaluated-- relative to later education. Individuals, their dropout rates have been lower in conces- families, and communities should finance sion schools, and the competition has also moreof postcompulsoryeducation,particu- reduced dropout rates in nearby public larly tertiary education. Contributions from schools.58 Contracting for support services those able and willing to pay can promote (meal provision, facility maintenance) is engagement and accountability.53 To ensure used extensively, usually with good results. equitable access, fees and other cost-sharing Contracting for professional services (such mechanisms need to be accompanied by as curriculum design) is also easy to specify well-designed and balanced demand-side and monitor. Contracting for management financing packages (covered later in this services is difficult to implement, not least chapter). Such packages include needs- because of the challenge of identifying mea- based grants at the lower secondary level, surable and verifiable performance criteria. as well as grants that are needs- and merit- A few countries are experimenting with based, loans, and savings schemes at the contracts for private financing and con- upper secondary and tertiary levels. struction of schools. Learning for work and life 81 Better education choices dent on guardians, policies should also take by young people into account the preferences and parenting skills of guardians. As young people mature, they take more Parents affect student behavior, beyond control of their education, but some con- simply influencing the decision to attend straints prevent them from benefiting from school, by affecting a young person's prefer- their learning opportunities. They lack the ence for education through the home envi- motivation to learn because they have poor ronment, and by helping to develop their behavioral skills and not much of a say in children's behavioral skills. In India, part of their own education. They lack information the relationship between maternal literacy on postprimary education opportunities and child schooling reflects the effects of and the labor market--and have limited maternal schooling on the study hours of access to resources. They face competing children.59 In Rwanda, a guardian's prefer- options to schooling--work and family--as ence for education explains part of the cor- well as alternative learning options. Help- relation between the education of guardians ing young people make better education and children. The strongest evidence comes choices requires better decision-making from the positive impacts on education of skills, incentives to exert effort, and their programs that emphasize parenting quality involvement as stakeholders in their educa- to build behavioral skills of children, either tion. Financial incentives that permit better directly or through mentors.60 choices (such as conditional cash transfers or vouchers) and better information about Improving the incentives to learn. Social learning options and work possibilities mechanisms (such as peer influence and (such as school-based career advisory ser- the social accountability of teachers) and vices) lead to better education decisions. direct economic incentives can influence the effort to learn. Some evidence suggests Motivating and involving students that the efforts of students, teachers, and As young people reach adolescence, they start parents are jointly determined61; thus, to be deciding how much effort to spend on study- successful, interventions will often need to ing and whether to go to school (see figure take into account the role of peers, teach- 2.4). Their decisions indicate their prefer- ers, and parents and their preferences and ences for education--preferences influenced incentives. It is the independent influence by peers, parents, teachers, and schools that of peers on students, however, that has been shape their environment. Their ability to the most rigorously evaluated, and research successfully act on those preferences is deter- shows that even randomly selected peer mined by their behavioral skills and psy- groups (such as classmates or roommates) chological well-being. Many young people affect the behavior of students and educa- gradually develop control over their educa- tion outcomes.62 tion decisions, while others--young women, Influencing peer interactions by sorting orphans, those with disabilities, and those students into different groups can have an from stigmatized groups--face additional important impact on both academic and barriers to effective decision making. social outcomes.63 In fact, many countries Well-designed policies can affect both sort students into schools based on their the preferences for education and the incen- ability, and in countries where there has tives to learn. For example, outcomes can been ethnic conflict, students are sometimes be improved by affecting how peers influ- sorted into classes based on ethnicity. There ence each other, by increasing young peo- is some evidence indicating that diversity ple's connectedness to schools, by directly of student achievement or social groups rewarding effort, by developing their deci- within a school or classroom might nega- sion-making capacity, by responding to tively affect student achievement in second- their input in the school-level policy-mak- ary school. A study in China demonstrates ing process, and by reducing social exclu- that the diversity of academic achievement sion. For younger adolescents more depen- has a negative effect on the test scores of 82 WORLD DEVELOPMENT REPORT 2007 poorly performing secondary students.64 of education that are difficult to measure Other evidence suggests, however, that poli- (such as how teachers interact with stu- cies to sort students in classes by ability are dents). Data from more than 40 countries usually ineffective, because ability varies by show a strong association between school specific task and students progress at dif- climate and test scores,70 and evidence from ferent rates.65 Some types of social interac- the United States and the Caribbean shows tions between young people from different that school connectedness predicts school groups can also have positive effects on out- dropout and health outcomes.71 comes such as trust and tolerance.66 Can such factors be affected by policy? It is likely that what matters is not the Yes.Reforms specifically designed to increase level of diversity, but the classroom con- students' connectedness with schools and text-- whether it is competitive or coopera- teachers, such as the Seattle Social Develop- tive. Controlled experiments indicate that ment Project in the United States, improved classroom activities that require coopera- academic achievement and other student tion among students from different ethnic behaviors.72 (Poor schools in many coun- groups can improve tolerance (cross-group tries have high connectedness through sim- friendships) and empathy across groups. ple efforts to engage students.) Often such Such cooperative policies in socially and reforms are comprehensive and rely on a academically diverse classrooms also combination of policies similar to those rec- improve test scores, as well as behavioral ommended throughout this chapter: skills such as self-confidence.67 · Training teachers to enhance students' "Studying is only for the people of A merit scholarship program for girls in identification with school through stu- higher caste, not for ones of lower Kenya shows how economic incentives that dent encouragement, student input into caste like me." take social mechanisms into account can influence effort.68 The test scores of the girls lessons, group learning, and group aca- Young person, Nepal demic competitions. who had a good chance of getting a schol- January 2006 arship improved, which can be explained · Increasing youth participation in school by the economic incentive of the scholar- policy. ship. The test scores of boys and lower-abil- · Teachingbehavioralskills,whichcanincrease ity girls also improved as did student and student confidence and motivation. teacher attendance, perhaps as the result of · Teaching through drama, sports, and peer influence or the complementarity of arts. student and teacher effort in the classroom. Teacher attendance may have increased as Decision-making capability--building behav- a result of providing the correct economic ioral skills and reducing social barriers.Even incentives for parents to hold teachers if students value education and prefer to socially accountable. The program also can study, they may not have the capability to Figure 3.4 Influencing effort directly be more cost-effective than some typical act on their preferences, perhaps because can be a cost-effective way to improve education programs (figure 3.4). they lack behavioral skills or psychological learning in Kenya Increasing students' engagement with capacity. Yet motivation, persistence, self- Cost per pupil of 0.1 standard deviation schools is one way to raise student effort in discipline, cooperation, and effective deci- gain in test scores ($) 6 school and reduce the likelihood of drop- sion making are rarely taught in schools. $4.77 $5.61 ping out. If students identify with their This is a major gap, because the impact of 4 schools, the importance of other social behavioral skills on education outcomes $2.48 identities can be reduced allowing schools and the ability of schools to teach those 2 to shape student preferences for the effort skills have been empirically demonstrated 0 to study.69 This effect may be best quanti- (see box 3.1). Girls' Teacher Textbook fied by research on school climate and A student's ability to make decisions can scholarship incentives provision program school connectedness, often determined by also be affected by his or her mental health. (Busia district) asking students if they feel like they belong Research from developed countries indi- Source: Kremer, Miguel, and Thornton (2004). in school, feel connected to it, or like it. The cates the negative effect of mental health Note: The scholarship program had no effect in answer to this question matters, and ask- disorders on school attainment.73 While in neighboring Teso district, where it was associ- ated with a natural calamity. ing it can capture several important aspects general mental health has been less stud- Learning for work and life 83 ied in developing countries, posttraumatic Young people as stakeholders in school stress disorder is known to affect as much policy. The World Development Report as a third of the youth population in post- 2004 recognized "client power" in improv- conflict countries, and a study from Algeria, ing the delivery of basic public services Cambodia, Ethiopia, and Gaza demon- such as primary education, with parents strated its effect on schooling.74 Thus, safe- and communities viewed as the clients of guarding mental health can be an important the education system. Young people are the investment in school outcomes.75 beneficiaries of the education system and, Negative social perceptions can affect starting in adolescence, they often begin students' decision-making capabilities. to make important decisions about their Social and cultural norms in many coun- education, and so become actors in their tries exclude some groups from educa- own right (chapter 2). Also, secondary and tion, primarily girls and young women,76 tertiary institutions usually serve a much although ethnicity, caste, and disability broader area than primary schools, making are also dimensions of exclusion. Exten- traditional community management less sive ethnographic research has long found effective as an accountability tool. pervasive attempts by teachers and admin- Students can exercise their client power istrators to discourage students from con- through many institutional mechanisms. tinuing in school if they belong to tradi- They can set up students councils, be elected tionally excluded social groups (including to school boards, or be consulted more the poor).77 This can affect education broadly on the design of rules, policies, decisions by lowering young people's self- and curricula at the school or classroom perception or by leading them to believe level. Students can also provide feedback (sometimes accurately) that schools and about teachers. In most countries, however, labor markets will not reward their effort. student participation is limited or nonex- Evidence from Kenya demonstrates that istent. Nor are there many evaluations of teacher attitudes about adolescent girls' student participation. The studies typically ability to learn predicts their dropout deci- focus on the personal outcomes of partici- sions.78 In Uttar Pradesh, India, an experi- pants rather than on the impact of student- ment indicates that students from the low- influenced policies on the quality of school est castes perform worse when a student's life or academic outcomes.80 Student partici- caste is announced before taking a cogni- pation can also have an impact on behavioral tive skills test--indicating that low-caste skills (as students become decision makers) students believe that teachers who care and other forms of civic engagement. about caste will not reward them for high Student participation improves school performance.79 environment and relations. Many educa- Programs may aim to increase the ben- tion providers are reluctant to allow student efits to young women or their families of participation, but one careful analysis of 75 overcoming social barriers (through finan- studies (mostly in England) found that even cial incentives such as targeted merit schol- in secondary schools where school manage- arships and conditional cash transfers) or ment and parents were initially reluctant they may aim to reduce the effect of social to institute reforms, no school reversed the barriers on decisions by trying to raise reforms, and both teachers and students self-perception (through, for example, set- reported better interactions and improved ting up bank accounts for girls, as in Ban- student behavioral skills as a result of more gladesh). Programs may also be designed student participation.81 A major challenge to reduce the social barriers themselves is that some education providers are willing (information campaigns, training of teach- only to allow token or symbolic participa- ers). In some environments, especially tion, which prevents students from having a where physical violence is a concern, pro- substantial impact on policy. grams should be careful to target not only Some developing countries have sig- the excluded, but groups who perpetuate nificant student participation, at least in social exclusion. tertiary institutions. South Africa's Student 84 WORLD DEVELOPMENT REPORT 2007 Leadership Council defended student rights Young people partly base schooling during apartheid and provided training for decisions on expected economic returns, future political leaders (including Nelson particularly when it comes to the decision Mandela).82 Recent reforms of universities to attend university. Results from a study in Russia and secondary schools in the for- of Indonesia show that students choose to mer Yugoslav Republic of Macedonia have enter upper secondary school on the basis increased the role of students in school of expected returns.85 However, uncer- management, giving them as much as 30 tainty about future labor market returns percent of the voting rights on some uni- prevents some young people from mak- versity councils. In a survey of the alumni ing the education decision that is best for of one university in Russia, most listed them, generating inefficiency. If people such participation as the single most useful are risk averse, this uncertainty also leads aspect of their education.83 to underinvestment in education. The study from Indonesia also finds substan- Providing information on tial uncertainty about future returns to opportunities for learning and work upper secondary and higher education. Postprimary education decisions are based Those from poor families face more uncer- on private information on the supply and tainty than those from richer households. quality of education opportunities available Overall, 11 percent of young people would to young individuals, and an understanding change their education choices under full of how they translate into education and certainty--13 percent of the poor and 10 labor market outcomes. Imperfect infor- percent of the nonpoor. Because of this mation about education makes schools and uncertainty, the poor tend to underinvest teachers less accountable and young indi- in education more than the rich. viduals less able to make good choices and Providing young people with infor- influence school quality, while the uncer- mation on labor market opportunities tainty young people face about the future and payoffs to different levels and types returns to education leads to inefficiencies of schooling can allow them to make and underinvestments. more educated guesses about their future Good information on learning opportuni- returns, which makes their decisions more ties leads to better choices. It allows individu- efficient. An experiment in the Dominican als to monitor the performance of schools Republic increased the likelihood of con- and teachers and to influence school quality. tinuing in school by 12 percent, by provid- Information on learning options can also ing information to students enrolled in the have a real impact on choices, particularly for last year of primary education about the the poor. Yet there are very few information returns to secondary education.86 That the programs around the world. The U.K. Aim- poor are not more responsive to informa- higher program targets disadvantaged 13- to tion about future returns indicates that 19-year-olds,raising their awareness of higher direct and indirect pecuniary costs and education opportunities well in advance. It nonpecuniary costs such as low behavioral posts information on higher education on a skills (low aspirations, for example) do web portal, and it provides mentoring and much to limit access to postcompulsory visits to institutions of higher learning. An education in Indonesia.87 Thus, the poli- initial evaluation shows positive effects on cies needed for poor people to go to upper promotion rates and test performance--and secondary would include some combina- mixed results on intentions to go on for tion of grants, income-contingent loans, higher education.84 School-based career guid- mentoring, and information.88 ance services--which provide information on education and job market opportunities Providing financial incentives to as well as counseling--are being introduced alleviate constraints to better choices in a few developing countries (discussed later Resources matter for schooling decisions in the chapter in box 3.7). when young individuals and their fami- Learning for work and life 85 lies face imperfect credit markets that pre- look at the relationship between income vent them from borrowing against future and schooling,90 but few look explicitly at income. Young people may also be finan- how credit constraints affect the demand cially constrained when their parents have for education. Such constraints determine sufficient resources or access to credit but college attendance in Mexico, though the are unwilling to finance investments in effect is small, and in rural Pakistan they education. Also, young people in postpri- explain secondary school completion and mary education face competing options to postsecondary school attendance more than schooling--work and family--as well as secondary school entry.91 alternative learning options. Competing options and schooling. In Credit constraints in schooling. low- and lower-middle-income countries, In most developing countries there is a many of those of secondary school age work strong relationship between poverty and (figure 3.6). With age, school participa- school progress, particularly in the tran- tion diminishes while work participation sitions to lower and upper secondary increases. In poor countries such as Cambo- school (figure 3.5). Is that link between dia, the trade-off between school and work income and schooling explained by credit becomes more of a problem at the age of constraints? Not necessarily, or at least secondary school entry--43 percent of boys not entirely. Other factors correlated are already working by ages 10­12. Condi- with income also explain the differences tions have been improving there, however, in school progress between the poor and as young people stay longer in school and the rich, including inequality in access to start work later. Most working children can physical facilities. combine work responsibilities with school- The school attainment differences ing, though with age, working students between the poor and the rich are also become a smaller group relative to full- explained by the difference in skill readiness time workers--by ages 13­15, 62 percent for secondary and higher education. The of Bolivian working males are already out rich tend to attend schools of higher qual- of school.92 Young people from the poorest ity than the poor and have a better environ- quintile are less likely to attend school exclu- ment for skills formation. The quality of sively, and more likely to work, than those schooling affects the motivation for con- from the richest. In some countries, girls are tinuing in school and the academic readi- less likely to be working outside the home, ness for subsequent education. A large part of the association between income and col- Figure 3.5 More rich children than poor children completing primary school make it to higher lege attendance in the United States is due levels to long-term factors that affect readiness for Percentage of primary school graduates who reach each level college by increasing cognitive and behav- 100 ioral skills, not to short-term financial con- 90 straints at the time of the college decision.89 80 The policy implication is to shift the focus 70 from tuition subsidies to getting individuals 60 50 ready for college through learning invest- 40 ments before college. 30 However, more young people in devel- 20 oping countries are likely to face credit 10 0 constraints in gaining access to secondary Poorest Richest Poorest Richest Poorest Richest Poorest Richest and higher education than in developed quintile quintile quintile quintile quintile quintile quintile quintile Colombia Indonesia Thailand Zambia countries. Credit markets are less developed (2000) (2002) (2002) (2001/02) and direct financial support for schooling is First year lower secondary First year upper secondary First year tertiary often much more limited,justifying the need for policies to address them. Many studies Source: http://econ.worldbank.org/projects/edattain. 86 WORLD DEVELOPMENT REPORT 2007 Figure 3.6 Adolescents in many low-income and lower-middle-income countries combine school and work Richest quintile Poorest quintile Percent Percent 100 100 80 80 60 60 40 40 20 20 0 0 Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages Ages 10­12 13­15 16­18 10­12 13­15 16­18 10­12 13­15 16­18 10­12 13­15 16­18 10­12 13­15 16­18 10­12 13­15 16­18 Cambodia (2004) Brazil (2001) Bolivia (2002) Cambodia (2004) Brazil (2001) Bolivia (2002) School only School and work Work only Neither in school nor working Source: Authors' calculations based on Fares, Montenegro, and Orazem (2006a). but they are more likely to be engaged in Conditional cash transfers. Addressing domestic work (for the countries in figure credit constraints requires identifying the 3.6 the distribution of school and work is target population, and then designing similar for boys and girls).93 the appropriate package of demand-side Working can impair schooling and learn- financing mechanisms for postprimary ing, particularly at younger ages. A study education that also account for the oppor- "They [young people] don't go to conducted in Vietnam finds that if children tunity costs from competing choices. Credit school because they have a family to ages 8­13 work while in school, it reduces constraints on the demand for education at take care of, they cannot continue school enrollment and educational attain- the lower secondary level can be addressed to rely on the good will of their own ment five years later.94 In rural Bangladesh, through needs-based grants (scholarships, parents." working while attending primary school has conditional cash transfers, vouchers), given a sizable negative effect on the transition to the externalities associated with basic edu- Young person, Peru secondary school--and starting to work cation. Even when lower secondary educa- January 2006 while attending secondary school has even tion is free of school fees, students and their larger negative effects on secondary school families must incur other direct and indi- completion.95 A study of 11 countries in rect private costs, making a case for grants Latin America and the Caribbean finds siz- targeted to the poor. At the upper secondary able negative effects on both mathematics and tertiary levels, credit constraints can be and language test scores among primary alleviated with a combination of well-tar- school students.96 (For more on child labor, geted grants, loans, and savings schemes. see chapter 4.) In designing such a package, it is also Girls who marry early or bear children at important to think of the education system an early age are also more likely to have less as a whole--to avoid imbalances and bot- schooling.97 The secondary school dropouts tlenecks. For example, the success of Opor- attributed to pregnancy range between 10 tunidades in Mexico and Bolsa Familia in and 20 percent in most countries in Africa.98 Brazil increases the pressure on the post- The relationship also works the other way: compulsory system. It is thus necessary to in Guatemala, school attainment increases anticipate the rising number of individuals the age at which young women first marry completing basic education and willing to or become a parent.99 study further. Learning for work and life 87 Conditional cash transfers can increase marriage until age 18. The stipend has been the demand for schooling both directly associated with postponing marriage and (through increased income) and by reduc- increasing schooling.107 ing the incidence of work (by compensating for the forgone income from work). Intro- School vouchers. Apromisingtooltoaddress duced in the late 1990s, particularly in Latin credit constraints and school quality, school America, they provide cash to poor young vouchers are publicly provided for students people conditional on school attendance, to enroll in the school of their choice. Beyond and are quickly becoming popular in other the potential effect of vouchers on their ben- parts of the world.100 Mexico's Oportuni- eficiaries, vouchers can increase competition dades, the best documented, has increased among schools and thus increase the quality secondary school attendance rates by 8 per- of the system. They also allow enrollment cent, the transition to secondary school by to increase without additional public capital nearly 20 percent, and grade attainment by costs. However, they can have a detrimen- 10 percent, with significantly larger effects tal effect on the (lower quality) schools that for girls than for boys.101 The impact on (higher performing) voucher recipients leave, enrollment is mainly due to the condition at least in the short run. for attendance.102 Oportunidades is not A few developing countries have used inexpensive, but the net benefit is substan- vouchers, including Chile, Colombia, and tial.103 Efficiency can be increased by target- the Czech Republic. Colombia's voucher ing and calibrating the size of the grants.104 program offered vouchers to poor indi- Part of the positive impact of conditional viduals to attend private schools and had cash transfers on school enrollment comes a positive impact on learning, which per- from a reduction in work. But the transfers sisted over the long run.108 More evidence alone do not appear to be enough to reduce is needed, though, on the overall impact of work significantly, which is important for vouchers, including the impact on nonben- policy makers concerned about child labor eficiaries. Vouchers for secondary or even and how child labor affects academic per- tertiary education should be targeted to the formance. A substantial number of chil- poor (with possible refinement for merit at dren continue to combine both work and the upper levels). Choice should be open to school under Oportunidades. The pro- any accredited school (public or private), gram increases the number of children in supported by publicly provided informa- school and reduces the number of children tion on these schools. who are working, but does not necessarily reduce the hours worked of children who Loans. Well-designed loan programs, also attend school. Also, Oportunidades combined with grants targeted to need increases the likelihood that young people and merit, can allow postcompulsory edu- will stay in school when a household experi- cation to expand equitably. Grants, when ences a shock, but does not prevent parents affordable, are likely to be more appropri- from resorting to child work in response to ate for upper secondary education than a shock.105 Evidence from rural Brazil sug- for higher education--the ratio of social gests that after-school programs may be a to private benefits is likely to be higher good complement to the conditionality on in upper secondary, and grants in upper school attendance.106 secondary are likely to be more progres- Conditional cash transfers and other sive. To improve cost-effectiveness, they policies that reduce the price of schooling should be specifically targeted to the poor (fee waivers) have been used to weaken the who are most likely to benefit from the attraction of activities that compete with grant on the basis of merit.109 The higher girls' schooling. Under the secondary school cost of tertiary education makes stu- stipend for girls in Bangladesh (see box dent loans particularly useful, but well- 6.5), girls received a stipend conditional on designed loans are hard to implement in secondary school attendance and delaying low-income countries. 88 WORLD DEVELOPMENT REPORT 2007 Income-contingent loans are superior to ested in pursuing further education. The conventional loans (box 3.7). They gener- amount an individual is entitled to depends ate the needed resources in the face of lim- on the amount saved and the kind of train- ited fiscal capacity. By deferring payments ing desired. In Brazil, a graduation incentive until individuals start working and reach for primary and secondary education, Pou- a certain income, they have lower default pança Escola, was introduced as part of the rates, promote more equitable access and first version of Bolsa Escola in the Federal loan repayment, and increase efficiency by District.110 Oportunidades in Mexico intro- addressing uncertainty about future earn- duced Jovenes con Oportunidades, through ings and facilitating consumption smooth- which conditional cash transfer beneficia- ing. However, they are hard to implement ries accumulate points from the last year of and may be a realistic option only for some lower secondary until the end of secondary middle-income countries. school. Credit points are converted into a savings account and deposited into individ- Individual learning accounts. The easy ual accounts in the National Savings Bank, implementation and attractive features-- which beneficiaries can tap for further study induced savings, consumption smoothing, or to start a business if they complete upper and a low public burden--of individual secondary before turning 22. learning accounts make up a promising Young people lack adequate information, financing option for middle-income coun- financial resources, academic readiness, and tries. Individual learning accounts, which decision-making skills.Addressing these com- are becoming more popular in OECD coun- plementary constraints requires policies that tries, encourage savings for education while integrate information, mentoring, academic providing vouchers to individuals inter- support, and financial incentives (box 3.8). B O X 3 . 7 Income-contingent loans A well-designed loan system has three charac- subsides are also deeply regressive because · a reliable system of identifying individuals, teristics--income-contingent repayments, an graduates, rather than students, make repay- which is the responsibility of national government; efficient interest rate, and adequate size. ments.Targeted interest subsidies may still be an · the capacity to keep records (amount borrowed), Income-contingent repayments. A con- option, however, for people with low earnings or a responsibility of the loans administration; ventional loan involves repayments of a fixed who are out of the labor force.In some loan pro- · the capacity to collect repayments,ideally amount per month, with the risk that low- grams, unpaid debt is eventually forgiven. through the tax or the social security system; income borrowers may default, while income- Adequate size. Loans should be large enough and contingent repayments are a percentage of the to make education and training feasible, but · the capacity to track income, ideally through borrower's earnings--often collected alongside small enough so that most individuals can income taxes or social security contributions. income tax or social security contributions, repay, with a cap both in annual amount and and thus have built-in insurance against an duration. Income-contingent loans have a long Given the requirements,it is not surpris- inability to repay because the loan repayment repayment period and small monthly payments ing that successful income-contingent loans falls if a borrower's income falls.They thus assist so they are friendlier to other borrowing (say, in advanced economies--including Australia, borrowers by protecting them from uncertain to start a business), because lenders typically New Zealand,the Netherlands,Sweden,and the returns to human capital investments and pro- look at monthly income net of loan repayments. United Kingdom--are not echoed in poorer mote equity because this uncertainty is more The purpose of a loan is to allow young people countries.Chile and South Africa have such of a deterrent for the poor.They assist lenders to redistribute resources to themselves over schemes on a small scale,with repayments col- by using the government's power to collect their lifetimes, so loans should in principle be lected by universities,a method that has proven taxes as a substitute for physical collateral. available to all qualified applicants.Without a unsatisfactory.Both schemes have had some suc- After Australia introduced income-contingent government guarantee, private sector lenders cess,but would be fiscally costly on a larger scale. loans in 1989, participation in higher education would charge a high risk premium, so if govern- Thailand is planning to introduce an income- increased, particularly among women. ments face serious cash-flow constraints it may contingent loan scheme in 2006,the success of Efficient interest rate. The interest rate should be necessary to restrict loans to the poor. which will depend greatly on the effectiveness be broadly equal to the government's cost of Beware the implementation requirements. of income tax collection.Designing a cost-effec- borrowing.Some countries, such as Australia and Policy makers invariably underestimate the insti- tive repayment mechanism in poorer countries the United Kingdom, offer loans at a zero real tutional requirements.A particular error is to focus should be at the top of the policy-maker agenda. interest rate, but because of the resulting fiscal on lending policy, with inadequate attention,time, pressures loans are too small, access is limited, or resources for loan administration. A country and university income is reduced.Interest rate should not embark on a loan scheme without Sources: Barr (2004) and Chapman (forthcoming). Learning for work and life 89 Offering second chances Integrated approaches address the many Poverty, economic shocks, and bad schools B O X 3 . 8 force many young people to leave school constraints on young people without having acquired the basic skills Developed countries have integrated pro- School-based career guidance services they need for work and life. Large num- grams targeted to disadvantaged youth in can help students make better educational bers also begin school late or never start. secondary school to help them go to college, and career choices by providing them with Allowing these youths to remain illiterate and have found them to be effective.Similar information and skills. A review of the lim- or semi-literate and unskilled throughout to the U.K.'s Aimhigher,the U.S.'s Upper ited evidence on these programs in devel- Bound Program focuses more on preparing oped countries found positive effects for their lives is costly for them, their families, for college but does not provide financial career decision making and maturity. Some and their communities. Second-chance assistance.Its impacts on high school and middle-income and transitional economies programs have effects on behavior beyond college performance are generally limited are also introducing these services, namely but large for those with low educational Chile, the Philippines, Poland, Romania, schooling and work that should be con- expectations and at academically high risk. Russia, South Africa, and Turkey. Most com- sidered in evaluating their full effects. For In Mexico, a new World Bank-funded mon is to provide guidance counselors, example, reduced crime accounts for a sub- program recognizes that the limited access who have a broad mandate but focus on stantial part of the benefit of the Job Corps of disadvantaged youth to college needs student learning and behavioral problems. program in the United States.111 to be addressed by a combination of tar- The official ratio of counselors to students is, geted financial assistance (a mixture of however, very low (1:500 in the Philippines, Many countries operate a variety of pro- loans and grants), academic support for 1:800 in Romania), and counselors gener- grams to get out-of-school youths back into poor and talented students in secondary ally do not deal with education and career school or in informal training courses--and schools to prepare them for college, and choices. the diffusion of information about higher illiterate young adults into literacy programs. education opportunities and labor market Few countries, however, have a system of outcomes through a Web-based labor- Sources: Johnson (1996); Myers and Schirm second chances that meets the diverse needs market observatory. (1999); and Watts and Fretwell (2004). of young people. The high cost of operating such a system may be one reason. Another may be the need for innovative solutions and formal school system and informed by effective partnerships. Few countries have the demands from the labor market and pilot tested and evaluated such programs. society--is often missing. In its place are The needs of out-of-school youths are numerous programs that focus on disad- diverse because they leave school at different vantaged youths but are not linked to each points in the schooling cycle, with different other or to the school system. levels of skill attainment. Some have never attendedschoolandarefunctionallyilliterate. Young people still in school: Others have dropped out before completing Remedial education basic education. Still others have completed For those still in school, one policy response "If we could go to school in the basic education but did not acquire basic has been to offer poorly performing students evenings, or for a few hours in the skills. Even among youths who have the same supplementary instruction. Programs iden- middle of the day, we could pull skills (or lack thereof), the second chances tify at-risk students at the primary level and rickshaws during part of the day, likely to be appealing and effective depend provide better grounding for them when and go to school for the rest. We on the age of the young person.112 Second they reach the secondary--but identifying cannot be in school the whole day, chances also need to be tailored to the local who needs such instruction is a key step. In we need to earn money." environment, which might be rural or urban some developed countries, such as Australia, Basti boy, Kalayanpur, Bangladesh in either a low- or middle-income country. Canada, and the United States, the results of January 2006 This diversity adds to the challenge of standardized tests trigger a supplementary providing a system of well-targeted second tutoring program. This may be challenging chances. Addressing diversity, while scal- to implement in many developing countries ing up enrollments, is greatly facilitated because standardized testing is nonexistent or by reaching out to the private sector and much less frequent, so rather than using stan- NGOs. Government in these programs may dardized tests, the successful Balsakhi pro- be most effective as a standard setter, regu- gram in India allowed teachers to informally lator, and funder (along with international identify students that were falling behind. donors), and less as a provider. In many countries, however, the growing A policy and organizational framework numbers of students who take after-school for second chances--clearly linked to the tutoring may be driven largely by teach- 90 WORLD DEVELOPMENT REPORT 2007 ers who seek other sources of income, with · In India a large remedial education pro- families bearing the costs. Teachers have gram for younger children also had a perverse incentive to create demand for positive results.115 Young women from their time in after-school tutoring by resist- the community teach basic literacy and ing improvements in their regular classroom numeracy skills to primary school pupils hours. For this reason, Hong Kong (China) who have not mastered the expected com- and Turkey prohibit teachers from provid- petencies. It could be that class size mat- ing supplementary tutoring for their own ters more for children falling behind their students.113 Such a policy should accompany peers: not capable of following the stan- any effort to provide remedial education. dard curriculum, they need individual, Well-designed remedial programs have nonthreatening attention from a teacher. proven to be successful in improving school outcomes for students of different ages and Young people who are out of school in very diverse environments, and often Equivalency programs. To appeal to out- help the most disadvantaged students. of-school youth, second-chance programs · In 1999, Israel implemented a remedial must take into account why young people education program to boost the per- dropped out or never attended school, the centage of students in the secondary challenges they will face to stay in a pro- academic track who earn matriculation gram, and how they can be integrated into certificates. The program targeted 10th the formal education sector or find employ- through 12th graders who needed addi- ment. All these vary by age, skill, and the tional instruction to pass the matricula- local environment. Equivalency, literacy, and tion exams, usually determined by failing job training programs may serve different marks and teacher assessments. The pro- youth populations, but their common aim gram increased the probability of partic- of providing competencies for work and life ipating students earning a matriculation requires a more integrated approach: literacy certificate by 22 percent.114 and equivalency programs that include life skills and vocational training, and vocational training programs that include life skills. Equivalency systems use more practical B O X 3 . 9 Reaching out-of-school youth in curricula, more flexible schedules, and less Bangladesh formal instruction methods than regular schools. They depend on a strong partner- Two NGOs in Bangladesh--the Bangladesh curriculum--focusing on Bangla, math, and ship between the formal education sector, Rural Advancement Committee (BRAC) and social science--allows students to progress private providers of programs, and pro- the Underprivileged Children's Education at a fast pace, maintaining their interest and Program (UCEP)--have provided education permitting them to catch up to other young spective employers (box 3.9). Without this to many young people. In a comparative people. partnership, the graduates of equivalency study by UNICEF, students from both pro- BRAC and UCEP tailor their delivery systems will be left holding diplomas that grams performed much better on general to target young people of different ages allow neither reintegration into the regu- tests than did students in the Department and skill levels. BRAC's Non-formal Primary of Non-formal Education's own Hard-to- Education Program targets younger people lar school system nor employment in jobs Reach schools program, although more rig- (ages 8­10) and reintegrates them into the requiring a certain level of competency. orous impact evaluations are needed. formal education system.Their curriculum The mode of delivery must take into The two programs rely on many of the was adjusted to include English so that stu- account why young people dropped out. For same principles, but their mechanics differ. dents make a smooth transition to the for- Both target the poorest and thus provide mal education system. UCEP targets older example, to bring programs closer to homes schools and all materials in the areas students (ages 10­16), and thus emphasizes in rural areas of Mexico, the Telesecundaria where the poorest people live. If a school speed (providing two grades of education program offers lessons by video, while in is to be opened in an area, or students are in each year), completeness (providing five the urban slums of Bangladesh, programs targeted to begin the program, parents are or eight years of general schooling), and involved before school starts, and continu- feeds students into UCEP-run vocational rent rooms rather than build schools to ing parental involvement is expected.The programs, which then integrate students solve a supply shortage. Knowing that even feedback of parents, teachers, and students directly into the private sector. small costs can be a barrier to enrollment is regularly solicited, and the curricula have Source: Eusuf and Associates and Center on for the poorest, successful programs pro- undergone continuing revisions. A simpli- Social Research and Human Development fied version of the standard government (2002). vide textbooks, notebooks, and pencils. To accommodate the pressure for adolescents Learning for work and life 91 to work, the Tutorial Learning System in neglected by governments and donors, but Colombia allows students in rural areas and there are signs of renewed interest--literacy their facilitators to determine the preferred is now part of the Millennium Development schedule and pace. This greater flexibility Goals and the Education for All goals. should not, however, come at the expense of The past neglect is partly due to poor educational quality. results of literacy programs, but there is Aspectsof theschoolenvironment--social room for adapting programs to the needs of support, curriculum, learning methods-- participants. Many countries carried out lit- must be properly adapted because they eracy campaigns in the 1960s through 1980s affect both why young people dropped out through government-led, top-down brief of the formal school system, and whether courses that offered no follow-up. The uptake they will stay in an equivalency program. of eligible participants was limited, and about For younger adolescents the support of par- 50 percent of them dropped out. Of those ents can reduce attrition and boost student who stayed on, about half passed literacy performance, so involving parents in the tests,and about a quarter eventually regressed early stages of a program is likely to pay off. to illiteracy. So only a quarter of participants In addition, programs that emphasize social acquired stable literacy skills. Dropout rates support and emotional connections--by fell and completion rates improved for these keeping the same group of students and programs in the 1990s, but the literacy out- teacher together over a multiyear program comes were still modest.117 (Colombia's Tutorial Learning System)-- A key aspect of the improvement was the tend to have lower dropout rates than pro- shift to more contextual and demand-driven grams whose flexibility comes at the expense models. Attracting enrollees and keeping of such support and continuity (Nonformal them interested is a big challenge. Many Education Project in the Philippines).116 countriesoperateadultliteracyprogramsthat Successful equivalency programs that hope teach not only reading, writing, and arith- to reintegrate people in the formal educa- metic but also job and life skills relevant to tion system often use teaching methods that the local context. Several programs in Africa are similar to those recommended above for involve the local community, churches, and formal schools--student-centered learning, businesses. They apply active learning and regular assessment, and remedial sessions to other participatory methods for instruction involve students in their learning progress in local languages, and they include postpro- (box 3.9). Programs for older youth, however, gram follow-up, such as reading activities, to often use very different approaches. The Mex- solidify literacy skills. Examples include the ican National Institute for Adult Education Senegal Pilot Female Literacy Project and the (INEA) has developed an innovative educa- Ghana Functional Literacy Project, which tion model for out-of-school individuals 15 or require at least 300 hours of instruction over older to learn how to complete the equivalent 18­21 months. The cost of the Ghana pro- of primary,lower secondary,or upper second- gram is $24 per enrolled learner per cycle ary education. It provides a curriculum based and $43 per successful literate graduate per on acquiring skills for work and life through cycle. (None of these programs has been sub- a flexible system of modules--individuals can ject to a rigorous impact evaluation.) choose among the modules and the length of the program is attuned to their needs, cover- Job training. Vocational training programs ing subjects such as health and civic education for out-of-school youth can be more cost- and vocational skills. effective by improving targeting, and ensur- ing that programs are complemented with Literacy programs. The poorest countries other services and tailored to the needs of in the developing world, and large pockets of local labor markets. Evidence from devel- disadvantaged groups in better-off countries, oped and developing countries shows low have out-of-school youths who are illiterate. rates of return to most training programs, Despite the magnitude (137 million youth), and few programs pass cost-benefit tests.118 the illiteracy problem has been largely Vocational training is most effective for those 92 WORLD DEVELOPMENT REPORT 2007 at the high end of the wage distribution but employment and earnings prospects of par- often has less effect for those at the bottom, ticipants when training is provided as part illustrating the complementarity of skills.119 of a comprehensive package that includes Programs also tend to be more effective employment services, counseling, and life when they include on-the-job training and skills.120 Examples include the Jovenes and employer sponsorship. Vocational programs Entra 21 programs in Latin America (see for youth are most likely to improve the chapter 4 for more on training programs). Table 3.1 Summary of youth education policies Proven and successful Promising but unproven Unlikely to be successful Opportunities Universal lower Compulsory schooling laws (R. B. de Moving vocational tracks to upper secondary (Chile) Early tracking and selection secondary to provide Venezuela) (Tanzania and Tunisia) basic skills Diversification with Allowing private sector entry and private- No terminal vocational tracks (South Africa and Tunisia) Unregulated private sector flexibility of postbasic public partnerships (Colombia) Transferable credit-based courses (Thailand) (Cambodia) education Providing quality assurance and information Part-time schooling (Argentina and Russian Federation) (Chile and Rep. of Korea) Fostering competition (autonomy, performance-based funding) (Chile) Improving quality and Teaching quality Making the curriculum more relevant Teacher incentives based on relevance of education Continuous, needs-based teacher training Practical, thinking, and behavioral skills (South Africa) narrow test scores (Chile, for work and life with follow-up Kenya, and Mexico) Blending of vocational and general curricula (Chile) Well-designed and negotiated performance- Better connection to work and local economy (China's based pay (Chile) university-local economy linkages, U.S. Career Academies and Germany dual system) School Accountability Disseminating information on school performance (Chile) Capabilities Motivating students Developing behavioral skills (U.S. Big Including students in school policy decision making Brother/Big Sister and Philippines Make a (Georgia and Russian Federation) Connection) Young person­based conditional cash transfers Improving school connectedness with (Bangladesh stipends for girls) students (United States) Improving incentives to exert effort (merit scholarships for girls in Kenya) Providing better Information on education opportunities (U.K. School-based career guidance services (Poland and information Aimhigher) Turkey) Financial incentives to Conditional cash transfers (Mexico Income-contingent loans (Australia and Thailand) alleviate constraints to Oportunidades) Individual learning accounts (Mexico) better choices Vouchers (or beneficiaries) in Colombia Second chances Remedial education Testing to determine eligibility (Israel) Supplementary teaching Combining with other services (information, parallel to regular classes financial incentives, mentoring) (U.S. Upward provided by the same Bound Program) teachers (Cambodia) Equivalency programs Group classes with flexible schedules (Colombia) Individual, self-paced Simplified and practical curriculum, life skills (Mexico) programs (Philippines) Mechanisms to smooth transition to formal education or work (Bangladesh) Literacy programs Beneficiary participation in design (Bangladesh) Combination with skills training (Senegal and Ghana) Built-in follow-up mechanisms (Bangladesh) Partnerships with private sector and NGOs (Bangladesh and Brazil) Vocational training Combining with life skills (Jovenes and Entra 21 in Latin America) Learning for work and life 93 postbasic learning options; a relevant curriculum that teaches practical sub- Governments can help improve the skills of jects, thinking skills, and behavioral young people for work and life by imple- skills; and connection between school menting policies that meet their needs. The and work--all abetted by teachers who balance and sequencing of education poli- are well-prepared and motivated. This cies across the three dimensions--postpri- should be a priority for middle-income mary education opportunities, tools to countries where most young people are enhance education decision making, and equipped with adequate levels of basic second-chance education options--as well skills and there is an increasing demand as prioritization among them (basic skills for postbasic skills (from primary school rather than postbasic skills) depends on the or from the labor market). state of a country's education system (how ·To enable young people to make better it performs in preparing youth for work education choices, develop their behav- and life), its level of development, its overall ioral skills, involve them as stakehold- development priorities, and the priorities ers in their own education, and provide of its young people. For example, young them with financial incentives and good people consulted in Bangladesh demanded information. This applies to all coun- more part-time schooling, while those from tries, but financial incentives should be a Georgia set the teaching of decision-making priority for countries where many young skills as a priority.121 This Report proposes people do not go to secondary school the following key areas of policy action even though the facilities exist. (table 3.1): ·To meet the learning needs of young · Improve educational preparation for people who failed to acquire basic skills adolescence by building a strong founda- the first time around, develop a sys- tion and providing quality basic educa- tem of remedial education, equivalency tion (including lower secondary) for all. programs, literacy programs, and skills Relevant to most developing countries, training that takes into account their this should be a priority for low-income diverse needs and economic conditions. countries where large numbers of young This is a priority for countries with large people do not attain the basic skills numbers of young people who are out of needed for further study, work, and life. school and lack basic skills, particularly · To meet the growing demand for post- countries that have "lost generations" of basic skills, provide diverse and flexible young people. spotlight Vietnamese youth: Managing prosperity Vietnam has been experiencing a period of unprecedented economic growth and poverty reduction. The youth cohort today is larger, more educated, healthier, and more enthusiastic than ever before. However, growth has not fixed all problems. Indeed, it may have brought some new ones for Vietnamese youth: exposure to new health risks, difficult conditions among rural migrants, frustration from the inability to find jobs that match their higher levels of education, and the inadequacy of the skills produced by the education system relative to the changing needs of the labor market.1 Starting in 1986, Vietnam gradually shifted Emerging health risks ing the activities into the lower secondary from a centrally planned system to a social- Greater wealth and changing lifestyles have curriculum. It also includes community- ist market economy. It doubled its GDP in increased the exposure of youth to new based Healthy Living Clubs to reach out-of- the 1990s and more than halved the poverty technologies, mass media, and global cul- school adolescents and equip them with the rate from 58 percent in 1993 to 20 percent ture--45 percent of urban youth have used knowledge and practical skills to respond in 2004. Fueling these changes was a dis- the Internet. This is creating tension between to and cope with substance abuse, unpro- ciplined, hard-working, and fast-learning traditional and modern values. It has also led tected teenage sexual relationships, and the young population. More than half of its 83 to new health risks, such as drug use, HIV/ risk of HIV/AIDS. million people are under 25 years old, and AIDS, unwanted pregnancies and abortions, In the last 10 years, road deaths have 27 percent are between 12 and 24. and traffic accidents. Well over half of all increased fourfold--from 3,000 a year Youth in Vietnam today are more edu- reported cases of HIV infections are inject- to almost 13,000. Road accidents on a cated, healthier, and more optimistic than ing drug users. Youth make up a growing motorcycle are now the leading cause of ever before. The lower secondary school share of HIV/AIDS infections--from 10 death for youth ages 15­24. Motorbike completion rate increased from 25 per- percent in 1994 to about 40 percent today. racing and limited helmet use (only 25 cent in 1992 to 62 percent in 2002 (see the Information gaps--fewer than 60 per- percent of young drivers wear helmets) figure). Although disparities remain, the cent of rural youth had ever heard of syphi- are the main behavioral factors behind improvements have been widely shared, lis or gonorrhea, and 45 percent of youth these figures. The costs are also borne by with females, rural youth, ethnic minori- reported not knowing how to use a con- the society as a whole: road injuries con- ties, and the poor benefiting proportion- dom--and negative attitudes toward con- sume 75 percent of medical care budgets ally more. The first Survey Assessment of dom use--only about 15 percent of youth in urban hospitals. Vietnamese Youth--conducted in 2003 and have ever used a contraceptive method-- The Asia Injury Prevention Foundation consisting of a household-based sample make young people vulnerable to sex- is working with the Ministry of Education of 7,584 youth ages 14­25--shows that related health risks. and Training to introduce a Traffic Safety most Vietnamese youth are hopeful about Young women are especially vulnerable Education curriculum in primary schools. the future, believing that they have more to sex-related health risks because of their More is needed, however, on road safety opportunities and a brighter future than limited decision-making power and the enforcement. their parents.2 lack of comprehensive sex education. The cultural stigmatization of risky health and Managing rural-urban migration Vietnamese youth have become substantially social behaviors as "social evils," particu- The surge in business activity has led to a more educated larly as they relate to HIV/AIDS, has been a huge increase in the demand for labor, with 1992 2002 major impediment in delivering prevention major shifts from agriculture to nonagri- and care to vulnerable groups and in devel- cultural activities and migration from rural oping effective behavior change communi- to urban areas. Between 1994 and 1999, Total cation. Vietnam lacks youth-specific health more than 4 million people seeking better policies to address the impact of HIV/AIDS employment and economic opportunities Richest and substance abuse. moved across provincial borders, with more quintile A few programs have begun to fill this than 53 percent moving into urban centers, gap through life skills education and youth particularly Hanoi and Ho Chi Minh City. Poorest reproductive health services. UNICEF, in Over half these internal migrants were quintile partnership with the Ministry of Educa- younger than 25 years old, with the high- tion and Training, the Vietnam Women's est rate for those ages 20­24. Migration 100 80 60 40 20 0 20 40 60 80 100 Union, and the Vietnam Youth Associa- has been happening at a very fast pace: the Percentage of youth completing each tion, provides healthy living and life skills 2004 population census of Ho Chi Minh level of school for youth, focusing on ethnic minorities City uncovered 420,000 more people living Primary Lower secondary and young women. The program includes in the city than authorities had predicted. Upper secondary life skills education for 120 lower secondary Migrants there make up about 30 percent schools, and its success has led the Ministry of the population, and outnumber perma- Source: Staff estimates based on nationally representative household surveys in 1992 and 2002. of Education to work toward mainstream- nent residents in 7 of 24 districts. Vietnamese youth: Managing prosperity 95 This massive migration wave, by itself, Managing expectations Hanoi complained about the irrelevance of puts pressure on services and jobs and and improving the the curriculum and teaching being too pas- creates tension with the local population. relevance of education sive and not interactive. Under the registration system, migrants Although some training takes place in About 1.4 million young Vietnamese enter need to get permanent registration status firms, it does not cover the training needs the labor market each year. They are becom- in their new places before they can use such of many who struggle to be productively ing better educated--the relative supply of services as public schools, health insurance, employed. This is particularly the case of workers with primary education to those housing, and microcredit. Access to perma- youth who were left behind in the country's with upper secondary education or higher nent residence status, however, is very lim- bonanza, who find themselves without the is growing rapidly--and have high expecta- ited, putting migrants at high risk. basic skills for work and life.There are,how- tions about their futures. The large number In Ho Chi Minh City, about 40 percent ever, some examples of second-chance pro- of better prepared and more enthusiastic of children (ages 11­14) of short-term and grams run by nongovernmental organiza- youth entering the labor market creates seasonal migrants are out of school, com- tions (NGOs) that target these individuals. enormous opportunities,but also substantial pared with 15 percent of children of non- The Blue Dragon Children's Foundation risks if they are not productively employed. permanent migrants, who have resided for is an Australian organization that supports The increase in the return to upper over six months and can demonstrate per- children and youth ages 7­20 from disad- secondary and tertiary education relative manent employment. Older youth are at vantaged backgrounds (including street chil- to primary education between 1992 and even greater risk: 80 percent of short-term dren). A contract is set up with each child, 2002 indicates an increase in the relative and seasonal migrants and 53 percent of specifying objectives, the program to achieve demand for workers with upper secondary nonpermanent migrants 15 to 18 years old them, and the obligations. The program pro- education or higher. Returns between 2002 have dropped out of school, compared with vides integrated services, including scholar- and 2004 have dropped, however, suggest- 34 percent of permanent residents. ships, food and lodging, health assistance, ing that the supply of educated workers is Migrants tend to work in small firms and English and computer skills, recreational beginning to outpace demand. As a result, the informal sector where they enjoy little activities, and counseling. The program also many young people are taking on jobs well protection in terms of collective bargaining, facilitates access to other successful NGO below their education level or are underem- fair wages, and other benefits. Migrants also programs (e.g., KOTO) that provide voca- ployed. They risk becoming frustrated. lack access to public microcredit to start a tional training for disadvantaged youth with The key policy challenge is the inad- new business. at least lower secondary schooling. KOTO equacy of the skills produced by the educa- A revision or elimination of the house- combines hospitality training with life skills tion system to meet the changing needs of hold registration system has been debated training and hands-on experience. Trainees the labor market. About 50 percent of firms in the National Assembly. A proposal to tie are provided with housing, food, medical in the textiles and chemical sectors con- the budget for public services to actual (and insurance, and a training allowance. sider skilled labor to be inadequate for their frequently updated) population counts is needs. About 60 percent of young workers also being considered. Beyond improved with vocational and college education need Youth policy development access to general services, however, few further training right after they become Young people are increasingly seen as an social protection programs are targeted to employed. Software companies also report instrumental force for driving the country's vulnerable migrants. that local IT training institutions fail to future, but cultural perspectives that view Action Aid recently began offering produce qualified graduates, and that they this group as immature and needing guid- holistic assistance to the migrant commu- have to spend at least one year retraining ance and control, persist. This has impli- nity, including evening classes delivering 80­90 percent of recruits. cations for youth policies, which are often basic education for children out of school, Although Vietnam has greatly increased problem oriented. Some positive changes in HIV information, commercial sex worker the number of people that go through the youth policy are under way. The National outreach and services, and a microcredit school system, the curriculum and teaching Youth Development Strategy by 2010 and scheme to support livelihood develop- methods have not kept pace.The curriculum the first Law on Youth have recently been ment. Marie Stopes International provides in upper secondary and tertiary education adopted, and a Master Plan for Youth health services for youth migrants through (and even vocational education) remains Health is currently being developed. And mobile clinics and site-based clinics in too theoretical, providing little variety. some youth have been consulted on these industrial zones. Youth consulted in Ho Chi Minh City and and other government plans. Going to work 4 Once young people are in the labor mar- reintegrate the most vulnerable will allow ket, they begin to reap the benefits of ear- youth to gain productive employment. lier investment in education and health, and The chapter starts by documenting the c h a p t e r continue to develop the skills needed for a challenges youth are facing in their transi- productive livelihood. A successful transi- tion to work along with their consequences tion to work for today's many young people in the short and long terms. The next sec- can accelerate poverty reduction through tion shows why general policies will not better allocation of their labor, and boost solve all the constraints youth face so that economic growth. Some youth, however, a youth-specific approach is needed in the encounter roadblocks: some go to work too labor market. The last three sections discuss early, others cannot enter the work force, still the policy options that could support the others get stuck in low productivity work. opportunity, capability, and second-chance Because most learning occurs at the pillars of the Report's policy framework in beginning of the work life,initial experiences the transition to work. can have long-lasting effects. Difficulties in entering the labor market can persist and Youth challenges be very costly to mitigate. Poverty and slow economic growth can exacerbate poor youth in the labor market outcomes such as child labor, school drop- By age 24, almost all youth in developing out rates, and joblessness. The lack of access countries have left school and entered a new to insurance and information and the unin- stage in life--some to begin wage work, tended consequences of some labor market some to engage in home enterprise, some policies magnify the impact of household to form households and raise families, and poverty and slow economic growth. some to do combinations of these activi- Correcting for these failures requires ties. Over the past 30 years, the age of leaving the right mix of policies to ensure that school has risen in every region in the world enough opportunities are available for except Africa.1 The increased educational young people, that their skills match attainment associated with this should have employment opportunities, and that sec- improved the transition to work and led to ond-chance options protect those who fall greater success once employed. Indeed it has, behind. Some policies, such as improving but too many youth still face significant chal- the investment climate or enhancing the lenges in their path to work. The main obsta- functioning of the labor market, are not cles are starting too early, failing to enter the youth specific, but have a disproportion- labor market, and having difficulties moving ate impact on youth. Other interventions across jobs and up the skill chain. The effect are youth specific and should be designed of these obstacles on skill accumulation, to fit the country context. In most middle- future performance in the labor market, and income countries, the priority is to reform economic development are long lasting. labor market institutions and to build more bridges between school and work Starting too early to better accommodate new entrants. In Some young people never attend school, most low-income countries building on and many others begin working at very basic skills and providing a springboard to young ages. An average of 14 percent of 96 Going to work 97 Figure 4.1 Child labor is highest in Africa Percentage of children economically active 80 70 60 Regional average 50 40 Grand average 30 20 10 0 Africa East Europe & Latin America Middle South Asia & Central & the Caribbean East & Asia Pacific Asia North Africa Source: Fares and Raju (2006). Note: Regional (dotted horizontal lines) and grand (solid horizontal line) means are calculated by weighting country child economic activity rates by country child population (7­14 years). the population ages 10­30 in 82 develop- percent of working children were estimated ing countries have never attended school. to be attending school. Child labor is prevalent among this group, About 70 percent of child laborers are but it is also common among those who in agriculture, predominantly in unpaid entered school. The International Labour family work. Fewer than 10 percent are in Organization (ILO) estimates that despite manufacturing.4 For many poor families, an 11 percent drop in the incidence of child child labor represents a significant share of labor between 2000 and 2004, 218 million household income. For example, in Brazil- are still trapped in child labor.2 In 65 coun- ian households in which children work, tries, about 21 percent of children under child labor represents 17 percent of urban age 15 are economically active (figure 4.1). household income and 22 percent of rural The estimate likely understates child work household income. The unfortunate trade- because it is difficult to measure work out- off is that children who sacrifice schooling side the market and because child labor when young are likely to be poor as adults.5 is characterized by short spells missed by Recently completed research has surveys.3 The highest reported rates are in improved our understanding of the deter- Sub-Saharan Africa, averaging 35 percent. minants and consequences of child labor.6 In seven of 29 African countries, more that Many adolescents work while in school, but half the children between ages 7 and 14 are the effect of doing so is unclear (chapter working (figure 4.1). 3). For young adults, working may enable A consensus against the very visible and them to finance upper secondary and ter- worst form of child labor is reflected in the tiary education that would otherwise be large number of countries ratifying ILO unaffordable. The problem is with younger convention 182 on the Worst Form of Child people--work appears to be more damag- Labor. However, the majority of children ing to school attainment because prema- are not working under such harsh visible turely dropping out of school reduces the conditions. The reality is quite complicated. amount that youth learn while in school. Many working children combine school A negative association between work and with work. In 29 countries in Sub-Saharan the test scores of 8th graders is found in a Africa, an estimated 52 percent of children majority of countries. 7 Poorer schooling working were also attending school, while in outcomes also lead to poorer earnings later 19 countries in Latin America as many as 78 in life. In Brazil, boys who entered the work 98 WORLD DEVELOPMENT REPORT 2007 force before age 12 earned 20 percent less only to urban male youth, but to women per hour and were 8 percent more likely to and rural youth as well. be in the lowest income quintile than boys Children of parents who worked as who started working after age 12.8 children are more likely to work at young Leaving school too early is costly for ages, holding other household attributes later productivity. The forgone earnings constant, suggesting that child labor recurs and the lack of skill accumulation can across generations and may be a means by make it much more difficult to escape pov- which poverty is passed on from parent to erty as an adult. In 61 countries, the esti- child.10 Child labor can also have temporary mated average return per year of schooling or permanent adverse health consequences was 7.3 percent for men and 9.8 percent for that can hinder future earning capacity.11 women.9 The returns are highly correlated within countries, so markets that reward Failing to enter the labor market schooling for men also reward women, and Many young males and females face signifi- markets that reward urban residents also cant difficulties entering the labor market. reward those in rural areas. These returns This employment difficulty can be mea- suggest that across a wide array of coun- sured differently depending on the country tries at all stages of development, educa- context and youth characteristics (box 4.1). tion offers substantial wage returns--not According to most measures,youth are more B O X 4 . 1 Measuring youth activity in the transition to work Conclusions about youth labor market out- force that is unemployed.The employment rate the caveats of each indicator and suggests comes differ depending on how youth time is the share of the population that is employed. alternatives: allocations are measured. For example, youth Two other measures are used in developing- · The unemployment rate is a measure of dif- who are not employed in the formal market may country settings: the proportion of the popula- ficulty of finding work. In middle-income be spending time productively in school or in tion neither in the labor force nor in school, and countries the ratio of youth to adult unem- informal production activities or they may also the proportion of the population neither work- ployment rate is telling, but short-term and be actively seeking work (see figure below). ing nor in school (the jobless rate). long-term unemployment need to be distin- The standard labor market indicators in No one measure provides a complete pic- guished. In low-income countries, the youth developed countries are the unemployment ture of the labor market for youth, so multiple unemployment rate is very low, and relevant rate, the employment rate, and the labor measures are needed to analyze youth labor only for the more educated and better off force participation rate.Youth are considered markets in developing economies. For a sample portion of the population. employed if they work at least one hour for a of 91 developing countries, these indicators · The employment rate for youth does not wage in the weeks prior to the survey but also are not perfectly correlated.There is an inverse account for school enrollment and the type if they work in unpaid labor for an enterprise correlation between the unemployment rate of work. Using the population out of school as owned by their households.To be consid- and the employment rate, and a weaker inverse a reference group allows for better compara- ered unemployed, an individual must be not correlation between the unemployment rate bility with adults, while looking at the sector employed but actively seeking work.The labor and the labor force participation rate. According of work, hours of work, and measures of earn- force participation rate is the share of the popu- to the level of country development and the ing shed light on the quality of employment lation either employed or unemployed, and the gender and education of youth, the relevant and underemployment. unemployment rate is the share of the labor indicator could vary.The following list indicates · "Out of school and out of work"is a measure of unused human capital but not for girls involved in household activities.The relative ratio for males in this group indicates the An illustration of youth time use extent of discouraged youth who withdrew from the work force. In the labor force Not in the labor force · Youth employment is considered informal if the job is unpaid or if the job includes no Employed Not in the labor force and not in school benefits such as participation in the country's Formal market social security system. High rates of informal- ity are a signal that youth are finding less Informal Enrolled in school permanent, low-quality jobs. market Working Not working · Combining school and work is potentially harmful for the very young, and could be an Unemployed indicator of the risk of early school exit. Source: Fares, Montenegro, and Orazem (2006a). Going to work 99 likely to be unemployed than adults. Sig- Figure 4.2 Unemployment is higher for youths than nificant variation in unemployment exists for adults between urban and rural sectors, between Youth unemployment rate (percent) developed and developing countries, as well 40 as between poor and rich households.Young 2:1 females are more likely than young males to 3:1 stay out of the labor force. Early difficul- 30 ties in finding employment can have long- 1:1 lasting effects on employment later in life. Young people have a hard time finding 20 employment. Survey data from 60 devel- oping countries suggest that, after leaving school, youth spend an average of 1.4 years 10 in temporary or intermittent work and spells of joblessness before permanently entering stable employment.12 This estimated dura- 0 tion varies widely between countries and 0 10 20 30 40 estimation methodologies, but could reach Adult unemployment rate (percent) above four years in some instances. In many Source: Fares, Montenegro, and Orazem (2006a). countries in Eastern Europe, Latin America, Note: The 1:1 line represents the cases in which the estimated and the former Soviet Union, youth enter- unemployment among youth and adults is identical. The 2:1 (and 3:1) lines represent cases in which estimated youth unemployment ing the labor market experienced long spells is twice (three times) as high as adult unemployment. Each data of unemployment.13 Initial failure in finding point in the graph represents one country. a job can lead to persistent joblessness for young people, especially in weak economies. In most countries, the less skilled youth In every region the difficulty youth face are more likely to face difficulties in finding in entering the labor market is evident work compared with more skilled youth. in higher unemployment rates for young However, in some developing countries men and women than for older workers. the unemployment rate is very high even Youth make up 25 percent of the working- among educated youth, a great concern age population worldwide, but 47 percent to many countries in the Middle East and of the unemployed. The estimated global North Africa. In Tunisia, where the unem- unemployment rate for youth increased ployment rate for 20- to 24-year-olds is steadily from 11.7 percent in 1993 to 14.4 more than three times higher than the rate percent in 2003. It varies widely across for those over 40, it is more than 40 percent regions, from a low of 7.0 percent in East for youth with higher education compared Asia to 13.4 percent in industrial economies to about 25 percent for those with primary to a high of 25.0 percent in the Middle East education.17 The disadvantage faced by the and North Africa.14 Across all markets the most educated persists over several years youth unemployment rate is two to three after entry into the labor force. The cross- times higher than the adult unemploy- sectional pattern implies that it takes an esti- ment rate, regardless of the level of aggre- mated 10 years for the unemployment rate gate unemployment (figure 4.2). The high for Tunisians with postsecondary education unemployment rate among youth in some to drop below that of the less educated. countries has made unemployment in these Unemployment is only one symptom of countries a youth problem. In the Arab problems in the labor market. Within poor Republic of Egypt, Indonesia, Qatar, and countries, youth unemployment is concen- the Syrian Arab Republic, youth make up trated among those who are educated and more than 60 percent of the unemployed.15 from high-income families. Other dimen- In most developing countries, the youth sions in the labor market must be examined unemployment rate is higher in urban than to assess the ease or the difficulty for young in rural areas, and is estimated to be higher people to integrate themselves in the labor for young women than for young men.16 market. Some youth are neither working 100 WORLD DEVELOPMENT REPORT 2007 nor in school (box 4.2); other poor young Figure 4.3 Where young people cannot afford to people cannot afford to stay unemployed, be unemployed, youth unemployment is low and employment is high most have to work. So the incidence of unemployment may be low, although youth Employment/unemployment rate (percent) 50 are still mired in poverty. In the poor- est countries, youth unemployment rates Youth unemployment Youth employment are very low and employment rates are 40 very high (figure 4.3). As country income increases the incidence of unemployment 30 among youth also increases--a likely reflec- tion of available alternative income sources 20 and safety nets that make it possible to be unemployed. Youth employment rates also 10 fall as per capita income increases because youth devote more time to schooling. 0 Low-income Lower- Upper- High-income countries middle- middle- countries The effects of these difficulties are lasting. income income An initial period of unstable employment countries countries upon leaving school is common and not Source: Fares, Montenegro, and Orazem (2006a). of great concern if young people eventually move into more stable jobs, but this is not always the case. The duration of unemploy- ment for some is very high. For example, in 2000, more than an estimated 60 percent of unemployed youth remained unemployed B O X 4 . 2 Some youths are neither in the for more than six months in the Czech labor force nor studying Republic, Hungary, and the Slovak Repub- lic.18 In both Brazil and Chile, youth cohorts In many countries, the proportion of youth The incidence of young females neither in who are neither in the labor force nor in that entered the labor market during reces- the labor force nor studying is higher than school is too large to dismiss as a problem for young males, which is higher than for sions faced an atypically high likelihood of of measurement or as a temporary phe- adult males unemployment during the recession and nomenon. A cross-country comparison of persistently high unemployment for several young men and women reveals important Youth neither in labor force nor studying differences in these proportions across gen- (percent) more years even after recovery began.19 In 80 der and relative to adults (see the figure): Bosnia and Herzegovina, youth's difficult Males · Differences in levels. Female observations Females entry into the labor market led to low future almost always lie to the right of male 60 earnings (box 4.3). observations, implying a higher incidence Long spells of unemployment can dis- for females relative to males. 40 courage youth from remaining in the labor · Differences in ratios. Most observations for force, leading to a high incidence of youth men lie above the 45 degree line, where the estimated incidences for youth and out of school and work (see box 4.2). The 20 adults are identical, implying that young delays in finding work are important at this men are systematically more likely to fall age because young people need the early in this group than adult men.The pattern 0 is not replicated for females. 0 20 40 60 80 experience to build on their basic education and to continue to acquire skills relevant to Some of the high estimates may be Adults neither in labor force nor studying (percent) attributable to measurement problems, par- the labor market. Being either unemployed ticularly for young women working in their or out of the workforce for a long time can Source: Fares, Montenegro, and Orazem (2006a). households. In Tanzania, the main reason Note: The 45 degree line represents the cases in limit the accumulation of human capital young women said they were not looking which the estimated incidence of neither working young people need to get better integrated for work was their household responsibili- nor studying among young and adults are identical. Each data point in the graph represents one country. ties. For young men, it was the lack of mar- into the workplace and find productive ket work. nonmarket activities.Youth who are neither employment. The data reveal important differ- attaining marketable skills in school nor During repeated spells of unemployment, ences between males and females in this using those skills in productive work are a young men and women in Canada and the group--males are predominately discour- wasted resource in the economy, so mecha- aged workers, while females are engaged in nisms need to be found to tap that resource. United States increase their uptake of train- ing, stay longer in school, delay marriage, Going to work 101 B O X 4 . 3 Early unemployment persists in Bosnia and Herzegovina In Bosnia and Herzegovina,despite the end of the labor market.Among those 15­24 who were period were also found to have faced a greater the civil conflict in the late 1990s,youth have had unemployed in 2001,77 percent were jobless one likelihood of unemployment or joblessness significant difficulties entering the labor market year later,and 58 percent were still jobless three (both inactivity and unemployment) in 2004. and experienced excessive instability in their early years later.Even among youth employed in 2001, Among young workers, the experience of job- years of the transition to work.In 2004,the unem- a third of them were jobless in 2002,and a quarter lessness is associated with about 11 percent ployment rate was 62 percent for those between of them were still jobless in 2004.20 greater probability of unemployment and 30 15 and 19 years old,and 37 percent for those Controlling for young workers'character- percent greater probability of joblessness.The between 20 and 24 years old,compared to 22 per- istics (for example, gender, education, marital effect on earnings is also significant. For all cent for adults in the same year.These outcomes status), those who suffered a spell of unemploy- workers in Bosnia and Herzegovina, a spell of persist in the first few years of youth experience in ment or inactivity at any point over the 2001­02 joblessness is associated with lower wages. Poor transition probabilities in Bosnia and Herzegovina, 2001­04 (percent) Employment status in 2002 Employment status in 2004 Inactive Unemployed Employed Inactive Unemployed Employed Employment Status in 2001 All Inactive 73 12 14 63 15 21 Unemployed 32 34 34 30 27 43 All employed 13 07 81 13 08 79 Ages 15­24 Inactive 71 17 12 53 21 26 Unemployed 36 41 23 26 32 42 All employed 22 12 66 11 15 74 Source: Fares and Tiongson (2006). Note: "Inactive" is defined as being out of school and out of the labor force. and continue to live with their parents.21 Not youth are less likely to be employed com- all these options are available in developing pared to older men and women. Even after countries. When productive options are not adjusting for school enrollment, the differ- available for jobless youth, there is a greater ence between youths' and adults' employ- likelihood that they will enter activities dam- ment rates persists. Among youth, the aging to themselves and society. Youth diffi- employment rate for young men is always culties in the labor market increased crime higher than the employment rate for young rates in France, and increased the probabil- women, partly reflecting a stronger attach- "I am positive for my future. I'm ity of incarceration in the United States.22 ment to the labor force among males--but sure I will find a job sooner or later Similarly, in Sri Lanka, where the proportion also reflecting the additional difficulties and the first job doesn't mean a job of long-term unemployed young people many young women face in going to work that I will do my whole life. The exceeds that of adults, high youth unem- and the greater proportion of them engaged most important thing for me is to ployment was cited as the main cause for in home production, not included in mea- improve myself." large-scale unrest of Sinhalese youth from sured employment. Xiangju, university student, the rural south. The second insurgency from When working, youth often are found China 1987­91 brought the country to the verge in low-paying jobs or unpaid family work. December 2005 of collapse and left 40,000­60,000 dead or For 74 developing countries with data, only missing, most of them youth.23 25 percent of working youth in low-income countries are in paid work, with the pro- Getting stuck in jobs that portion rising with country income, to 57 do not build human capital percent for the middle-income group and Youth and adult employment are posi- 74 percent for the high-income group. Even tively correlated--as adult employment those who are paid are less likely to have rises, youth employment also increases. But access to social security compared with 102 WORLD DEVELOPMENT REPORT 2007 older workers. In these 74 countries, it is Figure 4.4 Returns to experience are highest for not uncommon to find the incidence of the young unpaid work to be two to four times higher Percent return to one year of additional experience for youths than adults (chapter 1). 7 Starting in a low-paying job, or being Burkina Faso Brazil mismatched early on with the wrong type 6 Chile of employment, would not have severe con- Indonesia sequences if youth can move to more pro- Mexico Pakistan ductive opportunities. Indeed, during the 5 Slovenia early transition to work, youth are expected Thailand Uganda to be experimenting with different types of 4 employment, and evidence indicates that such early turnover will enhance subsequent job matches, reflected in higher productivity 3 and higher earnings for youth.24 However, not all youth are mobile. In Burkina Faso, 2 more than 90 percent of teenagers between 15 and 19 with no education started working as family helpers in 1993,falling only to about 1 80 percent five years later.25 Higher educa- tion does increase mobility. More than 80 percent of teenagers between 15 and 19 with 0 20 30 40 some secondary education started working Age as family helpers in 1993, and this share fell Source: Fares, Montenegro, and Orazem (2006b). to about 40 percent five years later.While ini- tial differences were not as large, those with higher education moved out to find better On entering the labor market, youth may productive work much more quickly. have the opportunity to obtain formal and Does starting in informal or unpaid on-the-job training, with a large impact work rather than formal-sector work lead to on their eventual earnings. Young workers different employment and wage outcomes? have the fastest wage increases during this Where informality is widespread, the infor- period of learning on the job, and the rate mal sector is an important stepping stone in slows as workers age (figure 4.4). Returns the transition from school to work for those to an additional one year of experience at who have the choice. In Latin America the age 20 increase earnings by up to 6 percent. vast majority of apprenticeship occurs in However, holding everything else constant, the informal sector, and new entrants might an additional year of experience at age 40 choose the informal sector to acquire the increases earnings by less than 3 percent. skills needed for the labor market.26 In Alba- The more skills acquired in the early work nia, Argentina, Georgia, Hungary, Poland, career, the more the worker can earn later Russia, Ukraine, and República Bolivariana on. In more than four-fifths of the coun- de Venezuela, youth gain substantially when tries analyzed, earnings peaked after age 40, they move from the informal to the formal with an average peak in earnings at age 47. sector.27 The wage gains from the move At the peak, earnings were on average 2.5 are significantly greater for youths than times the starting wage,indicating consider- for older workers. Youths also benefit from able skill development after leaving school, faster wage growth once in the formal sec- most of it in the first few years on the job. tor, both in comparison with older workers Youth who lose the chance to acquire these and in comparison with their counterparts skills after leaving school because of early in the informal sector. Some, however, labor market difficulties may face a career become stuck in informal low-paying jobs of lower skills and poorer pay. Indeed, in that offer no opportunity to further develop Hungary, initial career success drives later their human capital. labor market outcomes.28 Going to work 103 The effect on poverty and social outcomes PETI (Program to Eradicate Child Labor) in should not be understated.In many countries, Brazil were shown to be successful already in some households with working youth are reducing child labor incidence by providing still poor, even after factoring in youth earn- support to income-constrained families con- ings. Because the most abundant asset of the ditional on their children attending school poor is labor, if poor households are unable to and after-school programs. escape poverty even when their youth work, it Young people suffer disproportionately is unlikely that they will do so through other from weakening labor markets. Cross-coun- means. This puts the policy issues for poverty try analyses from developed and developing alleviation in stark focus. The overwhelming countries show that increased labor demand evidence is that better schooling helps youth always improves employment and increases make an easier transition from school to labor force participation among youth.30 In work and enjoy greater success. More stable Brazil in the 1980s and 1990s, employment employment and earnings also ease the next rates for adults during economic downturns transition into marriage and household for- and expansions varied only slightly from the mation. However, the poorest households trend, but movements were much greater for cannot meet their current consumption needs young males and females.31 The differences without the income earned through their in employment fluctuations are even more children's labor, so their children's schooling apparent in Chile. The employment rate for and potential escape from poverty is sacrificed young males is consistently below that of for current subsistence. adults, partly because a large share of those between 15 and 24 years old are enrolled in school and also because of the greater diffi- What makes youth vulnerable culties youth face in their transition to work. in the labor market? However, the difference is not constant and Youth labor market outcomes are affected varies widely over time. During the economic by general trends in poverty and economic slowdowns of 1974, 1982, and 1998, the growth. Because of the severe income con- employment rate for young males, adjusted straints in low-income countries, house- for school attendance, declined at least twice holds may have no choice but to send as much as that for adult males (figure 4.5). children to work, while in slow-growing What is contributing to the vulnerability economies, youth have significant difficul- of youth in the labor market? Four factors: ties in finding work. · In some countries, large cohorts of new The incidence of child labor is high in the entrants and higher female participation Sub-Saharan Africa and South Asia regions, rates will continue to add pressure on characterized mostly by low-income coun- the youth labor market. tries (see figure 4.1). This relationship, how- ever, is not linear and indicates that at very low levels of income, the effect of changes in Figure 4.5 In Chile employment is more responsive per capita income on the incidence of child to demand fluctuations for young males than for adults labor is the highest. Significant variation also Employment rate (percent) occurs in the incidence of child labor even at 100 Males ages 25­49 similar levels of income, which indicates that 90 factors other than poverty could increase or reduce the incidence of child labor. Some 80 families and children have low perceived returns to education, while others face bor- 70 rowing (and other) constraints to finance 60 Males ages 15­24 their children's schooling. Microanalysis for Burkina Faso and Guatemala shows that the 50 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 incidence of child labor increases when poor families are faced with income shocks.29 Source: Fares and Montenegro (2006). Note: Shaded portions represent periods of economic slowdown. Conditional cash transfer programs such as Employment series are adjusted for school enrollment. 104 WORLD DEVELOPMENT REPORT 2007 · Poor access to information and credit Lack of access to information leads to premature exit from school, per- and credit petuating skill mismatches. Leaving school to start work before acquir- · Policy failures have unintended con- ing the relevant skills limits the ability of sequences on youth employment and youth to take advantage of future work widen the gaps between youth and adults opportunities (chapter 3). Poor households in the labor market. with limited access to credit, facing income · Social institutions hinder the full partici- or health shocks, might have no option but pation of many youth, particularly girls, to withdraw their children from school in skill acquisition and work. and send them to work. Low expected returns to education might also cause early Large youth cohorts school dropout and entry to work. Because of information failures, households may Several developing countries around the undervalue the potential returns from world are experiencing larger youth cohorts schooling, particularly when jobs requiring (see spotlight on differing demographics education are in urban areas and the house- following chapter 1). In some, particularly hold is rural. These information failures are in Sub-Saharan Africa, the share of youth greatest in households with poorly educated in the population will continue to rise for parents. the next few decades, adding pressure on Another reason for low expected returns the youth labor market. Indeed, earlier baby is that policies restrict youth from moving booms in developed countries increased easily from one job to another, from one area youth unemployment (see spotlight on to another, or from one industry to another. baby booms following chapter 4). TheHeritageFoundationEconomicFreedom In 32 developing and transition coun- Index measures how a country's economic tries, a 10 percent increase in the youth institutions allow people to work, produce, share of the population increased youth consume, and invest in the ways they feel unemployment by an estimated 6 percent between 1980 and 2000.32 In Ethiopia, local are most productive.36 Returns to schooling average 9.9 percent in the group of develop- labor markets with the largest share of youth ing countries where workers are freer to seek in the population had the lowest youth economic advantage, but 6.4 percent in the employment rates, with the effect more pronounced among uneducated youth.33 less mobile group (figure 4.6). This is consis- tent with theoretical work that ties returns to Thus, countries with rising youth cohorts human capital to economic mobility across will face increasing challenges in absorbing alternative sectors and occupations. youth in jobs. Lack of access to information reduces Even where the youth share of the popu- the effectiveness of job search and prolongs lation is decreasing, the underlying increase joblessness among youth. Not knowing the in female participation rates will limit the available opportunities in the labor market impact of slower population growth as and how to prepare for them reduce the like- larger shares of young females in these lihood of youth developing the right skills cohorts look for employment in the labor market.34 Rising educational attainment has and finding the appropriate job for their skills. The information asymmetry makes had a particularly important effect on the employers less confident in hiring new labor supply choices for women. As women entrants because they are not certain about acquire more education, they increasingly their productivity. It also increases turnover move out of traditional household or agri- as youth and employers learn more about the cultural production activities and enter quality of their employment relationship. wage work. Rising female education levels and the associated rise in female participa- tion in wage work have another effect--they Restrictive labor market institutions are strongly inversely correlated with coun- Labor market institutions--such as unem- try fertility rates.35 ployment insurance, employment protec- Going to work 105 Figure 4.6 Returns to education are higher for ment rates for young workers.40 The adverse workers who are more mobile impact of such regulations on young work- Return to education (percent) ers' employment rates was more than twice 20 that on prime-age male workers in 15 Latin 15 American and Caribbean countries and 28 More mobile Organisation for Economic Co-operation 10 Less mobile and Development (OECD) countries in the 5 1980s and 1990s.41 0 Because youth are more likely to be at the "Young people are deprived of secure ­5 bottom of the wage distribution, changes jobs; their unemployment rate is in the minimum wage will naturally have a well above the national average." ­10 0 2 4 6 8 10 12 14 16 larger impact (positive or negative) on them. Jérémie, law student, France Years of schooling In Brazil, an increase in the minimum wage Source: Fares, Montenegro, and Orazem (2006b). led to greater job loss for female, young, and Note: "More mobile" countries have an average return to educa- low-skilled workers whose wages were clus- tion of 9.9 percent; "less mobile" countries have an average return to education of 6.4 percent. Countries are classified as "more tered around the minimum. In Chile, mini- mobile" and "less mobile" according to the Heritage Foundation Economic Freedom Index. Each data point in the graph represents mum wages reduced the overall employment one country. probabilities of youth, particularly the unskilled.42 Even when the informal sector is tion laws, and the minimum wage--came large, as in Latin America, minimum wages into being in response to the suffering of in the formal sector spill over into wages the unemployed and the exploitation of in the informal sector.43 As a consequence, workers. Despite good intentions, those youth in the informal sector are also affected institutions are less than optimal in many by changes in the minimum wage. countries. Their design might have been When public sector wages and benefits poor to start with, circumstances and are more generous than private sector com- the economic environment might have pensation,astrongincentivearisesforyoung changed, and political considerations might (usually educated) school leavers to queue have given institutions a life and shape of for government jobs and stay unemployed their own. Reforming these institutions has for some time after graduation. Substantial long been on the agenda in many coun- wage premiums in the public sector--cou- tries.37 Some of these institutions have a pled with job security, tenure, prestige, and disproportionate effect on youth. other nonwage benefits--influence the deci- Employment protection laws are effective sion to voluntarily hold out until a public in protecting jobs and preventing job loss, sector job opportunity opens. In Morocco, butalsoraisehiringcosts,puttingyoungpeo- the starting hourly wage in the public sec- ple at a disadvantage.38 World Development tor is 42.5 percent higher than in the private Report 2005 and Doing Business 2006 show sector.44 This leads to a strong preference for that employment regulations can be more public employment among highly educated stringent in developing countries than in young Moroccans. In Tunisia, the public industrial countries. High firing costs reduce sector wage premium is 18 percent, again layoffs as well as job creation in firms and leading the young to queue for jobs in the limit the entry of new firms, disproportion- public sector rather than accept less attrac- ately burdening youth in the labor market.39 tive private sector jobs.45 In Ethiopia, a large Furthermore, reduced turnover increases share of the unemployed youth aspired to the duration of unemployment. This effect, work in the public sector because of the combined with the lasting impact of long, perceived high benefits.46 early unemployment spells, impairs youth These results are not unique to Ethiopia, outcomes and future prospects. In Chile, Morocco, and Tunisia. Earnings regressions where job security provisions depend on for 39 developing countries reveal a public job tenure, employment was biased against wage premium in 25 countries, on average young workers. Stricter employment protec- about 26 percent, controlling for individual tion laws meant lower wages and employ- characteristics. For other countries such as 106 WORLD DEVELOPMENT REPORT 2007 Cambodia and Vietnam, the public wage ket institutions is a priority. In low-income premium is negative.47 In Latin America the countries, with large informal sectors and public wage premium is much higher for dominance of the rural economy, reforming women than men. In several countries in institutions will have limited impact. Thus, Latin America and in Indonesia, the public expanding alternatives in the rural sector, wage premium also varies with skill levels.48 promoting sectoral and regional mobility, and reducing child labor are most urgent. Inhibiting social institutions, especially for young females Improving the investment climate In all regions over the past three decades, Economic growth and job creation benefit labor force participation rates have risen most participants in the labor market, youth for young women, coincident with rising included. When labor demand is strong, female educational attainment and falling youth employment and labor force partici- fertility rates. In some regions, however, ris- pation for both males and females increases ing female education levels have not trans- while the unemployment rate for youth goes lated into dramatic increases in labor force down.51 Because the private sector should participation rates for young women. Social lead in job creation, a good investment cli- institutions and norms could be a reason. mate is needed to allow firms to form and A simple cross-country regression for 128 expand. World Development Report 2005 countries shows that religion alone can argued that governments should create a explain about one-third of the variation in better investment climate by tackling unjus- female participation rates.49 tified costs, risks, and barriers to competi- Social norms can also affect whether tion. They can do this by ensuring political young women succeed in the labor market. stability and security, improving the regula- In Egypt, women, whose average education tory and tax climate for investment, provid- level has increased enormously, stay close ing needed infrastructure, and improving to home and refrain from driving--limit- information on vacancies for job seekers. ing their job mobility. While young males Expanding world trade has shifted pro- increased their commuting distance to work duction around the world. Because the between 1988 and 1998, young women did young are the most able to respond to the not, limiting their access to paid employ- growing demand for labor, these shifts favor ment outside government. The least edu- young workers.In Indonesia,industries with cated women are the most disadvantaged, youth employment shares more than twice confined either to domestic work or to non- the national average are concentrated in wage work in home-based enterprises and electronics and textile manufacturing, heav- subsistence agriculture.50 ily engaged in exporting. The young workers in these youth-intensive exporting sectors are disproportionately well-educated-- Broadening labor 47 percent have completed secondary market opportunities schooling, compared with 11 percent of In all countries, a good investment cli- other employed youth. Similarly, in Viet- mate lets the private sector expand, helps nam 20 percent of the young workers in trade flourish, and allows the country to those sectors have completed secondary attract foreign direct investment, all needed school, compared with 8 percent in other for job creation. Youth can contribute to sectors, and in Brazil 41 percent, compared the growth of these sectors, but steps are with 15 percent. needed to mitigate the effect of market and Youth-intensive exporting sectors seem policy failures that disproportionately affect to target young female workers, who make youth. The steps will differ depending on a up 74 percent of employed youth in those country's level of development. In middle- sectors in Indonesia and Vietnam, com- income countries and countries where labor pared with 46­50 percent in other sectors. market institutions are more binding and Young workers in exporting firms are bet- likely to be enforced, reforming labor mar- ter paid than their counterparts elsewhere. Going to work 107 Figure 4.7 Productivity and earnings increased faster for youth relative to older workers in Slovenia during the transition, (1992­2001) Relative productivity differential Relative wage differential Percent Percent 120 120 Ages 40­49 100 100 Ages 40­49 80 Ages 18­29 80 Ages 18­29 60 60 40 40 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Source: Vodopivec (2005). Note: The figures show relative productivity and earnings of different age groups compared to a reference group of people ages 50 and above, whose productivity and earnings were normalized to 100 percent. In Indonesia, young workers in the export cated labor force able to adapt to new technol- sector were paid 30 percent more than ogy with appropriate knowledge, skills, and young workers in other sectors and 20 per- behavior. In several Latin American countries, cent more than young workers in other this has boosted high-skilled wages.54 The manufacturing. Young workers may also be experiences of Thailand and Malaysia sug- attracted to exporting firms and those with gest that government has a role to play in at least some foreign ownership, because of reducing the negative production externali- the greater likelihood of training.52 ties caused by accelerating wage increases, The young may be particularly attractive in turn caused by shortages in quality to firms in the new and growing sectors of labor.55 Evidence suggests that opportuni- the economy because they are more adapt- ties in export-oriented sectors can provide able than older workers to new production incentives for youth to acquire more skills. methods. During the Estonian transition, Among 48 developing countries, increases the relative share of employment and the in apparel and shoe exports as a share of returns to experience rose faster for the GDP were found to be positively associated young.53 In Slovenia, wages and labor pro- with subsequent upturns in both male and ductivity rose faster for the youngest work- female secondary school enrollment. For ers (figure 4.7). In the early 1990s, younger the average country, a doubling of apparel workers were the least productive age and footwear exports as a share of GDP group in the labor market. The productiv- raises female secondary school attendance ity of those between 18 and 29 years old by 20­25 percent.56 was about 70 percent of the productivity of those 50 years and older. Around 1996, Reforming institutions young workers' productivity caught up with World Development Report 2006 discusses that of 40- to 49-year-olds, and by 2001 why governments intervene in the labor mar- youth had become the most productive. At ket and how poorly designed or inappropri- the same time, earnings rose at a pace faster ate policies can make conditions worse for for the young than for older workers, clos- equity and efficiency. As shown earlier, these ing significantly the gap in earnings with policies disproportionately affect youth. older workers. In high-income countries, the result is a Industrial growth led by foreign direct reduction in the employment rate for young investment was initiated partly due to the people and an increase in the incidence and availability of cheaper young labor. However, duration of unemployment (see spotlight the dynamic growth process increased the on baby booms following chapter 4). In low- demand for a highly skilled and highly edu- and middle-income countries, the result is a 108 WORLD DEVELOPMENT REPORT 2007 B O X 4 . 4 Reforming part of the labor market has been no substitute for comprehensive reform in Spain and France Spain and France have suffered from high costs proved to be effective in reducing overall the introduction of the fixed duration contract, unemployment among youth over the last two unemployment, particularly for the young. the Contrat à Durée Déterminée, CDD, a par- decades. Both governments have experimented Youth unemployment decreased from about 40 tial reform in 1979.The proportion of those with partial labor market reforms as a means to percent in 1995 to 20 percent in 2000. employed under the CDD rose from 3 percent reduce the youth unemployment problem. In France,youth unemployment rates have in 1983 to 17 percent in 2000.This substantially In Spain, the initial reform in the early remained above 20 percent since the 1990s.In increased turnover without a reduction in 1990s was intended to increase employment 2006,the government proposed a contract for first unemployment duration. flows among youth by making it less costly for employment (Contrat Première Embauche,CPE) The lesson: Because partial reform, as in firms to hire young workers. Lowering the fir- that would have allowed employers to fire workers Spain in the early 1990s, tends to reinforce ing costs for entry level jobs without changing under 26 within a two-year trial period without market segmentation, it is a poor substitute the costs for regular jobs may have increased giving a reason.It was hoped that employers for broad reform, and its public support and firms'willingness to hire new workers but also would be more likely to recruit young people if effectiveness are limited. Another alternative, made them reluctant to retain these workers. they knew they could be fired readily and that this suggested by analysts in France, could be a The result of this policy was a systematic rise would reduce youth unemployment.Students, more gradual approach that deemphasizes in the use of temporary contracts for young with the support of workers'unions and opposi- the segmented labor market for different age workers, without an increase in permanent and tion parties,demonstrated against the new law groups--perhaps a progressive contract in stable employment. Over 30 percent of employ- and forced a showdown with the government, which protection gradually increases as a ment became temporary, while unemployment leading to the withdrawal of the proposed law. worker's tenure rises. remained as high as 24 percent. Spain had Some decried the lack of consultation. Sources: Blanchard (2006); Blanchard and Landier to move to broader reforms in 1997. Policies Analysts also argued that the CPE reinforced (2001); Cahuc and Carcillo (2006); Kugler (2004); and such as reductions of payroll tax and dismissal the market segmentation already in place since Kugler, Jimeno, and Henanz (2003). segmented labor market--one comprising a and train young inexperienced workers. small number of workers who benefit from Analysts advise against jumps in protec- greater employment and income security tion from one type of employment to and another comprising a large number of another--jumps likely to segment the mar- young workers alternating between short ket (box 4.4). A more progressive contract spells of employment and joblessness, with is preferred, where protection increases little access to security and opportunities to gradually with tenure with no large discrete develop their human capital.57 changes in protection.59 Such contracts pro- vide incentives for firms to invest in young Broad, not piecemeal, labor market reform. workers as their productivity increases with To protect those currently employed, some longer tenure. governments have experimented with mod- A similar approach applies to the mini- est labor market reforms, but because partial mum wage, intended to protect workers' reforms tend to reinforce market segmen- wages from falling to very low levels. In tation, the results were not favorable (box many developing countries, however, the 4.4). For policy makers, a move from partial minimum wage is high. In Chile, Colom- reform toward more general and comprehen- bia, Costa Rica, Nicaragua, Panama, Peru, sive reform will benefit employment creation and República Bolivariana de Venezuela, with a disproportionate effect on youth. The it exceeded 50 percent of the median wage level of protection has to be balanced with for workers ages 26­40 between 1980 and the flexibility needed to encourage job cre- 2000.60 Yet many workers receive much less ation. In 1990, Colombia introduced a labor than the minimum wage because of weak market reform that substantially reduced enforcement. Young workers' wages tend to the costs of dismissing workers. The reform be concentrated at the lower end of the wage increased turnover for formal sector workers distribution anyway. For example, in Russia relative to nonformal but also reduced the the share of young workers with labor mar- length of unemployment spells, particularly ket earnings below 30 percent of the median for youth and for more educated workers.58 wage is estimated at 38 percent for 2002, in At any general level of protection, firms Indonesia at 35 percent for 2003, and in will need additional incentives to employ Ghana at 52 percent for 1998.61 Going to work 109 Several countries have differentiated Better design of wage subsidies and bet- the minimum wage by age to mitigate the ter targeting are needed to ensure that young effects on youth employment. Setting a hires do not gain employment at the expense lower minimum wage for youth reduces of other employees, and that the subsidy the disemployment effects by keeping it goes to those employers who would not profitable for firms to hire and train young have hired young workers in the absence of inexperienced workers.62 Below-minimum this additional financial incentive. Evidence apprenticeship wages significantly increased from the Czech Republic, Hungary, and the job opportunities for young graduates in Poland demonstrates that youth-specific Chile.63 Lower wages should be intended to wage subsidies can be of particular benefit if subsidize on-the-job training so youth can they are well targeted to the most disadvan- develop the skills needed to increase their taged, with females with lower educational productivity and future earnings. attainment benefiting the most. In OECD In many countries where the public sec- countries, wage subsidies work best for tor offers higher wages and more generous unemployed youth, especially those from benefits and employment security, educated more disadvantaged backgrounds.66 youth remain out of work while waiting for Differences in the target group, eligibility openings in the public sector. Closing the criteria, assignment to participation, type of gap in pay and benefits between the public jobs, and duration and amount of subsidies and the private sector, or at least reducing the will play a role in the effectiveness of these incentive to wait for a public sector opening, programs. These programs exhibit large will reduce youth unemployment among the variations. In Belgium, the "employment most educated. Closing the gap in informa- plan" offers two-year subsidies for employ- tion, promoting opportunities in the private ers through a reduction in social insur- sector, and aligning higher education more ance contribution, of up to 25 percent of with labor market demands will shorten the the gross wage in the first year and around long queues of young men and women wait- 17 percent of the gross wage in the second ing for public sector jobs. Governments also year of the subsidy. In Sweden, a program have to signal a willingness to move from targeting the long-term unemployed offers job guarantees (explicit or implicit) to a employment subsidies for six months. In more competitive process for entry into the the Slovak Republic, two-year subsidies were public sector, perhaps by making the appli- offered both in the public and private sector. cation process merit based. In Turkey, the Because wage subsidies are costly and less requirement of passing a set of examinations effective in economies with large informal to apply for a government post shortened sectors, their applicability is limited mostly the queue for public sector jobs and shifted to middle-income countries. Even in mid- graduates to the pursuit of other careers. dle-income countries, however, the financial constraints mean that programs have to be Direct employment creation. Wage subsi- well-targeted and be of limited duration.67 dies to private firms have sometimes encour- aged employers to hire new entrants.64 In a Expanding rural opportunities world where wages cannot adjust to com- In many developing countries, many youth "[Working in government] is stable, pensate for the risks firms might perceive still live in rural areas. For rural youth, with higher social status and more from hiring inexperienced young workers, employment opportunities are not only space for future development as a subsidies will have an effect on hiring and in agriculture but also off farm. Includ- government official." dismissal policies of employers. Several eval- ing rural towns, the rural nonfarm sector Jingxiao, university student, uations of wage subsidy schemes in European accounts for about 40 percent of full-time China countries are available.Almost all studies find rural employment in Asia and Latin Amer- a large beneficial impact on employment.65 ica and 20 percent in Sub-Saharan Africa.68 December 2005 While encouraging, these findings do not The history of economic development has usually take into account potential displace- shown that development of the nonfarm ment effects or deadweight loss that may be sector is tied to improved productivity on associated with wage subsidy schemes. the farm. As technological innovations raise 110 WORLD DEVELOPMENT REPORT 2007 including those in other countries (see chap- B O X 4 . 5 Off-farm opportunities for youth ter 8). So policies to develop the nonfarm in Palanpur, India sector will have a particularly pronounced effect on youth,even if not targeted at them.73 The nonfarm economy has expanded hard work and, in some cases, exposure to For example, the promotion of small and greatly in the North Indian village of Palan- health hazards. medium rural enterprises that use imported pur in the past decades. In the mid-1980s, Although employment outside Palan- more than a third of village income came pur is highly valued by villagers, particularly technologies could have a differential impact from nonagricultural activities, and more the young eager to venture beyond the vil- on youth, given their advantage in using new than 70 villagers were employed regularly lage, access remains limited.Why? Because technologies,as with Taiwan,China's promo- or semi-regularly in the nonfarm sector (of a of social status (proxied by caste and educa- tion of rural manufacturing. As subcontrac- working-age male population of about 250). tion levels), wealth (bribes need to be paid), Visits to the village in the 1990s and in 2005 and outside contacts (a"recommendation" tors, rural firms can acquire inputs, technical indicate that the expansion of nonfarm is often required). know-how, and links to external markets, occupations has not abated. Palanpur is located in socially con- increasing their attractiveness to young Many young male villagers from Palan- servative rural Uttar Pradesh, and outside pur hold semi-regular jobs in industrial workers.74 As education levels for rural youth employment opportunities are generally workshops and bakeries in the nearby confined to males. Elsewhere, nonfarm jobs improve, they can enter a broader range of towns of Chandausi and Moradabad. are also accessible to women. nonagricultural occupations. In rural India Employment contracts are often piecework, and in Brazil, better educated youth have a offering fairly high incomes in return for Source: Lanjouw and Stern (2006). higher likelihood of moving into highly pro- ductive nonfarm work.75 Some of these new productivity on the farm, labor is freed up off-farm opportunities involve physically to move to the nonfarm sector.69 demanding and hazardous work (box 4.5). The range of opportunities in rural As transportation and economic integration areas is far wider than might be apparent reach the small towns scattered throughout at first glance. The rural nonfarm economy rural areas, rural youth gain better access generates 30­50 percent of rural incomes to urban opportunities without the need to throughout the developing world, shares migrate. that continue to grow.70 In some instances, the high share is a result of crop failures Facilitating mobility or other adverse shocks to the farm sector. For 29 developing countries with data, youth In most cases, however, rising productiv- are 40 percent more likely than older people ity growth in the agriculture sector raises to move from rural to urban areas or to farm income and hence the demand for move across urban areas. Those who move goods produced outside agriculture. Rising have better employment outcomes, with agriculture labor productivity also frees up mobility correlated negatively with youth labor to work off farm. unemployment and positively with employ- The rural nonfarm economy, extremely ment and labor force participation.76 heterogeneous, provides an important The Chinese rural-urban migration typ- source of youth employment throughout ifies the pattern.77 Rural migrants tend to be the developing world. In Latin America, younger than 35. Two-thirds are male. Half about half the youth population ages 15­24 are single. Female migrants are younger and in rural areas, and more than 65 percent of less likely to be married. Rural migrants are those ages 25­34, work in nonagricultural more educated than rural residents who did activities. In 15 countries in this region, the not migrate: 66 percent had a lower second- higher share of youth employment in sev- ary school education,compared with 40 per- eral nonagricultural sectors compared to cent of rural residents who did not migrate. employment in agriculture bears out the Rural migrants are less educated than urban importance of the rural nonfarm econ- residents, and so tend to fill the least skilled omy.71 In rural India, the likelihood of mov- jobs available in cities. Even so, the incen- ing into nonfarm casual occupations peaks tives to migrate are clearly economic. Per at age 22, and in Brazil at around 33.72 capita urban incomes are more than three Because young people are the most times those in rural areas, and the gap is ris- mobile, they are the most likely to switch sec- ing. Even though rural migrants are paid tors to take advantage of new opportunities, half of what similarly skilled urban resi- Going to work 111 dents are paid, their pay is still well above Those who want to work on their own need that of workers in rural areas. Many rural information, mentoring, and credit. migrants send money back to their relatives in rural areas, representing 13 percent of Skill development beyond schools Chinese rural income in 2003. In middle-income countries with growing Rural migrants nevertheless face real industrial sectors, the bridges between school disadvantages relative to urban workers. and work, meant to address youth's lack of In China, migrant workers have little legal work experience and severe skill mismatches, recourse to compel payment, few receive take the form of formal apprenticeships or health insurance or pensions, and few have bringing work experience into the school good access to social services. China makes context. In low-income countries with lim- this explicit by the restriction on establish- ited formal schooling, traditional apprentice- ing legal residency in urban areas. These ships are more widespread and more likely to restrictions imply that total employment in provide the initial experience and skills youth the city is artificially smaller than its most need. In all countries, training systems must efficient size. Allowing employment to grow not only prepare youths for entry to work, to its peak efficient level could raise pro- but provide pathways for continual learning duction by as much as 35 percent.78 Rural over a lifetime in response to changing tech- migrants without residency permits have to nologies and global economic requirements. pay significantly more for education, health, Increasing the incentives for firms to train and other public services.79 The govern- and to reform training systems is essential. ment clearly recognized these issues, and is making policy changes to gradually delink Formal apprenticeship schemes. Germany's the residency permit system from its welfare "dual system" combines part-time schooling program, and to strengthen enforcement of with work and apprenticeship. Employer labor regulations.80 involvement ensures that the skills offered fit Restrictions on rural-urban migration, the needs of employers, reducing the likeli- not unique to China, show up in various hood of skill mismatches. The program has guises in many countries. Such restrictions wide coverage: more than half of all youth "Lack of experience is the main tend to depress wages in rural areas relative undertake an apprenticeship. As in Germany, barrier that young people face while to urban areas, whereas allowing population alternative school-based paths to qualifica- seeking employment, because most flows out of rural areas tends to raise wages tions in France, the United Kingdom, and employers prefer a few years of work for those remaining in rural villages, and the United States are associated with rather experience." benefits the country as a whole by improving selective improvements in early labor market efficient allocation of labor.81 Many youth experience.82 Overall, the strongest evidence Rahat, 24, Bangladesh also pursue opportunities to work overseas, favoring formal apprenticeships is the posi- whether they are from rural or urban areas. tive impact on employment for young men, Issues of international migration, including and on earnings for young women. country policies, are taken up in chapter 8. Do apprenticeships apply to developing countries? Probably not, in their current for- Choosing to work and mat,becauseof thesmallshareof employment in the modern wage sector, the slow growth of developing the skills to do so wage employment and jobs for new appren- Young people around the world ask, when tices, and the weakness of institutions.83 Some should I start to work? What kind of job developing countries have tried the dual sys- do I want? How do I look for it? How do tem, but with no clear pattern of success. The I get ready for work? Preparing youth for Mubarak-Kohl initiative in Egypt,launched to employment starts with general education introduce the dual system in 1995, illustrates that provides the foundation for later acqui- the challenges to starting such initiatives.Early sition of vocational skills. These vocational reports from the ILO indicated resistance in skills are acquired in the formal education the public education system and the absence system and beyond--in apprenticeship, of private sector umbrella organizations to work experience, and nonformal training. manage joint training courses.84 112 WORLD DEVELOPMENT REPORT 2007 To succeed, these programs need to move ing the quality and relevance of the training beyond the traditional craft and technical they could offer apprentices.89 The vouch- trades and provide more general content ers helped create a market for training that as a foundation for occupational special- encouraged new sources of supply and com- ization. This could reduce mismatches in petition. Providing literacy and basic educa- growing sectors, promote adaptability, and tion for apprentices and certifying their skills reduce gender bias. The United Kingdom's on completion also improve outcomes. Modern Apprenticeship program offers apprenticeships in nontraditional trades in Training by employers. Bringing work business administration, retailing, catering, experience into the schooling context can personal care, and information technology; improve the youth transition to work (chap- women constitute nearly half the appren- ter 3); at the same time employers provide tices. Australia's New Apprenticeships com- and finance training on and off the job long bine practical work and structured training, after youth exit school. As a source of skills leading to nationally recognized qualifica- for youths, employers are often overlooked tion in more than 500 occupations. It is not in favor of public training programs. Sur- clear how well these types of programs will veys conducted by the World Bank in 37 perform in low-income countries.85 countries covering 18,217 manufacturing firms show that enterprises are active train- Traditional apprenticeships in low-income ers.90 Nearly 60 percent of firms in East Asia countries. Formal apprenticeships in the and the Pacific provide training with the modern wage sector may be less relevant share falling to just under 20 percent in the in many developing countries, where self- Middle East and North Africa. employment and the growth of microenter- Leaving training to enterprises does not, prises in the informal sector have accounted however, ensure access for all to training. for an expanded share of employment over Enterprises often are less likely to invest in the past three decades. In Ghana, 80­90 per- skills widely used by other enterprises for cent of all basic skills training comes from fear of losing trained workers and their traditional apprenticeships, compared with investment. Thus, not all firms will train, 5­10 percent from public training institutions nor will all workers in enterprises be trained. and 10­15 percent from nongovernment In Colombia, Indonesia, Malaysia, Mexico, sources.86 Across West Africa, it is common to and Taiwan, China, large manufacturing find more apprentices than wage employees firms are more likely to train than smaller in informal sector firms.87 The strengths of ones.91 Kenya, Zambia, and Zimbabwe traditional apprenticeships, while not care- show a similar pattern, with manufacturing fully evaluated, are their practical orienta- firms employing 151 or more workers being "To avoid jobs like house servants tion, self-regulation, and self-financing. They twice as likely to invest in external train- and cleaners, youth programs can cater to individuals who lack the educational ing for their workers as those employing include training in other jobs like requirements for formal training (rural and 51 to 150, and more than 10 times as likely catering, carpentry, and building." urban poor), and they are generally cost- as those with firms of 10 or fewer workers. effective. However, they favor young men, Firms with a higher likelihood of training Jack, 16, Zambia screen out applicants from very poor house- are those that export, have foreign invest- holds, perpetuate traditional technologies, ment, and adopt new technologies--and and lack standards and quality assurance.88 they are more likely to train workers with Steps to strengthen traditional apprentice- more education.92 Other firms will need ship include improving literacy and the basic financial incentives to train young inexpe- education of apprentices, opening access to rienced workers. Policies can also condition new technologies, improving the pedagogi- the participation in other programs (like cal and technical skills of master craftsmen, wage subsidies for new entrants) on the and certifying skills attained. In Kenya's Jua provision of training for young workers. Kali, the informal sector, vouchers enhanced the access of master craftsmen to new tech- Technical and vocational education and nologies and upgraded their skills, improv- training. Because not all firms can pro- Going to work 113 vide training and because not all young employers and competency-based delivery workers benefit from employer training, by a mixture of public and private provi- public interventions are needed. Training sion, measuring performance in terms of offered by employers is relevant and effec- job placement and increased worker pro- tive, but that provided by the public sector ductivity. Curricula developed in a modu- is subject to question. Rigid, low-quality lar fashion promote flexible entry and exit training systems disconnected from labor for training consistent with a lifelong learn- markets have led many countries to reform ing model. Sound monitoring and evalu- their programs. Nonformal training sys- ation programs are important in guiding tems outside formal education are changing reforms, policy development, and market the way providers are governed, managed, operations. and financed (chapter 3). China, Chile, the Starting work on their own: Islamic Republic of Iran, the Republic of Korea, Malaysia, Mozambique, and Singa- Self-employment and pore, recognizing the fiscal limits of public youth entrepreneurs provision, have opened the doors to public- Many young people in the labor market private partnerships to diversify financ- work in businesses they have started on ing for training, promote sustainability, their own.95 Some are entrepreneurs by and improve access and relevance. In Latin necessity, others by opportunity.96 In Latin America, but also other regions, the roles America, 13 percent of those 16­24 are in of government as financier and provider of entrepreneurial activities, the great majority training (Servicio Nacional de Aprendizaje (12 percent) self-employed; only 1 percent or SENA, for example) are being reassessed are employers. The self-employed are gener- for national training agencies.93 Specialized ally less educated and poorer than employ- training agencies, responsible for train- ers. Women make up about one-third of ing policies and strategies, are assuming a the self-employed entrepreneurs and about larger role in policy development and man- one-quarter of the employers. agement of training expenditures instead Of the unemployed in Peru in 1998, 18 of provision. These are also opening more percent became self-employed by 2001, competition between public and private compared with only 6 percent in Nicaragua providers to improve quality and relevance (table 4.1). The higher self-employment in of the training offered. Peru explains part of the lower persistence In Mauritius, the Industrial Vocational of joblessness in Peru. About half the young Training Board has split the financing and provision of training and adopted a compet- Table 4.1 Employment transitions for youth (ages 16­30) itive model for procuring training services. Status in 2001 Argentina and Chile have similar national bodies. In Chile, the Servicio Nacional de Unemployed or Employee (%) Self-employed Employer (%) Status in 1998 inactive (%) (%) Capacitación y Empleo (SENCE), a spe- cialized agency of the Ministry of Labor, Peru maintains no capacity for the provision of Unemployed or 33 24 18 2 inactive training and instead procures training ser- vices from other public and private provid- Employee 7 28 13 1 ers for target groups. Competition promotes Self-employed 13 25 52 1 efficiency in delivery and more closely links Employer 9 9 55 27 training to market demands, shifting the Nicaragua financing model for training from supply- Unemployed or 60 18 7 0 driven to demand-driven.94 inactive Overall, training systems are moving Employee 25 59 23 17 away from a narrow focus on inputs for Self-employed 11 23 45 9 training, with more instructors, workshops, Employer 3 17 40 31 and equipment--to a focus on outcomes, Source: Llisteri and others (2006). with attention to skills standards set by Note: Not included in the table are students and those in unpaid family work. 114 WORLD DEVELOPMENT REPORT 2007 people self-employed in 1998 were still self- entrepreneurs with high-growth potential, employed three years later. Nine percent of frequently founded by young middle-class the self-employed in Nicaragua had become people. Universities, business schools, pri- employers within three years, but only 1 per- vate foundations, incubators, angel inves- cent in Peru. And only a third of employers tor networks, and, more recently, some in 1998 continued to be employers in 2001, governments provide direct support to the "I do not want to be a babshahi the majority becoming self-employed or entrepreneur--networking, incubation, [Bengali word for businessman, paid employed. All in all, these patterns sug- mentoring, and financing. For example, implying small businessman]. I gest that self-employment is a faster route Endeavor (a program in Argentina, Brazil, want to be a bijnizman [after the to paid employment, but if it persists over Chile, Mexico, and Uruguay) helps young English word, implying large scale the medium term, it is not likely to create ventures in a second round of growth business]." additional jobs. mainly through networking (with private In Latin American countries, about half investors) and mentoring. Male young person, Bangladesh of entrepreneurs felt motivated to strike out The public sector in some countries has on their own just after secondary and tertiary started to support entrepreneurs close to or school and during their first labor experi- just after start-up. The Umsobomvu Youth ence, using the knowledge, skills, and con- Fund is a development fund in South Africa tacts they had acquired.97 They come mainly created by the government to support access from middle- or upper-middle-class families, to information, skills development, and about half of them from families with at least financial support for people under age 35. one entrepreneurial parent. More than half Softex in Brazil, a public-private partnership, are graduates starting their business within targets university students in software, pro- two years of leaving university. This small viding training courses, technical assistance, group is responsible for a disproportionate and networking support.In Chile,a seed cap- part of the jobs created by new companies. In ital program led by CORFO (Corporación de Argentina, for instance, five years after their Fomento de la Producción) provides finance creation, about 6 percent of the new firms and technical assistance to entrepreneurs, are responsible for 60 percent of the jobs in operating a public-private partnership with survivor firms from that cohort.98 universities and incubators to identify and These entrepreneurs face several con- evaluate the most promising ventures and straints to creating and growing a venture: prepare them for seed capital. This program access to financing, to formal networks, and was replicated by the Buenos Aires Emprende to clients,suppliers,and skilled workers.The 1 and scaled up by Emprende 2 by large pub- Global Entrepreneurship Monitor indicates lic universities and trade chambers fostering only a very small share of these entrepre- software entrepreneurs. neurships are able to succeed. Young Latin American entrepreneurs face higher trans- Providing a springboard to action costs than those in East Asia. Most of them use their networks (mainly pro- reintegrate the most vulnerable duction networks of clients and suppliers Vulnerable young people--those who and social networks of family and friends) started work too early, never attended to overcome obstacles and make their ven- school, failed to acquire literacy, or never tures grow. General policies that enhance made it to the workforce--need a second the environment for doing business are not chance. Some disadvantaged youth--such youth-specific but are needed to facilitate as those with disabilities, ethnic minori- entrepreneurship in general. ties, and orphans--never had even a first However, because youth lack the net- chance. Providing them with the relevant works, experience, and collateral of adults, skills to enter or reenter the workforce they face additional constraints. Several reduces inequities in the labor market and new programs to promote entrepreneur- increases their productivity and ability to ship have been initiated in Latin America, break out of poverty traps. Because second but they are fairly new and have not been chances are costly, they have to be well- formally evaluated. They are targeted to targeted, designed to increase youth skills, Going to work 115 and geared to labor market needs. Because young people in need of second chances B O X 4 . 6 Employing youth with disabilities are usually vulnerable along several dimen- In addition to the usual challenges youth Experience from low- and middle-income sions, programs have to be comprehensive. face finding employment, disabled youth countries also demonstrates that disabled face a lack of access to jobs and employ- youth can be integrated into the labor mar- Second-chance programs ment centers because of stigma and other ket.In Egypt,Ethiopia,and Uganda,organiza- barriers.In particular, disabled people have tions are empowering disabled youths and are costly but needed often been denied an education: About one- their parents to plan for and pursue employ- Policies and programs for second chances third of all children not in primary school ment.In Pakistan many disabled youth are have a disability.In Brazil, while 55 percent employed at Independent Living Centers. are typically costly and rarely successful. of 18- to 19-year-olds are employed, only 29 The Salva Vita Foundation in Hungary Meager and Evans (1998) observe that "it is percent of physically disabled youths and has run a program since 1996 to integrate rapidly becoming conventional wisdom in 24 percent of mentally disabled youths have the disabled into the general workforce: the policy evaluation literature that labor jobs.This lack of education and employment · The Supported Employment Service sets them up for a lifetime of poverty.In Ser- market training and re-training schemes for assists in job placement, offers training, bia and Montenegro, 70 percent of disabled the unemployed have not lived up to expec- and helps solve problems at work. people are poor and only 13 percent have · The Employees'Club provides individual tations."A recent review of 19 programs, five access to employment.In Sri Lanka, over 80 and group follow-up for clients who have of them in transition and developing coun- percent of the disabled are unemployed. found employment through the Sup- Evidence from OECD countries shows tries, shows that training programs rarely ported Employment Service. that disabled youth can be quite productive improve the employment and earnings of given the right attitudes and supports, dou- · The Work Experience Program integrates young participants.99 The results underline bly important because disabled youth have employment into the school curriculum. greater difficulty recovering from an unsuc- the importance of having universal access to cessful school-to-work transition. Some first-chance policies and programs. In some OECD countries have instituted national Sources: Bercovich (2004); World Bank (2004c); cases, however, the costs of not intervening policies on school-to-work transition for Stapleton and Burkhauser (2003);Tudawe are overwhelming, and if well designed, sec- disabled youth. (2001); and www.salvavita.hu. ond chances could be cost effective. An estimated 8.4 million children are engaged in what international conventions call the "unconditional worst" forms of guidance in selecting the right model.101 child labor, which include child trafficking, Experience suggests, however, that interven- prostitution, and other forms of extremely tions need to require that youth are either hazardous work. About 10­12 percent of working or actively searching for work, pro- the population in developing countries is vide the skills relevant to integrate or rein- estimated to be disabled, and some evidence tegrate them into work, and be delivered suggests they are disproportionately poor.100 efficiently in response to local demand. Young people with disabilities, as well as Beneficiaries have to work or look for work. "Discrimination forms the biggest youth from ethnic minority groups, invari- Unlike other interventions, such as cash challenge given the social exclusion ably face more difficulties finding employ- transfers or unemployment insurance, of disabled persons from society. ment, despite the evidence that they can be interventions targeting youth need to Most of the disabled youth haven't productive given the right support (box 4.6). include either work or active search as accomplished their education due From an equity perspective, public interven- conditions to benefit from the programs. to lack of support, hence they have tion is needed to support the most vulner- Programs in the public sector that focus on a skill deficit [relative to] the job able and to offer them a second chance to the provision of public works to produce market." reintegrate into the workplace. For the very needed public goods and services pro- young, some second-chance opportunities Frederick, 23, Kenya vide good opportunities for young work- could reintegrate them into the education ers, particularly the low skilled and rural, system (chapter 3). For the large pool of low- to acquire initial work experience. Few skilled unemployed youth, a second chance evaluations have tested whether these pro- could help them move into productive work. grams improve the chances of participants to enter the labor market and enhance What might make employment in the private sector. One for successful programs? positive example is Argentina's Trabajar Not enough evaluations of youth employ- program, which had a significant impact ment interventions are available to provide on participants' current income.102 There 116 WORLD DEVELOPMENT REPORT 2007 is also some evidence of lagged gains from private sector employment by 13 percent- past participation. Among continuing par- age points. Interestingly, effects for older ticipants in this program, about half felt cohorts were insignificant, so the successes that it improved their chances of getting were confined to youth. a job, two-thirds that it gave them a mar- Public employment services should also ketable skill, and about one-third that it require youth to be active in job searching. expanded their contacts.103 In Korea, the Philippines, and Thailand, The African AGETIP (Agence d'Exécu- however, young people make little use of tion des Travaux d'Intérêt Publique) pro- state labor offices.106 In those countries, grams combine efforts to build public employment offices have been transformed infrastructure such as roads, buildings, and into one-stop centers giving job seekers sanitation systems, with efforts to provide access to job search assistance and place- jobs and training for unemployed youth. ment in vocational training. In Korea, how- Construction firms that get the contracts ever, only 5.8 percent of the unemployed also agree to use relatively labor-intensive found jobs through the public employ- practices to use local inexperienced youth ment services, and even that may overes- who receive training funded by AGETIP.The timate the impact because there were no youth are hired on a temporary basis,but the proper controls to measure the placement training and work experience are important rate for people not using the service. inroads to later, more permanent employ- An interesting recent example of private ment. Because the public works projects are involvement from Brazil is the First Job local, they can be targeted geographically to program started in Curitiba municipality, assist relatively poor, uneducated, or unem- financed by the local government. It aims ployed areas of the population. to link youth with firms in the municipal- The evaluation of the first seven years ity (no evaluation is available yet). Another of the AGETIP program in Senegal found promising public employment program is that the number of engineering firms more JobsNet, a quickly growing job-matching than tripled, the number of construc- agency in Sri Lanka.107 tion firms increased fivefold, and 35,000 person-years of employment were gener- Programs should provide the relevant ated. Unfortunately, governance can be skills. Comprehensive programs that an issue.104 Public works projects require provide training as part of a package that transparency and oversight to ensure that includes basic education, employment ser- the projects are targeted to the poor, that vices, and social services are more likely to only worthy projects are funded, that the have better success. Entra 21, a global effort money is used wisely, and that inexperi- intended to prepare 19,000 disadvantaged enced youth are trained. youth for jobs requiring information and Public works provide good targeting communication technology in 18 countries for other youth interventions that could in Latin America, placed at least 40 percent increase the likelihood of youth finding bet- of the targeted youth in employment. The ter employment opportunities beyond the programs offer a complete range of services, program. Argentina's Proempleo experi- including not just technical and life-skills ment in 1998­2002 tested mechanisms to training but also job placement services, help participants in the public works pro- internships, and advice in developing self- gram (Trabajar) find employment in the employment initiatives. A meta-analysis of private sector. It assessed whether wage sub- six Entra 21 projects revealed a higher than sidies and specialized training could assist expected job placement rate among 2,890 participants in the transition from workfare youths.108 Employment rates rose from to regular work. The wage subsidy increased 15 percent at the start to 54 percent 6­12 the probability of becoming employed in months later. Most jobs were in the formal the private sector by 9 percentage points sector, permanent, and paid the minimum for young participants under 30.105 The wage or higher. Although most youth opted wage subsidy and training programs raised for salaried employment, in several coun- Going to work 117 B O X 4 . 7 Joven programs increased employment and earnings for some disadvantaged youths The Joven programs offer comprehensive train- employment conditions before the program.In and adult females. In Chile one study showed a ing to unemployed and economically disad- Argentina 83 percent of participants, and in Chile negative impact on wages of -8.8 percent, led vantaged youths 16 to 29 years of age, aiming 57 percent were unemployed.Women were fairly by a reduction of wages in the formal sector. to improve their human and social capital and equally represented in Chile, while Argentina Subsequent analyses found a positive impact employability.The demand-driven model has had the lowest female participation (about 40 on earnings approaching 26 percent, strongly been customized throughout Argentina, Chile, percent).Targeting focused on 16- to 24-year- significant for youths 21 and younger. In abso- Colombia, the Dominican Republic, Panama, Par- olds, about 70 percent of all participants.109 lute terms the wage impact was higher for men, aguay, Peru, and República Bolivariana de Ven- but in a comparison of pre- and postprogram Employment ezuela.Technical training and internship experi- earnings, women had a slightly higher increase The programs increased the probability of ben- ences with employers are combined with basic relative to men. eficiaries finding employment upon graduation, life skills and other support services to ensure especially for women. In Argentina, the program Costs and benefits social integration and job readiness (see table). increased the probability of employment for With the given underlying cost per trainee and Private and public institutions--contracted young adult women (21 years and older) by the impact on employment and earnings, the through public bidding mechanisms--provide about 10 percentage points over a control net present value (NPV) of the program can the training and organize the internships. group. In Chile the program increased the prob- be calculated (given a discount rate, usually The programs target the poor, and more ability of employment 21 percentage points, assumed to be 5 percent).While costly, these than 60 percent of participants come from low- with strongly significant results for youths 21 programs in Argentina and Chile have positive income families.The highest education level and younger. NPVs, with a higher NPV in Chile compared completed by beneficiaries was secondary, with to Argentina. It is important to note that this significant participation by school dropouts Earnings calculation does not take into account the (50 percent in Chile Joven).Other targeting In Argentina the program increased monthly externalities from the program such as bet- criteria, such as employment, gender, and age, wages by about 10 percent over a control group, ter health outcomes and reductions in risky also applied.Most beneficiaries had precarious with results more favorable for young males behavior among participants. In this sense, the estimates are likely to provide only a lower Costs and impact of programs varies across countries bound of the NPV. It is also important to note that with the exception of forgone earnings, the Argentina Chile party incurring the direct costs (public funds) Proyecto Joven Chile Joven is different from the party benefiting from the program--the participants. Coverage (people) 116,000 165,000 Cost per trainee ($) 2,000 730­930 Sources: Aedo and Nuñez (2001); Aedo and Private benefits Pizarro Valdivia (2004); de Moura Castro (1999); Impact on employment (percentage point increase) 10 21 Elias and others (2004); Inter-American Develop- ment Bank (2005); and Santiago Consultores Impact on earnings (percentage point increase) 10 26 Asociados (1999). tries as many as a quarter set up their own advantaged youth (box 4.7). Their target- micro businesses. ing of low-income youths has improved Employers surveyed by Entra 21 value labor placement and earnings for their the combination of life skills and techni- beneficiaries across Latin America. Criti- cal skills developed by the program, rating cal to their success is the nature of the youths'life skills as satisfactory to highly sat- training--from technical to life skills and isfactory in meeting their companies' needs. from lectures to internships--and the More than 70 percent of employers said the sound support services and course cer- graduates' potential as workers was equal to tifications that foster youth's continuing or greater than that of other employees in participation. similar positions, and more than 90 percent in four projects rated graduates' overall per- Skill development should respond to local formance as better than or equal to that of demand and promote competition among workers in similar positions. providers. Among the important factors In Argentina, Chile, Peru, and Uruguay behind the success of the Joven programs the Joven programs have been widely are that the demand-oriented approach recognized as successful in reaching dis- fosters private participation--and that 118 WORLD DEVELOPMENT REPORT 2007 Table 4.2 Summary of youth employment policy directions and examples of programs Proven and successful Promising but unproven Unlikely to be successful Opportunities Creating jobs Trade openness: youth-intensive exporting Wage subsidies and private sector incentives, Schemes guaranteeing public sector jobs sectors (Indonesia and Vietnam) targeted to unskilled and unemployed (Hungary, for the educated (Morocco, Egypt, Sri Market-oriented reform (Slovenia and Poland, and the Czech Republic) Lanka, and Ethiopia) Estonia) Wage-setting institutions that compress General labor market reform (Colombia) wages (overly high minimum wages, Chile and Brazil) When minimum wages are too binding, lowering youth minimum wage (Chile) Overly restrictive employment protection laws (Chile, Latin American and Caribbean countries, and OECD countries) Partial labor market reform (reducing employment protection for youth only, or temporary contracts) increases youth turnover, but segments the market (France and Spain) Reducing child Conditional cash transfers for children labor vulnerable to child labor (PETI in Brazil, PROGRESA in Mexico, and Human Development Bond Project (BDH) in Ecuador) Facilitating Support for rural nonfarm employment Restrictions on rural migrants mobility (Taiwan, China's promotion of rural manufacturing) Capabilities Skills Apprenticeship programs: successful New apprenticeship programs (the United Slovenia's capitalization program (for development in Germany, unproven in developing Kingdom and Australia offering apprenticeships in entrepreneurs) countries new service sectors) Traditional apprenticeship with access to Traditional apprenticeships in the informal sector new technologies for master craftsmen (mixed evidence from Sub-Saharan Africa) (Kenya's Jua Kali program) Jobs Net, Sri Lanka matching agency Enterprise-based training (Ghana, Kenya, Training vouchers (Malaysia) and Zimbabwe) Reforming training institutes to introduce competition among private and public providers (Mauritius Industrial Vocational Training Board, Chile Servicio National de Capacitación y Empleo) Self-employment Self-employment assistance (Hungary and Poland) Promotion of entrepreneurship (Endeavor program in Argentina, Brazil, Chile, Mexico, and Uruguay) Softex (Brazil) public-private partnership for entrepreneurs Chile CORFO and Buenos Aires Emprende 1 and 2 Empowering youth with disabilities (Pakistan's Independent Living Centers) Second chances Bundled programs (comprehensive) Entra 21 programs including training, soft skills, providing training, placement, mentoring, internship (18 countries in Latin America and the and the like (U.S. Job Corps, Joven Caribbean) programs in Argentina, Chile, Peru, and Reintegrating youth with disabilities into the Uruguay) labor market (Hungary Salva Vita Foundation on Public works programs (in low-income supported employment and work experience countries targeted to the low-skilled and program) unemployed youth) AGETIP (Senegal) public works targeting youth Proempleo program in Argentina combines Colectivo integral de Desarrollo in Peru public works and wage subsidies Public employment services (Korea, the Philippines, and Thailand) Going to work 119 competition is promoted among training benefit more than adults. Other interventions providers. Transferring the Joven model are youth-specific, designed to fit individual to other developing countries requires country contexts: strong institutions to manage a decentral- In middle-income countries, reform- ized program and to coordinate the goals · ing labor market institutions to better and operations of training institutions and accommodate new entrants and pro- participating companies. It also requires viding financial incentives for firms to continual evaluation of the quality of the hire young workers will broaden youth courses and internships and the perfor- opportunities. Building more bridges mance of training institutions. between school and work and increas- ing access to information will enhance In sum, roadblocks on the way to work have the ability of youth to take advantage of implications for youth themselves and on these opportunities. development and poverty reduction efforts. · In low-income countries, building on Policy makers need to consider strategies to basic skills through better-designed delay youth from going to work too early, formal and informal apprenticeship to smooth the entry to the workforce, and will improve relevance to the needs of to allow movement toward better quality a changing labor market and facilitate work (table 4.2). In all countries, a bet- youth transition to work. A mix of public ter investment climate, a well-functioning work programs, wage subsidies, intern- labor market, and an expanding nonfarm ships, and training provides a spring- rural sector broaden employment oppor- board to reintegrate the most vulnerable tunities for everybody, and youth could back into productive employment. spotlight Do baby booms lead to employment busts? Not in OECD countries The post­World War II OECD baby boom offers lessons on how to absorb large youth cohorts into the labor market. What groups are most affected? What policies help youth most? Do bad policy choices cause more damage to youth employment prospects than large youth populations? Youth unemployment in OECD unemployment rates in the G-7 countries cation, compounding disadvantages related countries is due to weak have risen less than 2 percentage points. to discrimination in the labor market. In demand and bad policy Youth unemployment rates did rise France, where government statistics do and not the baby boom modestly as the baby boom cohort entered not recognize ethnicity, youth unemploy- the labor market. However, the unemploy- ment rates in predominantly ethnic urban One year after the end of World War II ment rates are affected much more by the enclaves are around 40 percent, nearly twice and for about 20 years thereafter, the G-7 overall strength of the labor market.1 Youth the already high French average. countries experienced a surge in the birth unemployment is extremely sensitive to rate relative to periods before and since. As the business cycle: youth benefit greatly High youth unemployment can a result, an unusually large share of youth when labor demand is increasing, but suf- cause youth and the country entered the labor force from the mid-1960s fer disproportionately when the economy is permanent harm until the early 1980s, with the peak occur- in recession or growing slowly. One lesson ring between 1967 and 1973 (figure 1). Does early unemployment cause perma- from the Organisation for Economic Co- Since then, the youth share of the labor nent scarring of youth, resulting in employ- operation and Development (OECD) coun- market has fallen steadily. ment difficulties later in life? Answers vary. tries is that even a modest deterioration in One might expect that youth in the In the United States, most studies find that the strength of labor demand, measured baby boom cohorts would have had much spells of unemployment after leaving school by the rising unemployment rate for older more difficulty finding work than would do not result in persistent unemployment workers in OECD countries, has increased the relatively small youth cohorts entering later in life. This corresponds to fairly high the difficulty in the transition from school the labor market in the 1990s. The opposite transition rates from unemployment into to work. is true, however: average youth unemploy- employment: 46 percent of unemployed ment rates across the G-7 countries are 7 youth are employed one month later. The less educated and minority percentage points higher than during the Corresponding transition rates in youth have the greatest problems peak of the baby boom. In contrast, adult France, Germany, and the United Kingdom with unemployment are much lower, ranging from 4 to 14 per- Are youth receiving too much education so cent, and more evidence indicates that early Figure 1 Youth unemployment rates for the that they become overqualified for the jobs unemployment results in persistent unem- G-7 countries are higher now than during the baby boom that are available? No. Evidence suggests ployment. One-third of the unemployed that the least educated face the greatest mis- in France have been unemployed more Unemployment rate Labor force match between skills and job vacancies. In than a year, compared with 8.5 percent in (percent) ratio (percent) almost every industrial economy, average the United States. The persistent adverse 25 30 unemployment rates fall as years of school- effects of early unemployment on later ing increase. employment stability can last seven years 20 27 Nor does job training tend to reduce in France, compared with two in the United 15 24 the disadvantage faced by the less educated. States.2 The degree of persistence appears The gap in access to jobs continues as the to respond to business cycles, with less per- 10 21 cohorts age because the most educated get manent damage from early unemployment the most job training. College graduates in economies experiencing job growth. In 5 18 in OECD countries are seven times more addition, the scarring effect of early unem- likely to receive training than are high ployment tends to be greatest for the least 0 15 school dropouts. Similarly dramatic gaps in educated and for disadvantaged youth. 3 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 access to training exist between the highest Weak youth labor markets tend to delay Unemployment rate: ages 15­19 (left axis) and lowest literacy groups. other transitions. In Europe, the average age Unemployment rate: ages 20­24 (left axis) Unemployment rates are uniformly at which youth leave the home has increased, Ratio of population ages 15­24 to higher for minority youth in OECD coun- especially in southern European countries. population ages 25­64 (right axis) tries. Such groups are atypically disadvan- In Italy, 80 percent of males ages 18­30 still Source: Authors' compilations based on data provided by taged by recessions and by policies that live with their parents, compared with 25 OECD Database on Labour Force Statistics. tend to limit new job creation. They are also percent in the United States. Across OECD Note: Shaded portion of the figure indicates peak baby boom entry years. atypically disadvantaged in completing edu- countries, the average age of marriage has Do baby booms lead to employment busts? 121 increased while the average number of chil- Figure 2 Employment protection hurt OECD ment protection, which protects their jobs dren per household has fallen. youth, but not adults, in 2003 at the expense of youth and other outsid- Weakening youth labor markets have ers. Perhaps that is why all but a few coun- Unemployment rate (percent) at least a partial role in explaining these tries have found it so difficult to relax the 25 changes in life transitions. Youth tend to employment protection, even when their delay leaving their parents' homes dur- 20 youth unemployment rates are so high. ing recessions. Differences in the relative strength of country youth labor markets can 15 Efforts to fix the youth labor explain observed differences across coun- market have mixed success tries in the average age of home leaving.4 In 10 Youth OECD countries have used various poli- Germany and Spain, the likelihood of leav- 5 cies to try to fix youth unemployment. ing home increases significantly with youth Adults One option that appears unsuccessful is to employment status and labor earnings.5 0 try to "make room" for youth employment When youth face constraints in access 0 0.5 1 1.5 2 2.5 3 3.5 4 by encouraging older workers to retire. to legal employment, they may engage in OECD employment protection legislation The limited evidence suggests that older illegal activities. Studies in the United States (EPL) index and younger workers may be complements and the United Kingdom show that weak- Source: Authors' compilations based on information pre- and not substitutes in production. Coun- ening wages for low-skilled youth are cor- sented in OECD (2004). Note: In the figure, youth are those ages 15­24; adults are tries with higher retirement ages for men related with increases in criminal activity.6 those ages 25­54. Correlation with EPL index is 0.32 for and women have higher employment rates Less consistent evidence links long-term youth and 0.27 for adults. Each data point in the graph rep- resents one country. for male and female youth.11 Similarly, youth unemployment with crime, although efforts in France to limit hours of work discontent with high rates of youth unem- to force firms to hire additional workers ployment in minority communities has younger workers (figure 2). In eras of both appear not to have resulted in appreciable been cited as a contributing factor to unrest. large and small youth cohorts, stricter EPL job growth. One recent study in France shows that cit- is correlated with higher youth unemploy- The average OECD country spends ies with higher youth unemployment have ment rates. around 2 percent of GDP on active labor higher rates of burglaries, thefts, and drug Similar findings hold for legislation market policies, with training being the offenses.7 that diminishes wage flexibility, such as largest component of those expenditures. high minimum wages or extending union- Public expenditures are only about one- Efforts to protect job security do negotiated wages to nonunion workers. tenth of the total, however, and private not help and may hinder youth Such policies reduce wage inequality across training is weighted heavily toward the The youth unemployment problem appears workers, but risk making it too expensive to most educated. Private training is unlikely to be exacerbated by policies aimed at pre- hire those lacking schooling or prior labor to offer a significant second-chance option serving jobs. Many countries have enacted market experience. for those who failed to attain a sufficient Employment Protection Legislation (EPL) Many of the countries with the stron- level of prior education. Publicly subsidized that makes it more difficult or costly for gest youth labor market outcomes over training tends to have the greatest success firms to lay off workers. These policies are the past 15 years (Ireland, the Nether- with more-educated recipients. designed to insure workers against income lands, New Zealand, the United Kingdom, The experience of youth training pro- loss from fluctuations in labor demand. the United States) have tended to be those grams in Europe suggests that they have However, such legislation also makes it with rising wage inequality. The implica- improved the transition to employment more costly for firms to hire workers, thus tion is that wage flexibility has helped these but that the impact on earnings is more stricter EPL tends to depress the rate of new economies to adjust to shocks and to cre- mixed.12 job creation.8 ate new job opportunities for youth, but at Of other active labor market policies, These adverse consequences of EPL a cost of increased income disparities in the job search assistance and wage subsidies are borne mostly by groups that are dis- population.10 appear to be the most promising for raising proportionately first-time job seekers, so Countries with stronger EPL have expe- employment rates of disadvantaged youth, youth tend to be atypically disadvantaged.9 rienced growth in temporary and fixed- but public employment programs have not Because EPL appears to retard new job cre- term jobs that are frequently exempt from worked. Evidence also suggests that youth ation, it can also heighten the persistent firing restrictions. This allows new job are more successful in transitioning to effects of early unemployment on employ- growth, but it creates dual labor markets employment in countries where unemploy- ment prospects later in life. with protected jobs held predominantly by ment benefits are conditioned on active job Stricter employment protection legisla- "insiders" (older male workers) and tem- search and willingness to accept jobs when tion does not appear to create unemploy- porary jobs held by "outsiders" (women, offered.13 ment problems for older workers and may minorities, and youth). Insiders have an even insulate them from competition with incentive to maintain and expand employ- Growing up healthy 5 If death rates are the benchmark, the young puts their health at risk. Awareness of the are a healthy group, and today's young are consequences of decisions for health, and of healthier than at any time in history. This ways to avoid ill health, is very low among c h a p t e r presents an unprecedented opportunity for young people, especially girls, and only a further investment and growth.Young people small percentage of those aware actually in developing countries have a mortality rate adopt safe behavior. Young people are thus of less than 3 percent, down significantly over likely to make uninformed decisions about the past 20 years, and a fraction of that for behaviors that put their health at risk. infants and adults.1 Risky behavior during youth can deplete Average mortality is a misleading mea- productive human capital many years into sure, however, because it does not reflect the future. Long after tobacco smoking the behavior that puts health at risk later in peaked in the United States, tobacco was the life. A more appropriate benchmark would single largest cause of all lung cancer deaths, reflect such behaviors as tobacco use, drug and about half of those who died were still use, excessive alcohol consumption, sexual in middle age. In some developing coun- behavior that increases the likelihood of sex- tries today, close to half of all young men ually transmitted diseases, and inadequate are smokers. Similarly, HIV develops into diet and physical activity. These behaviors AIDS with a lag of up to 10 years, taking its affect youth while they are still young: for toll on people in their prime working ages. example, unprotected sex can lead to HIV In many developing countries, new HIV infection or an unplanned pregnancy. How- infections affect young people dispropor- ever, most of the adverse consequences tionately. The costs of treating AIDS and show up only when they grow older, in such such chronic diseases as cancers, diabetes, noncommunicable diseases as lung cancer, and heart disease are high, and the treat- diabetes, and heart disease. ments often ineffective. Good health is not equally available to all The best way to avoid the future loss young people. Although mortality among of productive human capital and steep young people is low on average, and young increases in future health care expenditure people are fairly healthy, their likelihood of is to modify health behavior during youth, premature death is much higher in poor when habits are still being learned. Policies countries. The average 15-year-old boy has to promote better health for young peo- a 90 percent chance of surviving to the age ple rest on three legs. First, give them the of 60 in Western Europe or North America, knowledge to help them make informed but only a 50 percent chance in Sub-Saharan choices about their behavior--and the skills Africa, primarily due to the spread of com- to negotiate safe behavior with peers and municable diseases such as AIDS. In coun- partners. Second, create an environment tries hardest hit by AIDS, the probability is for the young to practice healthful behavior, only 20 percent.2 making risky behavior costly, and limiting Young people today have access to a the opportunities for it. Third, for young much broader range of choices than pre- people harmed by poor health decisions viously available, in a vastly different envi- or environments, provide health services, ronment, which makes it harder to choose treatment, and rehabilitation. Broadening appropriately and to avoid behavior that access to these services--whether dealing 122 Growing up healthy 123 with unwanted pregnancies,obesity,or drug Figure 5.1 In the United States, mortality from lung addiction--will minimize the long-term cancer among men increased dramatically for nearly 40 years after smoking peaked consequences and lead to better health. The health concerns facing young peo- Deaths per 100,000 males 80 ple differ greatly around the world. As the Global Burden of Diseases project shows, 60 leading causes of death and disability for young people include injuries in Latin America and HIV/AIDS in Africa.3 This 40 Lung Oral cavity and pharynx chapter focuses on the factors that are com- Esophagus mon to all youth health issues. Rather than 20 Larynx focusing on specific health outcomes, the chapter presents a framework to develop 0 1950 1955 1960 1965 1970 1975 1980 1985 1990 policies that encourage healthier behavior Source: Centers for Disease Control and Prevention (1993). among youth. Note: Deaths per 100,000 population are standardized to the 1970 age distribution of the U.S. population. Promoting the health of young people stimulates just when they are entering the workforce. It can make a job search more difficult, and growth and reduces poverty being unemployed can make drug abuse and health care expenditures more attractive. Unemployment and illicit Although the death rate for young people drug-taking are strongly correlated, in both ages 12­24 is less than 3 percent, and young developed and developing countries. In people are generally healthy, their continu- Colombia, the prevalence of cocaine use ing health and survival into adulthood are was 4.1 percent for the unemployed, 0.4 at risk, largely because of their behavior as percent for the employed.6 In Russia, severe youth. It has been estimated that nearly two- alcohol consumption is associated with a thirds of premature deaths and one-third of higher probability of job loss.7 Countries thetotaldiseaseburdenof adultscanbeasso- with higher rates of alcohol consumption ciated with conditions or behavior begun among youth also report higher rates of in youth.4 Policies that encourage healthful motor vehicle deaths and suicides among behavior among youth, by improving their youth. More than half of deaths due to productivity and health as adults, will have homicide or traffic accidents in South ripple effects on the economy. Africa and Brazil had blood alcohol levels in excess of legal limits.8 Traffic accidents Impact on poverty are among the leading causes of death reduction and growth and disability in developing countries. Riskyhealthbehaviorduringyouthcandeplete Accidents are estimated to cost low- and the economy of productive human capital for middle-income countries $65 billion per many years into the future. The prevalence year, or between 1 and 1.5 percent of gross of smoking among U.S. males peaked before national product.9 1945,but since then,with smoking rates more Safe health behavior, by contrast, can or less unchanged, the deaths attributed to encourage greater investment in produc- lung cancer increased nearly fourfold, and tive human and physical capital. Longer life among smokers twentyfold (figure 5.1). Some expectancy may lead to higher investments developing countries, in which half of young in secondary school, just as higher mortality men are smokers, face the burden of enor- due to HIV/AIDS may reduce the gains from mous health costs in 20 or 30 years. investments in children accumulated during Heavy alcohol consumption and drug the past generation. HIV and AIDS reduce use reduce productivity and increase savings and investments in productive phys- absenteeism and other health-related costs ical capital among the poor, thus reducing to firms and individuals.5 Drug abuse is the likelihood that poor youths escape pov- concentrated among 18- to 25-year-olds-- erty.10 African parents have responded to the 124 WORLD DEVELOPMENT REPORT 2007 higher mortality risk from HIV/AIDS by 0.03 percent and 0.40 percent of GDP--costs having more children and providing each of expected to rise as young smokers age and them with less education.11 begin to suffer the consequences.18 Caring for AIDS patients is squeezing the Impact on health care expenditures resources and care available to HIV-nega- Addictions, exercise patterns, and eating tive patients. In South Africa, patients are habits are hard to change, harder as peo- turned away from hospitals because of lim- ple mature and become set in their ways. ited beds,19 and Kenya has seen higher mor- A longitudinal study from Indonesia finds tality among HIV-negative patients.20 From that the proportion of men who smoked in 1988 to 1992, the average number of people 2000 was almost identical to the proportion admitted per day to a Nairobi hospital who who had ever smoked since 1993, suggest- were not infected with HIV fell by 18 percent ing that very few smokers quit.12 Modifying whilethenumberof HIV-infectedmorethan health behavior during youth, when hab- doubled.21 Although there is no evidence its are still being learned, is a much more that health workers risk infection more than effective means of preventing unsustainable the general population, the fear of infection, increases in health care expenditures. Treat- plus the greater demand for health care and ment, especially for such noncommunica- the overwhelming tide of dying patients, has ble diseases as cancers, diabetes, and heart accelerated their burnout.22 disease, is expensive and often ineffective. Public intervention is needed In adulthood, when adverse health conse- quences become visible, both the behavior to promote youth health and the consequences are irreversible. Governments around the world try to Consider the consequences. Smoking reduce risky behavior--through tobacco increases the risk of general health prob- control; regulation of alcohol consumption; lems, the susceptibility to severe respira- and public messages about diet, nutrition, tory illnesses, and the risk of dying from and safe sex. Such public intervention is jus- lung cancer.13 Excessive alcohol consump- tified because of market failures that domi- tion leads to greater alcohol dependence, nate health behavior and cause individuals more alcohol-related injuries, and other to make privately and socially inferior health psychological and physical problems later decisions, and these market failures are in life. Substance abuse can lead to addic- magnified for young people. Private markets tions; to circulatory, respiratory, and diges- do not facilitate optimal decision making by tive diseases; to accidental overdose; and to individuals for a number of reasons. One the greater risk of acquiring HIV and AIDS. problem is that individuals have incomplete Poor nutrition and a lack of exercise can information about the consequences of lead to obesity, which causes hypertension, their behavior over time. They do not know cardiovascular diseases, type-2 diabetes, whether or when they might face the adverse and many other chronic diseases in youth consequences of their behavior. Unprotected and adulthood.14 Early sex, unprotected sex, sex and excess tobacco and alcohol con- and having multiple partners can lead to sumption are risky, but the risks are difficult HIV/AIDS and other sexually transmitted for young people to quantify (see boxes 2.7 diseases. Almost all HIV infections in 2001 and 2.9). They might find risky behavior in Africa and parts of Latin America, and attractive or convenient in the short run, or about a quarter in Eastern Europe, can be believe that experimenting with such behav- attributed to unsafe sex.15 ior is safe, or be coerced into engaging in it. Now consider the costs. Providing health Even if the consequences of poor envi- care to a drug addict costs about 80 percent ronments or decisions are purely private, more than providing it to an average person public action can still be justified--on the in the same age group,16 and by one estimate, grounds of merit or concern for equality. tobacco use results in a global net loss of $200 Good health is often accepted as a merit billion a year.17 Individual country studies good: something socially accepted as benefi- estimatethenetcostof tobaccouseatbetween cial for people, regardless of their feelings in Growing up healthy 125 the matter. Equity also matters, because the age at menarche (first menstruation) has poor are likely to be less healthy than the declined for girls, and the average age at nonpoor. This is most obviously manifest in marriage is increasing.25 A study based on differences in life expectancy. In Brazil, the 27 Sub-Saharan African countries shows life expectancy for young men between the that this rise in age at marriage is linked to ages of 10 and 19 from the poorest quintile is the increase in the percentage of young peo- 38.0 years, and for young women 35.8 years. ple who engage in premarital sex.26 For young men from the wealthiest quintile, Imperfect knowledge about conse- it rises to 49.7 and for young women to 53.1. quences can lead people to engage in unpro- Poverty cuts 12 years from the expected life tected sex. Fewer than half of sexually active of a young man, and more than 17 years young people use condoms, even though from the life of a young woman.23 unprotected sex is the greatest risk factor The rest of the section describes only for HIV transmission in most areas of the some of the risky health behaviors--unpro- world (figure 5.2). Even in countries where tected sex, tobacco use, drug use, excessive HIV prevalence is high, a large proportion alcohol consumption, physical inactiv- of young people engage in unprotected ity, and unhealthy diet--that arise from sex. These young people are at greater risk the market failures described above. While of HIV infection. In Mozambique, a coun- governments intervene in these areas in try with moderately high HIV prevalence, many ways, these measures must be bet- sexual activity among youth is common, ter designed to address the unique market but condom use is low. The share of sexu- failures young people face when making ally active boys using condoms ranges from choices about their behavior. 20 percent in Mali to about 50 percent in Zambia. Condom use is higher among Health behavior, imperfect unmarried sexually active girls than among information, and inequality married girls. In Uganda, slightly more than Individuals may not have good information 50 percent of all unmarried girls used con- about the risks they face today, even if they doms the last time they had sex, compared understand the risks over time. Some infor- with only 3 percent of married girls. mation, such as HIV status, is easily withheld Condom use may be lower among from partners,and during the early and more young married women than unmarried virulent stages of many infections, infected women because they are planning to bear persons may be unaware of their true status. children. But in Burkina Faso, Kenya, and These external risks are also present where Figure 5.2 Sexually active youth are unlikely to use condoms, even where HIV prevalence is individuals suffer from the behavior of oth- high ers over which they have no control. The most obvious case is second-hand tobacco Percent Percent smoke,but other ostensibly private behaviors All boys (left axis) can have consequences for others, especially Married girls (left axis) if the costs of later treatment are passed on to Unmarried girls (left axis) 60 HIV prevalence (right axis) 30 the wider community. Imperfect information and risky sex. More 40 20 than half the young in many countries are sexually active, and data from surveys con- 20 10 ducted between the late 1990s and 2004 show that the proportion who initiated sex- 0 0 ual activity before the age of 15 is increas- ic ing.24 A significant proportion of youth in Haiti Faso Mali Benin Zambia Malawi Kenya Uganda Ghana Bolivia Republ Cameroon developing countries--especially girls--are Zimbabwe Mozambique Burkina sexually active within marriage or informal Dominican unions, but many unmarried youth are also Source: Demographic and Health Surveys for surveys conducted between 1997 and 2004. sexually active. Evidence indicates that the Note: Left axis of figure refers to 15- to 24-year-olds who reported using a condom at most recent sexual intercourse. 126 WORLD DEVELOPMENT REPORT 2007 Zambia fewer than a third of married girls of experimentation; this experimentation who did not use condoms were planning a is partly intended to acquire information pregnancy within two years--the rest were about behavior, choices, and consequences, trying to avoid pregnancy.27 Unprotected as well as to form a sense of identity and sex increases the risk that married young belonging. People, young and old, choose girls will become infected, and recent data behaviors because of the pleasure and ben- from Sub-Saharan Africa show that HIV efits they yield. The pleasure from some of incidence is growing fastest among young these behaviors is fleeting, while the costs married women. This is primarily because can persist. Experimentation can lead to younger women are married to older men, habits and addictions,which can be destruc- who have a higher chance of being infected tive and extremely difficult to break. (through risky sex with partners outside Preferences may be time-inconsistent, marriage).28 One study in rural Uganda and behaviors can have irreversible conse- found that the HIV infection rate among quences. In the future, today's youth will married women under 20 was nearly three most likely wish they had made different times that of unmarried women under 20 decisions when they were young, especially (17 percent compared with 6 percent).29 if they begin to suffer the consequences. For many of these adverse consequences, it is Unequal power and risky sex. Risky sexual not possible in later life to undo the dam- behavior is more likely to occur among poor age caused by earlier behavior. Treatment, youth,whoareinaweakerpositiontonegoti- especially for such noncommunicable dis- ate safe sex, and are more likely to experience eases as cancer, diabetes, and heart disease, coerced sex and sex for exchange.30 Forced is expensive and often ineffective. sex exposes young women to the risks of HIV and other sexually transmitted diseases, Alcohol, tobacco, and drugs. Alcohol is risks heightened by injuries from physical the most widely consumed drug in the violence.31 More than 20 percent of women world: about half those 15 and older have attending antenatal clinics in Soweto, South consumed alcohol in the past year.36 Pat- Africa, reported having had sex with a "non- terns are difficult to interpret, because primary"male partner in exchange for goods moderate drinking--even by youth--is or money.32 Data from DHS surveys around accepted in many countries. The propor- the world indicate that 13 percent of unmar- tion of young people who report drinking ried women between the ages of 15 and 19 generally exceeds 60 percent, of whom 10 received money or gifts in exchange for sex in to 30 percent engage in binge drinking (fig- the four weeks preceding the survey.33 ure 5.3).37 In the United Kingdom, young The "sugar daddy" phenomenon is people between 16 and 24 are the heaviest widely observed in Africa and other set- drinkers in the population, and the least tings. A survey of 45 studies in Sub-Saharan likely to abstain from drinking.38 Limited Africa reports that sex with older unmar- data from developing countries suggest that ried partners is widely accepted among young people are beginning to drink alco- adolescent girls in many countries.34 There hol at earlier ages. Boys are more likely than is some evidence that the HIV epidemic has girls to drink alcohol, and to drink heavily, increased the incidence of sex between older though consumption among girls in some men and younger women, as men seek to countries (especially in Latin America) has avoid infection.35 Women who report trans- begun to approach or even surpass that of actional sex, controlling for age and num- young men. ber of partners, were 50 percent more likely Early initiation of alcohol use is corre- to be HIV-positive. lated with a greater likelihood of both alco- hol dependence and alcohol-related injury. Health behavior, habit formation, A study of hospitals in three cities in South and irreversibility Africa found that 61 percent of patients Young people lack information partly admitted to trauma units in these cities because they lack experience.Youth is a time were alcohol-positive, including 74 per- Growing up healthy 127 Figure 5.3 Alcohol consumption is common among Figure 5.4 Many young people, especially young men, consume tobacco 15- to 19-year-olds in some countries Indonesia Vietnam Drinking alcohol in past year Nepal Binge drinking at least three Russian Federation times in past month Armenia Poland Mexico United Kingdom Kenya Denmark Zambia Norway Uganda Females Males Slovenia Haiti Czech Republic 0 10 20 30 40 50 60 70 80 Russian Fed. (Moscow) Percent Croatia France Sources: Indonesia Adolescent and Youth Survey (2002­03) of 15- to 24-year-olds; Mexico current smok- ing prevalence is based on survey of urban 14- to 22-year olds; Vietnam Ministry of Health and General Cyprus Statistics Office, UNICEF, and WHO (2005) surveys of 14- to 25-year-olds; Russia Longitudinal Monitoring Hungary Survey (Round 6) (15- to 24-year-olds). Remaining data was obtained from MEASURE Demographic and Bulgaria Health Surveys conducted between 2000 and 2004 (15- to 24-year-olds). Ukraine Note: Figure refers to youth who report currently smoking cigarettes or pipes or using other tobacco Greece products. Lithuania Slovak Republic Few young people experiment with ille- Portugal gal drugs, and an even smaller number go Romania on to develop long-term chronic problems. 0 20 40 60 80 100 Even so, measures to prevent experimental Percent use are worthwhile to avoid addiction and Sources: Computed using data from Hibbell and others (2000) and the acute and possibly fatal reactions with Bloomfield and others (2003). Note: "Binge drinking" is defined as consuming five or more drinks even limited experimental use. Young peo- in a row. ple in developed and developing countries experiment with cannabis, amphetamines, cent of violence cases, 54 percent of traffic cocaine, heroin, and inhaling solvents, glue, collisions, and 30 percent of trauma from other accidents.39 Young people who abuse and gasoline. Inhaling volatile chemicals, relatively neglected by policy makers, is alcohol and drugs are more likely to com- extremely dangerous, and acute intoxica- mit crimes, and substance abuse is a major risk factor in violence.40 Examination of tion can be fatal. Young people are more likely to abuse solvents because they are eas- 960 people arrested in nine police stations ily available in homes and shops, and street in three cities in South Africa found that 22 children are especially vulnerable (box 5.1). percent were under the influence of alcohol when the alleged crime took place.41 Per capita consumption of tobacco is B O X 5 . 1 Street children abusing drugs declining in developed countries, but ris- ing in many developing countries, for both The World Health Organization's Substance hunger.There are no direct comparisons men and women. Between 1970 and 1990, Abuse Department identifies inhalation of among non-street youth; but a recent study volatile substances as a particular problem of Egyptian university students found that tobacco consumption is estimated to have of street children.It recommends that 7 percent had ever tried cannabis,and 18 increased by about 3.4 percent a year in low- prevention and interventions are urgently percent had ever used solvents. and middle-income countries,42 and people needed to deal with the almost universal A study of street children (ages 15­16) are beginning to smoke at younger ages.43 use of organic inhalants among street chil- in Karachi,Lahore,Peshawar,and Quetta in dren in developing countries. Pakistan found that of those who had used Reported use of cigarettes, pipes, and chew- Cairo has a large and rapidly growing drugs,90 percent inhaled glue,gasoline,or ing tobacco varies widely (figure 5.4). Most population of street children (150,000 in paint thinner,all easily available from the smokers in Indonesia consume clove ciga- 2001).Nearly two-thirds of those surveyed local market.Nearly two-thirds of them rettes, which contain twice the tar, nicotine, regularly abuse drugs or solvents.Of those reported having used these substances for who consumed illicit substances,97 percent more than two years.Roughly three-quar- andcarbonmonoxideofAmericancigarettes, reported sniffing glue.Other substances ters had never been to school.They beg, and smoking among 15- to 19-year-olds rose include cannabis,hashish,solvents,and clean cars,and scavenge through garbage, from 32 percent in 1993 to 43 percent in prescription medication.They take drugs spending half what they earn on drugs. 2000.44 Fewer girls than boys report tobacco because of peer pressure,to relieve the Sources: Refaat (2004); United Nations Office on pressures of the street,to help them sleep, Drugs and Crime (2004); UNODCCP (2002); and use, though it may be increasing among girls and to help them endure pain,violence,and WHO (1999). in developing countries.45 128 WORLD DEVELOPMENT REPORT 2007 The prevalence of illegal drug use is Figure 5.5 Many young women in developing highest in developed countries but increas- countries are overweight ing in developing countries. In many Arab Rep. of Egypt regions, especially Central Asia, prevalence Turkey now approaches developed country levels.46 Bolivia Nicaragua (Estimates of drug abuse by young people Colombia are available only from a few small studies, Cameroon mainly for school students.) There are an Armenia Zimbabwe estimated 13 million injecting drug users Morocco worldwide, 78 percent of them in develop- Haiti ing and transition countries, the majority Kenya young.47 Potentially deadly in itself, inject- Ghana Uganda ing drug use increases the risk of acquiring Kyrgyz Republic HIV through the sharing of infected needles Benin and the exchange of body fluids. Zambia Kazakhstan Overweight Mali Underweight Diet and exercise. More sedentary life- Malawi styles, along with high intakes of salt and Burkina Faso India saturated fats and low intakes of vegetables Nepal and fruits, can lead to obesity, high blood 0 10 20 30 40 50 pressure, high blood cholesterol, and such Percent noncommunicable diseases as heart disease Source: Demographic and Health Surveys conducted between and diabetes. The consequences of poor diets 2000 and 2003. Note: Anthropometric measurements were taken for all married can be passed from generation to generation: and unmarried women interviewed. BMI-for-age cutoffs from a ref- babies born to malnourished mothers are at erence population were used to classify 15- to 24-year-old women as overweight (above the 85th percentile cutoff) and underweight significantly higher risk of being overweight (below the 5th percentile cutoff). or obese in adulthood.48 The health consequences of poor diets cant sex-based differences in patterns of diet and physical inactivity have generally been and exercise. In urban Mexico, only a third of considered diseases of the affluent. How- young girls report exercising, compared with ever, these noncommunicable diseases, far more than half of all boys.53 from being affluent-country lifestyle dis- The health consequences of poor diets eases, appear with greater frequency in pop- and lack of exercise are becoming more ulations undergoing rapid socioeconomic prevalent in many developing countries. improvement with better access to food and In large cities in China, the prevalence of shifts in diets.49 Data from China, the Arab hypertension among those 18 years and Republic of Egypt, India, Mexico, the Phil- older increased from less than 12 percent in ippines, and South Africa reveal a marked 1991 to 19 percent in 2002, and the preva- shift over the last 20 years toward diets high lence of type-2 diabetes increased from 4.6 in saturated fat, sugar, and refined foods, percent to 6.1 percent. In Egypt and Mexico, while the share of cereals, legumes, pulses, the prevalence of diabetes has been esti- and nuts remained stable or declined.50 mated at 10 percent.54 Being overweight is rapidly becoming Nutritional habits and outcomes are sta- more prevalent in low- and middle-income ble over time, so there are great benefits to countries, where incidence is increasing intervening during youth when habits are especially rapidly among poor households.51 being formed. Longitudinal data in Guate- A recent survey from rural Mexico of low- mala show that being overweight or obese income households found that 60 percent of as a youth significantly raises the chance of adult women and more than 50 percent of becoming obese in adulthood (figure 5.6). adult men were overweight.52 In many coun- tries young people are more likely to be over- weightthanunderweight,withgirlsespecially In many ways the risks facing young people at risk (figure 5.5). This partly reflects signifi- today are greater, and consequences poten- Growing up healthy 129 Figure 5.6 Being overweight as a youth increases Figure 5.7 The proportion of 15- to 24-year-olds infected with HIV is high, the chances of being overweight as an adult especially among girls Percent Zambia (2001/02) 100 Kenya (2003) Cameroon (2004) Overweight as a youth Tanzania (2003/04) 80 Not overweight as a youth Mali (2001) Burkina Faso (2003) Ghana (2003) 60 Female Male Dominican Republic (2002) 0 2 4 6 8 10 12 40 HIV prevalence (percent) Sources: Demographic and Health Surveys and Tanzania HIV/AIDS Indicator Survey. 20 are delaying sex, and this has resulted in a 0 Male Female Male Female drop in HIV incidence.56 A few evaluated Overweight during Obese during programs suggest that providing culturally adulthood adulthood appropriate teaching about health risks and Source: Personal communication with Alexis Murphy, International Food Policy Research Institute, 2005 (using INCAP Longitudinal increasing the capability of young people to Study, Guatemala, 1987­2004, IFPRI). practice healthy behavior (including negoti- Note: Overweight is body mass index (BMI) greater than 25 and obese is BMI greater than 30, such that individuals who are obese ating safe sex with partners) are more likely are also considered overweight. to change behavior. Providing young people with school- tially more deadly, than for previous gen- ing can be an effective way to change their erations. This is most obvious in the spread behavior. Evidence shows that there is a of HIV and AIDS, increasingly prevalent strong relationship between education and among young people (figure 5.7), and dra- health. Healthy decisions are promoted by matically increasing the risks of unprotected education and economic growth, which sex. Today's young people also have access to raises the prospect of higher lifetime earn- more calorie-dense foods, which, combined ings and a better life in the future. A stron- with more sedentary lifestyles, are making ger sense of an attainable and prosperous young people obese and leading to hyper- future can encourage individuals to make tension, heart disease, and diabetes. Adding good health and survival more likely. Edu- to this is the aggressive marketing of tobacco cation, often called a "social vaccine," is products and the increase in smoking around considered by many to protect young peo- the world. So the broader opportunities open ple from engaging in risky behaviors (box to young people make it harder to choose 5.2).57 Policies that improve young people's appropriately and to avoid risky behavior. "We are aware that we drink alcohol access to education can also be effective and that we consume drugs and we behavior-change policies. Chapters 3 and 4 Strengthening young have sexual relations with different discuss ways in which young people's access people's capability to girls. We want to change but we do to education and labor market opportuni- not have a great deal of will to do practice healthy behavior ties can be improved. In this chapter, we that." A central element of health promotion is focus on investments in improving health providing health education to change youth behavior, of which formal health education Young male gang member, behavior and encourage adoption of healthy is only one--albeit important--determin- Comas, Peru behaviors. However, behavior change is one ing factor.58 January 2006 of the most difficult goals to achieve in health promotion. In recent years, experi- Information is necessary ence with health education, particularly to change behavior in the context of HIV prevention, shows Many countries have health education pro- changing knowledge alone may not change grams, especially for school-going youth. behavior.55 But evidence from Uganda and However, evidence of the efficacy of school- Eastern Zimbabwe suggests that behav- based health education programs is mixed, ior change is possible. Young people there with variations in effectiveness depending 130 WORLD DEVELOPMENT REPORT 2007 tion of safe sexual behavior.60 School pro- B O X 5 . 2 The role of education in behavior change grams can reach a large number of young people in countries where enrollment Studies from developing and developed Why does education lead to the adop- rates are high; and enrollments--espe- countries show that more education is tion of healthy behaviors? The observed associated with healthier lifestyle choices, relationship between schooling and less cially in primary school--have grown sig- be it smoking,drinking,sex,or use of seat- risky behavior is partly explained by the fact nificantly, increasing the potential audience belts when driving.In Indonesia,poor and that education raises the private returns to for school-based health information cam- uneducated youth are more likely to smoke remaining healthy.Education gives young cigarettes,and higher education is associated people a sense of the future and the ability paigns. The structured school environment with reduced smoking.In urban Mexico,youth to imagine themselves in the future in a way is conducive to teaching young people who expected to complete secondary school that has some value today.It also improves about their bodies and about safe health were less likely to smoke,engage in risky sex, their earning prospects and other life behavior. The programs offer a chance to or drink alcohol.In rural Kenya,where many opportunities.This gives educated young students do not expect to complete their edu- people a stronger incentive to engage in reach large numbers of young people and cation because of high costs,free uniforms safe behavior.Finally,education also gives their teachers, as well as an opportunity to (and sex education) significantly reduced people the ability to better process health institutionalize sex education and broaden risky sex,evidenced by a drop in pregnancies information.All of these reasons might its impact when ministries of education among schoolgirls.In the United States,those explain why educated people are more who attend and are more engaged in school likely to adopt healthy behaviors.But these make it official policy. No evidence indi- are less likely to engage in risky behavior. observed associations could also be partly cates that sex education increases sexual Better prospects for current and lifetime explained by a"selection effect"--those activity among youth.61 employment,earnings,and wages also reduce youth who value their future also attend A randomized evaluation in Namibia the chance that they engage in risky behavior. school. In South African communities where youth showed that health education given to chil- wages and employment are high,young peo- dren promoted safe sex behavior (box 5.3). ple are also more likely to use condoms.HIV Sources: Blum and Nelson-Mmari (2004); De A summary of 21 school-based sex edu- prevalence in Uganda,thought to be higher Walque (2004); Dupas (2006); Gertler and oth- ers (2006); Gruber (2001); Kaufman and Stavrou cation programs in developing countries among the educated in the early stages of the (2004); Kenkel (2000); Strauss and Thomas epidemic in Sub-Saharan Africa,is negatively found that nearly all had a positive influ- (1995);Witoelar,Rukumnuaykit,and Strauss related to education. (2005); and World Bank (1999b). ence on reproductive health knowledge and attitudes.62 However, not all of those stud- ies assessed behavior. Of the few that did, partly on the outcome evaluated. Health one found an increase in condom use in education can include information about the short term among sexually active youth clean water and sanitation, nutrition, ages 11­14 in Jamaica, but the effect disap- substance abuse, infectious diseases, vio- peared in the long term. Programs in Chile, lence, and sexual and reproductive health. Mexico, and Uganda also included younger A school-based program of the Partner- teenagers and found less sexual activity and ship for Child Development in Tanzania's greater use of contraceptives among those Lushoto district provided education for sexually active. helminth (worm) infection and personal "I know how to prevent risks. hygiene. At the outset of the project, no Using mass media and social marketing to However, many times we do not put schools provided drinking water or water reach all youth. School-based programs our knowledge into practice because for hand washing after using the latrine. By exclude those youth not in school. And [of] our rhythm of life." the end of the first year, all schools in the these programs are mostly provided in sec- intervention area were doing both. Knowl- ondary school, after many young people Young male gang member, edge and practices improved in the inter- have left school (chapter 3) and after many Comas, Peru vention schools but not in the comparison young people have already begun having January 2006 schools. A follow-up survey 15 months sex. Broadly based information, educa- later found that many of the healthy behav- tion, and communication programs aim to iors were maintained in the intervention change the health of young people through schools.59 social marketing and community-wide public information campaigns. In addition School-based health education programs. to curriculum-based health education and Most evaluations of school-based programs sex education programs in schools, such have focused on sex education, which can efforts include youth development pro- increase knowledge and encourage adop- grams through youth centers, and the use Growing up healthy 131 of mass media, to influence the knowledge, attitudes, and behavior of young people. B O X 5 . 3 Reducing HIV risk in Namibia Mass media can inform youth and the The Namibia Ministry of Youth and Sport After 12 months,participants who community about health issues and,in prin- developed My Future is My Choice,a cur- entered the program as virgins were more ciple, shape attitudes, beliefs, and behaviors. riculum-based program derived from the likely to have remained virgins than those Mass media campaigns raise knowledge Focus on Kids curriculum for African- in the control group (especially among significantly, but their impact on behavior American youth ages 9­15 in public hous- girls),though the percentage who reported ing in the United States.It included basic remaining virgins fell dramatically among is limited. Social Marketing for Adolescent information about reproductive biology, both groups.Condom knowledge and com- Sexual Health (SMASH) in Botswana, Cam- HIV/AIDS,substance abuse,and violence,as petence remained strong after 12 months: eroon, Guinea, and South Africa changed well as communication and decision-mak- participants were more likely to report that knowledge but not behavior.63 The Arte y ing skills,over 14 sessions.The intervention they knew where to find condoms and how was randomly assigned to young people to use them,and were more likely to report Parte program (Paraguay) and the Promo- in grades 9 to 11 (15- to 18-years-old) in 10 that they could successfully negotiate con- tion of Youth Responsibility project (Zim- secondary schools. dom use with their unwilling partners. babwe) reached similar conclusions. Participants displayed improved knowl- The results were based on self-reported edge of HIV/AIDS,reproduction,and the use behavior,which the authors acknowledge Broadcast media, with their wide reach, of condoms relative to the control group. is subject to bias.They note that it would be can increase awareness, though there is little A few indicators of attitudes toward sexual useful if future studies could also examine evidence that they change behavior. The intercourse also improved for participants. the effect on biological outcomes (such as television drama Sexto Sentido in Nicaragua They were more likely than controls to have pregnancy and HIV and sexually transmit- used a condom in the immediate follow-up ted infections rates). attracts 70 percent of the entire TV-viewing period and more likely to report that they audience, and 80 percent of 13- to 17-year- intended to use condoms in the next six Sources: Fitzgerald and others (1999) and Stan- olds.64 Even more impressive, the wildly months. ton and others (1999). popular dramas Soul City and Soul Buddyz in South Africa are watched by 13 million people each week, and the entire cam- the messages into television programs and paign (radio, television, and print) reaches advertisements, and asking pop stars to per- an estimated 16 million people each week. form specific HIV/AIDS songs.66 Evaluations of Soul City consistently show Peer programs recruit and train a core that viewers and readers are more aware of group of youth to serve as role models and to health risks and healthy choices. However, provide information,referrals to services,and the studies have not controlled for preexist- contraceptives to their peers. They typically ing differences between viewers and non- include several elements important to health viewers. Nor have they examined changes in promotion and development: strong identi- behavior or such objectively measured out- fication with the social and cultural environ- comes as pregnancy and incidence of sexu- ment of the target group, promotion of social ally transmitted infection (STI).65 norms and values supportive of positive atti- Youth development programs and peer tudes and health behavior, and involvement education programs vary widely in design of young people in programs designed for and goals. Youth development programs them. They take advantage of the fact that focus on life options and skills, educational many young people prefer to interact with aspirations, vocational opportunities, and others similar to themselves and commonly psychosocial development needs. They may identify peers as one of their primary sources or may not address reproductive health, for reproductive health information. but the many program components can act There are few well-controlled evalua- "We don't know how to get together to promote a healthy lifestyle. One tions of peer education programs. One in knowledge about sex. It has always group of young people consulted during the Peru improved knowledge and attitudes, been a topic that can't be talked preparation of this Report suggested both reduced the proportion of sexually active about in public." cultural and content-specific ways in which males, and increased contraceptive use. The Male university freshman, 18, information on HIV/AIDS and reproduc- West African Youth Initiative in Ghana and China tive health could be made more effective and Nigeria reduced risky sexual behaviors in December 2005 attractive to youth. These included making the intervention area, and the Entre Nous the content short and specific, keeping the Jeunes Program in Cameroon increased message "real, close to daily life," integrating condom use.67 132 WORLD DEVELOPMENT REPORT 2007 Information is not always Figure 5.8 Knowledge of the ABCs (Abstain--Be faithful--use Condoms) to avoid HIV prevention is enough--preferences matter too low among young people Interventions to change behavior--such Boys ages 15­24 as the Abstain, Be faithful, use Condoms (ABC) campaign--have been the main- Dominican Republic Nepal stays of HIV prevention since the 1990s. Haiti These programs rely on informing people Kenya about behaviors that can protect them from Zambia Armenia becoming infected. Uganda's success in Benin halting the rise of HIV infections has been Uzbekistan attributed to a combination of these inter- Indonesia ventions.68 In many other countries, knowl- Girls ages 15­24 edge of these approaches is low despite the Dominican Republic promotion of ABC messages (figure 5.8). Uganda Even among those who know the ABC Zambia messages of the HIV/AIDS campaigns, few Guyana Cambodia put their knowledge into practice. Among Kenya youth who report that condoms protect Bolivia against HIV, few actually use them (figure Vietnam Philippines 5.9). Low rates of condom use are found even Haiti among those who know how to obtain them. Benin Young people with more education are Albania more aware than those with less educa- Nepal Armenia tion that condoms can prevent HIV trans- Morocco mission, and they are more likely to use Uzbekistan condoms. Among all young people, how- Indonesia Turkmenistan ever, the wide gap between knowledge and Tajikistan behavior is not eliminated by education. In 0 10 20 30 40 50 60 70 80 fact, knowledge of condoms is more respon- Knowledge of HIV prevention sive to education than is condom use (figure methods (percent) 5.10), so that the gap between knowledge Sources: HIV/AIDS Indicator Database (MEASURE DHS) and sur- and behavior increases with education. veys conducted 2000­04. Note: Knowledge of HIV-prevention methods is the composite of People engage in risky behavior because two components (prompted). The figure represents the percentage of respondents who, in response to a prompted question, say that it yields benefits. Unprotected sex, smok- people can protect themselves from contracting HIV by using con- ing, and drinking alcohol can be perceived doms or having sex only with one faithful, uninfected partner. as providing short-term goods for the per- son engaging in these activities. There is also The influence of peers, as well as cul- a significant time lag between the activity tural and historical norms, may influence (which provides a benefit) and the mani- individual behavior, and peers may encour- festation of harmful consequences, and the age risk-taking or discourage risk aversion. risk that the individual faces is uncertain. For example, it may be unseemly for young The symptoms of AIDS, for example, are women to argue about sex with a partner, or apparent only some years after exposure to to insist on condoms,or even to discuss sex at the virus, and the risk of being infected with all.And even occasional behavior can quickly each act of sexual intercourse is on the order become habitual.70 That is why it is essential of 0.1 percent, with the probability varying to intervene when people are young and in enormously across individuals.69 Low values the process of forming habits and identities. on the future relative to the present, and per- ceptions of invulnerability can affect the will- Changing preferences to narrow ingness of young people to translate better the knowledge-behavior gap knowledge about safe health practices into Even when young people receive informa- safer behavior. This is clearly evident in sex, tion on healthy behavior, they may make but it applies to any other risky behavior. choices that put their health at risk. Inter- Growing up healthy 133 ventions to change behavior have tradition- Figure 5.9 Knowing that condoms prevent HIV does not always lead to condom use ally focused on the design and content of Dominican Republic Dominican Republic information campaigns, on the assump- Uganda Zimbabwe tion that with the right information young Mozambique Uganda people will make better choices. But as this Burkina Faso Malawi Zimbabwe Burkina Faso chapter has shown, it is also necessary to Malawi Mali address the formation of preferences and Cameroon Cameroon decision-making capabilities. Mali Bolivia Zambia Haiti Kenya Zambia The design and content of health informa- Benin Benin Bolivia Mozambique tion campaigns. Many issues relating to Ghana Kenya health education, particularly sex educa- Haiti Ghana tion, still need to be resolved. Programs 0 20 40 60 80 100 0 20 40 60 80 100 vary widely in what is taught, at what age, Percentage of unmarried girls Percentage of married and ages 15­24 unmarried boys in what setting, by whom, and in what man- ages 15­24 ner. Often, funding is low, and teachers are Know about and use condoms Know about but do not use condoms not trained to deliver the information effec- tively. In many cases programs are offered Source: Demographic and Health Surveys conducted between 1997 and 2003. only in high school after many youth are Note: The total length of each bar represents the percentage of sexually active young people who know that condoms can prevent HIV transmission. already sexually active. Despite the fact that no evidence suggests that sex educa- tion increases sexual activity among youth, Figure 5.10 Knowing that condoms prevent HIV increases with education, but so does the gap between knowledge and behavior not all countries include sex education as a part of their school health program.71 Most Kenya Dominican Republic countries provide information to secondary Percentage of married and unmarried young Percentage of married and unmarried young students, fewer to primary students (table women ages 15­24 women ages 15­24 100 100 5.1). Many provide information on nutri- Knowledge that condoms prevent HIV tion, but in only a very few does this include 80 80 information on obesity. Knowledge that condoms prevent HIV Providing accurate and specific informa- 60 60 tion is more effective than providing vague 40 40 or general information. School-based HIV/ STI programs were more likely to have an 20 Use condoms 20 Use condoms impact on behavior than general reproduc- tive health programs.72 A school-based sex 0 0 No Incomplete Complete No Incomplete Complete education intervention in Kenya that pro- education primary primary education primary primary vided young girls with information about Sources: Demographic and Health Surveys: Kenya (2003) and Dominican Republic (2002). the higher prevalence of HIV infection Note: "Know that condoms prevent HIV" includes those who know condoms can prevent HIV transmission. "Use con- doms" is based on the sample of those surveyed who know condoms can prevent HIV transmission. among older men reduced the incidence of intergenerational sex and significantly reduced pregnancies among girls--in a set- young people in schools do not learn basic ting where age-mixing is quite common.73 literacy and numeracy skills, raising ques- Messages must provide a range of options: tions about the effectiveness of the health programs providing only one message--say, education curriculum (chapter 3). An HIV on abstinence--will not reduce STIs. The information campaign in Uganda benefited information must be reinforced by repeated better-educated participants more than it exposure, too. Those that include multiple did those with less education.75 If all young sessions, over a number of years, are more people are to be reached, the content of the effective than those that merely provide program and the way it is delivered must be information once or a few times.74 tailored to reach them. Health information campaigns must take into account the ability of the target Changing attitudes and preferences directly. audience to absorb the messages. Many In recent years, evidence that information is 134 WORLD DEVELOPMENT REPORT 2007 Table 5.1 Proportion of countries that include specific health topics in the school curriculum (percent) HIV/AIDS HIV/AIDS in in primary secondary Substance abuse Specifically Number of Region school school (including alcohol) Tobacco Nutrition obesity countries Sub-Saharan Africa 93 86 46 54 79 7 28 South Asia (except India) 0 100 75 75 75 25 4 Indian states 13 100 13 13 100 13 8 East Asia and Pacific 83 100 100 83 100 50 6 Latin America and the Caribbean 0 100 85 62 85 46 13 Middle East and North Africa 33 67 0 0 67 33 3 Source: Partnership for Child Development (2006). Note: Sub-Saharan Africa includes Benin, Burkina Faso, Cameroon, Côte d'Ivoire, Chad, the Democratic Republic of Congo, Eritrea, Ethiopia, Gabon, The Gambia, Ghana, Guinea, Kenya, Liberia, Madagascar, Malawi, Mali, Mauritania, Mozambique, Niger, Nigeria, Rwanda, Senegal, Sierra Leone, Tanzania, Togo, Uganda, and Zambia. South Asia includes Bangladesh, Bhutan, Nepal, and Sri Lanka. Indian states include Andhra Pradesh, Gujarat, Karnataka, Maharashtra, Manipur, Nagaland, Tamil Nadu, and Uttar Pradesh. East Asia and Pacific includes Cambodia, Indo- nesia, the Lao People's Democratic Republic, the Philippines, Thailand, and Vietnam. Latin America and Caribbean includes Argentina, The Bahamas, Barbados, Belize, Brazil, Chile, Guyana, Jamaica, Mexico, Panama, St. Vincent and the Grenadines, Trinidad and Tobago, and Uruguay. Middle East and North Africa includes Egypt, Islamic Republic of Iran, and Israel. necessary but not sufficient to change health As box 5.2 shows, education has at least behavior has led to interventions designed two beneficial consequences for health to change preferences, either by fiat or by behavior, in addition to simply provid- persuasion. The experience to date suggests ing a conduit for information. First, it can that dictating behavior change, that is, forc- enhance the ability of individuals to absorb ing people to make certain choices, is not the information, and second, it can change likely to be successful. Virginity pledges and expectations and attitudes about the future. abstinence-until-marriage programs have For example, people with more education garnered much popular attention, compli- are more likely to value the future more cating evaluation and interpretation. A sur- highly and discount the future more slowly "Many in Vietnam think that sex vey of published research from developed (see box 2.7); and those with higher dis- education is giving youth the `key' to countries in 2001 found no evidence that count rates--or a higher relative value of certain behaviors/activities, but it is they had any lasting impact on sexual activ- the present--are more likely to be obese.81 better to give [them] the key rather ity or risks.76 Other research has found that Enhancing a young person's sense of than let [them] get lost in difficult virginity pledges may delay first sex,but have the future is one of the prime objectives situations." no impact on STI incidence or pregnancy, of health interventions that focus on "life because those who pledged were much less skills." These include the ability to think Young notetaker at the likely to use contraception than nonpledg- critically, to be assertive, and to understand youth consultation in ing youth once they engaged in sex.77 the influence of community, family, and Ho Chi Minh City, Vietnam Mental health services can change the gender in decision making. One life-skills April 2006 incentives for youth to make healthy deci- program among youth in Kenya found that sions, because depression and low self- male participants were more likely to report esteem can lead to excessive risk-taking. condom use, and female participants were Similarly, many studies, primarily from more likely to report fewer sexual partners, developed countries, suggest that regular than among the control group.82 Many life- physical activity and sports can lower the skills programs involve young people as risk of many chronic diseases of adulthood, peer educators to provide information (box reduce stress and depression, and improve 5.4). Preliminary anecdotal results from the self-esteem.78 Sports programs also dis- ongoing Regai Dzive Shiri project in Zim- seminate information, empower the young, babwe include the observation that "girls and keep idle youths occupied and away walk with their heads high" and that boys from crime.79 The Mathare Youth Sports complain that "they are more difficult to Association in Kenya, in addition to orga- seduce these days"!83 nizing sports, engages thousands of young slum residents in community-development The need for well-designed evaluations. activities and provides them with informa- Behaviorthatisprivate,hardtoverify,andthat tion on HIV and other issues (chapter 7).80 is the subject of strong cultural norms is diffi- Growing up healthy 135 culttochange.Forthesereasons,theimpactof alcohol consumption,taxes,advertising bans, health education programs designed to affect and sales restrictions can reduce demand. this behavior has proven extremely difficult to evaluate. For example, sex education may Improving access to health services affect both knowledge and some self-reported Informing youth about practicing safe sex behavior,84 but there is little evidence that can increase condom use, if they have access these programs have a beneficial impact on to pharmacies or clinics that distribute con- objectively measured outcomes. Most studies doms.86 Social marketing of condoms can tend to evaluate changes in knowledge, rather improve young people's uptake of condoms than changes in behavior. through pharmacies and retail outlets. The Alford, Cheetham, and Hauser (2005) Horizon Jeunes program in Cameroon com- reviewed nearly 200 studies of youth- bines health education efforts with condom oriented health interventions, of which 10 distribution by peer educators. As part of a met their criteria for success in both inter- countrywidesocialmarketingcampaign,the ventionandevaluation.Of those10,onlytwo program increased condom use by young showed any positive impact on objectively women (box 5.4).87 Condoms can also be measured health status. A series of reviews distributed to young people where they get conducted by Kirby, Lepore, and Ryan together, such as at youth centers. The goal (2005) and Kirby, Laris, and Rolleri (2005) of this approach is to reach young people in found strong evidence of program impact a comfortable environment where they also on knowledge, values, and self-reported have access to recreational activities. How- behavior.Of the 83 studies reviewed in Kirby, ever, experience from interventions in Latin Laris, and Rolleri (2005), two-thirds found a America and Africa have shown that such significant positive impact on self-reported centers are not effective at changing the sexual behaviors or outcomes. However, behavior of sexually active youth.88 only 9 of those 83 include biomarkers, and The coordination of multiple condom of those 9, only 3 report a positive impact on promotion strategies is needed to reach all health outcomes. youth--married and unmarried, in and Studies that do look at behavior tend to out of school. Often, however, reproduc- rely on self-reported behavior, which can be tive health programs are separate from subject to reporting bias. A three-year eval- programs to prevent STIs. Funding as well uation of a school-based, family-life educa- as interventions are often separate for con- tion program in Jamaica found that young doms intended for contraceptive use (dis- people in the schools that received the tributed by family planning services) and intervention were more likely than those in those to prevent STIs and HIV transmission a comparison group to report their sexual (distributed through STI services, HIV test- activity inconsistently.85 For the effects of ing and counseling services, and condom health education programs to be reliably social marketing campaigns). estimated, objective outcomes, such as preg- nancies or prevalence of STIs, must be used. Those outcomes will be evident only with a Social marketing can change behavior-- lag, and studies must be prepared to assess B O X 5 . 4 behavior over a suitable length of time. Horizon Jeunes in Cameroon Horizon Jeunes,a reproductive health pro- serve unmarried young women,often denied Enhancing opportunities gram targeted to urban youth ages 12 to 22 services by reproductive health providers. to make healthy choices both in and out of school,was framed within Horizon Jeunes added to young the national social marketing campaign. It people's knowledge of reproductive health Having better access to health services can sent out two main messages to youth through and changed behavior.The proportion of encourage young people to practice safe various channels:to delay the initiation of sex, females who ever used condoms rose from behavior. Conversely, restricting opportuni- and if they chose to have sex to use condoms 58 percent to 76 percent in the treatment ties to make poor choices may also be benefi- to prevent HIV and STIs.Youth-friendly,the pro- group,compared with a decline from 53 gram had peer educators distribute the Pru- percent to 50 percent in the control group. cial. The set of available opportunities can be dence Plus condoms promoted in the national altered by changing their availability directly, campaign.It also encouraged providers to Source: Alford, Cheetham, and Hauser (2005). or by changing prices. For tobacco use and 136 WORLD DEVELOPMENT REPORT 2007 This separation is a problem because sumption by 4­5 percent, and binge drink- young married women are less likely to use ing by 20 percent. There is similar evidence condoms, which are widely regarded as less on the consumption of illicit drugs: a 10 effective at preventing unwanted pregnan- percent increase in the price of marijuana cies than other contraceptives. This leaves will reduce marijuana use by 5 percent; and young married couples vulnerable to STIs. price increases in marijuana, cocaine, and However, young people who perceive the heroin reduce both arrests and hospital risk of acquiring STIs to be low may not admissions associated with drug consump- use condoms and therefore leave them- tion. Changes in prices can explain most selves vulnerable to unwanted pregnancies. of the observed changes in binge drinking Promoting condoms as "dual protection" and marijuana use by high school seniors rather than only as "safe sex" may increase between 1975 and 2003.94 the uptake of condoms and protect both Cigarette smoking tends to be more sen- married and unmarried young people from sitive to price in low-income countries than STIs and unintended pregnancies.89 in high-income countries. For example, it is estimated that a price rise of 10 percent for Changing prices and incentives a pack of cigarettes reduces demand for cig- Young people's choices respond to changes arettes by 6­10 percent in China, and only 4 in prices and incomes, as well as to the percent in the United States.95 One reason existence of health services. In rural Kenya, for the difference could be that low-income a randomized controlled experiment pro- countries have a larger share of young peo- viding free uniforms (along with sex edu- ple than high-income countries, and young cation) significantly reduced risky sex, as people are more price-sensitive than adults. evinced by a drop in pregnancy incidence Poorer people are also more price-sensitive among schoolgirls.90 General poverty than wealthier.96 alleviation programs targeted at youth In addition to raising prices through or families with youth can increase the taxation, comprehensive bans on advertising opportunities available to young people, and product promotions, age restrictions on and conditional cash transfers can provide sales, and prominent health warning labels additional incentives for healthy choices. can reduce the consumption of tobacco and The Oportunidades program in Mexico alcohol. Comprehensive bans on cigarette provided incentives for young people to advertising and promotion reduced smoking remain in school, where they received in some high-income countries, although health information and periodic health partial bans had little or no effect.97 Studies services (see chapter 3). In addition to the based on cross-country analysis find no link beneficial effects on schooling, the pro- between advertising and sales restrictions gram led to reduced smoking and alcohol and reduced smoking.98 However, a study consumption for all youth, and an increase of 100 countries comparing consumption in the age of sexual debut among girls.91 trends over time found that consumption Most governments levy taxes on tobacco fell much more steeply in countries that had and alcohol, which increases prices. In gen- nearly complete bans on advertising, com- eral, young people are more price-sensitive pared with countries with no such bans. than adults.92 If the price of cigarettes rises, Health warning labels on cigarette packs, they are less likely to take up smoking, and though effective in reducing tobacco con- those who have begun smoking are more sumption among adults, may not discourage likely to quit. In Indonesia, where the preva- youth from smoking, because they are more lence of smoking among men is high, 15- to likely to buy single cigarettes than packs.99 24-year-old males were more responsive to Young people are exposed to a wide vari- cigarette prices than older males.93 Alcohol ety of tobacco control policies, including consumption also declines with increases in advertising restrictions, health warnings, and price. Among high school students in the prohibitions on the sale of tobacco to minors. United States, a 10 percent increase in the There is little consistency in policies: some price of alcohol will reduce alcohol con- countries ban advertising without restricting Growing up healthy 137 sales to minors; others ban sales to minors but do not restrict advertising. The inde- B O X 5 . 5 Russia limited the sale of alcohol, and deaths pendent effect of each policy is difficult to and illnesses fell identify. Interventions to reduce the con- sumption of these potentially harmful sub- In 1985,President Gorbachev implemented Overall,life expectancy among males anti-alcohol legislation to limit alcohol improved.Even so,the law was repealed stances are more effective if implemented sales and raise the legal purchasing age to after three years.With the liberalization of jointly: for example, tobacco control is more 21.Alcohol consumption fell,but only by alcohol sales and relaxation of administra- effective if it includes both advertising bans 26 percent,despite a 63 percent decrease tive controls,alcohol psychosis increased. and higher taxes.100 in sales.The consumption of unregistered Surveys show a substantial increase in alco- alcohol,including moonshine,nearly hol consumption between 1992 and 1994, Policies to reduce consumption of harm- doubled,and increases were also seen in consistent with a sharp price reduction. ful substances can have unintended conse- substance abuse and related poisoning, quences. In 1985, Russia restricted alcohol especially among youth. Source: World Bank (2005d). sales and raised the legal age for alcohol consumption. This dramatically improved life expectancy among men, but it also dence should be part of a comprehensive increased the use of harmful alcohol substi- tobacco-control policy. Many people are tutes (box 5.5). able to quit smoking successfully on their own and with some guidance (box 5.6).102 What if prevention fails? However, some young smokers might have Helping young people developed a strong dependence on nicotine deal with the adverse and could benefit from nicotine replace- ment interventions to address the physio- consequences of poor logical aspects, such as tobacco dependence health behavior or misfortune and tobacco withdrawal.103 Despite good information and the freedom Strong evidence from controlled trials to make independent decisions, we are all indicate the benefits of treatment for drug susceptible to poor health outcomes. Occa- abuse, including significantly fewer drug- sionally we make poor choices: in spite of related health and social problems. Those the overwhelming evidence of the harm- improvements translate into substantial ful effects of tobacco, more than a billion reductions in social problems and societal people smoke worldwide.101 Occasionally costs,includingreducedcrime,violence,and we fall victim to the external effects of the incarceration.104 Many countries--primar- private decisions of others, or to the failure ily in Asia and the former Soviet Union-- of government to provide services, or we are are introducing methadone maintenance coerced into making risky choices. A vital therapy and needle and syringe exchange function of youth health services is to help young people overcome the adverse conse- quences of risky health behavior. They may be addicted to tobacco or drugs, or infected B O X 5 . 6 Technology can help change young people's with sexually transmitted diseases, or have behavior: Using text messages in an unintended pregnancy. They need sup- New Zealand to reduce smoking port to recover to a healthy life: abortion services, sexual health services, maternal In New Zealand,as in many countries around received general information only once in the world,mobile phone ownership is two weeks.All participants were between and child health services, drug rehabilita- high and text messaging very popular.An 19 and 30,half of them female. tion services. They also need cessation and intervention in New Zealand used text mes- After six weeks,28 percent in the treat- rehabilitation programs to help them stop saging to help youth quit smoking.Informa- ment group had quit smoking,more than the behavior that is causing the harm. tion on individual participants was used to twice the percentage in the control group. create personalized text messages providing Given the widespread use of mobile tele- advice,support,and distraction.Participants phone technology,this type of interven- Programs to help youth could also communicate with others with tion might be easily replicated in different overcome addiction similar characteristics ("quit buddies"). regions. Participants randomly assigned to a A supportive environment is needed to treatment group received five personalized Sources: Rodgers and others (2005) and Inter- encourage tobacco consumers in their messages per day,those in a control group net Safety Group (2005). attempts to quit. Treating tobacco depen- 138 WORLD DEVELOPMENT REPORT 2007 programs. There is similarly strong evi- STI in adolescents increased the uptake of dence that these programs reduce the risk STI services among sexually experienced of HIV infection.105 Two international students and significantly reduced the inci- studies found that average seroprevalence dence of STIs.109 decreased in cities with exchange programs, Unmarried adolescents are often denied and increased in cities without them.106 services in countries where premarital sex is frowned upon. In Ghana, services were Improving access to health denied to young or unmarried clients, and services for treatment of to married women who could not demon- sexually transmitted infections strate the consent of their spouses.110 Even More than 100 million STIs other than HIV where young people are legally protected, occur every year around the world among reproductive health services may be out people under 25 (figure 5.11).107 Most are of reach. In South Africa, many reproduc- easily treated, without severe or lasting con- tive health services are not easily accessible sequences, if diagnosed and treated early. by youth, and young people feel that facil- Treating STIs prevents HIV/AIDS trans- ity staff are judgmental and hostile.111 In mission, because STIs facilitate the sexual Nigeria, adolescents who contracted an STI transmission of HIV, particularly syphilis, would go to a traditional healer rather than chancroid, and genital herpes, which pro- use formal reproductive health services duce genital ulcers.108 Many infections,how- because of the high cost and low quality.112 ever,go unnoticed,especially among women For these reasons, making clinics youth- and girls, who may show no symptoms or friendly--training doctors and nurses to signs so mild that they are unrecognizable. deal with young clients, having clinic hours In Nigeria, providing school students with that are convenient for youth, and offer- STI health education and training providers ing space where young people can consult (pharmacists and private doctors) to treat with providers in privacy--may encour- age young people to use health services for treatment. Evaluations of youth-friendly Figure 5.11 Reported STI incidence varies among youths in Sub-Saharan Africa health services have so far not shown any evidence that they increase the use of health Benin services by young people.113 The interven- Cameroon tions that do seem to increase their use of Central African Republic treatment services are health information Chad that teaches them to recognize symptoms of STIs and referrals to trained providers, as in Côte d'Ivoire the Nigeria program. Gabon Antiretroviral therapy (ART) can turn Guinea AIDS from a certain death sentence to a Kenya manageable (though severely unpleasant) Malawi chronic condition. Treatment remains pro- Mali hibitively expensive, and current interna- Niger tional efforts to provide ART are already Nigeria overwhelming the capacity of many hard- Togo hit developing countries, even though only Uganda a small percentage of those clinically eligible Zambia for treatment actually receive it. Public pro- Males vision of ART may be feasible and affordable Zimbabwe Females in cases where the epidemic is limited.114 0 5 10 15 20 There have been no well-controlled stud- Incidence of STIs among sexually active 15- to ies of the impact of treatment on preven- 29-year-olds (percent) tion, but some simulation models show that Source: Bankole and others (2004). treatment can enhance the effectiveness of Growing up healthy 139 prevention.115 Other simulations caution Not all unprotected sex has to result that an exclusive focus on treatment will in an unwanted pregnancy (where young increase incidence, partly due to the neglect women are forced to consider abortion as of other, more conventional interventions, an option). Access to emergency contra- and partly due to the incentive effects of ception can effectively prevent unwanted treatment.116 pregnancy in such cases, but it is typically SimulationsinSouthAfricashowthatfull not provided as part of reproductive health coverage of ART could avert up to 1.7 mil- services in developing countries.124 Even lion deaths and 860,000 orphans by 2010.117 when young people know about emergency Some empirical evidence suggests that ART contraception, few are well informed about will reduce the costs associated with hospital its use.125 care and treatment of opportunistic infec- Safe abortion services and emergency tions due to AIDS. One study from Brazil contraception are, however, highly contro- estimated that ART averted 358,000 hospi- versial in many countries. Where abortions tal admissions between 1996 and 2002, sav- are illegal or restricted, the percentage of ing US$2.2 billion.118 HIV-positive injecting unsafe abortions and maternal deaths aris- drug users, predominantly young, also ben- ing from them tends to be very high. This efit from ART treatment,119 but little atten- is evident in Romania, where maternal tion goes to treatment and care for injecting death dropped dramatically when abortion drug users living with HIV/AIDS.120 Among was legalized.126 Where legalizing abortion those who do receive treatment, allocations is not feasible, the access for young people do not favor youth.121 to modern contraceptive methods must be strengthened. Providing access to emergency Ensuring that girls have access to safe contraception and safe abortion services and postabortion care abortion services will greatly reduce the associated health One potential outcome of unprotected sex risks of unplanned pregnancy. One pos- is unplanned pregnancy. For unmarried sible approach is to link contraceptive ser- young women, such pregnancies can bring vices (including emergency contraception), immense social costs, especially in countries abortion, and postabortion care, making where family networks do not support out- them part of maternal health service deliv- of-wedlock births. It is not surprising, then, ery.127 Access to postabortion care can be that each year millions of young women strengthened through outreach by trained undergo unsafe and illegal abortions. midwives. Where the public sector cannot Women opting to terminate pregnancies provide abortion or postabortion care, pri- using unsafe methods are predominantly vate clinics are viable alternatives, if they are young and unmarried, particularly in Sub- regulated to ensure safe procedures. Saharan Africa and Latin America and the Caribbean. In Sub-Saharan Africa, about 60 Feedback effects of percent of women who have unsafe abor- treatment programs tions are 15­24 years old. In Latin America Do programs to cope with or mitigate the and the Caribbean, young women make consequences of bad choices affect the deci- up about 43 percent of those who undergo sion to engage in risky behavior in the first unsafe abortions.122 In Kenya, Nigeria, and place? Can treatment programs have posi- Tanzania, adolescent girls make up more tive feedback for prevention? Or might there than half of the women admitted to the be negative feedback, or moral hazards? If hospital for complications following illicit so, can we design "incentive-compatible" abortions, adding to the costs of already interventions to minimize poor choices and under-resourced health systems.123 Access still protect those with poor outcomes? to safe abortion services is thus critical for There is some legitimate concern that young women to avoid further damage to providing second chances will encourage their health. moral hazard--that is, if the costs of some 140 WORLD DEVELOPMENT REPORT 2007 risky actions are not borne by individuals, Evidence indicates that young people do they will behave more carelessly, or in a way respond to policies that alter the set of avail- that increases risks to themselves or others. able opportunities. This is most clearly seen For instance, health insurance may cause in the impact on young people of changes people to take greater risks with their health, in prices and incomes, such as the imposi- or consume more health services than they tion of cigarette taxes and the distribution would in the absence of insurance.128 of conditional cash transfers. This chapter No evidence suggests that the availability has argued, however, that the most sig- of emergency contraception will increase nificant determinant of healthy behavior the incidence of unprotected sex. Of the in youth--and of health in adulthood--is women attending a family planning clinic the capability of young people to make the in Pune, India, those who received pills and right decisions. information were not significantly more Some health promotion interventions likely to have unprotected sex--and not one have changed behavior, such as sanitation of these woman used emergency contracep- and hand-washing--as has information tion more than once over the course of the provided to parents to improve the health year.129 In the United Kingdom, only 4 per- of young children, such as breastfeeding. cent of those using emergency contracep- Behavioral interventions have empha- tion reported using it more than twice in a sized health education, on the assumption year, so they presumably were not using it as that individuals lack information and that a substitute for regular contraception.130 more information will allow them to make It has been suggested that ART will dis- healthier choices. A key policy recommen- courage risk aversion and increase risky sex- dation in countries where information is ual behavior.131 Studies of men who have sex missing, or misinformation rife, is to estab- with men in Europe and the United States lish standards and provide information have shown that potent ART may induce about behavior and consequences. As table both HIV-positive and HIV-negative men to 5.2 shows, school health education varies engage in riskier behavior than they would greatly around the world. In some regions, in the absence of such therapies.132 A long- information on HIV is provided only in term study of condom use in Kenya--dur- secondary schools, after many young people ing a period when two reported "cures" for have become sexually active. AIDS were announced and widely touted As this chapter has shown, information by government leaders--found that after is essential to healthy decision making; but each announcement, sex workers reported a it may not be sufficient, especially for deci- substantial drop in condom use. Eventually sions about such private issues as food and the ineffectiveness of these drugs against sex. Behavior in these areas is extremely AIDS became apparent, and condom use difficult to influence. More evidence will increased.133 improve understanding and the design of policies. In these difficult areas, experi- ence provides three clear directions about The interventions affecting the health of health information campaigns. First, target young people can be loosely distinguished younger people rather than older youths: as those that affect the opportunities and information about safe sex has a stronger capabilities for healthy choices and those and more lasting impact if delivered prior that offer second-chance services for those to the initiation of sexual activity. Second, who require assistance to cope or recover. repeated exposure to health messages is Table 5.2 summarizes the evidence of inter- more effective than one single exposure. ventions known to work in promoting Third, provide specific information that youth health and those that are promising, addresses the real problems facing young but for which claims of effectiveness have people. This is best seen in the program in yet to be verified. Kenya that provided specific information Growing up healthy 141 Table 5.2 Summary of policies to improve young people's health and health behavior Proven successful Promising but unproven Unlikely to be successful Opportunities Opportunities to make healthy choices Providing services to Treatment for depression and mental health Youth centers (more likely to change behavior attract boys, older youth) Sports programs and programs promoting physical fitness Using taxes, advertising Price controls and taxes (for tobacco Partial advertising bans on bans, and sales and alcohol) tobacco and alcohol restrictions Total advertising bans, restrictions on sales to minors (tobacco and alcohol) Capabilities Strengthening ability to practice healthy behavior School-based health Curriculum-based sex education combining Health education on tobacco, alcohol, and drug abuse Abstinence-only programs education basic reproductive health information with (to delay transmission of Health education promoting healthy diet, physical communication and decision-making skills STIs, HIV, and pregnancy) activity (Namibia's My Future is My Choice) Abstinence-only programs (to delay initiation of sexual Curriculum-based programs providing activity) culturally relevant information about risky sex and specific information about health risks (rural Kenya, randomized evaluation) Mass media and social Mass media campaigns combined with peer Social Marketing for Adolescent Sexual Health Program promoting only marketing education and trained providers in Cameroon (SMASH) program in four African countries--program abstinence (to delay (Horizon Jeunes) effective in increasing condom use only in Cameroon transmission of STIs, HIV, (no evidence of behavior change in Botswana, Guinea, and pregnancy) and South Africa) Radio, television plus telephone hotline and peer education-- Sexto Sentido (Nicaragua), LoveLife (South Africa) Second chances Minimizing consequences of risky behavior HIV/AIDS and sexual STI treatment and counseling--increased ART (to prevent HIV transmission) and reproductive health use of private physicians to treat STIs in two "Youth-friendly" sexual health and family planning services cities in Nigeria (Benin City and Ekpoma) health services Providing emergency contraception and abortion services ART (to minimize consequences of HIV infection) Providing postabortion care Treatment of addictions "Harm reduction"--needle exchange, and cessation programs methadone substitution to young women on the higher HIV preva- environments. The benefits of these second- lence in older partners.134 chance services accrue both to the young Second-chance health services--espe- people served and to the broader society, in cially legalized abortion, harm reduction terms of reduced expenditure on curative for injecting drug users, and STI treat- care, lower infectious disease prevalence, ment--are essential to minimize the adverse and spillover effects on welfare, security, consequences of poor decisions and poor and economic growth. spotlight Addressing disparities among Brazil's youth Persistent inequality is one of the most important concerns in an economy that is otherwise growing and making progress in reducing poverty. Inequality in human development outcomes among Brazil's youth endangers future progress. In particu- lar, a shortage of labor market and educational opportunities for vulnerable groups reinforces the intergenerational trans- mission of poverty and leads to continuing inequality. Brazil is developing a cross-sectoral approach to break this cycle. Brazil, an industrial power with the largest Targeting disadvantaged youth value of the transfer for older youth to stay population in Latin America and the Carib- Brazil has a rich portfolio of public and in school (recognizing the higher opportu- bean, has made big strides in reducing the private programs to expand opportuni- nity cost), or (c) extending attendance con- poverty that continues to afflict millions ties, enhance capabilities, and provide sec- ditions to youth ages 16­18, which would of its people. Brazil's literacy and tertiary ond chances to its youth. It is guaranteeing capture secondary school enrollment (or a enrollment rates are on par with other Latin universal access to antiretroviral medica- combination of the three). American countries such as Colombia and tion, part of an HIV/AIDS strategy that is Mexico, but averages mask the disparities ProJovem considered to be an international model. that affect millions of poor young people.1 In education, it is providing more funds to The Brazilian government recognizes that local municipalities to address high repeti- young people who have left school may · Youth (15­24 years old) from the poor- tion rates and the poor quality of second- wish to further their education. In fact, est decile (10 percent) of families have ary school services. It has also begun to look approximately 20 percent of working youth a formal sector employment rate of 4 for responses that cut across transitions and return to school.8 The Educação de Jovens e percent, one-eighth the national average sectors, with the Bolsa Familia, ProJovem, Adultos program is an adult education class for the age group--and barely a tenth of Abrindo Espaços, and health education that focuses on literacy for adults and youth the national average employment rate of programs for males and females. who have left school. In addition, a new pro- adults.In contrast,those from the richest gram--ProJovem--is being piloted. Going decile have a formal sector employment Bolsa Familia beyond literacy for youth (ages 18­24) rate of 50 percent, one-third higher than For so many overage students, the oppor- who have left school, it instead offers a full the national average for the age group.2 tunity cost of staying in school is high. curriculum that covers mathematics, lan- · Youth from the poorest decile of families The perceived benefit of staying in school guages, job preparedness, and citizenship, have a 14 percent illiteracy rate,three times beyond a few grades of primary is low and among other topics; a two-week volunteer the national average (by age 12, half the the forgone earnings can be significant.5 To project; and career and general support ser- youth from the poorest families have left alleviate the opportunity and direct costs vices to youth while they are participating school). The corresponding rate for those to attending school, Brazil was one of the in, and shortly after leaving, the program. from the wealthiest decile is 0.3 percent. first countries to experiment with condi- Abrindo Espaços · By age 14, young women from poor tional cash transfers for school attendance. neighborhoods have begun their tran- Bolsa Escola (School Scholarship), emerged Social exclusion is hypothesized to be a driv- sition to motherhood--compared with first at the state level, and paid families a ing factor behind youth violence. The Open age 17 among girls from the wealthiest monthly stipend if all of their children ages School Program in Pernambuco, started in households.3 7­14 attended school.6 School attendance 2000 as a partnership between local govern- increased more for beneficiaries than for ment and UNESCO, keeps schools in the Because inequalities of opportunity a control group. In 2001, the program was poorest and most violent neighborhoods are easily transmitted across generations, expanded to the national level and in 2004 open during weekends, offering children investing in its young people is key to Bra- the federal government launched the Bolsa and youth an array of cultural and athletic zil's long-run poverty reduction strategy. Família (Family Grants) program merging activities to keep them off the streets and A recent analysis of the main household the Bolsa Escola with other conditional cash to allow them to express themselves in a survey indicates that more than a fifth of transfer programs.7 While an impact evalu- peaceful manner. A UNESCO study shows the total earnings inequality in Brazil can ation of the Bolsa Família program is just that participating schools have experienced be explained by four variables: parental starting, a 2005 study concludes that among a 60 percent reduction in violence.9 The schooling, father's occupation, race, and the lower-income deciles, the stipend can program, now known as Abrindo Espaços region of birth,4 with the human capital of make a difference of 11.5 percent in school (opening spaces), has since been expanded the parents most important. enrollment.To expand the program to youth, to Rio de Janeiro, Bahia, São Paulo, and Brazil is trying to address these inequali- the government has been discussing the pos- Rio Grande do Sul, and is showing positive ties by targeting disadvantaged youth directly sibility of adapting Bolsa Família's education results.10 Also, schools that entered the pro- and by coordinating the actions of diverse incentives by (a) providing bonuses for sec- gram earlier had greater success, suggesting providers. ondary school graduation, (b) increasing the increased impact over time.11 Addressing disparities among Brazil's youth 143 Health education The program was evaluated in three of Rio's els. The recently organized Vozes Jovens Given early sexual initiation among youth favelas, two of which had the program and have developed a proposal for a national combined with risky sexual behavior (as the third of which did not. The evaluation youth policy. This group of leaders from defined by the lack of use of contracep- found that six months after the program youth nongovernmental organizations tives), programs addressing youth are of ended, there was greater condom use among (NGOs) both strengthens the NGO particular importance in preventing teen- program participants, less incidence of new movement and gives youth voice on the age pregnancies and the spread of sexually sexually transmitted infections, and a sig- national stage. transmitted diseases. In 2003 the Ministries nificant improvement on the Gender Equi- · State and local governments have devel- of Health and Education launched a con- table Men scale,13 relative to the control site. oped youth strategies and are channel- troversial pilot program of condom dis- While there were some shortcomings in the ing federal and their own resources to tribution to schools in five municipalities. evaluation methodology, Program H's expe- local civil society organizations and In 2004 the program was extended to 205 rience shows that its approach is promising private sector firms to implement pro- municipalities responsible for almost half of for healthy sexual behavior.14 grams. Better coordination between the all HIV/AIDS cases in Brazil. The program state and local levels in terms of defin- has the additional benefit of preventing ing target groups, priorities, and the Coordinating policies teenage pregnancies, now accounting for 25 division of labor at each level of govern- percent of all births in Brazil. The program to target poor youth in ment would allow for greater efficiency was expected to reach 900 public schools a highly decentralized state in the delivery of services. For example, attended by about a half million students. Recently, Brazil has taken steps to enhance the federal government of Brazil is While this particular initiative has not been coordination among the various players in using tax and expenditure incentives evaluated, similar programs for the general youth policy: to municipalities and states to increase population have been part of Brazil's suc- secondary school enrollment via a · The recent creation of the Secretaria da cessful strategy of curbing the fast spread of financing mechanism known as the Juventude (Youth Secretariat) allows HIV/AIDS. FUNDEB (Fundo de Desenvolvimento for a central guiding body strategically e Manutenção do Ensino Básico e Valo- Gender: Bring the boys in positioned within the Secretaria Geral to rização do Magisterio). Unsafe and early sexual behavior is often facilitate collaboration across ministries · NGOs are already very active in imple- attributed to gender-related roles: boys and develop a national strategy.By focus- menting programs and giving feedback attempting to prove their masculinity and ing on developing national priorities and to government at all levels. Helping to girls without the bargaining power to nego- guidelines to enable actions at the local align local priorities, further encour- tiate the situation. Rather than leaving the level through technical and financial aging the work of the NGOs through responsibility to girls, as in so many pro- support, it aims to leverage public and incentives and support from govern- grams, Program H in Brazil has instead private budgets for maximum impact. ment (financial and technical), and worked to give greater agency to boys by Other ministries are also doing this with improving program design through the altering the way they think about gender their own youth strategies. development of monitoring and evalu- roles and behaviors, with the hope of chang- · The youth themselves are mobilizing at ation systems will further enhance civil ing their sexual choices and expectations.12 both the community and national lev- society's role. Forming families 6 As young people form families, their ability to nutrition services. Programs aimed at to plan safe childbearing and raise healthy delaying marriage can also give young girls children depends on their education, nutri- the opportunity to avoid entering mother- c h a p t e r tional status, and health knowledge--and hood too early. Because a young woman's on their use of health services. Many young nutritional status before pregnancy can men and women are not well prepared.They significantly affect the baby's health, nutri- lack knowledge of good health practices, tional services should reach young women and available maternal and child health before and during pregnancy. services may not fully meet the needs of Strengthening decision-making capa- first-time parents. Malnutrition, especially bilities, particularly in reproductive health micronutrient deficiencies, are common and the nutrition and care of infants, will among young women, who in many parts help young men and women prepare for of the world become mothers when they are parenthood. Health education can stimu- still teenagers, elevating the health risks for late demand for child health and nutrition both mother and baby. services, particularly if it also targets young Young parents' decisions about the tim- men. Teaching life skills to young people ing and number of children affect popula- can encourage them to delay marriage and tion growth and so directly affect economic to use health services. Early child develop- development. If young people choose to have ment programs that promote parenting and smaller families, the decline in births can child care skills can also develop the deci- bring about a rise in the share of the working sion-making skills of young parents. age population, a potential bonus for coun- Second-chance programs can help teen- tries with the right supporting policies. age mothers overcome obstacles posed by Nutrition and reproductive health ser- low education and poor employment oppor- vices are among the most important human tunities. Because most teenage mothers are capital investments that prepare young peo- from poor households, such programs must ple to become the next generation of par- address the disadvantages of poor socioeco- ents, helping them plan births and ensure nomic status. the health of mother and child. Failing to provide a young mother with adequate Preparing for family nutrition before and during pregnancy formation is good for increases the risk of low birth weight infants. Low birth weight infants are less likely to growth and poverty reduction survive the first year of life. Low birth weight Young people's transition to parenthood can also causes irreversible damage to a child's have a lasting impact on the economy and ability to learn in school and be productive demographic trends in a country because in the labor force. It increases susceptibility in most countries, first births--the entry to chronic health conditions in adulthood, into parenthood--take place during youth. such as coronary heart disease. Nearly 60 percent of girls in developing Policies to broaden the opportunities for countries become mothers before age 25. youngmenandwomentobebetterprepared Boys make this transition a bit later, becom- for parenthood include improving access to ing fathers between 25 and 29.1 This differ- reproductive and child health services and ence largely reflects gender differences in the 144 Forming families 145 age of marriage.2 For many reasons, includ- ing societal pressures, newlyweds make a B O X 6 . 1 The sequencing of marriage and childbearing swift transition to parenthood (box 6.1). Marriage may be a precondition for child- example, in Mali, nearly 36 percent of young With the marriage age increasing for women bearing in many parts of the world, but women were married by age 15. and men in most countries, the interval the trend varies across countries. In many Although the causal effect of the age between marrying and having the first child developed countries, such as the United at marriage on various outcomes remains is becoming shorter: most become parents States and France, and in some parts of to be resolved (especially in developing Latin America and Africa, out-of-wedlock countries), its association with negative out- within a year and a half of marriage.3 childbearing is common. For example, in the comes for women is well documented. Early United States, births to unmarried women marriage is associated with early childbear- Impact on growth account for nearly 30 percent of all births, ing and higher fertility.Women who marry and poverty reduction which is the result of both an increased early are also more likely to have less say in proportion of unmarried women and their decision making in a marriage. Increased Preparing youth for the transition to family higher fertility. risk of domestic violence is also associated formation so that they can plan childbear- In many other parts of the world, such with early marriage. ing, have a safe pregnancy, and raise healthy as Africa, South Asia, and Muslim societies, Strikingly, studies report early marriage marriage is the only recognized state in as a risk factor in acquiring HIV/AIDS for children has an impact on productivity which childbearing is permitted.The timing girls. In Kenya and Zambia in 1997­98, HIV and savings, which affect economic growth and cultural norms associated with mar- infection rates were 48­65 percent higher and poverty reduction. Parents' labor sup- riage can significantly affect the quality of among married girls than sexually active ply and productivity increase because they married life.The biggest concern is physical unmarried girls of similar ages. A similar pat- abuse by partners. In South Asia, violence by tern is found in other Sub-Saharan African can plan childbearing more effectively. husbands, sometimes linked to dowry pay- countries. Studies find that high HIV/AIDS Young children deter mothers' participa- ments, is one of the major causes of death prevalence among young married girls tion in work, particularly paid work, as in among young women. is associated with greater frequency of urban Morocco.4 So having fewer unin- Early marriage, and hence early child- unprotected sex with an older partner who bearing, is prevalent in many regions of is more likely to be infected by HIV than tended births can facilitate young women's the world.The proportion of girls marrying younger men. participation in the labor force. Helping before age 18 ranges from less than 20 per- Sources: Bruce and Clark (2004); Buvinic (1998); couples attain their desired family size-- cent in Central Asia to more than 60 percent Clark (2004); Eltigani (2000); Jensen and Thorn- both through reduced child mortality and in Bangladesh, Guinea, and Mali. A very ton (2003); Lesthaeghe and Moors (2000); early transition to marriage, before age 15, is Singh and Samara (1996); Upchurch, Lillard, and through planned births--also increases also notable in some parts of the world. For Panis (2002); and Willis and Haaga (1996). parental investment in their children's edu- cation, nutrition, and health. Because the survival of children can encourage couples to save, these higher savings can have an is incomplete. The risks are heightened for additional impact on growth. girls who become pregnant very soon after Avoiding pregnancies at a very young menarche. Pregnancy-related illnesses asso- age, having fewer unintended pregnancies, ciated with early, frequent, or closely spaced and spacing births prevent the depletion pregnancies drain women's productivity, of mothers' health and reduce the risk of jeopardize their income-earning capacity, maternal and child mortality and ill health. and contribute to their poverty.6 Female genital mutilation, practiced in Pregnancy-related illnesses are a signifi- parts of Sub-Saharan Africa, can also exac- cant cause of death among young women erbate the health risks to young mothers.5 ages 15­29. In South Asia, Sub-Saharan Young women in many countries face the Africa, and the Middle East and North risk of becoming mothers at a very young Africa, regions with high fertility, between age, largely because of early marriage (box 15 and 20 percent of all female deaths arise 6.1). More than 10 percent of 15- to 19- from pregnancy-related causes. Reduc- year-old females are mothers in Sub-Saha- ing those deaths has strong intergenera- ran Africa, South Asia, and Latin America tional benefits. Children benefit from lower (figure 6.1). In Bangladesh and Mozam- maternal mortality because those who lose bique, more than 30 percent of 15- to 19- a parent are much more vulnerable, partly year-old females are mothers or pregnant. because of the loss of resources and partly Pregnant adolescents face higher risks of because of the lack of parental care. In maternal mortality, delivery complications, Indonesia, children whose mothers die are obstructed labor, and premature delivery, less likely to start school and are less healthy mainly because their own physical growth than other children.7 In Ethiopia, children 146 WORLD DEVELOPMENT REPORT 2007 Figure 6.1 Teenage motherhood is common in some Figure 6.2 The share of youth fertility is high in regions countries with low fertility Sub-Saharan Africa Total fertility rate of women ages 15­49 8 Ghana Kenya Nigeria Cameroon 6 Zambia Chad Mali 4 Mozambique India Europe & Central Asia and 2 Middle East & North Africa Dominican Republic Armenia Jordan Turkmenistan 0 20 30 40 50 60 Armenia Morocco Share of young womens' fertility (ages 15­24) Egypt, Arab Rep. of in overall fertility (percent) Sources: ORC Macro (2005) and MEASURE DHS STAT Compiler South Asia and (surveys conducted between 2000 and 2004). East Asia & Pacific Vietnam Philippines the start of childbearing. In some countries Cambodia evidence indicates that women are stop- Indonesia ping childbearing earlier than did previ- Nepal Bangladesh ous cohorts of women. In India in the late 1990s, there was a roughly one year drop in Latin America & the Caribbean the age when women stopped childbearing, Peru from 30.2 years (among women ages 45­49) Bolivia to 28.7 years (among women ages 40­44).10 Haiti Colombia The compression of childbearing during Dominican Republic youth is visible in the large share of births Nicaragua to young women in countries where fertil- 0 10 20 30 40 50 ity is low (figure 6.2). In many countries, Percent of females ages 15­19 who are mothers or are pregnant with births among 15- to 24-year-olds account for first child 30­50 percent of all births. In the Dominican Sources: ORC Macro (2006) and MEASURE DHS STAT Compiler Republic and India, where fertility rates are (surveys conducted between 2000 and 2005). lower than three births per woman, births among 15- to 24-year-old women constitute whose mothers succumbed to AIDS show close to 50 percent of all births. In Armenia, less psychological well-being and lower par- with fewer than two births per woman, the ticipation in schooling.8 youth share is 60 percent. Because of their larger share in fertil- Impact on growth through ity, young people's decisions about par- demographic trends enthood will shape future demographic The fertility decline witnessed around the trends: as more couples are better able to world in the past 20 years has led to a con- plan their births, both fertility and mortal- centration of births among women ages 15­ ity will decline, and the share of working- 24.Asfertilitybeginstodecline,childbearing age population will rise. With the right set patterns change in three ways: women may of supporting conditions, an economy can delay their first birth, space their births, or reap the benefits for growth of having more stop having children at an earlier age than workers with fewer dependents (chapter 1). previous cohorts. Even though women's In some African countries, such as Chad, marriage age has increased, the average gap however, young women can expect to have between marriage and the first birth has six children or more during their lifetime.11 fallen,9 suggesting very little net delay in In these countries, preparing young couples Forming families 147 for family formation will spur the decline in fertility and in dependency ratios, offering B O X 6 . 2 Education shapes family formation a window of opportunity to benefit from a Young women and men today are more tional attainment of teenage mothers in larger working-age population. educated than previous generations were some countries. In Bangladesh during the when they become parents. Better edu- 1990s, the percentage of teenage mothers Preparation for family cated parents plan safer childbearing and who had ever enrolled in secondary school invest more in their children's education rose from 16 percent to 26 percent.To the formation is poor and health. Compared with the past, young extent that mothers'education is associated Many factors determine when young men people today are also more likely to marry with better health outcomes, children of and women become parents, the number later and have more say in whom and when teen mothers may suffer fewer disadvan- they marry. Some attribute the decline in tages today. of children they have, and how they raise arranged marriages and the shift of mar- Expanding schooling and employment their children. In some settings young peo- riage decision making from parents to opportunities can delay entry into mother- ple make these decisions, while in others, young people to the increased education hood. A study from Guatemala spanning 35 parents or extended family make the deci- of women. years suggests that women delayed child- There is also a strong link between bearing because of increased schooling. sions for them (chapter 2). To some extent, female education and reduced childbearing Using rich panel data, the study shows that these are purely private decisions that do in almost all countries. In some countries, education has a significant causal effect not merit public intervention. However, even a few years of attending primary on age at parenting for women but not school reduces the number of children for men. Every additional grade of school there is a role for public investment in areas ever born, and secondary education has a attainment delayed the mean age of first that will ensure safe passage through a first stronger impact. Mothers'education is also parenthood for females by 0.52­0.87 years. pregnancy and beyond, because young peo- associated with reductions in desired family It also reduced the probability of becoming ple may underinvest in family planning or size and increases in contraceptive use. teenage mothers (before age 18) by 14 to In some countries, the push toward uni- 23 percent. maternal health services relative to the level versal primary education in the 1990s and that might be socially optimal. Sources: Behrman and others (2006); Mensch, the incentives to girls to continue beyond Singh, and Casterline (2005); and National Most governments finance maternal primary have stimulated girls'enrollment Research Council and Institute of Medicine health services because of the positive exter- in secondary school, evident in the educa- (2005). nalities of the improved health of the (as yet unborn) child. Governments also provide these services because of the low demand from the United States suggests that labor on the part of prospective parents who have market outcomes for men are linked to par- little education, little information, and are enthood: fathers' earnings increase when poor. A further justification for government they have children.14 investment in supporting the transition to While the nature of unions may vary, parenthood is to ensure equity. Publicly marriage or forming a union is a key tran- financed or provided services broaden sition in life. For many young people and access to those who would otherwise not be their families, timing of marriage is affected able to avail themselves of the services, par- by economic pressures and expected gender ticularly women and adolescent girls.12 roles (box 6.3). Once married, newlyweds Among the many factors influencing make joint decisions about contracep- family formation is the position of young tive use and the timing of births. Discord women in their parents' households. A posi- between spouses in the demand for children tion of disadvantage can push them into can also affect the couples' use of contra- pregnancy at a very young age, and it can ceptives or maternal health care. also lead to lower investments in educa- In addition to these factors, nutrition and tion, with significant consequences for the reproductive health are important for a suc- transition to parenthood (box 6.2). Young cessful transition to family formation. Good men's labor market outcomes are associated nutrition and reproductive health have big with the timing of transition to marriage payoffs for the young when they make the and parenthood. Evidence from the Arab transition to parenthood.Young women who Republic of Egypt, the Philippines, Thai- are underweight or who suffer from micro- land, and Vietnam suggests that poverty nutrient deficiencies before pregnancy are and a lack of financial security are reasons more likely to have low birth weight infants.15 for men to delay marriage. In rural Ethio- Preventing low birth weight infants brings pia, the unavailability of land is associated very high returns for the child: lower infant with delayed marriage for men.13 Research mortality,better cognitive ability,and reduced 148 WORLD DEVELOPMENT REPORT 2007 young as it once was,but micronutrient defi- B O X 6 . 3 Voices of Bangladeshi youth: ciencies remain common. While the use of Searching for the ideal spouse family planning, maternal, and child health services has increased in many countries, There is considerable agreement on what a lot more discussion on what constitutes in others it remains low. Even where the constitutes an"ideal"partner--of either a good woman, and they seemed quite use of services has increased, women may sex.For both parties, education and good concerned about the"morality"of women character is considered desirable.However, today--particularly of their"easy"female not receive all services, particularly hurt- boys seek girls with good looks and girls colleagues in the garment factories.The ing first-time mothers. Moreover, young seek boys with family wealth or a job.In Kalyanpur basti girls, however, were quite women and men are poorly informed about Sylhet, an"ideal"husband is somebody who cynical in their comments about a good sex and child health.Governments intervene is established and honest and an"ideal"wife husband: "a bad husband is someone who is somebody who is shongshari (good at beats you in public, in front of everyone; a in nutrition, family planning, and maternal household tasks), has good character, and good husband is someone who beats you and child health in almost all countries. The is good looking.The Hindu male group in quietly, at home, so no one realizes." gaps described here suggest that these inter- Chittagong is cynical about what brides look for in a boy: "all the girls want in a husband ventions have to be more effective in reach- Source: Ali and others (2006); Consultation is money--none of the other characteristics ing young men and women. meetings carried out with 23 youth groups matter.If a boy has money, he will get a (ages 10­27) in Chittagong, Dhaka, Rajshahi, bride." The Kalyanpur basti (slum) boys had and Sylhet, Bangladesh, January 2006. Youth suffer from nutritional deficiencies A young woman's height can indicate whether she risks having a difficult deliv- chance of acquiring noncommunicable dis- ery, because small stature is often related ease in adulthood. It also improves labor to small pelvic size. The risk of having a productivity in adulthood, with the eco- baby with a low birth weight is also higher nomic benefits close to $510 per infant pre- for mothers who are short. Low maternal vented from falling into low birth weight.16 weight and micronutrient deficiency before Fathers' nutrition can indirectly affect child and during pregnancy can cause low birth health through the effect on household weight infants.19 income. Well-nourished fathers are more In most developing countries, young productive in the labor market and have girls appear on average to be well nour- higher earnings.17 ished, with heights greater than levels that First-time parents experience health signal obstetric risks (140­150 centimeters). risks--for the mother and the baby. For Other than South Asian countries and a few example, firstborn children have a greater countries in Latin America, the proportion likelihood of dying within the first four of 15- to 24-year-olds shorter than 145 cen- weeks of life, perhaps due to a lack of health timeters is 3 percent or less. Bangladesh, knowledge and to inappropriate care. HIV/ India,and Nepal have the highest prevalence AIDS poses an additional concern for young of young girls who are stunted, ranging people starting a family. Because young from close to 16 percent of 20- to 24-year- women are more likely to marry older men, olds in Bangladesh and Nepal to 13 percent they face a greater risk of acquiring HIV.18 in India (figure 6.3).Among Latin American That is why the incidence of HIV is higher countries for which data are available, Peru among young women than young men in has the highest prevalence, with close to 14 populations where the disease is prevalent, percent of 15- to 24-year-olds shorter than as in some Sub-Saharan African countries 145 centimeters. The prevalence of under- (chapter 5). This pattern will also emerge in weight young women ages 15­24 is less low HIV-prevalence countries,such as India, than 3 percent in Colombia, Egypt, Nicara- where infections are spread mostly through gua, and Turkey, but is high in South Asia. sexual contact. Not having access to relevant In most countries, the percentage of young information, counseling, and testing during people who are overweight is greater than pregnancy increases the risk of mother-to- the proportion underweight (chapter 5). child transmission of the virus. In contrast to their generally good nutri- As the rest of the section shows, under- tional status, young people suffer from nutrition is not as widespread among the micronutrient deficiencies. Anemia, the Forming families 149 Figure 6.3 Young women of short stature risk Figure 6.4 Anemia is highly prevalent among young developing obstetric complications women Bangladesh Benin India Mali Peru Haiti Nepal India Madagascar Burkina Faso Bolivia Ghana Cambodia Cameroon Nicaragua Uganda Ages 15­19 Mozambique Kazakhstan Ages 20­24 Colombia Kyrgyz Republic Bolivia 0 5 10 15 20 Armenia Percentage of women shorter than 145 centimeters 0 10 20 30 40 50 60 70 Prevalence of anemia among females Sources: ORC Macro (2006) and MEASURE DHS STAT Compiler ages 15­24 (percent) (surveys conducted between 1998 and 2005). Source: Demographic and Health Surveys conducted between 1998 and 2004. Note: Anemia is defined as hemoglobin (Hb) content in blood of outcome of multiple micronutrient defi- less than 12 grams/deciliter (includes mild, moderate, and severe ciencies, is prevalent among young people anemia). Adjustments in these cutoff points were made for women living at altitudes above 1,000 meters and for women who smoke, in most developing countries (box 6.4). since both groups require more hemoglobin (Centers for Disease During adolescence, the nutritional require- Control and Prevention (1998). Figure represents married and unmarried young women ages 15­24. ments for iron increase because of rapid growth and so does the risk of iron defi- knew the signs, most reported that friends ciency. Among boys the risk subsides after were the source of the information. their growth spurt.Among girls and women, In Bangladesh few teenage mothers could however, menstruation increases the risk of identify life-threatening conditions dur- iron deficiency throughout the childbearing ing pregnancy. Only about 5 percent knew years.20 Anemia is highly prevalent among about conditions such as severe headaches, young women ages 15­24, including those high blood pressure,and pre-eclampsia,that who are pregnant (figure 6.4).21 In Benin, might threaten the life of the mother during Mali, Haiti, and India, more than 50 per- pregnancy or delivery. Nearly 50 percent of cent of girls are anemic. In Egypt, close to teenage mothers reported not seeking any 30 percent of boys ages 11­19 suffer from assistance for maternal complications.26 anemia.22 In the United States and Europe, Young couples are less likely to use con- the prevalence of anemia among women traceptives than older couples, evident in and children is 7 to 12 percent. the percentage of women by age who report using any method of contraception (figure Young people are not well 6.5). In Peru, 64 percent of 30- to 34-year- informed--and are less old women use contraceptives, nearly seven likely to use key services Sexual and reproductive health knowledge is low among young people.Among sexually active youth in Nigerian schools, awareness B O X 6 . 4 Anemia: The outcome of multiple deficiencies of the risk of pregnancy from the first sex- Anemia, characterized by a fall in the con- Women with severe anemia can experi- ual encounter is very low.23 Nor are young centration of hemoglobin in the blood, ence difficulty meeting oxygen transport people able to identify the time of month arises from a deficiency of iron, folate requirements near and at delivery, especially when the risk of pregnancy is highest. Even (vitamin B9), vitamin B12, and other nutri- if their blood loss is severe.This may be an ents. Vitamin A deficiency is also known underlying cause of maternal and infant married girls, who are most likely to be reg- to increase the risk of anemia. Many deaths.Iron deficiency also affects physical ularly engaging in sex, were no more knowl- other causes of anemia--hemorrhage, work capacity, in both men and in women, edgeable than unmarried girls.24 Of young infection, genetic disorders, and chronic but no studies have focused on adolescents. people ages 15­24 in Indonesia,25 21 per- disease--have been identified. However, Sources: DeMaeyer (1989); International Nutri- nutritional deficiency, primarily due to a cent of girls and 28 percent of boys did not tional Anemia Consultative Group (1979); Inter- lack of iron in the everyday diet, accounts national Nutritional Anemia Consultative Group know any of the signs of puberty's physical for most cases. (1989); Li and others (1994); and Yip (1994). changes for the opposite sex. Of those who 150 WORLD DEVELOPMENT REPORT 2007 Figure 6.5 Young women are less likely to use Figure 6.6 Young mothers do not get full care during contraceptives than older women antenatal visits Percentage of sexually active young women, married Rwanda and unmarried, who use contraceptives Chad Burkina Faso Dominican Republic Uganda 80 Peru Benin Turkmenistan Ethiopia Kenya Mali 60 Cambodia Indonesia Haiti Kenya Cameroon 40 Morocco Egypt, Arab Rep. of Zimbabwe 20 Namibia Philippines Bangladesh 0 Armenia 15­19 20­24 25­29 30­34 Mozambique Age Nepal Ghana Sources: ORC Macro (2006) and MEASURE DHS STAT Compiler Dominican Republic (surveys conducted between 2000 and 2005). Bolivia Malawi Nicaragua times the proportion for 15- to 19-year-olds, Colombia perhaps because young couples want to have Peru children soon after marriage. Or it could be 0 10 20 30 40 50 60 70 80 90 that family planning services emphasize Percentage of 15- to 24-year-old women who have received methods that help women stop births (for antenatal care and were informed example, sterilization) rather than space of pregnancy complications births, so older women ready to stop may Source: Demographic and Health Surveys conducted between 1998 and 2003. "I don't think information about be more likely to use these services than Note: Data shown are restricted to women who used antenatal bearing and fostering children I younger women just starting their families. care for their latest infants during the three years prior to the survey. have obtained is enough, and we Another constraint for unmarried women can't obtain this information from is the difficulty of obtaining contraceptives. school and family." In some countries, laws regulate young In many countries in Latin America and women's access to contraceptives. Restric- Sub-Saharan Africa, commensurate with University student, China tions may include minimum age require- the general high use of antenatal care, the December 2005 ments and requirements that young women percentage of young women receiving such be married and have spousal approval.27 care is also high, ranging from 80 percent In Bangladesh, India, Mali, and Pakistan, to 100 percent.30 However, even where the the use of maternal health services, particu- use of antenatal care is high, young mothers larly antenatal care, is low among young who receive antenatal care do not receive full women. Price, including travel and wait- care. First pregnancies are at a higher risk of ing time, determines health care use by all neonatal mortality, and informing women of women.28 For young women and first-time potential complications is an important com- parents, a lack of knowledge about the need ponent of care for young mothers. In most for preventive care during pregnancy could countries for which data are available, young be an additional factor. Girls' lower bargain- mothers who used antenatal care were more ing power also reduces their ability to nego- likely to get checkups, such as measurements tiate with their husbands and in-laws about of blood pressure, but unlikely to be told the need for care during pregnancy. In rural about pregnancy complications (figure 6.6). Pakistan, the mobility of adolescent girls is In many countries where anemia is prevalent, highly constrained, making it difficult for iron supplements are offered as part of ante- them to seek services.29 natal care. But in Cambodia, where anemia Forming families 151 affects more than half of all young women, vices friendlier for young mothers has been fewer than a quarter of mothers received iron tried in several countries, but no rigorous supplements during antenatal care.31 evaluations are available.33 An example of Parenting skills help parents interpret a promising program is a hospital-based infant and young child behaviors, as does breastfeeding program for adolescent knowledge about their health, nutrition, mothers in Mexico that was associated and developmental needs. Young mothers with increased antenatal visits by program and fathers tend to be less aware of signs recipients.34 of childhood illnesses and of ways to treat Because mobility can be a constraint, them. Knowledge of oral rehydration ther- outreach services (rather than fixed-site apy (ORT), a simple and effective response delivery) can also make a difference. In to a child's dehydration during episodes of such settings, providing mass-media health diarrhea, remains low in many countries, information campaigns that reach all in particularly among young men and teen- the community can stimulate the demand age mothers. In Peru, only about half of all for services. There are no health outreach teenage mothers were aware of ORT salts, programs that target youth, but the success compared with 80 percent of 25- to 29- of programs targeting all couples in their year-old mothers. Even in Indonesia, where childbearing years suggests that outreach awareness is close to 100 percent among can be effective.35 The doorstep delivery older mothers, only 85 percent of teenage program in rural Bangladesh (Matlab dis- mothers knew about ORT salts. In Kenya, trict) significantly increased the uptake of only 40 percent of young men ages 15­24 antenatal and postnatal care services.36 The had heard of ORT, compared with 60 per- Lady Health Worker Program in Pakistan cent of older men.32 has been effective for women of all repro- ductive age groups.37 Providing opportunities for Increasing men's uptake of reproduc- youth to prepare for parenthood tive health and family planning services improves their health and the likelihood Opportunities for young men and women that couples are protected from sexually to become better prepared for family life transmitted infections (STIs) and unin- "If young people have free access to can be broadened by improving their access tended pregnancies (see the spotlight on family planning methods, we could to family planning, maternal, child health, Brazil). Men who need care for sexually understand the consequences of our and nutrition services. Financial incentives transmitted diseases may not seek it from actions and could be more conscious can also increase the opportunities for peo- maternal and child health and family plan- about our behavior, and could be ple, particularly the poor, to use health ser- ning clinics, largely viewed as women's more careful." vices. Most of the programs have not been services. They may be more attracted to specifically targeted to young people, but Young person, Cuzco, Peru separate facilities that can provide them they offer promising approaches to prepar- January 2006 with STI and family planning services. ing young people for parenthood. Efforts to Engaging the private sector in public- prevent early marriage can broaden young private partnership can improve the avail- girls' opportunities to avoid early mother- ability of reproductive health services for hood and help them avoid all the associated women, with limited involvement and adverse health consequences. resources from the government. If govern- ments find it controversial to deliver such Improving access to health services services to young people, contracting them Young men and women are less likely to out can reach young women, as in Colom- use family planning, maternal, and child bia.38 The social marketing of condoms health services--access being a key issue. prevents STIs and HIV, but the few evalu- One way to improve access is to provide ations available do not provide evidence "youth-friendly" services where providers on preventing unintended pregnancies.39 are trained in catering to young people's Promoting condoms as "dual protection" needs. Making antenatal and postnatal ser- rather than only as "safe sex" may increase 152 WORLD DEVELOPMENT REPORT 2007 the uptake of condoms and protect both acid as supplements to pregnant women to married and unmarried young people from prevent anemia during pregnancy. Because sexually transmitted infections and unin- pre-pregnancy nutritional status has a sig- tended pregnancies.40 nificant effect on the newborn's health, As discussed in chapter 5, integrating STI nutritional measures should also target girls "I fear that now, especially living in and HIV services with reproductive health before they become mothers. Although pro- rural communities, people do not services can encourage greater use of both. viding supplements over a long period has have access to much information. This is particularly important in Sub-Saha- proven difficult, because they have to be They are less privileged and all ran Africa, where HIV prevalence is already taken daily and they sometimes have side the time it is only the urban high--and in India, where prevalence, effects, school-based iron-supplementation communities that have the though low, is increasing among young programs have been found to be effective.44 sensitization campaigns." married women.Integration can ensure that Results from a recent survey by the Part- young women receive counseling about HIV nership for Child Development of school Jestina, rural youth activist, using a and mother-to-child transmission of the health policies in selected countries shows nontraditional music campaign, virus when they go for antenatal checkups. that although a number of countries offer Sierra Leone Few women, however, receive such counsel- iron supplementation to school children, February 2006 ing or even opportunities for testing. The many countries such as Benin and Camer- feasibility and effectiveness of integration, oon, where close to 50 percent or more of given the resource constraints facing most young girls are anemic (figure 6.4), do not developing countries, are debatable--and offer such services in schools.45 no evaluations are available.41 Nutrition services must be an impor- tant dimension of antenatal and postnatal Improving access to care services for teen mothers. Pregnant nutrition services teenagers are at high obstetric risk, par- The World Health Organization recom- ticularly if short or underweight before mends that if an adolescent is still growing, pregnancy. Close monitoring of teen- adequate weight gain and nutrient intakes age mothers' nutritional status has been must be ensured to prevent poor pregnancy recommended, because adequate weight outcomes. Because iron deficiency is often gain may even be more critical for them accompanied by other micronutrient defi- than for older mothers.46 Nutrition and ciencies (vitamin A, folate), food-based weight monitoring are not always easily approaches are likely to improve young implemented, and health providers must people's diets. Fortifying foods and provid- be able to give women context-specific ing supplements are fairly inexpensive--and dietary advice. In addition to iron supple- successful--ways of reducing micronutrient mentation, vitamin A, zinc, and calcium deficiencies. Where anemia is highly preva- supplementation can also be particularly lent, food fortification may not be sufficient, beneficial for teen mothers, because they and iron supplements may be necessary.42 are at a higher risk of pregnancy-induced School-based iron supplementation pro- hypertension and pre-eclampsia. Teen grams can be effective in reaching adoles- mothers might also need postpartum cents in countries where enrollment rates are nutritional care, such as diet counseling high. It has been estimated that the benefit- and support for breastfeeding. cost ratio of iron supplementation for sec- ondary school students ranges between 26 Offering financial incentives and 45--that is, one dollar invested in iron Conditional cash transfers have been effec- supplementation will yield 26 to 45 dollars tive in increasing the use of preventive in return.43 health care by poor households. Although Information on anemia in the population these programs have not been targeted is often lacking, and surveys that measure to young parents, they offer a promising anemia prevalence can help in developing approach to increasing the uptake of pre- health intervention programs to prevent ventive health services by first-time parents. it. Many countries provide iron and folic Mexico and Nicaragua have provided cash Forming families 153 transfers conditional on household mem- monetary investment of Rs. 2,500 ($80) in bers' participation in health and nutrition government securities within three months workshops and on visits to the health center. of her birth, which she can claim when she Mexico's Oportunidades increased nutri- turns 18, if still unmarried. In 1995, the tion monitoring, immunization rates, and Haryana government expanded the scheme antenatal care visits. Growth-monitoring by offering a higher maturity amount (from visits increased by an estimated 30 to 60 Rs. 25,000 [$800] to Rs. 30,000 [$960]) for percent, and children under 5 had fewer ill- girls who agree to defer cashing in their nesses than children outside the program.47 securities. The program has not been evalu- Nicaragua's Red de Protección Social also ated for short-term outcomes or long-term promotes children's participation in nutri- objectives.51 tion monitoring and the timely immuniza- tion of children.48 Strengthening young people's Preventing early marriage decision-making capabilities to prepare for parenthood A delay in early marriage, one way to prevent teen pregnancies, is likely to yield benefits Young people need good information to for child health as well. For example, in Gua- make better choices about the timing of temala delaying marriage for girls improves births,the health services to use,and the right the chances of their children surviving past child care and feeding practices. Programs to age 5.49 Many countries have laws specify- strengthen their decision-making capabili- ing the minimum age for girls and boys to ties in reproductive health, nutrition, and the marry, with or without parental consent. In care of young children include health educa- 50 of 81 countries examined, the minimum tion, parenting and early child development age for marriage is at least 18 for both males services, and life skills education--for young and females,50 and in 32 countries it is lower men as well as young women. for girls than for boys. Over time the legal minimum age at marriage for girls has risen, Providing health information but it is difficult to enforce where vital regis- to young men and women tration systems are weak. A minimum age is Informing young people can be effective in also more likely to be effective when young preparing them for the transition to parent- girls, particularly those from poorer house- hood. Many governments offer such infor- holds, have opportunities to attend school mation as part of their school health program and improve their livelihoods. as well as under broader nutrition programs. Norms about marriage age are culturally sensitive issues, and where there are strong Sex education to prevent early childbearing. taboos against premarital sex, daughters' Sex education programs delivered to unmar- early marriage might appear to be a desir- riedyouth--whetherschoolbasedorthrough able option for parents. Efforts to prevent massmedia--canincreaseknowledgeamong early marriage must therefore involve par- young women and men. Knowledge, how- ents and the community as well as young ever, may not be sufficient to change behav- people themselves. One such program is ior (chapter 5). As described in chapter 5, the Apni Beti, Apna Dhan ("our daughter, most evaluations of sex education programs our wealth") scheme in the Indian state of have relied on self-reported behavior, which Haryana, launched in October 1994 to raise may not reveal the true program impact. awareness about the importance of the girl- However, evidence from impact evaluations child and to reverse gender discrimination. suggests that sex education can be effective It honors mothers of girls with a small mon- in changing behavior. In Kenya, an impact etary award (Rs. 500, or $16) to cover post- evaluation of an intervention that provided delivery needs of the mother for the birth female primary students with sex education of a daughter, paid within 15 days of birth. that included specific information about the It also endows each girl with a longer-term risk of getting HIV from sex with older men 154 WORLD DEVELOPMENT REPORT 2007 reduced teenage childbearing.52 In Chile, from low or middle socioeconomic status, the school-based sex education intervention and promoted their involvement in the preg- Adolescence: Time of Choices increased the nancy, preparation for the delivery, and any use of contraceptives and reduced the inci- potential emergency through various mass dence of teen pregnancy.53 media campaigns and training programs. Desa Siaga focused on getting the whole Reproductive health education programs for community involved in safe motherhood, couples. Reproductive health programs arranging transport to hospitals, providing provide health education to married cou- funds, donating blood, and being alert to ples, but few evaluations are available.54 emergencies during childbirth. Although not rigorously evaluated for impact, moni- · A program in Bangladesh provides new- toring reports show that husbands and lyweds with reproductive health infor- wives exposed to the Siaga programs were mation and services before they have more likely to have more knowledge of signs children. All newly married couples are of emergency than their unexposed coun- "In order to help our students to registered by a family planning field- terparts and more likely to have delivered at gain better knowledge, we organize worker during a home visit, establishing health facilities or with midwives.55 events at SOS village [space for a relationship with the couple and pro- free discussion on reproductive viding the opportunity to deliver family Nutrition education. School-based health health issues]." planning information. The fieldworker education programs can encourage healthy also provides referrals to health clinics Do, 22, male university student eating and physical activity. Such programs for maternal and child health care. in Hanoi and chairperson must include messages promoting the con- The Population Council's First-Time of the Reproductive Health · sumption of a variety of fruits and vegeta- Parents Project in two cities in India Club of his university bles--and moderation in saturated fats.56 provides reproductive and sexual health March 2006 Young people must also be encouraged to knowledge targeted not only to married consume foods rich in iron.Educating moth- young girls but also to their husbands, ers about rest during pregnancy and appro- mothers, mothers-in-law, health care priate child feeding practices have been part providers, and the community. of successful nutrition programs in India · A community-based approach to mar- (Tamil Nadu), Indonesia, and Thailand. ried girls' reproductive health in the Like all health education programs, bring- Indian state of Maharashtra tests the ing about change in diet habits through nutri- effectiveness of delivering information tion education is difficult (chapter 5). An through community-based organiza- evaluationoftheBangladeshIntegratedNutri- tions along with improving the quality tion Program (BINP) raised knowledge about and content of public services by train- appropriate nutrition-related behaviors, but ing health providers. It also targets girls' most mothers did not practice them.57 Pov- husbands and mothers-in-law. erty and time constraints were a major reason. Nor did the program adequately involve hus- Information targeted to men. Providing bands and mothers-in-law, perhaps prevent- information to men about safe motherhood ing many young mothers from practicing new and child health services can increase the behavior. Some behavioral change efforts, couples' uptake of maternal and child health such as hand washing to prevent diarrhea, services (box 6.5). The Suami Siaga ("alert were found to be effective.58 This program husband") campaign in Indonesia shows also had innovative components, such as tar- that mass media campaigns can increase geting adolescent girls through the creation of husbands' involvement in safe motherhood. Adolescent Girls Forums. In one subdistrict Suami Siaga and Desa Siaga ("alert village") covered by the BINP, newlyweds were tar- were part of public awareness campaigns geted to test whether it is more cost-effective implemented with the five-year safe moth- to address first pregnancies and work with erhood program. Between 1998 and 2002, couples until their child's 24th month, rather Suami Siaga targeted husbands ages 15 to 45 than aiming to cover all young children and Forming families 155 pregnant women in a community. Results from an evaluation of this newlyweds initia- B O X 6 . 5 Grameen Bank's "Sixteen Decisions"-- tive is not yet available.59 convincing men to have fewer children A study estimating the impact of male and that are part of the microcredit programs, Provide parenting skills and early female participation in microcredit pro- activities that may have altered men's childhood development services grams in rural Bangladesh found that men's attitudes. Grameen Bank teaches its partici- participation in the program reduced fertil- pants the value of small families (among Parenting skills can improve child develop- ity. Among the four microcredit programs other social issues, such as girls'education). ment. One such skill is knowledge of when an in the study, participation in Grameen Bank Having a small family is one of the"sixteen infant is ready for complementary feeding. In had the largest effect--surprising because decisions"that members must promise to addition to this skill, parents must also follow men spend more time working and less obey.The effect of men's exposure to such time childrearing. So the effect for men messages highlights the importance of tar- practices such as active or interactive feeding, could not be the result of greater livelihood geting men in information and education selecting foods suited to the child's emerg- opportunities from microcredit. campaigns for family planning and repro- ing motor capacities and taste preferences, Authors attribute the finding to men's ductive health. and talking and playing with the child dur- exposure to social development activities Source: Pitt and others (1999). ing the meal. Early childhood development programs in Ecuador and Jamaica show that responsive and interactive parenting to sup- women. For example, participation in port psychological development in children microcredit programs has increased the use can offset many of the adverse consequences of formal health care by women. In urban of childhood malnutrition on cognitive Malaysia, women's unearned income (a development.60 In Jamaica, nutritional sup- proxy for their bargaining power) increased plementation for undernourished children the demand for maternal care.65 and psychosocial stimulation improved men- Evaluations based on randomized con- tal development. trol experiments of youth-focused inter- Formal child care services can also support ventions in this area are rare, but impact mothers who want to work.Extensive research assessments of a few programs provide from the United States suggests that the price some guidance for effective interventions. of child care affects mothers' labor force par- One is the Better Life Options program in ticipation.61 In urban Guatemala, the higher India,66 initiated by the Centre for Devel- price of formal child care facilities reduces opment and Population Activities in 1987. hours worked by mothers. Children attending Targeting out-of-school young women ages the Hogares Comunitarios child care facilities 12­20, it offers various services in periur- in urban Guatemala had better dietary and ban slums and rural areas--offering knowl- micronutrient intakes than their counter- edge of reproductive health and services, parts who did not.62 In Vietnam, where aver- providing vocational training, and promot- age fertility and family size are low, 41 percent ing women's empowerment through recre- of urban mothers use formal sources of child ational events. Results from treatment and care (schools and institutional care) while 46 control group comparisons indicate that percent of rural mothers rely on extended the program improved the welfare of young family members.63 In Kenya, the high costs of women--delaying marriage, increasing early childhood development programs dis- knowledge of reproductive health, strength- courage households from using formal child ening decision-making skills, and increasing care facilities and reduces mothers' participa- the use of health care services (figure 6.7). tion in work. The school enrollment of older A quasi-experimental study in Nepal children, mainly girls, is also affected.64 reveals that an integrated reproductive health program targeted to youth ages Teaching young women life skills 14­21 had a large significant impact on Most life skills programs for girls, married behavior, such as the use of reproductive or unmarried, also provide health informa- health care services, teenage pregnancy, tion, family life education, and livelihood and marrying young. A distinctive feature training. Gaining such skills empowers of the intervention was that it also involved 156 WORLD DEVELOPMENT REPORT 2007 Figure 6.7 In India, acquiring life skills can Supporting those who become stimulate young women's demand for health services mothers at an early age Did not receive life Girls who become mothers at a very young skills training age need to overcome consequences such as Received life skills training interrupted schooling. Young mothers may Received antenatal discontinue school because of lack of fam- care ily or community support and the physical Received two doses of tetanus toxoid demands of pregnancy and childbirth. Oth- Received postnatal ers may drop out of school when they marry, care then later face difficulties in finding paid Married at age 18 work and earning a living. Because most or above teenage mothers are from poor households, Delivered baby in second-chance programs must address their health institution disadvantages. Currently use contraception Flexible school policies 0 20 40 60 80 100 Flexible school and social policies can miti- Percentage of married women gate the adverse effects of teen pregnancy. In Source: Center for Development and Population Activities (CEDPA) the United States, where teen pregnancy is (2001). among the highest in the developed world,71 it is often regarded as a public health prob- adults, teachers, and health care providers in lem because teen mothers and their children assessing the needs of youth and designing are also more likely to have higher poverty delivery mechanisms. This may have con- rates and greater dependence on the welfare tributed to the success of the program. 67 system. Like the Better Life Options program, Some U.S. studies find a significant causal the life skills program of the Indian Institute impact of teenage childbearing on schooling for Health Management in Pachod in rural and earnings, while others find that a good Maharashtraoperatesinruralareasandurban part of the consequences can be attributed slums and targets out-of-school females ages to prior social and economic disadvantages 12­18. Offering a one-year course one hour and not to teenage childbearing. The results, each weekday evening, led by women trained rather than being contradictory,might reflect in health and nutrition, literacy, and life skills, different periods of time.72 In the 1960s and it had a significant impact on delaying mar- 1970s, when social conditions made it dif- riage for the young women.68 ficult for girls to cope with pregnancy, teen mothers faced irreversible consequences. Giving young women Over time, better access to second-chance resources to delay marriage programs ensuring school continuation for Interventions that encourage girls' school- teen mothers may have reduced the causal ing--scholarships, vouchers, free books, and impact of teen pregnancy on a range of out- uniforms--can also discourage early marriage comes. High school equivalency programs and hence early pregnancy. The well-known and welfare programs help teen mothers secondary school stipend program in Ban- make up for their low income and catch up gladesh (box 6.6) is promising because girls' with their schooling. School systems also average age at marriage is so low there. But it adapted to the education of pregnant and is not clear, because of the lack of a compre- parenting teenagers, and this might have hensive evaluation, whether it delayed mar- kept them in school. riage for girls--a study of two villages found Evidence from South Africa suggests that that it did.69 Better employment opportuni- such supportive schooling policies helped ties for young women, such as the increased teen mothers catch up and complete their job opportunities in garment factories in Ban- education.73 More countries in Sub-Saharan gladesh, can also delay marriage.70 Africa and Latin America allow for more Forming families 157 liberal reentry policies, and some even allow pregnant girls to remain in school during B O X 6 . 6 Cash transfers conditional on delaying pregnancy (Burkina Faso, Cameroon, Chile, marriage to promote school attendance Peru). This is an improvement from the for girls in Bangladesh 1990s, when most countries in Sub-Saharan In 1977 a local NGO in Bangladesh began a provide occupational training to girls leav- Africa required expulsion of pregnant girls. small project to provide secondary school ing school. The implementation of these policies has stipends to girls who had completed pri- According to an operational evaluation, not been documented, and their effects have mary school, on the condition that their the project increased girls'enrollments. not been evaluated.74 parents agreed to delay their marriage. In There were serious concerns, however, 1994, with support from the International about the impact on educational perfor- Development Association, the program mance. In 1999, only a quarter of the girls Integrated programs to meet the evolved into the Female Secondary School who received stipends in grade 10 passed diverse needs of a teen mother Assistance Project (FSSAP), covering all 460 the secondary school certificate exam, less rural subdistricts in the country. than the nationwide secondary school com- Since 1977, the Women's Center of Jamaica The conditions for continuing participa- pletion rate for girls.This could be because Foundation has supported unmarried teen tion were that girls would agree to some schools may have inflated enrollment and attendance data to meet performance mothers in an integrated program that · attend school for at least 75 percent of targets.The evaluation also found no evi- meets many needs of very young mothers.75 the school year, dence that the program led to a rise in girls' · obtain at least 45 percent marks on aver- Offering teen mothers a chance to complete age at marriage. age in final examinations, and their education, it encourages young girls The first phase of the FSSAP neither · remain unmarried until completing the collected baseline data nor established to avoid repeating pregnancy during their secondary school certificate exam. an external control group, which made it teenage years. It also offers them vocational Each recipient was allowed to with- impossible to carry out an impact evalua- training and day care. draw cash from the bank independently. tion.To remedy this, a rigorous evaluation The program provides formal schooling An extensive information campaign was component was added in the second phase of the FSSAP (initiated in March 2002). for pregnant girls ages 12­16--and personal conducted to raise public awareness of the importance of female education.The proj- and group counseling about the challenges Sources: Bhatnagar and others (2003); Khand- ect also took steps to enhance the school ker, Pitt, and Fuwa (2003); and World Bank of teen pregnancy and motherhood. It infrastructure, recruit female teachers, and (2003c). makes referrals to local hospitals and clinics for health services, including family plan- ning. It also offers practical services to sup- port young mothers during and after their pregnancies, such as day care for infants, of countries and that iron supplementation classes in parenting and child nutrition, and programs may not always reach them. Few information about women's and children's countries where anemia is a problem have a legal rights. It also provides job skills train- national program of iron supplementation. ing and vocational training and placement Use of antenatal care, during which iron for women ages 18­24. No rigorous evalua- supplements are typically offered to ane- tions are available, but the program appears mic mothers, is far from universal in some to have improved the lives of teen mothers regions, particularly South Asia. Even where in Jamaica. Program benefits were transmit- most mothers use antenatal care (and where ted across generations: all children of pro- anemia is common),only a small percentage gram participants were enrolled in school, report receiving iron supplements. There is and none of the teenage daughters of par- also scope to improve young girls' access ticipants had pregnancies. Most girls in the to iron supplementation through school- program had only one child.76 based health programs. This chapter has also identified countries that need to pay more attention to interven- Many policies and programs for nutrition tion in increasing knowledge on reproduc- and reproductive health can prepare young tive health. On average, a high percentage of people to form families (table 6.1). Coun- young people use reproductive health ser- tries with widespread anemia must give vices such as antenatal care in medical clin- priority to nutritional interventions. This ics. However, of those who used the service, chapter has shown that anemia is highly the percentage with critical knowledge about prevalent among young women in a number topics such as pregnancy complications and 158 WORLD DEVELOPMENT REPORT 2007 Table 6.1 Programs and interventions that prepare youth for transition to family formation Proven successful Promising but unproven Unlikely to be successful Opportunities Improving access to Conditional cash transfers for use preventive Reorienting reproductive health, family planning services health services (Mexico and Nicaragua) (first- services, and safe motherhood services to youth needs time parents were not the focus of program) · Training providers to deal with youth Micronutrient supplementation and food · Family planning outreach (doorstep delivery) to youth fortification for children and for young women before and during pregnancy Engaging the private sector Family planning and maternal and child health · Contracting out family planning services in some programs (not targeted to young mothers) countries (Profamilia in Colombia) · Public-private partnerships · Social marketing of contraceptives Integrating STI and HIV services with family planning and maternal and child health (integrating condom distribution) Increasing men's uptake of reproductive health and contraceptive services Preventing early Legislation setting a minimum age at marriage; banning marriage child marriage Delaying girls' marriage by offering financial incentives to parents (for example, Our Daughter, Our Wealth program in Haryana, India) Capabilities Providing health and School-based sex education to prevent teen Reproductive health education and education about Programs offering nutrition education pregnancy (Chile, Adolescence: Time of Choices) safe motherhood and child health to information that is a) general in content; b) not Nutrition education to mothers to improve child · Young pregnant girls (Mexico, hospital-based culturally relevant nutritional status through feeding practices programs) (hand washing in rural Bangladesh) (not · Newlyweds (Bangladesh Integrated Nutrition Program, necessarily targeted to young mothers) Bangladesh Newlyweds Program; India, First-time Parents Project, Community Based Approach to Married Girls' Reproductive Health Project) · Men (Suami Siaga in Indonesia) Nutrition education to improve young people's dietary intakes, especially those programs directed to teenage mothers Teaching parenting Early child services and responsive parenting skills skills (Jamaica and Ecuador) (not targeted to young or first-time parents) Empowering young Conditional cash transfers to young women (Bangladesh women Female Secondary School Stipend Program) Life skills plus livelihood training--(Better Life Options Program in India) Second chances Supporting teen School policies allowing pregnant girls to continue in mothers with flexible school or to return after delivery school policies School equivalency programs Integrating programs Combining child care and the opportunity to learn livelihood skills (Women's Center of Jamaica Foundation) Forming families 159 infant care is low in many African countries grams have three features in common (table and some countries in Latin America and 6.1). First, they target youth and have youth- South Asia (see figure 6.6). These countries friendly components. Second, they focus must invest further in quality and delivery on more than the transition to parenthood mechanisms for reproductive health care because this transition is interlinked with services. other transitions such as school and work, Most programs that have been shown to both associated with socioeconomic back- enhance reproductive health effectively do ground. Third, because transition to forming not explicitly focus on youth (table 6.1). This families involves multiple decision makers in chapter has highlighted some of the inter- various cultural settings, they involve not just ventions that can help to prepare youth for the young couple, but also parents, teachers, the transition to parenthood. Promising pro- caregivers, and the community. Exercising citizenship 7 Consider what young people can accom- be a citizen is to be a member of a political plish: Days after the October 2005 earth- community and to enjoy the privileges and quake in Pakistan, news began circulating protections, as well as the incumbent obli- c h a p t e r that coordination problems among relief gations, associated with community mem- agencies were keeping aid from reaching bership.2 Citizenship has both passive and those who needed it most. A group of 24 active dimensions. Individuals, simply by students from Lahore University of Man- being community members, receive rights agement Sciences volunteered to be the first and privileges--the right to a free educa- surveyors of devastated villages. Sleeping in tion, the right to a legal identity, and, in lib- tents, traveling by foot and in borrowed cars, eral democracies, the right to vote, to a fair these young men and women--between the trial, and to associate with others. They also ages of 18 and 22--surveyed 3,500 house- take on obligations to pay taxes, and to serve holds, assessing needs and delivering sup- in the military where required. Active citi- plies. Since then they have conducted a zenship emphasizes how individuals should second, more in-depth survey of 200 house- hold public officials accountable for their holds--covering some 32,000 individuals actions, demand justice for themselves and to date. Their data helped donors and relief others, tolerate people who are ethnically or agencies target their efforts and save lives.1 religiously different, and feel solidarity with It is during their youth that people start their fellow citizens and human beings. to participate in social and political life on Public action to nurture good citizens their own,and decide which skills to acquire, is important because markets cannot do it where to work, and with whom to develop alone--nor can mere social participation intimate relationships. They can vote for (too many social institutions are exclusion- the first time. Many choose, or are required, ary or worse). It is also important because to serve in the military. Some join clubs or collective action, public accountability, sports teams. They might decide, along with caring for kin and community, and envi- others in their religious institution, to pro- ronmental stewardship are so much more vide care for extremely sick AIDS patients difficult without an active citizenry. Even if in their community. They might deliberate Kant was right when he wrote that the prob- over whether to assume the debt a neighbor lem of good government can be solved even owes to the village moneylender. Some have "for a race of devils," the solution would be staged protests because fees at their schools expensive and unpleasant (box 7.1). were too high, others because community This chapter examines the transition leaders ostracized a supposedly immodest that young people undergo as they become young woman. All these roles--both social citizens and the policies that might support and, in the broad sense of the word, politi- them, first reviewing data on youth citizen- cal--are aspects of citizenship. ship at the global level. Although young Citizenship is an ideal in social move- people might be growing less interested in ments and political life, but the meaning politics and more disaffected from main- of the term is elusive because almost any stream institutions in high-income and many relationship between individuals and com- middle-income countries, that does not munities can be cast as an aspect of citi- appear so in several low-income countries, zenship. The simplest definition is that to where interest in politics, and confidence 160 Exercising citizenship 161 B O X 7 . 1 What is citizenship? Historically, the status of citizens was con- goods, such as employment and access to Zealand, Uruguay), as the numbers of undocu- trasted to the state of slaves and others sup- basic services--and not only political and civil mented aliens, asylum seekers, and refugees posedly without the capacity to use reason to immunities and privileges.3 increase, and as countries carve out export pro- shape the common destiny of communities. That is the ideal of citizenship.The reality, cessing zones and other areas with different The ideal of citizenship has also, at various of course,has always been,and continues to legal rights and obligations.5 times, and in various places, emphasized equal become,more complex.For example,de facto The ideal of universal, equal citizenship participation in and obedience to the com- legal rules throughout much of Africa ascribe does not specify the state's relationship to munity, advocated participation in social life, rights--regarding not only religion and family ethnic minorities. Should subnational electoral and described a zone of private life that the matters but also land ownership and economic and administrative boundaries correspond to state ought not violate. Citizenship also has opportunities--to the chiefs of ethnic and lin- ethno-linguistic divisions? And what languages been used to refer to a shared way of life and guistic groups,not to individuals.This practice should be taught in schools?6 Some states the willingness to defend it against outsid- reflects the colonial distinction between indi- have stripped ethnic minorities of national citi- ers. Citizenship is now often used to suggest vidualist rights for urban residents and group- zenship to shed the costs of providing services human dignity, and many social and political based privileges for"peasants."4 to groups that do not support the dominant movements, such as movements for health and The two main elements of contemporary regime, making hundreds of thousands state- education rights, are cast as efforts to enhance citizenship--shared identity and rights--are less in the last decade.7 In many countries, citizenship.The rights associated with citizen- becoming unbundled as legal resident aliens women continue to face legally sanctioned ship, particularly in developing countries, have receive rights of access to basic public services and cultural barriers to own and inherit in recent decades expanded to encompass the in many countries (and even the right to vote property, participate in politics, or receive an requisites of freedom--social and economic in national elections in Chile, Malawi, New education. in the civil service and private business are tion for youth and that a young person's increasing. identity emerges through recognition from The chapter next explains why youth those who count. Possessing a legal identity citizenship is crucial for development out- and having work are both important. The comes. The youth experience of citizenship clearest and most equitable policy for pro- is formative and has lasting effects on the moting the capability of young people is extent and kind of political participation to make sure that all young people possess throughout life. Citizenship affects devel- legal identities as full citizens. Two specific opment outcomes through three channels: policies can promote youth capabilities in by enhancing the human and social capital citizenship: civic education and programs of individuals, by promoting government of youth development and youth action. accountability for basic service delivery, and The absence of agency can lead youth to by enhancing the overall climate for invest- choose negative social roles, including gang ment and private decision making. membership and participation in personal The chapter then moves to the three pol- and political violence. icy areas developed throughout this Report: Because many young people are attracted opportunities, capabilities, and second to, and experiment with, social defiance-- chances. The opportunities available for and because many governments and societies youth to develop active citizenship depend fail to protect youth--legally recognized sec- on the principles and customs that structure ond chances are crucial. The chapter analyzes the political and social spaces throughout a policies to give second chances to young peo- nation. Countries have promoted youth cit- ple who have committed crimes and to child izenship in several ways, including lowering soldiers; young offenders can benefit from the voting age and establishing youth coun- restorative justice programs, and former child cils and consultative bodies, military ser- combatants from assistance in reintegrating vice, and national and community service. with their home communities. Such opportunities for political and social participation, if flexible and well-designed, Youth participation: can support active citizenship. The section on capabilities develops the Rising,declining,or both? notion that the adoption of political and Concerns about the civic dispositions of social roles is a process of identity forma- young people are not new. In the eighth 162 WORLD DEVELOPMENT REPORT 2007 century BCE Hesiod observed: "I see no Figure 7.1 Interest in politics is on the rise among hope for the future of our people if they are youth ages 18­24 in low-income countries dependent on the frivolous youth of today, Percentage of youth very or rather interested in politics for certainly all youth are reckless beyond 80 words." Condorcet argued in 1782 that, as a 1990 2000 rule, every generation appears less virtuous 60 than its predecessors. As always, the virtue of young people is a preoccupation in many, 40 probably most, places in the world. How well-founded are concerns about 20 "Although youths want to contemporary youth? Are young people less participate in politics and raise involved citizens than their parents? Citizen- 0 their concerns and priorities, they ship is a composite of complex and cultur- a* China India* Fed.* are unable to do so because political ally differentiated identities, attitudes, and Nigeria* Turkey* Finland* Argentin Slovenia*Kingdom leaders have their own agendas and behaviors. For youth in high-income coun- Russian United self-interests, which detaches youths tries, some measurable declines--in politi- Low-income Middle-income High-income from active involvement in politics cal participation, interest in politics, and countries countries countries and social life." membership in civic organizations--have Source: Authors' calculations from World Values Survey 1990­2000 (Inglehart and others (2004)). been widely documented. Youth sections of *Difference significant at < 5 percent. Chandan, 20, Nepal political parties in Belgium have lost more January 2006 than 60 percent of their numbers since the 1980s. Membership in youth organizations your life? How often do you discuss poli- in Sweden fell from 220,000 in 1972 to less tics with friends?--exhibit the same trends. than 50,000 in 1993. Almost all the decline Another way to look at this is to compare in Canadian voter turnout can be attributed political interest among young people with to the lower rate of voting among young that of older age groups. The proportion of people today, compared to their counter- young people in most middle- and high- parts 30 or 40 years ago.8 income countries who think that politics A three-generation, longitudinal analy- is important is about half that for older age sis that separates life-cycle and generational groups, or even less. But in China, India, effects finds a sharp decline in social trust Nigeria, Vietnam, and Zimbabwe, young among American youth.9 In recent decades people are at least as interested in politics newspaper reading, watching politics on TV, as older people (table 7.1). In Indonesia knowledge of current events, voting, and and the Islamic Republic of Iran, interest the belief that voting is a civic duty have in politics is highest among the young, and all fallen among youth in almost all estab- steadily declines with age. lished democracies.10 These changes reflect a These differences in participation coin- decline in both opportunities for participa- cide with an equally distinct pattern in tion and civic interest on the part of youth, youth attitudes. In the low-income coun- although alternative forms of civic par- tries sampled, there is evidence of growing ticipation may be emerging. Indicators of confidence in many national institutions, environmental activism and participation in confidence that appears to be falling in protests, for instance, are up among young many middle- and high-income countries. people in established democracies.11 Whereas youth in middle- and high-income Is declining interest visible in develop- countries have less confidence in the civil ing countries? Apparently not, at least not service than a decade ago, the reverse is true in low-income countries. Analyses of data in low-income countries (figure 7.2). Con- from the World Values Survey suggest that, fidence in the press among youth is down for low-income countries, youth interest in or unchanged over the last decade in mid- politics might actually be rising. It has been dle- and high-income countries, but rising rising in China, India, and Nigeria, but fall- in low-income countries. Strikingly, in low- ing elsewhere (figure 7.1). Related questions and lower-middle-income countries--such in the survey--How important is politics in as China, India, Russia, and Vietnam-- Exercising citizenship 163 young people are most likely to believe that Table 7.1 Do young people care less about politics than older groups? Not everywhere business should be privately owned, in con- Age group trast to the pattern in high-income coun- 18­29 30­44 45­64 65+ tries--France, Japan, and the United States Low-income countries (table 7.2). Indonesian youth in 2000 were less likely to believe that business should be India 46.3 45.3 43.5 40.5 privately owned, but this might reflect their Indonesia 53.3 41.7 28.4 18.9 heightened political awareness during the Nigeria 54.9 50.3 47.5 53.9 financial crisis. Uganda 45.7 54.4 58.5 -- Girls are less likely than boys to partici- Vietnam 76.9 79.4 82.0 80.7 pate in political activities. Gender dispari- Zimbabwe 31.6 30.3 33.7 25.3 ties in political interest (How often do you discuss politics with friends?) are generally Middle-income countries larger in low-income countries (table 7.3). Albania 26.1 44.3 47.6 44.0 This is related to the large gap in educa- Argentina 13.0 19.4 19.9 25.8 tional and participatory opportunities for Bosnia and Herzegovina 29.3 39.7 44.6 51.7 girls and young women in low-income China 74.7 65.3 74.8 83.3 societies (including informal opportunities, Iran, Islamic Rep. of 61.9 57.4 45.5 32.1 such as spaces to play). That leaves girls less interested in public life, which in a vicious Korea, Republic of 38.5 47.2 60.3 71.4 circle leaves the interests of girls and young Poland 30.6 37.4 53.2 41.5 women underrepresented in public institu- Russian Federation 28.8 37.8 47.8 39.8 tions. Recent data from Sierra Leone show Venezuela, R. B. de 22.2 25.3 27.6 24.2 that girls are much less likely than boys to High-income countries attend community meetings, and when Canada 34.9 46.2 56.7 55.5 they do attend, they are much less likely to speak (see spotlight on Sierra Leone follow- Finland 22.1 21.5 35.1 32.8 ing chapter 7). Among urban slum dwellers France 27.2 34.7 40.5 42.6 in Rio de Janeiro, boys score significantly Iceland 36.8 53.3 53.4 62.9 higher than girls on every dimension of Japan 43.7 48.9 75.7 84.1 citizenship, including political participa- United States 58.5 59.3 71.6 83.5 tion, membership in community or civic Source : Authors' calculations from World Values Survey 2000 (Inglehart and others (2004)). organizations (excluding churches), seeking Note: -- = Not available. Table represents percentage of respondents in each age group reporting very or rather out government agencies, and having offi- interested in politics. cial legal documents.12 Figure 7.2 Young people's confidence in the civil The declining interest and confidence service is increasing in low-income countries in mainstream political institutions among Percentage of youth reporting great or a lot of confidence youth in middle- and high-income coun- in civil service tries may be due to other coincident trans- 80 formations that are less pronounced in 1990 1995 2000 low-income countries. The technologies 60 that have lowered information and coordi- nation costs might also have increased the 40 relative power of firms, civil society orga- nizations, and other nonstate actors, and 20 in the process reduced the power, prestige, and legitimacy of the state--and the incen- 0 tive to participate in traditional politics. Low- Lower- Upper- High-income income middle- middle- countries Young people might have less interest and countries income income trust in political life as aging populations countries countries push political debates toward the concerns Source: Authors' calculations from World Values Survey 1990­2000 of older citizens, and as income inequality (Inglehart and others (2004)). Note: The sample of countries is restricted to those with data from increases.13 Substitute forms of participation all three waves of the survey. 164 WORLD DEVELOPMENT REPORT 2007 Table 7.2 Should private ownership of business be increased? A more meaningful set of indicators of youth Age group citizenship, which countries could assemble with new household and official data, would 18­29 30­44 45­64 65+ include the share of youth without identity Low-income countries papers, the share of youth in presentence India 19.3 18.2 19.4 16.5 detention, measures of local political or civic Indonesia 4.9 8.0 13.3 18.0 knowledge, and the percentage of youth Uganda 39.4 40.1 41.5 -- who answer yes to the question, "Have you ever worked together with someone or some Vietnam 15.0 14.3 11.6 10.0 group to solve a problem in the community Zimbabwe 33.8 34.9 43.2 30.0 where you live?" Middle-income countries Albania 21.0 22.4 18.7 19.8 What youth citizenship Argentina 13.4 15.2 20.9 19.9 means for adult citizenship Bosnia and Herzegovina 19.8 15.3 15.2 13.6 and development China 10.5 8.1 4.8 -- Iran, Islamic Rep. of 8.6 10.3 8.0 3.7 Patterns of behavior endure: political partic- ipation in adulthood is largely determined Korea, Republic of 7.0 9.4 14.3 11.1 by participation in youth.15 Young people Poland 12.1 9.1 9.6 4.6 learn political beliefs and behavior from Russian Federation 14.9 8.4 6.0 5.6 those around them, and over time these Venezuela, R. B. de 17.0 21.0 25.0 23.0 orientations become habits, even if young High-income countries people leave their socializing group behind. Canada 14.3 18.8 17.1 25.1 Consider voting habits, which are stable over time.16 The first voting experience is Finland 5.7 8.2 11.9 13.2 challenging: young people might not know France 14.7 17.4 21.9 25.2 how to register to vote, where polling places Iceland 13.5 18.4 19.9 21.4 are located, and perhaps have not devel- Japan 4.9 6.5 8.3 17.5 oped an understanding of where candidates United States 24.8 24.3 26.4 34.2 and parties stand on issues. Moreover, their Source: Authors' calculations from World Values Survey 2000 (Inglehart and others (2004)). peers, from whom they learn, are typically Note: -- = Not available. Table represents percentage in each age group reporting that private ownership of business nonvoters. Some young citizens overcome should be increased. these obstacles and become habitual voters, that are more opportunistic and less stable but others do not. Whereas parental edu- might have increased but are not yet being cation and income, as well as peer effects, measured--such as "monitoring citizen- help young citizens overcome these voting ship," in which individuals evaluate govern- "start-up" costs, these socioeconomic and ments from afar through electronic media; demographic effects diminish over time as targeted protests; or "checkbook activism," voting (or nonvoting) tendencies become in which individuals spend money on con- habitual.17 sumer goods and nongovernmental organi- Conversely, political exclusion during zations (NGOs) that match their values.14 youth has lifelong consequences. Although Whether trends in youth participation voting rates among women in the United and attitudes in low-income countries will, States gradually approached those among as incomes rise, begin to reverse themselves men over the 20th century, the group of and resemble the declines visible in richer women who came of age before female countries remains to be seen, as do the enfranchisement in 1920 exhibited lower effects of these declines on youth engage- voting rates than their male counterparts ment in politics and social relationships in over their entire lives.18 middle- and high-income countries. Note The lasting impact of early political behav- that these trends of youth political engage- ior is visible in other areas, though it is gener- ment hardly encapsulate political citizenship, ally stronger for symbolic attachments (party let alone the social dimensions of citizenship. affiliation) and the extent of participation Exercising citizenship 165 than for other political variables, such as loca- Table 7.3 Women (ages 18­29) are less likely to discuss politics with friends tion on the left-right political scale or attitudes Overall Men Women toward specific policies. A study comparing young individuals who participated in intense Low-income and dangerous political activism with indi- India 60.3 76.0 40.4* viduals who were going to participate, but for Indonesia 82.4 88.0 77.0** some reason did not, found that those who Nigeria 74.7 82.9 66.0* participated were more involved in politics Uganda 73.0 73.6 72.4 and displayed more concern for civil rights Vietnam 75.8 82.1 69.9* issues over the course of their lives. Another study tracking lifelong participation among Zimbabwe 44.0 58.6 30.9* successive cohorts of high school students Middle-income found significant continuities in civic engage- Albania 70.5 81.9 60.9* ment and social trust over the life span.19 Argentina 46.3 49.2 43.3 Bosnia and Herzegovina 60.8 69.2 53.0* Participation in civic life promotes shared growth China 82.4 87.7 78.3 Iran, Islamic Rep. of 76.7 79.7 73.5* Active citizenship can broaden the access of previously excluded groups to oppor- Korea, Republic of 70.4 69.9 70.8 tunities for growth and higher living stan- Poland 72.0 72.6 71.4 dards, most obviously in the empowerment Russian Federation 70.3 74.0 66.4 of women. Participants in the Women's Venezuela, R. B. de 53.6 57.1 49.6 Empowerment Program in Nepal were High-income more likely than nonparticipants to initi- Canada 57.9 64.4 51.0* ate community development activities and campaigns against domestic violence, alco- Finland 62.7 64.2 60.8 hol, and gambling. They had more influ- France 52.0 62.9 42.5* ence on household expenditures, and they Iceland 66.3 60.9 72.5 better understood the importance of keep- Japan 45.0 58.5 34.6* ing their daughters in school. Legal and United States 65.1 67.1 61.9 political empowerment includes inform- Overall 64.4 69.7 59.2* ing people of their rights and providing the Source: Authors' calculations from World Values Survey 2000 (Inglehart and others (2004)). disadvantaged with opportunities to effect Note: Table represents percentage of respondents in each age group reporting that they sometimes discuss politics and exercise these rights. The Panchayat Raj with friends. * Difference between men and women significant at < 5 percent. program in India has empowered women ** Difference between men and women significant at < 10 percent. and previously marginalized groups (dalits) and led to some increase in participatory democracy.20 Citizen participation is greater in Active citizenship also facilitates collec- democracies than nondemocracies, almost tive action, which can yield more effective by definition, and some evidence indicates and better targeted public services.21 Com- that democracies, on balance, have better munity involvement is particularly effec- development outcomes than authoritar- tive in managing such local public goods ian governments. However, untangling the as water supply, sanitation, forests, roads, effects of democracy and economic growth schools, and health clinics.22 In some areas, runs into problems of identifying causal the participation of older youths in deci- pathways, imperfect measures of democ- sion making enhances service quality. Stu- racy and participation, selection bias related dent-university co-management councils in to differences in regime durability, and the Russia aim to reduce corruption in higher absence of reliable data. Some observers education (chapter 9). The municipality of conclude that democracy has a modest or Fortaleza,Brazil,improved budget processes near-zero net effect on growth, with a possi- and outcomes by including young people in bly positive effect during initial democrati- the deliberations (box 2.4). zation and a possibly negative effect later on 166 WORLD DEVELOPMENT REPORT 2007 Alluring, Fascinating as interest groups mobilize.23 More secure enrollment is lower among households in Beautiful, Charming property rights, associated with constitu- Colombian municipalities where homicide Mere descriptions all tional limits on the state and vigilance on rates are above the national median. Crime That we have never seen the part of rival institutions, seem to pro- and violence can have international spill- All around us there is but one image mote growth.24 over effects: tourism in Turkey has been sig- Tell me, is this to be my fate? Democratic participation also enhances nificantly reduced by violence in Greece.26 No sir! No sir! Accountability and development outcomes indirectly--reduc- Significant group-based social exclusion nothing but Accountability ing corruption, improving governance, is also a source of violence and conflict. . . . increasing the demand for human capital Some observers express concerns that the investment, and preparing for and prevent- size of youth cohorts in many developing How long will it take? ing disasters. Democracies--insofar as they countries predisposes countries toward war, For this dream to be realized improve governance--reduce corruption, but the evidence on this is mixed (box 7.2). whichinturnstimulatestechnologicalchange Most kinds of political violence--whether The answer to all these questions... and spurs productivity. In countries with the interstate conflict or war, civil war, riots, Accountability, only Accountability best civil liberties, public investments have or terror--have roots in grievances and Accountability, Accountability an economic rate of return between 8 and perceived injustices. As Trotsky observed, Accountability, only Accountability 22 percentage points higher than the rate in however, "The mere existence of privations Junoon, Ehtesaab27 countries with the worst civil liberties.Voting is not enough to cause an insurrection; if it rights and participation explain which coun- were, the masses would always be in revolt." tries expand access to education, itself crucial Political violence requires a motive, but for economic growth. Famously, democratic also a group identity and the subsequent countries avoid calamitous outcomes, such socialization of individuals into a fight- as famines.25 ing mode--and then an opportunity to Crime and the fear of crime and violence engage in violence. Democracy may at first are widely acknowledged to depress private increase the opportunities to stage conflict investment among both households and as the repressive powers of the state are dis- firms. Between 1984 and 1996, civil war cost mantled, and only subsequently reduce the Sri Lanka most of its tourists, and the esti- motive to fight, after democratic institutions mated equivalent of $1.6 to $2.8 billion, or channel and satisfy group-based grievances. between 13 and 23 percent of GDP. The total The result would be an inverted U-shaped cost of the war--in lost human capital, law relationship between democracy and the and order, and investments--weighs most risk of civil war, a proposition supported by heavily on the poor and the young. School recent findings.28 B O X 7 . 2 Do large youth cohorts cause violence? Maybe, if economic growth rates are low Samuel P.Huntington and Robert D.Kaplan have knowledge.32 While cohort effects have been conflict.Similarly,others argue that"multiple both argued that demographic dynamics por- found for political views and behavior as well as demographic stress factors tend to exacerbate tend conflict and violence across the develop- for civic knowledge in some contexts,they are each other's effects,expose more of a popula- ing world.29 Their argument is that members of mitigated by national sociopolitical factors in tion and more geographic areas to tensions, large cohorts,relative to those in their parents' others.The correlation between youth cohort and test developing country governments cohort, experience reduced opportunities in life: size and crime and violence is stronger in rap- with complex challenges."34 Thus, the risk of more childhood poverty,less parental attention idly growing cities,exacerbated by HIV/AIDS civil conflict for countries in the early or middle and supervision,and greater influence of peers and competition for cropland and fresh water.33 phases of their demographic transition may be relative to adults.30 Their lower economic sta- Urdal (2004) finds no evidence for Hun- heightened by an interaction of demographic tus, in turn, leads to lower fertility rates, higher tington's claim that societies with larger youth factors with each other and with nondemo- female labor force participation rates,later mar- cohorts are particularly war-prone,nor do large graphic factors. riage, and higher rates of divorce and out-of- youth cohorts appear to lead to anarchy.Urdal Very little of this research focuses on the wedlock birth when compared to the preceding does find that cohort size can increase the pro- effects of cohort sizes for developing coun- cohort.31 pensity for conflict among countries with poor tries.35 In recent decades,the importance of Researchers across diverse disciplines have economic performance.It seems that a large cohort size has diminished due to changing looked for the effects of cohort size on crime, youth cohort can aggravate the tensions caused sociopolitical and demographic dynamics, such drug use,wars,political alienation,and civic by poor growth but does not by itself lead to as shifts in gender roles and values. Exercising citizenship 167 Youth can be political actors Figure 7.3 Young people are more willing than older people to give up income to prevent while still in their youth environmental pollution Investing in youth citizenship affects pat- Percentage willing to give up income terns of participation, development pri- 100 orities, and thus development outcomes, as Ages 18­24 young people age. But youth are important 90 Ages 25 and above not merely because they are future adults: 80 they can define and achieve positive change today. The political, moral, and even stylistic 70 choices of youth help society see what is cul- turally important and achieve what is politi- 60 cally possible. Writing between the World 50 Wars, Mannheim noted that the emergence East Latin South Europe & Sub- of youth permits a society to achieve "fresh Asia & America Asia* Central Saharan Pacific* & the Asia* Africa contact"with its cultural and social possibili- Caribbean* ties, and this fresh contact"facilitates re-eval- Source: Authors'calculations based on World Values Survey uations of our inventory and teaches us both 1990­2000 (Inglehart and others (2004)). to forget that which is no longer useful and * Difference significant at < 5 percent. to covet that which has yet to be won."36 It is no accident that a new generation of lish (or deny) opportunities for young peo- young leaders has been involved in the tran- ple to participate in public life. This section sition to democracy and economic openness addresses the participatory aspect of institu- in Latin American countries, the political tions--the next, how institutions shape the reforms in Eastern and Central Europe, social identities of young people. Every insti- and the adoption of new information tech- tution does both, but for ease of exposition nologies everywhere. Less embedded in this section focuses on the opportunities to older patronage and exchange networks participate in elections, youth councils, the than adults, they are positioned to exploit military, and national or civic service. new political, social, and economic condi- tions.37 It also makes sense that young peo- The quality of participatory opportuni- ties depends on a society's constitutional ple everywhere tend to be more receptive to framework--the formal rules and infor- emerging values and worldviews, such as mal practices that suffuse its political and environmentalism (figure 7.3). Young peo- social spaces. Institutions based on liberal ple have been crucial participants in China's democratic principles teach young people anti-imperialist and democratic movement the beliefs, skills, and habits of active citi- that began in Beijing on May 4, 1919; the zenship--through the ways nondiscrimina- 1942 Quit India movement; the prodemo- tion and personal liberty are incorporated cracy movement in South Africa following in schools, legal systems, health clinics, and the Soweto uprising in 1976; and the Otpor village leadership councils. Rules for union- youth movement in the former Yugoslavia ization gave rise to labor rights activism in between 1998 and 2000 (box 7.3). Europe. Social rules associated with plan- Opportunities for political tation economies promoted forced labor arrangements and limited political activ- participation and ism in Guatemala, Guyana, and elsewhere active citizenship in Latin America. More recently, in Brazil, As young people encounter their society's mobilization to oust the military govern- main social institutions, they learn the privi- ment led to a new constitution in 1988 with leges and protections their communities a specific right to health care, which in turn provide, the tasks their communities require, promoted citizen mobilization to provide and what they can do to improve those insti- antiretroviral drugs to all AIDS patients.38 tutions. Social institutions both teach young Young people have more opportunities people how others regard them and estab- to participate in public life than ever before, 168 WORLD DEVELOPMENT REPORT 2007 ing generation became better educated and B O X 7 . 3 The Otpor youth movement in the more politically astute, and more committed former Yugoslavia to democratic ideals, it also became more aware of its exclusion from citizenship. It When older generations in positions of port for the movement,starting within their grew more cynical and less willing to partici- power are beholden to vested interests and families,schools,and communities,under pate in what it perceived to be a closed and are morally compromised,young people the radar of Serb authorities.They commu- can have the independence and moral nicated their slogans and symbols through corrupt system run by and for the elite. For courage to stand up to these institutions-- graffiti,badges,T-shirts,and other media. many, particularly for youth, the main con- whether parents or government officials or Belgrade was covered with Otpor slogans tact with the state is now the police, widely corporations.Young people's lack of expe- and its protest symbol,the black fist. considered more disrespectful and needlessly rience can be a political asset.Free from Gathering strength through their grass- prevailing norms and protocol,they can roots work and partnerships with student violent than drug dealers. As drug factions challenge failed policies and procedures. groups and trade unions,Otpor's demon- emerged and began to negotiate votes on the The Otpor ("Resistance") youth move- strations against military tribunals and a community's behalf, the few tangible benefits ment in the former Yugoslavia,which played public communications campaign spread a key role in removing Slobodan Milosevic the movement into the provinces and to the communities received from old-style cli- from power, relied on these traits.Otpor older age groups.This mounting pressure entelist politicians disappeared (box 7.4). began in four universities,where young brought early elections.When Milosevic Promoting active citizenship among people used simple protest tactics and tried to annul the election results in Sep- youth requires more than allowing more principles of nonviolence to express their tember 2000,Otpor conducted national dissatisfaction with the Milosevic regime protests that led to the installation of a new young people to vote and hold office. The and their disillusionment with political par- legitimately elected president.40 constitutional framework involves the broad ties and the political process.39 Participants opportunity structure of the society, not just had a hands-on approach to building sup- Source: La Cava and others (2006). elections. Steps to enhance that opportunity structure include establishing institutions in the sense that there is "more democracy of accountability in government, widening in more places" than at any time in human access to justice, and enhancing civil society history. About 60 percent of the world's advocacy and participation. countries are democracies (121 of 193), up from about a quarter in 1974 (41 of 150).41 Participating in elections In addition, the recent trend toward politi- The opportunity to vote is perhaps the most cal decentralization has expanded opportu- hallowed form of political participation in nities to participate in subnational politics. electoral democracies. Recognizing this, and Brazil, India, Indonesia, Italy, Mexico, Nige- aiming to enhance the involvement of young ria, Uganda, and the United Kingdom people in public life, most democracies now are among the many countries that have set the voting age at 18 (107 of 121 countries recently taken significant decentralization with available data), and as low as 15 (in the initiatives.42 New technologies have also Islamic Republic of Iran). Because of varying lowered the cost of acquiring information rules for the accessibility and voluntariness of on social and political life. voting, and the timing of national elections, However, the evidence that youth partici- it is difficult to compare voting rates across pation and interest in politics has declined as countries. It is widely recognized, however, democratic institutions have become more that young people are less likely to vote than widespread suggests that competitive elec- adults, and that youth voting rates have been tions are not enough to genuinely expand declining in many middle- and high-income the opportunities for active citizenship. In countries.43 The percentage of urban youth many countries, the democratic transitions in Chile registered to vote steadily declined are incomplete, with persistent authoritar- between 1997 and 2003 (figure 7.4). This ian enclaves, such as secret police; continu- may be related to global patterns, as well as ing clientelism; concentrations of power to potential fines and imprisonment in Chile that limit participation, political competi- for failing to vote once registered. tion, and accountability; and the emergence There are two alternative explanations of lawless zones where the state is weak. for the difference in voting rates between These themes emerge clearly in a 30-year young people and adults. First, young study of favelados--urban slum dwellers-- people may vote less frequently because in Rio de Janeiro, Brazil. As each succeed- they have less experience with politics and Exercising citizenship 169 are less socially and politically integrated than adults. Alternatively, young people B O X 7 . 4 The citizenship of Big George-- may be decreasingly interested in and more from youth to adulthood excluded from political life. The decline in overall turnout observed in many countries Jorge Paivo Pinto (not his real name),known poor favelas,the locations of choice for the as"Big George,"came to Rio de Janeiro from drug trade. between 1965 and 1999 may be due to life- a small city in the Northeast when he was 16 The drug trade attracted both money cycle effects, paradoxically exacerbated by years old.He is the fifth of 19 children,nine and weapons to the favelas,and drug deal- the decision to lower the voting age to 18. of whom died from malnutrition.His parents ers began to exert more influence,taking It may be that well-prepared voting cohorts were illiterate rural workers,and he never over the Residents'Association,community went to school.He traveled all over the coun- organizations,and even the local school. are being replaced by younger, less well-pre- try during his military service,an experience George was forced to move to a more pared ones, who first vote just when they that developed his political awareness. remote area of the favela,but he continued leave home.44 Overall, young people are While still young,he was among the to upgrade community life,fighting for a voting at lower rates than they did in previ- early invaders of a squatter settlement in health clinic,daycare,and better school ous generations.45 the industrial North Zone of Rio, and he led quality--and to be a respected leader in his the fight against eviction.By 1968 he was own right,angering the dealers even more. While developing countries such as a highly respected leader within the com- The lights on his street were shot out each Bolivia and India have lowered their voting munity, and he spearheaded the struggles night.Bullet holes punctured the water tank ages, no studies have been found that track to obtain electricity,water,sewerage,paved on his roof and riddled his house with pock- pathways,and concrete steps on the muddy marks.Youth gangs doing drugs congre- the subsequent effect on youth or adult hillsides. gated in front of his entranceway.Finally,in turnout. Because there is evidence that vot- Rio's citizens had earlier lost the right 2004,after several death threats and much ing is a habit, reforms that lower the vot- to vote for mayor,governor,and president, pleading by his family,he moved out. but they could still elect their own city The majority of the people interviewed ing age should be combined with efforts to councilmen (vereadores).George remained in Rio's slums do not feel like citizens.They incorporate young people into public life, president of the Residents'Association and do not even feel like gente (people).Experi- and perhaps to ritualize their first voting negotiated with the city council on behalf ence has taught them to be cynical.The experience. For example, a proposal to lower of the community.Democracy was restored police are not held accountable for extra- in 1984­85,and citizen action,nonprofit judicial action,including murder--and the the voting age to 16, combined with more organizations,and political parties blos- justice system and the political system are intensive "citizenship education" in schools, somed.This was also when the drug traffic "complicitous in maintaining the privilege is being discussed in the United Kingdom.46 started to appear within the favelas.It was of the privileged,"as George's son explained. Turnout is also related to illiteracy and eco- felt that the police protected the rich neigh- borhoods,but looked the other way in the Source: Perlman and Anthony (2006). nomic indicators, but evidence also sug- gests that relative educational levels, not the absolute amount of education, determine Figure 7.4 Fewer youth are registering to vote in voting behavior for individuals.47 urban Chile Randomized trials of "get-out-the-vote" Percentage registered to vote campaigns in the United States show that 80 phone canvassing increased young voter Ages 18­20 turnout by 5.0 percentage points, and Ages 21­24 60 Ages 25­29 face-to-face canvassing by 8.5 percentage points. The cost per vote ranged from $12 40 to $20.48 In developing countries, capacity constraints often translate into out-of-date 20 voter rolls and cumbersome registration processes. However, these constraints may 0 be overcome in ways that are not prohibi- 1997 2000 2003 tively expensive: a voter roll of 75 million Year of survey people in 83 electoral districts in Bangla- Source: Authors' calculations based on Chilean National Youth desh was digitized and reproduced on CD- Surveys 1997­2003. ROMs to allow voters to check their names before the elections--a first in the country's Youth councils promote citizenship history.49 Some countries have legally codi- Many countries do not permit young people fied youth involvement in governance, such to hold national political office. For instance, as the Philippines through the Youth in the Philippines requires individuals to be Nation-Building Act--an important first 40 years of age to serve as president, 35 as a step in encouraging youth to vote. senator, and 25 as a member of the house of 170 WORLD DEVELOPMENT REPORT 2007 representatives. (Some offices of local gov- Perhaps the most important drawback ernments in the Philippines are open to indi- is that these initiatives are necessarily small, viduals as young as 18). Even if young people typically targeting those already motivated to could hold national office, either through a participate. In Nepal, boys were more likely to lowering of age cutoffs or official set-asides participate than girls, and higher caste youths for youth, it is not obvious that national pri- more likely than others.53 Moldova's youth orities would move in a direction favorable to parliament (implemented with the support youth interests. Youth representatives could of an NGO) involved 450 children in four be too few in number, possess heterogeneous years, and because of concerns about who priorities, or cease to identify with other was included and the small scale of involve- youth once they obtain positions of influence ment, "was eventually phased out in favor of (chapter 9), with no change in the outcome. concentrating on the development of effective Many countries promote advisory youth models of local youth participation capable of assemblies, councils, or parliaments. Some, reaching greater numbers of young people."54 such as those in Slovenia and Zimbabwe, Moldova then moved to youth councils as bring together local youth representatives forums for youth representation and empow- at the national level. Regional structures erment--the councils were operating in 25 include the European Youth Forum, the percent of all localities in Moldova in 2005. Latin American Youth Forum, and the Afri- can Youth Parliament. Assemblies also oper- Military service has disparate impacts ate in many locales, including the Youth Wars and armies have been crucibles of Council of Catalunya and the Youth Parlia- national identity, particularly in mul- ment of Ryazan, Russia. In the municipal- tiethnic societies. The former Yugoslav ity of Barra Mansa, in Rio de Janeiro, Brazil, People's Army was one of that country's children ages 9­15 help set expenditure pri- only national institutions, and conscrip- orities. In the Philippines, the councils are tion was one of the few common national open to youths ages 15­21. experiences. When the armed forces attach These initiatives can promote civic themselves to the nation and not to a par- engagement--research shows that partici- ticular regime or social group, they can be pation in student government (and other liberal, modernizing institutions. For young extracurricular activities and volunteering) people, military service can provide oppor- is strongly correlated with other civic behav- tunities to learn skills and integrate into iors, such as volunteering and voting in national society, particularly for disadvan- adulthood.50 But there are few data on youth taged ethnic and racial minorities who have participants in representative bodies--or on few other opportunities for advancement. those whom they represent. Preconditions In the United States, active duty service has for successful youth involvement in decision large positive returns for African-Ameri- making include commitment from the top cans, while it depresses earnings for whites leadership to youth involvement, prior sup- and women. Among almost all countries port by the organization for young people sampled in a recent international survey, in decision-making roles, strong advocacy the majority of young people (ages 15­24) by adult leaders for youth participation in remain favorably disposed to military ser- decision making, and pressure by young vice, even more than to mandatory national people to increase their involvement in gov- service (table 7.4). And in most of these ernance.51 Without a strong sense of pur- countries, young women tend to favor both pose or a clear set of objectives, a sense of national and military service at higher rates disempowerment and tokenism can set in.52 than young men.55 The objective of some of these councils is to Although some have argued that required advise governments on how to improve ser- military service can promote citizenship, vices provided for youth, but there is little conscription is unevenly applied in both evidence on whether any youth parliaments developed and developing countries, favor- have enhanced the quality of youth services ing the wealthier classes. The result is that it or the governance of service delivery. is more likely to harm rather than promote Exercising citizenship 171 Table 7.4 Youth opinions on military and national service in 2005 (percentage agreeing or strongly agreeing) The government should require all young people to perform one or two years of national service Albania Bangladesh Ethiopiaa Iraq Malaysia Romania Tajikistan Ages 25+ 77 84 54 38 76 75 90 Ages 15­24 51 84 44 46 69 59 84 Male 43 86 44 44 67 55 79 Female 60 82 43 48 72 64 89 Working 53 88 43 46 70 60 83 Not working 50 82 44 46 69 58 92 High socioeconomic status 57 79 -- 21 93 54 86 Low socioeconomic status 48 86 -- 39 70 60 82 Military service is a good experience for young people Albania Bangladesh Ethiopiaa Iraq Malaysia Romania Tajikistan Ages 25+ 89 86 54 43 88 81 90 Ages 15­24 70 90 46 54 84 69 85 Male 60 91 43 48 85 61 79 Female 81 90 49 60 83 78 91 Working 71 94 42 56 85 63 91 Not working 70 89 49 52 84 70 84 High socioeconomic status 81 89 -- 60 93 61 89 Low socioeconomic status 65 92 -- 80 87 72 78 Source: WDR 2007 InterMedia surveys. -- = Not available. a. Not a representative sample--Addis Ababa and Tigray regions are not included in the survey. a democratic conception of citizenship pre- The effects are compounded if one con- mised on equal rights and obligations.56 The siders the risk of military service for health military draft is equivalent to an income tax and well-being: even in the absence of com- on conscriptees, with an associated decline bat participation, military service can result in annual earnings of as much as 15 percent in rape, physical assault, and psychological and a reduction in the incentive for families trauma.58 Data from the United States show to invest in their children's education. The that military service also has significantly tax is also regressive because many devel- adverse consequences for later (post-service) oping countries establish exemptions to health. Cohorts with higher participation in conscription, such as medical eligibility and military service subsequently suffered higher deferments for continuing education, and premature mortality, primarily due to isch- because wealthier families are better posi- emic heart disease and lung cancer.59 tioned to take advantage of these exemp- Community and national service: tions, either directly or through connections and bribes. In Russia, poor, low-educated, Involve young people in design, and rural households were much more likely and give them choices to have their sons enlisted, and the lifetime Some countries, such as Brazil, France, Ger- losses they incurred were large--about 15 many, and Israel, offer young people service percent of annual income. The costs are not programs as alternatives to voluntary or limited to conscripts themselves. Research required military duty. In other countries, conducted for this report suggests that mili- among them Ghana, Indonesia, Nigeria, tary service significantly increased the likeli- South Africa, and a number in Latin Amer- hood of subsequent criminal activity.57 ica, governments have required skilled young 172 WORLD DEVELOPMENT REPORT 2007 people, such as newly educated physicians or ticipants know each other and staff, and the university graduates, to perform national presence of accountability standards.61 service. In some cases, NGOs sponsor vol- untary service programs (Servicio País in Acquiring an identity and Chile), and in others the government con- tributes financing (Green Corps in Austra- a sense of belonging lia). The Mathare Youth Sports Association Youth is the period of acquiring an identity. in Nairobi, Kenya, and the Cambodian Vol- Erik Erikson put the challenge of identity unteers for Community Development were this way: "from among all the possible and both established by young people. Service imaginable relations, [a young person] must programs also vary in the length of service, make a series of ever-narrowing selections compensation for participants, the extent of personal, occupational, sexual, and ideo- to which different social groups mix, the logical commitments."62 Children typically activities youth engage in, the structure make any number of imaginary choices, but of supervision, and extent of youth input. for young people the choices are real and Their objectives are to provide youth with in some respects irreversible. Success in the opportunities for civic participation, instill transition to active citizenship entails, in civic virtues, build livelihood skills, and Erikson's words, feeling "at home" in one's contribute to community well-being. society, knowing "where one is going," pos- Voluntaryserviceprogramspromotecivic sessing the "inner assuredness of anticipated engagement. A rigorous longitudinal study recognition from those who count." For of the AmeriCorps National Civilian Com- youth, who are those who count? Families are munity Corps program in the United States important, of course, as are peers. As young compared civic and political outcomes for people grow older,however,those who count program participants with those for young are increasingly drawn from institutions of people who applied to the program but did society--teachers, police, employers. not participate. It found that the program A young person, viewed favorably, is increased civic engagement (though not more likely to feel invited to participate in the likelihood of voting). Systematic evalu- public and economic life--social recogni- ations of community service programs are tion promotes active citizenship. A longitu- difficult to conduct because the very charac- dinal study of youth in Estonia found that teristics of successful programs--organiza- a measure of a young person's self-esteem tional autonomy and youth initiative--are predicted the likelihood of starting a busi- confounded by selection effects that make ness years later. Similarly, having a goal of impact hard to measure. Nevertheless, many attending secondary school, controlling for service programs from around the world, family background, was related to second- such as VivaRio in Brazil, have produced ary school completion. Conversely, social passionate advocates and highly supportive invisibility, discrimination, mistreatment former participants.60 at the hands of powerful institutions, and One risk for service programs is that man- stark inequalities of opportunity can lead dates applied to young people can become young people to suffer not only in material mandates for institutions, such as schools, to terms but in their sense of who they are and produce services (in this case opportunities what they can do. Commenting on the need for civic engagement), with many of the gen- for spending money, a young Malagasy man eral problems associated with the delivery of said, "If I go out without shoes, it's as if I'm public services. This underscores the impor- not a man." A sense of belonging is impor- tance of flexibility, youth input, and organi- tant to the performance of any organization, zational autonomy in the design of service whether family, firm, or nation: "from the programs (box 7.5). In addition, elements classroom to the boardroom, . . . a sense of common to successful youth service pro- identity and attachment to an organization grams appear to include recognition afforded is critical to well-functioning enterprises."63 to youth for socially valued (as opposed to At the extreme, young people can assume token) work, a manageable size so that par- that the labels society applies to them are in Exercising citizenship 173 fact true because, for most people, it is bet- ter to have a negatively valued identity than B O X 7 . 5 Promoting voluntary and no identity. In addition, when states fail to independent opportunities establish, or actively repress, inclusive iden- tities for youth, or fail to provide for the More research is needed on the kinds of destroyed,rather than created,the kinds opportunities that best stimulate youth of state-society relationships that facilitate needs of citizens, resentment and opposi- citizenship and on government initiatives accountability and transparency.Less well tional identifications emerge. For example, promoting them.The research so far sug- known, but perhaps for that reason more the marginalized young descendants of gests that conscription,while an instrument striking, were youth-led groups, such as the North African immigrants to France speak a of nation-building in some contexts, is Edelweiss Pirates and the White Rose,who often applied inequitably,with negative despite the threat of execution,resisted the countercultural antilanguage, verlan, which consequences for poverty alleviation and National Socialists.65 The exploitation of the arose in the politicized banlieues (suburbs) solidarity.Service programs to promote energy and strength of young people for of the 1970s, because it deliberately obscures civic engagement are more effective when state power is evident in the youth wings of meaning, and carries elements of defiance, young people have choices about the political parties--associated with attacks on activities they want to participate in and are rival groups in Côte d'Ivoire, Indonesia, Rus- indifference, and heroism. On the computer involved in their design. sia,Rwanda,Zimbabwe,and elsewhere.66 of a young man arrested in Italy for alleged Too often, political leaders have used Precisely because state-sponsored youth political extremism was a song that repeated the energy and vitality of young people as participation has been (and can be) exploit- the words "I am a terrorist, I am a terrorist," instruments of state power.The Hitlerju- ative,participatory programs must promote gend (Hitler Youth) in Germany was a noto- youth involvement in their design and over and over, suggesting an aspect of glory rious instance of state-sponsored oppor- objectives,and emphasize organizational even in that label.64 tunities for youth civic engagement that autonomy for the coordinating institutions. Social organizations provide spaces for young people to develop a sense of belong- ing. They include social and civil orga- services. According to UNICEF, more than nizations, such as religious groups, rural half of all births in developing countries associations in West Africa (kafoolu), samba are unregistered.67 A study in Brazil using schools and sports teams in Brazil, clubs data from 2001 found that of children not such as 4-H and Scouts, and environmental enrolled in school, 8 percent did not enroll "When a youth gets citizenship, he protection movements. To promote a sense for lack of documentation, about the same or she needs to realize that he or she of belonging and active citizenship, public percentage who did not do so because they has now transited from a nobody to policy for social organizations should be lacked money for school-related expenses.68 a somebody." permissive, with few requirements for for- At older ages, the lack of documentation Young person, Nepal mation and reporting. impedes young people's entry into the formal January 2006 This section examines four institutions labor market (chapter 4). The high cost of that young people encounter,through which obtaining passports in many countries bears they learn how others regard them, and especially heavily on poor and credit-con- with or against which they begin to identify strained young people (chapter 8). Internal themselves: the state and the documents of migration in China, Vietnam, and elsewhere formal citizenship, schools, employers, and results in "floating populations," composed official and NGO youth groups. Young peo- significantly of young people who lack resi- ple sometimes turn, when these and other dence cards and other identity cards legally formal institutions fail to recognize them, required to obtain housing, education, to informal groups whose identities include and health care--and without which they opposition to society's formal institutions. remain excluded from society (see spotlight So the section also examines the motives on Vietnamese youth--managing prosper- and consequences of gang membership and ity, following chapter 3).69 Taking steps to political extremism. make basic legal identity more available to young people could increase their sense of A legal identity belonging to society by opening crucial ser- Actions, or omissions, on the part of the vices and social institutions to them. state affect whether, and how, a young per- son comes to feel recognized in society. The mixed impact of These include a legal identity--a passport, schools on citizenship an identification card, and other basic doc- Aspects of civics in school policies­-- umentation often crucial for access to basic whether to pledge allegiance to the flag or 174 WORLD DEVELOPMENT REPORT 2007 sing the national anthem--are political flash- the stated goal of weekly sessions; and crime points in Japan and elsewhere. The reason: and political struggles within the provin- schooling is inherently political in the sense cial and local governments. Those obsta- that, over the long term, it establishes a par- cles suggest a general problem that might ticular understanding of citizenship and the explain why civics courses, although able to nation. Both directly through school policies promote civic knowledge, have almost no and social science classes--and indirectly impact on "the development of democratic through the practices and choices of edu- attitudes and behaviors." 75Students learn cators, communities, and peers--schools as much, and probably more, about citizen- endorse certain virtues for students (chapter ship from the broader school culture than 3).Lee KuanYew of Singapore,in a discussion from civics classes, and the broader school of multiculturalism, education, and nation- culture usually replicates the patterns of building, argued that a good citizen defends exclusion and hierarchy in society. his country, protects his wife and children, In service-learning programs, students respects elders, is a good neighbor, and is work outside the school to meet real com- "clean, neat, punctual, and well mannered." munity needs. Teachers incorporate those Advocates of democratic education empha- outside activities into the curriculum, and size nondiscrimination, respect for the rights students examine what they have experi- of others, holding public officials accountable enced and receive recognition for their con- for actions, the ability to deliberate and state tributions. Service learning can promote publicly the reasons for choices, respect for social awareness, increase social connected- the rights of others, and shared solidarity.70 ness, and reduce smoking, alcohol abuse, Although schools promote national iden- and unwanted pregnancies. The programs tity over the long run, it is not clear whether appear promising, but almost all available civic education promotes citizenship in the studies have been in high-income countries, short run. Most recent studies on the impact and outcomes seem to vary with the quality of civic education rely on self-reported stu- of the program offered.76 dent behaviors and have difficulty establishing Religious schools, which typically enjoy causal relationships. Data from the Interna- a degree of curricular autonomy from the tional Association for the Evaluation of Edu- state, often address values and social ideals cational Achievement Civic Education Study, more directly than state schools. The Jesuit based on nationally representative samples Fé y Alegria schools in República Bolivariana of 14-year-olds in 28 countries, showed that de Venezuela and elsewhere in Latin America increased civic knowledge was correlated with attempt to integrate community building, self-reported engagement and citizenship- skills training, and leadership development oriented attitudes.71 In the United States, stu- into many of their programs. In many of the dents who took at least three courses in social Mujahid group of madrassas in the Indian studies were more likely to register to vote, to state of Kerala, girls outnumber boys, and the vote, and to perform volunteer work.72 schools stress women's rights and empower- Evaluations have repeatedly found, how- ment for girls.77 There is a danger, however, ever, that civic education classes have a weak that some religious schools indoctrinate stu- effect on school-age children.73 Out-of-school dents, vilify outsiders, or undermine equality civic education in Zambia changed knowl- of opportunity for boys and girls. In Jordan, edge rather than behavior, and the impact the government is promoting a new religious was mediated by educational attainment. curriculum and textbooks that would high- Civics lessons in South Africa were effective light human rights and democratic ideas only if the methods were participatory, if civ- within Islam and eliminate negative refer- ics classes met more than once a week, and if ences to adherents of other religions. students found their teachers charismatic.74 In the South African program, obsta- Recognition through work cles to participation included resistance by Employment can instill a sense of compe- school officials and teachers to allow an tence, autonomy outside the home, and NGO to conduct the civics sessions, despite social standing. It facilitates the develop- Exercising citizenship 175 ment of social capital and the means to start Youth development programs have been one's family, which itself promotes social implemented in municipalities in Colombia belonging and confers a protective effect and more widely in FYR Macedonia,but eval- on youth (chapter 6). In rural Botswana, uations are not yet available. Obstacles to the working is the crucial element of go itirela, successful application of youth development or "making oneself socially" a part of the programs in developing countries include community. There is evidence that higher skepticism of parents, a tendency to focus on rates of youth unemployment lead to more young men in urban areas, and a social reluc- burglaries, thefts, and drug offenses. Unem- tance in many places to mix with individuals ployed young people are more likely to feel of other classes and ethnic groups. alienated, express less confidence in exist- Youth action programs encourage social ing political systems, talk less about politics, activism and community involvement more and more frequently support revolutionary explicitly than youth development programs ideas than their employed peers.78 (spotlight on youth action following chapter Work is particularly important for young 9). Young people have been key participants women, for whom it is sometimes the only in political movements as diverse as street culturally acceptable experience in the public demonstrations in support of the adoption sphere. Expanding women's access to credit of the Convention of the Rights of the Child andtargetingagriculturalextensionandtech- in Brazil and school-based clubs that advo- nology to women can enhance their sense of cate evaluations of teacher performance in belonging and value (chapter 4). Working Romania. But programs encouraging activ- adolescents, especially young women who ism are fairly new and remain unproven. are engaged in intense or solitary occupa- Many rely on education, and while there is tions, such as domestic or household labor, evidence that peer education benefits the may feel isolated if they are unable to social- educators, there is less evidence that it bene- ize with their peers.79 fits the target group.82The programs tend to focus on a small core of youth, such as those Youth development and with proven leadership skills, and might not youth action programs reach those most in need. Youth development programs combine sports, mentoring, theater, life skills, leader- Youth, gangs, and crime ship training, peace building, and livelihood Identity, status, and belonging are impor- skills, usually in a defined geographic area. tant reasons for young people, usually Typically, their goals are to build self-con- young men, to join gangs. Membership can fidence, trust, and problem-solving skills. provide prestige or status among friends, However, there are few persuasive evalua- opportunities for association, excitement, tions of these programs, and most do not and money, and a sense of belonging and link program characteristics to the assets identity for marginalized young people. and developmental processes believed In some areas young people are actively important for youth development. recruited into gangs.83 Fieldwork from Short-term or intermittent contact, such urban Nicaragua illustrates the commonly as that characterizing"part-time,uniformed observed phenomenon that poor and mar- clubs," may not provide the sustained envi- ginalized young people band together to ronment to develop relationships of trust, create opportunities and identities denied perhaps because of the lack of opportuni- them by prevailing social structures.84 ties for one-to-one contact.80 A U.S. pro- Some young people join gangs to gram that focused on poor adolescents in rebel against authority. Others want to high-risk neighborhoods--and included be accepted by a group of peers. Still oth- education, community service, skills devel- ers are attracted to the group's rituals and opment, and financial incentives over four roles. Gang members may feel better after years--increased positive attitudes and joining--with more self-esteem, fewer community service, though it was prohibi- symptoms of depression or anxiety, and an tively expensive for developing countries.81 improved sense of physical health, as well. 176 WORLD DEVELOPMENT REPORT 2007 They may also feel competent, optimistic, in selves begun to commit organized crimes, control, and accepted by their peers.85 including kidnapping and extortion.89 The few long-term studies of gang Serious offenses by gang members are membership have identified "risk factors" infrequent--on average, fewer than 10 per- for belonging to a gang and committing a cent of boys are charged with violent crimes, crime. These include community character- and an even smaller number of boys (6­7 istics (weak social integration and the prev- percent) is responsible for the majority of alence of violence and availability of illegal serious violent crimes. Most offenses involve drugs), family characteristics (poverty, poor boys who commit minor crimes against parental supervision, or parental absence), property, and most boys grow out of it.90 and individual characteristics (depression, But gang membership is dangerous--the poor commitment to school, illicit drug use, likelihood of being killed is many times and peers who are gang members).86 Young higher for gang members than for the gen- women, in particular, might turn to gangs eral population.91 for protection, even where gangs are not Young gang members commit a dispro- widespread. A recent survey reveals that 88 portionate share of offenses, both violent percent of young sex workers in Nicaragua and nonviolent, and the influence of gang reported being friends with a gang member, membership on violence is greater than the and in Panama 92 percent. Among those influence of violent nongang peers. Youths who were not sex workers, only 37 percent commit more serious and violent acts while in Nicaragua reported having gang mem- they are gang members than they do after bers as friends, and 47 percent in Panama.87 they leave the gang. In some countries, the The state, or its local police and politi- number and share of crimes committed by cians, can fuel gang activity. In the 1980s, young people have increased significantly in politicians in Rio de Janeiro armed sup- recent years. In Eastern and Central Europe, porters and gangs in their garrison com- youth crime more than doubled in the first munities, and local police were caught six years after transition. Similarly, juvenile negotiating an arms deal with drug traffick- offenses increased after apartheid restric- ers in 2004. In other countries gang leaders tions were lifted in Namibia.92 pay extortion money or bribes to police. In Evidence of the efficacy of interventions-- Jamaica, Nigeria, the Philippines, and else- diversionprograms--topreventyoungpeople where, local governments, the military, and from joining gangs, to encourage them to senior politicians have directly armed and leave, or to prevent gang violence is limited. collaborated with ganglike militias.88 The few programs properly evaluated show Where the state or local authority and ambiguous effects. Criminalization and other formal institutions appear to have bro- suppression are the most common official ken down, and where the incidence of crime responses to gangs,but they are the least effec- is high, young people may band together tive. Successful interventions must address for protection, or to provide services. In the underlying marginalization, discrimina- some cases, these groups--which can tion, lack of opportunities, and hopelessness include "youth patrols" and other service- that afflict young people.93 oriented associations--can evolve into For those who have not yet joined a gang criminal gangs. The Bakassi Boys began in or even committed a crime, but are at sub- 1998 in the Nigerian city of Aba as an offi- stantial risk of doing so, several diversion cially sanctioned response to petty crime in programs have been designed and evaluated the market. A group of young people was in the United States. Some have even shown organized by local traders, with the sup- measurable benefits. However, the resources port of local politicians, to patrol the market and time they require--in costs and in and chase out criminals. This group quickly trained individuals to provide services to at- turned vigilante and took the initiative to risk youth and their families--make them summarily execute those they suspected of less than appropriate for most developing- crime. By 2002 the Bakassi Boys had them- country contexts. Exercising citizenship 177 Peace education programs promote tol- child abuse prevention; and alternative con- erance and conflict resolution skills among flict resolution.97 youth living in areas of potential conflict, whether among rival gangs or rival ethnic Young people and political violence or religious groups. One such program in There appears to be no single reason for Ecuador provided students with extracur- individuals to join organizations promot- ricular training in creative arts as well as ing political violence. Most who engage workshops on the risks of teenage preg- in political violence are young men, and nancy and drug and alcohol use. Interviews many--though not all--are students. Many with beneficiaries of this multipronged members of the Taliban (taliban means approach revealed that students who par- "students") were recruited from madrassas ticipated in the project developed an aware- in Pakistan and Afghanistan, although the ness of the sociopolitical landscape around vast majority of madrassas do not foment them and had much greater confidence in conflict or hatred, and only a small minor- expressing their opinions and a higher sense ity of madrassa students become involved of self-esteem than before.94 in political violence. Studies conflict on the Juvenile crime is positively associated propensity of young people to be involved with local unemployment and poverty, and in terrorism. Some argue that the majority decreases with family income and educa- are under 25, others that young people are tion.95 Youth crime also responds to jobs and not more likely to be involved. Yet even in changes in wages. Falling wages in the United widespread violent political movements, States in the 1970s and 1980s were partly young people play a small role. Of course, responsible for the increase in youth crime.96 the vast majority of young people are not Programs to combat crime are less effective if involved. And as with crime, their involve- they do not take into account the alternative ment in radical movements, even with ter- opportunities for schooling and work. rorism, is often temporary.98 The widespread availability of guns A study of 250 West German terrorists increases death rates from violence. Reduc- found that as many as a quarter had lost at ing the spread of small arms would have least one parent by age 14, suggesting that a significant impact on violence. Most of the disposition toward political violence the trade in guns among youths is already might be related to a kind of psychological illegal, but possible interventions include loss. Other studies argue, however, that nei- licensing, regulation, and a ban on carry- ther psychopathology nor socioeconomic ing guns. Youth homicide rates are high- deprivation drive people toward politi- est in Latin America. In 1995, there were cal violence. In some cases, participants nearly 13,000 homicides among young in political violence were more educated people ages 10 to 29 in Colombia, or 84 and wealthier than their counterparts in per 100,000. The municipal governments the general population, and in others they of Cali and Bogotá, Colombia, banned the were more likely to come from working- carrying of guns on certain days known to class backgrounds.99 Terrorist organizations have higher homicide rates (weekends and might be selecting more educated individu- holidays), and there were fewer homicides als for difficult operations or leadership. when the ban was in effect. In Bogotá that This suggests that interventions to pre- effort was part of a comprehensive set of vent political violence need to go beyond interventions to reduce youth violence providing educational and economic that included administrative reform and opportunities. Social and political inclu- municipal accountability; public awareness sion, in addition to improved economic and antiviolence mobilization; reduction opportunities, can drain crucial support for in alcohol consumption; public order and violent groups. Organizations promoting the restoration of urban spaces; local com- political violence resemble gangs and cults munity-based security councils; additional in the socialization process: social contacts funds for policing, domestic violence, and and the social environment promote a sense 178 WORLD DEVELOPMENT REPORT 2007 of belonging to the organization. This has as combatants, measures to reconcile them been observed among immigrants in Eng- to their home communities, and to assist in land, global terrorists, and the Weather their reintegration, are critical for them and, Underground in the United States. In addi- in some cases, for regional stability. tion, the calling to political violence might, like gang membership, be a way to overcome Restorative justice for young people a fragmented inner identity with some- Experimenting with social defiance is almost thing believed to be more transcendent and universal. Between 70 percent and 80 percent higher. Young people "want to believe, with of children have committed at least one--usu- every sinew of their existence."100 ally petty--offense, most often unreported and undiscovered. Young people are often Young people need legally vulnerable to arrest and detention for "status recognized second chances crimes," offenses that stem from the status of Young people develop their identities by the offender rather than the offense commit- joining an organization, trying a job, or fall- ted. Most common among these are statutes ing in with a crowd. They discover how well against vagrancy or loitering.A 1997 report by those roles fit with their self-conception and Human Rights Watch found that 1,800 young their aspirations, trying on new roles that fit people were imprisoned in Kenya for"destitu- better, and then adjusting their aspirations tion and vagrancy,"and a further 500 for being again. Sometimes the most dangerous roles "beyond parental control." Egyptian police and identities seem most real, a precarious often arrest children they deem "vulnerable moment for youth. Erikson puts it this way: to delinquency"or"vulnerable to danger,"cat- if societal authorities "diagnose and treat egories delineated in Egypt's Child Law.102 as a criminal, as a constitutional misfit, as Criminal behavior does begin in youth.103 a derelict doomed by his upbringing, or Data from South Africa show that 60 per- indeed as a deranged patient a young per- cent of repeat offenders committed their son who, for reasons of personal or social first crimes by age 19, and 82 percent by age marginality, is close to choosing a negative 25.104 But the evidence is clear that the way identity, that young person may well put his to prevent continued criminal behavior is energy into becoming exactly what the care- not to punish young criminals excessively: less and fearful community expects him to do not impose harsh penalties, do not incar- be--and make a total job of it."101 cerate youths with adults, provide access "Youth rebel because we have a Two of the most dangerous roles for youth to justice, and promote restorative justice different perspective, different tastes, are criminals and soldiers. And two practices rather than incarceration. and because we've grown up within of legally recognized second chances for different cultures." their social belonging and political inclusion Avoid harsh penalties. Many countries are criminal due process and the reintegra- impose harsh penalties for the young. Young person, Honduras tion of child soldiers. It is important not to Between 1996 and 2001, 11 individuals are January 2006 criminalize young people's experimentation. known to have been executed for crimes It is also important to avoid, whenever pos- committed when they were under age 18, sible, their incarceration--not only because eight in the United States. (In 2005 the U.S. it promotes stigma and a negative identity Supreme Court abolished the death pen- but because of the effects on the health and alty for offenders younger than age 18.) The well-being of young inmates and detainees. Democratic Republic of Congo, the Islamic Where criminal justice systems are under- Republic of Iran,and Nigeria each carried out developed, many youth wait in overcrowded one such execution; each has since expressly and dangerous prison for months,even years, renounced the practice, which violates the before seeing a lawyer or a judge. Many child Convention on the Rights of the Child and soldiers joined insurgent groups because the International Covenant on Civil and they were abducted or because their choices Political Rights.105 Other harsh penalties and abilities were severely circumscribed by often inflicted on young people include poverty and social dislocation. While they severe corporal punishment and exces- engage in (and suffer) extraordinary violence sive, long-term imprisonment, particularly Exercising citizenship 179 for victimless crimes. The rate at which Figure 7.5 Countries incarcerate young people at very different rates young people are imprisoned varies enor- Australia mously across countries (figure 7.5). Some Ghana countries have fewer youth in prison than Finland Norway expected given the size of the youth popu- Peru Japan lation, average income, and schooling. This Slovenia may be because youth crime rates are low, Nicaragua Sweden or because they have established alternative Lesotho Ireland methods to deal with young criminals. Switzerland Denmark Young people respond to increases in the Zambia severity of punishment and the likelihood Zimbabwe Bangladesh of punishment by reducing criminal behav- Lebanon Côte d'Ivoire ior. But juvenile punishment does not deter Macedonia, FYR later criminal behavior, and even in the Trinidad and Tobago Indonesia short run individuals with either the most Angola Netherlands minor or the most serious criminal histories Philippines Bolivia are not significantly deterred. Premature or Spain excessive punishment, including incarcera- France Namibia tion and social stigma, can lead young peo- Albania Greece ple to continue to participate in criminal Belgium activity or violence. Harsher prison condi- Malaysia Iceland tions are associated with higher recidivism Armenia Morocco rates. Reintegration, treatment, and restora- Thailand Iran, Islamic Rep. of tion help young people find ways to belong, Jamaica to feel both personally and socially valued. Turkey Bulgaria Obviously, policies allowing second chances United Kingdom New Zealand need to be balanced with the legitimate pur- Korea, Rep. of pose of deterring violence.106 Czech Republic Argentina Slovak Republic Singapore Do not incarcerate youths with adults. Azerbaijan Algeria Many countries have laws or regulations Mexico Portugal forbidding the incarceration of young peo- Canada ple with adults, but the laws are routinely Estonia Moldova ignored--either deliberately or because of Germany South Africa insufficient capacity in juvenile or adult Lithuania Romania prisons. Estimates suggest that more than Georgia 10,000 U.S. juveniles are housed in adult Hong Kong (China) Mozambique criminal justice settings each year. Juvenile Panama Chile offenders sentenced to adult prison are Latvia more likely than both their peers within United States Ukraine juvenile facilities and adults serving time Belarus Russian Federation alongside them to re-offend on release from ­4 ­2 0 2 4 6 prison. Peer effects for various categories of Deviation from international average theft, burglary, and felony drug and weap- Source: Authors' calculations based on data from International Center for Prison Studies (2003). ons crimes suggest limiting the exposure of Note: This figure shows international differences in incarceration rates among youth, in standard deviations from the less experienced criminals--the young--to international mean, controlling for youth population size, GDP per capita, and gross secondary enrollment. those with more "criminal capital."107 Prisons are also extremely high-risk envi- and unsafe injecting practices. Syringe-shar- ronments for the transmission of HIV and ing rates are higher in prisons than among other communicable diseases. They are over- injecting drug users outside prison.108 Juve- crowded, provide poor nutrition and limited nile inmates in adult facilities were five times health care, promote unprotected sex and more likely to be sexually assaulted, twice as unsafe tattooing,and continue illicit drug use likely to be beaten by staff, and eight times 180 WORLD DEVELOPMENT REPORT 2007 more likely to commit suicide than juveniles appear to do more harm than good.A review held in juvenile facilities.109 of nine randomized controlled trials finds that, relative to doing nothing, these pro- Provide access to justice. Many young grams are likely to increase delinquency.111 people in developing countries languish in Similarly, alternative residential camps that prisons for months, sometimes years, before impose military discipline on residents encountering a lawyer, a judge, or anyone ("boot camps") do not reduce recidivism.112 who might establish their innocence. Most Restorative (in contrast to retributive) countries do not have even basic data on justice provides opportunities for victims access to justice for youth,such as the number and offenders to meet face-to-face, talk "Ocassionally they call us rebels of young people in pretrial detention. Initia- about the crime, express concerns, and work because we dress poorly or tives that attempt to overcome limited court out a plan for restitution (box 7.6). Almost ridiculously." capacity and bring legal support to incarcer- unknown 30 years ago, it is widespread ated youth--such as the Justice on Wheels today, both for adults and young people, Young person, Honduras Program in the Philippines, where judges with thousands of programs in individual January 2006 travel to correctional facilities and process jurisdictions in more than 80 countries.113 detainees--appear promising. Also impor- These programs, and their requirements for tant are efforts, such as those recommended offenders, differ across jurisdictions--for in a South African Criminal Justice System example, they are not universally a substi- White Paper, to separate those awaiting trial tute for incarceration. Best known among from sentenced criminals, to separate orga- them is the South African "Truth and Rec- nizational responsibility for detainees from onciliation Commission" following the responsibility for inmates, and to monitor transition to democracy in the 1990s.114 the length of stay of detainees.110 Other examples include victim-offender mediation programs in Bucharest and Promote restorative justice rather than Craiovia, Romania, and a public-private incarceration. Other interventions make alliance supporting alternative justice pro- young offenders aware of the future they grams managed by the AlvarAlice Founda- face if they continue their criminal behavior. tion in Cali, Colombia. The most famous are the "Scared Straight" Recent reviews of the impact of restor- programs, which expose young people to ative justice programs on participant wel- inmates in adult prisons. However, they fare and the likelihood of recidivism show B O X 7 . 6 Last chance in Texas Candace's parents were drug dealers.By the Guided by therapists, youth immerse Texas Department of Criminal Justice costs the time she was nine,her father was dead and themselves in detailed accounts of their lives state $626,000,not including the cost of crimes she had been raped and injected with heroin. and crimes.Students repeatedly reenact their committed by those who reoffend.The average Forced to steal to keep her younger siblings crimes, playing both themselves and their vic- cost of rehabilitation at Giddings for that same from starvation,Candace ended up with a tims.Participants are thus taught to empathize person would come to about $160,000--a quar- friend of her mother's who exposed her to crack both with their victims and their inner selves. ter of the cost of incarceration. cocaine and tried to lure her into prostitution. Students spend their time in Giddings learning Candace spent 70 months in Giddings. At 13,she ran away with her 23-year-old boy- things that were not taught in their homes-- After she had struggled to complete the Capital friend; together, they robbed 120 convenience communication,introspection,and the ability to Offenders program the first time,she asked to stores in six months.Candace was arrested accept criticism without reacting angrily. repeat it.The parole committee at Giddings and sentenced to the Giddings State School in Youth who go through the Capital Offend- asked her,"How are you a different person than Texas--home to the Capital Offenders program ers program demonstrate lower recidivism when you arrived?"She replied"I came here so for rehabilitative youth justice. rates than those incarcerated elsewhere.Only locked in my feelings,there was no way I could The school seeks to make youth offenders 10 percent of students released from Giddings understand them.Everything Giddings has to confront defining events in their lives and the had been rearrested in the 36 months after their offer,I took advantage of.I earned myself some crimes they have committed.It teaches students release,compared to 74 percent of all youth distance from myself." to take responsibility for their lives.This can parolees in a recent study in California. be difficult for students whose traumatic early While these findings are impressive,the pro- experiences have shaped their senses of right gram is costly.However, the average young per- Source: Candace's life story and Giddings program and wrong (see box 2.9 on brain development). son incarcerated for a 40-year sentence in the details excerpted from Hubner (2005). Exercising citizenship 181 B O X 7 . 7 Private sector interventions to deter youth crime On February 15,2003,three members of a "No,no,red light.Bring him over here."They didn't want murder.They agreed to work for the company youth gang known as La Placita entered the to give him over.But Jimin finally bought the guy for during the mornings;in the afternoons they were 50,000 bolivares.That's something like 23 bucks.Amaz- grounds of C.A.Ron de Santa Teresa (CARST), a given basic education,instruction in values and ing, no? Twenty-three dollars, the difference between rum distillery in the town of El Consejo,Aragua life and death.116 legal issues,drug counseling,and sports activities. State,República Bolivariana de Venezuela,and At the end of the first three months,the assaulted a security guard.One of the youths Alberto met with the young man,and project recruited a second cohort from a rival was found by the company's security manager. proposed that the youths pay for the damage gang,and then a third gang came asking to join The chief executive officer of CARST, 34-year-old they had caused by working for the company the project.More than 100 young people have Alberto Vollmer,describes the event: without pay for three months.He accepted the gone through the program.State Police report a offer, and on the following Monday showed up 35 percent drop in crime in the district;but as of Our head of security,Jimin,caught one of the guys after at the company's doors together with 22 other three days,and took him to the police.But the police April 2004,only one of the project's graduates had here,it's not like the States.They look on the computer: members of the gang. found outside employment.The rest were work- "Wanted for this,this,and this.Ah."The worst prison you The company responded by starting the ing temporary jobs at Hacienda SantaTeresa. can imagine is the best alternative.Otherwise,they take Alcatraz Project.Many of the young people were you out in the jeep--which means you're dead.Jimin calls me up and says,"Listen,the police are taking this poorly educated,many had drug problems,and Sources: Brandt (2005); de Cordoba (2004); and Gon- guy out to execute him.Green light or red light?"I said, many had committed serious crimes,including zalez and Marquez (2005). fairly consistent results, both from random- Between 30,000 and 50,000 child soldiers ized controlled trials and other methods. are engaged in the conflicts in the Demo- Victims and offenders who participated in cratic Republic of Congo, 30 percent of all restorative processes were more satisfied combatants there. About 100,000 children than those who went through the courts. In have fought in Sudan's two-decade civil general, offenders in restorative justice pro- war. Some 100,000 Iranian children fought grams were more likely to complete restitu- and died in the war with Iraq. Myanmar tion agreements, and less likely to reoffend, has more than 75,000 child soldiers serving than those in control groups. Of the studies both in the state army and with its armed that matched participants to nonpartici- rivals. Some 70­80 percent of combatants pants,only one found that participants had a in Colombian guerilla and paramilitary slightly higher risk of reoffending than non- units were under the age of 25. The UN esti- participants. The evaluation and adoption mates that more than 50 states have actively of restorative justice programs are compli- recruited another 500,000 children into cated by the fact that they are almost always military and paramilitary forces.117 voluntary. Those who refuse to participate A large percentage of child combat- may not have benefited from the program ants--as many as one-third--are abducted even if they had participated. Some may or otherwise pressed into fighting. Other refuse out of fear that these extrajudicial young people join because of family pov- programs do not guarantee legal due pro- erty and social marginality: armies and cess: the accused must admit guilt to avoid militias offer youth employment, food, trial, and may not be informed of their legal shelter, and social membership--and in rights.115 Restorative justice programs need many cases the promise of booty, including not be administered by the state. They can sex, drugs, and material goods. A third of be initiated by private firms or individuals, the fighters in the civil war in Sierra Leone as a complement or alternative to official had lost at least one parent, and 60 percent justice systems (box 7.7). had been displaced from their homes before the war started; most were uneducated and Rehabilitation for young combatants poor. Political marginalization and the An estimated 300,000 children under age lack of economic opportunities encour- 18, who represent 10 percent of global com- aged the formation of factions among rural batants, either are fighting in wars or have youth. Some young people are motivated been recently demobilized. (If the age group by revenge: 15 percent of young recruits were expanded to include youths to age 24, in Colombia had a sibling who was killed the figures would be substantially larger.) before their recruitment.118 182 WORLD DEVELOPMENT REPORT 2007 The range of pain and loss for surviv- vey of child soldiers in Africa found that ing ex-combatants includes almost every 50 percent regularly had severe nightmares, dimension of social and economic well- and 25 percent suffered some form of mut- being: injury, exposure to disease, psycho- ism. At camps in Uganda, 70 to 80 percent logical trauma,sexual abuse,social isolation, of female child soldiers and 60 percent of poverty, lost education. Strikingly, there is males tested positive for one or more sexu- some evidence that these effects might be ally transmitted diseases. Drug addiction worse for youths than for children, possibly and battle injuries such as amputation are because children are more resilient, or per- common, as are the sexual abuse and rape haps because they suffer fewer stigmas, than of girl soldiers, often recruited to serve the youths (box 7.8). militia leaders.119 Programs to give these soldiers a second Linking programs to development plan- chance in life usually comprise disarma- ning is particularly important in countries ment, demobilization, and rehabilitation such as Liberia, where as many as 10 per- (DDR). To prevent re-recruitment, revenge, cent of the male working-age population and abuse, it is important during the demo- will go through such programs. Ex-combat- bilization process to house underage com- ants and potential new recruits need help in batants separately from older youth and obtaining skills, jobs, and self-employment adults. Rehabilitation should also include opportunities consistent with foreseeable medical and psychosocial support. A sur- labor market trends and the needs of their B O X 7 . 8 War-affected youth in Uganda Youth suffer more than any other age group Figure 1 Among ex-combatants in Figure 2 Postconflict earnings in from war violence,yet not all war-affected Uganda, adult literacy rates are lower Uganda decrease more for young adult youth receive equal attention and resources.In than those of adolescents combatants than for child combatants northern Uganda, for instance, services for chil- dren are more common than those for young Percentage of respondents who can read Earnings gap between combatants or write adults.However,according to a recent survey of and noncombatants ($) 750 youth combatants and noncombatants in northern Uganda,young adults are at least as 90 0 badly affected by war as children (and in some cases more so). 85 ­1 The rebel Lord's Resistance Army (LRA) has terrorized northern Uganda for two decades. 80 ­2 More than 1.5 million people have been dis- placed; tens of thousands have been attacked, 75 maimed,or killed; and almost one-third of the ­3 population has lost a family member to war violence.At least 66,000 youth are thought 70 ­4 15 20 25 30 15 20 25 30 to have been forcibly recruited into the LRA. Age Age at start of combat activity The rebels have focused on abducting males between the ages of 13 and 18, but people of all ages and both sexes have been taken.The dura- tion of abductions ranges from one day to 10 only seven years--and they are twice as likely to opportunities to return to rebuild their human years.Two-thirds of them are severely beaten,a be illiterate.Those who were abducted earn sig- capital. fifth are forced to kill,and nearly 10 percent are nificantly lower wages.Abduction,however,has There are few resources available to young forced to murder a family member or friend to few long-term psychological consequences on adults affected by the conflict.Those abducted bind them to the group. the majority of youth, and is only weakly associ- after the age of 17 were less likely to have The consequences of abduction and forced ated with symptoms of emotional distress. passed through a reintegration center (the prin- soldiering on youth are severe.Those who had Although most programs and policies are cipal intervention available in the area) before been abducted are more than three times as focused on children, young adults who were going back to the community.And fewer young likely to have a serious physical injury or illness abducted seem to have fared at least as poorly adults than children report receiving assistance that impedes their ability to work.Abductees as those under age 18. And by some measures, from NGOs.Some NGO staff complain that are twice as likely to report difficulties in family young adults are doing even worse than chil- donor funds are more forthcoming for child sol- relations.In a society where kin are crucial to dren.Young adults are less likely to be literate diers than young adults. success,such discord can be ruinous.Abductees (figure 1); they also earn lower wages (figure have nearly a year less education--a substantial 2), most likely because they were pulled out of Source: Annan, Blattman, and Horton (2006), avail- amount when median educational attainment is school at a critical stage and have had fewer able at www.sway-uganda.org. Exercising citizenship 183 local communities. Also needed are com- birth. A society that cuts off from its youth plaint mechanisms--to begin to reverse the severs its lifeline."124 political exclusion that was itself a source The task of nation-building is never com- of the conflict and to address injustices plete; it must be renewed for every gen- that occurred in the conflict. Some of these eration. Countries can promote youth rights-based approaches could emphasize citizenship not only by establishing broad the needs of youth "floating populations," liberal democratic principles, but by mak- such as those who worked in the diamond ing every institution with which youth mines in Sierra Leone, by granting com- puter-generated identity cards.120 come into contact a venue for inclusion, solidarity, and participation. The participa- If large benefits are targeted toward ex- tion of young people is important because combatants and little to the larger rural pop- it builds the capabilities of future decision ulations, young people will have an incentive makers, and because their involvement can to become new combatants. Similarly, focus- improve the quality of services that govern- ing exclusively on collecting the weapons of ments provide. ex-combatants without also taking steps Which institutions are most important to limit arms trading ignores the fact that for youth citizenship, and how can they be most postconflict environments are awash improved? For many of the policies discussed in weaponry. DDR programs thus need to in this Report, there are few evaluated inter- think not only about ex-combatants but ventions for youth citizenship. This makes about the broader young population and the large reservoir of potential new recruits.121 direct comparisons, cost-benefit estimates, and priority setting difficult.Youth programs Female ex-combatants have a distinct set suffer from the weaknesses of "youth" as an of medical, psychosocial, education, and interest group--widespread stigmatization, employment needs, yet DDR programs tend constrained voice,and short-term identifica- to tailor services for young men. tion on the part of its members (chapter 9). The few evaluations of rehabilitation and Even so, some general lessons emerge from reintegration programs show mixed suc- the evidence presented in this chapter (table cess. The Emergency Demobilization and 7.5). Lowering the voting age, particularly if Reintegration Project in Bosnia and Herze- combined with social and educational sup- govina focused on training and counseling port, might help young people develop a for employment, and included a quasi- pattern of participation at the polls that will experimental evaluation component. It persist over their lifetimes. Voluntary service increased wage employment and earnings, opportunities might be combined with life- even among youth, and the largest impact skills training, vocational training, and pub- on employment was among participants with very little education.122 In Liberia and lic works programs (chapters 3, 4, and 5) to develop not only civic attitudes, but actual Sierra Leone, the vast majority of DDR par- economic opportunities for young people, ticipants reported successful reintegration which can give them the sense that their into economic, social, and political life, but lives have a purpose and direction. Military nonparticipants in the DDR process fared service, while widely praised as developing as well as participants. Young ex-combat- the skills of disadvantaged youth, carries ants, like young noncombatants, still face long-term economic and health risks for considerable problems, primarily due to the young people that participate and is not their lack of education and skills and the absence of job opportunities.123 obviously superior to nonmilitary economic opportunities. Military conscription tends to be inequitably enforced. Many young people lack any legal iden- Perhaps Kofi Annan put it best: tity whatsoever, without which they cannot "No one is born a good citizen; no nation use basic services, and which contributes to is born a democracy. Rather, both are pro- the sense of exclusion and personal inefficacy cesses that continue to evolve over a life- that they experience. Ensuring that every time. Young people must be included from young person possesses the documenta- 184 WORLD DEVELOPMENT REPORT 2007 Table 7.5 Summary of citizenship policy directions and examples of programs Proven successful Promising but unproven Unlikely to be successful Opportunities Fostering active youth participation Lowering the voting age to 15 or 16 with social support (United Kingdom); local youth councils (Moldova) Not all opportunities are equal Military conscription (U.S. minorities) Military conscription (Russian Federation) Giving youth choices Voluntary service opportunities run by civil society (Kenya, Cambodia, and Chile) Capabilities Building safe spaces and trust Ban on firearms Youth development programs (Colombia and FYR In-school traditional civics (Colombia) Macedonia), but there are problems of scale education (South Africa and Zambia) Encouraging activism Youth action programs including issue advocacy (Romania and Brazil), but there are problems of scale Recognizing youth as individuals and Legal recognition and documentation (Brazil) leaders Service learning (United States) Second chances Providing alternatives to incarceration Restorative justice Access to justice for youth (Philippines) Harsh penalties, such (South Africa) as capital punishment (Democratic Republic of Congo, Islamic Republic of Iran, Nigeria, and the United States) and prolonged imprisonment for victimless crimes Providing resources needed for Disarmament, demobilization, and rehabilitation programs Excessive imprisonment reintegration into society (Sierra Leone and Uganda) (Russian Federation, Belarus, Ukraine, and the United States) tion to secure the rights and privileges of ship learning and socialization, as might national citizenship ought to be a prior- service-learning programs. ity for governments. An additional prior- Criminal justice reform should be a top- ity is to create secure and safe spaces for level priority for governments. Too many young people. The most successful anti- young people are incarcerated for indefinite violence and crime prevention efforts have periods, without access to legal assistance, included a comprehensive list of interven- on the basis of obscure or vague charges. For tions, including the enforcement of a ban young people, restorative justice programs on firearms in particularly volatile locales should be used in place of incarceration and events. Traditional civics classes tend wherever possible.In societies emerging from to have little effect on civic attitudes, prob- wars or conflicts, programs that demobilize, ably because those classes are swamped by disarm, and reintegrate ex-combatants, and the wider school culture. School safety and that provide opportunities to all youth, are inclusion (chapter 3) can enhance citizen- crucial for security and development. spotlight Rebuilding lives and institutions in Sierra Leone Rebuilding lives and institutions in Sierra Leone 185 After a brutal civil war, young people in Sierra Leone are trying simultaneously to build their lives and their country. Insti- tutions and infrastructure are now being rebuilt, but opportunities are still limited. The country's 1.5 million young people need a second chance to build their skills; they need opportunities to engage in productive employment; and they need the opportunity to help rebuild social institutions for better governance. Youth make up more than a quarter of the population; they are desperate to learn, to work, to start families, and to contribute to their country's growth and development. Sierra Leone has emerged from a decade- With substantial assistance from the many rural locations do not have quali- long conflict that displaced nearly half of international community, the government fied teachers, resulting in overcrowding the population, destroyed much of the is trying to establish three pathways for its and frustration. The government's target economy and its productive capacity, and 1.5 million young people: improve basic is to provide universal primary education halted any progress in the development of human capital services, enhance oppor- by 2015. Access to secondary education has human capital. The social costs have been tunities for productive employment, and also doubled since 2001, but the number incalculable. Sierra Leone is today relatively encourage civic participation to rebuild of places has not kept pace with demand.7 stable, but overcoming the legacy of the war social capital. Besides rebuilding schools, as elsewhere in and of prewar mismanagement remains an the developing world, the quality of educa- enormous challenge. The country's renewal Human development--basic tion must be improved so that graduates are must address widespread corruption, inef- health and education for youth equipped with the skills the market seeks. ficient public services, and low investment Private and public investment in human in critical economic and social areas. development is increasing, and the govern- Shared growth--youth Nowhere is this more evident than ment, together with local communities, has employment and opportunities in the indicators of human capital. The rebuilt many of the schools and clinics that Sixteen percent of the population lives in country ranks 176 of 177 in the human were destroyed during the war. To respond Freetown; the vast majority live in small development index, and 70 percent of the to the high demand for second-chance agricultural communities. Many rural population lives in poverty. Life expectancy education, programs such as Complemen- youth are migrating to urban areas, espe- at birth declined to 34 years in 2002, from tary Rapid Education for Primary Schools cially those who feel they have limited 42 years in 1990.1 Maternal mortality rates (CREPS) provides condensed education opportunities in rural areas. Underemploy- are among the highest in the world; teen- to youth forced to leave school during the ment is common among both urban and age pregnancy rates are high, as are rates conflict, benefiting about 110,000 pupils in rural youth, and formal-sector employment of sexually transmitted infection among 2004. is extremely rare: 3 percent of youth are youth. HIV prevalence is low (1.5 percent), The government introduced free pri- employed in the public sector, and 2 per- although many of the factors that facilitate mary education for all in 2001, but many cent in the nonagricultural private sector. explosive increases in HIV can be found in schools are supported by contributions While some young people would like Sierra Leone, including widespread sexual from parents and their communities.6 more training, especially in areas such as abuse, high unemployment, chronic pov- These contributions supplement official business development, many others are erty, and commercial sex work and informal resources--paying for supplies and repairs concerned about access to land and credit. exchanges of sex for goods and services.2 to buildings, and even supporting addi- Some young people who have received skills Roughly 40 percent of 12- to 24-year-olds tional teachers--but they can also deter training lack the tools required to practice and 63 percent of 25- to 35-year-olds have students of poorer families from attending their trades and have no resources to pur- never attended school. Only 20 percent of school. The vast majority of teachers report chase them. In a survey of ex-combatants 25- to 35-year-olds have finished primary that they are rarely paid on time; this may who had received some training and were school.3 All people, young and old, consis- contribute to a teacher absence rate of 22 working, only 28 percent had used their tently rank education among their highest percent and to the practice of holding pri- new skills to secure a job.8 Young people priorities, as do ex-combatants in rehabili- vate tutoring sessions outside of school. in focus groups also express the belief that tation programs.4 Progress on education is being made "connections" are needed for formal sector Sierra Leone defines youth as the period in spite of these many obstacles: primary employment. And while land is abundant, between ages 15 and 35; according to this enrollment has doubled since 2001; gross elders in many rural areas maintain strict definition, 34 percent of the population primary enrollment is more than 150 per- control over land allocation, deterring are youth. Using this Report's definition of cent. About half of sixth-graders are older youth from farming.9 youth, 26 percent are between 12 and 24, than 13, reflecting the return to schooling The lack of opportunities for young and 16 percent are between 25 and 35. Only of many who were denied education when people in Sierra Leone has already had 39 percent of these older youth are male, younger, as well as grade repetition. While devastating consequences. Are there les- reflecting the impact of violence and migra- access to primary school has increased sons from elsewhere? In many postwar tion on male youth during the conflict.5 significantly in the last couple of years, economies, such as Somalia, Angola, and 186 spotlight WORLD DEVELOPMENT REPORT 2007 Spotlight Title Mozambique, much of the population is Participation and voice are lower among youth ized young people, especially from rural likely to be engaged in "gray" or informal and women in Sierra Leone areas, who lacked education and access to economic activities as the primary means livelihood opportunities.13 The rebel forces for survival.10 These activities are essential Percentage took advantage of the "void of opportu- 100 for young people who have few alternatives nities," as one young person described it, Male Female for sustainable livelihoods or entry into the 80 to recruit soldiers for the war. Another labor market. Participation in the infor- youth, when asked why he joined the reb- mal sector is also a response to the lack of 60 els, responded, "I had never been offered credit, information, or institutions that are anything, they provided me new shoes and required for a formal sector to function. As 40 clothes."14 one review of the rehabilitation experiences The indescribable violence of the civil 20 of youth in Croatia concluded, it would war created suspicion and fear, particu- be unfortunate if the continued lack of larly of ex-combatants, making it difficult 0 opportunities left youth "with little else but Ages Ages Ages Ages for some to return home. Reintegrating dreams."11 18­24 25 and 18­24 25 and ex-combatants and other displaced people above above These examples show that it is important has been an enormous challenge, yet mil- to focus on today's youth through short- Attended a community meeting in the lions have returned home, and almost all past year term interventions, such as public works ex-combatants have been reintegrated into Spoke at a community meeting in the programs. But ultimately investments and past year society. Many communities have developed policy changes are required for broadly strong informal networks and local institu- based economic growth in the medium and Source: Miguel, Glennerster, and Whiteside (2006) and tions, partly as a reaction to the breakdown Whiteside and others (2006). longer term. Key among these investments of national structures; and there is some evi- is rehabilitating the country's physical dence that these social networks and insti- infrastructure and social institutions. Opportunities for the expression of tutions are strongest in those parts of the voice are growing: many paramount chiefs country that were hardest hit by the war.15 Rebuilding trust--youth were chosen in competitive elections fol- Experience in other postconflict coun- voice and governance lowing the end of the war, and the country tries suggests that the active participation Traditional society is strictly hierarchi- held local elections in 2004--the first in a of young people is a significant determi- cal. Village elders maintain control over generation. Participation among young nant of their satisfaction with reintegration. land, the allocation of labor, and marriage. people was very high in these elections, and Increased participation may also enhance Young people feel excluded from decision many local councilors represent a new gen- access to basic services, psychosocial well- making in many communities. They are eration of politicians. being, and social capital.16 In turn, young significantly less likely to believe that they Elders and policy makers often use the people are key to postwar economic and can change unjust policies or laws; they term "youth" with disdain and trepidation, politicalreforms,improvingtheeffectiveness attend fewer community meetings, and are to describe those--especially male--youth of rehabilitation efforts, and the transition less likely to speak during meetings. This is who are unable to provide for a family and from crisis to development. The experiences especially true among young women (see are a potential threat to peace and stability. of countries as diverse as Mozambique, Sri the figure). Community youth leaders, who There is widespread concern that without Lanka, and Vietnam show that addressing are appointed rather than chosen by youth better opportunities, the resentment may the needs of youth and managing the tran- themselves, may not represent the interests result in renewed violence. A recent social sition from crisis to development requires of young people: half are over 35, and a assessment for Sierra Leone found that the flexibility, local knowledge, and the inclusive tenth are over 50.12 conflict was fought primarily by marginal- participation of all stakeholders.17 In FYR Macedonia, university students were tired of paying bribes and having their pro- fessors accept them. They launched a public campaign to raise awareness of the levels of corruption, attract other students to their anticorruption campaign, and lobby for reform that would foster a more transpar- ent university environment. The media embraced their campaign, and reform is under way to change the higher education law. (http://www.studentitezasebe.org.mk/) Across transitions and next steps INCREASINGLY, YOUTH'S TRANSITIONS ALSO CROSS national borders. This is occurring physically, through the large representation of young people among migrants, and virtually, through the young's dispro- P A R T III portionately high and ever increasing use of new technologies. Both present worrisome risks and enormous opportunities. What next? Implementing policies that affect young people face three challenges. One is that the youth "lens" is a cross-sectoral one, but most policies that affect youth are set sectorally. Another is that young people are a weak constituency for reform because, underrep- resented in both civic forums and by their families, they lack voice. Finally, there is a paucity of proven success stories. Moving and communicating across borders 8 Young people today live in a world inte- temporary worker programs and by letting grated by faster movements across bor- the youth who do migrate use and build ders--movements of goods, capital, their human capital. Sending countries c h a p t e r information, technology, ideas, and people. can also do more to increase the develop- This chapter focuses on the two interna- ment impact of youth migration. The ben- tional movements in which youth play the efits from existing young migrants can be most major roles: international migration increased--by lowering the costs of sending and the spread of information and ideas remittances and facilitating return migra- through information and communication tion. They can also expand the opportuni- technologies (ICTs). Youth involvement in ties for other youth to migrate by avoiding these two global movements can enhance hefty passport costs and restrictive legal growth and alleviate poverty. It can also conditions on emigration--and setting up broaden their opportunities, enhance their more agreements for labor migration. And capabilities, and give them second chances they can mitigate trafficking and illegality when things go wrong in their many transi- by providing more information on the risks tions. of moving and living abroad and by imple- Young people's opportunities widen menting policies that foster more domestic when they can migrate to work abroad or opportunities for work. use today's technologies to acquire new A youth lens on ICTs suggests that gov- skills and get better jobs at home. More ernments need to pay more attention to developing-country students are studying particular types of regulations, in addition overseas and at home through online edu- to their broad regulatory and competition cation programs. New interactive technolo- policies. Communal access to new ICTs gies are providing unprecedented amounts is more important for younger individu- of information to youth, allowing them to als than older, so regulations that allow become more informed decision makers easy entry for prepaid phone card opera- and to communicate more with youth in tors, Internet cafés, and village phones can other countries. have large payoffs for youth. Policy makers One problem is that young people in should do more to use ICTs to communi- many developing countries have few legal cate and interact with youth on government options to migrate, leading to illegal migra- policy and to promote local language con- tion and trafficking. A second is that the tent. Policy makers also need to experiment rapid expansion in mobile phone and ICT with helping the first generation of youth use has yet to reach many young workers. using these new technologies to do so in The challenge for policy is to extend the a responsible and safe way, mitigating the benefits of migration and ICTs to more risks of child pornography, cyber bullying, developing country youth--and to enhance and other such dangers. their development impact while mitigating the new risks. Youth and Receiving countries can do more for poverty reduction and development by pro- international migration viding more opportunities for less-skilled In 2005, an estimated 190 million of the young migrants--through seasonal and world's people lived outside their country 188 Moving and communicating across borders 189 of birth, 49.6 percent of them women, 50.4 Figure 8.1 Young people add disproportionately to the flow of developing country migrants percent men.1 Of the world's migrants, 82 Canada South Africa percent come from developing countries, Percent of immigrants Percent of immigrants with Bangladesh, China, India, Mexico, 4 6 Russia, and Ukraine sending the largest Females numbers.2 Small island countries have the 3 largest proportions of migrants (box 8.1). 4 New analysis for this Report shows the 2 propensity to migrate increasing over the Males 2 teenage years, peaking in the early twen- 1 ties in many destination countries, such as Spain and the United States (figure 8.1). 0 0 Migrants to developing countries, such 0 20 40 60 80 0 20 40 60 80 as South Africa, are more heavily concen- Age Age trated among youth. Countries with very Spain United States skill-intensive immigration criteria, such Percent of immigrants Percent of immigrants 4 4 as Canada, receive fewer youth migrants. Twelve- to fourteen-year-olds are less likely 3 3 to migrate to developed countries than younger children, who are more likely to be 2 2 accompanying their parents. Young people make up a higher propor- 1 1 tion of the flow of international migrants than the stock (table 8.1). Thus, the aver- 0 0 age youth immigrant is much more likely 0 20 40 60 80 0 20 40 60 80 to have recently arrived in the host coun- Age Age try than older migrants. The proportion of Source: McKenzie (2006a). youth migrants varies across destination Note: Shaded portions indicate the 12­24 age range. Height of each curve represents percent of the flow of total migrants into the country who are of a given age. countries, ranging from a low of 17­20 per- cent of the flow into Canada and Russia to a high of 50 percent of Nicaraguans migrat- ing to Costa Rica and women migrating return from higher incomes.4 Young peo- to Côte d'Ivoire. Less age-specific infor- ple are likely to face lower costs in moving mation is available on refugees, but youth and have higher lifetime returns. Expected are also a large share of asylum seekers in returns can be higher because they have some countries.3 Overall, about a third of more of their human capital in education the migrants from developing countries than in job-specific skills than do older are youths, perhaps 20­25 percent of the workers--and longer working lives.The for- "[I]mmigration by youth should stock. Broadening the definition of youth gone earnings from migrating are also likely be viewed as resourceful and not as to also include 25- to 29-year-olds gives to be lower for youth, especially in countries opportunistic." them half the migrant flow and a third of with high levels of youth unemployment Young person, Kenya the stock. Based on these patterns, 32­39 and strict seniority rules that reduce wages November 2005 million youth migrants are from develop- more for the young. ing countries. While the motive for migrating for work may be greater for the young, policy condi- tions and personal circumstances determine Youths are more likely to migrate the ability to act on it. When the only legal because of individual, family, options for migration are through high- and community factors skilled immigration categories--requiring Why are people most likely to migrate when tertiary education or substantial job experi- young? The classic economic explanation is ence--migrants are less likely to be youth. that migration is an investment, requiring Only 17 percent of the flow of Chinese individuals to incur costs to generate the immigrants to the United States are ages 190 WORLD DEVELOPMENT REPORT 2007 B O X 8 . 1 Small islands, large migrations Youths growing up in small island states have Table 1 Migrants living abroad among the highest likelihood of migrating of Population Percent Population Percent youth in any country.The average island coun- (thousands) migrants (thousands) migrants try with a population under 1.5 million has 17 percent of its citizens overseas, though several Africa Pacific Islands have more than 30 percent abroad (table 1). Cape Verde 470 18.7 Fiji 835 13.5 Migration is even more dramatic from some of the smallest islands.Niue has a resident popula- Comoros 600 3.2 Kiribati 96 2.4 tion of 1,761, with 5,328 Niue-born living in New Mauritius 1,222 6.9 Marshall Islands 53 13.0 Zealand.Comparison of the 1997 and 2001 Niue censuses suggests that 28 percent of all 15- to São Tomé and 157 8.5 Micronesia, 125 12.2 24-year-olds left the country in these four years. Principe Federated States of Many of these small islands have high rates Seychelles 84 8.7 Palau 20 20.2 of youth unemployment that, together with the need to obtain tertiary education abroad, Caribbean Samoa 178 35.1 drive migration.Many of the wage jobs are in Antigua and 79 28.9 Solomon 457 0.5 the public sector, which often places a premium Barbuda Islands on seniority, limiting entry positions for young Dominica 71 32.0 Tonga 102 31.1 people. Other aspects of living in a small country Grenada 106 23.8 Vanuatu 210 1.0 can also spur youth migration.A 2005 survey St. Kitts and Nevis 47 38.5 South Asia of Tongans asked about the importance of dif- ferent reasons for choosing to apply (or not) to St. Lucia 161 17.5 Maldives 293 0.8 a special immigration quota that New Zealand St. Vincent and 109 31.1 reserves for 18- to 45-year-old Tongans each Grenadines year (table 2).Youth who applied to migrate gave better public services in New Zealand Trinidad and 1,313 18.8 as the most important reason for migrating, Tobago along with joining family networks and earning Source: McKenzie (2006b). higher wages.However, 82 percent of youth also reported that the chance for a better social life Table 2 Reasons for Tongan youth to apply, or not apply, to migrate to New Zealand in New Zealand was an important or somewhat important reason for applying. Percent saying reason Percent saying reason Whereas paying for education is an impor- is very important is somewhat important tant motive, fewer youth are interested in Main reasons given for applying to migrate migrating to overcome credit constraints associ- ated with buying a house or starting a business. Better public services such as health 71 25 Of youth who did not apply to migrate, 100 per- care in New Zealand cent say that a lack of information was the main To be with family members already in 68 21 reason, while concerns about the cost of airfares New Zealand and English language abilities are also impor- To earn higher wages in New Zealand 43 50 tant.Broadening the opportunity for youth to migrate therefore requires better quality English Better social life 43 39 language education in schools--and perhaps To earn money for school fees in Tonga 11 64 loan programs.And providing them with more information on migration opportunities can To earn money to build a better house 7 25 enhance their ability to choose. in Tonga Many of these small island states benefit To earn money to start a business in Tonga 7 7 heavily from the remittances sent by interna- tional migrants: 31 percent of GDP in Tonga, 12 Main reasons given for not applying to migrate percent in Samoa, and 11 percent in Kiribati. Did not know the requirements 100 0 But there are fears that many of these young migrants will never return, and that countries Do not want to move away from family 22 17 members will lose many of their most dynamic workers. Recent developments in information and com- Cannot afford the cost of an airfare to 22 11 munication technologies may increase options New Zealand at home through distance education and access Do not feel my English is good enough 17 39 to job opportunities across borders. Source: Pacific-Island New Zealand Migration Survey, Tonga. Sources: McKenzie (2006b) and World Bank (2005i). Note: Applicant results for 18- to 24-year-olds, nonapplicant results for 18- to 30-year-olds due to small sample size. Moving and communicating across borders 191 Table 8.1 Youth make up a large share of developing country migrants Proportion of migrants who are youth Proportion of youth migrants Migrant flow Migrant stock who are female Destination Froma Source Age range Males Females Males Females Flow Stock Argentina Developing c 12 to 24 31.7 41.9 8.4 9.6 65.3 57.3 Brazil All c 10 to 24 20.7 29.8 5.8 6.7 56.8 50.3 Canada Developing c 12 to 24 19.7 20.0 15.9 14.3 51.1 49.2 Chile All c 10 to 24 32.9 31.6 31.5 29.9 51.4 50.9 Costa Rica Nicaragua c 10 to 24 50.1 49.4 34.6 34.5 53.6 50.0 Côte d'Ivoire All c 12 to 24 34.2 50.2 17.2 27.1 48.3 46.4 Oman All c 10 to 24 -- -- 9.5 19.7 -- 42.4 Russian All l 14 to 24 18.7 16.7 -- -- 33.5 -- Federation South Africa Developing c 12 to 24 46.0 44.8 20.4 23.3 38.8 41.4 Spain Developing c 12 to 24 26.7 26.9 46.7 45.5 47.9 48.1 United Kingdom Developing c 12 to 24 30.9 34.5 14.7 14.4 55.1 50.2 United States Developing c 12 to 24 36.7 31.1 19.5 16.6 41.6 45.4 Source: McKenzie (2006a). Note: -- = Not available, c = census or survey-based measure, l = legal flow of permanent residents. a. "Developing" indicates developing country migrants only. "All" indicates all immigrants to this country. 12­24, compared with more than 40 per- social network, which lowers the costs and cent from Mexico, Honduras, Guatemala, increases the benefits of migrating. Because and El Salvador, for whom family migration youth are more likely to migrate, a young and illegal channels are more important.5 potential migrant is more likely to have a recent migrant in his or her peer network Family factors. The decision to migrate than an older individual, and so may be is often a decision of the family, not just an more likely to benefit from the migrant net- individual, particularly in developing coun- work. Over time, a culture of migration can tries, where imperfect credit and insurance develop in a community, with migration markets create a rationale for migrating to becoming a rite of passage for youth, and diversify risk and finance costly household with those not migrating considered lazy investment activities. Households can send and unenterprising.6 one of their members and count on remit- tances to help them cope with financial cri- ses and natural disasters. They will select Why is youth migration the member not just based on who has the a development issue? greatest individual gains from migrating-- This Report argues that investing in but also consider the household functions youth is crucial for a country's long-run the member performs and the likelihood of development. Migration offers a way for remitting money. In many societies parents young people to earn a higher return on exhibit greater control over daughters than these investments, and to gain more skills sons, so young women may be especially through education and work experience likely to be sent for family reasons. abroad. However, while migration usu- ally results in large increases in income Community factors. Once some young for the individual,7 governments may be people have migrated, community factors concerned that many of the externalities make it more likely that other youth will associated with having a more educated migrate too. One reason is the migrant and skilled population will be lost if youth 192 WORLD DEVELOPMENT REPORT 2007 migrate. The degree of concern depends on ers to free trade. Because many of these new whether young migrants ever return, the migrants would be young people, much age at which they return, and the skills they of this aggregate impact will be the result bring with them. of youth migration. The return of young Many migrants do return. Studies of legal migrants with education, skills, and income migrantstotheUnitedStates,WestGermany, acquired abroad is likely to have a larger and Switzerland in the 1960s and 1970s impact on development than the return of found that between half and four-fifths of older individuals, because young migrants migrants returned to their country of origin. will still have most of their working lives A more recent study of Mexican migrants ahead of them in their home countries. to the United States estimated that half the Even migrants who do not return can migrants returned within two years and continue to have sizable impacts on the almost 70 percent within 10 years.8 More- development of their home countries. over, new research for this Report shows that Remittances sent back to developing coun- migrants tend to return at relatively young tries amounted to $167 billion in 2005,10 ages (figure 8.2), giving them long periods with a large share coming from young of productive life in their home countries to migrants. One recent review concluded that use the skills and wealth acquired abroad. remittances are especially likely to be higher For example, the average youth migrant when migrants are young but married, with from Mexico returns at age 24 after spending family behind at home.11 Over longer peri- three years abroad, while the average youth ods young migrants who stay in their des- migrant from Albania returns at age 25 tination countries can continue to support after seven months abroad. Although return development at home through involvement migrants may migrate again, the likelihood in diaspora networks that facilitate trade of further migration falls after age 25--lead- and technology transfers. ing many return migrants to work for most Even so, some migrants will not return. If of their lives in their home countries. they are highly skilled, this gives rise to fears This temporary migration of youth can of brain drain in the sending countries. This have large impacts on poverty reduction and is most evident in the health sector, where development. The World Bank's Global Eco- the migration of nurses and doctors has had nomic Prospects 2006 estimated that a 3 per- large impacts in some countries. For exam- cent increase in the global stock of migrants ple, in 2002 at least 11,000 Sub-Saharan by 2025 would boost global income by $356 physicians were licensed and practic- billion and developing country incomes by ing in Canada, the United Kingdom, and 1.8 percent a year.9 This is more than the the United States.12 As a result, the public gains from removing all remaining barri- health sectors in several countries have large vacancy rates, hampering efforts to scale Figure 8.2 Migrants from developing countries tend up health interventions. The fear of brain to return home at a relatively young age drain is much lower for youth migration, because the majority of youth migrants Percentage of returnees 5 are not highly skilled. Governments may Mexico still be concerned, however, if the state 4 South Africa has financed expensive tertiary education, Philippines only to see students leave after graduation. 3 Albania Pakistan Where higher education is mostly financed 2 privately, as in the Philippines, this is not as much of a concern. 1 How does international migration 0 0 10 20 30 40 50 60 70 80 affect youth transitions? Age Youth migration also matters for devel- Source: McKenzie (2006a). opment because it provides new oppor- Note: Height of each curve represents percent of the flow of total migrants into the country who are of a given age. tunities--and new risks--for the youth Moving and communicating across borders 193 transitions. Young people may migrate to Figure 8.3 Patterns of migration and youth transitions vary according to age, gender, and work, to attend school in another coun- circumstance try, to get married or follow a spouse, or to Working Attending school become a citizen of another country. The Percentage of immigrants Percentage of immigrants prospect and process of migration may, in 100 100 turn, affect decisions about schooling in Ages 12­14 80 80 Ages 15­17 the potential migrant's home country. It Ages 18­24 may also affect the health knowledge, fer- 60 60 tility decisions, and health behavior of the migrant. It may affect the transition to work 40 40 of young people who do not migrate. It may 20 20 also affect the degree of civic engagement. The exact interactions between migra- 0 0 tion and youth transitions vary according to Males Females Males Females Males Females Males Females the age and gender of the migrant and to United States Côte d'Ivoire United States Côte d'Ivoire the circumstances in the sending and receiv- Living with a parent Married ing countries (figure 8.3). The likelihood of Percentage of immigrants Percentage of immigrants 100 80 working and being married increases with age, while younger migrants are more likely 80 to be attending school and accompanied by 60 parents. Female migrants are less likely to 60 be working and more likely to be married 40 40 than males of the same age. 20 20 Migrant youths tend to work in a small number of occupations and not use all their 0 0 Males Females Males Females Males Females Males Females skills. Migration broadens the opportu- United States Côte d'Ivoire United States Côte d'Ivoire nity to work but only in certain occupations. Source: McKenzie (2006a). The most frequent jobs for young men are Note: Work and marriage rates were not asked of 12- to 14-year-olds in the United States, and are presumed to be physically intensive, as construction labor- close to zero. ers and agricultural workers. Young migrant women are most likely to work as domes- some of these occupations than are migrant tics, cashiers, sales clerks, and waitresses and youth. In Spain, 35 percent of recent female cooks. While many of these jobs are stereo- youth migrants do domestic work, com- typically immigrant jobs in some countries, pared to less than 5 percent of native female youth are more likely to be working in many youth workers. In Argentina, 25 percent of of these jobs than older migrants. For exam- male migrant youth work in construction ple, young female migrants to the United and building, compared to less than 7 per- States are 2.7 times more likely to be a wait- cent of native male youth.13 ress as recent female migrants ages 35 to 50 Many developing country youth migrants and 3.4 times more likely to be a cashier. have less education than the average in the The few occupations that young migrants country they are moving to, but even youth tend to cluster in have low barriers to entry who migrate with high levels of education and require little previous experience and may have difficulty obtaining jobs in the education. Many of these jobs, considered fields in which they are trained. The prob- of low status in developed countries, offer ability of a migrant with a bachelor's degree little career advancement. Such jobs were finding a skilled job varies considerably traditionally filled by teenagers in devel- by country of origin, reflecting in part the oped countries, whose declining labor-force quality and language of education in the participation creates additional demand for home country.14 Thus, educated Indians immigrant youth to fill them. The native- have much higher probabilities of being born youth who do work in receiving coun- employed in skilled jobs in the United States tries are much less inclined to engage in than educated Mexicans. 194 WORLD DEVELOPMENT REPORT 2007 A further barrier is widespread occupa- educating the children remaining at home. tional licensing, which makes it illegal for Studies in El Salvador and the Philippines anyone without a license to perform the have found remittances to lower the prob- job. Such restrictions are estimated to cover ability of children dropping out of school.18 18 percent of American workers and many Although older migrant youth are more occupations in the European Union. They likely to be working than in school, tertiary cover not only professional occupations study abroad is a large and growing indus- such as medicine and law, which youth try. It often serves as one of the only legal are less likely to be trained in, but also less ways for youth from developing countries skilled occupations such as hairdressing to enter developed countries--and pro- and cosmetology.15 The difficulties young vides opportunities to turn migration for skilled migrants face in transferring quali- education into that for work. The number fications across countries can push even of non-OECD students studying in OECD highly skilled migrant youth to work in countries shot up by 59 percent between many of the same occupations as their less 2000 and 2002.19 Although only a small skilled counterparts. number of developing countries send a large Migration by some young people can number of students abroad, opportunities improve the labor market prospects of to migrate to obtain a tertiary education other youth in sending countries, a safety are very important for individuals in coun- valve in times of high unemployment. Alba- tries with limited domestic tertiary educa- nia, Algeria, El Salvador, Jamaica, Mexico, tion systems. An extreme example is Niger, the Philippines, and Turkey all have more which has more tertiary students in France than 10 percent of their labor force abroad, than in its seven domestic tertiary institu- and evidence indicates that these large out- tions. The number of students abroad from flows increase the wages and employment Albania, Cameroon, Jamaica, Kenya, and prospects of those left behind. One study Malaysia exceeds 20 percent of the number found that a 10 percent reduction through of tertiary students at home.20 emigration in the number of male Mexi- The prospect of migration may also can workers in a skill group increased the affect the incentives to become educated at average wage of remaining workers in that home. Recent literature suggests the possi- skill group by 4 percent.16 However, because bility of a "brain gain," in which migration individuals with more education were more improves the incentives to acquire educa- likely to migrate, emigration increased the tion for the pool of workers considering wages of remaining workers more for indi- migrating. Because some of these individu- viduals with higher levels of education, als will not migrate, the average human increasing wage inequality among remain- capital level in a country may, in theory, ing workers. be higher than in a situation where no one migrates. In some contexts, education deci- Migrating for education--and educating to sions do seem to be very closely geared to migrate. The chance to obtain an educa- the requirements of the global labor market. tion in another country is one of the main An example is the Philippines, with high motives for youth migration. Parents who rates of private education in fields that shift migrate with their children often consider in response to international demand.21 The schooling to be one of the most important prospect of migration is also driving part of reasons for their decision. A recent survey the demand for learning global languages in of Tongan migrants to New Zealand found many countries. that 87 percent of parents reported the Conversely, when legal channels for desire to have their children educated in migration are limited and domestic educa- New Zealand as a very important motive tion is poorly rewarded in overseas labor for migrating, even more than the 76 per- markets, potential migrants may choose less cent viewing the opportunity to earn higher education. There is evidence for this among wages as very important.17 Parental migra- Mexican migrants. A survey of students in tion can also broaden the opportunities for the state of Zacatecas revealed that students Moving and communicating across borders 195 with migrants in their families expressed migrant member knew more about meth- "Here in New Zealand there's heaps less desire for continuing their education to ods of contraception than women in non- of places where you can smoke and university.22 This appears to outweigh any migrant households,with the effect stronger drink, whereas if you were in Samoa remittance effect on the education of 16- to if the household had a female migrant. This and you drink and smoke and one 18-year-olds, so the net effect of having a greater health knowledge, coupled with the person knows you, then the whole migrant parent is to lower the education of higher household income from migration, village will know that you smoke, their children.23 improved birth outcomes. Children born to because everyone knows each other mothers in migrant households had higher really well." Young migrants are especially vulnerable birth weights and were less likely to die in Ann, 20, Samoan migrant to HIV/AIDS, but migration to developed their first year of life.26 August 2005 countries also offers possibilities for greater health knowledge. Migration has been Migrationandcivicengagement--uncertain identified as a key factor in the spread and identities, but potential exists for engage- prevalence of HIV/AIDS in Southern and ment and positive change. Migration can Western Africa. Migrants also have higher disrupt the process of becoming an active incidence rates than the general population citizen in one's home country. Young peo- in other parts of the world. It is estimated ple who plan to migrate permanently or that returned migrant workers accounted for long periods face conflicting desires to for 41 percent of all diagnosed HIV/AIDS assimilate and to maintain their culture and cases in Bangladesh, 32 percent in the Phil- national identities, while youth migrating ippines, and 25 percent in Sri Lanka in for shorter periods may find themselves 2004.24 Major reasons for the greater vul- isolated from opportunities to take part nerability of migrants are their tendency to in society in either country. Even so, there engage in risky sexual behavior and their are avenues for migrant youth to take part lower access to information and prevention in community organizations and civil soci- services. ety. Governments are also providing more Migration for many young migrants options for formal participation in home involves prolonged periods of separation country politics through absentee voting from their spouses and the watchful eyes of and dual citizenship. family. For example, more than 90 percent Many formal immigrant groups, such of African migrant mine workers in South as Latino hometown associations in the Africa live in single-sex hostels, with easy United States, tend to be led by older and access to commercial sex workers.25 Lone- more established migrants, with youth less liness, separation, and the anonymity of likely to be involved unless their parents are being a foreigner can increase the chance of around. But such societies can offer social risky sexual activities. Trafficking victims in opportunities and a sense of community the sex industry are also at high risk of HIV/ for recent arrivals. They may be particu- AIDS. But because of cultural and language larly important for young women working barriersandalackof financialresources,they as domestic workers or in other occupa- have little access to information and pre- tions that offer little contact with the out- vention services. Undocumented migrants side world. Like youth at home, immigrant are especially vulnerable because they may youth are often more inclined to participate avoid contact with any official government in less structured community activities. services for fear of deportation. A study of Haitian youth in Miami found Although the circumstances accompa- that the most common activities were help- nying migration often have the potential to ing non-English speakers or senior citizens increase health risks, migration to countries in their neighborhoods, and helping peers with good infrastructure can lead migrants through counseling or tutoring.27 How- to acquire more health knowledge. Research ever, many admitted ignorance of both the shows a strong effect of Mexican migration opportunities for civic participation and to the United States on the acquisition of how to take advantage of them. knowledge about different contraceptive While public opinion and news cover- practices. Women in households with a age often focus more on the minority of 196 WORLD DEVELOPMENT REPORT 2007 immigrant youth who join gangs, a review less likely to migrate illegally than men in of empirical evidence from the United Albania. Even in Mexico, when women do States found immigrants to be generally less migrate illegally, they do it in a safer way involved in crime than similarly situated than men, traveling with others and using groups of natives.28 This was attributed to a paid guide.31 Less systematic recent evi- the greater likelihood of immigrants to be dence comes from newspaper stories of employed than poor native workers living young African men trying to cross into in similar areas. Immigrants may also face Europe through the cities of Melilla and "Illegal immigrants can get injured high costs of being caught, such as being Ceuta in October 2005. Such illegal migra- or even die when trying to cross deported,making them less likely to commit tion brings the risk of arrest, robbery, and the American border. If they make crimes. However, some sending countries death. In 2005 at least 460 people died it, they might improve their lot a have faced a rise in gang activity because crossing the border from Mexico to the bit, but over time and due to the of the deportation of young migrants. An United States, 75 percent of them male, 35 little communication with them, example is the emergence in El Salvador of percent ages 12 to 24.32 they don't feel close to their families the Mara Salvatrucha gang, founded by Sal- Young people also constitute the major- anymore. I don't know if that's vadoran immigrants in the United States, ity of victims of human trafficking.33 worth it in the end." many of whom were deported back to El Measurement is difficult because of the Salvador after committing crimes in the clandestine nature of trafficking, but it Guadalupe, 18, Honduras United States. is estimated that between 600,000 and January 2006 Enhancing the opportunity for youth 800,000 people are trafficked across bor- migrants to participate formally as citizens ders each year.34 The majority of victims in their home countries is the increase in are female, although an increasing number absentee voting rights and dual citizen- of males are also affected. Data on victims ship in many countries. In 1998, at least assisted by the International Organiza- 43 countries allowed their citizens to vote tion for Migration indicate that 81 per- from abroad, although many migrants face cent between 2001 and 2005 were female, logistical and informational obstacles in and 71 percent ages 14 to 25. A study in Figure 8.4 Youth are more likely than exercising this right.29 A number of devel- South Eastern Europe found that the vast older people to migrate illegally oping countries have decided to allow their majority of girls are recruited by personal First-time Mexican emigrants emigrants to vote since then, including contacts or newspaper advertisements, Percentage entering destination illegally Mexico, Mozambique, and the Philippines. offering them work. Younger victims may 100 The votes of migrants can push for insti- be volunteered by parents in exchange for 90 tutional change in their home countries. a monthly allowance, without the parents Analysis of the votes cast by Czech and Pol- necessarily understanding the conditions 80 Females ish migrants in their recent national elec- in which their child will be working. Many tions found evidence that migrants' voting victims ending up in the sex industry are 70 behavior was influenced by the institu- recruited under false promises of employ- Males 60 tional environment in their host country, ment in other industries.35 10 20 30 40 50 with those living in western democracies Age more likely to favor center-right parties.30 With the demand for youth First-time Albanian emigrants This could be even more so for youth, who migration growing, how can policies Percentage entering destination illegally may be less invested in the institutions of enhance the development impact? 80 Males their home countries. A large number of youth in many devel- 60 oping countries express a strong desire to Youth migration can also be riskier migrate, especially for short durations: 91 40 The strong motives for youth to migrate percent of Albanians, 88 percent of Roma- 20 result in a demand to migrate that exceeds nians, 80 percent of Ethiopians, 78 per- Females the supply of legal opportunities. There- cent of Bangladeshis, 76 percent of Iraqis, 0 fore, youth may consciously choose to 60 percent of Tajiks, and 57 percent of 10 20 30 40 50 Age migrate illegally, seen in the higher prob- Malaysians say they would migrate if they abilities of younger migrants being undoc- had the legal opportunity, but fewer than Note: Calculated from the Mexican Encuesta Nacional de la Dinámica Demográfica 1997 and umented in data from both Mexico and 23 percent would move permanently (fig- the Albanian Living Standards Measurement Survey 2005. Albania (figure 8.4). Young women are ure 8.5). The pressure is reflected in long Moving and communicating across borders 197 wait lists for some of the legal opportuni- Figure 8.5 Leave a light on for them--most young people wish to migrate temporarily ties available. In October 2005, brothers and sisters of migrants from China, India, Move permanently Move temporarily Try it out Not move Mexico, and the Philippines all had waits Males Albania in excess of 10 years to be able to enter the Females United States through the family reuni- Males fication category.36 Given limited legal Bangladesh Females options, some of this pressure spills into Males illegal immigration. Ethiopia Females The youth bulge in many develop- Males ing countries (see spotlight on differing Iraq Females demographics following chapter 1) and Males aging in most developed countries is likely Malaysia Females to increase the demand for international Males migration over the coming years. With- Romania Females out further migration the labor forces in Europe, Russia, and high-income East Asia Males Tajikistan Females and the Pacific is projected to fall by 43 million between 2005 and 2025, while it is 0 20 40 60 80 100 Percentage of youth projected to rise by 19 million in China, 77 million in Latin America and the Caribbean, Source: WDR 2007 InterMedia surveys. In the survey, youth ages 15­24 were asked "If it were possible for you legally to move to another country to work, would you?" 82 million in the Middle East, North Africa, and Turkey, 93 million in low- and middle- income East Asian and Pacific countries, of cohort size is lower when policies restrict 211 million in Sub-Saharan Africa, and 292 migration, but a study of migration into the million in South and Central Asia.37 United States over 1971­98 reveals that the This will increase demand for interna- share of the sending-country population tional youth migration for three main rea- in the 15­29 year age group is a significant sons. First, because youth have the highest predictor of migration, raising emigration propensity to migrate, the rising num- rates from Latin America by 11 percent over ber of youth in developing countries will those from Western Europe.38 Based on increase the number of potential migrants. these historical estimates and demographic Second, a larger youth cohort can increase projections, it is predicted that the annual unemployment and lower wages in devel- flow of emigrants out of Sub-Saharan Africa oping countries, while the smaller youth will rise by between 1.5 million and 2.1 mil- cohort in developed countries can push lion by 2025.39 up wages (chapter 4). So the economic gains from migration will rise, increasing Receiving countries can increase the benefits the likelihood of migrating for each youth. by providing more opportunities for youth to Third, as more youth migrate, the size of migrate, perhaps through temporary worker the migrant network will increase, further programs--and by letting migrants develop increasing the motive for other youth to and use their human capital. Many devel- migrate. oped countries have targeted their immi- History provides support for the pre- gration systems at highly skilled workers, dictions that a larger youth cohort results providingdevelopingcountryyouthwithfew in more outmigration. Migration patterns options for legal migration. With aging soci- from Western Europe to the New World eties and rising incomes fueling demand for between 1820 and 1913 show a large and the services that young, less-skilled migrants positive effect of youth cohort size on emi- typically perform,developed countries stand gration, with almost half the additional to gain from more youth migration. How- births ultimately emigrating. Similar size ever, receiving countries typically worry that effects have been calculated for intra-Africa an influx of migrants reduces employment migration in the late 20th century.The effect or lowers wages for native workers. 198 WORLD DEVELOPMENT REPORT 2007 A very large literature tries to measure the ing permanent migrants to integrate. This effect of immigration on the wages of natives. provides a further reason for bringing global A recent meta-analysis of 344 estimates con- languages into the curriculum in developing cluded the average effect on the wages of countries with high outmigration (chapter 3). natives is significant but small, with a 1 per- Temporary and seasonal work programs offer cent increase in immigrants lowering wages one avenue for receiving countries to reap of native workers by 0.11 percent.40 While many of the benefits of more youth migration other studies have found larger effects, the while lessening the perceived costs of assimi- most recent research from the United States lating migrants. Many youth in developing and the United Kingdom also finds little countries express a desire to migrate for a "For the third time in one year, I evidence that immigration has had adverse short period (see figure 8.5), allowing them to have been denied [a] visa by the U.S. effects on native workers in these countries.41 save money to buy a house,open a business,or embassy in Nigeria. I was meant Cross-country evidence in Europe has found achieve other goals in their home countries. to attend the inaugural meeting more negative effects in countries with more Several temporary worker programs are of the UNFPA [United Nations rigid labor markets, which restrict hiring youth-specific. The Seasonal Agricultural Population Fund] youth advisory and firing, with the effect greater for young Workers Scheme in the United Kingdom panel as a member of the panel. My men. As a result, youth immigration into allows full-time students ages 18 and over visa application was denied...it is Denmark, Switzerland, and the United King- who live outside the European Economic sad to know that even if you are to dom, which have fairly flexible markets, has Area to work for six months in seasonal attend a UN meeting either as a less effect on native workers than immigra- agricultural work. Participants can apply speaker or participant with all your tion into France, Germany, Italy, and Spain, to take part again after three months out arrangement[s] made by the UN, which have higher business entry costs and of the country. The Working Holidaymaker you might still be denied visas." more labor market restrictions.42 Scheme in the United Kingdom allows Young person, Nigeria The early experience following the acces- 17- to 30-year-olds from Commonwealth June 2006 sion of eight Eastern European countries to countries to work for up to 12 months of the European Union (EU) shows broadly a two-year stay.45 Australia, Canada, and positive impacts of increased migration for New Zealand also employ working holi- the receiving countries. Twelve of the EU15 day schemes open to 18- to 30-year-olds, membercountrieselectedtomaintainrestric- but these cover only a few middle-income tions on migration from these new member countries. Such schemes provide some states during a transitional period, while opportunities for youth migration, but Ireland, Sweden, and the United Kingdom requiring participants to be full-time stu- allowed immediate free movement of work- dents or have enough funds to support ers. The majority of workers moving under themselves for their first few months of this scheme were young: 83 percent of new "holiday" limits poor, less-skilled youth registered workers from the accession coun- from participating. tries in the United Kingdom were 18­34, and The main concern of developed countries 44 percent were 18­24.43 Initial assessments is whether temporary workers will return have concluded that the main impact of these home at the end of the work period. Expe- flows was to increase output and employ- rience is mixed, and further policy experi- ment, relieving labor market shortages and mentation is needed. One of the key factors having little impact on native workers.44 appears to be whether workers can return A second concern of receiving countries with a reasonable expectation of being able is the potential difficulties associated with to work again in a subsequent year. Cana- assimilation of migrants from different cul- da's seasonal agricultural workers' program tural backgrounds. This is an area of active is a possible model, with employers free to policy debate in many developed countries, request the same workers again the next and is largely beyond the scope of the devel- year.46 Of the 15,123 workers who entered in oping country focus of this Report, except to 2004,only 1.5 percent went absent before the note that youth migrants generally are able to end of the contract, and almost all are esti- assimilate more easily, and that acquisition mated to have returned home. In contrast, of the host country language is widely found the previous version of the United King- to be of fundamental importance in allow- dom's seasonal agricultural worker program Moving and communicating across borders 199 did not allow the opportunity to return, and financial institutions and obtain credit for had estimated overstays of 5­10 percent. business start-ups, so programs that enhance access to credit can be beneficial. Developingcountriescanmaximizethedevel- Several countries have tried to attract opment impact of youth migration through back their most highly skilled migrants, with policies that increase the benefits from exist- mixed success.51 Such programs are rarely ing migrants. . . There is considerable scope directedatyoungmigrants,whoarelesslikely for policy interventions in sending countries to be highly skilled. One broader issue that "I am a girl. I can't go working in to increase the development benefits from does affect youth is the difficulty migrants other cities. My brothers could never their existing stocks of youth migrants. One often find in having the qualifications they tolerate it." area of recent policy focus is the high cost gained abroad recognized back home. This Female, 20, Morocco53 of sending remittances, which reduces the also affects youth indirectly through their amount received by relatives back home and tertiary education systems. For example, acts as a disincentive to remit. Because young Romanian academics often find it hard to get migrantsaremorelikelytobemigratingwith- masters and doctoral degrees earned in Ger- out the papers to establish bank accounts and many and the United States recognized, and have less experience with the financial system, so are reluctant to return.52 high remittance costs may be even more of a barrier. Only 31 percent of 18- to 24-year- old recent migrants in the United States have bank accounts, compared with 65 percent of B O X 8 . 2 Poor job prospects fuel migration in Morocco 25- to 50-year-old recent migrants.47 Policies to lower the remittance costs In the nine communities visited in Morocco who identify migration as one of the main for the Moving out of Poverty study, youth factors helping the best-off households, include providing information to migrants express great frustration at the difficulty of and a way out of poverty."The only period on the costs of using different methods obtaining good jobs locally. "Graduates are of moving up in our life was between 2000 (as with Mexico's consulates in the United a thousand times more numerous than the and 2005, and the reason for this was my States), delivering financial education to existing posts,"exclaims a young man from son's emigration to Spain,"says one woman Foum Zaouia, while in the opinion of one in Foum Zaouia, who now plans to send her migrants before leaving (as in the Philip- young woman from Bir Anzarane, the few other son abroad.If youth feel pushed by pines), and developing the financial infra- good jobs to be had are"via corruption, via their families to migrate, they do not speak structure for receiving remittances through mediation, via favoritism, and via closeness." of it in those terms, although a joke from competition policy and the lack of onerous Training or job placement services seem Tamessa-Tissyan-Azendo may be revealing to be far out of reach for most youth, even on this subject: "A mother sends her son regulations.48 Such policies offer the poten- those with college degrees.As for setting to work in a faraway city.When he comes tial for considerable gains in remittances: a up their own businesses, young men and back after three years, the mother, instead study of Tongan migrants in New Zealand women express interest but also discour- of welcoming and hugging him, says to him: agement due to weak markets, high taxes, `Oh! Why do you come back so early, did you estimated that reducing the cost of sending complex procedures, the inability to access forget something?'" money to that prevailing in the more com- credit, and the lack of their own funds. Youth reveal more mixed emotions about petitive world markets would increase their Faced with these conditions, youth migration than adults.Several express uneasi- remittances by 28 percent.49 widely identify migration overseas as the ness about having to leave their villages,and best, if not the only, means to get ahead. the resulting separation of families.Those Sendingcountriescanalsobenefitthrough Young men say their goal is"to save with insufficient funds to migrate interna- policies that facilitate the successful return of regularly to be able to buy a work contract tionally may migrate internally and speak of young migrants. One part of this is sound abroad at 60,000 DH [Moroccan dirhams] difficult working conditions in the cities of macroeconomic policy, which can generate (more than $6,550) or to migrate illegally, Morocco."We suffer and the conditions of which costs between 20,000 and 30,000 work are very hard.Most of us work only in expectations of improving conditions. Not DH per person."Over 70 percent of Moroc- construction yards ...whatever the job we do, surprisingly, youth are more likely to migrate can 18- to 24-year-old migrants in Spain the income is minimal."Even so,others wel- when facing poor domestic labor markets are men, and over half of the women who come the potential new surroundings,and (box 8.2). Many temporary migrants decide migrate are married to another migrant.54 the escape from family tensions over being Young women face more restrictions, out of work and clashes between modern to return to enjoy the higher purchasing saying parents are reluctant to let them and traditional ways. "What all of us really power at home. Evidence also suggests that work outside the home, let alone in a big wish is to go to Europe.Here we can't even the return decisions of some migrants are city or another country. "The girl is always communicate with our parents.Also,there is driven by the desire to achieve a certain target supervised,"explains a young women from nothing for us to do,"says a young man from Igourramene-Tizi. the village of Bir Anzarane. level of savings abroad, which can be used to Migration is viewed extremely favor- start businesses on returning home.50 Young ably by adults in these study communities, Source: Narayan and Petesch (2006). people find it particularly hard to access 200 WORLD DEVELOPMENT REPORT 2007 . . . through policies that expand the opportu- of all female migrants, and 15 percent of nities for other youth to migrate. . . Despite all male migrants.56 The Philippine gov- the benefits from migration, country poli- ernment licenses recruitment agencies and cies can inhibit the opportunities for young markets its workers worldwide, signing 56 people to emigrate.55 One barrier is the bilateral treaties with receiving countries. cost and time of obtaining a passport, the A network of attachés and welfare officers most fundamental document of legal iden- operates worldwide, acting as resources tity for migrants (chapter 7). Data on pass- and advocates for the overseas workers. port costs in 127 countries show that one In addition to broadening opportunities in every 10 charges more than 10 percent for migration, the government provides of average annual per capita income for a pre-employment orientations. Potential passport. Reducing passport costs by 1 per- migrants are told about cultural differences cent of per capita income is associated with in the country they are considering--and a 0.75 percent increase in emigrants per given information on illegal recruitment, capita. Because young people are likely to methods for sending money, and phone have less accumulated savings, high pass- numbers to use in case of grievances. port costs may be more of a barrier than for older migrants. . . . and through policies that mitigate Several countries inhibit emigration the risks associated with international through legal barriers on the right of women migration. The risks of trafficking and to emigrate, which in some cases apply only illegal migration can be mitigated, first, by to young women (table 8.2). These coun- broadening opportunities for other forms tries have 5­6 percent fewer migrants per of work, both domestically and abroad. Pro- capita than countries with similar levels of moting the entry of disadvantaged youth income, population, and governance that into the labor force in developing coun- do not employ such restrictions. Faced with tries can lessen the desire to leave (chapter these restrictions, young women who wish 4), but large income differentials, missing to migrate must do so through alternative credit markets, and a desire to experience channels, increasing the risk of trafficking. life in other countries will still provide In addition to removing restrictive poli- strong incentives to leave. Providing legal cies, countries can take more active mea- forms of temporary migration thus acts as sures to broaden the range of migration an alternative to illegality, the only form of opportunities available to youth. The best migration available to many youth. Second, established example is the Philippines, information campaigns and legal steps can which in 2005 sent 1 million of its citizens promote the agency of youth, helping them overseas as contract workers. On average, become less vulnerable to the false promises youth ages 18­24 constituted 31 percent of traffickers and giving them legal recourse when being sold by family members. Third, victims of trafficking can be given a second Table 8.2 Countries that restrict the right of women to emigrate chance for life back home--by working with developing country governments to ensure Married women require their Restrictions on both married husband's permission but no and unmarried women that victims are not treated as criminals and restriction on unmarried women can receive help in returning. Afghanistan Congo, Dem. Rep. of Iran, Islamic Rep. of Sending countries should lessen the Gabon Jordan risks of HIV/AIDS affecting their migrants, Uganda (when travelling with children) Kuwait (unmarried women under 21) especially given the potential spread of the Libya disease to the nonmigrant population. The Unmarried women require their Qatar (women under 30) father's permission but no two main policies are to promote informa- restriction on married women Saudi Arabia tion and prevention activities in the sending Sudan Egypt, Arab Rep. of (women under 21) areas and to work with receiving countries Swaziland United Arab Emirates to create an environment for migrants less Yemen, Republic of conducive to the spread of HIV/AIDS. For Source: McKenzie (2005). example, predeparture orientation semi- Moving and communicating across borders 201 nars in Bangladesh, Indonesia, the Philip- messaging, also known as short messag- pines, and Vietnam all inform migrants ing service (SMS). In a typical age pattern, D E F I N I T I O N 8 . 1 about HIV/AIDS, while prevention activi- youth were the first adopters of the Inter- ICTs ties in Thailand are held in some of the net in the Kyrgyz Republic and account for main sending villages, allowing spouses and most of the growth in users between 2001 Information and communica- tion technologies (ICTs) consist other family members to obtain informa- and 2005 (figure 8.6). Data from surveys of hardware, software, networks, tion as well. in 2005 show that youth accounted for 43 and media for the collection, The higher risk associated with the percent of all Internet users ages 15 and storage, processing, transmission, separation of migrants and their partners older in China, 50 percent in Armenia, 53 and presentation of information (voice, data, text, images), as well can be reduced by working with receiving percent in Bolivia, 60 percent in Egypt, 61 as related services.Communica- countries to enable the spouse or partner to percent in the Kyrgyz Republic, and 70 per- tion technologies consist of a accompany the migrant. While single-sex cent in Indonesia. These proportions, simi- range of communication media hostel environments for migrants are the lar to those for 2002 and 2003, suggest that and devices, including print, telephone, fax, radio, television, norm in mining and construction in some approximately 130­160 million of the 269 video, audio, computer, and the receiving countries, studies have shown the million new Internet users between 2000 Internet. potential of family housing to dramatically and 2003 were ages 15 to 24. Source: Neto and others (2005). reduce the incidence of HIV.57 Cameroon Although youth are more likely than provides an example, with villages built to older age groups to use the new ICTs, the support immigrants working and living use among youth varies dramatically. with their families at an oil pipeline con- Across countries surveyed in 2005, the share struction site. of 15- to 24-year-olds who have ever used the Internet varies from less than 1 per- Youth and the global flow cent in Ethiopia to 12 percent in Indonesia, 13 percent in Ghana, 15 percent in Egypt, of information and ideas 29 percent in Armenia, and 53 percent in The first few years of the new millennium China. The digital divide also occurs within saw extremely rapid increases in Internet, mobile phone, and computer use in devel- oping countries. Between 2000 and 2003, Table 8.3 Catching up fast: The rise of new technologies the developing world gained more than Low Middle High one-quarter of a billion Internet users and EAP ECA LAC MENA SA SSA income income income almost half a billion mobile phones. These Usage rate per 1,000 people new technologies are growing much faster than older information and communica- "Old" ICT tion technologies (ICTs) such as television, Daily newspapers 60 -- 61 -- 59 12 44 55 -- radio, mainline telephones, and newspapers Radios 287 447 410 273 112 198 137 344 425 (definition 8.1 and table 8.3).Mobile phones Telephone mainlines 161 228 170 133 39 11 32 177 393 have overtaken mainline phones in cover- Television sets 314 408 290 205 81 63 78 319 362 age in many parts of the world, and there "New" ICT are more Internet users per 1,000 people than there are daily newspapers purchased Internet users 68 161 106 46 10 20 16 117 279 in every region except South Asia. Even so, Mobile phones 195 301 246 85 23 51 23 224 785 Internet use remains low in poorer develop- Personal computers 26 73 67 31 7 12 7 42 284 ing countries, and radios and televisions are Annual per capita growth since 2000 (%) much more prevalent. Internet users 41 59 38 39 20 32 63 46 13 Rapid growth in ICT use Mobile phones 51 48 27 52 87 42 83 43 17 among young people Personal computers 28 18 17 9 27 11 24 20 12 Although young and old alike watch televi- Telephone mainlines 21 1 5 15 12 3 14 12 0 sion and listen to the radio, young people Television sets 10 -- -- 5 5 10 4 5 0 are the main users of the new ICTs, espe- Source: World Bank (2006h). cially the Internet and more advanced Note: Data are generally for 2002­03, except for newspapers (2000) and radio (1997). High-income countries are non-OECD high income. -- = Not available. EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin features of mobile phones such as text America and the Caribbean; MENA = the Middle East and North Africa; SA = South Asia; SSA = Sub-Saharan Africa. 202 WORLD DEVELOPMENT REPORT 2007 "The digital divide must be defined countries (table 8.4). Computer and mobile Internet cafés (figure 8.7). Access at school as a wider concept than the access to phone ownership and Internet and SMS varies considerably across countries. Some the Internet. Even if you have it, if usage are highest among youth in urban richer, developing countries have con- you do not know how to navigate it areas and with more education and higher nected many schools, with Chile having 75 you will find nothing in it." household incomes. In Indonesia, 59 per- percent of schools online. In contrast, data cent of university students had used the from six Sub-Saharan African countries Young person, Argentina Internet and 95 percent SMS, compared reveal that fewer than 1 percent of schools January 2006 with 5 percent or less among youth with are covered.58 Mobile phone use can also be only primary education. communal, especially in rural areas. Wide- The use of these new ICTs is a more spread access to phone resellers in many communal experience in developing coun- countries has reduced the barrier to access tries than in developed. Many youth do not for young people. have computers in their own homes, and In some countries, young women access instead access the Internet at school or at the Internet less through these public access points than do young men (figure 8.7). In Figure 8.6 In the Kyrgyz Republic, young people use the Internet Ghana, 16.5 percent of male youth use Inter- more than older people and account for much of its growth net cafés, more than twice the 6.6 percent for female youth. Women may not feel com- Percentage who used the Internet in the past four weeks 25 fortable or may be restricted from attend- 2005 ing these public points alone or after certain 20 hours. Even at school, girls may find it harder to gain access. In Sub-Saharan Africa, enroll- 15 ment rates of boys greatly exceed those of 10 girls, so girls compete with a large number 2001 of boys for scarce computer resources.59 5 In contrast, young women do not appear 0 to have less access to mobile phones than 15­19 20­24 25­29 30­34 35­39 40­44 45­49 50­54 55­59 60­64 65+ young men, and may actually use them Age group more in some countries. Source: InterMedia national audience surveys. Young people are more likely to adopt these new technologies for economic, phys- Table 8.4 The digital divide among Indonesian youth iological, and social reasons. As with migra- tion, longer working lives mean that young Youth with Youth with Internet use computer in mobile phone SMS use people have more time to gather the benefits (percent) home (percent) (percent) (percent) from investing in new technology. The cost All 15- to 24-year-olds 12 5 26 24 of investing in the skills required to learn how to use the new ICTs is also likely to be Among 15- to 24-year-olds less for youth, who are better educated than Males 16 5 22 22 older generations and may receive train- Females 8 5 31 27 ing through school. Moreover, youth find it Urban youth 16 7 28 27 easier to acquire complex information-pro- Rural youth 6 2 21 18 cessing tasks. The tendency of youth to use Primary education or less 3 1 9 5 these technologies is amplified by the desire to use these technologies for entertainment, Secondary education 7 4 19 19 and reinforced through peer learning and Senior education 20 8 37 35 network effects: the value of a mobile phone College or university education 59 28 96 95 or Internet connection increases when more Monthly household income of one's peers are using it. Above Rp 1,250,000 29 19 57 55 As a result of this rapid expansion in Rp 600,000 to 1,250,000 10 3 33 22 ICTs, young people around the world are Less than Rp 600,000 5 0 7 7 more able to access information and con- nect to ideas and people outside their coun- Source: InterMedia national audience surveys. Note: Rp = Indonesian rupiah; 1,250,000 Rp is about $128; 600,000 Rp is about $61. tries. In 2005 it was estimated that there Moving and communicating across borders 203 were close to 1 billion Internet users world- Figure 8.7 Public Internet access points are important for young people wide.60 A social experiment involving users in 166 countries measured the number Albania of steps required to connect to designated targets and found that the popular notion China of "six degrees of separation" between any two people in the Internet world is not too Internet café far wrong: the median number of steps Ghana Home required to connect users in different coun- Work tries was seven.61 Surveys for this Report School Tajikistan show youth to be more likely than 25- to 50-year-olds to communicate with people Venezuela, in other countries (figure 8.8). A remark- R.B. de able 44 percent of Romanian youth and 74 50 40 30 20 10 0 10 20 30 40 50 percent of Albanian youth reported having Males Females communicated with someone abroad in the Percentage of each mechanism used last month. Telephone is the most common to access the Internet means of communication, but SMS and Source: InterMedia national audience surveys. e-mail are also very popular. Global connectivity and Figure 8.8 Youth are more likely than older people to communicate with people abroad, especially using new technologies the youth transitions Percentage of people using each method Although the main reason for many youth of communication to use computers, the Internet, and mobile 80 phones is entertainment--playing games, Telephone downloadingmusic,andtalkingwithfriends Letter E-mail (table 8.5)--the new ICT technologies 60 SMS are having wide-ranging effects on youth transitions. New opportunities for work and study are opening up, and the interac- 40 tive and decentralized nature of these new technologies is providing youth with many more opportunities to obtain information outside the traditional channels, enhancing 20 their agency. While the majority of youth in many developing countries still do not use the Internet or mobile phones, the experi- 0 Ages Ages Ages Ages Ages Ages Ages Ages ence of those who do shows the possibilities 15­24 25­50 15­24 25­50 15­24 25­50 15­24 25­50 and potential benefits of increased access. Albania Bangladesh Iraq Romania Because the spread of these technologies Source: WDR 2007 InterMedia surveys. is very new in many developing countries, Note: The figure reflects communication with people abroad in the past month. much of the impact has yet to be carefully evaluated. Thus, in many cases this Report can describe only how the new ICTs are being used to enhance youth transitions, sourcing employed approximately 695,000 without providing systematic evidence on people in India in 2004­05.62 One estimate the magnitude of the effects. Even so, the suggests that 11 percent of all service jobs rapid and continuing growth of ICTs in worldwide, amounting to 160 million jobs, developing countries suggests that their could be carried out remotely.63 However, importance for youth will increase. actual offshore employment is predicted to only reach 4.1 million by 2008, suggest- Broadening opportunities and providing sec- ing considerable scope for future growth. ondchancesforwork. Business process out- Such employment acts as an alternative to 204 WORLD DEVELOPMENT REPORT 2007 Table 8.5 What do youth do online? migration, allowing workers to sell their muting, this opens options for youth with Percentage of Chinese Internet labor overseas without having to leave their disabilities.68 However, young people with users ages 16­25 reporting that they country. disabilities often have among the least sometimes, often, or always use the The average age of a call center employee access, due to higher likelihoods of low Internet to: in India is 23, with employees more likely income and education, and to physical Play games 72 to be male and urban and to have upper barriers such as Internet cafés that are not Download music 70 secondary or tertiary qualifications. While wheelchair accessible or equipped with the Do general browsing 69 wages are much lower than in developed necessary technologies. One example of a Read news 61 countries,they are high by developing coun- self-sustainable business model providing Search entertainment information 61 try standards, creating a new generation of opportunities for disabled youth is Digi- young professionals who are often the first tal Divide Data (DDD), a data outsourc- E-mail 53 in their families to have a debit card, ben- ing center in Cambodia that employs only Online chat 50 efits, and to live alone or with roommates. youth who are orphans, land mine victims, Online study 35 Other jobs created for youth by ICTs include physically disabled, or trafficked, with each Work 31 employment as programmers, Internet café of the more than 100 employees receiving Check product information 30 workers, local language Web site developers, vocational training and scholarships to con- Search for medical information 20 and village phone operators.65 tinue their education.69 In addition to directly creating jobs, ICTs Blog 15 provide information about non-ICT job ICTs diversify the range of learning oppor- Online professional training 11 openings to youth. Online job databases, tunities, but lack of education can be a bar- Source: The Chinese Academy of Social such as one run by the Philippines Depart- rier to their use. Distance education has Sciences Internet Survey taken in five cities, January 2005.64 ment of Labor and Employment, offer incorporated television and radio for more information to those with Internet access. than 60 years, and these traditional ICTs are In poorer developing countries, mobile still the most cost-effective ICT educational phones are particularly important for job interventions for secondary schooling in information. In South Africa and Tanzania, many developing countries, helping to meet many respondents identified mobile phones the challenge of extending schooling beyond as essential for contacting employers and primary (chapter 3). For example, Mexico's getting contacted about job openings, par- Telesecundaria program gives those finish- ticularly in remote areas and areas of high ing primary school in rural areas a way to crime.66 The newer uses of mobile technol- continue their schooling without having to ogy are also proving useful for job informa- travel long distances.70 More than 1.2 mil- tion. OKN Mobile in Kenya provides a job lion students in 16,500 locations receive information service called Kazi560, which televised lessons, followed by in-class work sends SMS (text message) job advertise- guided by a teacher. One teacher is used to ments to job-seekers, who pay a small fee cover all subjects, rather than the subject per listing received. The service, with more matter specialists used in general second- "Go to your nearest ICT building, than 30,000 subscribers,is targeted at poorer ary schools--cutting per student costs is it accessible to wheelchair users? workers for whom the cost of job informa- in half. But concerns remain about qual- Will the blind visitors be able to tion has been prohibitive--the SMS infor- ity, with Telesecundaria students scoring use at least one of the computers? mation is marketed at one-tenth the cost of poorly on the international Programme There may be yes for those from a newspaper or bus fare into town.67 for International Student Assessment tests. the developed countries but in New ICTs also offer the potential for a The challenge is thus to raise quality while developing countries especially in second chance at work for youth with dis- maintaining the low cost. Africa it is a BIG NO." abilities, but this promise has yet to be For tertiary education, the Internet's achieved for many disabled youth. Speech capacity for two-way interaction offers the Ambrose, Uganda synthesizers and text magnifier programs greatest promise for improving access and October 2005 can allow visually impaired youth to use affordability and for providing flexibility to ICTs for work, while e-mail and SMS offer combine work with further study. Several greater flexibility in work-related commu- developing countries already cater to sub- nication needs for the hearing impaired. stantial numbers of online students, while Many ICT jobs do not require mobility, developing country students also take online and coupled with possibilities for telecom- classes from developed country universities Moving and communicating across borders 205 without having to migrate. For example, information gains from Internet searches "I use ICTs to educate children and close to 1 million students are studying are naturally less for youth who are unable adolescents about environmental online in China, while the U.S.-based Uni- to read, process, and choose among differ- protection and sustainable versity of Phoenix had students from about ent sources of information, or even spell development as well as to give a 90 countries in 2003.71 Such programs can the words they are looking for. The difficul- possibility to educators and civil expand access and save on costs. Mexico's ties are compounded for many developing society organizations to share Tec Milenio uses professors from its par- country youth by the lack of access to con- visions, information, experiences, ent university to deliver online courses to tent in their native languages. In 2002, 72 etc. regarding this subject." modestly equipped satellite campuses for percent of the world's Internet pages were Cecilia, Argentina one-third the original cost, opening access in English; 7 percent in German; 6 percent October 2005 to young working adults. In other countries, in Japanese; 3 percent in Spanish; 3 percent however, poor infrastructure, low incomes, in French; 2 percent in Italian, Dutch, and and government regulation limit the access Chinese; and 1 percent or less in any other of youth to online education. One approach language.77 Education in global languages, in these circumstances is to create learning especially English, is thus key to expand- centers that combine online classes with ing access to global content, together with local moderators and technology (box 8.3). development of local language Web sites. Considerable debate surrounds the cost- effectiveness and justification for public Facilitating more informed reproductive provision of specialized classes in com- health decisions. The private and anony- puter skills in poor developing countries. mous nature of the Internet offers youth the Some initiatives, such as the World Links program in Uganda, lessen digital divides in access and have spillover benefits, with B O X 8 . 3 Moving in fits and starts with technology-- 80 percent of secondary school students in the African Virtual University the program indicating that they had taught friends or family members some computer Tertiary education in many Sub-Saharan face-to-face in-class sessions with teaching skills.72 Until costs fall, however, computer African countries is hampered by limited assistants.Supplementary use of the Internet resources,empty libraries,and excess demand lowered costs significantly,but satellite tech- provision is unlikely to be financially pos- for classes.The AfricanVirtual University (AVU) nology is still needed because of poor tele- sible in many poorer developing coun- uses new technologies to help remedy this communication infrastructure in the region. problem,increasing access to quality tertiary try educational systems, with the cost of a The AVU has provided courses to over education in the region by tapping into 24,000 participants.Degree, diploma, certifi- computer lab amounting to between 2 and global knowledge and educational institu- cate, and short-course programs are offered 21 times the total discretionary budget per tions.But its experience illustrates the travails in a range of subjects, including computer primary student, according to one calcula- of working with evolving technologies and science, public health, languages, journal- tion.73 Even when computers are provided, a the challenges currently facing online educa- ism, accounting, and business administra- tion in developing countries. tion.Current joint university programs lack of infrastructure and trained personnel The AVU grew out of a World Bank pilot include business studies offered through can inhibit use: in the Dominican Repub- project initiated in 1997.Its rocky start Curtin University in Australia, and computer lic computers sat in their boxes for more raised concerns about its viability.Because science offered through RMIT University in the ICT infrastructure in Africa was in its Australia and Laval University in Canada. than four years at some schools because infancy, the initial delivery approach used AVU also provides a digital library, offering of inadequate or absent electrical capac- digital video broadcasting over satellite access to international journals and e- ity.74 Using computers already in school networks, which was very expensive and books, substituting for empty libraries. for computer-assisted learning can help--a offered only limited interactivity with teach- The AVU, a work in progress, will need to ers.Rapid advances in Internet protocol continue to evolve with technology.African program run by the nongovernmental orga- standards during 1998­2001 made online universities still are likely to pay 100 times nization Pratham in India resulted in sizable learning feasible--and African Virtual more for Internet service than institutions in improvements in mathematics skills.75 University's 100 percent satellite-based North America.The remaining challenge is Although specialized education in ICTs approach outdated and inefficient. finance.The AVU pilot relied too heavily on AVU reassessed its technology options donor financing and private sector subsi- may not be required, a lack of education in 2001 to reduce costs and improve the dies.The learning centers are now financed hampers the use of the new ICTs. A study connectivity and efficiency of networks.The through course fees and educational grants of mobile phone use in several African delivery approach now consists of a mixed from local universities and governments. mode methodology,incorporating online countries found that rural uptake of SMS and satellite video broadcast courses,pre- messaging was low because of illiteracy and Sources: www.avu.org; Halewood and Kenny packaged learning materials on CD-ROMs (2006); International Telecommunication Union indigenous language, even though text mes- and DVDs,chat sessions with the lecturer,and (ITU) (2005); and Prakesh (2003). sages were cheaper than making a call.76 The 206 WORLD DEVELOPMENT REPORT 2007 possibility to discretely access information while abroad. High prices for international about reproductive health and sexuality calls are becoming less of a problem as that they may be otherwise too embarrassed prepaid phone cards and voice over Inter- to ask or unable to talk about for cultural net protocol (VoIP) calls lower the costs of reasons. One-quarter of young Internet connecting home. Calls to migrant family users in Kathmandu, Dakar, and São Paulo members are also one of the most common reported using the Internet to find informa- uses for village mobile phones. Online dis- tion about sex education and health topics cussion boards and migrant Web sites pro- (box 8.4).78 This is particularly important vide a way for migrants to connect with and for young women in traditional societ- meet others from their community and to ies, who tend to have few other opportu- foster expatriate civic associations.The Haiti nities for obtaining this information. All Global Village Web site receives 500,000 hits young women who had access to the Inter- a month, with 80 percent from outside of net through the World Links program in the country, acting as a central forum for Mauritania reported obtaining informa- those in the diaspora to discuss community tion on sexuality, puberty, and HIV/AIDS affairs and ways to help their country.81 prevention.79 Although developing coun- What policies enhance the try evidence is not available, a randomized experiment among young women at family development impact of planning clinics in the United States offers youth use of ICTs? evidence that computer-based aids to con- Youth use of ICTs matters indirectly for traceptive decisions can improve health development outcomes through the impacts knowledge, increase take-up rates, and on youth transitions--and directly through reduce adolescent pregnancies.80 the large youth contribution to overall ICT use. A few transition and newly industrial Helping migrants stay connected as citizens. countries, such as the Czech Republic, the New ICTs are lowering the barriers between Slovak Republic, Hong Kong (China), the migrants and their home communities, Republic of Korea, and Singapore, have seen enabling them to connect with one another economic growth directly driven by the pro- duction of ICTs. But for most developing countries, ICT use rather than ICT produc- B O X 8 . 4 Staying alive: HIV prevention using ICTs tion is likely to have a much bigger impact on growth. Substantial evidence from devel- More widespread use of television and dialogues far too explicit."The Senegalese oped countries now shows a strong effect radio makes these older ICTs the main com- organizers also focused on radio stations for information technology use on produc- ponents in widespread information cam- rather than cable television--the primary paigns to prevent the spread of HIV/AIDS. vehicle for the global campaign.Radio is tivity and growth, but this occurred only The 2002 global HIV-prevention campaign the most popular and widely available with a substantial lag after the introduction Staying Alive was broadcast on television electronic medium in Senegal--96 percent of these technologies.82 stations that reached nearly 800 million of youth surveyed in Dakar have access to The more recent introduction and rela- homes, as well as radio stations in 56 coun- radio compared with 39 percent to cable tries. Survey results from three cities sug- programming.The proportion of surveyed tively low use rates in many developing gest that people exposed to the campaign youth who knew about the campaign in countries suggest that the contribution of were more likely to talk to others about Dakar was 82 percent, but less than one- ICTs to growth is currently lower than in HIV/AIDS and more likely to understand the quarter in São Paulo and Kathmandu, where developed countries,83 but that the rapid importance of using condoms, discussing the campaign was limited to cable. HIV/AIDS with sexual partners, and getting The Staying Alive campaign continues current expansion should contribute to tested for HIV. to produce content for television and radio, future growth. Positive effects are already The campaign was particularly effec- but it has also embraced the new ICTs, pro- beginning to be seen. Recent cross-country tive where adapted to local conditions. viding an online Web site (http://www.stay- work has found that access to the Internet Although there was a considerable body of ing-alive.org/) in 10 languages with infor- material from the United States, the Senega- mation provided in languages and formats spurs the export performance of develop- lese participants decided to localize their designed to appeal to young people, links ing country firms.84 At an even more micro content based on the fact that, according to a variety of help lines, online discussion level, several studies have documented to one participant,"[t]he countryside and boards, and downloads for mobile phones. the clothes were too exotic, the references improvements in prices received by farm- too westernized [and] the images and the Source: Halewood and Kenny (2006). ers and fishermen, thanks to better access to mobile telephony-- fishermen in India, for Moving and communicating across borders 207 example, use mobile phones to get informa- ber of Internet cafés grew from 100 in 1998 "The use of ICT[s] [has] contributed tion about prices at different ports before to 4,000 in 2000, dramatically expanding to the success of community-driven deciding where to land their catch.85 youth access and generating many Internet- development initiatives by making related jobs.87 Similarly, regulations allowing it possible for project coordinators to The most important government policies to easy entry for prepaid phone card operators source funds . . . through the Internet. foster ICT use are the core elements of any and long distance phone calls over the Inter- ICT[s] ha[ve] also been used in infrastructure policy: sound economic con- net can have large payoffs for youth. exchange programs, whereby youth ditions, regulatory policy promoting compe- Regardless of whether the government is exchange ideas on projects that have tition, and complementary infrastructure. involved in Internet provision, governments [been implemented] and projects Yet uncertain market demand and network can help stimulate demand for new services that are currently being carried out." externalities may lead the private sector to by providing public service content online. Thomas, Zimbabwe underprovide access, providing a rationale Governments can reach youth through the October 2005 for further government intervention to serve media they use. They can also kick-start local rural areas. The case is clearest for cellular language content, preventing a vicious cycle telephony, due to mounting evidence linking in which those who do not speak global lan- greater access to telephones to several devel- guages do not use the Internet because of a opment outcomes. dearth of content, while the lack of users acts The Internet is a newer technology, and as a disincentive to local-language Web site less evidence is available, making it still too creation. The government of Tamil Nadu early to recommend direct government pro- offers one such example, providing seed sup- vision of Internet infrastructure. However, port to online initiatives and working with because the costs of delaying the introduc- the private sector to decide on a standardized tion of ICTs are also difficult to measure, Tamil keyboard and Tamil character encod- and the development of ICT skills is seen by ing scheme. As a result, use of Tamil on the many to be necessary for workers to take part Internet was reported to be far greater than in the global economy, governments may any other Indian language.88 want to speed the diffusion of this technol- The current generation of youth is the ogy. Governments have a mixed record in first experiencing the Internet in many this area, and those that do choose to directly countries,with all the pros and cons.Parents provide access to underserved areas can learn unfamiliar with the new technology and not from countries like Chile, where the Enlaces present when it is being used thus have little program combined infrastructure provision ability to protect young people from some with teacher training and decentralized sup- of the dangers. This raises issues of how to port, leading to widespread use in schools. In teach young people to be safe and responsi- the Dominican Republic, however, the provi- ble users of this new technology, protecting sion of computers was not accompanied by them from some of the risks of unfettered complementary infrastructure and person- access, such as child pornography, hate nel, resulting in unused computers in some groups, stalkers, pedophiles, and cyber bul- locations and lack of use for educational lies. In early December 2005, three of the purposes in others.86 top five search terms on the Internet, and 68 of the top 200, were sexual.89 This presents Regardless of their position on direct pro- a problem for youth who wish to use the vision of Internet access, governments can Internet to seek reproductive health infor- increase the benefits of ICTs for youth. A mation: web-filtering programs can block youth perspective on ICTs reveals that gov- useful content, while unfiltered searches for ernment regulation affecting communal teen sex are likely to result in pornographic modes of access determine youth access. content. Moreover, parents and society may Regulation can have dramatic effects on the consider some content appropriate for an incentives for private entrepreneurs (often 18-year-old but not for a 12-year-old. youths) to set up Internet cafés. A reform Given the vast amount of information of the licensing process in Algeria made it available, many youth may be unprepared extremelyaffordable($13)toobtainauthori- to sort through and judge what is reliable zation to provide Internet service. The num- and what is not. There is thus a need to help 208 WORLD DEVELOPMENT REPORT 2007 youth become safer and more effective users youth to migrate--in a way that is benefi- of the Internet. The natural place for this is cial for both countries and for the migrants in schools, but in many countries access to themselves. Temporary worker programs the Internet is available only out of school.So offer one promising avenue. Governments experimentation is needed with alternative also need to experiment more with pro- mechanisms for teaching youth how to use grams to prevent trafficking, and carry out these new ICTs safely, perhaps through gov- careful evaluations of these policies. ernment partnerships with telecenters. Little The main ICT priority for governments is known about what works in this area. is to ensure a good investment climate that allows private companies to serve the grow- ing demand for ICT services, by enacting Young people are extremely active partici- regulations that provide for easy entry and pants in the global flows of migration and competition. For youth it is particularly information. What then should be the pri- important to also provide good regulatory orities for governments to take full advan- conditions for modes of communal access, tage of this involvement? Table 8.6 provides such as village phones and Internet cafes. some first steps.Developing country govern- Governments also need to experiment with ments can try to increase the opportunities ways to provide youth with the skills needed for their existing migrants by allowing them to best take advantage of new technologies, to retain citizenship links with their home through teaching global languages, pro- countries, making it easier and cheaper for viding support for local language content them to send money home, and removing development, and developing ways to teach barriers to their return. But the main prior- youth responsible and safe use. Rigorous ity is for sending countries to work together evaluations of such policies are needed to with receiving countries to work out bilateral find out what works and to share lessons arrangements that expand opportunities for across countries. Table 8.6 Policies for youth in a global world Proven successful Promising but unproven Unlikely to be successful Opportunities Expanding opportunities for Bilateral agreements (the Temporary worker programs (Canada's seasonal Skill-based points systems developing country youth to migrate Philippines) agricultural workers program) (Australia and Canada)90 Increase opportunities for existing Regulations that allow open Absentee voting and dual citizenship (Mexico, Tax incentives to encourage migrants entry and competition (multiple Mozambique, and others) migrant return (Malaysia) countries) Enhancing access to information and Easy licensing of communal Government support for local language content Equipping schools with communications access provision (Internet cafés (Tamil Nadu government, India) computers without in Algeria) incorporating them into the curriculum (Dominican Republic) Creating new opportunities to work Teaching global languages (India) Online tertiary education (China and Tec Milenio in and study in developing countries Mexico) SMS and online job listings (OKN Mobile, Philippines and Sri Lanka online job listings) Capabilities Improving knowledge and context Family housing for young male Pre-orientation seminars for migrants (the Strong censorship of sexual surrounding HIV/AIDS migrants (Cameroon oil pipeline Philippines, Thailand) content on the Internet project) Computer-based contraceptive decision making (U.S. family planning clinics) Creating responsible Internet users Working with telecenters on education programs Second chances Preventing trafficking and help victims Information campaigns and rehabilitation of victims Restrictions on young (GTZ program against trafficking in women, BMZ, women's right to exit (Sudan Germany) and the Republic of Yemen) Providing work opportunities for Use of ICT to provide jobs for disabled youth (DDD disabled youth in Cambodia) spotlight What donors can do What donors can do 209 Donors can help governments undertake policies to broaden opportunities, enhance capabilities, and provide second chances to young people. Key actions for donors: applying the youth lenses of opportunities, capability, and second chances to their own policies, and funding the evaluation of the promising but unproven policies identified in this report. Determining the extent to which donors are evolving from mainly formal education in grams. But it would also be useful to have involved in supporting youth outcomes in the late 1990s to areas such as promotion of an overarching strategy to ensure prioriti- developing countries is a Herculean task-- healthy behavior, livelihoods and employ- zation and to harvest lessons across coun- one that almost defeated the writers of this ment, nonformal education, and family and tries. Donor agencies do this relatively well report! As with national governments, very community support--which now account for well-established sectors such as educa- fewdonorstrackprogramandprojectactivity for around 40 percent of lending for youth tion and health, but are only beginning to systematically by age range. The suggestions initiatives. Such an assessment can be used do so for youth (see the box). here are based on information and feedback to address key questions at the country from a limited number of donors--a com- level. Are there gaps? Are there overlaps? Are Developing the capability of young prehensive review is left for others. interventions sufficiently prioritized? people as decision-making agents Answering these questions would require One of the key issues discussed in this Report Donors through the youth lenses consideringsynergiesacrosstheinterventions is whether young people are recognized as How do donor policies and programs fare arising from the multisector nature of youth agents who have voice and who make deci- when seen through the "youth lenses" that outcomes--a difficult task for most agencies. sions that shape their human capital. There this Report has used to evaluate country For example, many interventions encourag- are two activities donors can undertake to policies? There appear to be large gaps, with ing young people to adopt healthier lifestyles promote voice and enhance the decision- considerablescopeforgreaterdonorinvolve- need to occur outside the health sector (chap- making capabilities of young people. One is ment to expand opportunities, enhance ter 5). Vocational education needs to provide to do more as global advocates for youth,par- agency, and extend second chances. a broader set of skills than those required for ticularly the most disadvantaged. The other immediate employment in an identified sec- is to involve young people more directly in Broadening opportunities to develop tor or industry (chapters 3 and 4). policy and program development. human capital Much of this work needs to take place in As global advocates, donors can fur- Much international investment in human the groups within donor agencies and other ther promote the adoption of international capital is focused on improving opportu- organizations that determine country pro- conventions, such as the United Nations nities for children in education and health in developing countries. This is consistent with the focus on preventive approaches and investment at early ages--a theme Organizing bilateral and multilateral youth programs emphasized by this report. However, the focus on children and basic services means In 1997, the German federal ministry for eco- of different projects and greater sharing of that youth issues are sometimes neglected. nomic cooperation and development (BMZ) knowledge across sectors affecting youth. Although those over age 12 form around became one of the first international donors · New approaches have been tested, includ- 30 percent of those in the age range 0­18 to make children and youth a major thematic ing attempts to integrate different fields in developing countries as a whole, there is area for development assistance. Over the past (education and conflict transformation, little consistent focus on this group. decade, BMZ has supported a wide range of information and communication technolo- Given the global size of this youth projects intended to benefit young people. gies to support vocational training, and peer- cohort and the changing nature of the risks One of its implementation agencies, German to-peer education to prevent HIV/AIDS). Technical Cooperation (GTZ), uses several and opportunities they confront (chapter Some multilaterals have also begun to take mechanisms to promote the integration of 1), complementary programs are needed action.The Inter-American Development Bank youth development into its overall strategy, that build on the earlier investments and was the first development finance agency to test new youth-oriented projects, and enable address the needs of older children and establish a youth strategy, and it has become a coordination across sectors: major funder of youth employment ventures. youth. What more can be done? · The organizational structure includes health, In 2003, the World Bank appointed a Children An important task is a comprehensive education, and youth as sectors under the and Youth adviser to develop a more youth- assessment of how present country assis- same division. Regional meetings of manag- focused approach and facilitate coordination tance strategies serve young people. The ers in these three areas are frequently held. across the Bank. youth-content of World Bank loans, for · Heavy investment in knowledge manage- Source: Author's discussions with officials at BMZ example, has grown over the past few years ment allows better access to the experiences and the Inter-American Development Bank. with the focus of youth-oriented lending 210 WORLD DEVELOPMENT REPORT 2007 conventions on trafficking and child labor, could also be supported through holding of older learners.2 Other second-chance which have increased international atten- regular consultations with youth as a part programs that could benefit youth include tion on these issues.1 By coming together of preparing youth-focused reports or proj- retraining programs well linked to the labor under the Youth Employment Network, ects. In the "Youth as Partners" work of the market and cost-effective restorative justice donors have played an important role in World Bank,a representative group of youth and rehabilitation programs. raising awareness and building commit- from different sectors of Brazilian society is ment to addressing youth unemployment. regularly convened to provide feedback on Moving forward: However, advocacy needs to be comple- the country program. Young people were mented with specific policies and programs also directly involved in the preparation Funding evaluation of promising to be more effective. For instance, donors of the youth development projects in the but unproven programs,and (BMZ, Sida), NGOs (Oxfam, Coalition Dominican Republic and FYR Macedonia. sharing lessons across countries against Trafficking in Women), and inter- Many programs highlighted in this Report Supporting second-chance programs national organizations (ILO, International are promising rather than proven, and most Organization for Migration, and UNICEF) Many young people need second chances, need to be adapted to the particular circum- have all supported programs to reduce illustrated most starkly by countries that stances of individual countries and their human trafficking and assist the victims. have "lost generations" due to conflict or youth. Donors can promote adaptation and However, there is very little knowledge shar- political and economic calamities (see spot- innovation in program areas that may be ing, and few of these programs have been light on Sierra Leone following chapter 7). too new or too risky for governments; they evaluated for their effectiveness, making it Youth literacy is often low in poor coun- can also support evaluation. World Devel- hard to know what works well. Donors can tries, but it is much lower in poor coun- opment Report 2004: Making Services Work invest more in pooling knowledge and sup- tries recovering from years of conflict, as for Poor People, highlighted the fact that porting more rigorous evaluations of their illustrated by Afghanistan and Sierra Leone many donors, including the World Bank, programs. in the figure below. While donors have often have resources for evaluation that go There has also been action on child responded strongly by rebuilding schooling unused. A good use of resources would be labor. The Understanding Children's Work infrastructure, often much more is needed to evaluate the promising programs identi- joint research program of the ILO, UNI- to take into account the very different needs fied in this report, such as the following: CEF, and the World Bank, initiated under the Oslo Agenda for Action, has advanced · Life-skills education (chapters 3, 4, 5, the global agenda on children's work. In Youth literacy is low in many countries, and 6) addition to improving understanding of especially those emerging from conflict · Provision of information on the returns children's work, it has resulted in greater to education to increase retention in coordination of the efforts of different Percentage of 15 ­ 24-year-olds who are literate schools (chapter 3) 100 government departments and agencies · Programs that enhance youth citizen- involved in addressing child labor in coun- ship such as school student councils, tries such as the Republic of Yemen. Similar 80 national service and other service learn- joint efforts could be developed to address ing programs, student parliaments, and the gap in data on youth in general, and the 60 youth leadership development programs shortage of rigorously evaluated interven- (chapters 3 and 7) tions (discussed in chapter 9). 40 · Programs that build skills among poor Donors could also draw upon young Sierra Leone youth through public works or wage people more directly in policy and program subsidies to firms to hire young new development. This could be through finan- 20 Afghanistan workers (chapter 4) cial and technical support to young people or their organizations to give them greater · Restorative justice programs for young 0 means for getting involved in international, 0 offenders and mobile court systems that national, and local development processes. 2,000 4,000 6,000 8,000 10,000 12,000 14,000 deal with young offenders (chapter 7) Inviting youth delegates to the General GDP per capita, PPP (constant 2000 $) · Use of the Internet and SMS to provide Assembly of the United Nations is one such job listings, as in Kenya, the Philippines, Source: Authors' calculations based on Fares, Montenegro, initiative. Another is the training provided and Orazem (2006a) and World Bank (2006h). and Sri Lanka (chapter 4 and chapter 8) by the German government to young pro- Note: Data shown are for 92 country-year observations from 64 countries. The line shown is the fitted line from a two- fessionals and future leaders. Youth voice tailed tobit regression of literacy on log GDP per capita. Youth policy: Doing it and getting it right 9 In a world that demands more than basic dents. Second-chance programs will still be skills, expanding the opportunities for needed, but not on the scale of low-income young people to learn means going beyond countries. In addition to a macroeconomic c h a p t e r primary schooling. Because building climate conducive to growth, countries may human capital takes more than schooling, need to reform labor market institutions to it also means including young people more better accommodate new entrants. in their communities and societies so that Knowing what to do is not enough-- they have opportunities to use their talents policies directed at young people often fail. at work and to participate as active citizens. Why? For three main reasons: First, influ- Informing and facilitating their decision encing youth transitions requires work- making--so that they choose well--adds to ing across many sectors, yet few countries their success as workers and entrepreneurs, take a coherent approach to establish clear as parents, and as citizens. And for those lines of accountability for youth outcomes. who have to recover from poor decisions Many countries have a vision for young or poor circumstances providing second people--typically articulated in a "national chances--to make up for missed opportu- youth policy" that fails to set priorities or nities--can keep young people from being coordinate action. These policies need to irrevocably left behind. be revisited and revamped. Even where For many low-income countries, the pri- policy is well articulated, it may stand alone ority is to provide quality basic education from national development policy. Poverty (including lower secondary education) for Reduction Strategy Papers in low-income young adolescents and, where HIV/AIDS countries are a case in point. Until recently, prevalence is high, such as in parts of Sub- there was little integration of young peo- Saharan Africa, to safeguard health by pre- ple's issues as part of a wider societal effort venting transmission among young people to combat poverty. This is changing, but who are beginning to be sexually active. not fast enough. Greater capacity is also Both will help in the transition to work. needed--for analysis, for integration with For many young people in these countries, national policy planning and budgeting basic skills for life and work will have to be processes, for coordinated implementation, acquired through second-chance programs. and for monitoring and evaluation. Making full use of young people's skills and Second, young people often lack voice in providing them further opportunities to the design and implementation of policies build skills on the job will require maintain- that affect them.1 Governments at all levels ing an overall macroeconomic environment also need to be more open to listening to conducive to growth. young people,particularly older youth begin- For middle-income and high-growth ning to engage more formally as citizens. As countries that have already provided mass important clients of many public services, basic and then secondary education, like young people can improve quality by par- those in East Asia and Eastern Europe and ticipating in implementation and provid- the former Soviet Union, the challenge is ing feedback. Promising experiences of this to develop a tertiary education system and kind of voice are having an impact on service to address new health threats to youth, delivery, as in the anticorruption efforts of such as tobacco use, obesity, and road acci- tertiary level students in Eastern Europe. 211 212 WORLD DEVELOPMENT REPORT 2007 Third, there are few success stories--few tive capacity to implement policies and pro- policies and programs that have proven grams. Managing income-contingent loan effective. Too many policies directed at programs to encourage the take-up of higher young people, including many discussed in education requires high levels of adminis- this Report, are promising but unproven, trative capacity (especially in tax adminis- and this is a serious impediment to wider tration), making such programs difficult to replication. More needs to be done to find implement in a low-income country. Other out which policies and programs improve financial incentives--such as conditional youth outcomes and why. This is not easy, cash transfers--are found to be effective in because of spillovers from one youth tran- a wide range of contexts where many young sition to others, and because some policies people do not go to secondary school even may be more effective in combination than though the facilities are there. on their own. Knowing what works, what The third dimension is the stage of the does not, and in what circumstances would demographic transition. Is the country be of tremendous value in improving policy proceeding through the demographic tran- for all countries. sition, and is there a window of opportu- nity from declining dependency? Or is the demographic transition yet to start? If the Youth policy priorities latter, the resources available for investment vary by country context in the human capital of children and youth Previous chapters focused on policies to are steadily falling (Chad, Niger). In these influence investment decisions across five countries, an emphasis on basic child and youth transitions: learning, working, stay- maternal health services is essential to lower ing healthy, forming families, and exercis- fertility and improve the lives of children ing citizenship. This entire range of policies and youth. At the other end of the spectrum that influence youth outcomes constitutes are countries where the window of oppor- what shall be referred to as "youth policy" tunity from falling dependency will soon in this chapter. close (Armenia, China). These countries We will not repeat these policies here should act now before rapid aging results in but ask instead: how are the policy recom- sharper trade-offs between the needs of the mendations discussed in chapters 3­8 and young and those of the elderly.2 summarized in the overview of this Report Fourth is the degree to which young to be customized to each country context? people influence decisions concerning It would seem that at least four dimensions themselves, discussed in chapter 2. In soci- should be considered. First, as discussed eties where decision making is communal, in chapter 1, the starting levels of young the family or community may have as much people's skills and capabilities (their human say, or more, in decisions affecting young capital) is the basis on which countries can people (as in Bangladesh). In societies that build. This can be determined from answers are more individualistic, the views of young to questions such as: what are primary and people may be more important to the final secondary completion rates? How healthy decision (as in Malaysia or Romania). While are young people entering adolescence? autonomy in decision making bears some What are the main threats to their health in relationship to income, some young people the short and long terms? Are young people in poor countries report as much autonomy building skills through productive employ- in decision making (Ethiopia) as their peers ment in the labor market? Are they prepared in richer countries (Iraq).3 for parenthood? And are there opportuni- While they may overlap, these dimen- ties for them to participate in community sions are distinct. In general, starting levels and society? of human capital and income are impor- Second, country income determines the tant in setting priorities for what countries extent to which young people and their fam- should focus on. The stage of the demo- ilies can afford to pay for these investments graphic transition indicates the urgency for themselves. It is also a proxy for administra- country action. The autonomy of young Youth policy: Doing it and getting it right 213 people in decision making influences whom A fast-growing economy to target and the type of intervention. Vietnam is a low-income country that has gone beyond the basics (see spotlight on A very low-income country Vietnamese youth following chapter 3). "We are not respected in the Take Sierra Leone (see spotlight following Enjoying its demographic dividend,second- community; they call us rude and chapter 7). Typical of many countries in ary completion rates are high, and a rapidly unruly ex-combatants." Sub-Saharan Africa with very low levels of growing economy readily employs young Young, commercial bike income and human capital, it is also shifting people. Dependency rates will continue rider and ex-combatant, from postconflict reconstruction to long- to decline for the next two decades, giving Bombali District, Sierra Leone term development. However, it has yet to scope to address the challenge of improv- February 2006 undergo the demographic transition. So, in ing the quality of higher education and the addition to maintaining sound fundamen- risks to young people from the rapid pace of tals for growth, improving access to and the economic and social change. quality of primary education and primary Vietnam now needs to expand access to health care must be a priority. This will give a more diverse and flexible upper secondary young people the basics for life and work as and tertiary education system, reorienting well as create the conditions for the transi- curricula so that young people learn relevant tion to lower fertility. practical skills (information technology, Getting children into school is not languages) and life skills (problem-solving, enough, as many of today's youth in Sierra working in teams) while creating stronger Leone did not go to school and are now connections between school and work (as illiterate. Offering them a second chance at with the university-industry link in China). school (as undertaken in Ghana's functional Tertiary education could be expanded by literacy program) will improve their chances reaching out to the private sector (as in of doing more remunerative work. Build- Chile). But to ensure its quality--something ing skills through employment in public that many developing countries often get works programs in conjunction with or as wrong--Vietnam will need good accredita- a complement to basic literacy should also tion and evaluation systems (Chile and the be considered (as with Senegal's Agetip). Republic of Korea provide good examples). For many young men directly involved in Creating opportunities may not be enough, conflict, rehabilitation that leads to work is especially for deserving students from poor essential. or rural backgrounds. So loans or scholar- The basic skills for good health and par- ships should be targeted to the needy or those enting can be imparted through a combi- who face the greatest conflict with work (as nation of life-skills training in school (as in in Mexico's individual learning accounts). Namibia's "My future, my choice" program), Health risks in Vietnam are growing due through second chance programs, and to alcohol and drug abuse, sexually transmit- through social marketing and media cam- ted diseases, and other risky behavior. So, paigns attractive to young people outside young people could benefit from more infor- schools (as in Horizon Jeunes in Cameroon). mation and stronger incentives to make the Even as young people take greater respon- right decisions. Providing life skills necessary sibility themselves, information and out- for better health in lower secondary schools "The shortest adult is taller than the reach--on health risks, on birth spacing, and could help (South Africa recently reformed tallest youth." on other reproductive health issues--should its secondary curriculum to include these be targeted not just at young people but more skills). Given the substantial increase in road Young person, broadly at families and communities. accidents among young men, traffic safety is Tonkolili District, Sierra Leone In a society rebuilding itself, institutions a life skill worth emphasizing. February 2006 should be genuinely open to the participa- Outreach and dissemination of informa- tion of young people. Ensuring their legal tion on sexual and reproductive health for identity and their representation in local youth, the primary decision makers in these community organizations and imposing matters,need to be undertaken with a respect milder penalties on young offenders can do for confidentiality. Even with the best infor- much to promote a sense of solidarity. mation, however, many young Vietnamese 214 WORLD DEVELOPMENT REPORT 2007 will stumble, making second-chance pro- education. The success of conditional grams important. They are also important cash transfers--such as Oportunidades in for vulnerable youth such as street children Mexico, not the standard education sector and commercial sex workers. policy instrument--lies precisely in the fact "Teachers. . . . teach theoretically, so Vietnam is increasingly providing more that they address competing demands on this disadvantages youth in learning opportunities for youth to participate in pub- young people's time (see chapters 3 and 6). well and being able to find work." liclife,butitcoulddomoretolegallyrecognize And with individual transitions as diverse as Young person, Hanoi, Vietnam young migrants to urban areas, whose access they are, young people can be hard to reach. March 2006 to public services is often compromised. Health information and education cam- As different as Sierra Leone and Vietnam paigns have to work across a broad range of may be, they present only a small part of the sectors to reach young people because they diversity among developing countries in the can be in school or out of school, or in work five transitions of young people, diversity or out of work (see chapters 4 and 5). that makes it difficult to draw more general Even where there are no spillover effects conclusions about where priorities should from one transition to another, there may lie. That is why each country needs to still be a need for coordination across sec- choose policies based on its circumstances. tors. Making lower secondary education universal (chapter 3) will in many countries Youth policy often require improvements in transport and other fails young people infrastructure to increase physical access to Knowing what to do is in many ways the schools. In Vietnam, addressing the high easy part of devising policy. Making sure it death rate from traffic accidents may require gets done, and done well, is more difficult. better road safety rules and stronger enforce- Youth policy has had at best mixed suc- ment (see spotlight on Vietnamese youth fol- cess in ensuring smooth transitions for the lowing chapter 3). For many young women, young, for at least three reasons: taking advantage of opportunities to learn or work may also require relief from some ele- · Poor coordination among policies and ment of domestic drudgery through invest- sectors that affect youth and limited ments in efficient sources of energy and water accountability for youth outcomes (see spotlight on gender following chapter 2). · Weak voice of young people in monitor- In most countries, sector ministries are ing and providing feedback on the qual- responsible for the bulk of policies that ity of policy and service delivery affect youth, while ministries or depart- · The paucity of proven successes ments of youth are responsible for youth outcomes (box 9.1). This does not make Poor coordination and for clear lines of accountability, because limited accountability departments of youth typically have little Youth outcomes, extending across the five influence over their sector counterparts. In transitions, are influenced by more than just some countries, youth ministries have trig- the health sector or the education sector. gered opposition from sector ministries to Because the transitions overlap--outcomes the detriment of youth policies. In others, in one sector are influenced by policies leaders of youth departments are political in another--many sectors need to work appointees with no particular interest in together when devising policies to influence developing or monitoring a medium- or outcomes. Addressing the difficulties in long-term vision for youth. With a range of going to work requires looking beyond the sectors responsible and the lines of account- labor market to macroeconomic policy and ability weak,no sector is fully accountable-- the investment climate and to the quality and youth outcomes suffer as a result. of education and training (as discussed in chapter 4). Inculcating citizenship involves Weak voice in policy formulation learning in school (chapter 7). Raising and service delivery enrollments requires reconciling the con- Although the interests of young people flicting choices of work and marriage against may be reflected by other, older groups in Youth policy: Doing it and getting it right 215 decision making at the local and national levels,4 it is striking how seldom young B O X 9 . 1 What do ministries of youth do? people are consulted on policies that affect Youth issues typically do not command an people and their organizations. them. According to one estimate, about entire ministry of their own.In most countries, The role of departments of youth is two-thirds of countries with national youth youth issues are dealt with by departments of partly a function of how far countries have councils, bodies intended to reflect the youth housed within other ministries-- progressed in developing and implementing views of different stakeholders, do not listen typically education (Jamaica and Sweden) or policies for the young.In many Organisation in omnibus ministries responsible for youth for Economic Co-operation and Development to youth nongovernmental organizations as well as sports,culture,social affairs,or local countries where policy development has (NGOs).5 Very few young people or youth government (India and Singapore). been well established over several decades, NGOS are consulted as a part of the poverty Their functions vary but broadly consist the role of youth departments is mainly policy of developing a framework for youth issues development,coordination,and monitoring. reduction strategy process in low-income and formulating policy,fostering coordi- In many developing countries only countries (discussed further below). nation,collaboration,advocacy,research, beginning to develop coherent poli- Service delivery is another area in which technical assistance,and monitoring.The cies toward young people,there is more youth are not visible. Even though young departments sometimes support special emphasis on advocacy and advice.In others, programs,outside the traditional line min- departments have moved to policy develop- people are important clients of public istries,directed to youth at the national or ment (Chile,El Salvador,and Nicaragua) and services, they often are not consulted. Yet local level.They may also maintain contact coordination (Chile). as direct clients, they can be crucial to and channels of communication with young Source: Mattero (2006). improving the quality of service delivery (see box 2.2).6 Many health services suf- fer from a very youth-unfriendly image, paying little attention to young people's Except for the Jovenes programs from Latin "Many organizations have visited concerns and needs (see chapter 5). Few America (which train out-of-school youth), us, but after we explained our school systems recognize young people as almost no second-chance programs are rig- situation to them we do not see stakeholders who can improve the qual- orously evaluated. Even where evaluations them again [there is no follow-up]." ity of education (see chapter 3). Yet young exist, they may be on narrow outcomes. The Young person, Kissy, Sierra Leone people have every incentive to hold public Jovenes evaluations look largely at impacts February 2006 services accountable, something they get on employment and wages, and not the bet- better at as they mature. ter health or lower crime that may go hand in hand with employment. Paucity of proven successes Few solid evaluations of youth programs in Getting it right-- developing countries unambiguously iden- by developing a coherent tify the causality from policy to program framework and integrating to effect.7 This gives youth policy the aura it with national policy of being soft, lacking in rigor. Much more What can be done to ensure that policy is not is known about what affects outcomes in destined for failure? To increase the chances childhood, thanks to a large and growing of success,policy makers must as a start artic- body of solid evaluations of the impact of ulate a coherent view of desired outcomes "Youth participation is token early childhood development programs, for young people, integrating that view with and . . . it is frustrating when school nutrition programs, school man- national planning and implementation [the authorities] send you . . . to a agement reforms, remedial primary edu- cation, and so on.8 As the tables at the end mechanisms. This will improve accountabil- meeting and they are not bothered ity for outcomes. The capacity to implement by the sets of recommendations and of the chapters 3­8 detail, many programs this strategy at all levels is also required. that are important for expanding oppor- follow up. . . ." tunities and building capabilities fall into Rotimi, 22, Nigeria the "promising but unproven" camp. They A coherent national June 2006 include most life-skills programs (which framework for youth many countries are investing more in) and Accountability is easier to assign if there is most programs for promoting youth citi- a well-articulated set of national objectives zenship, including student councils, youth for youth, developed with key ministries parliaments, and service learning (includ- and stakeholders. This is especially impor- ing national service). The lack of evidence tant where countries are decentralizing is not confined to expanding opportunities. public service delivery, a growing trend (see 216 WORLD DEVELOPMENT REPORT 2007 chapter 1). Many initiatives that focus on HIV/AIDS campaigns). A clear national youth--from smaller schemes responding framework can guide coordination. to the needs of special groups to programs However, the policy vision at the country that tackle national issues--often intersect level--often articulated through national at the local level, where most services that youth policy--is usually very narrow in influence youth are provided. Appropri- scope. In 2001, 82 percent of all countries ate enough in federal systems, they can be had a national youth policy, 89 percent had plagued by fragmented effort and poor a designated national youth coordination coordination, especially when services are mechanism, and 60 percent were imple- provided by more than one level of govern- menting a national youth program of action. ment (such as education in Brazil) or where But in 70 percent, national youth policy was effects overlap across jurisdictions (as in focused on the narrow range of concerns of the department of youth, with few links to other youth-serving ministries, such as edu- cation, health, or labor.9 Even where broad, B O X 9 . 2 Does addressing gender issues hold lessons it may not be very strategic, reading more for youth policy? The view from East Asia like a laundry list of desirable policies than Since the Fourth World Conference on responsibilities of different government a strategic program of action. Women in Beijing in 1995,governments in departments. Revisiting these policy statements can East Asia have set up legal and institutional Build the capacity to mainstream. The sharpen the focus on youth in conjunc- frameworks to promote gender equality. agency responsible for making sure that Most have defined and approved gender gender concerns are part of sector policy tion with building a broad constituency for plans of action,establishing mechanisms needs strong capacity,as do the sector min- youth outcomes. Having the key economic and bodies to ensure that gender concerns istries and other agencies--through suit- and finance ministries on board obtains are adequately addressed at all stages able mainstreaming mechanisms,adequate wider ownership of policy among public of policy planning and implementation. budget allocations,and technical training. Departments dedicated to women's empow- In some countries (Cambodia),the appoint- agencies, as evident from the progress with erment have been elevated to ministries,as ment of gender focal points responsible policy development in areas such as gender, in Cambodia and Indonesia,and high-level for incorporating gender issues into sector which--like youth--cuts across sector lines intersector coordinating bodies have been agendas has been only partly effective,and (box 9.2). And the development of youth formed,as in the Lao People's Democratic alternative or complementary mechanisms Republic,the Philippines,and Vietnam. (interministerial working groups) are being policy would greatly benefit from consulta- pursued.In most countries,technical train- What needs to be improved? tions with young people.10 ing on gender issues has been easily incor- Weaknesses in implementation and in In a national framework, many policies porated into general training programs. monitoring and evaluation are common Strengthen the women's movement in civil affecting youth outcomes are best addressed throughout the region,because many society.Although the position of women in within individual sectors. For example, relevant public institutions are still weak, national parliaments has hardly changed lacking resources and capacity,and strug- expanding the range of options open to over the decade,there has been significant gling to define their roles with the more young people in upper secondary school progress with policies that affect women. established sectors. Much of this is the result of women's stron- and increasing flexibility among them, What has been learned? ger voice in civil society to exert pressure for recommended in chapter 3, can be tackled Four lessons: change.In Indonesia,Korea,Mongolia,the largely within the education sector. Mak- Design a comprehensive gender strategy Philippines,and Thailand,women's move- ing preventive health services more youth involving all key ministries.A comprehen- ments have brought about changes in poli- sive gender strategy is needed so that cies and helped implement laws on labor, friendly,recommended in chapter 5,can also gender concerns are addressed up front in family,domestic violence,and trafficking. be handled within the health sector. What a country's overall development program. Monitor and evaluate.Few countries is required for many policies is that minis- All line ministries and specialized agencies have put adequate systems for impact tries or agencies help rather than hinder the should be involved, so that gender equality evaluation in place on time.Suitable indi- falls to the Ministry of Women's Affairs.The cators and data collection and analysis efforts of others. In Chile, the Department involvement of the Ministries of Economy mechanisms need to be developed from of Youth conducts a survey of youth condi- and Finance ensures that gender concerns the beginning,involving national statistical tions every three years, providing valuable are effectively addressed in the formula- institutions and research institutes.In Cam- feedback to line ministries on the efficacy tion and execution of the national budget. bodia and Lao PDR,national statistical sys- For instance, the interministerial working tems are being adapted to incorporate dis- of their programs. Departments of youth groups set up in Cambodia involve line aggregation by sex into national statistics. have limited expertise implementing health ministries in the design of the gender strat- Sources: Asian Development Bank and World or education policies. Some, though, have egy and in specific gender actions. And the Bank (2005); Brown,Al-Hamad,and De Paz 2002 National Strategy for the Advance- considerable expertise convening groups of Nieves (2005); Javate de Dios (2002);World ment of Women in Vietnam sets out the Bank (2005g); and World Bank (2005h). young people and their organizations. This could be put at the disposal of traditional Youth policy: Doing it and getting it right 217 line ministries that may wish to consult or implemented by traditional line ministries. obtain feedback from young people on the Ensuring that they understand their roles quality of services provided to them. and responsibilities builds and maintains a Coordination is needed, however, in constituency for youth issues. developing an overarching strategy for Integrating youth policies with national addressing youth outcomes; assigning clear development frameworks is in its infancy in roles and responsibilities for implementa- most developing countries. In low-income tion; getting ministries and agencies, civil countries, youth issues often sit outside the society organizations, and the private sec- poverty reduction strategy (PRS) process, tor to work with each other; monitoring failing to tap the various constituencies progress with implementation; and policy that support PRS outcomes. Although more development. Coordination in this sense recent PRS papers do a better job than older is no different from what is required for ones of integrating youth outcomes, cover- other cross-sector objectives of national age is still far from universal (box 9.3). policy, such as improving gender out- Many middle-income countries suf- comes. Because youth transitions affect one fer from the same poor integration with another, the evaluation of policies and pro- national development frameworks, per- grams also needs to be coordinated. haps attributable to the strong regional and historical flavor of national youth policy. Integration with national policy According to one view, the European Union planning and implementation and the British Commonwealth have tradi- mechanisms tionally promoted cross-sector, integrated Policies are more likely to be success- approaches, while the Arab Council of Min- ful if youth issues are well integrated into isters of Youth and Sports and the Confer- national policy planning and implementa- ence of Ministers of Youth and Sports of tion mechanisms, because most of them are the French-speaking Communities have B O X 9 . 3 Neither seen, nor heard--youth in the poverty reduction strategy process For a group that makes up a large share of Youth in poverty diagnostics viduals to poverty over the long run,such as the the population in most low-income countries, Only one in five papers (6 PRSPs) identified failure to acquire an education. young people are underrepresented in the pov- young people as a group vulnerable to poverty. Youth in poverty action plans erty reduction strategy process. A little over a third (12 PRSPs) identified youth Few PRSPs take an integrated view of the needs vulnerability to poverty in a minor way,while of young people in their action plans. Just Youth in consultations another one-fifth (6 PRSPs) identified youth as three-quarters of the papers (24 PRSPs) men- In a study of 31 poverty reduction strategy one of several groups vulnerable to poverty (see tion young people in their action plans, only papers (PRSPs) completed by September 2003, table 1).In part an issue of how people in this a half address youth issues as a key goal. A around half of the papers (14 PRSPs) were found age range get classified,many PRSPs also take a quarter of action plans do not mention youth at to be drawn up without consulting youth groups relatively static view of poverty.A more dynamic all (see table 2). This is not to say that policies at all as part of their overall consultation process. view would look at factors that predispose indi- that affect youth outcomes, such as education, do not figure prominently.They do, but not as Table 1 Are youth identified as a group vulnerable to poverty? part of a coherent cross-sector approach to addressing the needs of the young. Even where Major focus Minor focus One of many No mention Total there is reference to young people in the PRSP Number 6 12 6 7 31 action plans, there is little follow through in specific targets or budgetary allocations for Percent 19 39 19 23 100 implementation. Progress since 2003 is mixed. Provisional analysis of 55 PRSPs completed by April 2006 suggests that more PRSPs now identify young Table 2 Are youth in PRSP action plans? people as a group vulnerable to poverty. How- Major focus in key goal Minor focus in key goal No mention Total ever,a coherent cross-sector approach to youth issues is still lacking. Number 17 7 7 31 Percent 55 23 23 100 Source: Curtain (2006) and United Nations (2003). 218 WORLD DEVELOPMENT REPORT 2007 given more sustained attention to youth and Building this capacity is clearly criti- sports than to policy integration.11 Making cal and will take resources, but given what progress on integration with national policy is spent on youth through traditional line is thus relevant in a wide range of contexts. ministries,the additional resources are likely Integration with budgeting frameworks to be a small proportion of what is already is equally weak. Only a handful of devel- expended. The ability to hire staff with the oping countries systematically track how right skills is absolutely vital. In addition much is spent on young people, let alone to analytical skills, staff need to communi- relate it to youth outcomes.12 This kind of cate effectively with young people and their integrated view is essential to making prog- organizations and work across traditional ress with youth policy. sector boundaries. The ability to work with those outside government--including the Building capacity corporate sector and not-for-profit agen- Implementing a coherent and integrated cies, both engaged in improving outcomes national framework requires strong capaci- for young people--is equally important. ties--for analysis, policy development, The work of youth departments is likely to implementation, coordination, monitoring, be enhanced by having strong inputs from and evaluation. However, those capacities organizations that represent the interests have traditionally been weak in developing of young people. So building capacity in countries.In countries with a long history of youth-serving NGOs is also important. youth policy, the department of youth plays With youth policy still new in many the role of "champion" and "policeman" of developing countries, there are few good youth policy effectively, as in Sweden (box examples of where capacity has been built 9.4); but in many developing countries, effectively. However, in some countries in departments of youth are under-resourced Latin America, youth agencies are mov- both in financial resources and in person- ing in the right direction, focusing less nel--and thus lack the clout to be effective. on implementing programs and more on B O X 9 . 4 How do developed countries handle youth issues? Consider Sweden Sweden has a coherent national framework self-support,and culture and leisure--the five Capacity for youth,well integrated with national policy transitions as it were,Swedish style. The implementation of national youth policy planning and implementation mechanisms,and is aided by strong institutions.The Division with a strong capacity for implementation. of Youth Policy at the Ministry of Education, Integration with national planning Research and Culture coordinates youth policy Policy framework and implementation mechanisms within the government.The Minister for Pre- Swedish youth policy stretches back more than Swedish youth policy is highly"mainstreamed," School Education,Youth Affairs,and Adult 50 years. At the beginning, the main ambition well integrated with national policy planning Learning leads a system in which 20 or more was to create good scope for after-school and budgeting processes.Ten or so ministries governmental authorities monitor the living activities. Policy has since moved on.The foun- take responsibility for realizing the established conditions of young people. dations of an integrated national youth policy youth policy goals.As in most other countries, Most of the responsibility for living condi- were laid in the mid-1980s, with continuous government activities are divided into various tions of young people in Sweden rests with evolution since. sectors,each with its own goals and its own municipalities.A special agency,the Swedish The latest policy revision--set out in a new budget.All sectors that affect young people are National Board for Youth Affairs,supports the youth policy law,Power to Decide,enacted in expected to help achieve national youth policy development of municipal youth policy and 2004--provides a new structure for national objectives. monitoring of the objectives of national youth youth policy with the aim of giving all young Goals are established within each of the policy.It supplies up-to-date knowledge of people equal opportunities to develop,to be sectors for each of the five transitions identified young people's living conditions both locally empowered,to gain influence over their every- in the new law.This integrates youth policy into and nationally.It also provides support for the day lives,and thus to be able to realize their the relevant sectors and facilitates realizing and development of knowledge-based intersector dreams.The target is young people between the monitoring the overall objectives within the youth policy at the municipal level with a high ages of 13 and 25.Five main fields have been framework of the normal day-to-day activities in degree of youth influence. created to facilitate analysis and coordination-- the area.Each sector contributing to the national learning and personal development,health youth policy has its own indicators to measure Source: Forum 21, European Journal on Youth Policy, and vulnerability,influence and representation, how it is doing from a youth perspective. available on line at www.coe.int/youth/forum21. Youth policy: Doing it and getting it right 219 policy development, coordination, and Where departments of youth are headed: monitoring (box 9.5). Governments clearly B O X 9 . 5 need to experiment and innovate, especially Evidence from Latin America where they may be starting to build capac- Over the past two decades,most Latin complements the longer-term vision and ity from scratch. The lessons from attempts American countries have aimed to proposes interventions that are immediate, to address other cross-sector issues, such as strengthen youth departments (called measurable,and capable of being evaluated. gender equality, suggest that much can be secretariats,vice ministries,or presidential Each intervention has an assigned budget programs).The record is mixed.A general and relies on implementation by designated achieved over a decade (see box 9.2). lesson from the more promising reforms sector ministries and agencies. Because most is that youth departments should focus programs are implemented by line ministries, on guiding--rather than implementing-- the youth agency ,INJUV (Instituto Nacional de The payoff from having a coherent national national youth plans. la Juventud),focuses on coordination among In countries such as El Salvador and Nica- sectors and critical follow-up.INJUV's national framework for youth integrated into ragua,youth departments have achieved survey on youth,conducted every three years, national policy with capacity for imple- substantial success in contributing to the for- provides sector ministries with critical knowl- mentation is well illustrated by the model mulation of national plans,defining priorities, edge upon which to base programs targeting described in chapter 1, projecting the and helping to align sector policies. youth.Through its evaluations and research In many countries where sector min- on youth programs,INJUV also contributes impact of HIV/AIDS in Kenya. As discussed istries already identify their own priorities to the formulation of policies affecting youth. in chapter 1, the model predicts that the for youth policy,the most appropriate con- The next step will require further collaboration HIV/AIDS epidemic will take a heavy toll tribution of youth departments is to help with sector ministries to address issues identi- on growth in Kenya because of its impact coordinate such policies. fied in the evaluations related to youth.For In countries that have set forth ambi- many youth agencies,including INJUV,reach- on human capital accumulation and in tious long-term visions for youth develop- ing out to youth will also require collaboration particular on secondary school comple- ment,such plans need to be translated into with NGOs and youth organizations. tion rates. To combat the impact of the short- and medium-term action if there is to epidemic on school enrollment, the govern- be a realistic attempt at implementation. Source: Instituto Mexicano de la Juventud (IMJ), In Chile,a two-year youth action plan Organización Iberoamericana de la Juventud ment can either act directly on enrollment (Chile se Compromete con los Jovenes) (OIJ) (2006) and authors. by providing a subsidy to encourage school attendance or act indirectly to improve young people's health. The model suggests that at plausible discount rates, the net pres- affect them. Many national youth policies "As youth we are passionate and ent value of the benefits from the education are drawn up without adequate consulta- eager to make a difference. We are subsidy lies between 1.7 and 3.5 times that tion with young people. Even the institutions like warriors; all we want is just to of the costs. Under a combined program-- meant to reflect the voice of young people, be heard and given the chance to in which the educational subsidy is halved such as national youth councils, often fail make a difference." and the other half is spent on measures to to do so.13 Many low-income countries do combat the epidemic and treat its victims-- 21-year-old Liberian-American not consult with youth groups as a part of the net present value of the benefits would June 2006 the PRSP process (see box 9.3). This should be 8.9 to 13.3 times greater than the costs be remedied, particularly for older youth, (box 9.6). This striking improvement is due who may be exercising more indirect voice not only to saving lives under the combined through the ballot box. Efforts to give young intervention but also to the fact that the people voice need to go beyond the tokenism resulting reduction in expected mortality that often characterizes such attempts. Not provides an additional potent incentive to only does there need to be a process for lis- invest in education. tening to young people--there also needs to be a process for careful consideration of the Getting it right-- suggestions and feedback that emerge. by listening to young people An often-voiced criticism of more partici- As the most important clients of policies and patory approaches is the danger that youth services directed at them, the ability of young elites or other unrepresentative groups will people to exercise voice, or client power, can capture the process.With adequate planning, be invaluable to ensuring quality. Govern- however, a wide representation of young ments should thus be open to consultation people can be ensured, bringing benefits and feedback from young people on the both to policy making and to young people design and implementation of policies that (box 9.7).Denying young people an adequate 220 WORLD DEVELOPMENT REPORT 2007 B O X 9 . 6 Successful policy coordination and implementation: How health and education policies can work together to combat the AIDS shock Box 1.3 in chapter 1 discussed one model of Figures 1 and 2 plot the impact of these Both interventions hasten the recovery of how HIV/AIDS could affect human capital accu- interventions on postprimary enrollment and per capita income to 1990 levels--that is,they mulation and growth,calibrated using data from on per capita income.Both measures raise reduce the impact of the AIDS shock.Recovery Kenya.Here,this model is taken a step further postprimary enrollments,but with different time is fastest under the education subsidy.Even and asks what would be the impact on human profiles.The education subsidy,which costs 0.9 though the combined intervention leads to capital accumulation and growth of policies to percent of GDP in 2000,rising to 1.8 percent of higher postprimary enrollment and greater increase enrollments and reduce infection rates GDP in 2020,has the greatest impact on raising human capital in the long run,the lags in the among the young. enrollments up to 2020.If some of these funds system are such that in 2040,the outer year of Education subsidies are modeled assuming were allocated instead to health promotion, the projections,per capita output is higher for that 50 percent of the cost of secondary educa- all the effects of reduced mortality--and the the education subsidy. tion is borne by the state and funded by donors. expectation of reduced mortality--on invest- In view of the very different time paths of An alternative to this is to reduce the subsidy to ment in secondary education come into play.In the streams of costs and benefits for the two education by half and allocate the same stream comparison with the direct working of educa- interventions,an evaluation of their social profit- of outlays to combating the epidemic.Mea- tion subsidies,these effects are small at first,but ability requires that they be discounted at an sures that promote education directly,such as they accumulate over time,so much so that,in appropriate (real) rate.Assuming a discount rate school attendance subsidies,increase income by the model as calibrated,enrollments are higher of 4 percent,both programs yield significantly increasing human capital.Measures to combat under this program from 2020 onward. The higher benefits than costs (see table below). the epidemic,primarily through reducing infec- combined program would,in fact,cost slightly However,the intervention that involves the tion rates among 15- to 24-year-olds,not only less--0.8 percent of GDP in 2000,rising to almost promotion of health has a benefit-cost ratio reduce the toll of suffering and death but also 1.8 percent in 2020.Yet it yields higher second- roughly three to four times that of the interven- promote investments in education by increasing ary enrollments by 2030 than an education sub- tion involving educational subsidies alone. the families'lifetime resources and the expected sidy alone.This points to an important,long-run returns to education investments. synergy between education and health policy. The social profitability of education and health programs Figure 1. Impact of interventions on per Figure 2. Impact of interventions on Aggregate Per capita capita income postprimary enrollment GDP GDP benefits/ benefits/ Per capita income Postprimary enrollments Index costs costs Index of per capita income (1990 = 100) Postprimary enrollment rate (percent) School attendance 3.5 2.9 125 80 subsidy 115 Education subsidy Combined 13.3 11.7 60 plus health promotion program--school attendance and 105 health promotion 40 95 Education subsidy 20 85 No intervention Source: Bell, Bruhns, and Gersbach (2006). 75 0 Note: All benefits and costs discounted at the rate 1990 2000 2010 2020 2030 2040 1990 2000 2010 2020 2030 2040 of 4 percent a year. hearing can lead to less constructive expres- ing tertiary education, introducing student sions of voice. rights and mechanisms for student protec- The voice of young people can also tion, and reducing corruption. The results improve service delivery by monitoring and so far--more effective students' ombuds- "The lack of youth participation giving feedback on the quality of services. men are being instituted in several univer- leads to a lack of sense of belonging For example, students and student orga- sities; the monopoly of old-time student to the society which in turn leads to nizations in Eastern Europe have notched unions (often dominated by ruling elites) further alienation." important early wins in the fight against has ended; and dialogue has been initiated corruption in tertiary education through with parliaments, education ministries, Young person, Thailand the exercise of voice. In 2003, a regional net- and university administrators and faculties December 2005 work of student NGOs from Albania, Bul- to reduce corruption in higher education. garia, Croatia, FYR Macedonia, Moldova, While it is too early to say if this process and Serbia and Montenegro was formed to will lead to sustained improvements in improve governance in higher education. It issues that particularly affect students, this focused on changing regulations govern- is a start.14 Youth policy: Doing it and getting it right 221 B O X 9 . 7 Vozes Jovens: Opening national youth policy to youth voice and participation in Brazil It should come as no surprise that Brazil's strong National Youth Council,the governing body notes, "We were able to form our own youth civil society has given birth to an energetic of the National Youth Secretariat,made up of network,Rede Nacional de Juventude-RENAJU, youth movement.As part of the government's government and civil society representatives. which is represented in the National Youth effort to open up to civil society,it took on the State governments are now applying the Vozes Council.Vozes Jovens provided young lead- challenge of incorporating the voice of youth Jovens methodology,with the states of Ceará ers with the opportunity to sit on the National into policy development.As a complement to and Pernambuco recently hosting Vozes Jovens Youth Council,where they can help build the this process,the World Bank initiated a dialog events to enhance local policies.Today,more youth policies Brazil needs."16 with youth,Vozes Jovens (Youth Voices),to assist than 300 youth organizations participate in Challenges,however,remain.A truly national in the development of the World Bank's country local,subnational,and national policy processes. youth policy in Brazil requires harmonizing a assistance strategy in Brazil. This process has brought benefits to both diverse array of social and political interests (see Over time the Vozes Jovens grew into a the government and young people. According spotlight on Brazil following chapter 5).Young broad dialog on youth issues,independent of to Minister Luiz Dulci,of the General Secretariat people also introduce new ways of doing busi- the World Bank,involving representatives of the of the Presidency, "Vozes Jovens has been fun- ness--different language,different organizing Brazilian government and congress,political damental in contributing to the dialogue the methods,and different ways of presenting cases parties,other civil society organizations,and President has launched to ...empower Brazil- and results--that need to be incorporated into the private sector.It provided valuable input ian youth,and has resulted in the creation of policy development and programming.There is for establishing the National Youth Secretariat, sound policies for the country's youth,such as also a need for greater policy integration at vari- which coordinates youth policies at all levels of the second chance program PRO JOVEM,the ous levels of government,better priority setting government. National Youth Council and the National Youth across policies,and even more recognition of Vozes Jovens allows youth to express their Secretariat,within two years of President Lula the diversity of needs among Brazil's youth. needs,consolidates efforts of many youth taking office."15 organizations,and most important,institutional- The young civil society leader and National izes youth voice through participation on the Youth Council member Josbertini Clementino Source: Authors. Young people can also do a great deal to They can be used to arrive at an assessment create opportunities to exercise voice. The of the main issues confronting the young, in spotlight that follows this chapter, by young some cases, serving as an "early warning" on people and for young people, provides problems (risky health behavior,lack of legal examples of how to go about this. identity). Where there are large disparities within countries, it would be important to Getting it right--through go deeper than these countrywide measures monitoring and evaluation to subnational or local estimates. Owing to the importance of learning in Youth policy, as mentioned previously, is school, several measures are recommended bedeviled by the shortage of solid evaluations to monitor schooling, including school par- establishing the impact of policies and pro- ticipation, school completion, and school- grams on youth outcomes. There is also little ing quality.At a minimum, the preparedness consensus on what the most relevant out- of children for postprimary learning could comes might be. So establishing priority out- be assessed on the basis of literacy rates at comes and developing a body of evidence on the end of the primary (or early second- what works, what does not, and in what cir- ary) cycle. More sophisticated measures of cumstances are critical to advancing policy. learning could also be used, such as those derived from the Programme for Interna- Monitoring tional Student Assessment (PISA) and the Because many interrelated outcomes are Trends in International Mathematics and relevant to poverty reduction and growth, Science Study (TIMSS), which monitor this chapter lists a range of indicators that learning at the end of the primary cycle and emerge from the discussions in chapters at the end of compulsory schooling (lower 3­8 (box 9.8).17 (The tables at the end of secondary). This would require a substan- this Report present data on many of these tial expansion in coverage of these tests in indicators for a large number of countries, developing countries. disaggregated by sex). The indicators allow Because the productive use of young benchmarking and country comparisons to people's labor is important for both indi- monitor progress in different dimensions. viduals and society,detrimentally early tran- 222 WORLD DEVELOPMENT REPORT 2007 B O X 9 . 8 A youth scorecard? The many indicators of youth development In selecting indicators,the focus is on those relevant from a poverty reduction and growth perspective.Some may not be readily available from existing data sources,and are marked with an asterisk;most could be collected with little effort by adding questions to existing survey and census instruments.18 Learning 1. School enrollment by gender 2. Primary, lower secondary, and upper secondary completion rates among youth, by gender 3. Learning achievement--end-primary and end-secondary--by gender * Going to work 4. Incidence of child labor, by gender 5. Labor force participation rates, by gender 6. Unemployment rates, by gender, for rural and urban youth 7. Percentage not at work and not in school, by gender Staying healthy 8. Percentage currently using tobacco, by gender 9. Percentage currently using drugs including inhalants, by gender * 10. Body mass index, by gender * 11. Percentage of sexually active youth engaging in unprotected sex, by gender and by marital status 12. Percentage of youth with knowledge of how to prevent HIV/AIDS, by gender 13. Percentage of 15-year-olds who will die before reaching their 60th birthday, by gender Forming families 14. Age specific fertility rates 15. Percentage of young women giving birth before 18 16. Percentage of young women using antenatal care Citizenship 17. Percentage of youth who have worked together with someone else or some group to solve a problem in the community where they live, by gender * 18. Percentage of youth who correctly answer a question concerning political knowledge appropriate to the country, by gender * 19. Percentage of youth without identity papers, by gender * 20. Number of youth in presentence detention, per 100,000 population, by gender* Moving across 21. Percentage of youth studying abroad, by gender* borders and 22. Percentage of youth working abroad, by gender* communicating 23. Percentage of youth migrants returning within 10 years of migrating, by gender* 24. Percentage of youth who have used the Internet in the past month, by gender* Source: Authors' recommendations based on discussions in chapters 3­8 of this book. * The indicator is not currently collected on a regular basis. sitions to the labor force could be monitored delivery for this age group is important for through the incidence of child labor. Post- influencing risk-taking behavior, indicators poned transitions, also costly from a human are proposed for both knowledge (of safe capital development perspective, could be sexual practices and methods of preventing monitored through measures of unemploy- HIV/AIDS) and use (antenatal care) among ment and discouragement (being not at the young. The focus in family formation work and not in school). It is not enough to would be on the incidence of early mother- know whether the young are employed. The hood (before the age of 18). quality of employment also matters, espe- Encouraging young people to partici- cially in low-income countries where unem- pate in the broader political and democratic ployment may be something of a luxury, but processes of society builds a responsible it is notoriously difficult to pin down. citizenry, so participation in the local com- Young people incur multiple threats to munity and broader political awareness are their health, so monitoring tobacco use, proposed as measures of active and passive drug and inhalant use, unsafe sexual activ- citizenship. It would also be important to ity, body mass, and the probability of dying document government failures to provide before the age of 60 can provide estimates of legal identity to youngsters--and harsh the extent of these risks. Of these measures, punishments for young people (presentence statistics on drug and inhalant use and body detention). mass index are not collected systemati- Under migration and communication, cally in large numbers of developing coun- the proportion of young people abroad for tries. Because the efficacy of health service work and for study is of interest, as is the Youth policy: Doing it and getting it right 223 proportion who return after a reasonably but also the mother's health (through likely short time. Access to the Internet would lower fertility) and her children's health also need to be monitored. and mortality.19 For another example, the All these indicators are either collected impact of reducing crime among youth may routinely or could be collected routinely at have returns that go beyond direct improve- little marginal cost. Better monitoring of ments to the safety of local citizens and may outcomes for young people is thus within include indirect benefits through increases easy reach,and could contribute enormously in international tourism and foreign invest- to the analytical basis for improving policy. ment. Fully accounting for these spillover effects is important for understanding the Evaluation full benefits of investing in young people. Evaluation is not within easy reach because These spillover effects also make it more it is more intensive in both skills and difficult to evaluate youth investments resources. However, concerted efforts could using traditional cost-effectiveness analysis. improve the knowledge available for policy Such analysis--of, say, cost per disability making. life year saved or cost per beneficiary placed Spillovers from one youth transition to in employment--is widely used to evalu- others make impact evaluation difficult. For ate investments in health, training, and example, the impact of investing in an addi- other activities. It requires a single effec- tional year of schooling for girls beyond tiveness measure for comparing alternative primary school would need to take account investments, impractical where outcomes not only of potential additional earnings are multiple and varied.20 The alterna- (which may be small in societies where tive would be to use cost-benefit analysis, female labor force participation is low), comparing cost with the monetized sum of B O X 9 . 9 Credible proof of a program's success can ensure continuity: The case of Oportunidades Oportunidades (formerly Progresa),one of Mex- Why has it been successful? ship or close ties with the administration in power. ico's best known"brands,"gives cash benefits to Multisector programs are notoriously difficult It is run by a new agency that coordinates with poor households conditional on their maintain- to implement because of the complexity of the ministries responsible for various government ing the use of education and health services, coordinating operations across ministries,with services.The national parliament establishes particularly for children and youth.Introduced different mandates and sometime conflicting the program's budget and annually publishes in 1997 as a part of an effort to break the inter- objectives.How have programs such as Opor- its operating rules and the number of families generational transmission of poverty,it now tunidades managed successful implementation? enrolled.Enrollment is closed for several months covers some 25 million of the poorest people Central backing. From the beginning, Opor- before national elections,and payments are not throughout the country,with a budget of about tunidades had very strong central backing, made before voting days in any jurisdiction.A 0.36 percent of GDP in 2005.The evaluation including the full confidence of the executive transparent and nonpolitical system for allocating program was designed at the same time as its arm of the government,from the president to benefits is one of the program's hallmarks.21 implementation,permitting rigorous evaluation the finance ministry official who was one of its Rigorous monitoring and evaluation. From of a complex program with many components. cofounders.This allowed it to break through its inception the program has emphasized bureaucratic logjams and to secure adequate monitoring and rigorous evaluation,collect- What has it achieved? funding despite the usual vicissitudes of annual ing baseline data and longitudinal household Oportunidades has boosted enrollments and budget appropriations.Fiscal space permitted and service-provider data with treatment and narrowed the gender gap in secondary school. the program to grow over time. control groups.Initial evaluation was con- The bulk of the positive impact on school enroll- Piloting and outreach. Many operational tracted to an outside agency (International ment lies in higher continuation rates rather issues that needed to be addressed--such as Food Policy Research Institute),and continues than in returns of children who had left school. efficient targeting,the need for better monitor- to this day involving academics from three Grade repetition is lower,as are dropout rates, ing and evaluation,and interagency coordina- continents.Credible empirical proof of the especially in the critical transition from primary tion--were identified in an early pilot.Great program's achievements has been essential to to secondary school.Adolescents engage in less effort before the launch of the program built the program's longevity.The program survived risky health behavior. Qualitative analysis sug- consensus and support for the program.This a change in presidents,and was even expanded gests that the program is raising the aspirations effort continues,particularly through educating by the new government. of young people,especially those of girls to beneficiaries about their rights. Sources: Bate (2004); Behrman, Sengupta, and Todd become teachers and nurses and to put off mar- Apolitical allocations.The program goes to (2001); Levy and Rodriguez (2004); Murray (2004); rying until they can acquire the necessary skills. great lengths to avoid any impression of partisan- and Schultz (2000). 224 WORLD DEVELOPMENT REPORT 2007 benefits, but this requires knowledge of the gram on a specified outcome. Perhaps more effectiveness of the interventions across the relevant for young people would be, what range of potential outcomes, which is rarely is the most cost-effective way of reaching available. For example, no evaluations of the a given outcome? This may require coordi- Jovenes training programs in Latin America nated evaluations across a range of relevant look into the impact on health, civic partici- interventions. pation, or crime. Meeting the challenge is essential to Another aspect of youth policies is that advancing policy development. True, evaluat- they may be more effective when adminis- ing is expensive, but because the knowledge tered jointly, because outcomes may have generated is a public good, there is a strong common determinants that need to be tar- case for greater donor effort to assist in evalu- geted together. For example, health infor- ation. Donor assistance has already advanced mation and job skills training may be more the evaluation of policies and programs, but effective in ensuring that young people much more remains to be done (see spotlight obtain and retain a job when offered jointly on donors following chapter 8). than when either is offered on its own-- On top of adding to the pool of policy- because obtaining a job is a function both of relevant knowledge, one neglected outcome being healthy and being qualified. Similarly, of rigorous evaluation is the possible benefit discouraging girls (and their parents) from of insulating good programs from politics. early marriage may be more effective when Oportunidades, the conditional cash trans- girls are in school. In the model discussed in fer program in Mexico, provides a good box 9.6, targeting both education and health example. Found to be effective in boosting outcomes of young people is more effective secondary school enrollment, the evalu- in encouraging human capital development ation of its outcomes was built in at the than education interventions alone. outset. This helped ensure the program's Both these features--spillover effects and longevity (box 9.9). Changes in political complementarities across transitions--sug- leadership during program implementa- gest that when it comes to youth there may tion did not derail the program because its be a case for coordinated evaluations of established effectiveness in addressing the outcomes. Traditional evaluations focus on long-run causes of poverty was hard for the programs and judge the impact of the pro- new government to question. spotlight It's up to you(th)--taking action for development It's up to you--How young people can take action for development 225 This Report's simple message to governments and policy makers is that investing in young people is essential for develop- ment--and for those investments to be most effective, young people must be included as stakeholders in decisions that affect them. But what can young people do? This spotlight addresses young people directly about how to develop their own capa- bilities so they can seize the opportunities provided to them, but especially, so they can create opportunities for themselves, for other young people, and for everyone. The spotlight benefited greatly from consultations with youth from many coun- tries who have successfully taken action for development. It was written by young people, for young people. hold governments accountable--as young Young people make a difference people have done in Brazil (see box 2.4), FYR Macedonia, and Russia (chapter 3). People will decide their lives Powerful older villagers haven't always And "government" can be anything from a People will identify their problems appreciated their involvement:when Chamila local teacher to a global institution. People will lead to self-help (age 19) was elected to the village council,it People will share the fruits would not allow her to take office because of Going beyond the system her youth.But the community,remembering Many young people decide to start some- This is the motto of Chamila,Kamili,Wacheeri, her past hard work and cooperation while vol- and Milmini--four young women from villages unteering on village projects,forced the whole thing new, be it an activist campaign for in Sri Lanka who are helping young people council to resign instead.Chamila was then social justice or a motorbike transport busi- and all people in their villages overcome pov- reinstated with a new council. ness (chapter 7). These youth-led projects erty.Kamili (age 22) says,"Our problems are vil- These four women are just a few of the usually work best when young people work lage problems,and the village's problems are hundreds of poor young people (ages 16­25) together, and train the next "next genera- ours."They have seen resources wasted and who are volunteering and working together tion" to continue their efforts. Students at misguided projects implemented by those try- for community development in Sri Lanka. ing to help their villages,but instead of becom- one college in the United States started the ing disillusioned with development work they first student-run think tank in 2004, and became engaged in it.They sum up their work already, students at 50 other colleges have Source: Interview with Samantha M.Constant and as"Strengthen the village by increasing your Sarath K.Guttikunda.Social Accountability Work- joined the network, and lawmakers have individual capacity." shop.Gemi Diriya Project,Sri Lanka.March 2006. begun to use their research. Young people face many challenges, but every challenge to taking action is an opportunity to take action. Most young This Report recommends that governments Action can be as simple as developing people see a lack of jobs as just a challenge should invest in and recognize young people. one's own capabilities--learning about and to employment. Several large groups of For it to become reality, young people must taking advantage of a training program, a young ex-combatants in Sierra Leone took seize the initiative and invest--by develop- health service, career options--or finding it as an opportunity and joined together ing your own capacity, and taking action to out about and discussing with friends an across opposing sides of the conflict to start make youth investments successful. And as issue that matters to them (the environ- a motorbike transportation business. By young people in Sri Lanka recognize, (see ment, HIV/AIDS). What you decide to do starting a new trade (motorbike transpor- the box above), all meaningful investments and how you do it will depend on your skills, tation did not exist before the war), they in development can positively affect young your personality, the issues that inspire you, avoided the challenges faced by youth break- people. What can be done and how to do it? the resources you have available to you, and ing into established trades and instead were the circumstances in which you live. very successful. Plus, they found a job that Take action: What can be done Working with the system they particularly enjoy as young people. As No matter who you are, you can take action Few societies readily include young people one bike rider (age 18) says, "The war was as young people to make life better for in policy decisions (chapters 7 and 9), but exciting, but bad. To be a bike rider is excit- yourself, and at the same time, for others many young people overcome that chal- ing and good." Still they had to work to gain around you. Some of the poorest boys and lenge to volunteer or work for existing the trust of clients, and learn to cooperate girls (ages 12­15) in Malawi and Nepal, for NGOs, or cooperate with governments to with the traditional system: rather than example, have formed sanitation clubs-- improve policies. Through youth service reacting violently to perceived unfair treat- sometimes in spite of ridicule from their organizations in countries such as Kenya, ment by police, they hired a lawyer.2 peers--to build latrines and maintain water Pakistan, and the Philippines, youth have pumps for their schools, improving their developed and carried out active solutions Tools for action: How to get it done daily experience and the health of their to community and national problems. In Young people often face complex challenges, communities. Mexico 2,000 young people volunteer to such as lack of information and experience, The many things young people have improve literacy and civic participation skeptical and resistant authorities, bureau- accomplished fill books and reports,1 and through one of many large programs that cracies that favor qualifications over hard there are few limitations on the ways young involve young people. Young people have a work and skills, lack of resources, and lack people can contribute to development. key role in exercising their "client power" to of recognition for their work. Despite these 226 WORLD DEVELOPMENT REPORT 2007 obstacles, young people can still take action cies and actions of those who are supposed Voice to develop their capability--through per- to serve you and your community. Recon- You can hold civil servants, social workers, sistence, a willingness to experiment, and sider what you already know and what you and policy makers accountable. Another four basic tools. hear. As you take action, do research to important way to use your voice is to inform evaluate whether your efforts are making a Learn and learn from other young people. difference for those you want to help. One young activist from Italy found an Find out as much you can about how to alternative way to express her voice in the build skills for your work and your life, · "Asking for advice is the most important face of challenges: about both sides of development issues you tool that we have as youth. . . . My advice are passionate about, about different ways to other youth leaders is not to ignore the I was tired of taking part in pro-youth of taking action. small people . . . sometimes they know consultation processes ending up in One youth UN delegate from the Neth- more than those at the top," says a Libe- nothing, and was tired to keep talking erlands, spent"half a year"lobbying his gov- rian youth leader. but not having anyone listen. . . . I then ernment to participate in the Millennium · Check out TakingITGlobal's youth action decided that I [would] try to change it, Summit, and relied on his expertise on the guide, search for others, or create your and worked toward the aim of setting a issues to convince them to allow him to go: own.* youth platform, Youth Action for Change, to make my voice heard--and every other After a certain time they knew me, they · Talk with an open mind to people who young person for that matter--heard.... knew that I wasn't just smiling and repre- you think your action will help, and find The project proved very successful . . .5 senting the youth, but also, I could really out what they need--and then adapt help them with experts on different youth your plan to make it work A youth leader in Singapore says: issues. What I know a lot about because of · Find out about youth involvement in my work at the Red Cross is HIV/AIDS. monitoring and evaluation of programs About voice: know how it works. It's dif- But if they wanted to talk with child sol- that are designed to serve youth. ficult to effect any real change if you don't diers, I could help them. . .3 know the political infrastructure....Talk Cooperate to youth-friendly people in civil service How can this be done? or NGOs which effectively engage with If you are going to make change effective, · "Use the power of ICTs to get informa- government. you must work with others: form alliances, tion," writes a young volunteer in Nige- collaborate, work collectively. Just as impor- These four tools are just the beginning. ria (chapter 8). To start, learn from Web tant, learn work to with people who have Use these and others to motivate your plans sites such as the Development Gate- the power to help. and develop capabilities for action. way, the UN, and the World Bank, and A young woman from Kenya became the many sites for specific issues and the first UN youth delegate from Kenya in Who can help national organizations.* 2005 by taking initiative to voice important Many groups exist to help young people · Take advantage of programs that pro- issues and cooperate with Kenyan officials: take action. If you would like their help, vide information to young people, such From the beginning of the year, from my make your voice heard. As in the story from as Umsobomvu Youth Fund in South organizational level, I lobbied and advo- Sri Lanka, young people can often have a Africa, or those that provide informa- cated for the youth to be included in all big impact at the local level. Global youth tion about services for young people the processes on the MDGs. . . . We never organizations (mentioned above) are a net- (education, jobs), such as the ASER missed their meetings, we worked so work for many small, local organizations.At project in India (chapter 3).* hard to have our thoughts and ideas put both the local and global levels the UN has · Learn how to apply for small grants in. When it was time for the ministry to worked to include young people. The World for projects, work with people, and appoint the delegation, they appointed Bank has begun to involve youth organiza- write proposals for the many resources the most active and persistent.4 tions by providing courses, and by consult- provided by youth organizations, for ing with youth about its programs. At the example, "Global Youth Coalition on Join a local or national youth organiza- local level, the World Bank has civil society HIV/AIDS (GYCA) conducts online tion, or a global network of young people and youth liaisons in its country offices. At courses in proposal writing, advocacy, (such as TakingITGlobal, Youth Employ- the global level, see the work of the Youth, etc." advises one youth leader.* ment Summit campaign [YES], the Global Development and Peace Network. Youth Action Network) who are taking In the end, it is up to you--your involve- action. This is an opportunity to find other Evaluate ment is crucial to increasing the role of young people that you can learn from and young people and youth issues in efforts to Be critical. Question your interests, the work with and a way to make your voice alleviate poverty. circumstances around you, and the poli- heard.* *See the following Web sites: Development Gateway: www.developmentgateway.org/youth, GYCA: www.youthaidscoalition.org, Taking IT Global: www.takingitglobal.org, Umsobomvu Youth Fund: www.uyf.org.za, UN: www.un.org/youth, World Bank: www.worldbank.org/childrenandyouth, www.youthink.org, Youth Employment Summit Campaign: www.yesweb.org. Bibliographical note This Report draws on a wide range of World Bank docu- David Bloom, Daniel Kwabena Boakye, Christian Bodewig, ments and on numerous outside sources. Background papers Jan Bojö, Gillian Brown, Peter Buckland, Don Bundy, Dani- and useful comments were prepared by Arvil Van Adams, elle Carbonneau, Nadereh Chamlou, Nazmul Chaudhury, Pablo Angelelli, Sarah E. Anthony, Kaushik Basu, Clive Bell, Samantha M. Constant, Luis Constantino, Richard Curtain, Eric Bettinger, Ramona Bruhns, Pedro Carneiro, Esther Anis Dani, Amit Dar, Jishnu Das, Maitreyi Das, Monica Das Duflo, Pascaline Dupas, Tanja El-Cherkeh, Elizabeth Fussell, Gupta,JoannadeBerry,JaikishanDesai,ShantayananDevara- Sebastian Galiani, Elena Galliano, Hans Gersbach, Elizabeth jan, Karen Dillard, Sophia Drewnowski, Antonio Estache, Gomart, Juan Pablo Gutierrez, Naomi Halewood, Hugo Kan- Shahrokh Fardoust, Tazeen Fasih, Manuela Ferro, Armin tis, Charles Kenny, Michael Kremer, Hyejin Ku, Gloria La Fidler, Ariel Fiszbein, Constance A. Flanagan, Shubashis Cava, David Lam, Juan José Llisterri, Minna Mattero, Sarah Gangopadhyay, Marito H. Garcia, Ines Garcia-Thoumi, Alan Michael, Edward Miguel, Alexis Murphy, Zeynep Ozbil, Jan- Gelb, Tina George, Maninder Gill, Elena Glinskaya, Rachel ice E. Perlman, Agnes Quisumbing, Dhushyanth Raju, Usha Glennerster, Peter Gluckman, Daniela Gressani, Karen Gross, Ramakrishna, Martin Rossi, Juan E. Saavedra, Ernesto Schar- Jon Gruber, Sarath K. Guttikunda, Andrew Hahn, Minna grodsky, Samuel Simei, Elena Stirbu, Mio Takada, Luis Teje- Hahn, Keith Hansen, Inaam Haq, Karla Hoff, Leif Holmberg, rina, Erwin Tiongson, Andrea Tolciu, Gerold Thilo Vollmer, Camilla Holmemo, Robert Holzmann, Macartan Hum- Kathyrn Young, and Homa Zarghamee. phreys, Ingo Imhoff, Farrukh Iqbal, Shweta Jain, Evangeline Background papers for the Report are available either Javier, Malathi S. Jayawickrama, Rajendra Dhoj Joshi, Ursila on the World Wide Web www.worldbank.org/wdr2007 or Jung, Jan Kasprzycki-Rosikon, Omer M. Karasapan, Mats through the World Development Report office. The views Karlsson, Iqbal Kaur, Kei Kawabata, Homi Kharas, Douglas expressed in these papers are not necessarily those of the Kirby, Robert Krech, Angela Langenkamp, Frannie Léautier, World Bank or of this Report. Danny M. Leipziger, Peter Levine, Santiago Levy, Jeffrey D. In addition to commenting on the Report, the following Lewis, Samuel Lieberman, David Lindauer, Rick Little, Ben- also contributed substantially to preparing boxes and spot- jamin Loevinsohn, Elizabeth Laura Lule, Ante Lundberg, lights. Contributions for the spotlights and boxes were pro- Shelly Lundberg, Akiko Maeda, Sam Maimbo, Gift Manase, vided by Ragui Assaad, Nicholas Barr, Mayra Buvinic, Le Thi Viviana Mangiaterra, Ali M. Mansoor, Katherine Marshall, Minh Chau, Ernesto Cuadra, Wendy Cunningham, Carmen John May, Linda McGinnis, John McIntire, Rekha Menon, De Paz Nieves, Ana Cristina Torres Garcia, Andrea Guedes, Pradeep Mitra, Nancy Murray, Mustapha Kamel Nabli, Juan Carlos Guzman, Richard R. Hopper, Sandra Huang, Ambar Narayan, Claudia Nassif, Sophie Naudeau, Kenneth Peter F. Lanjouw, Kathy Lindert, Cynthia Lloyd, Daniel Mont, J. Newcombe, Akihiko Nishio, Tara O'Connell, Eric Olson, Juan Manuel Moreno, Andrew Morrison, Deepa Narayan, Patrick Osewe, Egbe Osifo, Mead Over, Pierella Paci, John Harry Patrinos, Pia Peeters, Patti Petesch, Jenny Petrow, Page, Guillermo Perry, Djordjija Petkoski, Susana Pezzullo, Susana Puerto, Martin Rama, Jamil Salmi, Noala Skinner, Lant Pritchett, Christine Zhen-Wei Qiang, G N V Ramana, Kristian Thorn, Carolyn Turk, Dorte Verner, Zeze Weiss and William Reese, Ana Revenga, Linda Richter, Pia Rockhold, Yesim Yilmaz. Ernesto Rodriguez, Halsey Rogers, Khama Rogo, James Many people inside and outside the World Bank gave Rosen, David Rosenblatt, David Ross, James Sackey, Junko comments to the team. Valuable comments and contribu- Saito, Hussain Samad, Stefano Scarpetta, Norbert Schady, tions were provided by Charlie Abelmann, Daron Acemo- Julian Schweitzer, Shekhar Shah, William Shaw, Carlos Silva- glu, Ahmad Ahsan, Harold Alderman, Santiago Levy Algazi, Jauregui, Bachir Souhlal, David Steel, Miguel Szekely, Kaleb Caridad Araujo, Ana Arjona, Jeffrey Jensen Arnett, Tamar Tamiru, Stephanie Tam, Thaisa Tiglao, P. Zafiris Tzannatos, Manuelyan Atinc, Arup Banerji, Sergio Bautista, Kathleen Bernice Van Bronkhorst, Dominique van der Mensbrug- Beegle, Jane Bertrand, Myra L. Betron, Mohini Bhatia, Nancy ghe, Willem Van Eeghen, Tara Vishwanath, Milan Vodop- Birdsall, Andreas Blom, John D. Blomquist, Anthony Bloom, ivec, Hermann von Gersdorff, Adam Wagstaff, Mike Walker, 227 228 WORLD DEVELOPMENT REPORT 2007 Michel J. Welmond, L. Alan Winters, Katherine Whiteside, Fares, Jean, and Dhushyanth Raju. "Child Labor across the Devel- Jorge Wong-Valle, David Woollcombe, Ruslan Yemtsov, Deji oping World: Patterns, Correlations and Determinants." Young, Tanju Yurukoglu, and Hassan Zaman. Fares, Jean, and Erwin Tiongson. "Entering the Labor Market and Early Mobility of Youth: Evidence from Panel Estimates in Bos- Nazanine Atabaki, Phil Hay, and Christopher Neal assisted nia and Herzegovina." the team with consultations and media dissemination. Other Galiani, Sebastián, Martín Rossi, and Ernesto Schargrodsky. "Con- valuable production assistance was provided by Madhur scription and Crime." Arora, a former WDR program assistant, Gytis Kanchas, Gomart, Elizabeth. "Supporting Youth Citizenship: Review of Evi- Polly Means, Nacer Mohamed Megherbi, Anthony Nathan, dence." Shunalini Sarkar, and Roula I. Yazigi. Gorpe, Mehmet Ziya, Jean Fares, and Claudio E. Montenegro. Despite efforts to compile a comprehensive list, some who "Public Wage Premium Around the World." contributed may have been inadvertently omitted. The team Gutierrez, Juan Pablo. "Oportunidades for the Next Generation: apologizes for any oversights and reiterates its gratitude to Effects of a Conditional Cash Transfer on the Wellbeing of Mexi- can Youths." all who contributed to this Report. Halewood, Naomi, and Charles Kenny. "Young People and Com- munications Technologies." Background papers La Cava, Gloria, Zeynep Ozbil, Sarah Michael, Elena Galliano, Adams, Arvil V. "The Role of Skills Development in the Transition Minna Mattero, and Mio Takada. "Youth and the Transition to to Work: A Global Review." Citizenship: The Role of the State in Promoting Positive Youth Basu, Kaushik, Hyejin Ku, and Homa Zarghamee."Determinants of Religious and Ethnic Identities and Young People's Role in Peace- Youth Behaviour and Outcomes: A Review of Theory, Evidence Building and Social Change." and Policy Implications." Lam, David. "The Demography of Youth in Developing Countries Behrman, Jere R., Alexis Murphy, Agnes Quisumbing, Usha Ramak- and its Economic Implications." rishna, and Kathyrn Young. "What is the Real Impact of Educa- Llisterri, J., H. Kantis, P. Angelelli, and L. Tejerina."Youth Entrepre- tion on Age of First Parenthood and Family Formation?" neurship in Latin America." Bell, Clive, Ramona Bruhns, and Hans Gersbach."Economic Growth, Mangiaterra, Viviana, and Gerold Vollmer. "Youth Consultations Education and Aids in Kenya Model: A Long-run Analysis." for the WDR 2007: Synthesis Report of Country and Grassroots Carneiro, Pedro, and Cristobal Ridao-Cano. "Heterogeneity and Consultations." Uncertainty in Returns to High School: The Case of Indonesia." Mattero, Minna."Youth Policies and the Institutional Framework." Carneiro, Pedro, and Cristobal Ridao-Cano. "The Role of Short McKenzie, David J. "A Profile of the World's Young Developing Term Liquidity Constraints in Explaining Educational Invest- Country Migrants." ments in Indonesia." Perlman, Janice E., and Sarah E. Anthony."Citizenship and Youth in Duflo, Esther, Pascaline Dupas, Michael Kremer, and Samuel Simei. the Favelas of Rio de Janeiro." "Education and HIV/AIDS Prevention: Evidence from a Ran- domized Evaluation in Western Kenya." Background notes El-Cherkeh, Tanja, Elena Stirbu, and Andrea Tolciu. "Youth Migra- Arjona, Ana. "Understanding Recruitment in Civil Wars." tion to Europe: Potential Impact on the Labour Market of the Countries of Origin." Blum, Robert. "The Adolescent Brain." Fares, Jean, and Claudio E. Montenegro. "Youth Unemployment's Filmer, Deon, Emmanuel Jimenez, and Annette Richter."Simulating Dynamics: Evidence from Brazil (1978­2002) and Chile (1957­ the Returns to Youth Opportunity Agency and Second Chances." 2005)." Graham, Carol and Matthew Hoover. "An Exploration of Civic Fares, Jean, Claudio E. Montenegro, and Peter F. Orazem. "How Are Activity in Latin America." Youth Faring in the Labor Market? Evidence from Around the Hasan, Amer."Youth Dispositions Towards Citizenship: Do Consti- World." tutional Frameworks Matter?" Fares, Jean, Claudio E. Montenegro, and Peter F. Orazem. "Varia- Ridao-Cano, Cristobal."Vietnamese Youth: Managing Prosperity." tion in the Returns to Schooling Across and Within Developing Economies." Endnotes Overview 25. World Bank (2003f). 1. This range encompasses those who are recognized officially 26. Kabeer (1999) and Sen (1985). by the UN as "youth," those ages 15­24, as well as those many clas- 27. Vietnam Ministry of Health and General Statistics Office, sify as adolescents. The wider range is necessary to enable us to dis- UNICEF, and WHO (2005). cuss transitions from puberty to full-time work. 28. Boyer and Shafer (1997); Cáceres and others (1994); and 2. World Bank (1990) and World Bank (2001d). Eggleston and others (2000). 3. World Bank (2005j). 29. World Bank (2004a). 4. National Research Council and Institute of Medicine (2005) 30. Centre for Development and Population Activities (CEDPA) p. 73. (2001). 5. Acemoglu (2003). 31. De Ferranti and others (2003) p. 97. 6. World Bank (2005r). 32. Jacoby and Skoufias (2002). 7. Cohen and Bloom (2005). 33. Arends-Kuenning and Amin (2000). 8. World Bank (2004e). In a celebrated longitudinal study of Rio 34. Jimenez and Lockheed (1989). de Janeiro's slums (favelas), anthropologist Janice Perlman recounts 35. Akerlof and Kranton (2000). that the youth there had more education than their parents but 36. Gruber and Zinman (2001) and World Bank (2005j). without perceptibly better jobs. In the late 1960s, parents would 37. Hahn and Leavitt (2003). warn their children that if they did not stay in school they would 38. While most of this research has been done in developed end up as garbage collectors. In July 2003, the city opened competi- countries such as Germany, the experimental result is robust across tion for 400 garbage collector jobs and 12,000 people applied. A a number of settings. See Dohmen and others (2005). high school diploma was the prerequisite (Perlman (2005)). 39. Hanushek and Wößmann (2005). 9. Bloom and Canning (2005). 40. Singer (2005) and Stohl (2001). 10. By 2050, 4 of every 10 people will come from today's cohort 41. Jha and others (2001). of those ages 12­24, their children, or their grandchildren. See Lam (2006). Chapter 1 11. Bell, Devarajan, and Gersbach (2006) p. 80. 1. World Bank (2006b). 12. See World Bank (1993a). Precise estimates vary, but about a 2. World Bank (1990); World Bank (2001d); and World Bank third of the growth rate of the East Asian tigers over 1960 to 1985 (2005r). was attributed in this study to sound investments in primary edu- 3. National Research Council and Institute of Medicine (2005). cation alone. 4. It has proven difficult to convincingly establish causality 13. De Ferranti and others (2003). from education to lower fertility. However, the role of mater- 14. Bell and others (2004) p. 44. nal schooling in improving child health has been extensively 15. Behrman and others (2005a). documented. Grossman (2005a); Knowles and Behrman (2005); 16. Abeyratne (2004). Schultz (2002); and World Bank (2001c) are some recent install- 17. See box 2.1 for a fuller discussion. ments to this literature. 18. UNESCO (2004b). 5. World Bank (2001c). 19. Examples are the Trends in International Mathematics and 6. Paxson and Schady (forthcoming). Science Study and Progress in International Reading Literacy Study 7. World Bank (2005r). tests. 8. Jimenez (1995). 20. Garces, Thomas, and Currie (2000); Glewwe, Jacoby, and 9. World Bank (2004e) and World Bank (2005f). King (2001); Kagitcibasi, Sunar, and Bekman (2001). 10. De Ferranti and others (2003). 21. UNESCO (2005). 11. Estimates are from a wide range of modeling approaches 22. World Bank (2005o). including computable general equilibrium simulations and various 23. Policies to sustain growth are studied at great length in other growth models. See Bell, Devarajan, and Gersbach (2006), table 1. work and are not discussed here. For a recent review, see World 12. World Bank (2002b). Bank (2005e). 13. Barro (1999) analyzing panel data from 100 countries during 24. Bloom and Canning (2005). 1960­95 finds growth to be positively related to the starting level of 229 230 WORLD DEVELOPMENT REPORT 2007 secondary schooling. Pritchett (2001), however, finds no impact of 37. National Research Council and Institute of Medicine (2005) education growth on growth of GDP per capita or growth of total and UNICEF (2000). factor productivity using data on 90 countries over the period 1960­ 38. The definitions of child labor have evolved over time to 85. Neither Barro nor Pritchett take account of schooling quality. reflect refinement of norms and standards. In its current definition 14. Hanushek and Kimko (2000). of child labor, the ILO includes all children under age 11 who are 15. World Bank (2001b). economically active, all 12- to 14-year-olds who are economically 16. See Panton quoted in World Bank (2001b). active for more than 14 hours per week, and all those age 17 and 17. World Bank (2001b). under who are engaged in hazardous activity and the worst forms 18. For Latin America and the Caribbean see Behrman, Birdsall, of labor including trafficking, bonded labor, armed conflict, prosti- and Szekely (2003) and Sánchez-Páramo and Schady (2003); for tution and pornography, and illicit work (National Research Coun- the Middle-East, Abu-Ghaida and Connolly (2003); for Sub-Saha- cil and Institute of Medicine (2005)). ran Africa, Schultz (2003); for East Asia, Abu-Ghaida and Connolly 39. For sources, see chapter 5. (2003); for China, Heckman and Li (2004); and for India, National 40. National Research Council and Institute of Medicine (2005). Research Council and Institute of Medicine (2005). 19. Berman and Machin (2000). Spotlight on demographics 20. Acemoglu (2003). 1. Unless otherwise stated, we use the UN World Population 21. Vietnam Ministry of Health and General Statistics Office, Prospects, (United Nations (2005)) for all demographic analysis. UNICEF, and WHO (2005). See http://www.un.org/esa/population/publications/WPP2004/wpp2004 22. National Research Council and Institute of Medicine (2005). .htm for definitions, assumptions, and other details. 23. World Bank (2000). 2. For this classification we assume that a country has reached its 24. National Research Council and Institute of Medicine (2005). peak if the peak youth population occurs between 2000 and 2010 or 25. National Research Council and Institute of Medicine (2005). the peak population occurs after 2010 but the growth rate between 26. Projections are based on the United Nations (2005), and 2005 and 2015 is below 0.5 percent per year. See Lam (2006) for assume that fertility follows the medium variant. further details. 27. These projections incorporate the impact of HIV/AIDS. Pro- 3. For this classification, we assume that a country has reached jections generally suggest that AIDS will hasten the fertility decline its peak if the peak youth population occurs between 2010 and 2030 and thus hasten the peak in the youth populations. For example, in and the growth rate between 2005 and 2015 is above 0.5 percent per Botswana the youth population is projected to peak around 2005 year, or the peak youth population occurs after 2030 but the growth but under the alternative "No AIDS" projection it would have con- rate between 2025 and 2035 is below 0.5 percent per year. tinued increasing for several more decades. 28. Johnson (2000) and Lam (2006). Chapter 2 29. Lam (2006). 1. World Bank (2003f). 30. Because the analysis is based on projections from the United 2. An exception is Moehling (2005) who studies the bargaining Nations' World Population Prospects, it does not cover countries power of working children in the United States during the turn of that are missing from this data base. Examples include countries the last century. with a population less than 100,000 in 2005 and Timor Leste. 3. These ideas have been developed by several Nobel Laure- 31. Another way of thinking of the pressure of new entrants is in ates--George Akerlof, James Heckman, Daniel Kahneman, Michael terms of the growth rate of youth cohorts. This, too, peaked in most Spence, and Joseph Stiglitz--among others. developing countries in the 1960s and 1970s, at the height of the 4. For an extensive review, see chapter 7 of World Development population explosion, and has been declining since. So although it Report 2006 (World Bank (2005r)). is still high in many developing countries--at 2­3 percent a year-- 5. UNESCO (2004b), Summary, p. 35. it is on a downward trend and thus points to an easing of pressure. 6. Kurz, Peplinsky, and Johnson-Welch (1994). In the few countries where the fertility transition is not under way, 7. Strauss and Thomas (1995). growth rates are not declining. 8. Perhaps best documented are the long-term effects of early 32. There is a long literature going back to the classic work of childhood development programs in the United States. Analysis of Coale and Hoover (1958) and Leff (1969) on how countries with high the long-term effects of the national Head Start program indicates population growth rates suffered from low savings rates because of that it increases the probability of graduating from high school and the high ratio of children and youth to the working-age population. attending college. Among African Americans the program also low- 33.BloomandSachs(1998);BloomandWilliamson(1998);Bloom, ers the likelihood of having been charged or convicted of a crime Canning, and Malaney (2000); and Mason (2001). The literature (Garces, Thomas, and Currie (2000). largely focuses on Japan, Hong Kong (China), Singapore, and Korea. 9. Glewwe, Jacoby, and King (2001). Bloom and Williamson also include China and Taiwan (China). 10. Kagitcibasi, Sunar, and Bekman (2001). 34. Deaton and Paxson (1997). 11. World Bank (2005f). 35. Bloom and Canning (2005). 12. It is difficult to know how much this would cost, but a very 36. Countries where dependency ratios are rising are a subset of rough order of magnitude can be inferred from a recent costing the ones where relative cohort size is rising. They are Chad, Equato- exercise by Binder (2005), who estimated the annual global cost of rial Guinea, Guinea-Bissau, Liberia, Niger, Republic of Congo, and universalizing secondary education to be $28 billion annually over Sierra Leone. a 25-year horizon ($22 billion if efficiency gains from the experi- Endnotes 231 ence of "best practice" countries are adopted universally). If lower 47. Pitt and Khandker (1996). secondary is about half as long as upper secondary, an annual tab 48. Jimenez and Lockheed (1989). of $11­14 billion may not be far from the mark. Cohen and Bloom 49. O'Donoghue and Rabin (2001). (2005) conclude that these amounts are affordable. 50. Gruber and Zinman (2001). 13. World Bank (2003f). 51. World Bank (2005p). 14. Studies show that in most countries, the rate of human capi- 52. Bruns, Mingat, and Rakotomalala (2003). tal accumulation, measured by its effect on productivity and wages, 53. Strauss and Thomas (1995). is at its highest during youth and decreases by more than a third 54. Groce (2003). (half) by the age of 30 (40). See chapter 4. 55. Levitt and Lochner (2001). 15. UNESCO (2005) p. 174. 56. Hahn and Leavitt (2003). 16. Lee and Rhee (1999). 57. World Bank (2003f). 17. McKenzie (2004). 58. O'Donoghue and Rabin (2001). 18. Bloom and Canning (2005). 59. National Youth Council of Ireland (2001). 19. In Malaysia, labor force participation of 20- to 24-year- 60. Calculated from figures obtained from Economic Research old women increased 23 percent between 1970 and 2000. Also see and Consulting, Swiss Reinsurance Company, Zurich. Joekes (1995). 61. Knowles and Behrman (2003) pp. 39­40 show that a $1,000 20. Maloney and Nuñez Mendez (2004) and Montenegro and investment in adult basic education and literacy could produce Pagés (2004). about 10.23 trainees They compare the benefits of this as equiva- 21. Heckman and Pagés (2000). lent to one year of primary schooling (lower bound) or four years 22. UNDP (1995) estimates that two-thirds of women's work is of primary schooling (upper bound). If one were to use the unit unpaid and outside national accounts. cost of education in Bangladesh (about $31, seeWorld Bank, Unesco 23. Computed from household survey "Enquete Niveau de Vies Institute for Statistics (UIS), and OECD (2006)), a similar invest- des Menages, Institut National de la Statistique, Republique de Côte ment would yield 32.3 primary school students. Thus, the bounds d'Ivoire" (2002). are roughly 0.8 to 3 times the cost, without opportunity costs. 24. World Bank (2004e). 62. See, for example, Goldscheider (2000). 25. World Bank (2003f). 63. Span Jr. (2000). 26. This definition of "agency" is used by Kabeer (1999). The 64. Hanushek and Wößmann (2005). concept has been popular among social scientists for some time, 65. Philippine News online. but it has been given great impetus by Sen (1985). 66. Filmer, Jimenez, and Richter (2006). 27. Arnett (2000). Social psychologists have distinguished 67. Jensen (2006). two contexts. Broad socialization, characteristic of many Western industrial societies, is consistent with emphasizing independence, Spotlight on gender individualism, and self-expression. Narrow socialization empha- 1. In Muslim societies restrictions take the form of purdah, in sizes conformity to expectations (Shanahan and others (2005). others early curfews or prohibitions against traveling alone. See 28. These findings are consistent with an earlier UNICEF (2001) World Bank (2004b) and World Bank (2005a). survey of East Asian countries, not comparable with the one above. 2. Singh (1998). 29. Iversen (2002) p. 821. 3. WHO (2002b). 30. UNESCO (2005). 4. Buvinic and others (2006). 31. Basu and Foster (1998). 5. National Research Council and Institute of Medicine (2005). 32. Basu, Ku, and Zarghamee (2006). 6. Lloyd, Mensch, and Clark (2000). 33. Vietnam Ministry of Health and General Statistics Office, 7. Grown, Rao Gupta, and Kes (2005). UNICEF, and WHO (2005). 8. Ritchie, Lloyd, and Grant (2004). 34. World Bank (2006g). 9. Levison, Moe, and Knaul (2001). 35. O'Donoghue and Rabin (2001). 10. Enrollment rates for primary school have decreased for 36. Knowles and Behrman (2005). boys in 24 countries: Azerbaijan, China, Gabon, Georgia, Indone- 37. Dupas (2006). sia, the Islamic Republic of Iran, Jamaica, Jordan, Latvia, Macedo- 38. Jensen (2006). nia, Malaysia, Maldives, Mauritius, Mexico, Moldova, Myanmar, 39. Vegas and Umansky (2005). Namibia, Oman, Peru, the Slovak Republic, Sri Lanka, St. Lucia, 40. Akerlof and Kranton (2000). West Bank and Gaza, and Zimbabwe. This trend is repeated for sec- 41. See Shanahan and others (2005) for evidence from the ondary school in 12 countries: Albania, Armenia, Belarus, Georgia, United States. Kazakhstan, the Kyrgyz Republic, Moldova, Niger, Romania, Sey- 42. Ali and others (2006) p. 11. chelles, Ukraine, and Uzbekistan. Calculations based on UNESCO 43. Billari, Philipov, and Baizán Munoz (2001). Institute of Statistics database. 44. De Ferranti and others (2003). 11. The percentages of idle girls in the other countries are 10 45. Jacoby and Skoufias (2002). percent in Brazil, 23 percent in Cameroon, 27 percent in Turkey, 46. Mathur, Greene, and Jamhotra (2003). This kind of "entrap- and 28 percent in Guatemala. ment" limits social mobility of young brides by forcing them to be 12. Lavy (1996) and World Bank (2005a). apprentices for domestic labor. 13. Amin and others (1998). 232 WORLD DEVELOPMENT REPORT 2007 14. World Bank (2005a). 27. Lee and Wong (2005). 15. Buvinic (2005). 28. UNESCO (2004b). 16. Ministerio del Trabajo Peru (1998). 29. After experimenting with other nontraditional teaching methods, England implemented a structured teaching model in the Chapter 3 1990s. In addition, some of the worst performing schools in the 1. Carneiro and Heckman (2003) and Cunha and others (2005). United States have implemented it with great success. The success- 2. Knudsen (2004); Newport (2002); and Shonkoff and Phillips ful Balsakhi remedial education program in India also uses similar (2000). methods. For review see Gauthier and Dembélé (2004). 3. Experimental evidence from the United States. Heckman, 30. BRAC's Non-Formal Primary Education Schools in Ban- LaLonde, and Smith (1999). gladesh are one example of combining interactive pedagogy with 4. Currie and Thomas (1995). routine assessments of student progress, teaching methods, and 5. See Alderman, Hoddinott, and Kinsey (forthcoming); Paxson curricula that invite the input of students and teachers themselves and Schady (forthcoming); and Carneiro and Heckman (2003) for (see box 3.11). An experiment in the Philippines that provided reviews of the literature in developed and developing countries. pedagogical materials and training to teachers improved learning 6. Paxson and Schady (forthcoming). outcomes. See Tan, Lane, and Lassibille (1999). 7. Experimental studies in India and Indonesia found large 31. This can be achieved by reducing the time allocated to low impacts on cognitive development and school performance of iron priority subject areas or increasing instructional time. See UNESCO supplementation among anemic children (Soewondo, Husaini, and (2004b) for a detailed discussion of time use. Pollitt (1989) and Seshadri and Gopaldas (1989)). For programs 32.World Bank (2003b). Empirical studies in developed countries in Jamaica, Argentina, and the Philippines, see, respectively Walker show strong wage premiums associated with computer use (Autor, and others (2005); Berlinsky, Galiani, and Gertler (2006); and Katz, and Krueger (1998)). Studies that control for unobserved het- Armecina and others (2006). erogeneity find positive but smaller effects and show that what really 8. UIS-UNESCO data: http://stats.uis.unesco.org/. matters is computer skills (Pabilonia and Zoghi (forthcoming)). 9. Canals-Cerda and Ridao-Cano (2004). This effect works 33. Kuku, Orazem, and Singh (2005) use data from nine transi- mainly by increasing the probability of working during primary tion economies to show that English proficiency is positively asso- and secondary school. The effect of grade repetition is only defined ciated with computer use. Worker-firm matched data for Malay- for secondary school completion. sia show that English language proficiency is the skill that workers 10. These tests include the Anti-Social Behavior, Rosenberg Self- feel they lack the most (48 percent), followed by information and Esteem, and Rotter Locus of Control indexes. technology skills. There is a labor market premium associated with 11. Woessmann and Hanushek (forthcoming). languages of international commerce (Munshi and Rosensweig 12. Compulsory school laws have been found to increase edu- (2003)). See UNESCO (2004b) on literacy and local languages. cational attainment (Lochner and Moretti (2004)), adult earnings 34. Evidence on these programs from the United States suggests (Patrinos and Sakellariou (2005)), and health and employment that financial literacy education improves young people's financial (Oreopoulos (2005)); and to lower crime (Lochner and Moretti knowledge and behaviors, by making money go further (Varcoe (2004)) and teenage pregnancy (Black, Devereaux, and Salvanes and others (2005)). (2004)). 35. Cox (2004). 13. Knight and Sabot (1990) and Mete (2004). 36. Kemple and Scott-Clayton (2004). 14. World Bank (2005r). 37. Johanson and Adams (2004). 15. Filmer (forthcoming). 38. Chen and Kenney (2005). 16. UNESCO Institute for Statistics (2005). 39. UNESCO (2004b). 17. World Bank (2002a) and UNESCO Institute for Statistics 40. For teacher absence, see Chaudhury and others (2006). (2005). Teacher shortages are particularly a problem in mathematics, sci- 18. Most private schools in tertiary education are independent ence, and technology as well as in rural areas. World Bank (2005f). (91 percent) relative to government-dependent private schools, 41. See, for example, Jacob and Lefgren (2002). whereas in secondary education there is a fairly even split. UNESCO 42. For more on teacher training, see UNESCO (2004b) and Institute for Statistics (2005). World Bank (2005k). 19. http://www.onderwijsraad.nl/Doc/English/masterofmarket 43. See Vegas and Umansky (2005) for evidence in Latin Ameri- .pdf. can and Caribbean countries. 20. World Bank (2002a). 44. Banerjee and Duflo (2006) and Vegas and Umansky (2005). 21. UNESCO Institute for Statistics (2005). One of the challenges of implementing benficiary control pro- 22. For example, the California higher education system com- grams with parents has been low demand for education, but involv- bines selective admissions to centers of excellence with more open ing young people in holding teachers accountable has yet to be tried admissions to other two- and four-year teritiary institutions. See (except at the tertiary level), and in some cases they may have a http://www.cpec.ca.gov/. higher demand for education than their parents. 23. World Bank (2005f). 45. Lavy (2002). 24. Benavot (2004). 46. "Basti" is Bangla for shantytown. 25. World Bank (2002a) and World Bank (2005l). 47. World Bank (2003b). 26. Levy and Murnane (2004). 48. Pratham Resource Centre (2005). Endnotes 233 49. World Bank (2005k). This project also included provision of 83. Pro-Rector for Student Affairs at the Plekhanov Academy of textbooks and facilities improvements. Moscow, Professor Oleg Cherkov, March 13th, 2006 seminar. 50. See, for example, Caldwell, Levacic, and Ross (1999). 84. Main portal: www.aimhigher.ac.uk, with a parallel site for 51. Fuchs and Woessmann (2004). those ages 14­16 called "Don't Stop" (http://www.aimhigher.ac.uk/ 52. World Bank (2006d) and Reinikka and Svensson (2002). dontstop/home/). The evaluation is based on one year exposure to However, the rule-based nature of formula funding does not make it among 16-year-old individuals (Emmerson and others (2005)). A it immune to capture. Appropriate control and monitoring mecha- more comprehensive evaluation of the program is under way. nisms need to be put in place. School grants need to include a fixed 85. Carneiro and Ridao-Cano (2005). The same is being done component to account for the fixed cost of running a school. Also, for Mexico. Also see Aakvik, Salvanes, and Vaage (2003) for Nor- indivisible inputs, such as teachers, are better handled through way. However, expected returns are just one factor in determining other funding channels. education attainment (see Carneiro and Lee (2005) for the United 53. Bray (2004). States and Fleisher and others (2004) for China. 54.Woessmann (2003); Woessmann and Hanushek (forthcoming). 86. This was a randomized experiment in urban areas (Jensen 55. World Bank (2005b) and Patrinos (2005). (2006)). 56. Angrist and Lavy (2002) and Angrist, Bettinger, and Kremer 87. These costs can offset the effect of certainty on risk-averse (forthcoming). individuals, which would tend to increase educational investments. 57. Filer and Münich (2002). There is robust experimental evidence showing that individuals 58. Patrinos (2005) and Barrera Osorio (2005).. from poor family backgrounds are more risk averse than those 59. Behrman and others (2002). from better-off families (Dohmen and others (2005)). Thus, the 60. De Walque (2005). poor would tend to underinvest in education as a result of greater 61. De Fraja, Oliveira, and Zanchi (2005). uncertainty, higher risk aversion, lower aspirations, and greater 62. Sacerdote (2001) and Kremer and Levy (2003). Establish- liquidity constraints. ing the causal effect of peer groups on the behavior of individuals 88. This is because repayment of income-contingent loans is has been a challenge, particularly because peer group formation is contingent on the ex post realization of earnings. Income-contin- often endogenous. gent loans are covered later in this chapter. 63. See Lazear (2001) and Fertig (2003) for how optimal class 89. Carneiro and Heckman (2002). However, credit constraints size or composition can vary with heterogeneity. See Fryer and may be more important than suggested by the authors, for two Torelli (2005) and Akerlof and Kranton (2005b) for the effects of reasons. Individuals may be credit constrained even when their student heterogeneity at the school level. families are not but are not willing to finance their education. Also, 64. Ding and Lehrer (forthcoming). the result is conditional on a given policy environment: part of the 65. Slavin (1996) cited in UNESCO (2004b). reason for the small role of short-run credit constraints may be the 66. Boisjoly and others (2004). success of policies to address them. 67. Slavin and Cooper (1999) and Steinberg and Cauffman 90. See, for example, Lillard and Willis (1994), who find an (1996). insignificant association between income and the transition to sec- 68. Kremer, Miguel, and Thornton (2004). ondary and tertiary, and Behrman and Knowles (1999) who find 69. Akerlof and Kranton (2002). strong income effects for children ages 6­17 in Vietnam. Behrman 70. OECD (2005b). and Knowles (1999) summarize the findings of 42 studies in 21 71. Blum and Libbey (2004). countries. Estimates of the relationship between household income 72. Catalano and others (2004). and schooling are significant but small in most cases. 73. Kessler and others (1995). 91. See Jacoby and Skoufias (2002) for Mexico and Sawada and 74. de Jong and others (2001). Lokshin (2001) for Pakistan. 75. Blum and Libbey (2004). 92. The extent to which working children are able to combine 76. See both UNESCO (2004a) and National Research Council work with school depends on household demand factors (poverty), and Institute of Medicine (2005) for extensive reviews. job opportunities for children, and the institutional differences in 77. See Hoff and Pandey (2004) and Dréze and Gazdar (1997) education systems (length of the school day). on caste in India; National Research Council and Institute of Medi- 93. Levison, Moe, and Knaul (2001). cine (2005) on girls in many countries; Akerlof and Kranton (2002) 94. Beegle, Dehejia, and Gatti (2004). for several different groups in the United States. 95. Canals-Cerda and Ridao-Cano (2004). 78. Lloyd, Mensch, and Clark (2000). The study controls for a 96. Gunnarson, Orazem, and Sánchez (2006). variety of factors that typically predict drop-out including parental 97. Greene and Merrick (2005); Singh (1998); and National characteristics and typical school quality variables such as parent- Research Council and Institute of Medicine (2005). teacher ratios. 98. Eloundou-Enyegue (2004). 79. Hoff and Pandey (2004). 99. Behrman and others (2006). 80. Davies, Williams, and Yamashita (2006). In one study from 100. See Rawlings and Rubio (2005) and Morley and Coady England, 12 schools with higher participation had higher outcomes (2003) for reviews. than would be predicted by student characteristics (Hannam (2001)). 101. Skoufias and Parker (2001); Schultz (2004); and Skoufias, 81. Davies, Williams, and Yamashita (2006). Davis, and de la Vega (2001). 82. Brett and Specht (2004). 102. Todd and Wolpin (2003). 234 WORLD DEVELOPMENT REPORT 2007 103. Coady (2000). cise definition of child labor that is a subset of economically active 104. Sadoulet and de Janvry (2006) show that efficiency gains in children, depending on age and hours-of-work thresholds. the program could be achieved by selecting among the poor those 3. Duryea and others (forthcoming). children induced to go to school with the scholarship and calibrat- 4. Fares and Raju (2006). ing the size of the transfer so that it is just sufficient to induce chil- 5. Yap, Sedlacek, and Orazem (2001) and Emerson and Souza dren to go to school. (2002). 105. Skoufias and Parker (2001). A similar program in Nicara- 6. Kaushik and Tzannatos (2003) and Betcherman and others gua was also found to decrease the incidence of work among 12- (2001). to 13-year-olds (Maluccio (forthcoming)). Ravallion and Wodon 7. See Gunnarson, Orazem, and Sánchez (2006) for evidence on (2000) found that the Food for Education program in Bangladesh child labor and school performance in Latin America; Rosati and reduced child labor but the effect only accounted for 25 percent Rossi (2003) for Pakistan and Nicaragua; and Heady (2003) for of the increase in the enrollment of boys. See de Janvry and others Ghana. (2006) for shocks. 8. Ilahi, Orazem, and Sedlacek (2005), Emerson and Souza (2006). 106. Yap, Sedlacek, and Orazem (2001). 9. These estimates were obtained from a simple Mincer type 107. Arends-Kuenning and Amin (2000). earning model using data from 61 household surveys (Fares, Mon- 108. Angrist and others (2002); Angrist, Bettinger, and Kremer tenegro, and Orazem (2006b)). (forthcoming). 10. Emerson and Souza (2003). 109. The impact evaluation of the means-tested scholarship 11. Kassouf, McKee, and Mossialos (2001). program in Indonesia (Sparrow (2004)) shows no impact on upper 12. Fares, Montenegro, and Orazem (2006a). secondary school enrollment. 13. Borgarello and others (2005). The country sample includes 110. The program was discontinued, and it was not emulated Albania, Argentina, Georgia, Hungary, Poland, the Russian Federa- when Bolsa Escola was adopted as a national program. Lavinas, tion, Ukraine, and República Bolivariana de Venezuela. Barbosa, and Tourinho (2001). 14. ILO (2004). The ILO produces global and regional estimates 111. Carneiro and Heckman (2003). based on several national surveys and the ILO Global Employment 112. This is partly due to the brain development process. Differ- Trends model. ent skills have different critical stages in the learning process, and 15. Kabbani and Kothari (2005). when these are missed later remedy is not possible. 16. ILO (2005). 113. Bray (2004). 17. World Bank (2003d). 114. Lavy and Schlosser (2005). 18. O'Higgins (2003). 115. Banerjee and others (2005). 19. Fares and Montenegro (2006). 116. For a series of studies on emotional connections to school 20. Fares and Tiongson (2006). and dropout rates, see Blum and Libbey (2004); World Bank 21. Mroz and Savage (2001) and Card and Lemieux (2000b). (2006d) for the Philippines; and CRECE (2005) for Colombia. 22. Freeman (2000) and Fougère, Kramarz, and Pouget (2006). 117. Abadzi (2003). 23. Hettige, Mayer, and Salih (2004). 118. Heckman, LaLonde, and Smith (1999) and Betcherman, 24. Topel and Ward (1992). Olivas, and Dar (2004). 25. Parent (2006). 119. Heckman, Smith, and Clements (1997). 26. Hemmer and Mannel (1989). 120. Betcherman, Olivas, and Dar (2004). 27. Borgarello and others (2005). 121. Reports on youth consultations conducted for the World 28. Audas, Berde, and Dolton (2005). Development Report 2007. 29. Guarcello, Mealli, and Rosati (2003) and Parent (2006). 30. O'Higgins (2003) and Korenman and Neumark (2000). Spotlight on Vietnam 31. Fares and Montenegro (2006). 1. This spotlight is based on material in Asian Development 32. O'Higgins (2003). Also Korenman and Neumark (2000) for Bank (2005); Nguyen Anh, Duong, and Hai Van (2005); Nguyen developed countries. Anh (2005); Parliamentary Committee for Social Affairs (2005); 33. See Rosati (forthcoming). Vietnam Ministry of Health and General Statistics Office, UNICEF, 34. National Research Council and Institute of Medicine and WHO (2005); Lautrédou (2005); World Bank (2003e); and (2005). World Bank (2005q). 35. Mammen and Paxson (2000); Huffman and Orazem (2004); 2. Vietnam Ministry of Health and General Statistics Office, and Greenwood, Seshadri, and Vandenbroucke (2005). UNICEF, and WHO (2005). 36. The index includes measures of trade; fiscal, monetary, labor, and regulatory policies; state ownership; government intervention Chapter 4 in finance and capital flows; property rights; and the importance of 1. National Research Council and Institute of Medicine (2005). the gray economy. 2. ILO (2006). In this chapter child labor and economically active 37. World Bank (1995); World Bank (2004e); and World Bank children are used interchangeably. Economically active children are (2005c). defined as those who performed at least one hour of work in the 38. See spotlight on baby booms following chapter 4 and Lazear reference week during the regular school year. Work comprises paid (1983); Bentolila and Bertola (1990); and Bertola, Blau, and Kahn and unpaid work in home-owned enterprises. The ILO has a pre- (2002). Endnotes 235 39. Hopenhayn and Rogerson (1993). 78. Au and Henderson (forthcoming). 40. Pagés and Montenegro (1999) and Montenegro and Pagés 79. The 2000 Census found that nearly 12 percent of the popu- (2004). lation did not have legal residency status. 41. Heckman and Pagés (2000). Jimeno and Rodríguez-Palen- 80. State council policy paper, "Opinion of the state council on zuela (2002), using a panel of OECD countries, also found that the issue of rural migrant workers." Beijing (2006). institutional settings (including employment protection laws) that 81. Pritchett (2006). increase labor market rigidity tend to increase the youth unem- 82. Ryan (2001). ployment rate. 83. Johanson and Adams (2004). 42. Cunningham and Siga (2006) for Brazil; Montenegro and 84. van Eekelen, de Luca, and Ismail (2001). Pagés (2004) for Chile; and Neumark and Wascher (1999) for cross- 85. Adams (2006). country comparison for OECD. 86. Atchoarena and Delluc (2001). 43. Maloney and Nuñez Mendez (2004). 87. Haan and Serriere (2002). 44. Boudarbat (2005). 88. Johanson and Adams (2004). 45. World Bank (2003d). 89. Riley and Steel (1999). 46. Serneels (2004). 90. Tan (2005). 47. Gorpe, Fares, and Montenegro (2006). 91. Tan and Batra (1995). 48. Paniza (2000) shows that the public wage premium is posi- 92. Johanson and Adams (2004). tive for low-skilled and negative for high-skilled male workers, with 93. World Bank (2006e). the opposite trend for females. Filmer and Lindauer (2001) show 94. Johanson and Adams (2004) and Cinterfor/ILO (2001). that low-skilled workers in Indonesia have a higher public wage 95. Llisterri and others (2006). premium, but the differences were not statistically significant. 96. "Entrepreneurs by necessity or by opportunity" is a defini- 49. Tzannatos (1999). tion adopted by Global Entrepreneurship Monitor to differentiate 50. Ragui and Arntz (2005). the motives of people to become an entrepreneur (Reynolds and 51. O'Higgins (2003) for cross-country comparison; Fares and others (2002)). The same concepts are used here applied to those Montenegro (2006) for Brazil and Chile; Rosati (forthcoming). people that are already entrepreneurs while Global Entrepreneur- 52. All figures above come from authors' calculation using ship Monitor uses them to classify those who are engaged in creat- Investment Climate Surveys for Brazil, Indonesia, and Vietnam. ing a business. 53. Noorkôiv and others (1998). 97. The InterAmerican Development Bank database includes 54. Sánchez-Páramo and Schady (2003) . Other results from data from Argentina, Brazil, Chile, Costa Rica, El Salvador, Mexico, East Asia and Sub-Saharan Africa also show these effects. Peru, and Uruguay. A broad study comparing the emergence and 55. Japan Bank for International Cooperation (2006). development of dynamic ventures in Latin America, East Asia, and 56. Gruben and McLeod (2006). South Europe can be found in Llisterri and others (2006). 57. World Bank (2004e). 98. Kantis (2003). 58. Kugler (2004). 99. Betcherman, Olivas, and Dar (2004). 59. Cahuc and Carcillo (2006) and Blanchard (2006). 100. Elway (1999). 60. World Bank (2004e). 101. Betcherman and others (2006). This inventory of existing 61. Social Protection Note, World Bank, forthcoming, "More interventions targeting youth in the labor market would provide and better jobs in developing and emerging economies." needed future guidance on what works and what does not. 62. Neumark and Wascher (1999). 102. Ravallion and others (2005). 63. Gill, Montenegro, and Dömeland (2002). 103. Galasso, Ravallion, and Salvia (2001). 64. OECD (2005a). 104. Pound and Knight (2006). http://www.usnews.com/ 65. Kluve (2006). usnews/biztech/articles/060403/3worldbank.htm. 66. Fretwell, Benus, and O'Leary (1999) and Kluve (2006). 105. Galasso, Ravallion, and Salvia (2001). 67. Cockx and Göbel (2004). 106. UNICEF (2000), table 4.11. 68. Hazell and Haggblade (1993). 107. See http://www.jobsnet.lk/. 69. Huffman and Orazem (2004). 108. Lasida and Rodríguez (2006). 70. Haggblade, Hazell, and Reardon (forthcoming). 109. Devia (2003) and Santiago Consultores Asociados 71. Köbrich and Dirven (2006). (1999). 72. Ferreira and Lanjouw (2001) and Kijima and Lanjouw (2004). 73. Haggblade, Hazell, and Reardon (forthcoming). Spotlight on baby booms in OECD countries 74. Otsuka (forthcoming). 1. Korenman and Neumark (2000). 75. Kijima and Lanjouw (2004), Ferreira and Lanjouw (2001). 2. Ryan (2001). 76. Authors' calculation based on Probit models for employ- 3. Burgess and others (2003). ment, unemployment, and labor force participation. The models 4. Card and Lemieux (2000a). were estimated for those 15­24 years old. The specifications include 5. Blanco and Kluve (2002). education, gender indicator, and country-specific effects. The data 6. Gould, Weinberg, and Mustard (1998); Grogger (1998); and were pooled from 29 country household surveys. Machin and Meghir (2004). 77. World Bank (2005n). 7. Fougère, Kramarz, and Pouget (2006). 236 WORLD DEVELOPMENT REPORT 2007 8. Blanchard and Wolfers (2000); Nickell and Layard (1999); 34. Luke (2003). and Blau and Kahn (1999) provide comprehensive reviews of the 35. Human Rights Watch (2002b). literature on how labor market institutions affect the labor market. 36. WHO (2004). 9. Jimeno and Rodríguez-Palenzuela (2002). 37. Hibbell and others (2000). 10. Neumark and Wascher (1999) found that minimum wages 38. Singleton, Lee, and Meltzer (2001) and Institute of Alcohol had the most severe effects when imposed in combination with Studies (2006). other employment protection rules. 39. Peden and others (2000). 11. Bertola, Blau, and Kahn (2002). 40. World Bank (1999c). 12. Ryan (2001). 41. Parry and others (2000). 13. Kluve (2006) and Martin and Grubb (2001). 42. Gajakashmi and others (2000). 43. Jha and Chaloupka (2000). Chapter 5 44. Witoelar, Rukumnuaykit, and Strauss (2005). Based on anal- 1. National Research Council and Institute of Medicine (2005). ysis of Indonesia Family Life Survey, 1993­2000. 2. WHO (2002a). The survival probabilities for girls are gener- 45. National Research Council and Institute of Medicine (2005). ally higher, but the cross-country comparisons are similar. 46. United Nations Economic and Social Council (2001). 3. López and others (2006). 47. UNICEF, UNAIDS, and WHO (2002). 4. Lule and others (2005). 48. Osmond and Barker (2000); Martorell and others (1998); 5. See Mullahy and Sindelar (1991), DeSimone (2002). Ramakrishnan (2004). 6. Rodríguez Ospina, Duque Ramirez, and Rodríguez García 49. Gluckman and others (2005) and Sawaya and others (2003). (1993). 50. FAO (2006). 7. World Bank (2005d). 51. Popkin (1994); Popkin (2002); and Popkin and Gordon- 8. Hibbell and others (2000) and Bloomfield and others (2003). Larsen (2004). 9. WHO and World Bank (2005). 52. Fernald and others (2004). 10. Lundberg, Over, and Mujinja (2003) and Yamano and Jayne 53. Gertler and others (2006). (2004). 54. World Bank (1993b); Reddy (2002); Deparment of Health 11. Kalemli-Ozcan (2001); Kalemli-Ozcan (2005). Services Sri Lanka (2002); and FAO (2006). 12. Witoelar, Rukumnuaykit, and Strauss (2005). 55. Hubley (2000) and Gatawa (1995). 13. Ezzati and Lopez (2004). 56. Cohen (2004) and Gregson and others (2006). 14. Ebbeling, Pawlak, and Ludwig (2002) and Koplan, Liverman, 57. Bloom (2005a) and Blum and Ireland (2004). and Kraak (2005). 58. Kenkel (2000). 15. Slaymaker and others (2004). 59. Lansdown and others (2002). 16. World Bank (1999c). 60. Cáceres and others (1994); Eggleston and others (2000); and 17. World Bank (1993b). Boyer and Shafer (1997). 18. World Bank (1999b). 61. Knowles and Behrman (2005). 19. Russell (2000). 62. James-Traore and others (2002). 20. UNAIDS (2000). 63. Agha (2002). 21. World Bank (1999a). 64. Abaunza (2002). 22. Bateman (2001); Chandra, Jairam, and Jacob (2004); and 65. See Muirhead, Kumaranayake, and Watts (2001). Visinntini and others (1996). 66. Focus group discussion with youth in Ho Chi Minh City in 23. Montoya Díaz (2002). January 2006, Vietnam Youth Association. 24. Singh and others (2000). Trend analysis based on MEASURE 67. James-Traore and others (2002). Demographic and Health Survey data from African countries where 68. Cohen (2004). surveys were carried out between 2000 and 2004. 69. Estimates of transmission probabilities per act vary from 25. McCauley and Salter (1995). 0.0001 to 0.0014 in U.S. and European studies, 0.002 in Thailand, 26. Mensch, Singh, and Casterline (2006). and 0.0001 to 0.004 in Uganda. Higher transmission probabilities 27. Demographic and Health Surveys. (up to 0.10) have been reported among men who had contacts with 28. Clark (2004). prostitutes in Thailand and Kenya (Gray and others (2001)). 29. Konde-Lule, Sewankambo, and Morris (1997). 70. See Jha and Chaloupka (2000) for tobacco; Moore and Cook 30. National Research Council and Institute of Medicine (2005). (1995) for alcohol; Aarts, Paulussen, and Schaalma (1997) for exer- 31. Family Health International (2005). cise; and Shafii and others (2004) for condom use. 32. Dunkle and others (2004). 71. Knowles and Behrman (2005). 33. Jejeebhoy and Bott (2003). Note that researchers and young 72. James-Traore and others (2002). people may differ in their definition of transactional sex. Focus 73. Dupas (2006). group discussions among young people in Durban, South Africa, 74. Kirby, Laris, and Rolleri (2005). found that gift-giving among same-age adolescents is common in 75. De Walque (2004). sexual relationships. Adolescents view the exchange of cash for sex 76. Kirby (2001). as prostitution but do not consider noncash gifts in the same way 77. Brückner and Bearman (2005); Bearman and Brückner (Kaufman and Stavrou (2004)). (2001); and Fortenberry (2005). Endnotes 237 78. WHO (2003b). 128. Pauly (1968). 79. Evidence of the impact of sport participation on empower- 129. Population Council (2002). ment and criminal or delinquent activity among youth is extremely 130. Rowlands and others (2000). weak Coakley (2002). See chapter 7 of this Report. 131. See Over and others (2004) and the papers cited therein. 80. Brady (1998). 132. Marseille (2003). 81. Smith, Bogin, and Bishai (2005). 133. Jha and others (2001). 82. Erulkar and others (2004). 134. Dupas (2006). 83. Student Partnership Worldwide (2005) p. 2. 84. Cáceres and others (1994); Eggleston and others (2000); and Spotlight on Brazil Boyer and Shafer (1997). 1. All figures in bullets come from World Bank (2006i). The 85. Eggleston, Leitch, and Jackson (2000). country's high levels of inequality, based on Gini coefficients, is sec- 86. James-Traore and others (2002). ond only to South Africa. 87. Alford, Cheetham, and Hauser (2005). 2. Bonelli and Veiga (2004). 88. Mensch, Hewett, and Erulkar (2001); Erulkar and Mensch 3. World Bank (2006i). (1997); and James-Traore and others (2002). 4. Bourguignon, Ferreira, and Menendez (2005). 89. Berer (2003). 5. Verner, Blom, and Holm-Nielsen (2001). 90. Dupas (2006). 6. Some states paid per child, others paid if all children went to 91. Gutierrez (2006). school, to prevent parent's choosing between children. World Bank 92. World Bank (1999b). (2001a). 93. Witoelar, Rukumnuaykit, and Strauss (2005). 7. By merging cash-transfer programs that promoted schooling, 94. Grossman (2005b). health, food consumption, and compensation for adjustment, the 95. World Bank (1999b). government seeks to increase the efficiency and effectiveness of the 96. See Townsend, Roderick, and Cooper (1994) for evidence many transfer programs. from the United Kingdom. 8. World Bank (2006i). 97. Rehm (2003) and World Bank (1999b). 9. The program has also dramatically reduced the rates of suicide, 98. Nelson (2003) and Fichtenberg and Glantz (2002). theft, drug use, sexual aggression, and armed robbery by students. 99. World Bank (1999b). 10. A program analysis conducted in 2002 compared schools 100. World Bank (1999b). under the Programa Abrindo Espaços and those that were not and 101. Jha and others (2005). found that schools that participated in the program had a lower 102. World Bank (1999b) and Fiore and others (2000). violence index (sum of violent acts, weighted by their severity) than 103. World Bank (1999b). those that did not, by 16 percent in Rio and by 14 percent in Per- 104. UNDCP (2003). nambuco (Waiselfisz and Maciel (2003)). 105. Drucker and others (1998) and WHO (2005a). 11. World Bank (2006a). 106. Hurley, Jolley, and Kaldor (1997) and MacDonald and oth- 12. The target population is men 14- to 25-years-old and the ers (2003). mean age of participants in the three evaluation sites, Bangu, Maré, 107. UNICEF, UNAIDS, and WHO (2002). and Morro dos Macacos, was 17. 108. See, for example, Grosskurth and others (1995). 13. The scale is a composite of qualitative questions that reflect 109. Okonofua and others (2003). whether a respondent disagrees with "traditional" gender norms, 110. Stanback and Twum-Baah (2001). such as, "there are times that a woman deserves to be beaten." 111. Dickson-Tetteh, Pettifor, and Moleko (2001). 14. Barker (2003). 112. Temin and others (1999). 113. Knowles and Behrman (2005). Chapter 6 114. Thailand Ministry of Public Health and World Bank 1. National Research Council and Institute of Medicine (2005). (2005). 2. In this chapter, the term "marriage" captures actual marriages 115. Salomon and others (2005). and consensual unions. 116. Over and others (2004). 3. National Research Council and Institute of Medicine (2005). 117. Health and Treasury Task Team (2003). Calculations of first birth intervals for ever-married girls' (ages 20­ 118. Teixeira, Vitória, and Barcarolo (2003). 24) using birth histories and age at marriage drawn from Demo- 119. There is evidence that continuing drug use is a behavioral graphic and Health Surveys conducted between 1998 and 2004. For deterrent to ART adherence (Lucas and others (2001)). these countries, the average first birth interval was 1.5 years. 120. UNAIDS and UNODCCP (2000). 4. Assaad and Zouari (2003). 121. Celentano and others (2001). 5. Zabin and Kiragu (1998); Study Group on Female Genital 122. Shah and Ahman (2004a) and Shah and Ahman (2004b). Mutilation and Obstetric Outcome (2006). 123. WHO (1998). 6. Conde-Aguledo, Rosas-Bermúdez, and Kafury-Goeta (2006); 124. Ellertson and others (1995). Hediger and others (1997); Makinson (1985); Miller (1991); Scholl 125. Aziken, Okonta, and Ande (2003). and others (1989); Scholl and others (1992); and Zabin and Kiragu 126. World Bank (1993b). (1998). 127. Berer (2003). 7. Gertler, Levine, and Martinez (2003). 238 WORLD DEVELOPMENT REPORT 2007 8. Bhargava (2005). 53. King (2005). 9. National Research Council and Institute of Medicine (2005). 54. Dupas (2006). 10. International Institute for Population Sciences (IIPS) and 55. Reported in Alford, Cheetham, and Hauser (2005). ORC Macro (2000). 56. FOCUS (1998) and Graft, Haberland, and Goldberg (2003). 11. Based on 2004 round of Demographic and Health Survey 57. Nanda, Switlick, and Lule (2005). The SIAGA Campaigns for Chad. Indonesia Web site, http://www.comminit.com/experiences/ 12. Jowett (2000) and Berman and Rose (1996). pds62004/experiences-1983.html. JHPIEGO News Release (May 13. Fafchamps and Quisumbing (2005). 5, 2004) Indonesian mothers surviving childbirth more often with 14. Lundberg and Rose (2002). support from husbands, community, faith-based groups, http:// 15. Galloway and Anderson (1994). www.jhpiego.org/media/releases/nr20040505.htm. 16. Alderman and Behrman (2006). 58. Delisle, Chandra-Mouli, and de Benoist (2000). 17. Strauss and Thomas (1995) and Behrman and Deolalikar 59. White (2005). (1988). 60. Loevinsohn (1990). 18. Clark (2004). 61. SCN News (2006). 19. Galloway and Anderson (1994). 62. Grantham-Mcgregor and others (1991) and Paxson and 20. Centers for Disease Control and Prevention (1998) and Yip Schady (forthcoming). and Dallman (1996). 63. Hallman and others (2005). 21. Tests using Hemocue. 64. Ruel and others (2002). 22. El-Zanaty and Way (2001). 65. Committee for Population and ORC Macro (2003). 23. International Nutritional Anemia Consultative Group (1989). 66. Lokshin, Glinskaya, and Garcia (2004). 24. Li and others (1994). 67. Nanda (1999). 25. Oyediran, Ishola, and Adewuyi (2002). 68. Centre for Development and Population Activities (CEDPA) 26. Mensch, Bruce, and Greene (1998). (2001). 27. See BPS-Statistics Indonesia and ORC Macro (2006) for a 69. Mathur, Mehta, and Malhotra (2004). description of the 2002­03 Indonesia Young Adult Reproductive 70. Institute for Health Management-Pachod (IHMP) and Health Survey. International Center for Research On Women (ICRW) (2003). 28. Siddiqua and Kabir (2002). 71. Arends-Kuenning and Amin (2000). 29. Senderowitz (1995). 72. Amin and others (1998). 30. Akin and others (1984); Hodgkin (1996); Nanda (1999); and 73. Singh and Darroch (2000). Wong and others (1987). 74. Geronimus and Bound (1990); Geronimus and Korenman 31. Khan (1998); Mumtaz and Salway (2005); and World Bank (1992); Hoffman (1998); and Klepinger, Lundberg, and Plotnick (2005a). (1997). 32. National Research Council and Institute of Medicine (2005). 75. Hofferth and Reid (2001). 33. Based on data from Measure DHS surveys in which ques- 76. Kaufman, de Wet, and Stadler (2001) and National Research tions were asked about components of antenatal care. Council and Institute of Medicine (2005). 34. Calculated using data from the Kenya Demographic and 77. Program description based on WHO (2003a). Health Survey, 2003. 78. WHO (2003c). 35. Senderowitz and Paxman (1985). 36. Senderowitz (1999). Chapter 7 37. World Bank (2005a). 1. Authors' interviews with student volunteers. For more infor- 38. Joshi and Schultz (2005). mation, see www.risepak.com. 39. World Bank (2005m). 2. Walzer (1989). 40. See Rosen (2000). 3. Marshall (1950). 41. See National Research Council and Institute of Medicine 4. Mamdani (1996). (2005). 5. Benhabib (2005) and Bauböck (2005). 42. Berer (2003). 6. Kymlicka (2001). 43. See Caldwell and Caldwell (2002) and Askew and Maggwa 7. Harrington (2005). (2002). 8. Hooghe, Stolle, and Stouthuysen (2004). 44. World Bank (2006f). 9. Jennings and Stoker (2002). 45. Knowles and Behrman (2003). 10. Wattenberg (2006). 46. Delisle, Chandra-Mouli, and de Benoist (2000) and World 11. Norris (2005). Bank (2005m). 12. Miguel, Glennerster, and Whiteside (2006). 47. Partnership for Child Development (2006). 13. Schmitter and Treschel (2005). 48. Delisle, Chandra-Mouli, and de Benoist (2000). 14. Stolle and Hooghe (2004). 49. Gertler (2000) and Skoufias (2001). 15. Macedo and others (2005) p. 48. 50. Rawlings and Rubio (2005). 16. Gerber, Green, and Shachar (2003). 51. Behrman and others (2005b). 17. Plutzer (2002). 52. Lloyd and Grant (2004). 18. Firebaugh and Chen (1995). Endnotes 239 19. Alwain and Krosnick (1991); Jennings and Stoker (2002); 49. Navarro and Hasan (2003). and McAdam (1988). 50. Verba, Schlozman, and Brady (1995). 20. Chattopadhyay and Duflo (2003); Dhakal and Misbah 51. Zeldin and others (2000). (1997); and Pande (2005). 52. Matthews (2003). 21. Stern, Dethier, and Rogers (2005). 53. Rajbhandary, Hart, and Khatiwada (2001). 22. Besley and others (2004); Jimenez and Sawada (1999); Nara- 54. In Thorup and Kinkade (2005) p. 77. yan (1995); and Venkatraman and Falconer (1998). 55. Edmunds, Forster, and Cottey (2002); Hirsch and Mehay 23. Przeworski and others (2000). (2003); and WDR 2007 InterMedia surveys. 24. Acemoglu, Johnson, and Robinson (2001). 56. Although national service is ostensibly universal in Scandi- 25. Isham, Kaufmann, and Pritchett (1995); Lindert (2003); navian countries, budget cuts have meant that less than one-third Rivera-Batiz (2002); and Sen (1999). of the eligible population actually serves. 26. Arunatilake, Jayasuriya, and Kelegama (2001); Barrera and 57. Angrist (1990); Galiani, Rossi, and Schargrodsky (2006); Ibáñez (2004); and Kutan and Drakos (2003). Imbens and Van Der Klaauw (1995); Lokshin and Yemtsov (2005). 27. Ehtesaab used by permission of Salman Ahmad. To learn 58. In a study of women who served in the U.S. military dur- more about Junoon, visit http://www.junoon.com. ing the Vietnam era, 30 percent reported having been raped, and 28. Fearon and Laitin (2003) and Hegre (2003). 35 percent reported being otherwise physically assaulted (Sadler, 29. Steinberger (2001). Booth, and Doebbeling (2005)). Another survey of more than 1,500 30. Stockard and O'Brien (2002) define relative cohort size as female veterans applying for posttraumatic stress disorder disabil- the ratio of the size of the younger generation (ages 15­29) to the ity benefits found that 71 percent had experienced some form of size of the older generation (ages 30­65). A cohort effect is distinct sexual assault while in service (Murdoch and others (2004)). from age or period effects in that it reflects influences unique to a 59. Bedard and Deschênes (2006). particular group of people, such as those born between 1950 and 60. Abt Associates (2004); Gal and Eberly (2006); Obadare 1954 or young men who came of age between 1940 and 1944. (2005); and Viva Rio (2005). 31. A brief overview of the Easterlin hypothesis in Pampel and 61. Personal communication with Susan E. Stroud, Executive Peters (1995), pp. 164­169. Director, Innovations in Civic Participation, June 2006. 32.Stockard and O'Brien (2002) find that birth cohorts that are rel- 62. Erikson (1994) p. 245. atively less socially integrated and regulated have higher suicide rates. 63. Akerlof and Kranton (2005a) pp. 10­11; Cole (2004b) p. 896; However, collective institutions such as those that support families and Pals and Tuma (2004); and Titma and Tuma (2005). children can moderate the effect. Jacobson (2004) finds evidence that 64. Erikson (1994); Roy (2004); and Sciolino, Wingfield, and larger cohorts increase marijuana use by lowering the risk of a sales Povoledo (2005) p. A8. arrest and by generating informational economies. Jacobson (2004) 65. Neuhaus (2005). and Levitt (1999) find little evidence of the effect of relative cohort 66. Kriger (2005) and Moscow Human Rights Bureau (2005). size on murder, violent crime, and property crime rates. The effects of 67. UNICEF (2005c). cohort size and educational and labor markets are analyzed elsewhere 68. Duryea, Olgiati, and Stone (2006). in this report. 69. Solinger (1999). 33. See Cincotta, Engleman, and Anastasion (2003) on these 70. Balcells i Ventura (2005); Cogan, Morris, and Print (2002) demographic stress factors and how they interact. Mesquida and p. 6; and Kymlicka (2001). Wiener (1999) point out that while a high proportion of young 71. Torney-Purta and others (2001). males is a necessary condition for the emergence of violent conflict, 72. Hahn (2005) p. 835. it is not a sufficient one. 73. Finkel and Strumbas (2000). 34. Cincotta, Engleman, and Anastasion (2003) p. 77. 74. Finkel and Strumbas (2000) p. 105. 35. For an exception, see Hudson and den Boer (2004) on China 75. Hahn, Dilworth, and Hughes (1998) in Gibson (2001) p. 7. and India. 76. Billig (2006). 36. Mannheim (1972) p. 294, as cited in Cole (2004b). 77. Dalrymple (2005). 37. Cole (2004a). 78. Bay and Blekesaune (2002); Durham (forthcoming); 38.Acemoglu,Johnson,and Robinson (2001);Gauri and Lieberman Fougère, Kramarz, and Pouget (2006); and Reiss and Roth (1993). (2004); Mamdani (1996); and Steinmo, Thelen, and Longstreth (1992). 79. National Research Council and Institute of Medicine (2005) 39. Chiclet (2001). p. 405. 40. La Cava and others (2006). 80. Furlong and others (1997). 41. Diamond (2003), citing Freedom House data. 81. Hahn, Leavitt, and Aaron (1994). 42. Rodríguez-Pose and Gill (2003). 82. Elliot and Tolan (2005). 43. For country-specific youth voter turnout rates, see Pintor 83. The Office of Juvenile Justice and Delinquency Prevention and Gratschew (2002) and Franklin (2004). (OJJDP) (1998). 44. Franklin (2004). 84. Maclure and Sotelo (2004). 45. Wattenberg (2006). 85. Levine (1999). 46. Wintour (2006). 86. Thornberry (1998). 47. Franklin (2004). 87. Rubio (2005). 48. Green and Gerber (2001). 88. Dowdney (2005). 240 WORLD DEVELOPMENT REPORT 2007 89. Smith (2004). 3. Government of Sierra Leone (2002), p 1. Cited in Ginifer 90. Shaw (2001). (2003). Based on Sierra Leone Integrated Household Survey, 2004. 91. Decker and Van Winkle (1996). 4. Women's Commission for Refugee Women and Children 92. Howell (1998); Thornberry (1998); and UNICEF (2005a). (2002). 93. Sherman and others (1998) and Rodgers (1999). For exam- 5. Based on Sierra Leone Integrated Household Survey, 2004 ple, Krug and others (2002). 6. Glennerster, Imran, and Whiteside (2006). 94. Authors' observations from the Ecuador Law and Justice for 7. World Bank (forthcoming) Education Sector Review. the Poor Program. 8. Simon (2003). 95. Mocan and Rees (1999). 9. ENCISS and World Bank (2006). 96. Grogger (1998). 10. Goovaerts, Gasser, and Belman Inbal (2005). 97. Krug and others (2002); Llorente and Rivas (2005); Sheley 11. Turniski (2004). and Wright (1993); and Villaveces and others (2000). 12. Miguel, Glennerster, and Whiteside (2006). 98. Rashid (2000); Sageman (2004); and Taylor (1988). 13. Richards, Bah, and Vincent (2004). 99. Berrebi (2003); Post (1998); and Sageman (2004). 14. ENCISS and World Bank (2006). 100. Berman and Laitin (2005); Bloom (2005b); Levine (1999) 15. Bellows and Miguel (2006). p. 342; Paxson (2002); Post (1998); Sageman (2004); Sprinzak 16. Hart (2004). (1998); and Stern (2004). 17. Bannon, Holland, and Rahim (2005). 101. Erikson (1994) p. 196. 102. Human Rights Watch (1997a); Human Rights Watch Chapter 8 (2003); and UNICEF (2005a), Innocenti Brief 3e. 1. Population Division of the Department of Economic and 103. Greenwood and others (1998). Social Affairs of the United Nations Secretariat, Trends in Total 104. Matthews, Griggs, and Caine (1999). Migrant Stock: The 2005 Revision http://esa.un.org/migration, 105. Human Rights Watch (2001a). April 7, 2006. 106. Akpokodje, Bowles, and Tigere (2002); Chen and Shapiro 2. World Bank staff calculations from Global Trade Analysis (2004); and Levitt (1998). Project (GTAP) database of Parsons and others (2005). 107. Bayer, Pintoff, and Pozen (2005) and Woolard and others 3. The United Nations High Commissioner for Refugees (2005). (UNHCR (2005)) reports that 34 percent of refugees are ages 5 to 108. WHO (2005b). 17 and 47 percent 18 to 59. In the United Kingdom, 15- to 24-year- 109. Beyer (1997). olds make up 32 percent of female asylum seekers and 38 percent of 110. Department of Corrections (2003). male asylum seekers (Heath and Jeffries (2005)). 111. Petrosino, Turpin-Petrosino, and Buehler (2005). 4. Sjaastad (1962). 112. Wilson, Mackenzie, and Mitchell (2005). 5. World Bank staff calculations from the 2000 United States 113. Van Ness (2005). Census public use sample. 114. Miers (2001). 6. Massey and others (1998) p. 47. 115. Latimer, Dowden, and Muise (2001); Akpokodje, Bowles, 7. See McKenzie, Gibson, and Stillman (2006) for recent evidence and Tigere (2002); and Roche (2006). of how large the gain in income can be, based on a migration lottery. 116. Brandt (2005). 8. See Dustmann (2001) for a review of these older studies and 117. AFP (2001); Arjona (2006); BBC (2001); Brown (1990); Reyes (1997) for Mexico. Coalition to Stop the Use of Child Soldiers (2000) in Singer (2005); 9. World Bank (2005i). Coalition to Stop the Use of Child Soldiers (2002); Coalition to 10. World Bank (2005i). Stop the Use of Child Soldiers (2003); Coalition to Stop the Use 11. Black (2003). of Child Soldiers (2004); Human Rights Watch (2002a); Leopold 12. Hagopian and others (2005). (2001); McGirk (2001); Seyboldt (2000); and Singer (2005). 13. McKenzie (2006a). 118. Arjona (2006); Humphreys and Richards (2005); Hum- 14. Mattoo, Neagu, and özden (2005). phreys and Weinstein (2003); and Singer (2005). 15. Kleiner (2000). 119. Human Rights Watch (1997b); ILO (2003); World Bank 16. Mishra (forthcoming). (2005t). 17. World Bank Staff calculations from the Pacific Island­New 120. Humphreys and Richards (2005). Zealand Migration Survey 2005. 121. Humphreys and Richards (2005). 18. Yang (2004) for the Philippines and Cox Edwards and Ureta 122. Benus, Rude, and Patrabansh (2001). (2003) for El Salvador. 123. Humphreys and Weinstein (2005) and Utas (2004). 19. World Bank Staff calculations from OECD (2005c) and 124. Kofi Annan in a speech to the 1st World Conference of OECD (2003). Ministers Responsible for Youth, Lisbon, Portugal, 1998. 20. Davis (2003). 21. Lucas (2004). Spotlight on Sierra Leone 22. Kandel and Kao (2001). 1. Government of Sierra Leone (2005a). Government of Sierra 23. McKenzie and Rapoport (2006). Leone (2005b). 24. International Organization for Migration (IOM) (2005b); 2. The Global Fund for AIDS (2005); and UNICEF (2005b). Simonet (2004); and Survelliance Unit of the NSACP, Sri Lanka. Endnotes 241 Note that in these countries migrants are more likely to be tested substantial return, helped in large part by their booming economies for HIV/AIDS than nonmigrants, as part of the health tests needed (International Organization for Migration (IOM) (2005b)). for employment abroad. As a result, their share of diagnosed cases 52. El-Cherkeh, Stirbu, and Tolciu (2006). Romania does not probably exceeds their share of all cases. have a mutually agreed convention with the United States or Ger- 25. International Organization for Migration (IOM) (2005a). many as it does with several other countries, but students can have 26. Hildebrandt and McKenzie (2005). their diplomas recognized through a special office. 27. Stepick, Stepick, and Kretsedemas (2001). 53. Narayan and Petesch (2006). 28. Martinez and Lee (2000). 54. World Bank staff calculations from Spain 2001 Census pub- 29. Australia Immigration Visa Services (1998). lic use sample. 30. Fidrmuc and Doyle (2005) include measures of inequal- 55. McKenzie (2005) provides the analysis cited in this paragraph. ity and regional controls to try to isolate the political socialization 56. Average over 1993­2000 based on the Survey on Overseas effect of migration from the self-selection effect. Filipinos of the National Statistics Office of the Philippines. Data 31. Donato and Patterson (2004). was kindly supplied by Dean Yang for this purpose. 32. See http://sandiego.indymedia.org/en/2005/10/111331.shtml 57. Gebrekristos and others (2005). and the database of deaths at the Arizona border provided by the 58. Enlaces Centro de Educación y Tecnología (2005) and Hale- Coalición de Derechos Humanos at http://www.derechoshumano- wood and Kenny (2006). saz.net/deaths.php4 [both accessed November 19, 2005]. 59. Mar Gadio (2001). 33. "Trafficking in persons" shall mean the recruitment, trans- 60. Internet World Statistics, http://www.internetworldstats. portation, transfer, harbouring or receipt of persons, by means of com/stats.htm, November 21, 2005, update [accessed December 5, the threat or use of force or other forms of coercion, of abduction, 2005]. of fraud, of deception, of the abuse of power or of a position of vul- 61. Dodds, Muhamad, and Watts (2003). nerability or of the giving or receiving of payments or benefits to 62. Halewood and Kenny (2006). achieve the consent of a person having control over another person, 63. McKinsey Global Institute (2005). for the purpose of exploitation (United Nations, Palermo Protocol 64. Special tabulations for youth provided by Guo Liang. See 2000) Liang (2005) for full details of the survey. 34. U.S. Department of State (2005a). 65. Curtain (2001). 35. Omelaniuk (2005) for the IOM data; Clert and others (2005) 66. Samuel, Shah, and Hadingham (2005). for the study in Southeastern Europe. 67. Mungai (2005). 36. U.S.Department of State (2005b). 68. ILO (2001). 37. Holzmann (2005), table A.2., based on UN medium variant 69. http://www.digitaldividedata.com. [accessed December 3, projections. 2005]. 38. Hatton and Williamson (2005). 70. Instituto Nacional para la Evaluación de la Educación de 39. Hatton and Williamson (2003). Mexico (2005). 40. Longhi, Nijkamp, and Poot (2005). 71. Perkinson (2005). 41. Borjas (2003) finds a 0.3­0.4 reduction in wages from a 1 72. World Links Impact Evaluation Series (2002). percent increase in immigration, while the more recent studies by 73. Halewood and Kenny (2006). The discretionary budget covers Card (2005) and Dustmann, Fabbri, and Preston (2005) find less all costs apart from teacher salaries, including supplies, teaching equip- effect. ment, utility bills, building maintenance, and other classroom needs. 42. Angrist and Kugler (2003). 74. Neto and others (2005). 43. United Kingdom Home Office (2005). 75. Banerjee and others (2005). 44. Portes and French (2005). 76. Gough and Grezo (2005). 45. However, in March 2006 the United Kingdom home office 77. Online Computer Library Center (OCLC) (2005). announced plans to eliminate these programs as part of the intro- 78. Geary and others (2005). duction of a points system. 79. Mar Gadio (2001). 46. Maclellan and Mares (2005) nicely summarize this program 80. Chewning and others (1996). and the lessons for other countries. 81. Parham (2004). 47. Data calculated by Anna Paulson from the Survey of Income 82. See Jorgenson (forthcoming) for a recent review. and Program Participation (SIPP) 1996­2000 for immigrants who 83. See World Bank (2006c) for an assessment of the growth came to the United States between 1990 and 1996. impact. 48. See World Bank (2005i) for a thorough discussion and 84. Clarke and Wallsten (2004). examples. 85. Jensen (2006). 49. Gibson, McKenzie, and Rohorua (2006). 86. Neto and others (2005). 50. See Mesnard (2004) for Tunisia; Yang (2005) for the Philip- 87. Guermazi and Satola (2005). pines. 88. Rao (1999). 51. In Malaysia, a scheme targeting the 250,000 skilled workers 89. http://www.wordtracker.com unfiltered list of top 500 search overseas with tax exemptions and other incentives to return only terms for December 6, 2005. led to the return of 104 expatriates in the first two years of opera- 90. Note that although these point systems typically give more tion; whereas South Korea and Taiwan (China) have seen more points to young migrants than to older migrants, the other skill 242 WORLD DEVELOPMENT REPORT 2007 criteria needed to obtain points typically disadvantage youth. As a 10. It should be borne in mind that some youth movements may result, most developing country youth are not provided an oppor- not articulate strong demand for policies supporting youth because tunity to migrate under the existing points systems. it is a transitory phase. By the time such policies would be imple- mented, many of the leaders of these movements would no lon- Spotlight on donors ger be young. This lack of a permanent interest group can weaken 1. The most significant international commitments for youth political support for youth-focused policy. include the 1989 UN Convention on the Rights of the Child, the 11. German Technical Cooperation and German Technical 1999 ILO Convention on Child Labor, and the 2000 UN Program Cooperation and International Council on National Youth Policy of Action on Youth updated in 2005. In addition, there are regional (2005). This is not to detract from the fact that many regional orga- commitments such as that of the Council of Europe and the Ibero- nizations play an important role promoting dialogue across coun- American Convention on Youth Rights. tries on issues of common regional concern, including issues with 2. In Timor Leste, in the first two years after conflict about half implications beyond a member country's immediate borders. of government spending on education was financed by external 12. Instituto Mexicano de la Juventud (IMJ) and Organización sources, allowing many children and young teenagers to attend Iberoamericana de la Juventud (OIJ) (2006). school. See World Bank (2004d). 13. German Technical Cooperation and German Technical Cooperation and International Council on National Youth Policy Chapter 9 (2005). As mentioned previously, about two-thirds of countries 1. Instituto Mexicano de la Juventud (IMJ) and Organización with national youth councils do not give voice to youth NGOs. Iberoamericana de la Juventud (OIJ) (2006). 14. Dulci (2005). 2.Although the stage of the demographic transition is correlated 15. Clementino (2006). with income, there is enormous diversity among both low- and 16. World Bank (2006g). middle-income countries. Chad and Tajikistan are at very similar 17. For a list of all indicators on youth that are collected by the levels of income per capita, but in Chad fertility is approximately UN system, see http://www.un.org/esa/socdev/unyin/documents/ 6.0 births per woman, while in Tajikistan it is 3.5. youthindicatorsexist.pdf. Also, for the status of discussions on 3. The discussion in this paragraph is based on special surveys which indicators should be used as a part of a worldwide effort to commissioned for this Report from InterMedia (WDR 2007 Inter- monitor youth outcomes, see http://www.un.org/esa/socdev/unyin/ Media surveys). See methodological note at the beginning of the documents/youthindicatorsreport.pdf. Report. Also, figure 2.4. 18. For example, questions on the citizenship block could be 4. Washington (2006) demonstrates how having female children obtained from more detailed criminal justice records, the inclusion influences legislators' voting records on women's issues in the United of questions on legal identity and citizenship in household surveys States, making them significantly more in favor of women's rights. and censuses, and the incorporation of modules on political and 5. German Technical Cooperation and German Technical Coop- social participation and knowledge in existing surveys. eration and International Council on National Youth Policy (2005). 19. See, for example, Summers (1992); Summers (1994); and 6. World Bank (2003f). Van der Gaag and Tan (1998). 7. In addition to this Report, see National Research Council and 20. Knowles and Behrman (2005). Institute of Medicine (2005), and Knowles and Behrman (2005) 21. Discriminant analysis was used on census data in the initial for a discussion of this point. There is a large developed-country stages to identify target communities and households within those literature; however, the very different circumstances of developed- communities. See Knowles and Behrman (2005). country youth and differences in capacity for implementation limit the applicability of these studies. Spotlight on youth action 8. See, for example, Vegas and Umansky (2005) on school 1. Two such sources are "Youth and the Millennium Develop- management; Banerjee and others (2005) on remedial education; ment Goals" (Ad Hoc Working Group for Youth and the MDGs Miguel and Kremer (2004) on deworming; and Glewwe, Jacoby, (2005)) and Kinkade and Macy (2005). and King (2001) on early childhood development. 2. Peters (2006). 9. German Technical Cooperation and German Technical Coop- 3. Ogar (2005). eration and International Council on National Youth Policy (2005). 4. Ekehaug (2005). References The word "processed" describes informally reproduced works that may not be commonly available through libraries. Aakvik, Arild, Kjell G. Salvanes, and Kjell Vaage. 2003."Measur- Aedo, Cristián, and Marcelo Pizarro Valdivia. 2004."Rentabilidad ing Heterogeneity in the Returns to Education in Norway Using Económica del Programa de Capacitación Laboral de Jóvenes Education Reforms." Bonn: Institute for the Study of Labor `Chile Joven'." INACAP and Mideplan. Santiago de Chile. (IZA), Discussion Paper Series 815. Processed. Aarts, Henk, Theo Paulussen, and Herman Schaalma. 1997. AFP. 2001."Amman Conference to Seek Ban on Use of Child Sol- "Physical Exercise Habit: On the Conceptualization and Forma- diers in Region, World." AFP, April 7. tion of Habitual Health Behaviours." Health Education Research Agha, Sohail. 2002."A Quasi-experimental Study to Assess the 12(3):363­74. Impact of Four Adolescent Sexual Health Interventions in Abadzi, Helen. 2003. Improving Adult Literacy Outcomes: Lessons Sub-Saharan Africa." International Family Planning Perspectives from Cognitive Research for Developing Countries. Washington, 28(2):67-70­& 113-118. DC: World Bank. Akerlof, George A., and Rachel E. Kranton. 2000."Economics and Abaunza, Humberto. 2002."Puntos de Encuentro: Communica- Identity." Quarterly Journal of Economics 115(3):715­53. tion for Development in Nicaragua." Sexual Health Exchange ------. 2002."Identity and Schooling: Some Lessons for the 2002(1):2­3. Economics of Education." Journal of Economic Literature Abeyratne, Sirimal. 2004."Economic Roots of Political Conflict: 40(4):1167­201. The Case of Sri Lanka." World Economy 27(8):1295­314. ------. 2005a."Identity and the Economics of Organizations." Abt Associates. 2004. Serving Country and Community: A Longitu- Journal of Economic Perspectives 19(1):9­32. dinal Study of Service in AmeriCorps. Washington, DC: Corpora- ------. 2005b."Social Divisions within Schools: How School Poli- tion for National and Community Service. cies Can Affect Students' Identities and Educational Choices." Abu-Ghaida, Dina, and Marie Connolly. 2003. Trends in Relative In Christopher B. Barrett, (eds.), The Social Economics of Pov- Demand for Workers with Secondary Education: A Look at Nine erty: On Identities, Groups, Communities and Networks. London: Countries in East Asia, Africa and MENA. Washington, DC: Routledge. Background Paper prepared for Expanding Opportunities and Akin, John S., Charles C. Griffen, David K. Guilkey, and Barry M. Building Competencies for Young People: A New Agenda for Sec- Popkin. 1984. The Demand for Primary Health Services in the ondary Education, World Bank. Third World. Totowa, NJ: Littlefield Adams. Acemoglu, Daron. 2003."Patterns of Skill Premia." Review of Eco- Akpokodje, Joseph, Roger Bowles, and Emmanuel Tigere. 2002. nomic Studies 70(2):199­230. Evidence-based Approaches to Crime Prevention in Devel- Acemoglu, Daron, Simon Johnson, and James A. Robinson. oping Countries - A Scoping Review of the Literature. York, 2001."The Colonial Origins of Comparative Development: United Kingdom: Centre for Criminal Justice Economics and An Empirical Investigation." American Economic Review Psychology. 91(5):1369­401. Alderman, Harold, and Jere R. Behrman. 2006."Reducing the Ackerman, Susan. 1996."Rebellion and Autonomy in Industrial- Incidence of Low Birth Weight in Low-Income Countries Has izing Penang: The Career History of a Young Malay Divorcee." Substantial Economic Benefits." World Bank Research Observer Southeast Asian Journal of Social Science 24(1):52­63. 21(1):25­48. Ad Hoc Working Group for Youth and the MDGs. 2005."Youth Alderman, Harold, John Hoddinott, and Bill Kinsey. Forthcoming. and the Millennium Development Goals: Challenges and "Long Term Consequences of Early Childhood Malnutrition." Opportunities for Implementation." Final report. Available Oxford Economic Papers. online at http://www.un.org/esa/socdev/unyin/documents/ Alford, Sue, Nicole Cheetham, and Debra Hauser. 2005. Science & youthmdgs.pdf. Success in Developing Countries: Holistic Programs That Work to Adams, Arvil V. 2006."The Role of Skills Development in the Prevent Teen Pregnancy, HIV & Sexually Transmitted Infections. Transition to Work: A Global Review." Background paper for the Washington, DC: Advocates for Youth. WDR 2007. Ali, Tariq Omar, Nuzhat Imam, Raihana Karim, and Nasheeba Aedo, Cristián, and Sergio Nuñez. 2001."The Impact of Training Selim. 2006. Voices of the Youth: Findings from Youth Consulta- Policies in Latin America and the Caribbean: The Case of Pro- tions in Bangladesh. Dhaka, Bangladesh: BRAC Research and grama Joven." ILADES - Georgetown University. Santiago de Evaluation Division. Available online at http://www.bracre- Chile. Processed. search.org/reports/final_youth_report_revised.pdf. 243 244 WORLD DEVELOPMENT REPORT 2007 Alwain, Duane F., and Jon A. Krosnick. 1991."Aging, Cohorts, and tion when Decisions are Inter related: Evidence from Urban the Stability of Sociopolitical Orientations Over the Life Span." Morocco." Topics in Middle Eastern and North African Econom- American Journal of Sociology 97(1):169­95. ics, Electronic Journal 5, Middle East Economic Association and Amin, Sajeda, Ian Diamond, Ruchira T. Naved, and Margaret Loyola University Chicago. Newby. 1998."Transition to Adulthood of Female Garment- Atchoarena, David, and Andre Marcel Delluc. 2001."Revisiting factory Workers in Bangladesh." Studies in Family Planning Technical and Vocational Education in Sub-Saharan Africa: An 29(2):185­200. Update on Trends, Innovations, and Challenges." Paris: Interna- Andersen, Susan L. 2003."Trajectories of Brain Development: tional Institute for Educational Planning, IIEP/Prg.DA 1.320. Point of Vulnerability or Window of Opportunity?" Neurosci- Au, Chun-Chung, and J. Vernon Henderson. Forthcoming."How ence and Biobehavioral Reviews 27(1-2):3­19. Migration Restrictions Limit Agglomeration and Productivity Angrist, Joshua, Eric Bettinger, Erik Bloom, Elizabeth King, and in China." Journal of Development Economics. Michael Kremer. 2002."Vouchers for Private Schooling in Audas, Rick, Eva Berde, and Peter Dolton. 2005."Youth Unem- Colombia: Evidence from a Randomized Natural Experiment." ployment and Labour Market Transitions in Hungary." Educa- American Economic Review 92(5):1535­58. tion Economics 13(1):1­25. Angrist, Joshua, Eric Bettinger, and Michael Kremer. Forthcom- Australia Immigration Visa Services. 1998. Immigration Laws: ing."Long-term Consequences of Secondary School Vouchers: December 1998. Sidney: Australia Immigration Visa Services. Evidence from Administrative Records in Colombia." American Available online at http://www.migrationint.com.au/news/ Economic Review. malta/dec_1998-03mn.asp. Angrist, Joshua, and Victor Lavy. 2002."New Evidence on Class- Autor, David H., Lawrence F. Katz, and Alan B. Krueger. 1998. room Computers and Pupil Learning." Economic Journal "Computing Inequality: Have Computers Changed the Labor 112(482):735­86. Market?" Quarterly Journal of Economics 113(4):1169­213. Angrist, Joshua D. 1990."Lifetime Earnings and the Vietnam Era Aziken, Michael E., Patrick I. Okonta, and Adedapo B. A. Ande. Draft Lottery: Evidence from Social Security Administrative 2003."Knowledge and Perception of Emergency Contraception Records." American Economic Review 80(3):313­36. among Female Nigerian Undergraduates." International Family Angrist, Joshua D., and Adriana D. Kugler. 2003."Protective or Planning Perspective 29(2):84­7. Counter-productive? Labour Market Institutions and the Effect Balcells i Ventura, Laia. 2005."Explaining Variation in the Salience of Immigration on EU Natives." Economic Journal 113(488): of Catalan Nationalism Across the France/Spain Border." Paper F302­F331. presented at the 7th Annual Retreat of the Society for Compara- Annan, Jeannie, Christopher Blattman, and Roger Horton. 2006. tive Research. March 6. Budapest. The State of Youth and Youth Protection in Northern Uganda: Banerjee, Abhijit, Shawn Cole, Esther Duflo, and Leigh Linden. Findings from the Survey of War Affected Youth. Uganda: UNI- 2005."Remedying Education: Evidence from Two Randomized CEF. Available online at www.sway-uganda.org. Experiments in India." London, U.K.: BREAD Working Paper 109. Arends-Kuenning, Mary, and Sajeda Amin. 2000."The Effects of Banerjee, Abhijit, and Esther Duflo. 2006."Addressing Absence." Schooling Incentive Programs on Household Resource Allo- Journal of Economics Perspectives 20(1):117­32. cation in Bangladesh." New York: Population Council Policy Bankole, Akinrinola, Susheela Singh, Vanessa Woog, and Deirdre Research Division, Working Paper 133. Wulf. 2004. Risk and Protection:Youth and HIV/AIDS in Sub- Arjona, Ana. 2006."Understanding Recruitment in Civil Wars." Saharan Africa. New York: The Alan Guttmacher Institute. Background paper for the WDR 2007. Bannon, Ian, Peter Holland, and Aly Rahim. 2005."Youth in Post- Armecina, G., Jere R. Behrman, P. Duazo, S. Chumanc, S. Gul- conflicts Settings." Washington, DC: World Bank, Youth Devel- tianoa, Elizabeth King, and N. Lee. 2006."Early Childhood opment Notes 1(1). Development through Integrated Programs: Evidence from the Barker, G. 2003."How Do We Know If Men Have Changed? Pro- Philippines." World Bank. Washington, DC. Processed. moting and Measuring Attitude Change with Young Men. Les- Arnett, Jeffrey Jensen. 2000."Emerging Adulthood: A Theory of sons from Program H in Latin America." Paper presented at the Development from the Late Teens through the Twenties." Amer- Expert Group Meeting on `The Role of Men and Boys in Achiev- ican Psychologist 55(5):469­80. ing Gender Equality'. November 21. Brasilia. Arunatilake, Nisha, Sisira Jayasuriya, and Saman Kelegama. 2001. Barr, Nicholas. 2004."Higher Education Funding." Oxford Review "The Economic Cost of the War in Sri Lanka." World Develop- of Economic Policy 20(2):264­83. ment 29(9):1483­500. Barrera Osorio, Felipe. 2005."Impact of Private Provision of Pub- Asian Development Bank. 2005. HIV/AIDS Prevention among lic Education: Empirical Evidence from Bogotá's Concessions Youth. Hanoi: Viet Nam Comission for Population, Family and Schools." Paper presented at the Mobilizing the Private Sector Children. for Public Education Conference. May 10. Harvard University, Asian Development Bank, and World Bank. 2005. Country Gender Boston, MA. Assessment. Manila and Washington, DC: Asian Development Barrera, Felipe, and Ana María Ibáñez. 2004."Does Violence Bank and World Bank. Reduce Investment in Education? A Theoretical and Empiri- Askew, Ian, and Ndugga Baker Maggwa. 2002."Integration of STI cal Approach." Universidad de Los Andes: Documentos CEDE Prevention and Management with Family Planning and Ante- 000582. Available online at http://economia.uniandes.edu. natal Care in Sub-saharan Africa: What More Do We Need to co/~economia/archivos/temporal/d2004-27.pdf. Know?" International Family Planning Perspectives 28(2):77­86. Barro, Robert J. 1999."Human Capital and Growth in Cross- Assaad, R., and S. Zouari. 2003."Estimating the Impact of Mar- Country Regression." Swedish Economic Policy Review riage and Fertility on the Female Labor Force Participa- 6(2):237­77. References 245 Basu, Kaushik, and James E. Foster. 1998."On Measuring Literacy." New Evidence from Guatemala." Emory University, University Economic Journal 108(451):1733­49. of Pennsylvania, and International Food Policy Research Insti- Basu, Kaushik, Hyejin Ku, and Homa Zarghamee. 2006."Determi- tute. Philadelphia, and Washington, DC. Processed. nants of Youth Behaviour and Outcomes: A Review of Theory, Behrman, Jere R., Alexis Murphy, Agnes Quisumbing, Usha Evidence and Policy Implications." Background paper for the Ramakrishna, and Kathyrn Young. 2006."What is the Real WDR 2007. Impact of Education on Age of First Parenthood and Family Bate, Peter. 2004. The Story Behind Oportunidades. Washington, Formation?" Background paper for the WDR 2007. DC: Inter-American Development Bank, Online Magazine: Behrman, Jere R., Piyali Sengupta, and Petra Todd. 2001. Progress- FOCUS. Available online at http://www.iadb.org/idbamerica/. ing through PROGRESA: An Impact Assessment of a School Sub- Bateman, C. 2001."Doctor Burnout Silent and Fatal." South Afri- sidy Experiment. Washington, DC: International Food Policy can Medical Journal 91(2):98­100. Research Institute. Bauböck, Rainer. 2005."Expansive Citizenship - Voting Beyond Bell, Clive, Ramona Bruhns, and Hans Gersbach. 2006."Economic Territory and Membership." Political Science & Politics Growth, Education and Aids in Kenya Model: A Long-run Anal- 38(4):683­87. ysis." Background paper for the WDR 2007. Bay, Ann-Helén, and Morten Blekesaune. 2002."Youth, Unem- Bell, Clive, Shantayanan Devarajan, and Hans Gersbach. 2006."The ployment and Political Marginalization." International Journal of Long-run Economic Costs of AIDS: A Model with an Applica- Social Welfare 11(2):132­39. tion to South Africa." World Bank Economic Review 20(1):55­89. Bayer, Patrick, Randi Pintoff, and David Pozen. 2005."Building Bell, Clive, Hans Gersbach, Ramona Bruhns, and Dagmar Volker. Criminal Capital Behind Bars: Peer Effects in Juvenile Correc- 2004."Economic Growth, Human Capital and Population in tions." New Haven, CT: Center for Economic Growth Working Kenya in the Time of AIDS: A Long-run Analysis in Historical Paper, Yale University 864. Perspective." University of Heidelberg. Heidelberg. Processed. BBC.2001."U.N.Finds Congo Child Soldiers."BBC News,February 21. Bellows, John, and Edward Miguel. 2006."War and Institutions: Bearman, Peter, and Hannah Brückner. 2001."Promising the New Evidence from Sierra Leone." American Economic Review Future: Virginity Pledges and First Intercourse." American Jour- 96(2):394­99. nal of Sociology 106(4):859­912. Benavot, Aaron. 2004. Comparative Analysis of Secondary Educa- Becker, Gary S. 1964. Human Capital. Chicago, IL: University of tion Curricula. Washington, DC: World Bank and International Chicago Press. Bureau of Education. Becker, Gary S., and Kevin M. Murphy. 1988."A Theory of Ratio- Benhabib, Seyla. 2005."Borders, Boundaries, and Citizenship." nal Addiction." Journal of Political Economy 96(4):675­700. Political Science & Politics 38(4):673­77. Bedard, Kelly, and Olivier Deschênes. 2006."The Long-Term Bentolila, Samuel, and Giuseppe Bertola. 1990."Firing Costs and Impact of Military Service on Health: Evidence from World Labour Demand: How Bad Is Eurosclerosis?" Review of Eco- War II and Korean War Veterans." American Economic Review nomic Studies 57(3):381­402. 96(1):176­94. Benus, Jacob, James Rude, and Satyendra Patrabansh. 2001. Bos- Beegle, Kathleen, Rajeev Dehejia, and Roberta Gatti. 2004."Why nia & Herzegovina: Impact of the Emergency Demobilization and Should We Care about Child Labor?" Washington, DC: World Reintegration Project. Washington, DC: US Department of Labor, Bank Policy Research Division Working Paper Series 3479. Bureau of International Affairs, Office of Foreign Relations. Behrman, Jere R., Nancy Birdsall, and Miguel Szekely. 2003."Eco- Bercovich, Alicia. 2004. People with Disability in Brazil: A Look at nomic Policy and Wage Differentials in Latin America." Wash- the 2000 Census Results. Rio de Janeiro: Instituto Brasilero de ington, DC: Center for Global Development Working Paper 29. Geografía e Estatistica. Available online at http://iussp2005. princeton.edu/download.aspx?submissionId=52108. Behrman, Jere R., and Anil B. Deolalikar. 1988."Health and Nutri- tion." In H. Chenery and T. N. Srinivasan, (eds.), Handbook of Berer, Marge. 2003."Integration of Sexual and Reproductive Development Economics vol. 1. Amsterdam: Elsevier. Health Services: A Health Sector Priority." Reproductive Health Matters 11(21):6­15. Behrman, Jere R., M. R. Foster, Mark Rosenzweig, and P. Vash- ishtha. 2002."Does Increasing Women's Schooling Raise the Berlinsky, S., S. Galiani, and P. Gertler. 2006."Public Pre-primary Schooling of the Next Generation?" American Economic Review Schooling and Primary School Performance." University Col- 92(1):323­34. lege, Universidad de San Andrés, and World Bank. London and Washington, DC. Processed. Behrman, Jere R., John Hoddinott, John A. Maluccio, Erica Soler-Hampejsek, Emily Berhman, Reynaldo Martorell, Agnes Berman, Eli, and David D. Laitin. 2005."Hard Targets: Theory and Quisumbing, Manuel Ramirez, and Aryeh D. Stein. 2005a. Evidence on Suicide Attacks." Cambridge, MA: National Bureau "What Determines Post-school Skills? Impacts of Pre-School, of Economic Research Working Paper Series 11740. School Years and Post School Experiences in Guatemala." Uni- Berman, Eli, and Stephen Machin. 2000."Skill-biased Technology versity of Pennsylvania, International Food Policy Research Transfer Around the World." Oxford Review of Economic Policy Institute, Middleburry College, Emory University and INCAP- 16(3):12­22. Guatemala. Philadelphia, PA. Processed. Berman, Peter, and Laura Rose. 1996."The Role of Private Pro- Behrman, Jere R., and James C. Knowles. 1999."Household viders in Maternal and Child Health and Family Planning Income and Child Schooling in Vietnam." World Bank Economic Services in Developing Countries." Health Policy and Planning Review 13(2):211­56. 11(2):142­55. Behrman, Jere R., A. Murphy, Agnes Quisumbing, Usha Ramak- Bernasconi, Andrés, and Fernando Rojas. 2004. Informe sobre la rishna, and Kathyrn Young. 2005b."What is the Real Impact Educación Superior en Chile: 1980-2003. Santiago, Chile: Edito- of Schooling on Age of First Union and Age of First Parenting? rial Universitaria. 246 WORLD DEVELOPMENT REPORT 2007 Berrebi, Claude. 2003."Evidence about the Link between Educa- Blanchard, Olivier. 2006."Emploi: La Solution passe par le CUP tion, Poverty, and Terrorism Among Palestinians." Princeton, (Contract Unique Progressif)." MIT. Cambridge, MA. Processed. NJ: Princeton University Industrial Relations Section Working Blanchard, Olivier, and Justin Wolfers. 2000."The Role of Shocks Paper 477. and Institutions in the Rise of European Unemployment: The Bertola, Giuseppe, Francine D. Blau, and Lawrence M. Kahn. 2002. Aggregate Evidence." Economic Journal 110(2000):C1­C33. "Labor Market Institutions and Demographic Employment Pat- Blanchard, Olivier J., and Augustin Landier. 2001."The Perverse terns." Cambridge, MA: National Bureau of Economic Research Effects of Partial Labor Markets Reform: Fixed Duration Con- Working Paper Series 9043. tracts in France." Cambridge, MA: MIT, Department of Eco- Bertrand, Marianne, Sendhil Mullanaithan, and Douglas Miller. nomics, Working Paper Series 01-14. 2003."Public Policies and Extended Families: Evidence from Blanco, Almudena, and Jochen Kluve. 2002."Why not Stay Home: Pensions in South Africa." World Bank Economic Review Nest-leaving Behavior in Western Europe." University of Cali- 17(1):27­50. fornia, Berkeley. Berkeley, CA. Processed. Besley, Timothy, Lawrence Haddad, John Hoddinott, and Michelle Blau, Francine D., and Lawrence M. Kahn. 1999."Institutions and Adato. 2004."Community Participation and the Performance of Laws in Labor Markets."In Orley Ashenfelter and David Card, Public Works Programs in South Africa." Dalhousie University. (eds.), Handbook of Labor Economics. Vol. 3A. Amsterdam: Elsevier. Halifax, Nova Scotia. Processed. Bloom, David E. 2005a."Education and Public Health: Mutual Betcherman, Gordon, Jean Fares, Amy Luinstra, and Robert Challenges Worldwide." Comparative Education Review Prouty. 2001."Child Labor, Education, and Children's Rights." 49:437­51. In Philip Alston and Mary Robinson, (eds.), Human Rights and Bloom, David E., and David Canning. 2005."Global Demographic Development: Toward Mutual Reinforcement. New York: Oxford Change: Dimensions and Economic Significance." Cambridge, University Press. MA: Harvard Initiative for Global Health Working Paper No. 1. Betcherman, Gordon, Martin Godfrey, Susana Puerto, Friederike Available online at http://www.hsph.harvard.edu/pgda/working/ Rother, and Antoneta Stavreska. 2006."Supporting Young working_paper1.pdf. Workers: Results of the Global Inventory of Interventions." Bloom, David E., David Canning, and Pia Malaney. 2000."Popula- World Bank. Washington, DC. Processed. tion Dynamics and Economic Growth in Asia." Population and Betcherman, Gordon, Karina Olivas, and Amit Dar. 2004."Impacts Development Review 26(supplement 2000):257­90. of Active Labor Market Programs: New Evidence from Evalu- Bloom, David E., and Jeffrey D. Sachs. 1998."Geography, Demog- ations with Particular Attention to Developing and Transition raphy, and Economic Growth in Africa." Brookings Papers on Countries." Washington, DC: World Bank, Social Protection Economic Activity 0(2):207­73. Discussion Paper Series 0402. Bloom, David E., and Jeffrey G. Williamson. 1998."Demographic Beyer, Marty. 1997."Experts for Juveniles at Risk of Adult Sen- Transitions and Economic Miracles in Emerging Asia." World tences." In Patricia Puritz, Alycia Capozello, and Wendy Shang, Bank Economic Review 12(3):419­55. (eds.), More Than Meets the Eye: Rethinking Assessment, Compe- tency and Sentencing for a Harsher Era of Juvenile Justice. Wash- Bloom, Mia. 2005b. Dying to Kill: The Allure of Suicide Terror. New ington, DC: American Bar Association Juvenile Justice Center. York, NY: Columbia University Press. Bhargava, Alok. 2005."AIDS Epidemic and the Psychological Well- Bloomfield, Kim, Tim Stockwell, Gerhard Gmel, and Nina Rehn. being and School Participation of Ethiopian Orphans." Psychol- 2003."International Comparisons of Alcohol Consumption." ogy, Health and Medicine 10(3):263­75. Alcohol Research & Health 27(1):95­109. Bhatnagar, Deepti, Ankita Dewan, Magüi Moreno Torres, and Blum, R. W., and M. Ireland. 2004."Reducing Risk, Increasing Pro- Parameeta Kanung. 2003."The Bangladesh Female Second- tective Factors: Findings from the Caribbean Youth Health Sur- ary School Assistance Project." World Bank. Washington, DC. vey." Journal of Adolescence Health 35(6):493­500. Processed. Blum, Robert. 2006."The Adolescent Brain." Background note for Billari, Francesco C., Dimiter Philipov, and Pau Baizán Munoz. the WDR 2007. 2001."Leaving Home in Europe: The Experience of Cohorts Blum, Robert W., and Heather P. Libbey. 2004."School Connected- Born Around 1960." International Journal of Population Geogra- ness - Strengthening Health and Education Outcomes for Teen- phy 7(5):339­56. agers." Journal of School Health 74(7):231­34. Billig, Shelley. 2006."Service Learning." In Lonnie R. Sherrod, Blum, Robert W., and Kristin Nelson-Mmari. 2004."The Health of Constance A. Flanagan, Ron Kassimir, and Amy K. Bertelsen, Young People in a Global Context." Journal of Adolescent Health (eds.), Youth Activism: An International Encyclopedia. Westport, 35(5):402­18. CT: Greenwood Press. Boisjoly, Johanne, Greg J. Duncan, Michael Kremer, Dan M. Levy, Binder, Melissa. 2005. The Cost of Providing Universal Secondary and Jacque Eccles. 2004."Empathy or Antipathy? The Impact of Education in Developing Countries. Cambridge, MA: American Diversity." Harvard University. Cambridge, MA. Processed. Academy of Arts and Sciences. Available online at http://www. Bonelli, Regis, and Alinne Veiga. 2004."Determinants of Educa- amacad.org/projects/ubase.aspx. tional Exclusion in Five States of Brazil." In Maria-Valeria Pena Black, Richard. 2003. Soaring Remittances Raise New Issues. Wash- and Maria Madalena, (eds.), Children's and Youth Vulnerabil- ington, DC: Migration Policy Institute. Available online at ity: Poverty, Exclusion and Social Risk in Five Brazilian States. http://www.migrationinformation.org/about.cfm. Brasilia: World Bank. Black, Sandra E., Paul J. Devereaux, and Kjell G. Salvanes. 2004. Borgarello, Andrea, Susan Duryea, Olgiati Scarpetta, and Stefano "Fast Times at Ridgemont High? The Effect of Compulsory Scarpetta. 2005."Early Experiences of Youths in the Labor Mar- Schooling Laws on Teenage Births." Cambridge, MA: National ket: Stepping Stones or Traps?" World Bank. Washington, DC. Bureau of Economic Research Working Paper Series 10911. Processed. References 247 Borjas, George J. 2003."The Labor Demand Curve IS Downward Philip R. Musgrove, (eds.), Disease Control Priorities in Develop- Sloping: Reexamining the Impact of Immigration on the Labor ing Countries. New York: Oxford University Press. Market." Quarterly Journal of Economics 118(4):1335­74. Cáceres, Carlos F., Anna M. Rosasco, Jeffrey S. Mandel, and Nor- Boudarbat, Brahim. 2005."Job-search Strategies and the Unem- man Hearst. 1994."Evaluating a School-based Intervention for ployment of University Graduates in Morocco." Paper presented STD/AIDS Prevention in Peru." Journal of Adolescent Health at the IZA-EBRD International Conference on Labor Market 15(7):582­91. Dynamics. May 5. Bologna, Italy. Cahuc, Pierre, and Stéphane Carcillo. 2006. Que Peut-on Attendre Bourguignon, Francois, Francisco Ferreira, and Marta Menendez. des Contrats Nouvelle Embauche et Premiere Embauche? Paris: 2005."Inequality of Opportunity in Brazil?" World Bank. Wash- Université de Paris. Available online at http://www.crest.fr/page- ington, DC. Processed. perso/cahuc/CNECahucCarcillo.pdf. Bowles, Samuel, and Howard Gintis. 1976. Schooling in Capitalist Caldwell, Brian J., Rosalind Levacic, and Kenneth N. Ross. 1999. America: Educational Reform and the Contradictions of Economic "The Role of Formula Funding of Schools in Different Edu- Life. New York: Basic. cational Policy Contexts." In Rosalind Levacic and Kenneth N. Boyer, Cherrie B., and Mary-Ann Shafer. 1997."Evaluation of a Ross, (eds.), Needs-Based Resource Allocation in Education: Via Knowledge -and Cognitive- Behavioral Skills-Building Interven- Formula Funding of Schools. Paris: UNESCO, International Insti- tion to Prevent STDs and HIV Infection in High School Stu- tute for Educational Planning. dents." Adolescence 32(125):25­42. Caldwell, John C., and Pat Caldwell. 2002."Is Integration the Brady, Martha. 1998."Laying the Foundation for Girls' Healthy Answer for Africa?" International Family Planning Perspective Futures: Can Sports Play a Role?" Studies in Family Planning 28(2):108­10. 29(1):79­82. Canals-Cerda, José, and Cristobal Ridao-Cano. 2004."The Dynamics of School and Work in Rural Bangladesh." Wash- Brandt, Pamela Robin. 2005."The Brotherhood." American Way ington, DC: World Bank Policy Research Working Paper Series Magazine. May 15, 2005. 3330. Bray, Mark. 2004."Sharing the Burden of Financing: Government Card, David. 2005."Is the New Immigration Really that Bad?" Eco- and Household Partnerships for Basic Education." Economic nomic Journal 115(507):F300­F323. Affairs 24(4):22­26. Card, David, and Thomas Lemieux. 2000a."Adapting to Circum- Brett, Rachel, and Irma Specht. 2004. Young Soldiers: Why They stances: The Evolution of Work, School and Living Arrangments Choose to Fight. Geneva: Lynne Rienner Publisher for the Inter- among North American Youth." In Dabid G. Blanchflower and national Labour Organization. Richard B. Freeman, (eds.), Youth Employment and Joblessness in Brown, Gillian, Laila Al-Hamad, and Carmen de Paz Nieves. 2005. Advanced Countries. Chicago, IL: University of Chicago Press. Gender Equality in East Asia: Progress, and the Challenges of ------. 2000b."Dropout and Enrollment Trends in the Post-War Economic Growth and Political Change. Washington, DC: World Period: What Went Wrong in the 1970's?" Cambridge, MA: Bank, East Asia and Pacific Region, Social Development Team. National Bureau of Economic Research Working Paper Series Brown, Ian. 1990. Khomeini's Forgotten Sons: The Story of Iran's Boy 7658. Soldiers. London, U.K.: Grey Seal. Carneiro, Pedro, and James J. Heckman. 2002."The Evidence on Bruce, Judith, and Shelley Clark. 2004. The Implications of Early Credit Constraints in Post-Secondary Schooling." Economic Marriage for HIV/AIDS Policy. Brief based on background paper Journal 112(482):705­34. prepared for the WHO/UNFPA/Population Council Technical ------. 2003."Human Capital Policy." In James J. Heckman and Consultation on Married Adolescents. New York: Population Alan B. Krueger, (eds.), Inequality in America: What Role for Council. Human Capital Policies? Cambridge, MA: MIT Press. Brückner, Hannah, and Peter Bearman. 2005."After the Promise: Carneiro, Pedro, and Sokbae Lee. 2005."Ability, Sorting and Wage The STD Consequences of Adolescent Virginity Pledges." Jour- Inequality." University College London. London. Processed. nal of Adolescent Health 36(4):271­8. Carneiro, Pedro, and Cristobal Ridao-Cano. 2005."Heterogeneity Bruns, Barbara, Alain Mingat, and Ramahatra Rakotomalala. 2003. and Uncertainty in Returns to High School: The Case of Indo- Achieving Universal Primary Education by 2015: A Chance for nesia." Background paper for the WDR 2007. Every Child. Washington, DC: World Bank. Catalano, Richard F., Kevin P. Haggerty, Sabrina Oesterle, Charles Burgess, Simon, Carol Propper, Hedley Reeys, and Arran Shearer. B. Fleming, and J. David Hawkins. 2004."The Importance of 2003."The Class of 1981: The Effects of Early Career Unem- Bonding to School for Healthy Development: Findings from the ployment on Subsequent Unemployment Experiences." Labor Social Development Research Group." Journal of School Health Economics 10(3):291­309. 74(7):252­61. Buvinic, Mayra. 1998."The Cost of Adolescent Childbearing: Evi- Celentano, David D., Noya C. Galai, Ajay K. Sethi, Nina G. Shah, dence from Chile, Barbados, Guatemala and Mexico." Studies in Steffanie A. Strathdee, David Vlahov, and Joel E. Gallant. 2001. Family Planning 29(2):201­9. "Time to Initiating Highly Active Antiretroviral Therapy among -------. 2005."Economic Opportunities for Women: Experiences HIV-Infected Injection Drug Users." AIDS 15(13):1707­15. from Latin America and the Caribbean." Paper presented at the Centers for Disease Control and Prevention. 1993."Mortality High-Level Policy Forum on Enhancing Women Economic Par- Trends for Selected Smoking-Related Cancers and Breast Cancer ticipation and Opportunities. El Cairo, Egypt. -- United States, 1950-1990." Morbidity and Mortality Weekly Buvinic, Mayra, Andre Medici, Elisa Fernandez, and Ana Cris- Report 42(44):857­863-6. tina Torres. 2006."Gender Differentials in Health." In Dean T. ------. 1998."Recommendations to Prevent and Control Iron Jamison, Joel Breman, Anthony R. Measham, George Alleyne, Deficiency in the United States, Recommendations and Report." Mariam Claeson, David B. Evans, Prabhat Jha, Anne Mills, and Morbidity and Mortality Weekly Report 47(RR-3):1­36. 248 WORLD DEVELOPMENT REPORT 2007 Centre for Development and Population Activities (CEDPA). Paradoxes of Youth and Sport. Albany: State University of New 2001. Adolescent Girls in India Choose a Better Future: An Impact York Press. Assessment. Washington, DC: Center for Development and Pop- Coale, Ansley J., and Edgar M. Hoover. 1958. Population Growth and ulation Activities (CEDPA). Economic Development in Low-Income Countries: A Case Study of Chandra, Prabha S., K. R. Jairam, and Anila Jacob. 2004."Factors India's Prospects. Princeton, NJ: Princeton University Press. Related to Staff Stress in HIV/AIDS Related Palliative Care." Coalition to Stop the Use of Child Soldiers. 2000."The Use of Indian Journal of Palliative Care 10(2):48­54. Children by OSCE Member States." Paper presented at the Chapman, Bruce. Forthcoming."Income Contingent Loans Human Dimension Seminar on Children and Armed Conflict. for Higher Education: An International Reform." In Eric A. Warsaw, Poland. Hanushek and Finis Welch (eds.) Handbook of the Economics of ------. 2002."DRC Child Soldiers." London, UK: Child Soldiers Education. Amsterdam: North-Holland. Newsletter 3. Chattopadhyay, Raghabendra, and Esther Duflo. 2003."Women as ------. 2003. The Use of Child Soldiers in the Americas: An Over- Policy Makers: Evidence from a India-Wide Randomized Policy view. London, U.K.: Coalition to Stop the Use of Child Soldiers. Experiment." MIT. Cambridge, MA. Processed. ------. 2004. Child Soldier Use 2003: A Briefing for the 4th UN Chaudhuri, Saulma, and Pratima Paul-Majumder. 1995. The Con- Security Council Open Debate on Children and Armed Conflict. ditions of Garment Workers in Bangladesh: An Appraisal. Dhaka: London, U.K.: Coalition to Stop the Use of Child Soldiers. Bangladesh Institute of Development Studies. Cockx, Bart, and Christian Göbel. 2004."Subsidized Employment Chaudhury, Nazmul, Jeffrey Hammer, Michael Kremer, Karthik for Young Long-term Unemployed Workers ­ An Evaluation." Muralidharan, and F. Halsey Rogers. 2006."Missing in Action: Université Catholique de Louvain. Louvain. Available online at Teacher and Health Worker Absence in Developing Countries." http://www.iza.org/en/webcontent/teaching/summerschool_ Journal of Economic Perspectives 20(1):91­116. html/7thsummer_school_files/ss2004_goebel.pdf. Processed. Chen, Kun, and Martin Kenney. 2005."Universities/Research Insti- Cogan, John J., Paul Morris, and Murray Print, eds. 2002. Civic tutes and Regional Innovation Systems: The Case of Beijing and Education in the Asia-Pacific Region: Case Studies Across Six Soci- Shenzhen." Berkeley, CA: BRIE Working Paper 168. Available eties. Oxford, U.K.: Rutledge. online at http://brie.berkeley.edu/publications/wp168revised.pdf. Cohen, Joel E., and David E. Bloom. 2005."Cultivating Minds: Chen, M. Keith, and Jesse M. Shapiro. 2004."Does Prison Harm Educating All Children is not only Urgent but also Fea- Inmates? A Discontinuity-Based Approach." Harvard University. sible within the Next Few Years." Finance and Development Cambridge, MA. Processed. 42(2):8­14. Chewning, Betty, Pat Mosena, Dale Wilson, Harold Erdman, San- Cohen, Susan A. 2004."Beyond Slogans: Lessons From Uganda's dra Potthoff, Anita Murphy, and Kathleen Kennedy Kuhnen. Experience with ABC and HIV/AIDS." Reproductive Health Mat- 1996."Evaluation of a Computerized Contraceptive Decision ters 12(23):132­35. Aid for Adolescent Ppatients." Patient Education and Counseling Cole, Jennifer. 2004a."Fresh Contact in Tamatave, Madagascar: 38(3):227­39. Sex, Money, and Intergenerational Transformation." American Chiclet, Christophe. 2001."Otpor: The Youth who Booted Milos- Ethnologist 31(4):573­88. evic." UNESCO Courier. ------. 2004b."The Jaombilo of Tamatave (Madagascar), 1992- Cincotta, Richard P., Robert Engleman, and Daniele Anastasion. 2004: Reflections on Youth and Globalization." Journal of Social 2003. The Security Demographic: Population and Civil Con- History 38(4):891­914. flict after the Cold War. Washington, DC: Population Action Committee for Population, Family and Children Vietnam, and International. ORC Macro. 2003. Vietnam Demographic and Health Survey Cinterfor/ILO. 2001. Modernization in Vocational Education and 2002. Calverton, MD: Committee for Population, Family and Training in the Latin American and the Caribbean Region. Mon- Children (Vietnam), ORC Macro. tevideo: Cinteford - ILO. Conde-Aguledo, Agustín, Anyeli Rosas-Bermúdez, and Ana Cecilia Clark, Shelley. 2004."Early Marriage and HIV Risks in Sub-Saha- Kafury-Goeta. 2006."Birth Spacing and Risk of Adverse Perina- ran Africa." Studies in Family Planning 35(3):149­60. tal Outcomes: A Meta-analysis." Journal of the American Medical Clarke, George R. G., and Scott Wallsten. 2004."Has the Inter- Association 295(15):1809­23. net Increased Trade? Evidence from Industrial and Developing Cox Edwards, Alejandra, and Manuelita Ureta. 2003."Interna- Countries." Washington DC: World Bank Policy Research Work- tional Migration, Remittances and Schooling: Evidence from El ing Paper Series 3215. Salvador." Journal of Development Economics 72(2):429­61. Clementino, Josbertini. 2006. Speech at Vozes Jovens Pernambuco Cox, Cristián. 2004."Policy Formation and Implementation in (part of WDR consultations), Recife, Brazil, January 17. Secondary Education Reform: The Case of Chile in the 1990s." Clert, Carine, Elizabeth Gomart, Ivana Aleksic, and Natalie Otel. Paper presented at the Second Regional Secondary Education in 2005."Human Trafficking in South Eastern Europe: Beyond Africa Conference. June 6. Dakar. Crime Control, an Agenda for Social Inclusion and Develop- CRECE. 2005. Sistema de Aprendizaje Tutorial. Columbia: CRECE. ment." World Bank. Washington, DC. Processed. Cunha, Flavio, James J. Heckman, Lance Lochner, and Dimitriy V. Coady, David P. 2000. The Application of Social Cost-benefit Analy- Masterov. 2005."Interpreting the Evidence on Life Cycle Skill sis to the Evaluation of PROGRESA. Report Submitted to PRO- Formation." Cambridge, MA: National Bureau of Economic GRESA. Washington, DC: International Food Policy Research Research Working Paper Series 11331. Institute. Cunningham, Wendy, and Lucas Siga. 2006."Wage and Employ- Coakley, Jay. 2002."Using Sports to Control Deviance and Vio- ment Effects of Minimum Wage on Vulnerable Groups in the lence among Youths: Let's Be Critical and Cautious." In Marga- Labor Market: Brazil and Mexico." World Bank. Washington, ret Gatz, Michael A. Messner, and Sandra J. Ball-Rokeach, (eds.), DC. Processed. References 249 Currie, Janet, and Duncan Thomas. 1995."Does Head Start Make DeMaeyer, E. M. 1989. Preventing and Controlling Iron Deficiency a Difference?" American Economic Review 85(3):341­64. Anemia through Primary Health Care: A Guide for Health Curtain, Richard. 2001."Promoting Youth Employment through Administrators and Programme Managers. Geneva: World Health Information and Communication Technologies (ICT): Best Organization. Practice Examples in Asia and the Pacific." Paper presented at Deparment of Health Services Sri Lanka. 2002. Annual Health the ILO/Japan Tripartite Regional Meeting on Youth Employ- Bulletin 2002. Colombo, Sri Lanka: Sri Lanka, Department of ment in Asia and the Pacific. Bangkok. Health Services. ------. 2005. Case for Investing in Young People as Part of a Department of Corrections. 2003. Draft White Paper On Correc- National Poverty Reduction Strategy. New York: UNFPA. Avail- tions in South Africa. Pretoria, South Africa: Department of able online at http://www.unfpa.org/upload/lib_pub_file/424_ Corrections. filename_Investing.pdf. DeSimone, Jeff. 2002."Illegal Drug Use and Employment." Journal Dalrymple, William. 2005."Inside the Madrasas." The New York of Labor Economics 20(4):952­77. Review of Books. Devia, Sergio. 2003. Exito o Fracaso de las Políticas Públicas de Davies, Lynn, Christopher Williams, and HiromiMan-Hing Aubrey Capacitación Laboral a Jóvenes? - Evaluación del Programa Tes- Ko Yamashita. 2006. Inspiring Schools: Taking up the Challenge of tigo: `Proyecto Joven' de Argentina (1993-2000). Geneva: Interna- Pupil Participation. London, U.K.: Carnegie Trust. tional Labour Office. Available online at http://www.ilo.org/. Davis, Todd M. 2003. Atlas of Student Migration. New York: Insti- Dhakal, Raju Malla, and M. Sheikh Misbah. 1997. Breaking Barri- tute of International Education. ers Building Bridges: A Case Study of USAID/Nepal's SO3 Wom- de Cordoba, José. 2004."AsVenezuela Tilts Left, a Rum Mogul Reaches en's Empowerment Program. Kathmandu, Nepal: USAID. Out to the Poor."Wall Street Journal, November 10. Page: A1. Diamond, Larry. 2003."Universal Democracy." Policy Review De Ferranti, David, Guillermo E. Perry, Indermit S. Gill, J. Luis 119(June & July):3­25. Guasch, William F. Maloney, Carolina Sánchez-Páramo, and Dickson-Tetteh, Kim, Audrey Pettifor, and Winnie Moleko. 2001. Norbert Schady. 2003. Closing the Gap in Education and Technol- "Working with Public Sector Clinics to Provide Adolescent- ogy. Washington, DC: World Bank. Friendly Services in South Africa." Reproductive Health Matters De Fraja, Gianni, Tania Oliveira, and Luisa Zanchi. 2005."Must 9(17):160­69. Try Harder. Evaluating the Role of Effort in Educational Attain- ment." London: Centre for Economic Policy Research Discus- Ding, Weili, and Stephen F. Lehrer. Forthcoming."Do Peers Affect sion Papers 5048. Student Achievement in China's Secondary Schools?" Review of Economic and Statistics. de Janvry, Alain, Frederico Finan, Elisabeth Sadoulet, and Renos Vakis. 2006."Can Conditionnal Cash Transfers Serve as Dodds, Peter S., Roby Muhamad, and Duncan J. Watts. 2003."An Safety Nets in Keeping Children at School and from Working Experimental Study of Search in Global Social Networks." Sci- when Exposed to Shocks?" Journal of Development Economics ence 301(5634):827­29. 79(2):349­73. Dohmen, Thomas, Armin Falk, David Huffman, Uwe Sunde, de Jong, Joop T. V. M., Ivan H. Komproe, Mark Van Ommeren, Jürgen Schupp, and Gert Wagner. 2005."Five Facts about Risk Mustafa El Masri, Mesfin Araya, Noureddine Khaled, Willem Attitudes: Evidence from a Large, Representative, Experimen- van de Put, and Daya Somasundaram. 2001."Lifetime Events tally-Validated Survey." Institute for the Study of Labor (IZA). and Posttraumatic Stress Disorder in 4 Postconflict Settings." Bonn. Processed. Journal of American Medical Association 286(5):555­62. Donato, Katharine M., and Evelyn Patterson. 2004."Women and de Moura Castro, Claudio. 1999. Proyecto Joven: New Solutions and Men on the Move: Undocumented Border Crossing." In Jorge Some Surprises. Washington, DC: Inter-American Development Durand and Douglas S. Massey, (eds.), Crossing the Border: Bank. Available online at http://www.colombiajoven.gov.co/ Research from the Mexican Migration Project. New York, NY: injuve/instit/bid/4_pjov.pdf. Russell Sage Foundation. De Walque, Damien. 2004."How Does the Impact of an HIV/ Dowdney, Luke. 2005. Neither War nor Peace: International Com- AIDS Information Campaign Vary with Educational Attain- parisons of Children and Youth in Organized Violence. Rio de ment? Evidence from Rural Uganda." Washington, DC: World Janeiro, Brazil: Children and Youth in Organized Armed Vio- Bank Policy Research Working Paper Series 3289. lence, Viva Rio, Instituto de Estudos da Religião. ------. 2005."Parental Education and Children's Schooling Out- Dréze, Jean, and Haris Gazdar. 1997."Uttar Pradesh: The Burden comes: Is the Effect Nature, Nurture, or Both? Evidence from of Inertia." In Jean Dréze and Amartya Sen, (eds.), Indian Devel- Recomposed Families in Rwanda." Washington, DC: World opment: Selected Regional Perspectives. New Delhi: Oxford Uni- Bank Policy Research Working Paper Series 3483. versity Press. Deaton, Angus S., and Christina H. Paxson. 1997."The Effects Drucker, Ernest, Peter Lurie, Alex Wodak, and Philip Alcabes. 1998. of Economic and Population Growth in National Saving and "Measuring Harm Reduction: The Effects of Needle and Syringe Inequality." Demography 34(1):97­114. Exchange Programs and Methadone Maintenance on the Ecol- Decker, Scott H., and Barrik Van Winkle. 1996. Life in the Gang: ogy of HIV." AIDS 12(Suppl A):217­30. Family, Friends and Violence. New York, NY: Cambridge Univer- Duflo, Esther, Pascaline Dupas, Michael Kremer, and Samuel sity Press. Simei. 2006."Education and HIV/AIDS Prevention: Evidence Delisle, Hélène, Venkatraman Chandra-Mouli, and Bruno de from a Randomized Evaluation in Western Kenya." Background Benoist. 2000."Should Adolescents Be Specifically Targeted paper for the WDR 2007. for Nutrition in Developing Countries? To Address Which Dulci, Luiz. 2005. Speech at seminar "Young Voices--A View of Problems, and How?" World Health Organization. Geneva. Youth Organizations and Movements on 21st Century Brazil," Processed. Brasilia, Brazil, May 24. 250 WORLD DEVELOPMENT REPORT 2007 Dunkle, Kristin L., Rachel K. Jewkes, Heather C. Brown, Glenda E. Elliot, Delbert S., and Patrick H. Tolan. 2005."Youth Violence Gray, James A. McIntryre, and Siobán D. Harlow. 2004."Trans- Prevention, Intervention, and Social Policy: An Overview." In actional Sex among Women in Soweto, South Africa: Prevalence, Daniel J. Flanner and C. Ronald Huff, (eds.), Youth Violence Pre- Risk Factors and Association with HIV Infection." Social Science vention, Intervention, and Social Policy. Washington, DC: Ameri- & Medicine 59(8):1581­92. can Psychiatric Press, Inc. Dupas, Pascaline. 2006."Relative Risks and the Market for Sex: Eloundou-Enyegue, Parfait M. 2004."Pregnancy Related Dropouts Teenagers, Sugar Daddies and HIV in Kenya." EHESS-PSE. Paris. and Gender Inequality in Education: A Lifetable Approach and Processed. Application to Cameroon." Demography 41(3):509­28. Durham, Deborah. Forthcoming."Empowering Youth: Making Eltigani, Eltigani E. 2000."Changes in Family-Building Patterns Youth Citizens in Botswana." In Jennifer Cole and Deborah in Egypt and Morocco: A Comparative Analysis." International Durham (eds.) Generations and Globalization: Family, Youth and Family Planning Perspectives 26(2):73­8. Age in the New World Economy. Bloomington, I.N.: Indiana Uni- Elway, Ann. 1999."Poverty and Disability: A Review of the Litera- versity Press. ture." Washington, DC: World Bank, Social Protection Discus- Duryea, Susan, Jasper Hoek, David Lam, and Deborah Levison. sion Paper Series 9932. Forthcoming."Dynamics of Child Labor: Labor Force Entry and Emerson, Patrick M., and André P. Souza. 2002."The Effect of Exit in Urban Brazil." In Peter F. Orazem, Guilherme Sedlacek, Adolescent Labor on Adult Earnings and Female Fertility in and Zafiris Tzannatos (eds.) Child Labor and Education in Latin- Brazil." University of Colorado. Denver, CO. Processed. America. Washington, DC: Inter-American Development Bank. ------. 2003."Is There a Child Labor Trap? Inter-generational Duryea, Suzanne, Analia Olgiati, and Leslie Stone. 2006."The Persistence of Child Labor in Brazil." Economic Development and Under-Registration of Births in Latin America."Washington, DC: Cultural Change 51(2):375­98. Inter-American Development Bank Working Paper Series 551. Dustmann, Christian. 2001."Why Go Back? Return Motives of Emerson, Patrick M., and André Portela Souza. 2006."Is Child Migrant Workers." In Slobodan Djajic, (eds.), International Migra- Labor Harmful? The Impact of Working Earlier in Life on Adult tion: Trends, Policies and Economic Impacts. Oxford: Routledge. Earnings." University of Colorado at Denver and Cornell Uni- versity. Denver, CO and Cornell, NY. Processed. Dustmann, Christian, Francesca Fabbri, and Ian Preston. 2005. "The Impact of Immigration on the British Labor Market." Eco- Emmerson, Carl, Christine Frayne, Sandra McNally, and Olmo nomic Journal 115(507):F324­341. Silva. 2005. Evaluation of Aimhigher: Excellence Challenge. The Early Impact of Aimhigher: Excellence Challenge on Pre-16 Out- Ebbeling, Cara B., Dorota B. Pawlak, and David S. Ludwig. 2002. comes: An Economic Evaluation. London: British Department for "Childhood Obesity: Public-health Crisis, Common Sense Education and Skills. Available online at http://www.dfes.gov. Cure." The Lancet 360(9331):473­482. uk/research/data/uploadfiles/RR652.pdf. Edmunds, Timothy, Anthony Forster, and Andrew Cottey. 2002. ENCISS, and World Bank. 2006."Youth in Sierra Leone: Focus "Armed Forces and Society: A Framework for Analysis." Bristol, Group Discussions." World Bank. Washington, DC. Processed. U.K.: The Transformation of Civil-Military Relations in Central and Eastern Europe 1.13c. Enlaces Centro de Educación y Tecnología. 2005. Estadisticas Nacionales Enlaces 2005. Santiago de Chile: Chile, Ministerio de Eggleston, Elizabeth, Jean Jackson, Wesley Rountree, and Zhiy- Educación. Available online at http://www.enlaces.cl/libro/esta- ing Pan. 2000."Evaluation of a Sexuality Education Program disticas.pdf. for Young Adolescents in Jamaica." Revista Panamericana de la Salud Publica 7(2):102­12. Erikson, Erik H. 1994. Identity: Youth and Crisis. New York, NY: W. Eggleston, Elizabeth, Joan Leitch, and Jean Jackson. 2000."Con- W. Norton & Company, Inc. sistency of Self-reports of Sexual Activity among Young Ado- Erulkar, Annabel S., Linus I. A. Ettyang, Charles Onoka, Frederik lescents in Jamaica." International Family Planning Perspectives K. Nyagah, and Alex Muyonga. 2004."Behavior Change Evalu- 26(2):79­84. ation of a Culturally Consistent Reproductive Health Program Ekehaug, Vidar. 2005. African Youth at the UN World Summit! for Young Kenyans." International Family Planning Perspectives World Wide Web: Panorama: A TakingITGlobal Online Publica- 30(2):58­67. tion. Available online at http://www.takingitglobal.org/express/ Erulkar, Annabel S., and Barbara S. Mensch. 1997. Youth Centers panorama/article.html?ContentID=6543. in Kenya: Evaluation of the Family Planning Association of Kenya El-Cherkeh, Tanja, Elena Stirbu, and Andrea Tolciu. 2006."Youth Programme. Nairobi, Kenya: Population Council. Migration to Europe: Potential Impact on the Labour Market of Eusuf and Associates, and Center on Social Research and Human the Countries of Origin." Background paper for the WDR 2007. Development. 2002. Non-formal Education Project-3: Mid-term El-Zanaty, Fatma, and Ann Way. 2001. Egypt Demographic and Evaluation Report. Dhaka: Eusuf Associates and Center on Social Health Survey 2000. Calverton, MD: Egypt Ministry of Health Research and Human Development for the Government of and Population, National Population Council and ORC Macro. Bangladesh. Available online at http://www.measuredhs.com/. Ezzati, Majid, and Alan D. Lopez. 2004."Smoking and Oral Elias, Victor, F. Ruiz-Nuñez, R. Cossa, and D. Bravo. 2004."An Tobacco use." In Majid Ezzati, Alan D. Lopez, Anthony Rodgers, Econometric Cost-benefit Analysis of Argentina's Youth Train- and Christopher J. L. Murray, (eds.), Comparative Quantification ing Program." Washington, DC: Inter-American Development of Health Risks: Global and Regional Burden of Disease Attribut- Bank, Research Network Working Paper R-482. able to Selected Major Risk Factors. Geneva: WHO. Ellertson, Charlotte, Beverly Winikoff, Elizabeth Armstrong, Sha- Fafchamps, Marcel, and Agnes R. Quisumbing. 2005."Marriage, ron Camp, and Pramilla Senanayake. 1995."Expanding Access Bequest, and Assortative Matching in Rural Ethiopia." Economic to Emergency Contraception in Developing Countries." Studies Development and Cultural Change 53(2):347­80. in Family Planning 26(5):251­63. Family Health International. 2005.Network 23(4) References 251 FAO. 2006. Assessment of the Double Burden of Malnutrition 2000."Treating Tobacco Use and Dependence: A Public Health in Six Case Study Countries. Rome: Food and Agriculture Service Clinical Practice Guideline." Paper presented at the Press Organization. Briefing, HHS Auditorium. June 27. Washington, DC Fares, Jean, and Claudio E. Montenegro. 2006."Youth Unemploy- Firebaugh, Glenn, and Kevin Chen. 1995."Vote Turnout of Nine- ment's Dynamics: Evidence from Brazil (1978-2002) and Chile teenth Amendment Women: The Enduring Effect of Disenfran- (1957-2005)." Background paper for the WDR 2007. chisement." American Journal of Sociology 100(4):972­96. Fares, Jean, Claudio E. Montenegro, and Peter F. Orazem. 2006a. Fitzgerald, A. M., B. F. Stanton, N. Terreri, H. Shipena, X. Li, J. "How are Youth Faring in the Labor Market? Evidence from Kahihuata, I. B. Ricardo, J. S. Galbraigth, and A. M. de Jaeger. Around the World." Background paper for the WDR 2007. 1999."Use of Western-Based HIV Risk-Reduction Interventions ------. 2006b."Variation in the Returns to Schooling Across Targeting Adolescents in an African Setting." Journal of Adoles- and Within Developing Economies." Background paper for the cent Health 25(1):52­61. WDR 2007. Fleisher, Belton M., Haizheng Li, Shi Li, and Xiaojun Wang. 2004. Fares, Jean, and Dhushyanth Raju. 2006."Child Labor across the "Sorting, Selection, and Transformation of the Return to Col- Developing World: Patterns, Correlations and Determinants." lege Education in China." Bonn: Institute for the Study of Labor Background paper for the WDR 2007. (IZA), Discussion Paper Series 1446. Fares, Jean, and Erwin Tiongson. 2006."Entering the Labor Mar- FOCUS. 1998. Reproductive Health Programs for Young Adults: ket and Early Mobility of Youth: Evidence from Panel Estimates Health Facility Programs. Watertown, MA: FOCUS on Young in Bosnia and Herzegovina." Background paper for the WDR Adults, Pathfinder International. 2007. Fortenberry, J. Dennis. 2005."The Limits of Abstinence-only in Fearon, James D., and David D. Laitin. 2003."Ethnicity, Insur- Preventing Sexually Transmitted Infections." Journal of Adoles- gency, and Civil War." American Political Science Review cent Health 36(4):269­70. 97(1):75­90. Fougère, Denis, Francis Kramarz, and Julien Pouget. 2006."Youth Fernald, Lia C., Juan Pablo Gutierrez, Lynnette M. Neufeld, Gus- Unemployment and Crime in France." Bonn, Germany: Discus- tavo Olaiz, Stefano F. Bertozzi, Michelle Mietus-Snyder, and sion Paper Series of the Institute for the Study of Labor (IZA) Paul J. Gertler. 2004."High Prevalence of Obesity Among the 2009. Poor in Mexico." Journal of the American Medical Association Franklin, Mark N. 2004. Voter Turnout and the Dynamics of Elec- 291(21):2544­5. toral Competition in Established Democracies since 1945. Cam- Ferreira, Francisco H. G., and Peter Lanjouw. 2001."Rural Non- bridge, United Kingdom: Cambridge University Press. farm Activities and Poverty in the Brazilian Northeast." World Freedom House. 2006. Freedom in the World 2005 : The Annual Development 29(3):509­28. Survey of Political Rights and Civil Liberties. Long Beach, CA: Fertig, Michael. 2003."Education Production, Endogenous Peer Rowman & Littlefield Publishers, Inc. Group Formation and Class Composition - Evidence from PISA Freeman, Richard B. 2000."Disadvantaged Young Men and 2002 Study." Bonn: Institute for the Study of Labor (IZA), Dis- Crime." In David G. Blanchflower and Richard B. Freeman, cussion Paper Series 714. (eds.), Youth Employment and Joblessness in Advanced Coun- Fichtenberg, Caroline M., and Stanton A. Glantz. 2002."Youth tries. Chicago and London: University of Chicago Press for the Access Interventions Do Not Sffect Youth Smoking." Pediatrics National Bureau of Economic Research. 109(6):1088­92. Fretwell, David H., Jacob Benus, and Christopher J. O'Leary. 1999. Fidrmuc, Jan, and Orla Doyle. 2005. Does Where You Live Affect "Evaluating the Impact of Active Labour Market Programme: How You Vote: An Analysis of Migrant Voting Behaviour. Results of Cross-Country Studies in Europe and Central Asia." Uxbridge: Brunel Business School. Washington, DC: World Bank, Social Protection Discussion Filer, Randall K., and Daniel Münich. 2002."Responses of Private Paper 9915. and Public Schools to Voucher Funding." The Center for Eco- Fryer, Roland G., and Paul Torelli. 2005."An Empirical Analysis of nomic Research and Graduate Education - Economic Institute. `Acting White'." Harvard University. Cambridge, MA. Processed. Prague. Processed. Fuchs, Thomas, and Ludger Woessmann. 2004."What Accounts Filmer, Deon. Forthcoming."School Availability and School Par- for International Differences in Student Performance? A Re- ticipation in 21 Developing Countries." Journal of Development Examination Using PISA Data." Munich, Germany: CESifo Studies. Working Paper Series 1235. Filmer, Deon, Emmanuel Jimenez, and Annette Richter. 2006. Furlong, Andy, Fred Cartmel, Janet Powney, and Stuart Hall. 1997. "Simulating the Returns to Youth Opportunity Agency and Sec- Evaluating Youth Work with Vulnerable Young People. Edinburgh: ond Chances." Background note for the WDR 2007. Scottish Council for Research in Education. Filmer, Deon, and David L. Lindauer. 2001."Does Indonesian Fussell, Elizabeth. 2006."Comparative Adolescences: The Transi- Have a `Low Pay' Civil Service?" Bulletin of Indonesian Economic tion to Adulthood in Brazil, Kenya, Mexico, the U.S., and Viet- Studies 37(2):189­205. nam." University of Tennessee. Knoxville, TN. Processed. Finkel, Steve E, and Sheryl Strumbas. 2000. Civic Education in Gajakashmi, C. K., P. Jha, K. Ranson, and S. Nguyen. 2000."Global South Africa: The Impact of Adult and School Programs on Demo- Patterns of Smoking and Smoking-Attributable Mortality." In cratic Attitudes and Participation. Washington, DC: USAID. Prabhat Jha and Frank J. Chaloupka, (eds.), Tobacco Control in Available online at http://pdf.dec.org/pdf_docs/PNACH577.pdf. Developing Countries. New York: Oxford University Press. Fiore, M. C., W. C. Bailey, S. J. Cohen, S. F. Dorfman, M. G. Gold- Gal, Reuven, and Donald Eberly. 2006. Service Without Guns. New stein, E. R. Gritz, R. B. Heyman, C. R. Jaen, T. E. Kottke, H. A. Zealand: Lulu Press. Lando, R. E. Mecklenburg, P. D. Mullen, L. M. Nett, L. Robinson, Galasso, Emanuela, Martin Ravallion, and Agustin Salvia. 2001. M. L. Stitzer, A. C. Tommasello, L. Villejo, and M. E. Wewers. "Assisting the Transition from Workfare to Work: A Random- 252 WORLD DEVELOPMENT REPORT 2007 ized Experiment." Washington, DC: World Bank Policy Research Gibson, Cynthia. 2001."From Inspiration to Participation: A Working Paper Series 2738. Review of Perspectives on Youth Civic Engagement." Paper pre- Galiani, Sebastian, Martin Rossi, and Ernesto Schargrodsky. 2006. sented at the Grantmaker Forum on Community and National "Conscription and Crime." Background paper for the WDR Service. Berkeley, CA 2007. Gibson, John, David J. McKenzie, and Halahingano Rohorua. Galloway, Rae, and Mary Ann Anderson. 1994."Prepregnancy 2006."How Cost Elastic Are Remittances? Evidence from Nutritional Status and Its Impact on Birthweight." SCN News Tongan Migrants in New Zealand." Pacific Economic Review 1994(11):6­10. 21(1):112­38. Garces, Eliana, Duncan Thomas, and Janet Currie. 2000."Longer Gill, Indermit S., and Ih Chon-Sun. 2000."Republic of Korea." In Term Effects of Head Start." Cambridge, MA: National Bureau Indermit S. Gill, Fred Fluitman, and Amit Dar, (eds.), Vocational of Economic Research Working Paper Series 8054. Education and Training Reform: Matching Skills to Markets and Budgets. New York: Oxford Univesity Press. Gatawa, B. G. 1995. Zimbabwe: AIDS Education for Schools. Harare, Zimbabwe: UNICEF. Gill, Indermit S., Claudio E. Montenegro, and Dörte Dömeland, eds. 2002. Crafting Labor Policy: Techniques and Lessons from Gauri, Varun, and Evan S. Lieberman. 2004."Institutions, Social Latin America. Washington, DC: Oxford University Press for the Boundaries, and Epidemics: Explaining Government AIDS Poli- World Bank. cies in Brazil and South Africa." Paper presented at the Annual Meetings of the American Political Science Association. Septem- Ginifer, Jeremy. 2003."Reintegration of Ex-Combatants." In Mark ber 2. Chicago, IL. Malan, Sarah Meek, Thokozani Thusi, and Jeremy Ginifer, (eds.), Sierra Leone: Building the Road to Recovery. Pretoria, Gauthier, Clermont, and Martial Dembélé. 2004. Quality of Teach- South Africa: Institute for Securities Studies. ing and Quality of Education: A Review of Research Findings. Paris: Background paper for the UNESCO Global Monitoring Glennerster, Rachel, Shehla Imran, and Katherine Whiteside. 2006. Report 2005. Available online at http://portal.unesco.org/. "Baseline Report on the Quality of Primary Education in Sierra Geary, Cindy W., Hally Mahler, William Finger, and Kathleen H. Leone: Teacher Absence, School Inputs, and School Supervision Shears. 2005. Using Global Media to Reach Youth: The 2002 MTV in District Education Committee and Government-Assisted Staying Alive Campaign. Arlington, VA: Family Health Interna- Schools in Sierra Leone." Processed. tional, YouthNet Program. Glewwe, Paul, Hanan G. Jacoby, and Elizabeth M. King. 2001. Gebrekristos, Hirut, Stephen Resch, Khangelani Zuma, and Mark "Early Childhood Nutrition and Academic Achievement: A Lon- Lurie. 2005."Estimating the Impact of Establishing Family gitudinal Analysis." Journal of Public Economics 81(3):345­68. Housing on the Annual Risk of HIV Infection in the South Gluckman, Peter D., Mark A. Hanson, Susan M. B. Morton, and African Mining Communities." Sexually Transmitted Diseases Catherine S. Pinal. 2005."Life-Long Echoes - A Critical Analysis 32(6):333­40. of the Developmental Origins of Adult Disease Model." Biology Gerber, Alan S., Donald P. Green, and Ron Shachar. 2003."Vot- of the Neonate 87:127­39. ing May be Habit Forming: Evidence from a Randomized Field Goldscheider, Frances K. 2000."Why Study Young Adult Living Experiment." American Journal of Political Science 47(3):540­50. Arrangements? A View of the Second Demographic Transition." German Technical Cooperation, and International Council on Brown University. Providence, RI. Processed. National Youth Policy. 2005. Comparative Analysis of National Gonzalez, Rosa Amelia, and Patricia Marquez. 2005. Ron Santa Youth Policy. Eschborn, Germany: International Council on Teresa's Social Initiatives. Washington, DC: Inter-American National Youth Policy (ICNYP). Available online at http://www. Development Bank, Social Enterprise Knowledge Network. icnyp.net. Goovaerts, Piet, Martin Gasser, and Alisa Belman Inbal. 2005. Geronimus, Arline T., and John Bound. 1990."Black/white Differ- "Demand Driven Approaches to Livelihood Support in Post- ences in Women's Reproductive-Related Health Status: Evidence War Contexts." Washington, DC: World Bank, Social Develop- from Vital Statistics." Demography 27(3):457­66. ment Papers 29. Geronimus, Arline T., and Sanders Korenman. 1992."The Socio- Gorpe, Mehmet Ziya, Jean Fares, and Claudio E. Montenegro. economic Consequences of Teen Childbearing Reconsidered." 2006."Public Wage Premium Around the World." Background Quarterly Journal of Economics 107(4):1187­214. paper for the WDR 2007. Gertler, Paul J. 2000. Final Report: The Impact of PROGRESA on Gough, Neil, and Charlotte Grezo. 2005."Africa: The Impact of Health. Washington, DC: International Food Policy Research Mobile Phones." Berkshire, England: Vodafone Policy Paper Institute. Series 2. Gertler, Paul J., Stefano F. Bertozzi, Juan Pablo Gutierrez, and J. Gould, Eric D., Bruce A. Weinberg, and David B. Mustard. 1998. Sturdy. 2006."Preliminary Results from Analysis of Poverty and "Crime Rates and Local Labor Market Opportunities in the Adolescent Risk Behavior in Mexico." World Bank. Washington, United States: 1979-1997." Review of Economic and Statistics DC. Processed. 84(1):45­61. Gertler, Paul J., David Levine, and Sebastian Martinez. 2003."The Government of Sierra Leone. 2002. Survey on Reinsertion & Rein- Presence and Presents of Parents: Do Parents Matter for More tegration Assistance to Ex-Combatants. Freetown, Sierra Leone: than their Money?" University of California. Berkeley, CA Avail- Government of Sierra Leone, report submitted to National Cen- able online at http://faculty.haas.berkeley.edu/. Processed. ter for the Dissemination of Disability Research. Gian, Cong, Doan Hong Quang, Nguyen Thi Lan Huong, and ------. 2005a."National Population Based HIV Seroprevalence Remco H. Oostendorp. 2006."Trade Liberalization, the Gender Survey of Sierra Leone". Freetown, Sierra Leone, Statistics Sierra Wage Gap and Returns to Education in Vietnam." Tinbergen Leone. Institute, Amsterdam Institute for International Development. ------. 2005b. Poverty Reduction Strategy Paper: A National Pro- Amsterdam. Processed. gramme for Food Security, Job Creation and Good Governance References 253 (2005-2007). Freetown, Sierra Leone: Government of Sierra Gruber, Jonathan, and Jonathan Zinman. 2001."Youth Smoking in Leone. the United States: Evidence and Implications." In Jonathan Gru- Graft, Auralice, Nicole Haberland, and Rachel Goldberg. 2003. ber, (eds.), Risk Behavior among Youths: an Economic Analysis. "Married Adolescents: A Review of Programs." Paper presented Chicago: Chicago University Press. at the WHO/UNFPA/Population Council Technical Consulta- Guarcello, Lorenzo, Fabrizia Mealli, and Furio Camillo Rosati. 2003. tion on Married Adolescents. Geneva. "Household Vulnerability and Child Labour: The Effects of Grantham-Mcgregor, S. M., C. A. Powell, S. P. Walker, and J. H. Shocks, Credit Rationing and Insurance."Washington, DC: World Himes. 1991."Nutritional Supplementation, Psychosocial Bank, Social Protection Unit, Human Development Network 0322. Stimulation, and Mental Development of Stunted Children: The Guermazi, Boutheina, and David Satola. 2005."Creating the `Right' Jamaica Study." Lancet 338(8758):1­5. Enabling Environment for ICT." In Robert Schware, (eds.), E- Gray, Ronald H., Maria J. Wawer, Ron Brookmeyer, Nelson K. development: from excitement to effectiveness. Washington, DC: Sewankambo, David Serwadda, Fred Wabwire-Mangen, Tom World Bank. Lutalo, Xianbin Li, Thomas van Cott, Thomas C. Quinn, and Gunnarson, Victoria, Peter F. Orazem, and Mario A. Sánchez. 2006. Rekai Project Team. 2001."Probability of HIV-1 Transmission "Child Labor and School Achievement in Latin America." World per Coital Act in Monogamous, Heterosexual, HIV-1-discordant Bank Economic Review 20(1):31­54. Couples in Rakai, Uganda." Lancet 357(9263):1149­53. Gutierrez, Juan Pablo. 2006."Oportunidades for the Next Genera- Green, Donald P., and Alan S. Gerber. 2001. Getting out the Youth tion: Effects of a Conditional Cash Transfer on the Wellbeing of Vote: Results from Randomized Field Experiments. Youth Vote Mexican Youths." Background paper for the WDR 2007. Coalition: Pew Charitable Trusts. Haan, Hans Christiaan, and Nicholas Serriere. 2002. Train- Greene, Margaret E., and Thomas Merrick. 2005."Poverty Reduc- ing for Work in the Informal Sector: Fresh Evidence from West tion: Does Reproductive Health Matter?" Washington, DC: and Central Africa. Turin: International Training Centre of World Bank Health, Nutrition and Population Discussion Paper the International Labour Organization. Available online July 2005. at http://siteresources.worldbank.org/INTLM/214578- Greenwood, Jeremy, Ananth Seshadri, and Guillaume Vandenb- 1103217503703/20295542/TrainingforWorkWCA.pdf. roucke. 2005."The Baby Boom and Baby Bust." American Eco- Haggblade, Steven, Peter B. R. Hazell, and Thomas Reardon. (eds.) nomic Review 95(1):183­207. Forthcoming."Transforming the Rural Nonfarm Economy." Balti- Greenwood, Peter W., Karyn E. Model, C. Peter Rydell, and James more, MD: John Hopkins University Press. Chiesa. 1998. Diverting Children from a Life of Crime: Measuring Hagopian, Amy, Anthony Ofosu, Adesegun Fatusi, Richard Birit- Costs and Benefits. Santa Monica, CA: RAND. wum, Ama Essel, L. Gary Hart, and Carolyn Watts. 2005."The Gregson, Simon, Geoffrey P. Garnett, Constance A. Nyamukapa, Flight of Physicians from West Africa: Views of African Physi- Timothy B. Hallett, James J. C. Lewis, Peter R. Mason, Stephen cians and Implications for Policy." Social Science & Medicine K. Chandiwana, and Roy M. Anderson. 2006."HIV Decline 61(8):1750­60. Associated with Behavior Change in Eastern Zimbabwe." Science Hahn, Andrew. 1999."Extending the Time of Learning." In Doug- 311(5761):664­66. las J. Besharov, (eds.), Ameruca's Disconnected Youth: Towards a Groce, Nora Ellen. 2003."Adolescents and Youth with Disability: Preventive Strategy. Washington, DC: Child Welfare League of Issues and Challenges." PhD thesis. Yale School of Public Health. America Inc. Grogger, Jeffrey. 1998."Market Wages and Youth Crime." Journal of Hahn, Andrew, Susan Lanspery, and Tom Leavitt. 2005. Docu- Labor Economics 16(4):756­91. mentation of Outcomes in the Philippines Make A Connection Grosskurth, Heiner, Frank Mosha, James Todd, Ezra Mwijarubi, Program. Baltimore, MD: International Youth Foundation and Arnoud Klokke, Kesheni Senkoro, Philippe Mayaud, John Chan- Nokia Corporation. galucha, Angus Nicoll, Gina ka Gina, James Newell, Kokugonza Hahn, Andrew, and Tom Leavitt. 2003. Joined-Up Government. Mugeye, David Mabye, and Richard Hayes. 1995."Impact of Coordination and Collaboration Opportunities to Strengthen Improved Treatment of Sexually Transmitted Diseases on HIV Multi-Sectoral Youth Policy Implementation in Jamaica. Washing- Infection in Rural Tanzania: Randomised Controlled Trial." ton, DC: World Bank. Lancet 346(8974):530­36. Hahn, Andrew, Tom Leavitt, and Paul Aaron. 1994. Evalua- Grossman, Michael. 2005a."Education and Nonmarket Out- tion of the Quantum Opportunities Program: Did the Program comes." Cambridge, MA: National Bureau of Economic Work? Waltham, MA: Brandeis University. Available online at Research Working Paper Series 11582. http://eric.ed.gov/ERICDocs/data/ericdocs2/content_storage_ ------. 2005b."Individual Behaviors and Substance Use: The Role 01/0000000b/80/27/39/7a.pdf. of Price." In Björn Lindgren and Michael Grossman, (eds.), Sub- Hahn, Carole L, Paulette Patterson Dilworth, and Michael stance Use: Individual, Behavioral, Social Interactions, Markets, Hughes. 1998. IEA Civic Education Project, Phase 1, The United and Policies. Amsterdam: Elsevier. States - A Review of Literature, Volume 1. Washington, DC: Grown, Caren, Geeta Rao Gupta, and Aslihan Kes. 2005. Taking International Association for the Evaluation of Educational Action: Achieving Gender Equality and Empowering Women. Achievement. Sterling, VA: Earthscan. Hahn, Carole L. 2005."School Influences and Civic Engagement." Gruben, William C., and Darryl McLeod. 2006."Apparel Exports In Lonnie R. Sherrod, Constance A. Flanagan, Ron Kassimir, and and Education: How Developing Nations Encourage Women's Amy K. Syvertsen, (eds.), Youth Activism: An International Ency- Schooling." Economic Letter - Federal Reserve Bank of Dallas clopedia. Westport, CT: Greenwood Publishing Group. 1(3):1­8. Halewood, Naomi, and Charles Kenny. 2006."Young People and Gruber, Jonathan, eds. 2001. Risky Behavior among Youths: An Eco- Communications Technologies." Background paper for the nomic Analysis. Chicago: University of Chicago Press. WDR 2007. 254 WORLD DEVELOPMENT REPORT 2007 Hallman, Kelly, Agnes R. Quisumbing, Marie Ruel, and Bénédicte Heckman, James J., Jora Stixrud, and Sergio Urzua. 2006."The de la Briere. 2005."Mothers' Work and Child Care: Findings Effects of Cognitive and Noncognitive Abilities on Labor Mar- from the Urban Slums of Guatemala City." Economic Develop- ket Outcomes and Social Behavior." Cambridge, MA: National ment and Cultural Change 53(4):855­86. Bureau of Economic Research Working Paper Series 12006. Halpern-Felsher, Bonnie L., and Elizabeth Cauffman. 2001."Costs Hediger, M. L., T. O. Scholl, J. I. Schall, and P. M. Krueger. 1997. and Benefits of a Decision: Decision-making Competence in "Young Maternal Age and Preterm Labour." Annals of Epidemi- Adolescents and Adults." Journal of Applied Developmental Psy- ology 7(6):400­6. chology 22(3):257­73. Hegre, Håvard. 2003."Disentangling Democracy and Develop- Hannam, Derry. 2001. A Pilot Study to Evaluate the Impact of Stu- ment as Determinants of Armed Conflict." Paper presented at dent Participation Aspects of the Citizenship Order on Standards the 44th Annual Convention of the International Studies Asso- of Education in Secondary Schools. London, U.K.: DFEE. Avail- ciation. February 25. Portland, O.R. able online at http://www.csv.org.uk/csv/hannamreport.pdf. Hemmer, Hans R., and C. Mannel. 1989."On the Economic Hanushek, Eric A., and Dennis D. Kimko. 2000."Schooling, Labor Analysis of the Urban Informal Sector." World Development Force Quality, and the Growth of Nations." American Economic 17(10):1543­52. Review 90(5):1184­208. Hettige, S. T., Markus Mayer, and Maleeka Salih. 2004."School- Hanushek, Eric A., and Ludger Wößmann. 2005."Does Educa- to-work Transition of Youth in Sri Lanka." Colombo: University tional Tracking Affect Performance and Inequality? Differences- of Colombo, Employment Policies Unit, Employment Strategy in-Differences Evidence Across Countries." Cambridge, MA: Department, Employment Strategy Paper 2004/19. Available National Bureau of Economic Research Working Paper Series online at http://www.ilo.org/public/english/employment/strat/ 11124. download/esp19.pdf. Harrington, Julia. 2005."Voiding Human Rights: Citizenship and Hibbell, Björn, Barboro Anderson, Salme Ahlström, Olga Bala- Discrimination in Africa." Justice Initiatives: Human Rights and kireva, Thoroddur Bjarnason, Anna Kokkevi, and Mark Mor- Justice Sector Reform in Africa February:23­28. gan. 2000. The 1999 ESPAD Report: Alcohol and Other Drug Hart, Jason. 2004. Children's Participation in Humanitarian Action: Use Among Students in 30 European Countries. Stockholm: The Learning from Zones of Armed Conflict. Oxford: Refugees Studies Swedish Council for Information on Alcohol and Other Drugs Center. (CAN) and the Pompidou Group at the Council of Europe. Hatton, Timothy J., and Jeffrey G. Williamson. 2003."Demo- Hildebrandt, Nicole, and David J. McKenzie. 2005."The Effects of graphic and Economic Pressure on Emigration out of Africa." Migration on Child Health in Mexico." Economia 6(1):257­89. Scandinavian Journal of Economics 105(3):465­486. Hirsch, Barry T., and Stephen L. Mehay. 2003."Evaluating the ------. 2005."What Fundamentals Drive World Migration?" In Labor Market Performance of Veterans Using a Matched George J. Borjas and Jeff Crips, (eds.), Poverty, International Comparison Group Design." Journal of Human Resources Migration and Asylum. New York, NY: Palgrave MacMillan. 38(3):673­700. Hazell, Peter B. R., and Steven Haggblade. 1993."Farm-Nonfarm Hodgkin, Dominic. 1996."Household Characteristics Affecting Where Growth Linkages and the Welfare of the Poor." In Michael Lip- Mothers Deliver in Rural Kenya."Health Economics 5(4):333­40. ton and Jacques Van der Gaag, (eds.), Including the Poor. Wash- Hoff, Karla, and Priyanka Pandey. 2004."Belief Systems and Dura- ington, DC: World Bank. ble Inequalities: An Experimental Investigation of Indian Caste." Heady, Christopher. 2003."What is the Effect of Child Labour on Washington, DC: World Bank Policy Research Working Paper Learning Achievement? Evidence from Ghana." Florence, Italy: Series 3351. Innocenti Working Papers inwopa00/7. Hofferth, Sandra L., and Lori Reid. 2001."The Effects of Early Health and Treasury Task Team. 2003. Full Report of the Joint Childbearing on Schooling Over Time." Family Planning Per- Health and Treasury Task Team Charged with Examining Treat- spectives, 33(6):259­67. ment Options to Supplement Comprehensive Care for HIV/AIDS Hoffman, Saul D. 1998."Teenage Childbearing Is Not So Bad After in the Public Health Sector. South Africa: Ministry of Health & All. Or is it? A Review of the New Literature." Family Planning Ministry of Treasury. Perspectives 30(5):236­39. Heckman, James J., Robert J. LaLonde, and Jeffrey A. Smith. 1999. Holzmann, Robert. 2005."Demographic Alternatives for Aging "The Economics and Econometrics of Active Labor Market Pro- Industrial Societies: Enhanced Immigration, Labor Force Partic- grams." In Orley Ashenfelter and David Card, (eds.), Handbook ipation, or Total Fertility." Washington, DC: World Bank, Social of Labor Economics, vol. 3A. Amsterdam: North-Holland. Protection Discussion Paper 0540. Heckman, James J., and Xuesong Li. 2004."Selection Bias, Com- Hooghe, Marc, Deitlind Stolle, and Patrick Stouthuysen. 2004. parative Advantage and Heterogeneous Returns to Educa- "Head Start in Politics: The Recruitment Function of Youth tion: Evidence from China in 2000." Pacific Economic Review Organizations of Political Parties in Belgium (Flanders)." Party 9(3):155­71. Politics 10(2):193­212. Heckman, James J., and Carmen Pagés. 2000."Regulation and Hopenhayn, Hugo, and Richard Rogerson. 1993."Job Turnover Deregulation: Lessons from Latin American Labor Markets." and Policy Evaluation: A General Equilibrium Analysis." Journal Economía 1(1):123­45. of Political Economy 101(5):915­38. Heckman, James J., Jeffrey A. Smith, and Nancy Clements. 1997. Howell, James C. 1998."Abolish the Juvenile Court? Nonsense!" "Making the Most out of Social Experiments: The Intrinsic Juvenile Justice Update 1:1­13. Uncertainty in Evidence from Randomized Trials with an Appli- Hubley, J. 2000. Interventions Targeted at Youth Aimed at Influenc- cation to the National JTPA Experiment." Review of Economic ing Sexual Behavior and AIDS/STDs. Leeds, U.K.: Leeds Health Studies 64(4):487­535. Education Database. References 255 Hubner, John. 2005. Last Chance in Texas: The Redemption of Imbens, Guido, and Wilbert Van Der Klaauw. 1995."Evaluating Criminal Youth. New York, NY: Random House. the Cost of Conscription in The Netherlands." Journal of Busi- Hudson, Valerie M., and Andrea M. den Boer. 2004. Bare Branches: ness and Economic Statistics 13(2):207­215. The Security Implications of Asia's Surplus Male Population. Inglehart, Ronald, Miguel Basáñez, Jaime Díez-Medrano, Loek Cambridge, MA: The MIT Press. Halman, and Ruud Luijkx, eds. 2004. Human Beliefs and Values: Huffman, W. E., and Peter F. Orazem. 2004."Agriculture and A Cross-cultural Sourcebook based on the 1999-2002 Values Sur- Human Capital in Economic Growth: Farmers, Schooling and veys. Mexico, D.F.: Siglo Veinteiuno. Health." Ames, Iowa: Iowa State University, Economics Working Institute for Health Management-Pachod (IHMP), and Interna- Papers 04016. tional Center for Research On Women (ICRW). 2003. Increasing Human Rights Watch. 1997a. Juvenile Justice: Police Abuse and Low Age at Marriage in Rural Maharashtra, India. Maharashtra, Detention of Street Children in Kenya. New York, NY: Human India: Institute for Health Management. Available online at Rights Watch. http://www2.phishare.org/files/1046_IHMP_marriagebrief_ ------. 1997b. The Scars of Death: Children Abducted by the Lord's Dec2003.pdf. Resistance Army in Uganda. New York, NY: Human Rights Institute of Alcohol Studies. 2006. Adolescents and Alcohol. St. Ives: Watch. Institute of Alcohol Studies. ------. 2001a. Easy Targets: Violence Against Children Worldwide. Instituto Mexicano de la Juventud (IMJ), and Organización New York, NY: Human Rights Watch. Iberoamericana de la Juventud (OIJ). 2006. Institucionalidad y ------. 2001b. Scared at School: Sexual Violence against Girls in Políticas Públicas de Juventud. Mexico City: Instituto Mexicano South African Schools. New York, NY: Human Rigths Watch. de la Juventud por encargo y con la colaboración de la Orga- Available online at http://www.hrw.org/reports/2001/safrica/. nización Iberoamericana de la Juventud. ------. 2002a. My Gun was as Tall as Me. New York, NY: Human Instituto Nacional para la Evaluación de la Educación de Mexico. Rights Watch. 2005."Las Telesecundarias Mexicanas: Un Recorrido sin Atajos." Online Magazine. ------. 2002b. Suffering in Silence: The Links Between Human Rights Abuses and HIV Transmission to Girls in Zambia. New Inter-American Development Bank. 2005. Ex-post Evaluation of York, NY: Human Rights Watch. Training Programs `Youth Labor Training Program' (PROJOVEN) in PERU. Ex-Post Project Report. Preliminary version. Washing- ------. 2003. Charged with Being Children: Egyptian Police Abuse ton, DC: Inter-American Development Bank. Available online at of Children in Need of Protection. New York, NY: Human Rights http://www.iadb.org/ove/Documents/uploads/cache/498982.pdf. Watch. Interagency Gender Working Group. 2005. An Education in Mak- Humphreys, Macartan, and Paul Richards. 2005."Prospects ing Schools Safe. Washington, DC: Interagency Gender Working and Opportunities for Achieving the MDGs in Post-Conflict Group. Available online at http://www.igwg.org/articles/safe- Countries: A Case Study of Sierra Leone and Liberia." NEPAD. schools.htm. New York, NY Available online at http://www.columbia.edu/ ~mh2245/papers1/HR.pdf. Processed. International Center for Prison Studies. 2003."World Prison Brief". London, UK, King's College London, University of Lon- Humphreys, Macartan, and Jeremy M. Weinstein. 2003. What the don. Available online at http://www.kcl.ac.uk/depsta/rel/icps/ Fighters Say: A Survey of Ex-Combatants in Sierra Leone June- worldbrief/world_brief.html. August 2003. Freetown, Sierra Leone: The Post-Conflict Reinte- gration Initiative for development and Empowerment (PRIDE). International Council on National Youth Policy. 2005. ICNYP Pro- files on Country Actions to Promote Implementation of National -------. 2005."Disentangling the Determinants of Successful Youth Policies. Vienna: International Council on National Youth Disarmament, Demobilization, and Reintegration." Paper pre- Policy (ICNYP). Available online at http://www.icnyp.net. sented at the 101st Meeting of the American Political Science International Institute for Population Sciences (IIPS), and ORC Association. September. Washington, DC Macro. 2000. National Family Health Survey (NFHS-2), 1998-99, Hurley, Susan F., Damien J. Jolley, and John M. Kaldor. 1997. India. Mumbai: IIPS. "Effectiveness of Needle Exchange Programmes for Prevention International Nutritional Anemia Consultative Group. 1979. Iron of HIV Infection." Lancet 349(9068):1797­800. Deficiency in Infancy and Childhood. Geneva: World Health Ilahi, Nadeem, Peter F. Orazem, and Guilherme Sedlacek. 2005. Organization. "How Does Working as a Child Affect Wages, Income and Pov- ------. 1989. Iron Deficiency in Women. Geneva: World Health erty as an Adult?" Washington, DC: World Bank, Social Protec- Organization. tion Discussion Paper Series 0514. International Organization for Migration (IOM). 2005a. HIV/ ILO. 2001. World Employment Report 2001: Life at Work in the AIDS, Population Mobility and Migration in Southern Africa: Information Economy. Geneva: International Labour Office. Defining a Research and Policy Agenda. South Africa: IOM ------. 2003. Wounded Childhood: The Use of Children in Armed South Africa Office. Available online at http://www.iom.org.za/ Conflict in Central Africa. Geneva: International Labour Office. Reports/PopulationMobilityReport.pdf. ------. 2004. Global Employment Trends for Youth. Geneva: Inter- ------. 2005b. World Migration 2005: Costs and Benefits of Inter- national Labour Office. national Migration. Geneva: International Organization for ------. 2005. Global Youth Employment Trends. Geneva: Interna- Migration. tional Labour Office. International Telecommunication Union (ITU). 2005. The African ------. 2006. The End of Child Labour: Within Reach. Geneva: Virtual University. Geneva: International Telecommunication International Labour Office. Available online at http://www.ilo. Union. Available online at http://www.itu.int/osg/spu/wsis- org/public/english/standards/relm/ilc/ilc95/pdf/rep-i-b.pdf. themes/ict_stories/Themes/Case_studies/AVU.html . 256 WORLD DEVELOPMENT REPORT 2007 Internet Safety Group. 2005. New Survey Illustrate the Profound Moses, Grace John, Ruth Nduati, Kelly MacDonald, and Seth Impact of Mobile Phones on Many New Zealand Youth. Auckland, Berkley. 2001."The Evidence Base for Interventions to Prevent New Zealand: Internet Safety Group. Available online at http:// HIV Infection in Low and Middle-Income Countries." Geneva: www.netsafe.org.nz/isgnews/text_generation.aspx. Commission on Macroeconomics and Health Working Paper Isham, Jonathan, Daniel Kaufmann, and Lant Pritchett. 1995. Series WG 5 Paper 2. "Governance and Project Performance." Washington, DC: World Jimenez, Emmanuel. 1995."Human and Physical Infrastructure: Bank Policy Research Working Paper Series 1550. Public Investment and Pricing Policies in Developing Coun- Iversen, Vegard. 2002."Autonomy of Child Labor Migrants." World tries." In J. R. Behrman and T. N. Srinivasan, (eds.), Handbook of Development 30(5):817­34. Development Economics vol. 3. Amsterdarm: Elsevier. Jacob, Brian A., and Lars Lefgren. 2002."The Impact of Teacher Train- Jimenez, Emmanuel, and Marlaine E. Lockheed. 1989."Enhanch- ing on Student Achievement: Quasi-Experimental Evidence from ing Girls' Learning Through Single-Sex Education: Evidence School Reform Efforts in Chicago."Cambridge, MA: National and a Policy Conundrum." Educational Evaluation and Policy Bureau of Economic Research Working Paper Series 8916. Analysis 11(2):117­42. Jacobson, Mireille. 2004."Baby Booms and Drug Busts: Trends Jimenez, Emmanuel, and Yasuyuki Sawada. 1999."Do Commu- in Youth Drug Use in the United States: 1975-2000." Quarterly nity-Managed Schools Work? An Evaluation of El Salvador's Journal of Economics 119(4):1481­512. EDUCO Program." World Bank Economic Review 13(3):415­41. Jacoby, Hanan G., and Emannuel Skoufias. 2002."Financial Con- Jimeno, Juan F., and Diego Rodríguez-Palenzuela. 2002."Youth straints on Higher Education: Evidence from Mexico." World Unemployment in the OECD: Demographic Shifts, Labour Bank and IFPRI. Washington, DC. Processed. Market Institutions, and Macroeconomic Shocks." Frankfurt: European Central Bank, Working Paper Series 155. James-Traore, Tijuana, Robert Magnani, Nancy Murray, Judith Joekes, Susan. 1995."Trade- Related Employment for Women in Senderowitz, Ilene Speizer, and Lindsay Steward. 2002."Inter- Industry and Services in Developing Countries." New York, NY: vention Strategies that Work for Youth: Summary of the FOCUS United Nations Development Program, Occasional Paper 5. on Young Adults End of Program Report." Arlington, VA: Family Available online at http://www.unrisd.org/. Health International, YouthNet Program, Youth Issues Paper 1. Available online at http://www.fhi.org/. Johanson, Richard K., and Arvil V. Adams. 2004. Skills Develop- ment in Sub-Saharan Africa. Washington, DC: World Bank. Japan Bank for International Cooperation. 2006."Meeting the Increasing and Changing Demand for Quality Labor in FDI-led Johnson, Amy W. 1996."An Evaluation of the Long-Term Impacts Growth: From Adaptability to Creativity." Japan Bank for Inter- of the Sponsor-A-Scholar Program on Student Performance." national Cooperation (JBIC). Tokyo. Processed. Mathematica Policy Research, Inc. Princeton, NJ. Processed. Javate de Dios, Aurora. 2002."From the Margins to the Main- Johnson, D. Gale. 2000."Population, Food, and Knowledge." Amer- stream: Making Government Work for Gender Equality." Paper ican Economic Review 90(1):1­14. presented at the ASEAN-World Bank High Level Conference on Jorgenson, Dale W. Forthcoming."Accounting for Growth in the Social Development in the National Development. January 16. Information Age." In Philippe Aghion and Steven Durlauf (eds.) Jakarta. Handbook of Economic Growth. Amsterdam: North-Holland. Jejeebhoy, Shireen J., and Sarah Bott. 2003."Non-consensual Sex- Joshi, Shareen, and Paul Schultz. 2005."Family Planning as an ual Experiences of Young People: A Review of the Evidence from Investment in Female Human Capital: Evaluating the Long Developing Countries." New Delhi: Populaiton Council South & Term Consequences in Matlab, Bangladesh."Yale University. East Asia Regional Working Papers 16. New Haven, CT. Processed. Jennings, Kent M, and Laura Stoker. 2002."Generational Change, Jowett, Matthew. 2000."Safe Motherhood Interventions in Low- Life Cycle Processes, and Social Capital." Paper presented at the income Countries: An Economic Justification and Evidence of Workshop on "Citizenship on Trial: Interdisciplinary Perspec- Cost Effectiveness." Health Policy 53(3):201­28. tives on the Political Socialization of Adolescents". June 20. Kabbani, Nader, and Ekta Kothari. 2005."Youth Employment in Montreal, Canada. the MENA Region: A Situational Assessment." Washington, DC: Jensen, Robert. 2006."Do the Perceived Returns to Education World Bank SP Discussion Paper 0534. Affect Schooling Decisions? Evidence from a Randomized Kabeer, Naila. 1999."Resources, Agency, Achievements: Reflections Experiment." John F. Kennedy School of Government, Harvard on the Measurement of Women's Empowerment." Development University. Cambridge, MA. Processed. and Change 30(3):435­64. Jensen, Robert, and Rebecca Thornton. 2003."Early Female Mar- Kagitcibasi, Cigdem, Diane Sunar, and Sevda Bekman. 2001. riage in the Developing World." Gender and Development "Long-term Effects of Early Intervention: Turkish Low-Income 11(2):9­19. Mothers and Children." Applied Developmental Pshychology Jha, Prabhat, and Frank Chaloupka, eds. 2000. Tobacco Control in 22:333­61. Developing Countries. New York: Oxford University Press. Kalemli-Ozcan, Sebnem. 2001."The Effect of Mortality on Fertil- Jha, Prabhat, Frank J. Chaloupka, James Moore, Vendhan Gaja- ity and Human Capital Investment: What Do We Learn from lakshmi, Prakash C. Gupta, Richard Peck, Samira Asma, and AIDS?" University of Houston. Houston. Processed. Witold Zatonski. 2005."Tobacco Addiction." In Dean T. Jamis- ------. 2005."AIDS, Reversal of the Demographic Transition and ion, David B. Evans, George Alleyne, Prabhat Jha, Joel G. Bre- Economic Development: Evidence from Africa." University of man, Anthony R. Measham, Mariam Claeson, Anne Mills, and Houston and NBER. Houston. Processed. Philip R. Musgrove, (eds.), Disease Control Priorities in Develop- Kandel, William, and Grace Kao. 2001."The Impact of Tem- ing Countries. Bethesda, MD: Disease Control Priorities Project. porary Labor Migration on Mexican Children's Educational Jha, Prabhat, Lara M. E. Vaz, Francis A. Plummer, Nico J. D. Aspirations and Performance." International Migration Review Nagelkerke, Bridget Willbond, Elizabeth N. Ngugi, Stephen 35(4):1205­1231. References 257 Kantis, H., eds. 2003. Estudios de Desarrollo Empresarial en Argen- ington, DC: National Campaign to Prevent Teen Pregnancy. tina: La Creación de Empresas y su Entorno Institucional. Buenos Available online at http://www.etr.org/recapp/theories/RiskPro- Aires: JICA. tectiveFactors/Matrix_Kirby.pdf. Kassouf, Ana Lucia, Martin McKee, and Elias Mossialos. 2001."Early Kleiner, Morris M. 2000."Occupational Licensing." Journal of Eco- Entrance to the Job Market and its Effect on Adult Health: Evi- nomics Perspectives 14(4):189­202. dence from Brazil."Health Policy and Planning 16(1):21­28. Klepinger, Daniel, Shelly Lundberg, and Robert Plotnick. 1997."How Kaufman, C. E., T. de Wet, and J. Stadler. 2001."Adolescent Preg- Does Adolescent Fertility Affect the Human Capital and Wages of nancy and Parenthood in South Africa." Studies in Family Plan- Young Women?"Madison,Wisconsin: Institute for Research on ning 32(2):147­60. Poverty, University of Wisconsin, Discussion Paper 1145-97. Kaufman, Carol E., and Stavros E. Stavrou. 2004."Bus Fare, Please: Kluve, Jochen. 2006."The Effectiveness of European Active Labor The Economics of Sex and Gifts Among Adolescents in Urban Market Policy." Bonn: Institute for Labor Policies (IZA), Discus- South Africa." Culture, Health and Sexuality 6(5):377­91. sion Paper 2018. Kaushik, Basu, and Zafiris Tzannatos. 2003."The Global Child Knight, John B., and Richard H. Sabot. 1990. Education, Productiv- Labor Problem: What Do We Know and What Can We Do?" ity, and Inequality: The East African Natural Experiment. Wash- World Bank Economic Review 17(2):147­73. ington, DC: World Bank. Kemple, James J., and Judith Scott-Clayton. 2004. Career Acad- Knowles, James, and Jere R. Behrman. 2003. Assessing the Economic emies: Impacts on Labor Market Outcomes and Educational Returns to Investing in Youth in Developing Countries. Washing- Attainment. New York: Manpower Demonstration Research ton, DC: World Bank, HNP Discussion Paper. Corporation (MDRC). Knowles, James C., and Jere R. Behrman. 2005."The Economic Kenkel, Donald S. 2000."Prevention." In Anthony J. Culyer and Returns to Investing in Youth in Developing Countries: A Joseph P. Newhouse, (eds.), Handbook of Health Economics. Review of the Literature." Washington, DC: World Bank: Health, Amsterdam: Elsevier. Nutrition, and Population Discussion Paper January 2005. Kessler, Ronald C., Cindy L. Foster, William B. Saunders, and Paul Knudsen, Eric I. 2004."Sensitive Periods in the Development E. Stang. 1995."Social Consequences of Psychiatric Disorders of the Brain and Behavior." Journal of Cognitive Neuroscience I: Educational Attainment." American Journal of Psychiatry 16(8):1412­25. 152(7):1026­32. Köbrich, Claus, and Martine Dirven. 2006."Características del Khan, Ayesha. 1998. Female Mobility and Social Barriers to Access- Empleo Rural no Agrícola en America Latina con Enfasis en los ing Health and Family Planning Services: A Qualitative Research Servicios." ECLAC. Santiago de Chile. Processed. Study in Three Punjabi Villages. Islamabad: Ministry for Popula- Kohlberg, Lawrence. 1973."The Claim to Moral Adequacy of tion Welfare. a Highest Stage of Moral Judgment." Journal of Philosophy Khandker, Shahidur R. 2005."Microfinance and Poverty: Evidence 70(18):630­46. Using Panel Data from Bangladesh." World Bank Economic Konde-Lule, J. K., N. Sewankambo, and M. Morris. 1997."Ado- Review 19(2):263­86. lescent Sexual Networking and HIV Transmission in Rural Khandker, Shahidur R., Mark M. Pitt, and Nobuhiko Fuwa. 2003. Uganda." Health Transition Review 7(Suppl):89­100. "Subsidy to Promote Girls' Secondary Education: The Female Koplan, Jeffrey P., Catharyn T. Liverman, and Vivica I. Kraak, Stipend Program in Bangladesh." World Bank. Washington, DC. eds. 2005. Preventing Childhood Obesity: Health in the Balance. Processed. Washington, DC: The National Academies Press. Available Kibria, Nazli. 1995."Culture, Social Class, and Income Control in online at http://www.nap.edu/catalog/11015.html. the Lives of Women Garment Workers in Bangladesh." Gender Korenman, Sanders, and David Neumark. 2000."Cohort Crowd- and Society 9(3):289­309. ing and Youth Labor Markets: A Cross-National Analysis." In Kijima, Yoko, and Peter Lanjouw. 2004."Non-farm Employment, David G. Blanchflower and Richard B. Freeman, (eds.), Youth Agricultural Wage, and Poverty Trends in Rural India." World Employment and Joblessness in Advanced Countries. Chicago and Bank. Washington, DC. Processed. London: University of Chicago Press for the National Bureau of Kim, Gwang-Jo. 2002."Education Policies and Reform in South Economic Research. Korea." In World Bank, (eds.), Secondary Education in Africa: Kremer, Michael, and Dan M. Levy. 2003."Peer Effects and Alco- Strategies for Renewal. Washington, DC: World Bank. hol Use Among College Students." Cambridge, MA: National King, Elizabeth M. 2005."Schemes to Acknowledge the Value of Bureau of Economic Research Working Paper Series 9876. Daughters in India." World Bank. Washington, DC. Processed. Kremer, Michael, Edward Miguel, and Rebecca Thornton. 2004. Kinkade, Sheila, and Christina Macy. 2005. Our Time is Now: "Sensitive Periods in the Development of the Brain and Behav- Young People Changing the World. New York, NY: Pearson ior." Cambridge, MA: National Bureau of Economic Research Foundation. Working Paper Series 10971. Kirby, Douglas. 2001. Emerging Answers: Research Findings on Kriger, Norma. 2005."ZANU(PF) Strategies in General Elec- Programs to Reduce Teen Pregnancy. Washington, DC: National tions, 1980-2000: Discourse and Coercion." African Affairs Campaign to Prevent Teen Pregnancy. 104(414):1­34. Kirby, Douglas, B. A. Laris, and Lori Rolleri. 2005."Impact of Sex and Krug, Etienne G., Linda L. Dahlberg, James A. Mercy, Anthony HIV Education Programs on Sexual Behaviors of Youth in Devel- B. Zwi, and Rafael Lozano. 2002. World Report on Violence and oping and Developed Countries."Research Triangle Park, N.C.: Health. Geneva, Switzerland: World Health Organization. Family Health International,Youth Research Working Paper 2. Kugler, Adriana D. 2004."The Effect of Job Security Regulations Kirby, Douglas, Gina Lepore, and Jennifer Ryan. 2005. A Matrix of on Labor Market Flexibility: Evidence from the Colombian Risk and Protective Factors Affecting Teen Sexual Behavior, Preg- Labor Market Reform." In James J. Heckman and Carmen Pagés, nancy, Childbearing and Sexually Transmitted Diseases. Wash- (eds.), Law and Employment: Lessons from Latin America and 258 WORLD DEVELOPMENT REPORT 2007 the Caribbean. Chicago and London: The University of Chicago Lazear, Edward P. 1983."A Competitive Theory of Monopoly Press. Unionism." American Economic Review 73(4):631­43. Kugler, Adriana D., Juan F. Jimeno, and Virginia Hernanz. 2003. Lee, Jong-Wha, and Changyong Rhee. 1999."Social Impacts of the "Employment Consequences of Restrictive Permanent Con- Asian Crisis: Policy Challenges and Lessons." New York: UNDP tracts." London: Centre for Economic Policy Research Discus- Occasional Paper 33. sion Paper 3724. Lee, Lena, and Poh Kam Wong. 2005."Entrepreneurship Educa- Kuku, Yemisi, Peter F. Orazem, and Rajesh Singh. 2005."Computer tion - A Compendium of Related Issues." Singapore: SSRN Adoption and Returns in Transition." Ames, Iowa: Iowa State Electronic Paper Collection Available online at http://ssrn. University Economics Working Paper 04021. com/abstract=856227. Kurz, Kathleen M., Nancy L. Peplinsky, and Charlotte Johnson- Leff, Nathaniel H. 1969."Dependency Rates and Saving Rates." Welch. 1994. Investing in the Future: Six Principles for Promoting American Economic Review 59(5):886­96. the Nutritional Status of Adolescent Girls in Developing Countries. Leopold, Evelyn. 2001."Congolese Kids Face Horrific Conditions." Washington, DC: International Center for Research on Women. Reuters, June 17. Kutan, Ali M., and Kostas Drakos. 2003."Regional Effects Lesthaeghe, R., and G. Moors. 2000."Recent Trends in Fertility and of Terrorism on Tourism." Journal of Conflict Resolution Household Formation in the Industrialized World." Review of 47(5):621­641. Population and Social Policy 9:121­70. Kymlicka, Will. 2001. Politics in the Vernacular: Nationalism, Mul- Levine, Saul. 1999."Youth in Terrorist Groups, Gangs, and Cults: ticulturalism, and Citizenship. New York, NY: Oxford University The Allure, the Animus, and the Alienation." Psychiatric Annals Press. 29(6):342­49. La Cava, Gloria, Zeynep Ozbil, Sarah Michael, Elena Galliano, Levison, Deborah, Karine S. Moe, and Felicia M. Knaul. 2001. Minna Mattero, and Mio Takada. 2006."Youth and the Transi- "Youth Education and Work in Mexico." World Development tion to Citizenship: The Role of the State in Promoting Positive 29(1):167­88. Youth Religious and Ethnic Identities and Young People's Role Levitt, Steven D. 1998."Juvenile Crime and Punishment." Journal in Peace-Building and Social Change." Background paper for of Political Economy 106(6):1156­1187. the WDR 2007. ------. 1999."The Limited Role of Changing Age Structure in Lam, David. 2006."The Demography of Youth in Developing Explaining Aggregate Crime Rates." Criminology 37(3):581­98. Countries and its Economic Implications." Background paper for the WDR 2007. Levitt, Steven D., and Lance Lochner. 2001."The Determinants of Juvenile Crime." In Jonathan Gruber, (eds.), Risky Behavior Lanjouw, Peter, and Nicholas Stern. 2006. Economic Development among Youths: An Economic Analysis. Chicago, IL: University of in Palanpur Over Five Decades. New York: Oxford University Chicago Press for the National Bureau of Economic Research. Press. Levy, Frank, and Richard Murnane. 2004. The New Division of Lansdown, R., A. Ledward, A. Hall, W. Issae, E. Yona, J. Matulu, M. Labor: How Computers Are Creating the Next Job Market. Princ- Mweta, C. Kihamia, U. Nyandindi, and D. Bundy. 2002."Schis- eton, NJ: Princeton University Press. tosomiasis, Helminth Infection and Health Education in Tan- zania: Achieving Behavior Change in Primary Schools." Health Levy, Santiago, and Evelyne Rodríguez. 2004. Economic Crisis, Education Research 17(4):425­33. Political Transition and Poverty Policy Reform. Washington, DC: Inter-American Development Bank, Regional Operations Lasida, Javier, and Ernesto Rodríguez. 2006."Entrando al Mundo Department II. Policy Dialogue Series. de Trabajo: Resultados de Seis Proyectos Entra 21." Baltimore, Li, R., X. Chen, Yan.H., P. Deurenberg, L. Garby, and J. G. Hautvast. MD: International Youth Foundation, Serie de Aprendizaje 2. 1994."Functional Consequences of Iron Supplementation in Latimer, Jeff, Craig Dowden, and Danielle Muise. 2001. The Effec- Iron-deficient Female Cotton Mill Workers in Beijing, China." tiveness of Restorative Justice Practices: A Meta-Analysis. Canada: American Journal of Clinical Nutrition 59(4):908­13. Research and Statistics Division, Department of Justice. Liang, Gui. 2005. The CASS Internet Survey Report 2005: Sur- Lautrédou, Gérard. 2005."Viet Nam: A Tale of Two Cities." The veying Internet Use and Impact in 5 Chinese Cities. Beijing: Magazine of the International Red Cross and Red Crescent Chinese Academy of Social Sciences. Available online at Movement. http://www.worldinternetproject.net/publishedarchive/ Lavinas, Lena, Maria Lígia Barbosa, and Octávio Tourinho. 2001. China%20Report%202005.pdf. Assessing Local Minimum Income Programmes in Brazil: ILO- Lillard, Lee A., and Robert J. Willis. 1994."Intergenerational Edu- World Bank Agreement. Geneva: International Labour Office. cational Mobility: Effects of Family and State in Malaysia." Jour- Available online at http://www-ilo-mirror.cornell.edu/public/ nal of Human Resources 29(4):1126­66. english/protection/ses/download/docs/2brazil.pdf. Lindert, Peter H. 2003."Voice and Growth: Was Churchill Right." Lavy, V., and A. Schlosser. 2005."Targeted Remedial Education Journal of Economic History 63(2):315­50. for Underperforming Teenagers: Costs and Benefits." Journal of Llisterri, J., H. Kantis, P. Angelelli, and L. Tejerina. 2006."Youth Labor Economics 23(4):839­74. Entrepreneurship in Latin America, Inter-American Develop- Lavy, Victor. 1996."School Supply Constraints and Children's ment Bank Forthcoming." Background paper for the WDR Educational Outcomes in Rural Ghana." Journal of Development 2007. Economics 51(2):291­314. Llorente, María Victoria, and Angela Rivas. 2005."Case Study: ------. 2002."Evaluating the Effect of Teachers' Group Perfor- Reduction of Crime in Bogota: A Decade of Citizen's Security Poli- mance Incentives on Pupil Achievement." Journal of Political cies."Washington, DC: World Bank Working Paper Series 35128. Economy 110(6):1286­317. Lloyd, Cynthia B. 2003."The Impact of Educational Qual- Lazear, Edward. 2001."Educational Production." Quarterly Journal ity on School Exit in Egypt." Comparative Education Review of Economics 116(3):777­803. 47(4):444­67. References 259 Lloyd, Cynthia B., and Monica J. Grant. 2004."Growing Up in What We Can Do About It. Washington, DC: Brookings Institu- Pakistan: The Separate Experiences of Males and Females." New tion Press. York: Population Council, Policy Research Division, Working Machin, Stephen, and Costas Meghir. 2004."Crime and Economic Paper Series 188. Incentives." Journal of Human Resources 39(4):958­79. Lloyd, Cynthia B., Barbara S. Mensch, and Wesley H. Clark. 2000. Maclellan, Nic, and Peter Mares. 2005."Labour Mobility in the "The Effects of Primary School Quality on School Dropout Pacific: Creating Seasonal Work Programs in Australia." Paper Among Kenyan Girls and Boys." Comparative Education Review presented at the Globalisation, Governance and the Pacific 44(2):113­47. Islands. State, Society and Governance in Melanesia Project Lochner, Lance, and Enrico Moretti. 2004."The Effect of Educa- (SSGM) Australian National University, Canberra, 25 - 27 Octo- tion on Crime: Evidence from Prison Inmates, Arrests, and Self- ber 2005. Reports." American Economic Review 94(1):155­89. Maclure, Richard, and Melvin Sotelo. 2004."Youth Gangs in Nica- Loevinsohn, Benjamin P. 1990."Health Education Intervention in ragua: Gang Membership as Structured Individualization." Jour- Developing Countries: A Methodological Review of Published nal of Youth Studies 7(4):417­32. Articles." International Journal of Epidemiology 19(4):788­94. Makinson, Carolyn. 1985."The Health Consequences of Teenage Lokshin, Michael, Elena Glinskaya, and Marito Garcia. 2004."The Fertility." Family Planning Perspectives 17(3):132­39. Effect of Early Childhood Development Programs on Women's Maloney, William F., and Jairo Nuñez Mendez. 2004."Measuring Labor Force Participation and Older Children's Schooling in the Impact of Minimum Wages: Evidence from Latin America." Kenya." Journal of African Economies 13(2):240­76. In James Heckman and Carmen Pagés, (eds.), Law and Employ- Lokshin, Michael, and Ruslan Yemtsov. 2005."Who Bears the Cost ment: Lessons from Latin America and the Caribbean. Chicago: of Russia's Military Draft?" Washington, DC: World Bank Policy University of Chicago Press. Research Working Paper 3547. Maluccio, John A. Forthcoming."Education and Child Labor: Longhi, Simonetta, Peter Nijkamp, and Jacques Poot. 2005."A Experimental Evidence from a Nicaraguan Conditional Cash Meta-analytic Assessment of the Effect of Immigration on Transfer Program." In Peter F. Orazem, Guilherme Sedlacek, Wages." Journal of Economic Surveys 19(3):451­77. and Zafiris Tzannatos (eds.) Eradicating Child Labor in Latin López, Alan D., Colin D. Mathers, Majad Ezzati, Dean T. Jamison, America in the 90s: The Promise of Demand Side Interventions. and Christopher J. L. Murray, eds. 2006. Global Burden of Dis- Washington, DC: World Bank and Inter-American Develop- ease and Risk Factors. Washington, DC: World Bank and Oxford ment Bank. University Press. Mamdani, Mahmood. 1996. Citizen and Subject: Contemporary Lucas, Gregory M., Laura W. Cheever, Richard E. Chaisson, and Africa and the Legacy of Late Colonialism. Princeton, NJ: Princ- Richard D. Moore. 2001."Detrimental Effects of Continued eton University Press. Illicit Drug Use on the Treatment of HIV-1 Infection." Journal of Mammen, Kristin, and Christina Paxson. 2000."Women's Work Acquired Immune Deficiency Syndromes 27(3):251­9. and Economic Development." Journal of Economic Perspectives Lucas, Robert E. B. 2004. International Migration Regimes and Eco- 14(4):141­64. nomic Development. Boston: Boston University, Department of Mangiaterra,Viviana, and Gerold Vollmer. 2006."Youth Consulta- Economics. tions for the WDR 2007: Synthesis Report of Country and Grass- Luke, Nancy. 2003."Age and Economic Asymmetries in the Sexual roots Consultations." Background paper for the WDR 2007. Relationships of Adolescent Girls in Sub-Saharan Africa." Stud- Mannheim, Karl. 1972."The Problem of Generations." In P Alt- ies in Family Planning 34(2):67­86. bach and R Laufer, (eds.), The New Pilgrims: Youth Protest in Lule, Elizabeth, James E. Rosen, Susheela Singh, James C. Knowles, Transition. New York, NY: David McKay and Company. and Jere R. Behrman. 2005."Adolescent Health Programs." In Mar Gadio, Coumba. 2001. Exploring the Gender Impacts of Dean T. Jamison, David B. Evans, George Alleyne, Prabhat Jha, World Links in some Selected Participating African Coun- Joel G. Breman, Anthony R. Measham, Mariam Claeson, Anne tries: A Qualitative Approach. Washington, DC: World Mills, and Philip R. Musgrove, (eds.), Disease Control Priorities Links. Available online at http://www.world-links. in Developing Countries. Bethesda, MD: Disease Control Priori- org/modules/Downloads/fileuploads/gender_study_v2.pdf. ties Project. Marseille, Elliot. 2003. The External Effects of HAART. A Back- Lundberg, Mattias K. A., Mead Over, and Phare Mujinja. 2003."Do ground Paper for HIV/AIDS Treatment and Prevention in India. Savings Predict Death? Precautionary Savings During an Epi- Washington, DC: World Bank. demic." UNAIDS. Geneva. Processed. Marshall, T. H. 1950. Citizenship and Social Class and Other Essays. Lundberg, Shelly, and Elaina Rose. 2002."The Effects of Sons and Cambridge, U.K.: Cambridge University Press. Daughters On Men's Labor Supply and Wages." Review of Eco- nomics and Statistics 84(2):251­68. Martin, John P., and David Grubb. 2001."What Works and for MacDonald, Margaret, Matthew Law, John Kaldor, Jim Hales, and Whom: A Review of OECD Countries' Experiences with Active Gregory J. Dore. 2003."Effectiveness of Needle and Syringe Pro- Labour Market Policies." Swedish Economic Policy Review grammes for Preventing HIV Transmission." International Jour- 8(2):9­56. nal of Drug Policy 14(5-6):353­57. Martinez, Ramiro Jr., and Matthew T. Lee. 2000."On Immigration Macedo, Stephen, Yvette Alex-Assensoh, Jeffrey M. Berry, Michael and Crime." In U.S.Department of Justice Office of Justice Pro- Brintnall, David E. Campbell, Luis Ricardo Fraga, Archon Fung, grams, (eds.), Criminal Justice 2000. Washington, DC: National William A Galston, Christopher F. Karpowitz, Margaret Levi, Institute of Justice. Meira Levinson, Keena Lipsitz, Richard G. Niemi, Robert D. Martorell, R., U. Ramakrishnan, D. G. Schroeder, P. Melgar, and Putnam, Wendy M. Rahn, Rob Reich, Robert R. Rodgers, Todd L. Neufeld. 1998."Intrauterine Growth Retardation, Body Size, Swanstrom, and Katherine Cramer Walsh. 2005. Democracy at Body Composition and Physical Performance in Adolescence." Risk: How Political Choices Undermine Citizen Participation, and European Journal of Clinical Nutrition 52(Supplement):S43­53. 260 WORLD DEVELOPMENT REPORT 2007 Mason, Andrew, eds. 2001. Population Change and Economic Mensch, Barbara S., Judith Bruce, and Margaret E. Greene. 1998. Development in East Asia: Challenges Met, Opportunities Seized. The Uncharted Passage: Girls' Adolescence in the Developing Stanford, CA: Stanford University Press. World. New York: Population Council. Massey, Douglas S., Joaquín Arango, Graeme Hugo, Ali Kouaouchi, Mensch, Barbara S., Paul C. Hewett, and Annabel S. Erulkar. Adela Pellegrino, and Edward Taylor. 1998. Worlds in Motion: 2001."The Reporting of Sensitive Behavior Amons Adoles- Understanding International Migration at the End of the Millen- cents: A Methodological Experiment in Kenya." Demography nium. Oxford: Oxford University Press. 40(2):247­68. Mathur, Sanyukta, Margaret Greene, and Anju Jamhotra. 2003. Mensch, Barbara S., Susheela Singh, and John B. Casterline. 2006. Too Young to Wed: The Lives, Rights and Health of Young Married "Trends in the Timing of First Marriage Among Men and Women Girls. Washington, DC: International Center for Research on in the Developing World." In Cynthia B. Lloyd, Jere R. Behrman, Women. Nelly P. Stromquist, and Barney Cohen, (eds.), The Changing Mathur, Sanyukta, Manisha Mehta, and Anju Malhotra. 2004. Transitions to Adulthood in Developing Countries: Selected Studies. Youth Reproductive Health in Nepal: Is Participation the Answer? Washington, DC: The National Academies Press. Washington, DC: International Center for Research On Women. Mesnard, Alice. 2004."Temporary Migration and Capital Market Available online at http://www.icrw.org/docs/nepal_0104.pdf. Imperfections." Oxford Economic Papers 56(2):242­262. Mattero, Minna. 2006."Youth Policies and the Institutional Frame- Mesquida, Christian G., and Neil I Wiener. 1999."Male Age Com- work." Background paper for the WDR 2007. position and Severity of Conflicts." Political and Life Sciences Matthews, Iole, Richard Griggs, and Glenda Caine. 1999. The 18(2):181­9. Experience Review of Interventions and Programmes Dealing with Mete, Cem. 2004."The Inequality Implications of Highly Selec- Youth Violence in Urban Schools in South Africa. Durban, South tive Promotion Practices." Economics of Education Review Africa: Independent Projects Trust. 23(3):301­14. Matthews, Limb H. 2003."Another White Elephant? Youth Coun- Miers, David. 2001."An International Review of Restorative Jus- cils as Democratic Structures." Space Polity 7(2):172­92. tice." London, U.K.: Crime Reduction Research Series 10. Mattoo, Aaditya, IIeana Cristina Neagu, and Caglar Özden. 2005. Miguel, Edward, Rachel Glennerster, and Kate Whiteside. 2006. "Brain Waste? Educated Immigrants in the U.S. Labor Market." "Civil Conflict and Local Collective Action in Sierra Leone." Washington, DC: World Bank Policy Research Working Paper Paper presented at the Annual Meeting Allied Social Science No. 3581. Associations. Boston, MA. McAdam, Doug. 1988. Freedom Summer. New York, NY: Oxford Miguel, Edward, and Michael Kremer. 2004."Worms: Identifying University Press. Impacts on Education and Health in the Presence of Treatment McCauley, A. P., and C. Salter. 1995."Meeting the Needs of Young Externalities." Econometrica 72(1):159­217. Adults." Baltimore, MD: Johns Hopkins School of Public Health, Miller, J. E. 1991."Birth Intervals and Perinatal Health: An Inves- Population Information Program Series J, No. 43. tigation of Three Hypotheses." Family Planning Perspectives McGirk, Jan. 2001."Brutality of Child Army Shocks Colombia." 23(2):62­70. The Independent, May 2. Ministerio del Trabajo Peru. 1998. La Evaluacion de Impacto del McKenzie, David J. 2004."Aggregate Shocks and Urban Labor Programa de Capacitación Laboral Juvenil Projoven. Lima, Peru: Market Responses: Evidence from Argentina's Financial Crisis." Ministerio del Trabajo de Peru. Economic Development and Cultural Change 52(4):719­58. Mishra, Prachi. Forthcoming."Emigration and Wages in Source ------. 2005."Paper Walls are Easier to Tear Down: Passport Countries: Evidence from Mexico." Journal of Development Costs and Legal Barriers to Emigration." Washington, DC: Economics. World Bank Policy Research Working Paper 3783. Mocan, H. Naci, and Daniel I. Rees. 1999."Economic Conditions, ------. 2006a."A Profile of the World's Young Developing Coun- Deterrence, and Juvenile Crime: Evidence From Micro Data." try Migrants." Background paper for the WDR 2007. Cambridge, MA: National Bureau of Economic Research Work- -------. 2006b."Remittances in the Pacific." Paper presented at ing Paper Series 7405. the Werner-Sichel lecture. February 15b. University of Western Moehling, Carolyn M. 2005."She Has Suddenly Become Pow- Michigan. erful: Youth Employment and Household Decision-Making McKenzie, David J., John Gibson, and Steven Stillman. 2006."How in the Early Twentieth Century." Journal of Economic History Important is Selection? Experimental vs Non-experimental 65(2):414­38. Measures of the Income Gains from Migration." Washington, Montenegro, Claudio E., and Carmen Pagés. 2004."Who Benefits DC: World Bank Policy Research Working Paper Series 3906. from Labor Market Regulations?" In James Heckman and Car- McKenzie, David J., and Hillel Rapoport. 2006."Can Migration men Pagés, (eds.), Law and Employment: Lessons from Latin Reduce Education? Depressing Evidence from Mexico." World America and the Caribbean. Chicago: University of Chicago Bank. Washington, DC. Processed. Press. Reprinted in Jorge Enrique Restrepo and Andrea Tokman McKinsey Global Institute. 2005. The Emerging Global Labor R. (eds.) "Labor Markets and Institutions," (2005), Santiago de Market: Part I - The Demand for Offshore Talent in Services. Chile: Banco Central de Chile. Washington, DC: McKinsey Global Institute. Available online Montoya Díaz, M. Dolores. 2002."Socio-economic Health at http://www.mckinsey.com/mgi/publications/emergingglobal- Inequalities in Brazil: Gender and Age Effects." Health Econom- labormarket/Part1/Index.asp. ics 11:141­54. Meager, Nigel, and Ceri Evans. 1998."The Evaluation of Active Moore, Michael J., and Philip J. Cook. 1995."Habit and Hetereo- Labour Market Measures for the Long-term Unemployed." geneity in the Youthful Demand for Alcohol." Cambridge, MA: Geneva: International Labour Office, International Migration National Bureau of Economic Research Working Paper Series Paper Series 16. 5152. References 261 Morley, Samuel, and David Coady. 2003. From Social Assistance National Assessment and Examination Center. 2005. Unified to Social Development: Targeted Education Subsidies in Devel- National University Entry Examinations. Tiblisi: Ministry of oping Countries. Washington, DC: Institute for International Education and Science, Government of Georgia. Economics. National Research Council and Institute of Medicine. 2005. Moscow Human Rights Bureau. 2005. Racism, Xenophobia, Anti- Growing Up Global: The Changing Transitions to Adulthood in Semitism and Ethnic Discrimination in Russia, January-June Developing Countries. Panel on Transitions to Adulthood in 2005. Moscow, Russia: Moscow Human Rights Bureau. Developing Countries. Cynthia B. Lloyd, ed. Committee on Mroz, Thomas A., and Timothy H. Savage. 2001. The Long-Term Population and Board on Children, Youth, and Families. Divi- Effects of Youth Unemployment. Chapel Hill, N.C.: University of sion of Behavioral and Social Sciences and Education. Washing- North Carolina at Chapel Hill and Welch Consulting Econo- ton, DC: The National Academies Press. mists. Available online at http://www.epionline.org/studies/ National Youth Council of Ireland. 2001. The Plunder Years: A mroz_10-2001.pdf. Report on Young Drivers and Motor Insurance in Ireland. Dublin: Muirhead, Debbie., Lilani Kumaranayake, and Charlotte Watts. National Youth Council of Ireland. 2001. Economically Evaluating the 4th Soul City Series: Costs Navarro, Napoleon, and Amer Hasan, eds. 2003. Good Practices in and Impact onHIV/AIDS and Violence Against Women. Johan- Asia and the Pacific: Expanding Choices, Empowering People. New nesburg; London: Institute for Health and Development Com- York, NY: UNDP. Available online at http://www.undp.org/rbap/ munication, and the London School of Hygiene and Tropical BestPrac/BGD_Elections.pdf. Medicine. Nelson, Jon P. 2003."Youth Smoking Prevalence in Developing Mullahy, John, and Jody L. Sindelar. 1991."Gender Differences Countries: Effect of Advertising Bans." Applied Economic Letters in Labor Market Effects of Alcoholism." American Economic 10(13):805­11. Review: Papers and Proceedings 81(2):161­65. Neto, Isabel, Charles Kenny, Subramaniam Janakiram, and Charles Mumtaz, Z., and S. Salway. 2005."I Never Go Anywhere: Extricat- Watt. 2005."Look Before you Leap: The Bumpy Road to E- ing the Links between Women's Mobility and Uptake of Repro- development." In Robert Schware, (eds.), E-development: From ductive Health Services in Pakistan." Social Science and Medicine Excitement to Effectiveness. Washington, DC: World Bank. 60(8):1751­65. Neuhaus, Tom. 2005."No Nazi: Youth Rebels of the Third Reich." Mungai, Wainaina. 2005."Using ICTs for Poverty Reduction History Today 55(11):52­7. and Environmental Protection in Kenya: The "M-vironment" Neumark, David, and William Wascher. 1999."A Cross-National Approach." In Maja Andejelkovic, (eds.), A Developing Connec- Analysis of the Effects of Minimum Wages on Youth Employ- tion: Bridging the Policy Gap between the Information Society and ment." Cambridge, MA: National Bureau of Economic Research Sustainable Development. Winnipeg, Canada: IISD. Working Paper Series 7299. Munshi, Kaivan, and Mark Rosensweig. 2003."Traditional Insti- Newport, Elissa L. 2002."Critical Periods in Language Develop- tutions Meet the Modern World: Caste, Gender and Schooling ment." In Lynn Nadel, (eds.), Encyclopedia of Cognitive Science. Choice in a Globalizing Economy." Cambridge, MA: Bureau London: Macmillan Publishers Ltd./Nature Publishing Group. for Research in Economic Analysis of Development(BREAD) Working Paper Series 038. Available online at http://www.cid. Nguyen Anh, Dang. 2005."Viet Nam Internal Migration: Oppor- harvard.edu/bread/papers/working/038.pdf. tunities and Challenges for Development." Paper presented at the Regional Conference on Migration and Development in Murdoch, Maureen, Melissa A. Polusny, James Hodges, and Nancy Asia. Lanwhou, China. O'Brien. 2004."Prevalence of in-service and Post-service Assault among Combat and Non-combat Veterans Applying for Depart- Nguyen Anh, Dang, Le Bach Duong, and Nguyen Hai Van. 2005. ment of Veterans Affairs Posttraumatic Stress Disorder Disabil- "Youth Employment in Viet Nam: Characteristics, Deter- ity Benefits." Military Medicine 169(5):392­5. minants and Policy Responses." Geneva: ILO Employment Strategy Papers 2005/9. Available online at http://www.ilo. Murray, Sarah. 2004."Investing in Young People: A Force with a org/public/english/employment/strat/download/esp2005-9.pdf. Huge Potential." Financial Times, January 23. Myers, David, and Allen Schirm. 1999."The Impacts of Upward Nickell, Stephen, and Richard Layard. 1999."Labor Market Insti- Bound: Final Report for Phase I of the National Evaluation tutions and Economic Performance." In Orley Ashenfelter and Analysis and Highlights." Mathematica Policy Research, Inc. David Card, (eds.), Handbook of Labor Economics. Vol. 3C. Washington, DC. Processed. Amsterdam: Elsevier. Nanda, Geeta, Kimberly Switlick, and Elizabeth Lule. 2005."Accel- Noorkôiv, Rivo, Peter F. Orazem, Allan Puur, and Milan Vodopivec. erating Progress towards Achieving the MDG to improve Mater- 1998."Employment and Wage Dynamics in the Estonia Transi- nal Health, Health Nutrition and Population." Washington, DC: tion, 1989-1995." Economics of Transition 6(2):481­503. World Bank, Health Nutrition and Population Discussion Paper Norris, Pippa. 2005. Democratic Phoenix: Reinventing Political 31969. Activism. Cambridge, U.K.: Cambridge University Press. Nanda, Priya. 1999."Women's Participation in Rural Credit Pro- O'Donoghue, Ted, and Matthew Rabin. 2001."Risky Behavior grammes in Bangladesh and their Demand for Formal Health Among Youths: Some Issues from Behavioral Economics." In Care: Is there a Positive Impact?" Health Economics and Econo- Jonathan Gruber, (eds.), Risky Behavior among Youths: An Eco- metrics 8:415­28. nomic Analysis. Chicago, IL: University of Chicago Press for the Narayan, Deepa. 1995. The Contribution of People's Participation: National Bureau of Economic Research. Evidence from 121 Rural Water Supply Projects. Washington, DC: O'Higgins, Niall. 2003."Trends in the Youth Labor Market in World Bank. Developing and Transition Countries." Washington, DC: World Narayan, Deepa, and Patti Petesch. 2006."Moving out of Poverty: Bank, Social Protection Discussion Paper Series 0321. Some Preliminary Results." World Bank, PRMPR. Washington, Obadare, Ebenezer. 2005. Statism, Youth and the Civic Imagi- DC. Processed. nation: A Critical Study of the National Youth Service Corps 262 WORLD DEVELOPMENT REPORT 2007 (NYSC) Programme in Nigeria. Missouri: Global Service Insti- Chile." Washington, DC: Inter-American Development Bank, tute, Washington University. Office of the Chief Economist Working Paper Series 398. OECD. 2003. Trends in International Migration: SOPEMI 2002 Pals, Heili, and Nancy Brandon Tuma. 2004."Entrepreneur- Edition. Paris: Organization for Economic Co-operation and ial Activities in Post-Soviet Societies: Impacts of Social Psy- Development. chological Characteristics." International Journal of Sociology ------. 2004. OECD Employment Outlook 2004. Paris: Organiza- 34(2):11­38. tion for Economic Co-operation and Development. Pampel, Fred C., and H. Elizabeth Peters. 1995."The Easterlin ------. 2005a. OECD Employment Outlook 2005. Paris: Organiza- Effect." Annual Review of Sociology 21:163­94. tion for Economic Co-operation and Development. Pande, Rohini. 2005."Can Mandated Political Representa- ------. 2005b. School Factors Related to Quality and Equity: tion Increase Policy Influence for Disadvantaged Minorities? Results from PISA 2000. Paris: Organization for Economic Co- Theory and Evidence from India." American Economic Review operation and Development. 93(4):1132­51. ------. 2005c. Trends in International Migration: SOPEMI 2004 Paniza, Hugo. 2000."The Public Sector Premium and the Gender Edition. Paris: Organization for Economic Co-operation and Gap in Latin America: Evidence from the 1980s and 1990s." Development. Washington, DC: Inter-American Development Bank, Research Ogar, Joel. 2005. A Friend's Slap. World Wide Web: Panorama: Department, Working Paper 431. A TakingITGlobal Online Publication. Available online at Parent, Daniel. 2006."Youth Labor Markets in Burkina Faso: http://www.takingitglobal.org/express/panorama/article. Recent Trends and Analysis." McGill University. Montreal. html?ContentID=6505. Processed. Okonofua, F. E., P. Coplan, S. Collins, F. Oronsaye, D. Ogunsakin, J. Parham, Angel Adams. 2004."Diaspora, Community and Com- T. Ogonor, J. A. Kaufman, and K. Heggenhougen. 2003."Impact munication: Internet Use in Transnational Haiti." Global Net- of an Intervention to Improve Treatment-seeking Behavior and works 4(2):199­217. Prevent Sexually Transmitted Diseases among Nigerian Youths." Parker, Susan. 2003."Evaluación de Impacto de Oportunidades International Journal of Infectious Diseases 7(1):61­73. sobre la Inscripción Escolar: Primaria, Secundaria y Media Omelaniuk, Irena. 2005."Trafficking in Human Beings." World Superior." Mexico, D.F.: Secretaria de Desarrollo Social, Docu- Bank. Washington, DC. Processed. mento de Investigación 6. Ong, Aihwa. 1987. Spirits of Resistance and Capitalist Discipline: Parliamentary Committee for Social Affairs. 2005. Assessment on Factory Women in Malaysia. Albany: State University of New Urban Migration Policy. Hanoi: UNFPA. York Press. Parry, C. D. H., A. Louw, E. Vardas, and A. Plüddemann. 2000. Online Computer Library Center (OCLC). 2005. Country and Lan- Medical Research Council and Institute for Security Studies: guage Statistics. Dublin, O.H.: Online Computer Library Center 3-Metros Arrestee Study (Phase 1). Parow: Medical Research (OCLC). Available online at http://www.oclc.org/research/proj- Council. ects/archive/wcp/stats/intnl.htm. Parsons, Christopher R., Ronald Skeldon, Terrie L. Walmsley, and ORC Macro. 2006."The Demographic and Health Surveys STAT L. Alan Winters. 2005. Quantifying the International Bilateral Compiler". www.measuredhs.com, ORC Macro. Movements of Migrants. Sussex University: The World Bank and Oreopoulos, Philip. 2005."Do Dropouts Drop Out Too Soon? the Development Research Centre on Migration, Globalisation International Evidence From Changes in School-Leaving Laws." and Poverty at Sussex University. Cambridge, MA: National Bureau of Economic Research Work- Partnership for Child Development, Working Group on the Nutri- ing Paper Series 10155. tion of the School Age Child. 2006. Survey of School Health and Osmond, C., and D. J. Barker. 2000."Fetal, Infant, and Childhood Nutrition Programs for the UN Subcommittee on Nutrition. New Growth are Predictors of Coronary Heart Disease, Diabetes, and York: United Nations. Hypertension in Adult Men and Women." Environmental Health Patrinos, Harry A. 2005."Education Contracting: Scope of Future Perspectives 108(Supp. 3):545­53. Research." World Bank. Washington, DC. Processed. Otsuka, Keijiro. Forthcoming."The Rural Industrial Transition in Patrinos, Harry A., and Chris Sakellariou. 2005."Schooling and East Asia: Influences and Implications." In Steven Haggblade, Labor Market Impacts of a Natural Policy Experiment." Labour Peter B. R. Hazell, and Thomas Reardon (eds.) Transforming the 19(4):705­19. Rural Nonfarm Economy. Baltimore, MD: John Hopkins Univer- Paul-Majumder, Pratima, and Anwara Begum. 2000."The Gen- sity Press. der Imbalances in the Export Oriented Garment Industry in Over, Mead, Peter Heywood, Julian Gold, Indrani Gupta, Sub- Bangladesh: Measures for Eliminating Gender Imbalances in hash Hira, and Elliot Marseille. 2004. HIV/AIDS Treatment Export-oriented Garment Industries." Washington, DC: World and Prevention in India: Modeling the Costs and Consequences. Bank, Policy Research Report on Gender and Development Washington, DC: World Bank; Health, Nutrition, and Popula- Working Paper Series 12. tion Series. Pauly, Mark V. 1968."The Economics of Moral Hazard." American Oyediran, Kola A., Gbenga P. Ishola, and Alfred A. Adewuyi. 2002. Economic Review 58(3):531­7. "Knowledge of Possible Pregnancy at First Coitus: A Study of Paxson, Christina H. 2002."Comment on Alan Krueger and In-school Adolescents in Ibadan, Nigeria." Journal of Biosocial Jitka Maleckova,"Education, Poverty and Terrorism: Is There Science 34(2):233­48. a Causal Connection?"." Princeton, NJ: Princeton University Pabilonia, Sabrina W., and Cindy Zoghi. Forthcoming."Returning Research Program in Development Studies Working Paper 207. to the Returns to Computer Use." American Economic Review. Paxson, Christina H., and Norbert Schady. Forthcoming."Cogni- Pagés, Carmen, and Claudio E. Montenegro. 1999."Job Secu- tive Development among Young Children in Ecuador: The Roles rity and the Age Composition of Employment: Evidence from of Wealth, Health and Parenting." Journal of Human Resources. References 263 ------. Forthcoming."Early Childhood Development in Latin Prakesh, Siddhartha. 2003."The African Virtual University and America and the Caribbean." Journal of Human Resources. Growth in Africa: A Knowledge and Learning Challenge." Wash- Peden, M., H. Donson, M. Maziko, and P. Smith. 2000."Substance ington, DC: World Bank, Human Development Findings 223. Abuse Trends Among Trauma Patients: The South African Available online at http://www.worldbank.org/afr/findings/eng- Experience." Paper presented at the 5th World Injury Prevention lish/find223.pdf. and Control Congress. New Delhi. Pratham Resource Centre. 2005. Annual Status of Education Perkinson, Ron. 2005."Beyond Secondary Education: The Promise Report: Aser 2005. Mumbai: Pratham Resource Centre. Available of ICT for Higher Education and Lifelong Learning." In Robert online at http://www.pratham.org/aserrep.php. Schware, (eds.), E-development: From Excitement to Effectiveness. Pritchett, Lant. 2001."Where Has All the Education Gone?" World Washington, DC: World Bank. Bank Economic Review 15(3):367­91. Perlman, Janice E. 2005."Policy Roundtable on the Policy Impli- ------. 2006."Boom Towns and Ghost Countries: Geography, cations of Rio Favela Re-Study." Paper presented at the World Agglomeration and Population Mobility." Harvard University. Bank Policy Roundtable on the Policy Implications of Rio Cambridge, MA. Processed. Favela Re-Study. June 9. Washington, DC Przeworski, Adam, Michael E. Alavarex, Jose Antonio Cheibub, and Perlman, Janice E., and Sarah E. Anthony. 2006."Citizenship and Fernando Limongi. 2000. Democracy and Development: Political Youth in the Favelas of Rio de Janeiro." Background paper for Institutions and Well-Being in the World, 1950-1990. Cambridge, the WDR 2007. U.K.: Cambridge University Press. Peters, Krijn. 2006."Footpaths to Reintegration: Armed Con- Ragui, Assaad, and Melanie Arntz. 2005."Constrained Geographic flict, Youth and the Rural Crisis in Sierra Leone." PhD thesis. Mobility and Gendered Labor Market Outcomes under Struc- Wageningen University. tural Adjustment." World Development 33(3):431­54. Petrosino, Anthony, Carolyn Turpin-Petrosino, and John Buehler. Rahman, Rushidan Islam. 1995."Formal Sector Emplyoment 2005."Scared Straight and Other Juvenile Awareness Programs Among Women in Bangladesh and Gender Composition of for Preventing Juvenile Delinquency". Chichester, U.K., John Industrial Workers." World Bank, Background paper for the Wiley & Sons, Ltd. 1995 World Development Report. Washington, DC. Processed. Pintor, Rafael López, and Maria Gratschew. 2002. Voter Turn- Rajbhandary, Jasmine, Roger Hart, and Chandrika Khatiwada. out Since 1945. Stockholm, Sweden: International Institute for 2001. Extracts from The Children's Clubs of Nepal: A Democratic Democracy and Electoral Assistance (International IDEA). Experiment. London, U.K.: International Institute for Environ- Pitt, Mark, and Shahidur Khandker. 1996."Household and Intra- ment and Development. Available online at http://www.iied.org/ household Impact of the Grameen Bank and Similar Targeted NR/agbioliv/pla_notes/documents/plan_04205.pdf. Credit Programs in Bangladesh." Washington, DC: World Bank Ramakrishnan, Usha. 2004."Nutrition and Low Birth Weight: Discussion Paper 320. From Research to Practice." American Journal of Clinical Nutri- Pitt, Mark M., Shahidur R. Khandker, Signe-Mary Mckernan, and tion 9(1):17­21. M. Abdul Latif. 1999."Credit Programs for the Poor and Repro- Rao, Madanmohan. 1999. Internet Content in India: Local Chal- ductive Behavior in Low Income Countries: Are the Reported lenges, Global Aspirations. The Hague: International Institute for Causal Relationships the Result of Heterogeneity Bias?" Demog- Communication and Development (IICD). raphy 36(1):1­21. Rashid, Ahmed. 2000. Taliban: Militant Islam, Oil and Fundamen- Plutzer, Eric. 2002."Becoming a Habitual Voter: Inertia, Resources, talism in Central Asia. New Haven, CT: Yale University Press. and Growth in Young Adulthood." American Political Science Ravallion, Martin, Gaurav Datt, and Dominique van de Walle. Review 96(1):41­56. 1991."Quantifying Absolute Poverty in the Developing World." Popkin, Barry M. 1994."The Nutrition Transition in Low- Review of Income and Wealth 37 (4): 345­61. income Countries: An Emerging Crisis." Nutrition Reviews Ravallion, Martin, Emanuela Galasso, Teodoro Lazo, and Ernesto 52(9):285­98. Philipp. 2005."What Can Ex-participants Reveal about a Pro- ------. 2002."An Overview on the Nutrition Transition and gram's Impact?" Journal of Human Resources 40(1):208­30. Its Health Implications: The Bellagio Meeting." Public Health Ravallion, Martin, and Quentin Wodon. 2000."Does Child Labour Nutrition 5(1A):93­103. Displace Schooling? Evidence on Behavioural Responses to an Popkin, Barry M., and P. Gordon-Larsen. 2004."The Nutrition Enrollment Subsidy." Economic Journal 110(462):C158­C175. Transition: Worldwide Obesity Dynamics and Their Determi- Rawlings, Laura B., and Gloria M. Rubio. 2005."Evaluating the nants." International Journal of Obesity 28(Supp. 3):2­9. Impact of Conditional Cash Transfer Programs." World Bank Population Council. 2002."Does Easy Accessibility of Emergency Research Observer 20(1):29­55. Contraception Increase Sexual Risk-Taking?" Momentum, Reddy, K. Srinath. 2002."Cardiovascular Diseases in the Devel- October. oping Countries: Dimensions, Determinants, Dynamics and Portes, Jonathan, and Simon French. 2005."The Impact of Free Directions for Public Health Action." Public Health Nutrition Movement of Workers from Central and Eastern Europe on the 5(1a):231­7. UK Labour Market: Early Evidence." London: United Kingdom Refaat, A. 2004."Practice and Awareness of Health Risk Behaviour Department of Work and Pensions Working Paper 18. among Egyptian University Students." Eastern Mediterranean Post, Jerrold M. 1998."Terror Psycho-Logic: Terrorist Behavior Health Journal 10(1/2):72­81. as a Product of Psychological Forces." In Walter Reich, (eds.), Rehm, Jürgen. 2003."Alcohol, Addiction and Public Health."Wash- Origins of Terrorism: Psychologies, Ideologies, Theologies, States of ington, DC: Disease Control Priorities Project, Working Paper 33. Mind. Washington, DC: Woodrow Wilson Center Press. Available online at http://www.dcp2.org/file/47/wp33.pdf. Pound, Edward T., and Danielle Knight. 2006."Cleaning Up the Reinikka, Ritva, and Jacob Svensson. 2002."Assessing Frontline World Bank." U.S.News and World Report, April 3. Service Delivery." World Bank. Washington, DC. Processed. 264 WORLD DEVELOPMENT REPORT 2007 Reiss, Albert J. Jr, and Jeffrey A Roth, eds. 1993. Understanding and Ruel, Marie T., Bénédicte de la Briere, Kelly Hallman, Agnes Preventing Violence. Washington, DC: National Academies Press. Quisumbing, and Nora Coj. 2002."Does Subsidized Childcare Reynolds, Paul D., William D. Bygrave, Erkko Autio, Larry W. Cox, Help Poor Working Women in Urban Areas? Evaluation of a and Michael Hay. 2002. Global Entrepreneurship Monitor. Execu- Government-Sponsored Program in Guatemala City." Washing- tive Report 2002. London: Babson College, London School of ton, DC: IFPRI, FCND Discussion Paper 131. Business and Ewing Marion Kauffman Foundation. Available Ruoen, Ren, and Chen Kai. 1995."China's GDP in U.S. Dollars online at http://www.kauffman.org/pdf/GEM2002.pdf. Based on Purchasing Power Parity." Washington, DC: World Richards, Paul, Khadija Bah, and James Vincent. 2004."Social Cap- Bank Policy Research Working Paper Series 1415. ital and the Survival: Prospects for Community-Driven Devel- Russell, Michele. 2000."Community Based Care and Support Ser- opment in Post-Conflict Sierra Leone." Washington, DC: World vices in South Africa." Paper presented at the The 13th Interna- Bank, Social Development Papers 12. tional AIDS Conference. July 7. Durban, South Africa. Riley, Thira, and William Steel. 1999."Kenya Voucher Program Ryan, Paul. 2001."The School-to-Work Transition: A Cross- for Training and Business Development Services." World Bank. National Perspective." Journal of Economic Literature Washington, DC. Processed. 39(1):34­92. Ritchie, Amanda, Cynthia B. Lloyd, and Monica Grant. 2004. Sacerdote, Bruce. 2001."Peer Effects with Random Assignment: "Gender Differences in Time Use among Adolescents in Devel- Results for Dartmouth Roommates." Quarterly Journal of Eco- oping Countries: The Implications of Rising School Enroll- nomics 116(2):681­704. ment Rates." Washington, DC: Population Council Working Sadler, Anne G., Brenda M. Booth, and Bradley N. Doebbel- Paper 193. Available online at http://www.popcouncil.org/pdfs/ ing. 2005."Gang and Multiple Rapes During Military Service: wp/193.pdf. Health Consequences and Health Care." Journal of American Rivera-Batiz, Francisco L. 2002."Democracy, Governance and Medical Women's Association 60(1):33­41. Economic Growth: Theory and Evidence." New York, NY: Sadoulet, Elisabeth, and Alain de Janvry. 2006."Making Condi- Columbia University Department of Economics Discussion tionnal Cash Transfers More Efficient: Designing for Maximum Paper Series 0102-57. Available online at http://www.columbia. Effect of the Conditionality." World Bank Economic Review edu/cu/economics/discpapr/DP0102-57.pdf. 20(1):1­29. Roche, Declan. 2006."Dimensions of Restorative Justice." Journal Sageman, Marc. 2004. Understanding Terror Networks. Philadel- of Social Issues 62(2):217­38. phia, PA: University of Pennsylvania Press. Rodgers, A., T. Corbett, D. Bramley, T. Riddell, M. Willis, R-B. Lin, Salomon, Joshua A., Daniel R. Hogan, John Stover, Karen A. Stan- and M. Jones. 2005."Do U Smoke after Text? Results of a Ran- ecki, Neff Walker, Peter D. Ghys, and Bernhard Schwartländer. domized Trial of Smoking Cessation Using Mobile Phone Text 2005."Integrating HIV Prevention and Treatment: From Slo- Messaging." Tobacco Control 2005(14):255­61. gans to Impact." PLOS Medicine 2(1):50­56. Rodgers, Dennis. 1999."Youth Gangs in Latin America and the Samuel, Jonathan, Niraj Shah, and Wenona Hadingham. 2005. Caribbean: A Literature Survey." Washington, DC: Latin Amer- "Mobile Communications in South Africa, Tanzania and Egypt: ica and the Caribbean Sustainable Development Working Paper, Results from Community and Business Surveys." Newbury, Urban Peace Program Series 4. Berkshire, U.K.: Vodafone Policy Paper Series 2. Rodríguez Ospina, Edgar, Luis Fernando Duque Ramirez, and Sánchez-Páramo, Carolina, and Norbert Schady. 2003."Off and Jesús Rodríguez García. 1993."National Household Survey on Running? Technology, Trade, and the Rising Demand for Skilled Drug Abuse". Bogota, Colombia, Escuela Colombiana de Medic- Workers in Latin America." Washington, DC: World Bank Policy ina y Fundación Santa Fé de Bogota. Research Working Paper Series 3015. Rodríguez-Pose, Andrés, and Nicholas Gill. 2003."The Global Santiago Consultores Asociados. 1999. Evaluación Ex-Post Chile Trend Towards Devolution and its Implications." Environment Joven Fase II. Santiago de Chile: Santiago Consultores Asocia- and Planning C: Government and Policy 21(3):333­51. dos. Available online at http://www.cinterfor.org.uy/public/ Rosati, Furio. Forthcoming."Child Labour and Youth Employ- spanish/region/ampro/cinterfor/temas/youth/doc/not/libro225/ ment: Ethiopia Country Study." In World Bank (eds.) Youth in libro225.pdf. Africa's Labor Market. Washington, DC: World Bank. Sathar, Zeba A., Minhaj ul Haque, Azeema Faizunnissa, Munawar Rosati, Furio Camillo, and Mariacristina Rossi. 2003."Chil- Sultana, Cynthia B. Lloyd, Judith A. Diers, and Monica Grant. dren's Working Hours and School Enrollment: Evidence 2002. Adolescents and Youth in Pakistan 2001-2002: A Nationally From Pakistan and Nicaragua." World Bank Economic Review Representative Survey. Islamabad: UNICEF. 17(2):283­95. Sawada, Yasuyuki, and Mikhail Lokshin. 2001."Household School- Rosen, James E. 2000."Contracting for Reproductive Health Care: ing Decisions in Rural Pakistan." Washington, DC: World Bank A Guide." Washington, DC: World Bank, Health Nutriton and Policy Research Working Paper Series 2541. Population Discussion Paper 28900. Sawaya, A. L., P. Martins, D. Hoffman, and S. B. Roberts. 2003. Rowlands, S., H. Devalia, R. Lawrenson, J. Logie, and B. Ineichen. "The Link Between Childhood Undernutrition and Risk of 2000."Repeated Use of Hormonal Emergency Contraception Chronic Diseases in Adulthood: A Case Study of Brazil." Nutri- by Younger Women in the UK." Journal of Family Planning and tion Reviews 61(5):168­75. Reproductive Health Care 26(3):138­43. Schmitter, Philippe C., and Alexander H. Treschel. 2005. Green Roy, Olivier. 2004. Globalised Islam: The Search for a New Ummah. Paper on the Future of Democracy in Europe for The Council of London, U.K.: C. Hurst & Co. Ltd. Europe. Paris: Council of Europe. Available online at http:// Rubio, Mauricio. 2005."La Mara, Trucha y Voraz." Inter-American www.coe.int/t/e/integrated_projects/democracy/05_Key_ Development Bank. Washington, DC. Processed. texts/02_Green_Paper/default.asp. References 265 Scholl, T. O., M. L. Hediger, J. Huang, F. E. Johnson, W. Smith, and Shanahan, Michael J., Erik J. Porfeli, Jeylan T. Mortimer, and Lance I. G. Ances. 1992."Young Maternal Age and Parity. Influences on D. Erickson. 2005."Subjective Age Identity and the Transition to Pregnancy Outcome." Annals of Epidemiology 2:565­75. Adulthood: When Do Adolescents Become Adults?" In Richard Scholl, T. O., R. W. Hediger, D. H. Salmon, H. Belsky, and I G. A. Settersten Jr., Frank F. Furstenberg Jr., and Rubén G. Rum- Ances. 1989."Association Between Low Gynaecological Age and baut, (eds.), On the Frontier of Adulthood: Theory, Research, and Preterm Birth." Paediatric and Perinatal Epidemiology 3:357­66. Public Policy. Chicago, IL: University of Chicago Press. Schultz, T. Paul. 2000. Final Report: The Impact of PROGRESA on Shaw, Margaret. 2001. Investing in Youth: International Approaches School Enrollments. Washington, DC: International Food Policy to Preventing Crime and Victimization. Montreal, Quebec, Can- and Research Institute. ada: International Center for the Prevention of Crime. ------. 2002."Why Governments Should Invest More to Educate Sheley, Joseph F., and James D. Wright. 1993. Gun Acquisition and Girls." World Development 30(2):207­25. Possession in Selected Juvenile Samples. Washington, DC: US ------. 2003."Evidence of Return to Schooling in Africa from Department of Justice, National Institute of Justice. Household Surveys: Monitoring and Restructuring the Mar- Sherman, Lawrence W., Denise C. Gottfreddson, Doris L. Macken- ket for Education." New Haven, CT: Yale University, Economic zie, Joun Eck, Peter Reuter, and Shawn D. Bushway. 1998. Pre- Growth Center Discussion Paper No. 875. Available online at venting Crime: What Works, What Doesn't, What's Promising: A http://www.econ.yale.edu/growth_pdf/cdp875.pdf. Report to the U.S. Congress. Washington DC: National Institute ------. 2004."School Subsidies for the Poor: Evaluating the Mexi- of Justice. can Progresa Poverty Program." Journal of Development Eco- Shonkoff, Jack, and Deborah Phillips, eds. 2000. From Neurons to nomics 74(1):199­250. Neighborhoods: The Science of Early Childhood Development. Schweinhart, Lawrence J., Helen V. Barnes, and David P. Weikart. Washington, DC: National Academy Press. 1993. Significant Benefits: The High-Scope Perry Pre-school Study Siddiqua, Yasmin, and M. Kabir. 2002."Adolescent Reproductive through Age 27. Ypsilanti, MI: High Scope Press. Health: What Are the Lessons Learned from the Intervention Sciolino, Elaine, Brian Wingfield, and Elisabetta Povoledo. 2005. Projects." Asia-Pacific Population Journal 17(3):79­100. "From Tapes, a Chilling Voice of Islamic Radicalism in Europe." Simon, Arthy. 2003. Ex Combatant Reintegration: Key Issues for Pol- The New York Times, November 18. Page: 1. icy Makers and Practitioners, Based on Lessons from Sierra Leone, SCN News. 2006."Adolescence." U.N.Standing Committee on Phase 3 evaluation. London: DFID. Nutrition 31 Simonet, Daniel. 2004."The AIDS Epidemic and Migrants in Sen, Amartya. 1985."Well-being, Agency and Freedom, The Dewey South Asia and South-East Asia." International Migration Lectures 1984." Journal of Philosophy 82(4):169­221. 42(5):35­67. ------. 1999. Development as Freedom. New York, NY: Random Singer, Peter Warren. 2005. Children at War. New York, NY: Pan- House, Inc. theon Books. Senderowitz, Judith. 1995."Adolescent Health: Reassessing the Pas- Singh, Susheela. 1998."Adolescent Childbearing in Develop- sage to Adulthood." Washington, DC: World Bank Discussion ing Countries: A Global Review." Studies in Family Planning Paper 272. 29(2):117­36. ------. 1999. Making Reproductive Health Services Youth Friendly. Singh, Susheela, and Jacqueline E. Darroch. 2000."Adolescent Washington, DC: Focus on Young Adults: Research Program Pregnancy and Childbearing: Levels and Trends in Developed and Policy Series. Countries." Family Planning Perspectives 32(1):14­23. Senderowitz, Judith, and John Paxman. 1985."Adolescent Fertility: Singh, Susheela, and Renee Samara. 1996."Early Marriage Among Worldwide Concerns." Population Bulletin 40(2):3­51. Women in Developing Countries." International Family Plan- Serneels, Pieter. 2004."The Nature of Unemployment in Urban ning Perspectives 22(4):148­75. Ethiopia." Oxford, U.K.: CSAE (Oxford) Working Paper Singh, Susheela, Deirdre Wulf, Renee Samara, and Yvette P. Cuca. 2004-18. 2000."Gender Differences in the Timing of First Intercourse: Seshadri, S., and T. Gopaldas. 1989."Impact of Iron Supplemen- Data from 14 Countries." International Family Planning Perspec- tation on Cognitive Functions in Preschool and School-aged tives 26(1):21­9. Children: The Indian Experience." American Journal of Clinical Singleton, Nicola, Alison Lee, and Howard Meltzer. 2001. Psy- Nutrition 50(3):675S­84S. chiatric Morbidity among Adults Living in Private Households Seyboldt, Taylor, eds. 2000. SIPRI Yearbook 2000: Armaments, Dis- 2000: Technical Report. London: HMSO Office for National armament and International Security. Oxford, U.K.: Oxford Uni- Statististics. versity Press. Sjaastad, Larry A. 1962."The Costs and Returns of Human Migra- Shafii, Taraneh, Katherine Stovel, Robert Davis, and King Holmes. tion." Journal of Political Economy 70(5):80­93. 2004."Is Condom Use Habit Forming? Condom Use at Sexual Skoufias, Emmanuel. 2001. Progressa and Its Impact on the Human Debut and Subsequent Condom Use." Sexually Transmitted Dis- Capital and Welfare of Household in Rural Mexico: A Synthesis of eases 31(6):366­72. the Results of an Evaluation by IFPRI. Washington, DC: Interna- Shah, I., and E. Ahman. 2004a."Age Patterns of Unsafe Abortion tional Food Policy Research Institute (IFPRI). in Developing Country Regions." Reproductive Health Matters, Skoufias, Emmanuel, Benjamin Davis, and Sergio de la Vega. 2001. 12(24 (Abortion law, policy and practice supplement)):9­17. "Targeting the Poor: An Evaluation of the Selection of House- ------, eds. 2004b. Unsafe Abortion: Global and Regional Estimates holds into PROGRESA." World Development 29(10):1769­84. of the Incidence of Unsafe Abortion and Associated Mortality in Skoufias, Emmanuel, and Susan W. Parker. 2001."Conditional 2000, 4th ed. Geneva: World Health Organization. Cash Transfers and Their Impact on Child Work and Schooling: 266 WORLD DEVELOPMENT REPORT 2007 Evidence from the PROGRESA Program in Mexico." Economia Stepick, Alex, Carol Dutton Stepick, and Philip Kretsedemas. 2001. 2(1):45­86. Civic Engagement of Haitian Immigrants and Haitian Ameri- Slavin, R. E. 1996. Success or Fall. Lisse, The Netherlands: Swets & cans In Miami-Dade County. Miami, Florida: Haitian American Zeitlinger. Foundation, Human Services Coalition of Dade County and Kellog Foundation. Slavin, R. E., and R. Cooper. 1999."Improving Intergroup Rela- tions: Lessons Learned from Cooperative Learning Programs." Stern, Jessica. 2004. Terror in the Name of God: Why Religious Mili- Journal of Social Issues 55(4):647­64. tants Kill. New York, NY: HarperCollins Publishers Inc. Slaymaker, Emma, N. Walker, B. Zaba, and M. Collumbien. 2004. Stern, Nicholas, Jean-Jacques Dethier, and F. Halsey Rogers. 2005. "Comparative Risk Assessment: Unsafe Sex." In M. Ezzati, A. Growth and Empowerment: Making Development Happen. Cam- Lopez, A. Rodgers, and C. Murray, (eds.), Comparative Quan- bridge, MA: MIT Press. tification of Health Risks: Global and Regional Burden of Dis- Sternberg, Robert. 1985. Beyond IQ: A Triarchic Theory of Human ease due to Selected Major Risk Factors. Geneva: World Health Intelligence. Cambridge: Cambridge University Press. Organization. Stockard, Jean, and Robert M. O'Brien. 2002."Cohort Effects on Smith, Daniel Jordan. 2004."The Bakassi Boys: Vigilantism, Vio- Suicide Rates: International Variations." American Sociological lence, and Political Imagination in Nigeria." Cultural Anthropol- Review 67(6):854­72. ogy 19(3):429­55. Stohl, Rachel. 2001. Global Report on Child Soldiers Released. Smith, P. K., B. Bogin, and D. Bishai. 2005."Are Time Preference Washington, DC: Center for Defense Information. and Body Mass Index Associated? Evidence from the National Stolle, Deitlind, and Marc Hooghe. 2004."Review Article: Inac- Longitudinal Survey of Youth." Economics & Human Biology curate, Exceptional, One-Sided or Irrelevant? The Debate 3(2):259­70. about the Alleged Decline of Social Capital and Civic Engage- Soewondo, S., M. Husaini, and E. Pollitt. 1989."Effects of Iron ment in Western Societies." British Journal of Political Science Deficiency on Attention and Learning Processes of Preschool 35(1):149­67. Children." American Journal of Clinical Nutrition 50(3):667­74. Strauss, John, and Duncan Thomas. 1995."Human Resources: Solinger, Dorothy. 1999. Contesting Citizenship in Urban China: Empirical Modeling of Household and Family Decisions." Peasant Migrants, the State, and the Logic of Market. Berkeley, In Jere R. Behrman and T. N. Srinivasan, (eds.), Handbook of CA: University of California Press. Development Economics Volume 3A. Amsterdam: Elsevier. Span Jr., Milton G. 2000. Remediation: A Must for the 21st Century Student Partnership Worldwide. 2005. The Regai Dzive Shiri Pro- Learning Society. Denver, CO: Education Commission of the gramme: 2002-2007. London: Student Partnership Worldwide. States. Available online at http://www.spw.org/. Sparrow, Robert. 2004."Protecting Education for the Poor in Study Group on Female Genital Mutilation and Obstetric Out- Times of Crisis: An Evaluation of the Scholarships Program in come. 2006."Female Genital Mutilation and Obstetric Out- Indonesia." Jakarta, Indonesia: East Asian Bureau of Economic come: WHO Collaborative Prospective Study in Six African Research, Development Economics Working Papers 96. Countries." Lancet 367:1835­41. Sprinzak, Ehud. 1998."The Psychopolitical Formation of Extreme Summers, Lawrence H. 1992."Investing in All the People." Pakistan Left Terrorism in a Democracy: The Case of the Weathermen." Development Review 31(4):367­93. In Walter Reich, (eds.), Origins of Terrorism: Psychologies, Ideolo- ------. 1994."Investing in All the People: Educating Women in gies, Theologies, States of Mind. Washington, DC: Woodrow Wil- Developing Countries." Washingon, DC: World Bank, Economic son Center Press. Development Institute Seminar Paper 45. Stanback, John, and K. A. Twum-Baah. 2001."Why Do Family Swedish International Development Agency (SIDA). 2005. Cedeca- Planning Providers Restrict Access to Services? An Examina- Ceara Project. Stockholm: Swedish International Development tion in Ghana." International Family Planning Perspectives Agency (SIDA). 27(1):37­41. Tan, Hong W. 2005."In-service Skills Upgrading and Training Pol- Stanton, Bonita F., Xiaoming Li, Joshua Kahihuata, Ann M. icy: Global and Regional Perspectives." Paper presented at the Fitzgerald, Simeone Neumbo, Geraldus Kanduuombe, Izabel MNA Job Creation and Skills Development Conference. Cairo. B. Ricardo, Jennifer S. Galbraith, Nancy Terreri, Irene Guevara, Tan, Hong W., and Geeta Batra. 1995."Enterprise Training in Hannu Shipena, Johan Strijdom, Rebecca Clemens, and R. F. Developing Countries: Incidence, Productivity Effects, and Pol- Zimba. 1999."Increased Protected Sex and Abstinence among icy Implications." Washington, DC: World Bank, Private Sector Namibian Youth Following a HIV Risk-Reduction Intervention: Development Department Working Paper 15373. a Randomized, Longitudinal Study." AIDS 12(18):2473­80. Tan, J. P., J. Lane, and G. Lassibille. 1999."Outcomes in Philippine Stapleton, David C., and Rirchard V. Burkhauser, eds. 2003. Decline Elementary Schools: An Evaluation of Four Experiments." World in Employment of People with Disabilities: A Policy Puzzle. Kal- Bank Economic Review 13(3):493­508. amazoo, MI: W.E. Upjohn Institute for Employment Research. Taylor, Maxwell. 1988. The Terrorist. London, U.K.: Brassey's. Steinberg, Laurence, and Elizabeth Cauffman. 1996."Maturity of Teixeira, Paulo R., Marco Antônio Vitória, and Jhoney Barcarolo. Judgement in Adolescence: Psychosocial Factors in Adolescent 2003."The Brazilian Experience in Providing Universal Access Decision Making." Law and Human Behavior 20(3):249­72. to Antiretroviral Therapy." In Dumoulin J. P. Moatti, B. Coriat, Y. Steinberger, Michael. 2001."So, Are Civilizations at War?" The Souteyrand, T. Barnett, and Y. A. Flori, (eds.), Economics of AIDS Observer, October 21.Available online at http://observer.guard- and Access to HIV/AIDS Care in Developing Countries, Issues and ian.co.uk/islam/story/0,1442,577982,00.html. Challenges. Paris: Agence Nationale de Recherches sur le Sida. Steinmo, Sven, Kathleen Thelen, and Frank Longstreth, eds. 1992. Temin, Miriam J., Friday E. Okonofua, Francesca O. Omorodion, Structuring Politics: Historical Institutionalism in Comparative Elisha P. Renne, Paul Coplan, H. Kris Heggenhougen, and Joan Analysis. New York, NY: Cambridge University Press. Kaufman. 1999."Perceptions of Sexual Behavior and Knowledge References 267 About Sexually Transmitted Diseases Among Adolescents in U.S.Department of State. 2005a. Trafficking in Persons Report 2005. Benin City, Nigeria." International Family Planning Perspectives Washington, DC: U. S. Department of State, Office of the Under 25(4):186­190. Secretary for Global Affairs. Thailand Ministry of Public Health, and World Bank. 2005. ------. 2005b. Visa Bulletin for October 2005. Washington, DC: Expanding Access to Antiretroviral Treatment in Thailand: United States, Department of State. Available online at http:// Achieving Treatment Benefits while Promoting Effective Preven- travel.state.gov/visa/frvi/bulletin/bulletin_2631.html. tion. Bangkok and Washington, DC: Thailand Ministry of Pub- UNAIDS. 2000. Report on the Global HIV/AIDS Epidemic 2000. lic Health and World Bank. Geneva: UNAIDS. The Global Fund for AIDS, TB and Malaria. 2005. Development of UNAIDS, and UNODCCP. 2000. Drug Use and HIV Vulnerability: a Comprehensive National Response to HIV/AIDS that Includes Policy Research Study in Asia. Bangkok, Thailand: UNAIDS. Adequate Prevention, Treatment, Care and Support for Those UNDCP. 2003. Investing in Drug Abuse Treatment: A Discussion Affected. Freetown: The Global Fund for AIDS, TB and Malaria. Paper for Policy Makers. New York: UNDCP. The Office of Juvenile Justice and Delinquency Prevention UNDP. 1995. The World's Women 1995: Trends & Statistics. New (OJJDP). 1998. Juvenile Justice Bulletin. Washington, DC: U.S. York: United Nations, Department of Economic and Social Department of Justice, Office of Justice Programs, Office of Affairs Statistics Division. Juvenile Justice and Delinquency Prevention. UNESCO. 2004a. Gender and Education for All Global Monitoring Thorn, Kristian, Lauritz Holm-Nielsen, and Samuel Jeppesen. 2004. Report 2003/4: The Leap to Equality. Paris: UNESCO. "Approaches to Results-Based Funding in Tertiary Education: ------. 2004b. Global Monitoring Report 2005. Education for All: Identifying Finance Reform Options for Chile."Washington, DC: The Quality Imperative. Paris: UNESCO. World Bank Policy Research Working Paper Series 3436. ------. 2005. World Education Indicators 2005. Paris: UNESCO, Thornberry, Terence P. 1998."Membership in Youth Gangs and Institute for Statistics. Involvement in Serious and Violent Offending." In Rolf Loeber UNESCO Institute for Statistics. 2005. Global Education Digest: and David P. Farrington, (eds.), Serious and Violent Offenders: Comparing Education Statistics Across the World. Montreal: Risk Factors and Successful Interventions. Thousand Oaks, CA: UNESCO Institute for Statistics. Sage Publications, Inc. UNICEF. 2000."Young People in Changing Societies." Florence: Thorup, Cathryn L., and Sheila Kinkade. 2005. What Works in UNICEF Innocenti Research Center Report 7. Available online Youth Engagement in the Balkans. Baltimore, MD: International at http://www.unicef-icdc.org. Youth Foundation. Available online at http://www.iyfnet.org/ section.cfm/31/223. ------. 2005a. Innocenti Digest: Juvenile Justice. Florence, Italy: Innocenti Research Center. Titma, Mikk, and Nancy Brandon Tuma. 2005."Human Agency in the Transition from Communism." In K. Warner Schaie and ------. 2005b. The Impact of Conflict on Women and Girls in West Glen Elder, (eds.), Historical Influences on Lives and Aging. New and Central Africa and the UNICEF Response. Paris: UNICEF. York, NY: Springer Publishing Company. ------. 2005c. The State of the World's Children 2006. New York, Todd, Petra, and Kenneth I. Wolpin. 2003."Using a Social Experi- NY: UNICEF. ment to Validate a Dynamic Behavioral Model of Child School- UNICEF, UNAIDS, and WHO. 2002. Young Peo- ing and Fertility: Assessing the Impact of a School Subsidy ple and HIV/AIDS: Opportunity in Crisis. New York: Program in Mexico." University of Pennsylvania. Philadelphia. UNICEF. Available online at http://www.unicef. Processed. org/publications/files/pub_youngpeople_hivaids_en.pdf. Topel, Robert H., and Michael P. Ward. 1992."Job Mobility and United Kingdom Home Office. 2005. Accession Monitoring Report. the Careers of Young Men." Quarterly Journal of Economics London: United Kingdom Home Office. Available online at 107(2):439­79. http://www.workingintheuk.gov.uk/ind/en/home/0/reports/ Torney-Purta, Judith, Rainer Lehmann, Hans Oswald, and Wol- accession_monitoring.Maincontent.0018.file.tmp/Accession_ fram Schulz. 2001. Citizenship and Education in Twenty-Eight Monitoring_ReportNW2%5B1%5D.2doc.pdf. Countries: Civic Knowledge and Engagement at Age Fourteen. United Nations. 2003. World Youth Report 2003: Global Situation Amsterdam, Netherlands: International Association for the of Young People. New York: United Nations, Department of Eco- Evaluation of Educational Achievement (IEA). Available online nomic and Social Affairs. at http://www.wam.umd.edu/~jtpurta/interreport.htm. ------. 2005. World Population Prospects: The 2004 Revision (CD- Townsend, J., P. Roderick, and J. Cooper. 1994."Cigarette Smok- ROM). New York: United Nations Population Division, Depart- ing by Socio-Economic Group, Sex and Age: Effects of ment of Economic and Social Affairs. Price, Income and Health Publicity." British Medical Journal United Nations Economic and Social Council. 2001. World Situa- 309:923­26. tion with Regard to Drug Abuse, with Particular Reference to Chil- Tudawe, Indra. 2001."Chronic Poverty and Development Policy dren and Youth. Vienna: United Nations. in Sri Lanka: Overview Study." Manchester, U.K.: CPRC Work- United Nations Office on Drugs and Crime. 2004. Solvent Abuse ing Paper 9. Available online at http://www.chronicpoverty.org/ Among Street Children in Pakistan. Islamabad: United Nations pdfs/09Tudawe.pdf. System in Pakistan. Turni_ki, Maja. 2004. The Place of Participation in the Recovery of UNODCCP. 2002. Street Children of Cairo and Alexandria: Drug Identity in Adolescents and Young Adults Affected by War and Dis- Abuse Trends, Consequences and Response. El Cairo: UNODCCP placement in Croatia. New York, NY: The Graduate Center, City Regional Office in El Cairo. University of New York. Upchurch, Dawn M., Lee A. Lillard, and Constantijn W. A. Panis. Tzannatos, Zafiris. 1999."Women and Labor Market Changes in 2002."Nonmarital Childbearing: Influences of Education, Mar- the Global Economy." World Development 27(3):551­69. riage, and Fertility." Demography 39(2):311­29. 268 WORLD DEVELOPMENT REPORT 2007 Urdal, Henrik. 2004."The Devil in the Demographics: The Effect Walker, S. P., S. M. Grantham-Mcgregor, C. A. Powell, and S. M. of Youth Bulges on Domestic Armed Conflicts, 1950-2000." Chang. 2005."Effects of Early Childhood Psychosocial Stimula- Washington, DC: World Bank, Social Development Papers, tion and Nutritional Supplementation on Cognition and Educa- Conflict Prevention and Reconstruction 14. tion in Growth-stunted Jamaican Children: Prospective Cohort Utas, Mats. 2004."Building the Future: The Reintegration and Study." Lancet 366(9499):1804­7. Marginalisation of Ex-combatant Youth in Liberia." In Paul Walzer, Michael. 1989."Citizenship." In Terrence Ball, James Farr, Richards, (eds.), No Peace, No War: An Anthropology of Contem- and Russell L Hanson, (eds.), Political Innovation and Concep- porary Armed Conflicts. Oxford, U.K.: James Currey. tual Change. New York, NY: Cambridge University Press. Van der Gaag, Jacques, and Jee-Peng Tan. 1998. The Benefits of Washington, Ebonya. 2006."Female Socialization: How Daugh- Early Child Development Programs, An Economic Analysis. Wash- ters Affect their Legislators Father's Voting on Women's Issues." ington, DC: World Bank, Human Development Network. Cambridge, MA: National Bureau of Economic Research Work- van Eekelen, Willem, Loetta de Luca, and Nagwa Ismail. 2001. ing Paper Series 11924. "Youth Employment in Egypt. InFocus Programme on Skills, Wattenberg, Martin P. 2006. Is Voting for Young People? New York, Knowledge, and Employability Skills." Geneve: International NY: Longman. Labour Office, Skills Working Paper 2. Available online at http:// Watts, Anthony G., and David H. Fretwell. 2004."Public Strate- www.ilo.org/public/english/employment/skills/youth/down- gies for Designing Career Information and Guidance Systems in load/skillswp2.pdf. Middle-Income and Transition Economies." World Bank. Wash- Van Ness, Daniel W. 2005."An Overview of Restorative Justice ington, DC. Processed. Around the World." Paper presented at the 11th United Nations White, Howard. 2005. Maintaining Momentum towards the MDGs: Congress on Crime Prevention, and Criminal Justice. Bangkok, An Impact Evaluation of Interventions to Improve Maternal and Thailand. Child Health and Nutrition Outcomes in Bangladesh. Washing- Varcoe, Karen P., Allen Martin, Zana Devitto, and Charles Go. ton, DC: World Bank, Operation Evaluation Division. 2005."Using A Financial Education Curriculum For Teens." Whiteside, Katherine, John Bellows, Mame Fatou Diagne, Benn Journal of Financial Counseling and Planning 16(1):63­71. Eifert, Rachel Glennerster, Edward Miguel, David Zimmer, and Vegas, Emiliana, and Ilana Umansky. 2005. Improving Teaching and Yongmei Zhou. 2006."Baseline Measures of Social Capital in Learning through Effective Incentives: What Can We Learn from GoBifo Communities: Report for IRCBP." Processed. Education Reforms in Latin America? Washington, DC: World WHO. 1998. The Second Decade: Improving Adolescent Health and Bank. Development. Geneva: World Health Organization. Available Venkatraman, Arjunamurthy, and Julie Falconer. 1998. Rejuvenat- online at http://www.who.int/reproductive-health/docs/adoles- ing India's Decimated Forests through Joint Action: Lessons from centhealth.html. Andhra Pradesh. Washington, DC: World Bank. ------. 1999. Volatile Solvent Abuse: A Global Overview. Geneva: Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady. 1995. World Health Organization. Voice and Equality. Cambridge, U.K.: Cambridge University ------. 2002a. The World Health Report 2002: Reducing Risks, Pro- Press. moting Healthy Life. Geneva: World Health Organization. Verner, Dorte, Andreas Blom, and Lauritz Holm-Nielsen. 2001. ------. 2002b. World Report on Violence and Health: Summary. "Education, Earnings, and Inequality in Brazil 1982-1998." Geneva: World Health Organization. Washington, DC: World Bank Policy Research Working Paper ------. 2003a. Adolescent Pregnancy: Unmet Needs and Undone Series 2686. Deeds. A Review of the Literature and Programs. Geneva: World Vietnam Ministry of Health and General Statistics Office, UNI- Health Organization. CEF, and WHO. 2005. Survey Assessment of Vietnamese Youth. ------. 2003b. Health and Development through Physical Activity Hanoi: Government of Vietnam. and Sport. Geneva: World Health Organization. Villaveces, Andrés, Peter Cummings, Victoria E. Espitia, Thomas ------. 2003c. Pregnancy, Childbirth, Postpartum and Newborn D. Koepsell, Barbara McKnight, and Arthur L. Kellermann. Care: A Guide for Essential Practice. Geneva: World Health 2000."Effect of a Ban on Carrying Firearms on Homicide Rates Organization. in 2 Colombian Cities." Journal of American Medical Association ------. 2004. Global Status Report on Alcohol. Geneva: World 283:1205­9. Health Organization. Visinntini, R., E. Campanini, A. Fossati, M. Bagnato, L. Novella, ------. 2005a. Effectiveness of Drug Dependence Treatment in Pre- and C. Maffei. 1996."Psychological Stress in Nurses' Relation- venting HIV among Injecting Drug Users. Geneva: World Health ships with HIV-infected Patients: The Risk of Burnout Syn- Organization. drome." AIDS Care 8(2):183­94. WHO, and World Bank, eds. 2005. World Report on Road Traffic Viva Rio. 2005. Youth Service: A Policy for Preventing and Providing Injury Prevention. Geneva and Washington, DC: World Health Alternatives to Youth Involvement in Urban Violence in Brazil. Rio Organization and World Bank. de Janeiro, Brazil: Viva Rio. WHO, Regional Office for Europe. 2005b."Status Paper on Pris- Vodopivec, Matija. 2005."Wage and Productivity Differentials ons, Drugs and Harm Reduction." Copenhagen: World Health during Slovenia's Transition: Matched Employer-Employee Evi- Organization Regional Office for Europe EUR/05/5049062. dence." Senior Honors thesis. Macalester College. Willis, Robert J., and John G. Haaga. 1996."Economic Approaches Waiselfisz, Julio Jacobo, and Maria Maciel. 2003. Revertendo Vio- to Understanding Nonmarital Fertility." Population and Develop- lências, Semeando Futuros: Avaliação de Impacto do Programa ment Review 22(Supp.):67­86. Abrindo Espaços no Rio de Janeiro e em Pernambuco. Brasilia: Wilson, David B., Doris L. Mackenzie, and Fawn Ngo Mitchell. UNESCO Office Brasilia. 2005. Effects of Correctional Boot Camps on Offending. Canberra, References 269 Australia: Campbell Collaboration's Crime & Justice Coordinat- ------. 2002b. Russian Economic Report # 3. Washington, DC: ing Group, Australian Institute of Criminology. World Bank. Wintour, Patrick. 2006."Brown Backs Votes at 16 in Radical ------. 2003a."Caribbean Youth Report: Issues and Policy Direc- Shakeup of Politics." The Guardian, February 27.Available online tions." World Bank. Washington, DC. Processed. at http://politics.guardian.co.uk/apathy/story/0,,1718744,00 ------. 2003b. Closing the Gap in Education and Technology. .html. Washington, DC: World Bank. Witoelar, Finman, Pungpond Rukumnuaykit, and John Strauss. ------. 2003c. Project Performance Assessment Report, Bangladesh 2005."Smoking Behavior Among Youth in a Developing Coun- Female Secondary School Assistance Project (Credit 2469). Wash- try: Case of Indonesia." Princeton University. Princeton, NJ. ington, DC: World Bank, Operations Evaluation Department. Available online at http://paa2006.princeton.edu/download. ------. 2003d. Republic of Tunisia: Employment Strategy. Report # aspx?submissionId=60756. Processed. 25456-TUN. Washington, DC: World Bank. Woessmann, Ludger. 2003."Schooling Resources, Educational ------. 2003e. Vietnam Development Report 2004: Poverty. Report Institutions, and Student Performance: The International Evi- No. 27130-VN. World Bank: Washington, DC dence." Oxford Bulletin of Economics and Statistics 65(2):117­70. ------. 2003f. World Development Report 2004: Making Services Woessmann, Ludger, and Eric A. Hanushek. Forthcoming."Does Work for Poor People. New York: Oxford University Press. Educational Tracking Affect Performance and Inequality? Differ- ------. 2004a. Addressing HIV/AIDS in East Asia and the Pacific. ences-in-Differences Evidence across Countries."Economic Journal. Washington, DC: World Bank. Women´s Commission for Refugee Women and Children. 2002. ------. 2004b. Gender and Development in the Middle East and Precious Resources Adolescents in the Reconstruction of Sierra North Africa: Women in the Public Sphere. Washington, DC: Leone, Participatory Research Study with Adolescents and Youth World Bank. in Sierra Leone. New York: Women´s Commission for Refugee Women and Children. Available online at http://www.reliefweb. ------. 2004c. Serbia and Montenegro: Poverty Reduction Strategy int/library/documents/2002/wcrwc-sie-31oct.pdf. Paper and Joint IDA-IMF Staff Assessment of the PRSP, 9. Wash- ington, DC: World Bank. Wong, Emelita L., Barry M. Popkin, David K. Guiley, and John S. Akin. 1987."Accessibility, Quality of Care and Prenatal Care in ------. 2004d. Timor-Leste Education Since Independence From the Philippines." Social Science & Medicine 24(11):927­44. Reconstruction to Sustainable Improvement. Washington, DC: World Bank, Human Development Sector Report EAP. Woolard, Jennifer L., Candice Odgers, Lonn Lanza-Kaduce, and Hayley Daglis. 2005."Juveniles within Adult Correctional Set- ------. 2004e. World Development Report 2005: A Better Invest- tings: Legal Pathways and Developmental Considerations." ment Climate for Everyone. New York: Oxford University Press. International Journal of Forensic Mental Health 4(1):1­18. ------. 2005a. Bridging the Gender Gap: Opportunities and Chal- lenges, Pakistan Country Gender Assessment. Washington, DC: World Bank. 1990. World Development Report 1990: Poverty. New World Bank. York: Oxford University Press. ------. 2005b. Colombia: Contracting Education Services. Wash- ------. 1993a. World Bank Policy Research Report 1993. The East ington, DC: World Bank. Asian Miracle: Economic Growth and Public Policy. New York: Oxford University Press. ------. 2005c. Doing Business in 2005: Removing Obstacles to Growth. Washington, DC: World Bank, International Finance ------. 1993b. World Development Report 1993: Investing in Corporation and Oxford University Press. Health. New York: Oxford University Press. ------. 2005d. Dying Too Young: Addressing Premature Mortality ------. 1995. World Development Report 1995: Workers in an Inte- and Ill Health Due to Non-Communicable Diseases and Injuries grating World. New York: Oxford University Press. in the Russian Federation. Washington, DC: World Bank. ------. 1999a. Confronting AIDS: Public Priorities in a Global Epi- ------. 2005e. Economic Growth in the 1990s: Learning from a demic. New York: Oxford University Press. Decade of Reform. Washington, DC: World Bank. ------. 1999b. Curbing the Epidemic: Governments and the Eco- ------. 2005f. Expanding Opportunities and Building Competen- nomics of Tobacco Control. Washington, DC: World Bank. cies for Young People: A New Agenda for Secondary Education. ------. 1999c. Proposed Learning and Innovation Loan in the Washington, DC: World Bank. Amount of $4.75 Million to the Argentine Republic for an Integrated ------. 2005g."Gender Assessment in Vietnam." World Bank. Drug Prevention Pilot Project. Washington, DC: World Bank. Washington, DC. Processed. ------. 2000. World Development Report 1999/2000: Entering the ------. 2005h."Gender Mainstreaming Process in the Lao PDR." 21st Century. New York: Oxford University Press. World Bank. Washington, DC. Processed. ------. 2001a. An Assessment of the Bolsa Escola Programs Report ------. 2005i. Global Economic Prospects 2006: Economic Impli- No. 20208-BR. Washington, DC: World Bank. cations of Remittances and Migration. Washington DC: World ------. 2001b. Caribbean Youth Development, A World Bank Bank. Country Study. Washington, DC: World Bank. ------. 2005j. Global Monitoring Report. Washington, DC: World ------. 2001c. World Bank Policy Research Report 2001: Engender- Bank. ing Development Through Gender Equality In Rights, Resources ------. 2005k. Implementation Completition Report on a Credit to And Voice. New York: Oxford University Press. the Republic of Guyana for the Secondary School Reform Project. ------. 2001d. World Development Report 2000/01: Attacking Pov- Washington, DC: World Bank. erty. New York: Oxford University Press. ------. 2005l. Malaysia: Firm Competitiveness, Investment Cli- ------. 2002a. Constructing Knowledge Societies: New Challenges mate, and Growth. Report N0. 26841-MA. Washington, DC: for Tertiary Education. Washington, DC: World Bank. World Bank. 270 WORLD DEVELOPMENT REPORT 2007 ------. 2005m."Pakistan Country Gender Report (Gray Cover)." World Bank. Available online at http://www1.worldbank. World Bank. Washington, DC. Processed. org/education/edstats/. ------. 2005n. Policy Note on Employment of Migrants in China. World Links Impact Evaluation Series. 2002. Uganda Tracer Study: Washington, DC: World Bank. An Impact Assessment of Information and Communications Tech- ------. 2005o. Republic of Uruguay. Policy Notes. Report No. nologies on World Links Participating Students. Washington DC: 31338-UY. Washington, DC: World Bank. World Links. ------. 2005p. Tobacco Use in Indonesia. Washington, DC: World Yamano, Takashi, and T. S. Jayne. 2004."Measuring the Impacts of Bank. Working-Age Adult Mortality on Small-Scale Farm Households in Kenya." World Development 32(1):91­119. ------. 2005q. Vietnam Business: Vietnam Development Report 2006. Report No. 34474-VN. World Bank: Washington, DC Yang, Dean. 2004."International Migration, Human Capital, ------. 2005r. World Development Report 2006: Equity and Devel- and Entrepreneurship: Evidence from Philippine Migrants' opment. New York: Oxford University Press. Exchange Rate Shock." Washington D.C: World Bank policy research working paper 3578. ------. 2005s."Youth Development in Kenya." World Bank. Washington, DC. Processed. ------. 2005."Why Do Inmigrants Return to Poor Countries? Evidence from Philippines Migrant's Responses to Exchange ------. 2005t."Youth in Post-Conflict Settings." Washington, DC: Rate Shocks." University of Michigan. Ann Arbor. Processed. Youth Development Notes Volume 1, Number 1. Yap, Yoon-Tien, Guilherme Sedlacek, and Peter F. Orazem. 2001. ------. 2006a."Crime, Violence, and Economic Development in "Limiting Child Labor Through Behavior-Based Income Trans- Brazil: Elements for Effective Public Policy." World Bank. Wash- fers: An Experimental Evaluation of the PETI Program in Rural ington, DC. Processed. Brazil." World Bank. Washington, DC. Processed. ------. 2006b. Global Monitoring Report 2006. Washington, DC: Yip, R. 1994."Iron Deficiency: Contemporary Scientific Issues and World Bank. International Programmatic Approaches." Journal of Nutrition ------. 2006c. Information and Communications for Development: 124(8):1479S­1490S. Global Trends and Policies. Washington, DC: World Bank. Yip, R., and P. R. Dallman. 1996."Iron." In E. E. Ziegler and L. J. Jr. ------. 2006d. Mobilizing Resources for Secondary Education in Filer, (eds.), Present Knowledge of Nutrition. 7th ed. Washington, EAP and LAC. Washington, DC: World Bank. DC: International Life Sciences Institute Press. ------. 2006e. Public Training Reform Issues in Colombia: The Zabin, Laurie Schwab, and Karungari Kiragu. 1998."The Health Case of SENA. Report # 27752. Washington, DC: World Bank. Consequences of Adolescent Sexual and Fertility Behavior in ------. 2006f. Repositioning Nutrition as Central to Development: Sub-Saharan Africa." Studies in Family Planning 29(2):210­32. A Strategy for Large Scale Action. Washington, DC: World Bank. Zeldin, Shepherd, Annette Kusgen McDaniel, Dimitri Topitzes, and ------. 2006g."School-to-Work Transition and Youth Inclusion Matt Calvert. 2000. Youth in Decision-Making: A Study on the in Southern Russia." World Bank. Washington, DC. Processed. Impacts of Youth on Adults and Organizations. Madison, W.I.: Inno- ------. 2006h. World Development Indicators. Washington, DC: vation Center for Community and Youth Development, University World Bank. of Wisconsin-Madison. Available online at http://www.theinnova- ------. 2006i. Youth at Risk in Brazil, Report No. 32310-BR. Wash- tioncenter.org/pdfs/Youth_in_Decision_Making_Brochure.pdf. ington, DC: World Bank. Zins, J. E., R. P. Weissberg, M. C. Wang, and H. J. Walberg, eds. 2004. World Bank, Unesco Institute for Statistics (UIS), and OECD. Building Academic Success on Social and Emotional Llearning: 2006. Education Statistics (EDSTATS). Washington, DC: What Does the Research Say? New York: Teachers College Press. Selected indicators Table A1. Learning Table A2. Program for International Student Assessment Table A3. Going to work Table A4. Child labor Table A5. Staying healthy (1) Table A6. Staying healthy (2) Table A7. Forming families (1) Table A8. Forming families (2) Technical notes Selected world development indicators Introduction Table 1. Key indicators of development Table 2. Poverty Table 3. Millennium Development Goals: eradicating poverty and improving lives Table 4. Economic activity Table 5. Trade, aid, and finance Table 6. Key indicators for other economies Table A1. Learning Enrollment rates Total Male Female Grade 9 completion ratea Survey ages ages ages ages ages ages ages ages ages year 12-14 15-17 18-24 12-14 15-17 18-24 12-14 15-17 18-24 Total Male Female Afghanistan 2003 36 25 11 54 42 18 19 8 3 20 31 6 Albania 2002 89 46 11 90 47 9 88 45 13 35 36 34 Angola 1999 55 45 18 56 48 20 53 41 16 10 11 10 Argentina 2001 97 86 46 97 84 43 98 87 48 78 76 81 Armenia 1999 98 73 19 98 68 13 99 77 25 .. .. .. Azerbaijan 2002 100 81 17 100 79 17 100 83 17 .. .. .. Bangladesh 2000 67 46 17 62 40 24 72 54 11 40 38 41 Benin 2003 65 49 23 72 56 34 56 41 14 48 55 39 Bhutan 2003 67 52 22 71 61 29 63 44 16 46 53 40 Bolivia 2002 54 51 34 54 52 35 55 49 32 31 31 31 Bosnia and Herzegovina 2001 98 90 28 98 93 22 99 87 34 .. .. .. Brazil 2001 95 81 34 95 82 34 95 80 34 57 57 57 Bulgaria 1995 92 77 24 92 80 22 93 74 25 53 56 50 Burkina Faso 2003 32 20 9 36 21 12 28 18 7 18 19 16 Burundi 1998 52 36 19 59 39 21 47 34 17 26 30 22 Cambodia 2004 88 61 15 89 66 21 86 55 11 38 43 34 Cameroon 2001 83 63 28 86 71 34 80 55 22 55 61 49 Cape Verde 2000 91 63 21 91 62 20 91 64 22 59 58 60 Chile 2003 98 91 40 98 91 41 99 91 38 79 80 79 Colombia 2000 85 66 27 84 64 28 86 68 26 50 49 51 Costa Rica 2001 87 66 37 88 64 36 87 68 38 60 58 62 Côte d'Ivoire 2002 61 39 17 69 50 23 51 29 12 36 46 26 Dominican Republic 2004 96 83 40 96 81 37 96 84 42 75 76 75 Ecuador 2004 85 68 33 85 66 32 85 69 35 62 61 63 Egypt, Arab Rep. 1998 84 69 26 88 73 29 81 65 23 64 68 61 El Salvador 2002 87 66 25 86 68 26 87 65 24 50 50 49 Estonia 2000 99 93 42 100 92 43 99 95 41 81 80 82 Fiji 1996 92 68 14 91 65 15 93 71 13 61 60 63 Ghana 1998 85 67 22 88 72 28 81 61 16 .. .. .. Guatemala 2002 71 46 20 79 52 26 63 40 15 35 41 29 Guinea 1994 36 29 13 44 37 22 26 19 6 34 44 24 Guyana 2000 94 64 0 92 62 0 95 66 0 45 42 47 Honduras 2003 77 48 23 77 45 21 78 52 25 47 45 50 Hungary 2002 100 98 50 100 98 50 100 97 51 .. .. .. India 2000 71 51 16 77 57 21 65 43 11 .. .. .. Indonesia 2002 88 62 16 87 62 18 88 61 15 43 44 43 Jamaica 2002 98 74 9 97 70 7 99 78 10 46 42 50 Jordan 2002 87 85 31 86 84 30 87 86 32 69 67 72 Kenya 1997 92 77 22 92 80 28 92 73 17 45 51 38 Kiribati 2000 89 58 9 87 52 8 91 65 9 .. .. .. Kyrgyz Republic 2002 96 90 30 96 89 24 96 90 35 .. .. .. Lesotho 2002 86 60 19 80 58 21 91 61 16 38 40 38 Malawi 1997 90 80 33 91 85 47 88 74 22 .. .. .. Maldives 1998 92 72 21 93 73 19 92 71 24 44 43 45 Marshall Islands 1999 86 70 29 85 69 32 87 70 27 55 55 56 Mauritania 2000 53 39 20 56 43 22 51 36 19 32 31 33 Mexico 2002 90 61 26 90 61 28 89 61 25 47 48 45 Micronesia, Fed. Sts. 2000 86 68 21 84 64 20 88 71 21 49 47 52 Moldova 2002 98 74 24 99 69 23 98 78 25 .. .. .. Mozambique 1996 58 37 10 64 44 17 52 28 4 10 13 8 Namibia 1993 94 84 40 94 83 43 94 84 37 .. .. .. Nepal 1995 61 43 14 71 51 18 51 36 9 0 0 0 Nicaragua 2001 82 59 27 79 54 25 86 63 29 48 42 53 Niger 2002 69 43 25 70 43 27 68 43 22 44 44 44 Nigeria 2003 64 58 36 64 59 42 65 58 30 44 44 43 Pakistan 2001 53 37 10 63 46 13 43 27 7 24 30 18 Palau 2000 96 88 31 95 86 30 96 91 31 76 75 78 Panama 2003 92 76 35 92 73 33 92 79 38 64 61 68 Paraguay 2001 87 64 28 87 61 28 87 67 29 55 53 58 Peru 2002 94 73 29 95 75 29 93 71 29 56 56 56 Poland 2002 100 99 54 100 98 52 100 99 57 .. .. .. Romania 1994 95 80 21 95 80 19 95 81 24 0 0 0 Rwanda 1997 77 40 14 77 42 15 78 39 13 23 23 23 São Tomé and Principe 2000 76 48 14 75 50 15 76 45 14 35 36 34 Senegal 1995 68 38 23 79 42 28 59 34 19 38 43 35 Sierra Leone 2003 77 61 31 81 68 43 72 52 21 51 59 43 Solomon Islands 1999 74 58 19 76 63 24 73 53 15 33 36 29 South Africa 2000 96 90 43 96 91 45 96 90 42 75 77 74 Swaziland 2000 90 76 25 89 78 34 91 75 18 60 64 57 Tajikistan 1999 91 63 10 94 72 15 88 54 6 .. .. .. Tanzania 2000 78 49 9 79 52 11 77 46 6 10 9 11 Thailand 2002 94 77 29 93 71 29 95 82 30 67 62 72 Tonga 1996 95 77 24 94 74 24 96 81 24 70 67 73 Trinidad and Tobago 1992 90 68 19 87 64 17 93 72 21 .. .. .. Turkey 2002 85 49 11 91 59 16 78 40 7 .. .. .. Uganda 2002 92 72 20 93 74 30 92 70 13 46 51 42 Uruguay 2003 96 81 44 95 78 41 96 84 48 72 69 76 Venezuela, RB 2004 93 76 26 92 74 23 95 78 28 59 57 62 Vietnam 2001 88 63 25 89 66 26 87 60 24 56 58 54 Zambia 2002 84 69 24 84 75 30 84 64 19 44 48 42 a. The statistic is computed using the information from nationally representative household surveys on last grade completed and current school participation of individuals ages 10-19 at the time of the survey. Estimates are are based on Kaplan-Meier method which takes into account whether the individual is still in school. 272 Table A2. Program for International Student Assessment Literacy, Math, Problem solving scale, age 15, 2003 age 15, 2003 age 15, 2003 Total Male Female Total Male Female Total Male Female Australia 526 506 545 524 527 522 530 527 533 Austria 491 467 514 506 509 502 506 505 508 Belgium 508 489 526 529 533 525 525 522 527 Brazil 402 384 419 357 365 348 371 374 368 Canada 530 514 546 535 541 530 529 533 532 Czech Republic 489 473 504 516 524 509 516 520 513 Denmark 492 479 505 514 523 506 517 519 514 Finland 543 521 565 544 548 541 548 543 553 France 495 476 514 511 515 507 519 519 520 Germany 492 471 513 503 508 499 513 511 517 Greece 472 453 490 445 455 436 448 450 448 Hong Kong, China 510 494 525 550 552 548 548 545 550 Hungary 483 467 498 490 494 486 501 499 503 Iceland 493 464 522 515 508 523 505 490 520 Indonesia 382 370 394 360 362 359 361 358 365 Ireland 516 501 530 503 510 495 498 499 498 Italy 475 455 495 466 475 457 469 467 471 Japan 498 487 509 534 539 530 547 546 548 Korea, Dem. Rep. 536 526 547 540 552 528 550 554 546 Latvia 490 470 509 483 485 482 483 481 484 Mexico 399 389 410 385 391 380 384 387 382 Netherlands 513 503 524 538 540 535 520 522 518 New Zealand 522 508 535 524 531 516 533 531 534 Norway 500 475 525 495 498 492 490 486 494 Poland 497 477 516 490 493 488 487 486 487 Portugal 477 459 495 466 472 460 470 470 470 Russian Federation 442 428 456 468 474 463 479 480 477 Serbia and Montenegro 412 390 433 437 438 436 420 416 424 Slovak Republic 470 453 486 498 507 489 492 495 488 Spain 480 461 500 485 490 481 482 479 485 Sweden 514 496 533 509 512 506 509 504 514 Switzerland 500 482 518 526 535 518 521 520 523 Thailand 418 396 439 417 415 419 425 418 431 Tunisia 374 362 387 359 365 353 345 346 343 Turkey 443 426 459 423 430 415 408 408 406 United States 495 479 511 483 486 480 477 477 478 Uruguay 434 414 453 422 428 416 411 412 409 273 Table A3. Going to work Labor force Not in the labor force participation rate, ages 15­24 Unemployment rate and not in school, ages 15­24 by urban/rural, by age group by gender, ages 15­24 ages 15­24 Young, Adult, ages ages 15­24 25­49 Total Male Female total total Male Female Urban Rural Total Male Female Afghanistan 2003 39.9 60.8 17.3 7.5 3.9 6.0 12.8 .. .. 46.1 18.0 77.1 Albania 2002 42.5 42.0 43.0 14.2 9.6 16.3 12.1 45.2 4.6 36.2 37.7 34.6 Angola 1999 46.4 48.1 44.9 8.0 3.8 10.5 5.6 9.3 0.4 30.7 26.9 34.1 Argentina 2001 38.6 47.4 29.9 28.8 12.0 27.4 31.1 28.8 .. 13.1 6.8 19.4 Azerbaijan 2002 35.9 41.6 30.3 .. .. .. .. .. .. 25.6 19.2 31.9 Bangladesh 2000 33.9 58.9 7.3 4.1 1.1 3.2 11.6 6.8 3.3 39.9 11.8 69.9 Belarus 2002 30.4 30.5 30.3 21.8 6.3 22.9 20.7 21.9 21.3 .. .. .. Benin 2003 53.9 46.3 61.8 5.9 3.0 6.0 5.9 8.8 4.5 14.5 12.4 16.5 Bhutan 2003 55.8 51.8 59.3 1.4 0.5 1.1 1.5 5.1 0.8 11.6 7.2 15.4 Bolivia 2002 58.9 66.9 51.3 13.7 7.0 9.9 18.4 20.0 5.5 9.0 7.7 10.4 Bosnia and Herzegovina 2004 46.0 .. .. 43.0 19.9 42.5 43.6 .. .. .. .. .. Brazil 2001 61.8 72.7 51.2 17.9 7.2 14.6 22.4 20.5 5.2 13.5 5.8 21.0 Bulgaria 1995 39.7 40.9 38.4 38.6 12.7 37.2 40.1 37.4 40.7 19.0 17.6 20.4 Burkina Faso 2003 79.1 82.4 76.2 3.8 1.8 3.9 3.6 22.1 0.7 8.7 3.1 13.7 Burundi 1998 70.7 67.8 73.1 0.4 0.5 0.6 0.3 25.3 .. 3.9 3.8 3.9 Cambodia 2004 77.9 79.3 76.5 1.4 0.5 1.5 1.2 4.2 0.8 9.8 6.3 13.2 Cameroon 2001 48.7 51.0 46.7 12.6 5.8 14.7 10.6 29.5 4.5 17.0 8.4 24.8 Cape Verde 2000 50.8 58.7 43.1 26.1 8.7 24.5 28.3 30.5 21.1 15.1 8.2 21.8 Chile 2003 36.1 41.7 30.3 21.2 8.3 17.8 26.1 22.3 14.0 14.4 7.8 21.1 Colombia 2000 57.1 66.1 48.6 30.2 13.6 23.1 39.3 36.0 20.1 14.2 5.6 22.4 Costa Rica 2001 51.4 64.8 37.0 13.4 4.0 11.9 16.4 14.0 12.7 16.5 5.7 28.1 Côte d'Ivoire 2002 48.4 53.0 44.3 5.0 3.9 6.0 3.8 12.9 1.3 28.4 16.5 39.0 Croatia 2004 34.7 37.6 31.5 36.0 13.5 32.2 41.1 36.7 35.5 7.1 7.8 6.2 Dominican Republic 2004 51.3 67.5 35.0 20.6 8.8 16.6 28.2 22.0 17.6 48.7 32.5 65.0 Ecuador 2004 50.8 62.9 38.3 12.2 4.7 10.1 15.6 17.5 3.8 16.6 6.7 26.9 Egypt, Arab Rep. 1998 42.3 44.8 39.5 23.7 5.2 22.1 25.7 32.5 19.6 18.7 12.2 25.8 El Salvador 2002 45.7 60.8 31.4 11.5 4.9 13.1 8.4 11.7 11.1 22.5 7.8 36.6 Estonia 2000 38.8 44.3 32.7 21.2 11.5 24.2 16.9 19.1 25.6 9.1 4.2 14.6 Ethiopia 2000 62.0 75.0 50.1 3.9 1.8 2.8 5.3 27.9 0.6 .. .. .. Fiji 1996 35.7 49.7 21.2 16.2 4.3 12.9 24.2 18.1 14.9 30.9 17.6 44.7 Gambia, The 1998 36.4 32.4 40.2 9.2 3.4 12.9 6.4 26.6 2.9 34.6 30.7 38.4 Ghana 1998 49.1 47.5 50.8 15.7 3.4 16.1 15.2 26.0 11.5 22.6 20.4 24.9 Guinea 1994 71.1 67.9 74.2 2.1 2.2 2.5 1.6 7.4 0.2 10.5 5.1 15.8 Haiti 2001 39.0 42.9 35.2 54.6 22.5 48.8 61.7 70.6 39.5 .. .. .. Honduras 2003 52.2 73.5 31.5 7.9 4.7 6.1 12.1 12.0 4.2 31.6 8.5 56.6 Hungary 2002 33.6 34.6 32.5 11.5 7.6 12.6 10.2 9.7 14.6 4.4 3.9 5.1 India 2000 44.3 63.0 24.1 8.1 1.6 8.4 7.0 15.6 5.9 29.2 5.1 55.3 Indonesia 2002 49.7 60.1 39.4 22.0 3.3 20.6 24.1 28.1 16.8 21.0 9.2 32.6 Jamaica 2002 30.0 37.2 22.6 18.9 6.6 14.5 26.3 17.8 18.7 40.1 35.7 44.6 Jordan 2002 33.6 51.2 15.1 42.2 17.0 37.6 58.8 40.4 48.7 18.9 3.0 35.6 Kazakhstan 2002 43.3 48.3 37.9 20.8 7.3 19.0 23.4 17.6 24.3 .. .. .. Kenya 1997 39.7 43.1 36.5 20.7 6.3 22.1 19.1 31.5 17.4 24.7 15.9 33.0 Kiribati 2000 76.1 77.7 74.4 2.0 1.3 2.0 2.0 .. .. 4.2 4.3 4.2 Kyrgyz Republic 2002 35.1 39.3 30.9 16.7 3.9 18.4 14.5 37.8 10.5 14.5 13.6 15.4 Latvia 2004 30.6 36.0 24.9 21.7 11.4 22.0 21.2 .. .. 7.0 6.7 7.3 Lesotho 2002 47.7 48.7 46.8 52.4 32.4 46.7 58.0 42.2 56.3 21.2 18.5 23.8 Lithuania 2000 .. .. .. 9.2 10.5 11.1 7.2 9.6 8.2 .. .. .. 274 Table A3. Going to work (continued) Labor force Not in the labor force participation rate, ages 15­24 Unemployment rate and not in school, ages 15­24 by urban/rural, by age group by gender, ages 15­24 ages 15­24 Young, Adult, ages ages 15­24 25­49 Total Male Female total total Male Female Urban Rural Total Male Female Madagascar 2001 59.9 64.6 55.3 4.1 3.3 3.9 4.2 9.3 2.7 .. .. .. Malawi 1997 .. .. .. 7.9 2.1 10.6 4.9 21.0 5.5 31.3 17.4 44.2 Marshall Islands 1999 .. .. .. 21.2 5.7 16.4 28.9 24.2 16.9 44.0 39.4 48.6 Mauritania 2000 26.1 33.6 19.2 10.8 7.9 11.2 10.1 26.9 3.2 46.5 36.1 56.0 Mexico 2002 49.0 64.8 33.6 6.7 1.9 7.6 4.9 7.2 5.3 20.3 5.3 35.2 Micronesia, Fed. Sts. 2000 38.9 43.5 34.0 30.7 14.3 27.4 35.3 .. .. 28.3 25.2 31.6 Moldova 2002 66.8 66.1 67.6 5.7 4.4 5.8 5.6 12.5 0.7 11.9 13.0 10.9 Mozambique 1996 64.2 59.2 68.8 1.1 0.6 2.4 0.2 7.2 0.3 16.9 13.8 19.6 Namibia 1993 35.3 37.8 33.1 35.1 18.2 32.6 37.7 49.0 22.8 14.7 9.9 19.1 Nepal 1995 62.3 59.6 65.0 .. .. .. .. .. .. 23.0 22.4 23.7 Netherlands 1999 53.6 49.7 57.8 3.8 2.0 3.7 3.9 .. .. 1.9 1.5 2.3 Nicaragua 2001 50.2 70.2 30.1 5.2 3.3 4.4 7.0 7.3 2.6 24.0 9.6 38.5 Niger 2002 36.6 48.6 26.8 20.8 8.7 21.0 20.4 .. .. 35.3 22.6 45.7 Nigeria 2003 .. .. .. 5.6 1.1 5.9 5.3 10.2 3.5 35.4 30.9 40.1 Pakistan 2001 44.7 65.1 24.9 10.4 4.4 8.0 16.5 14.9 8.6 37.9 13.1 62.1 Palau 2000 27.5 29.0 25.8 9.9 3.4 9.5 10.4 .. .. 25.2 24.6 25.9 Panama 2003 46.1 60.2 31.5 21.4 7.8 18.4 27.4 27.3 11.3 15.9 5.4 26.8 Paraguay 2001 61.1 75.7 45.8 13.8 5.6 11.7 17.3 18.4 7.7 16.3 6.2 26.9 Peru 2002 57.4 64.3 50.3 13.5 6.9 13.0 14.2 .. .. 13.3 7.4 19.4 Poland 2002 55.6 56.4 54.8 19.4 15.3 19.9 18.8 22.1 16.2 10.2 10.2 10.2 Romania 2002 63.3 63.6 63.1 24.2 17.9 24.0 24.4 26.3 22.4 .. .. .. Russian Federation 2002 37.8 38.5 37.0 25.7 11.5 24.2 27.4 21.7 35.4 .. .. .. Rwanda 1997 72.9 71.5 74.1 9.6 9.1 10.1 9.2 27.6 8.0 5.9 5.6 6.1 São Tomé and Principe 2000 35.2 50.3 20.0 8.5 1.4 7.1 11.6 7.2 10.1 40.5 24.2 57.0 Senegal 1995 37.5 53.1 24.2 10.1 9.0 12.3 6.0 .. .. 33.6 12.6 51.0 Serbia and Montenegro 2005 36.6 41.1 31.9 61.0 26.0 64.3 56.5 .. .. .. .. .. Sierra Leone 2003 40.3 34.2 46.0 1.8 3.0 2.2 1.5 3.6 1.0 19.2 14.7 23.5 Slovak Republic 1992 41.7 47.2 36.2 19.6 7.3 20.5 18.4 15.3 19.8 15.5 7.9 22.9 Slovenia 1999 38.4 43.4 32.8 30.3 10.9 27.4 34.6 .. .. 0.6 0.7 0.5 Solomon Islands 1999 54.8 57.3 52.2 11.5 4.8 12.7 10.1 .. .. 20.7 14.5 27.0 South Africa 2000 27.8 29.5 26.1 54.1 25.4 49.6 59.1 53.8 54.9 16.2 13.4 18.9 Sri Lanka 2002 50.0 59.5 40.6 45.2 10.3 39.9 52.9 47.3 44.9 .. .. .. Swaziland 2000 34.6 37.2 32.1 13.8 5.8 16.6 10.6 10.2 15.4 22.8 14.3 30.9 Sweden 2000 55.8 57.9 53.6 32.0 6.1 33.3 30.4 29.6 34.4 1.2 1.1 1.2 Tajikistan 1999 48.9 54.5 43.6 25.8 12.9 27.9 23.4 23.8 26.3 28.1 18.5 37.2 Tanzania 2000 66.9 69.1 65.2 2.8 0.8 4.6 1.4 10.9 1.1 12.7 6.8 17.5 Thailand 2002 51.8 57.8 46.0 9.2 1.1 10.4 7.8 12.8 7.4 5.9 2.6 9.0 Tonga 1996 42.1 53.9 29.7 30.2 8.7 31.9 27.0 .. .. 16.4 6.2 27.1 Trinidad and Tobago 1992 38.9 50.9 26.8 19.6 12.9 18.2 22.4 21.2 18.5 29.8 21.4 38.2 Turkey 2002 45.3 58.8 34.1 18.2 5.5 21.5 13.6 12.5 23.7 31.9 11.8 48.7 Uganda 2002 55.0 49.1 60.1 6.0 1.8 4.2 7.2 23.2 2.5 8.6 5.5 11.3 Ukraine 2003 32.4 36.8 27.6 39.1 18.2 37.8 40.9 33.3 51.4 .. .. .. Uruguay 2003 52.6 59.8 45.3 38.0 13.4 33.5 44.0 .. .. 10.0 5.7 14.4 Venezuela, RB 2004 46.5 59.3 33.2 19.3 11.1 17.9 21.9 .. .. 20.2 10.7 30.1 Vietnam 2001 63.9 63.7 64.0 5.4 1.1 5.8 5.0 14.6 3.6 5.0 3.7 6.3 Yemen, Rep. 1998 26.3 37.4 14.9 14.5 5.3 17.9 5.7 34.7 10.9 .. .. .. Zambia 2002 34.8 33.8 35.8 25.0 6.1 28.4 21.8 46.2 4.2 26.2 21.0 31.2 275 Table A4. Child labor Economically active children % of children ages 7­14 % of children ages 7­14 Survey year Total Male Female Work only Work and study Albania 2000 36.6 41.1 31.8 43.1 56.9 Angola 1995 5.2 4.9 5.6 77.6 22.4 Argentina 1997 20.7 25.4 16.0 8.6 91.4 Azerbaijan 2000 9.7 12.0 7.3 4.2 95.8 Bangladesh 2003 17.5 20.9 13.9 63.3 36.7 Bolivia 2000 19.2 20.4 18.0 19.7 80.3 Bosnia and Herzegovina 2000 20.2 22.8 17.6 4.0 96.0 Brazil 2003 7.1 9.5 4.6 5.8 94.2 Burkina Fasoa 1998 66.5 65.4 67.7 95.9 4.1 Burundi 2000 37.0 38.4 35.7 48.3 51.7 Cambodia 2001 52.3 52.4 52.1 16.5 83.5 Cameroona 2001 15.9 14.5 17.4 52.5 47.5 Central African Republic 2000 67.0 66.5 67.6 54.9 45.1 Chad 2000 69.9 73.5 66.5 44.6 55.4 Chile 2003 8.8 10.5 6.9 4.0 96.0 Colombia 2001 12.2 16.6 7.7 23.0 77.0 Congo, Dem. Rep. 2000 39.8 39.9 39.8 35.7 64.3 Costa Rica 2002 6.7 9.7 3.5 20.8 79.2 Côte d'Ivoire 2000 40.7 40.9 40.5 46.4 53.6 Dominican Republic 2000 12.5 16.7 8.1 7.2 92.8 Ecuador 2001 17.9 22.1 13.6 25.1 75.0 Egypt, Arab Rep. 1998 6.4 4.0 8.9 60.9 39.1 El Salvador 2003 12.7 17.1 8.1 19.5 80.5 Ethiopia 2001 57.1 67.9 45.9 63.5 36.5 Gambia, The 2000 25.3 25.4 25.3 41.6 58.4 Ghana 2000 28.5 28.5 28.4 36.4 63.6 Guatemala 2000 20.1 25.9 13.9 38.5 61.5 Guinea 1994 48.3 47.2 49.5 98.6 1.4 Guinea-Bissau 2000 67.5 67.4 67.5 63.7 36.3 Honduras 2002 11.4 16.5 6.1 41.9 58.1 India 2000 5.2 5.3 5.1 89.8 10.2 Iraq 2000 13.7 17.4 9.7 51.7 48.3 Kazakhstan 1996 29.7 30.3 29.1 4.4 95.6 Kenya 1999 6.7 6.9 6.4 44.8 55.2 Kyrgyz Republic 1998 8.6 9.7 7.6 7.0 93.0 Lesotho 2000 30.8 34.2 27.5 17.6 82.4 Madagascar 2001 25.6 26.1 25.1 85.1 14.9 Malawi 2000 10.6 9.4 11.6 17.1 82.9 Mali 2001 25.3 32.3 18.6 68.7 31.3 Mexicob 1996 14.7 20.0 9.5 45.6 54.4 Moldova 2000 33.5 34.1 32.8 3.8 96.2 Mongolia 2000 22.0 23.5 20.6 28.2 71.8 Morocco 1998/99 13.2 13.5 12.8 93.2 6.8 Namibia 1999 15.4 16.2 14.7 9.5 90.5 Nepal 1999 47.2 42.2 52.4 35.6 64.4 Nicaragua 2001 12.1 17.5 6.5 33.3 66.7 Panama 2000 4.0 6.4 1.4 37.5 62.5 Paraguay 1999 8.1 11.7 4.4 24.2 75.7 Peru 1994 17.7 20.4 15.2 7.3 92.7 Philippines 2001 13.3 16.3 10.0 14.8 85.2 Portugal 2001 3.6 4.6 2.6 3.6 96.4 Rwanda 2000 33.1 36.1 30.3 27.5 72.5 Senegal 2000 35.4 43.2 27.7 56.2 43.8 Sierra Leone 2000 74.0 24.7 72.7 53.8 46.2 South Africa 1999 27.7 29.0 26.4 5.1 94.9 Sudan 2000 19.1 21.5 16.8 55.9 44.1 Swaziland 2000 11.2 11.4 10.9 14.0 86.0 Tanzania 2001 40.4 41.5 39.2 40.0 60.0 Togo 2000 72.5 73.4 71.6 28.4 71.6 Trinidad and Tobago 2000 3.9 5.2 2.8 12.8 87.2 Turkey 1999 4.5 5.2 3.8 66.8 33.2 Uganda 2002/03 13.1 15.0 11.3 18.3 81.7 Uzbekistan 2000 18.1 22.0 14.0 4.1 95.9 Venezuela, RB 2003 9.1 11.4 6.6 17.6 82.4 Yemen, Rep. 1999 13.1 12.4 14.0 64.3 35.7 Zambia 1999 14.4 15.0 13.9 72.8 27.2 a. Data are for children ages 10­14. b. Data are for children ages 12­14. 276 Table A5. Staying healthy Probability that a 15-year-old Tobacco use in adolescents will die before the age of 60, 2003 Female Male per 1,000 % ages % ages Survey year 13­15 13­15 Female Male Afghanistan 448 510 Albania 2006 9 16 92 167 Algeria .. .. 125 155 Andorra .. .. 41 107 Angola .. .. 488 584 Antigua and Barbuda 2000 11 16 122 193 Argentina 2000 34 31 90 176 Armenia .. .. 108 240 Australia .. .. 51 89 Austria .. .. 59 115 Azerbaijan .. .. 120 220 Bahamas, The 2000 14 23 146 257 Bahrain 2001 12 34 81 117 Barbados 2002 13 16 106 189 Belize 2003 14 24 153 257 Bangladesh .. .. 258 251 Belarus .. .. 130 370 Belgium .. .. 66 125 Benin 2003 10 24 332 393 Bhutan .. .. 202 261 Bolivia 2000 24 35 180 247 Bosnia and Herzegovina 2003 12 19 89 190 Botswana 2001 12 17 839 850 Brazil 2002 18 21 129 240 Brunei .. .. 86 114 Bulgaria 2002 42 33 91 216 Burkina Faso 2001 7 18 462 533 Burundi .. .. 525 654 Cambodia 2003 3 11 285 441 Cameroon .. .. 461 503 Canada .. .. 57 93 Cape Verde .. .. 129 213 Central African Republic .. .. 590 641 Chad .. .. 444 513 Chile 2000 42 33 66 133 China 2003 6 14 103 164 Colombia 2001 29 30 97 231 Comoros .. .. 182 254 Congo, Dem. Rep. .. .. 452 578 Congo, Rep. .. .. 381 434 Costa Rica 2002 19 20 76 129 Côte d'Ivoire .. .. 450 558 Croatia 2002 15 19 70 173 Cuba 2001 18 17 87 137 Cyprus .. .. 47 99 Czech Republic 2002 33 36 74 166 Denmark .. .. 73 121 Djibouti .. .. 311 376 Dominica 2000 16 24 118 210 Dominican Republic .. .. 147 250 Ecuador 2001 17 21 127 212 Egypt, Arab Rep. 2001 16 23 157 242 El Salvador 2003 15 25 138 248 Eritrea .. .. 301 359 Estonia 2002 30 35 114 319 Ethiopia 2003 6 12 386 450 Fiji 1999 13 24 173 275 Finland .. .. 57 134 France .. .. 59 132 Gambia, The .. .. 262 332 Gabon .. .. 323 397 Georgia 2002 13 34 76 195 Germany .. .. 59 115 Ghana 2000 19 20 295 352 Grenada 2000 14 18 220 258 Greece .. .. 48 118 Guatemala 2002 12 18 165 289 Guinea .. .. 342 403 Guyana 2004 5 15 255 290 Guinea-Bissau .. .. 405 479 Haiti 2001 18 18 385 450 Honduras 2003 19 27 181 248 Hungary 2002 33 34 111 257 Iceland .. .. 53 81 India 2001 20 29 213 283 Indonesia 2000 5 37 204 241 Iran, Islamic Rep. 2003 5 14 125 201 Iraq .. .. 205 466 Ireland .. .. 60 100 Israel .. .. 51 92 277 Table A5. Staying healthy (continued) Probability that a 15-year-old Tobacco use in adolescents will die before the age of 60, 2003 Female Male per 1,000 % ages % ages Survey year 13­15 13­15 Female Male Italy .. .. 47 93 Jamaica 2001 15 24 123 165 Japan .. .. 45 96 Jordan 2004 12 28 120 189 Kazakhstan .. .. 187 419 Kenya 2003 14 21 521 495 Kiribati .. .. 191 304 Korea, Dem. Rep. .. .. 168 231 Korea, Rep. .. .. 61 155 Kuwait 2001 18 33 53 73 Kyrgyz Republic .. .. 160 339 Lao PDR 2003 4 18 303 335 Latvia 2002 33 41 120 306 Lebanon 2001 40 46 138 199 Lesotho 2002 20 32 781 912 Liberia .. .. 484 590 Libya 2003 9 19 101 172 Lithuania 2001 32 40 106 302 Luxembourg .. .. 63 115 Macedonia, FYR 2002 8 12 86 202 Madagascar .. .. 260 337 Malawi 2001 15 20 615 652 Malaysia .. .. 108 195 Mali 2001 13 45 427 486 Malta .. .. 49 84 Marshall Islands .. .. 280 333 Mauritania 2001 23 34 312 408 Mauritius .. .. 115 218 Mexico 2002 20 24 95 166 Micronesia, Fed. Sts. .. .. 172 206 Monaco .. .. 47 110 Moldova .. .. 152 303 Mongolia .. .. 179 310 Morocco 2001 9 17 103 159 Mozambique 2002 10 11 543 621 Myanmar 2001 5 37 222 337 Namibia .. .. 529 619 Nauru .. .. 303 448 Nepal 2001 6 15 284 290 Netherlands .. .. 66 93 New Zealand .. .. 65 98 Nicaragua .. .. 138 209 Niger 2001 14 27 477 508 Nigeria 2001 17 24 470 511 Norway .. .. 58 96 Oman 2003 9 27 91 163 Pakistan .. .. 199 225 Palau 2001 62 55 205 226 Panama 2002 16 19 84 146 Papua New Guinea .. .. 246 309 Paraguay 2003 23 24 119 171 Peru 2001 16 24 133 193 Philippines 2003 8 21 149 271 Poland 1999 24 33 81 202 Portugal .. .. 63 150 Qatar .. .. 76 93 Romania 2006 20 27 107 239 Russian Federation 2001 29 41 182 480 Rwanda .. .. 455 541 St. Kitts and Nevis 2002 16 20 145 200 St. Lucia 2001 10 19 131 224 St. Vincent and the Grenadines 2001 20 27 192 233 Samoa .. .. 203 235 San Marino .. .. 32 73 São Tomé and Principe .. .. 244 295 Saudi Arabia 2006 .. 5a 119 196 Senegal 2002 6 25 280 350 Serbia and Montenegro 2003 17 16 99 186 Seychelles 2002 25 36 92 235 Sierra Leone .. .. 517 597 Singapore 2006 8 11 51 87 Slovak Republic 2002 23 27 77 204 Slovenia 2003 29 27 69 165 South Africa 2003 27 38 579 642 Spain .. .. 46 116 Sri Lanka 2006 1 3 120 235 Sudan 2001 13 20 248 348 Swaziland 2001 10 21 790 894 Sweden .. .. 50 79 Switzerland .. .. 50 90 278 Table A5. Staying healthy (continued) Probability that a 15-year-old Tobacco use in adolescents will die before the age of 60, 2003 Female Male per 1,000 % ages % ages Survey year 13­15 13­15 Female Male Syrian Arab Republic 2002 15 24 126 188 Tajikistan .. .. 169 225 Tanzania .. .. 550 587 Thailand .. .. 153 267 Togo 2002 10 20 377 448 Trinidad and Tobago 2000 12 20 155 249 Tunisia 2001 7 29 113 167 Turkey .. .. 111 176 Turkmenistan .. .. 171 352 Uganda 2002 16 22 459 533 Ukraine 1999 35 46 142 384 United Arab Emirates 2002 13 30 121 168 United Kingdom .. .. 64 103 United States 2000 20 26 82 139 Uruguay 2001 26 22 87 180 Uzbekistan .. .. 142 226 Venezuela, RB 2003 12 15 97 181 Vietnam 2003 2 10 129 205 Yemen, Rep. 2002 14 21 227 298 Zambia 2002 24 25 685 719 Zimbabwe 2001 14 19 819 830 a. Data refer to capital cities only. 279 Table A6. Staying healthy Knowledge of HIV Condom use among sexually active youth prevention methods Female Male Female Male Survey % ages % ages % ages % ages Survey % ages % ages year 15­19 20­24 15­19 20­24 year 15­24 15­24 Albania .. .. .. .. 2000 39 .. Armenia 2000 0.2 4.0 .. .. 2000 37 52 Azerbaijan .. .. .. .. 2000 9 .. Benin 2001 3.6 4.5 16.8 27.0 2001 40 45 Bolivia 2003 1.2 4.3 .. .. 2000 52 .. Botswana .. .. .. .. 2000 71 .. Burkina Faso 2003 6.3 8.9 11.2 36.7 2003 45 54 Cambodia 2000 0.1 0.6 .. .. 2000 59 .. Cameroon 2004 11.3 16.0 .. .. 2004 64 73 Chad 2004 0.4 0.9 4.6 11.0 .. .. Colombia 2005 6.2 9.4 .. .. .. .. Congo, Dem. Rep. .. .. .. .. 2000 42 .. Côte d'Ivoire 1998/99 6.0 7.7 21.6 30.8 2000 46 .. Dominican Republic 2002 1.3 2.9 .. .. 2002 80 78 Eritrea 2002 0.4 0.4 .. .. 2002 58 .. Ethiopia 2000 0.3 0.7 1.6 3.4 .. .. Gambia, The .. .. .. .. 2000 45 .. Gabon 2000 10.9 13.0 40.8 48.4 .. .. Ghana 2003 5.2 7.7 .. .. 2003 72 75 Guatemala 1998/99 0.8 1.1 .. .. .. .. Guinea 1999 2.3 2.7 14.4 27.8 .. .. Guyana .. .. .. .. 2000 60 .. Guinea-Bissau .. .. .. .. 2000 24 .. Haiti 2000 2.0 4.9 .. .. 2000 44 69 Indonesia .. .. .. .. 2002/03 21a 32a Kazakhstan 1999 2.2 5.1 20.7 35.1 .. .. Kenya 2003 1.8 1.6 .. .. 2003 55 65 Lesotho .. .. .. .. 2000 47 .. Madagascar 2003/04 1.1 1.7 5.4 6.4 2000 32 .. Malawi 2000 2.8 2.9 .. .. 2000 57 65 Mali 2001 1.0 1.1 5.0 17.6 2001 33 43 Mauritania 2000/01 0.0 0.4 1.2 3.4 .. .. Moldova .. .. .. .. 2000 47 .. Mongolia .. .. .. .. 2000 71b .. Morocco 2003/04 0.0 0.4 .. .. 2003/04 36 .. Mozambique 2003 9.2 5.8 .. .. 2003 47 63 Namibia 2000 10.8 11.0 .. .. 2000 65 81 Nepal .. .. .. .. 2001 38c 75c Nicaragua 2001 1.0 2.6 .. .. .. .. Niger 1998 0.0 0.2 2.6 8.9 2000 28 .. Nigeria 2003 2.6 6.0 .. .. 2003 40 58 Peru 2000 0.9 3.3 .. .. .. .. Philippines 2003 0.1 1.0 .. .. 2000 46 .. Rwanda 2000 0.3 0.8 .. .. 2000 52 63 São Tomé and Principe .. .. .. .. 2000 20 .. Senegal .. .. .. .. 2000 46 .. Sierra Leone .. .. .. .. 2000 27 .. South Africa 1998 2.0 3.5 .. .. 1998 74 .. Tajikistan .. .. .. .. 2000 4d .. Tanzania 1999 3.7 6.0 10.6 20.9 2003/04 61 65 Togo 1998 6.9 6.7 12.8 30.4 2000 58 .. Trinidad and Tobago .. .. .. .. 2000 48 .. Turkey 1998 0.9 5.2 0.0 3.0 .. .. Turkmenistan 2000 0.1 0.4 .. .. 2000 15 .. Uganda 2000/01 5.7 4.9 .. .. 2000/01 66 77 Uzbekistan .. .. .. .. 2002 25 40 Vietnam .. .. .. .. 1997 52 .. Zambia 2001/02 4.0 5.2 .. .. .. .. Zimbabwe 1999 1.8 2.8 9.4 27.7 1999 64 69 a. Sample included ever married women and currently married men. b. Self weighting sample. c. Sample included ever married women and men. d. Sample was not random. 280 Table A7. Forming families Fertility rate per 1,000 women Women who gave birth before age 18 women ages 15­24 Survey year ages 15­24 Survey year % Armenia 2000 199 2000 8.0 Azerbaijan .. .. 2006 7.0 Bangladesh 2004 328 2004 45.5 Benin 2001 370 2001 23.8 Bolivia 2003 267 2003 19.0 Burkina Faso 2003 384 2003 27.1 Cambodia 2000 222 2000 12.1 Cameroon 2004 374 2004 33.0 Cape Verde .. .. 2006 24.0 Chad .. .. 2004 48.0 Colombia 2000 227 2005 19.5 Côte d'Ivoire .. .. 2006 .. Djibouti .. .. 2006 4.0 Dominican Republic 2002 306 2002 25.0 Ecuador .. .. 2006 18.0 Egypt, Arab Rep. 2000 247 2000 9.5 El Salvador .. .. 2006 24.0 Eritrea 2002 262 2002 25.4 Ethiopia 2000 335 2000 24.2 Gabon 2000 338 2000 35.4 Georgia .. .. 2006 11.0 Ghana 2003 250 2003 14.9 Guatemala 1998/99 393 2006 24.0 Guinea 1999 406 2006 47.0 Haiti 2000 269 2000 15.2 Honduras .. .. 2006 28.0 India 1998/99 317 2006 28.0 Indonesia 2002/03 182 2002/03 11.9 Jordan 2002 178 2002 5.0 Kazakhstan 1999 207 2006 6.0 Kenya 2003 357 2003 22.7 Madagascar 2003/04 395 2003/04 31.3 Malawi 2000 477 2000 30.3 Mali 2001 475 2001 45.0 Mauritania 2000/01 246 2006 25.0 Morocco 2003/04 136 2003/04 7.5 Mozambique 2003 424.2 2006 42.0 Namibia 2000 254 2000 20.5 Nepal 2001 358 2001 26.0 Nicaragua 2001 297 2001 28.1 Niger .. .. 2006 47.0 Nigeria 2003 355 2003 28.0 Paraguay .. .. 2006 13.0 Peru 2000 206 2000 14.2 Philippines 2003 231 2003 6.9 Romania .. .. 2006 5.0 Rwanda 2000 292 2000 9.2 Senegal .. .. 2006 27.0 South Africa .. .. 2006 20.0 Syrian Arab Republic .. .. .. .. Tanzania 1999 406 2006 26.0 Togo .. .. 2006 19.0 Turkey .. .. 2006 8.0 Turkmenistan 2000 214 2000 1.8 Uganda 2000/01 509 2000/01 42.0 Uzbekistan .. .. 2006 4.0 Vietnam 2002 163 2002 3.9 Yemen, Rep. .. .. .. .. Zambia 2001/02 426 2001/02 34.6 Zimbabwe 1999 311 2006 20.0 281 Table A8. Forming families Pregnant women not receiving prenatal carea not informed of the complicationsb ages 15­24 ages 15­24 Survey year % Survey year % Armenia 2000 7.0 2000 50.4 Bangladesh 2004 38.1 2004 50.1 Benin 2001 9.3 2001 17.3 Bolivia 2003 16.5 2003 62.9 Burkina Faso 2003 23.7 2003 15.1 Cameroon 2004 16.8 2004 36.5 Chad 2004 52.0 2004 14.1 Colombia 2000 10.1 2000 80.8 Côte d'Ivoire 1998 10.4 .. Dominican Republic 2002 1.0 2002 60.2 Egypt, Arab Rep. .. 2000 41.1 Ethiopia 2000 74.7 2000 20.4 Gabon 2000 3.1 .. Ghana 2003 5.1 2003 57.3 Guatemala 1998/99 11.7 .. Guinea 1999 18.8 .. Haiti 2000 18.6 2000 29.0 India 1998/99 30.3 .. Indonesia 2002/03 7.5 2002/03 28.6 Kazakhstan 1999 4.5 .. Kenya 2003 8.9 2003 34.2 Malawi 2002 4.1 2002 69.5 Mali 2001 40.5 2001 23.6 Morocco 2003/04 35.2 2003/04 37.3 Mozambique 2003 12.0 2003 51.6 Namibia 2000 5.0 2000 46.5 Nepal 2001 42.1 2001 52.0 Nicaragua 2001 11.6 2001 76.1 Peru 2004 5.0 2004 81.2 Philippines 2003 5.3 2003 46.9 Rwanda 2000 7.1 2000 5.7 Turkey 1998 28.0 .. Uganda 2001 4.8 2001 15.8 Vietnam 2002 20.2 .. Zambia 1001 4.2 .. Zimbabwe 1999 7.2 1999 43.8 a. Restricted to most recent births in the three years previous to the survey date. b. Restricted to those who used antenatal care for the latest infants during three years previous to the survey date. 282 Selected indicators 283 Technical notes and national statistical offices. These surveys yield a variety of data in education, employment, health, expenditure, and consump- Table A1. Learning tion that relate to child work; they do not provide information on unconditional forms of children's work. Enrollment rates by age group are based on nationally-representa- Household survey data generally include information on work tive household surveys. They measure the proportion of people in type--for example, whether a child is working for pay in cash or in a given age range reported to be attending school at the time of the kind or is involved in unpaid work, whether a child is working for household survey. someone who is not a member of the household, whether a child Grade 9 completion rates are computed using the informa- is involved in any type of family work (on a farm or in a business), tion from nationally representative household surveys on last grade and the like. The ages used in country surveys to define child labor completed and current school participation of individuals ages 10- range from 5 to 14 years old. The data in the table have been recal- 19 at the time of the survey. Estimates are based on Kaplan-Meier culated to present statistics for children ages 7­14. method which takes into account whether the individual is still in Although efforts are made to harmonize the definition of school. employment and the questions on employment used in survey questionnaires, some differences remain among the survey instru- Table A2. Program for International ments used to collect the information on working children. Dif- Student Assessment ferences exist not only among different household surveys in the The Program for International Student Assessment (PISA) same country, but also within the same type of survey carried out assesses the performance of 15-year-olds in reading, mathematical, in different countries. and scientific literacy as well as problem solving in terms of mastery Because of the differences in the underlying survey instruments of the school curriculum and the use of knowledge for everyday and in survey dates, estimates of the economically active child popu- tasks and challenges. lation are not fully comparable across countries. Caution should be exercised in drawing conclusions concerning relative levels of child Table A3. Going to work economic activity across countries or regions based on the pub- Labor force participation rate is the share of the youth population, lished estimates. either employed or unemployed, that is, economically active. Economically active children refer to children involved in eco- Unemployment rate is the share of the labor force that is unem- nomic activity for at least one hour in the reference week of the ployed. To be considered unemployed, an individual must be not survey. Work only refers to children involved in economic activ- employed but actively seeking work. ity and not attending school. Work and study refers to children Not in the labor force and not in school is the youth popula- attending school in combination with economic activity. tion that is neither in the labor force nor in school. Table A5. Staying healthy (1) Table A4. Child labor Tobacco use in adolescents is the percent of youth ages 13­15 who The data in the table refer to children's economic activity, a broader currently use tobacco. Data is based on World Health Organiza- concept than child labor. According to a gradually emerging con- tion (WHO) and Centers for Disease Control and Prevention's sensus, child labor is a subset of children's economic activity or Global Youth Tobacco Survey and augmented by Population Refer- children's work that is injurious and therefore targeted for elimina- ence Bureau's World's Youth 2006 Data Sheet. tion. Probability that a 15-year old will die before the age of 60 is In line with the international definition of employment, a child the probability of dying per 1,000 population ages 15­60. Based who spends at least one hour on economic activity during the refer- on figures computed by the WHO (The World Health Report 2005), ence week is classified as economically active. Economic activity is which may differ from official statistics that may use other rigorous as defined by the 1993 United Nations System of National Accounts methods. (revision 3) and corresponds to the international definition of employment adopted by the Thirteenth International Conference Table A6. Staying healthy (2) of Labor Statisticians in 1982. Economic activity covers all market Condom use among sexually active youth is the percent of cur- production and certain types of nonmarket production, including rently married or sexually active unmarried men and women ages production of goods for own use. It excludes household chores per- 15­19 and 20­24 using condoms. Data based on ORC Macro, 2006, formed by children in their own household. Some forms of eco- MEASURE DHS STAT compiler. nomic activity are not captured by household surveys and so are Knowledge of HIV prevention methods is the percent of not reflected in the estimates. These include unconditional forms of respondents who, in response to a prompted question, say that peo- child labor, such as child commercial sexual exploitation and child ple can protect themselves from contracting HIV by using condoms slavery, which require different data collection methodologies. or having sex only with one faithful, uninfected partner. Numera- The data used to develop the indicators are from household sur- tor is the number of respondents who, in response to a prompted veys conducted by the International Labour Organization (ILO), question, correctly identify using condoms or having sex only with the United Nations Children's Fund (UNICEF), the World Bank, one faithful, uninfected partner as means of protection against HIV 284 WORLD DEVELOPMENT REPORT 2007 infection. Denominator is the total number of respondents. Data Table A8. Forming families (2) based on ORC Macro, 2006, MEASURE DHS STAT compiler. Pregnant women not receiving antenatal care ages 15­24 is the per- centage of mothers aged 15­24 who used antenatal care for the latest Table A7. Forming families (1) infants born within the three years previous to the survey date. Based Fertility rate per 1,000 women ages 15­24 are age-specific fertility on Demographic and Health Surveys. rates (women ages 15­24) for the three years preceding the survey. Pregnant women not informed of the complications ages Based on ORC Macro, 2006, MEASURE DHS STAT compiler. 15­24 is the percentage of mothers aged 15­24 who used antenatal Women who gave birth before age 18 is the percentage of all care for the latest infants born within the three years previous to women ages 15­24 who gave birth by age 18. Based on ORC Macro, the survey date but were not informed of pregnancy complications. 2006, MEASURE DHS STAT compiler and Population Reference Based on Demographic and Health Surveys. Bureau World's Youth Survey 2006 Data Sheet. Symbols .. means that data are not available. 0 or 0.0 means zero or less than half the unit shown. / in dates, as in 1990/91, means that the period of time, usually 12 months, straddles two calendar years and refers to a survey year, an academic year, or a fiscal year. A blank means not applicable. Selected world development indicators In this year's edition of the Selected World Development Indica- in timeliness and reliability arising from differences in the capabili- tors, development data are presented in six tables presenting com- ties and resources devoted to basic data collection and compilation. parative socioeconomic data for more than 130 economies for the For some topics, competing sources of data require review by World most recent year for which data are available and, for some indica- Bank staff to ensure that the most reliable data available are pre- tors, for an earlier year. An additional table presents basic indica- sented. In some instances, where available data are deemed too weak tors for 75 economies with sparse data or with populations of less to provide reliable measures of levels and trends or do not adequately than 2 million. adhere to international standards, the data are not shown. The indicators presented here are a selection from more than The data presented are generally consistent with those in World 800 included in World Development Indicators 2006. Published Development Indicators 2006. However, data have been revised and annually, World Development Indicators reflects a comprehensive updated wherever new information has become available. Differ- view of the development process. Its opening chapter reports on ences may also reflect revisions to historical series and changes in the Millennium Development Goals, which grew out of agreements methodology. Thus data of different vintages may be published in and resolutions of world conferences in the 1990s, and were for- different editions of World Bank publications. Readers are advised mally recognized by the United Nations General Assembly after not to compile data series from different publications or differ- member states unanimously adopted the Millennium Declara- ent editions of the same publication. Consistent time-series data tion at the Millennium Summit in September 2000. In September are available on World Development Indicators 2006 CD-ROM and 2005, the United Nations World Summit reaffirmed the principles through WDI Online. in the 2000 Millennium Declaration and recognized the need for All dollar figures are in current U.S. dollars unless otherwise ambitious national development strategies backed by increased stated. The various methods used to convert from national currency international support. The other five main sections recognize the figures are described in the Technical notes. contribution of a wide range of factors: human capital develop- Because the World Bank's primary business is providing lend- ment, environmental sustainability, macroeconomic performance, ing and policy advice to its low- and middle-income members, the private sector development and the investment climate, and the issues covered in these tables focus mainly on these economies. global links that influence the external environment for develop- Where available, information on the high-income economies is also ment. World Development Indicators is complemented by a sepa- provided for comparison. Readers may wish to refer to national sta- rately published database that gives access to over 1,000 data tables tistical publications and publications of the OECD and the Euro- and 800 time-series indicators for 222 economies and regions. pean Union for more information on the high-income economies. This database is available through an electronic subscription (WDI Online) or as a CD-ROM. Classification of economies and summary measures Data sources and methodology The summary measures at the bottom of each table include econ- Socioeconomic and environmental data presented here are drawn omies classified by income per capita and by region. GNI per from several sources: primary data collected by the World Bank, capita is used to determine the following income classifications: member country statistical publications, research institutes, and low-income, $875 or less in 2005; middle-income, $876 to $10,725; international organizations such as the United Nations and its and high-income, $10,726 and above. A further division at GNI per specialized agencies, the International Monetary Fund (IMF), and capita $3,465 is made between lower-middle-income and upper- the Organisation for Economic Co-operation and Development middle-income economies. See the table on classification of econo- (OECD) (see the Data sources following the Technical notes for a mies at the end of this volume for a list of economies in each group complete listing). Although international standards of coverage, (including those with populations of less than 2 million). definition, and classification apply to most statistics reported by Summary measures are either totals (indicated by t if the aggre- countries and international agencies, there are inevitably differences gates include estimates for missing data and nonreporting coun- 285 286 WORLD DEVELOPMENT REPORT 2007 tries, or by s for simple sums of the data available), weighted averages / (w), or median values (m) calculated for groups of economies. Data in dates, as in 1990/91, means that the period of time, usually 12 for the countries excluded from the main tables (those presented in months, straddles two calendar years and refers to a survey year, an table 5) have been included in the summary measures, where data academic year, or a fiscal year. are available, or by assuming that they follow the trend of report- $ ing countries. This gives a more consistent aggregated measure means current U.S. dollars unless otherwise noted. by standardizing country coverage for each period shown. Where > missing information accounts for a third or more of the overall esti- means more than. mate, however, the group measure is reported as not available. The Statistical methods section in the Technical notes provides further < information on aggregation methods. Weights used to construct the means less than. aggregates are listed in the technical notes for each table. A blank means not applicable or, for an aggregate, not analytically From time to time an economy's classification is revised because meaningful. of changes in the above cutoff values or in the economy's measured level of GNI per capita. When such changes occur, aggregates based Technical notes on those classifications are recalculated for the past period so that a Because data quality and intercountry comparisons are often prob- consistent time series is maintained. lematic, readers are encouraged to consult the Technical notes, the table on Classification of Economies by Region and Income, and Terminology and country coverage the footnotes to the tables. For more extensive documentation see The term country does not imply political independence but may World Development Indicators 2006. refer to any territory for which authorities report separate social Readers may find more information on the WDI 2006, and or economic statistics. Data are shown for economies as they were orders can be made online, by phone, or fax as follows: constituted in 2004, and historical data are revised to reflect current For more information and to order on line: http://www.world- political arrangements. Throughout the tables, exceptions are noted. bank.org/data/wdi2006/index.htm. Unless otherwise noted, data for China do not include data for Hong To order by phone or fax: phone 1-800-645-7247 or 1-703-661- Kong, China; Macao, China; or Taiwan, China. Data for Indonesia 1580; fax 1-703-661-1501 include Timor-Leste through 1999 unless otherwise noted. To order by mail: The World Bank, P.O. Box 960, Herndon, VA 20172-0960, U.S.A. Symbols .. means that data are not available or that aggregates cannot be cal- culated because of missing data in the years shown. 0 or 0.0 means zero or less than half the unit shown. Classification of economies by region and income, FY2007 East Asia and the Pacific Latin America and the Caribbean South Asia High income OECD American Samoa UMC Argentina UMC Afghanistan LIC Australia Cambodia LIC Barbados UMC Bangladesh LIC Austria China LMC Belize UMC Bhutan LIC Belgium Fiji LMC Bolivia LMC India LIC Canada Indonesia LMC Brazil LMC Maldives LMC Denmark Kiribati LMC Chile UMC Nepal LIC Finland Korea, Dem. Rep. LIC Colombia LMC Pakistan LIC France Lao PDR LIC Costa Rica UMC Sri Lanka LMC Germany Malaysia UMC Cuba LMC Greece Marshall Islands LMC Dominica UMC Sub-Saharan Africa Iceland Micronesia, Fed. Sts. LMC Dominican Republic LMC Angola LMC Ireland Mongolia LIC Ecuador LMC Benin LIC Italy Myanmar LIC El Salvador LMC Botswana UMC Japan Northern Mariana Islands UMC Grenada UMC Burkina Faso LIC Korea, Rep. Palau UMC Guatemala LMC Burundi LIC Luxembourg Papua New Guinea LIC Guyana LMC Cameroon LMC Netherlands Philippines LMC Haiti LIC Cape Verde LMC New Zealand Samoa LMC Honduras LMC Central African Republic LIC Norway Solomon Islands LIC Jamaica LMC Chad LIC Portugal Thailand LMC Mexico UMC Comoros LIC Spain Timor-Leste LIC Nicaragua LMC Congo, Dem. Rep. LIC Sweden Tonga LMC Panama UMC Congo, Rep. LMC Switzerland Vanuatu LMC Paraguay LMC Côte d'Ivoire LIC United Kingdom Vietnam LIC Peru LMC Equatorial Guinea UMC United States St. Kitts and Nevis UMC Eritrea LIC Europe and Central Asia St. Lucia UMC Ethiopia LIC Other high income Albania LMC St. Vincent and the Grenadines UMC Gabon UMC Andorra Armenia LMC Suriname LMC Gambia, The LIC Antigua and Barbuda Azerbaijan LMC Trinidad and Tobago LMC Ghana LIC Aruba Belarus LMC Uruguay UMC Guinea LIC Bahamas, The Bosnia and Herzegovina LMC Venezuela, RB UMC Guinea-Bissau LIC Bahrain Bulgaria LMC Kenya LIC Bermuda Croatia UMC Middle East and North Africa Lesotho LMC Brunei Darussalam Czech Republic UMC Algeria LMC Liberia LIC Cayman Islands Estonia UMC Djibouti LMC Madagascar LIC Channel Islands Georgia LMC Egypt, Arab Rep. LMC Malawi LIC Cyprus Hungary UMC Iran, Islamic Rep. LMC Mali LIC Faeroe Islands Kazakhstan LMC Iraq LMC Mauritania LIC French Polynesia Kyrgyz Republic LIC Jordan LMC Mauritius UMC Greenland Latvia UMC Lebanon UMC Mayotte UMC Guam Lithuania UMC Libya UMC Mozambique LIC Hong Kong, China Macedonia, FYR LMC Morocco LMC Namibia LMC Isle of Man Moldova LMC Oman UMC Niger LIC Israel Poland UMC Syrian Arab Republic LMC Nigeria LIC Kuwait Romania UMC Tunisia LMC Rwanda LIC Liechtenstein Russian Federation UMC West Bank and Gaza LMC São Tomé and Principe LIC Macao, China Serbia and Montenegro LMC Yemen, Rep. LIC Senegal LIC Malta Slovak Republic UMC Seychelles UMC Monaco Tajikistan LIC Sierra Leone LIC Netherlands Antilles Turkey UMC Somalia LIC New Caledonia Turkmenistan LMC South Africa UMC Puerto Rico Ukraine LMC Sudan LIC Qatar Uzbekistan LIC Swaziland LMC San Marino Tanzania LIC Saudi Arabia Togo LIC Singapore Uganda LIC Slovenia Zambia LIC Taiwan, China Zimbabwe LIC United Arab Emirates Virgin Islands (U.S.) Source: World Bank data. Note: This table classifies all World Bank member economies, and all other economies with populations of more than 30,000. Economies are divided among income groups according to 2005 GNI per capita, calculated using the World Bank Atlas method. The groups are: low income (LIC), $875 or less; lower middle income (LMC), $876­3,465; upper middle income (UMC), $3,466­ 10,725; and high income, $10,726 or more. 287 Table 1. Key indicators of development Population age Gross national PPP gross national Population composition income (GNI)a income (GNI)b Life Adult Gross expectancy at Literacy Carbon domestic birth rate dioxide Average density product % ages emissions annual % people per % $ per $ per per capita Male Female 15 and per capita Millions growth sq. km Ages 0­14 $ billions capita $ billions capita % growth Years Years older metric tons 2005 2000­05 2005 2005 2005 2005 2005 2005 2004­05 2004 2004 2000­04 2002 Albania 3 0.5 114 27 8.1 2,580 17 5,420 4.9 71 77 99 0.8 Algeria 33 1.5 14 30 89.6 2,730 222c 6,770c 3.7 70 73 70 2.9 Angola 16 2.9 13 46 21.5 1,350 35c 2,210c 11.5 40 43 67 0.5 Argentina 39 1.0 14 26 173.0 4,470 539 13,920 8.2 71 79 97 3.5 Armenia 3 ­0.4 107 21 4.4 1,470 15 5,060 14.4 68 75 99 1.0 Australia 20 1.2 3 20 654.6 32,220 622 30,610 1.5 77 83 .. 18.1 Austria 8 0.5 100 16 303.6 36,980 272 33,140 1.4 76 82 .. 7.9 Azerbaijan 8 0.8 102 26 10.4 1,240 41 4,890 25.0 70 75 99 3.4 Bangladesh 142 1.9 1,090 35 66.2 470 296 2,090 3.5 63 64 .. 0.3 Belarus 10 ­0.5 47 15 27.0 2,760 77 7,890 9.8 63 74 100 6.0 Belgium 10 0.4 319 17 373.8 35,700 342 32,640 0.7 76 82 .. 8.9 Benin 8 3.2 76 44 4.3 510 9 1,110 0.7 54 55 35 0.3 Bolivia 9 2.0 9 38 9.3 1,010 25 2,740 2.1 62 67 87 1.2 Bosnia and Herzegovina 4 0.2 76 17 9.5 2,440 30 7,790 5.4 72 77 97 4.7 Brazil 186 1.4 22 28 644.1 3,460 1,534 8,230 0.9 67 75 89 1.8 Bulgaria 8 ­0.8 70 14 26.7 3,450 67 8,630 5.8 69 76 98 5.3 Burkina Faso 13 3.2 48 47 5.2 400 16c 1,220c 1.6 47 49 22 0.1 Burundi 8 3.1 294 45 0.7 100 5c 640c ­2.6 43 45 59 0.0 Cambodia 14 2.0 80 37 5.3 380 35c 2,490c 5.0 53 60 74 0.0 Cameroon 16 1.9 35 41 16.5 1,010 35 2,150 0.8 45 47 68 0.2 Canada 32 1.0 4 18 1,051.9 32,600 1,040 32,220 2.0 77 83 .. 16.5 Central African Republic 4 1.3 7 43 1.4 350 5c 1,140c 0.9 39 40 49 0.1 Chad 10 3.5 8 47 3.9 400 14 1,470 2.3 43 45 26 0.0 Chile 16 1.1 22 25 95.7 5,870 187 11,470 5.2 75 81 96 3.6 China 1,305 0.6 140 21 2,263.8 1,740 8,610d 6,600d 9.2 70 73 91 2.7 Hong Kong, China 7 0.8 .. 14 192.1 27,670 241 34,670 6.3 79 85 .. 5.2 Colombia 46 1.6 44 31 104.5 2,290 338c 7,420c 3.6 70 76 93 1.3 Congo, Dem. Rep. 58 2.8 25 47 6.9 120 41c 720c 3.5 43 45 67 0.0 Congo, Rep. 4 3.1 12 47 3.8 950 3 810 6.0 51 54 .. 0.6 Costa Rica 4 1.9 85 28 19.9 4,590 42c 9,680c 2.3 76 81 95 1.4 Cote d'Ivoire 18 1.6 57 42 15.3 840 27 1,490 ­1.9 45 47 49 0.4 Croatia 4 0.2 80 16 35.8 8,060 57 12,750 4.2 72 79 98 4.7 Czech Republic 10 ­0.1 132 15 109.2 10,710 205 20,140 6.2 73 79 .. 11.2 Denmark 5 0.3 128 19 256.8 47,390 182 33,570 2.8 75 80 .. 8.8 Dominican Republic 9 1.5 184 33 21.1 2,370 64c 7,150c 3.0 64 71 87 2.5 Ecuador 13 1.5 48 32 34.8 2,630 54 4,070 2.5 72 78 91 2.0 Egypt, Arab Rep. 74 1.9 74 34 92.9 1,250 329 4,440 2.9 68 73 71 2.1 El Salvador 7 1.8 332 34 16.8 2,450 35c 5,120c 1.0 68 74 .. 1.0 Eritrea 4 4.4 44 45 1.0 220 4c 1,010c 0.8 53 56 .. 0.2 Ethiopia 71 2.1 71 45 11.1 160 71c 1,000c 6.8 42 43 .. 0.1 Finland 5 0.3 17 17 196.5 37,460 163 31,170 1.8 75 82 .. 12.0 France 61 0.6 110 18 2,177.7e 34,810e 1,855 30,540 0.9 77 84 .. 6.2 Georgia 4 ­1.1 64 19 6.0 1,350 15c 3,270c 10.4 67 75 .. 0.7 Germany 82 0.1 236 14 2,852.3 34,580 2,409 29,210 0.9 76 81 .. 10.3 Ghana 22 2.2 97 39 10.0 450 52c 2,370c 3.7 57 58 58 0.4 Greece 11 0.3 86 14 218.1 19,670 262 23,620 3.4 77 81 96 8.5 Guatemala 13 2.4 116 43 30.3 2,400 56c 4,410c 0.8 64 71 69 0.9 Guinea 9 2.2 38 44 3.5 370 21 2,240 0.8 54 54 29 0.1 Haiti 9 1.4 309 37 3.9 450 16c 1,840c 0.5 51 53 .. 0.2 Honduras 7 2.3 64 39 8.6 1,190 21c 2,900c 2.3 66 70 80 0.9 Hungary 10 ­0.2 110 16 101.2 10,030 171 16,940 4.3 69 77 .. 5.6 India 1,095 1.5 368 32 793.0 720 3,787c 3,460c 7.1 63 64 61 1.2 Indonesia 221 1.3 122 28 282.2 1,280 820 3,720 4.2 66 69 90 1.4 Iran, Islamic Rep. 68 1.2 41 29 187.4 2,770 545 8,050 4.9 69 72 77 5.5 Ireland 4 1.7 60 20 166.6 40,150 144 34,720 2.6 76 81 .. 11.0 Israel 7 1.9 318 28 128.7 18,620 175 25,280 3.5 77 81 97 10.6 Italy 57 ­0.1 195 14 1,724.9 30,010 1,657 28,840 0.2 77 83 98 7.5 Jamaica 3 0.5 245 31 9.0 3,400 11 4,110 1.3 69 73 80 4.1 Japan 128 0.2 351 14 4,988.2 38,980 4,019 31,410 2.6 78 85 .. 9.4 Jordan 5 2.6 61 37 13.5 2,500 29 5,280 4.5 70 73 90 3.3 Kazakhstan 15 0.3 6 23 44.4 2,930 117 7,730 8.4 60 71 100 9.9 Kenya 34 2.2 60 43 18.0 530 40 1,170 0.4 49 48 74 0.2 Korea, Rep. 48 0.5 489 19 764.7 15,830 1,055 21,850 3.5 74 81 .. 9.4 Kuwait 3 2.9 142 24 59.1 24,040 59c 24,010c 5.3 75 80 93 25.6 Kyrgyz Republic 5 0.9 27 31 2.3 440 10 1,870 ­1.8 64 72 99 1.0 Lao PDR 6 2.3 26 41 2.6 440 12 2,020 4.6 54 57 69 0.2 Latvia 2 ­0.6 37 15 15.5 6,760 31 13,480 10.8 66 78 100 2.7 Lebanon 4 1.0 350 29 22.1 6,180 21 5,740 ­0.0 70 75 .. 4.7 Lithuania 3 ­0.5 55 17 24.1 7,050 49 14,220 8.0 66 78 100 3.6 Macedonia, FYR 2 0.2 80 20 5.8 2,830 14 7,080 3.8 71 76 96 5.1 Madagascar 19 2.8 32 44 5.4 290 16 880 1.8 54 57 71 0.1 Malawi 13 2.3 137 47 2.1 160 8 650 0.4 40 40 64 0.1 Malaysia 25 2.0 77 32 125.8 4,960 262 10,320 3.4 71 76 89 6.3 Mali 14 3.0 11 48 5.1 380 14 1,000 2.3 48 49 19 0.0 Mauritania 3 3.0 3 43 1.7 560 7c 2,150c 2.3 52 55 51 1.1 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 288 Table 1. Key indicators of development (continued) Population age Gross national PPP gross national Population composition income (GNI)a income (GNI)b Life Adult Gross expectancy at Literacy Carbon domestic birth rate dioxide Average density product % ages emissions annual % people per % $ per $ per per capita Male Female 15 and per capita Millions growth sq. km Ages 0­14 $ billions capita $ billions capita % growth Years Years older metric tons 2005 2000­05 2005 2005 2005 2005 2005 2005 2004­05 2004 2004 2000­04 2002 Mexico 103 1.0 54 31 753.4 7,310 1,034 10,030 1.9 73 78 91 3.8 Moldova 4 ­0.3 128 18 3.2f 880f 9 2,150 7.3 65 72 98 1.6 Mongolia 3 1.3 2 30 1.8 690 6 2,190 4.6 62 68 98 3.4 Morocco 30 1.7 68 31 52.3 1,730 132 4,360 0.4 68 72 52 1.5 Mozambique 20 2.0 25 44 6.1 310 25c 1,270c 5.7 41 42 .. 0.1 Namibia 2 1.4 3 42 6.1 2,990 16c 7,910c 2.4 47 48 85 1.1 Nepal 27 2.1 190 39 7.3 270 42 1,530 0.3 62 63 49 0.2 Netherlands 16 0.5 482 18 598.0 36,620 530 32,480 0.8 76 81 .. 9.3 New Zealand 4 1.4 15 21 106.7 25,960 95 23,030 0.7 77 82 .. 8.6 Nicaragua 5 2.0 45 39 5.0 910 20 3,650 1.9 68 73 77 0.7 Niger 14 3.4 11 49 3.3 240 11c 800c 1.1 45 45 29 0.1 Nigeria 132 2.3 144 44 74.2 560 137c 1,040c 4.7 43 44 .. 0.4 Norway 5 0.6 15 20 275.2 59,590 187 40,420 1.7 78 82 .. 13.9 Oman 3 1.0 8 35 23.0 9,070 37 14,680 .. 73 76 81 12.1 Pakistan 156 2.4 202 38 107.3 690 366 2,350 5.2 64 66 50 0.7 Panama 3 1.8 43 30 15.0 4,630 24c 7,310c 4.5 73 78 92 2.0 Papua New Guinea 6 2.1 13 40 3.9 660 14c 2,370c 1.0 55 57 57 0.4 Paraguay 6 2.4 16 38 7.9 1,280 31c 4,970c 0.4 69 74 .. 0.7 Peru 28 1.5 22 32 73.0 2,610 163 5,830 5.1 68 73 88 1.0 Philippines 83 1.9 279 35 108.3 1,300 440 5,300 3.3 69 73 93 0.9 Poland 38 ­0.2 125 16 271.4 7,110 515 13,490 3.3 70 79 .. 7.7 Portugal 11 0.6 115 16 170.7 16,170 208 19,730 ­0.2 74 81 .. 6.0 Romania 22 ­0.7 94 15 82.9 3,830 193 8,940 4.4 68 75 97 4.0 Russian Federation 143 ­0.4 9 15 639.1 4,460 1,523 10,640 6.9 59 72 99 9.8 Rwanda 9 2.3 366 43 2.1 230 12c 1,320c 3.2 42 46 65 0.1 Saudi Arabia 25 2.7 11 37 289.2 11,770 362c 14,740c 3.9 70 74 79 15.0 Senegal 12 2.4 61 43 8.2 710 21c 1,770c 3.7 55 57 39 0.4 Serbia and Montenegro 8 0.1 80 18 26.8g 3,280g .. .. 5.7 71 76 96 .. Sierra Leone 6 4.2 77 43 1.2 220 4 780 3.8 40 43 35 0.1 Singapore 4 1.4 6,495 20 119.6 27,490 130 29,780 3.7 77 81 93 13.7 Slovak Republic 5 0.0 112 17 42.8 7,950 85 15,760 5.9 70 78 100 6.8 Slovenia 2 0.1 99 14 34.7 17,350 44 22,160 3.8 73 81 .. 7.7 South Africa 45 0.5 37 33 224.1 4,960 548c 12,120c 5.6 44 45 82 7.6 Spain 43 1.4 87 14 1,100.1 25,360 1,120 25,820 1.7 77 84 .. 7.4 Sri Lanka 20 0.5 303 24 22.8 1,160 89 4,520 4.4 72 77 91 0.5 Sudan 36 1.9 15 39 23.3 640 72 2,000 5.9 55 58 61 0.3 Sweden 9 0.4 22 17 370.5 41,060 284 31,420 2.3 78 83 .. 5.8 Switzerland 7 0.7 186 16 408.7 54,930 276 37,080 1.2 79 84 .. 5.6 Syrian Arab Republic 19 2.5 104 37 26.3 1,380 71 3,740 1.7 72 75 80 2.8 Tajikistan 7 1.1 47 39 2.2 330 8 1,260 6.2 61 67 99 0.7 Tanzania 38 2.0 43 43 12.7h 340h 28 730 5.0 46 47 69 0.1 Thailand 64 0.9 126 24 176.9 2,750 542 8,440 3.6 67 74 93 3.7 Togo 6 2.7 113 43 2.2 350 10c 1,550c 0.2 53 57 53 0.3 Tunisia 10 0.9 65 26 29.0 2,890 79 7,900 3.3 71 75 74 2.3 Turkey 73 1.5 94 29 342.2 4,710 612 8,420 6.0 69 71 87 3.0 Turkmenistan 5 1.4 10 32 .. ..i .. .. .. 59 67 99 9.1 Uganda 29 3.5 146 50 7.9 280 43c 1,500c 1.9 48 50 67 0.1 Ukraine 47 ­0.9 81 15 71.4 1,520 317 6,720 3.3 63 74 99 6.4 United Kingdom 60 0.2 249 18 2,263.7 37,600 1,968 32,690 1.2 76 81 .. 9.2 United States 296 1.0 32 21 12,969.6 43,740 12,438 41,950 2.5 75 80 .. 20.2 Uruguay 3 0.7 20 24 15.1 4,360 34 9,810 5.8 72 79 .. 1.2 Uzbekistan 27 1.5 63 33 13.5 510 54 2,020 5.5 64 70 .. 4.8 Venezuela, RB 27 1.8 30 31 127.8 4,810 171 6,440 7.5 71 77 93 4.3 Vietnam 83 1.1 255 30 51.7 620 250 3,010 7.4 68 73 90 0.8 West Bank and Gaza 4 4.1 .. 45 3.8 1,120 .. .. .. 71 75 92 .. Yemen, Rep. 21 3.2 40 46 12.7 600 19 920 1.0 60 63 .. 0.7 Zambia 12 1.7 16 46 5.7 490 11 950 3.4 39 38 68 0.2 Zimbabwe 13 0.6 34 40 4.5 340 25 1,940 ­7.6 38 37 .. 1.0 World 6,438s 1.2w 50w 28w 44,983.3t 6,987w 60,644t 9,420w 2.4w 65w 69w 80w 3.9w Low income 2,353 1.9 83 36 1,363.9 580 5,849 2,486 5.6 58 60 62 0.8 Middle income 3,073 0.9 45 26 8,113.1 2,640 22,115 7,195 5.4 68 73 90 3.3 Lower middle income 2,475 1.0 63 25 4,746.5 1,918 15,622 6,313 5.9 68 73 89 2.6 Upper middle income 599 0.6 21 24 3,367.9 5,625 6,541 10,924 5.0 66 73 94 6.2 Low & middle income 5,426 1.3 56 30 9,476.8 1,746 27,954 5,151 5.2 63 67 80 2.2 East Asia & Pacific 1,885 0.9 119 24 3,067.4 1,627 11,149 5,914 7.8 68 72 91 2.4 Europe & Central Asia 473 0.0 20 20 1,945.0 4,113 4,324 9,142 5.9 64 73 97 6.7 Latin America & Caribbean 551 1.4 28 30 2,209.7 4,008 4,472 8,111 3.1 69 75 90 2.4 Middle East & North Africa 305 1.9 34 33 684.6 2,241 1,856 6,076 2.8 68 71 72 3.2 South Asia 1,470 1.7 308 33 1,005.3 684 4,618 3,142 6.4 63 64 60 1.0 Sub-Saharan Africa 741 2.3 31 44 552.2 745 1,469 1,981 3.1 46 47 .. 0.7 High income 1,011 0.7 31 18 35,528.8 35,131 32,893 32,524 2.1 76 82 .. 12.8 a. Calculated using the World Bank Atlas method. b. PPP is purchasing power parity; see Definitions. c. The estimate is based on regression; others are extrapolated from the latest Inter- national Comparison Programme benchmark estimates. d. Based on a 1986 bilateral comparison of China and the United States (Ruoen and Kai 1995), employing a different methodology than that used for other countries. This interim methodology will be revised in the next few years. e. GNI and GNI per capita estimates include the French overseas departments of French Guiana, Guadeloupe, Martinique, and Reunion. f. Excludes data for Transnistria. g. Excludes data for Kosovo. h. Data refers to mainland Tanzania only. i. Estimated to be lower middle income ($876­$3,465). 289 Table 2. Poverty National poverty line International poverty line Population below the poverty line Population below the poverty line Population Poverty Population Poverty below gap at below gap at Survey Rural Urban National Survey Rural Urban National Survey $1 a day $1 a day $2 a day $2 a day year % % % year % % % year % % % % Albania 2002 29.6 19.8 25.4 .. .. .. 2002a <2 <0.5 11.8 2.0 Algeria 1988 16.6 7.3 12.2 1995 30.3 14.7 22.6 1995a <2 <0.5 15.1 3.8 Argentina 1995 .. 28.4 .. 1998 .. 29.9 .. 2003b 7.0 2.0 23.0 8.4 Armenia 1998­99 50.8 58.3 55.1 2001 48.7 51.9 50.9 2003a <2 <0.5 31.1 7.1 Azerbaijan 1995 .. .. 68.1 2001 42.0 55.0 49.0 2001a 3.7 0.6 33.4 9.1 Bangladesh 1995­96 55.2 29.4 51.0 2000 53.0 36.6 49.8 2000a 36.0 8.1 82.8 36.3 Belarus 2000 .. .. 41.9 .. .. .. 2002a <2 <0.5 <2 <0.5 Benin 1995 25.2 28.5 26.5 1999 33.0 23.3 29.0 2003a 30.9 8.2 73.7 31.7 Bolivia 1997 77.3 53.8 63.2 1999 81.7 50.6 62.7 2002b 23.2 13.6 42.2 23.2 Bosnia and Herzegovina 2001­02 19.9 13.8 19.5 .. .. .. .. .. .. .. Botswana .. .. .. .. .. .. 1993a 23.5 7.7 50.1 22.8 Brazil 1996 54.0 15.4 23.9 1998 51.4 14.7 22.0 2003b 7.5 3.4 21.2 8.5 Bulgaria 1997 .. .. 36.0 2001 .. .. 12.8 2003a <2 <0.5 6.1 1.5 Burkina Faso 1998 61.1 22.4 54.6 2003 52.4 19.2 46.4 2003a 27.2 7.3 71.8 30.4 Burundi 1990 36.0 43.0 36.4 .. .. .. 1998a 54.6 22.7 87.6 48.9 Cambodia 1997 40.1 21.1 36.1 1999 40.1 13.9 35.9 1997a 34.1 9.7 77.7 34.5 Cameroon 1996 59.6 41.4 53.3 2001 49.9 22.1 40.2 2001a 17.1 4.1 50.6 19.3 Central African Republic .. .. .. .. .. .. 1993a 66.6 38.1 84.0 58.4 Chad 1995­96 67.0 63.0 64.0 .. .. .. .. .. .. .. Chile 1996 .. .. 19.9 1998 .. .. 17.0 2000b <2 <0.5 9.6 2.5 China 1996 7.9 <2 6.0 1998 4.6 <2 4.6 2001a 16.6 3.9 46.7 18.4 Colombia 1995 79.0 48.0 60.0 1999 79.0 55.0 64.0 2003b 7.0 3.1 17.8 7.7 Costa Rica 1992 25.5 19.2 22.0 .. .. .. 2001b 2.2 0.8 7.5 2.8 Côte d'Ivoire .. .. .. .. .. .. 2002a 14.8 4.1 48.8 18.4 Croatia .. .. .. .. .. .. 2001a <2 <0.5 <2 <0.5 Czech Republic .. .. .. .. .. .. 1996b <2 <0.5 <2 <0.5 Dominican Republic 1992 49.0 19.3 33.9 1998 42.1 20.5 28.6 2003b 2.5 0.8 11.0 3.6 Ecuador 1995 56.0 19.0 34.0 1998 69.0 30.0 46.0 1998b 15.8 6.3 37.2 15.8 Egypt, Arab Rep. 1995­96 23.3 22.5 22.9 1999­00 .. .. 16.7 1999­00a 3.1 <0.5 43.9 11.3 El Salvador 1992 55.7 43.1 48.3 .. .. .. 2002b 19.0 9.3 40.5 17.7 Eritrea 1993­94 .. .. 53.0 .. .. .. .. .. .. .. Estonia 1995 14.7 6.8 8.9 .. .. .. 2003a <2 <0.5 7.5 1.9 Ethiopia 1995­96 47.0 33.3 45.5 1999­00 45.0 37.0 44.2 1999­00a 23.0 4.8 77.8 29.6 Gambia, The 1992 .. .. 64.0 1998 61.0 48.0 57.6 1998a 26.5 8.8 54.3 25.2 Georgia 2002 55.4 48.5 52.1 2003 52.7 56.2 54.5 2003a 6.5 2.1 25.3 8.6 Ghana 1992 .. .. 50.0 1998­99 49.9 18.6 39.5 1998­99a 44.8 17.3 78.5 40.8 Guatemala 1989 71.9 33.7 57.9 2000 74.5 27.1 56.2 2002b 13.5 5.5 31.9 13.8 Guinea 1994 .. .. 40.0 .. .. .. Haiti 1987 .. .. 65.0 1995 66.0 .. .. 2001b 53.9 26.6 78.0 47.4 Honduras 1997 58.0 35.0 47.0 1999 58.0 37.0 48.0 1999b 20.7 7.5 44.0 20.2 Hungary 1993 .. .. 14.5 1997 .. .. 17.3 2002a <2 <0.5 <2 <0.5 India 1993­94 37.3 32.4 36.0 1999­00 30.2 24.7 28.6 1999­00a 34.7 8.2 79.9 35.3 Indonesia 1996 .. .. 15.7 1999 34.4 16.1 27.1 2002a 7.5 0.9 52.4 15.7 Iran, Islamic Rep. .. .. .. .. .. .. 1998a <2 <0.5 7.3 1.5 Jamaica 1995 37.0 18.7 27.5 2000 25.1 12.8 18.7 2000a <2 <0.5 13.3 2.7 Jordan 1991 .. .. 15.0 1997 .. .. 11.7 2002­03a <2 <0.5 7.0 1.5 Kazakhstan 1996 39.0 30.0 34.6 .. .. .. 2003a <2 <0.5 16.0 3.8 Kenya 1994 47.0 29.0 40.0 1997 53.0 49.0 52.0 1997a 22.8 5.9 58.3 23.9 Korea, Rep. .. .. .. .. .. .. 1998b <2 <0.5 <2 <0.5 Kyrgyz Republic 2000 56.4 43.9 52.0 2001 51.0 41.2 47.6 2003a <2 <0.5 21.4 4.4 Lao PDR 1993 48.7 33.1 45.0 1997­98 41.0 26.9 38.6 2002a 27.0 6.1 74.1 30.2 Latvia .. .. .. .. .. .. 2003a <2 <0.5 4.7 1.2 Lesotho .. .. .. .. .. .. 1995a 36.4 19.0 56.1 33.1 Lithuania .. .. .. .. .. .. 2003a <2 <0.5 7.8 1.8 Macedonia, FYR .. .. .. .. .. .. 2003a <2 <0.5 <2 <0.5 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 290 Table 2. Poverty (continued) National poverty line International poverty line Population below the poverty line Population below the poverty line Population Poverty Population Poverty below gap at below gap at Survey Rural Urban National Survey Rural Urban National Survey $1 a day $1 a day $2 a day $2 a day year % % % year % % % year % % % % Madagascar 1997 76.0 63.2 73.3 1999 76.7 52.1 71.3 2001a 61.0 27.9 85.1 51.8 Malawi 1990­91 .. .. 54.0 1997­98 66.5 54.9 65.3 1997­98a 41.7 14.8 76.1 38.3 Malaysia 1989 .. .. 15.5 .. .. .. 1997b <2 <0.5 9.3 2.0 Mali 1998 75.9 30.1 63.8 .. .. .. 1994a 72.3 37.4 90.6 60.5 Mauritania 1996 65.5 30.1 50.0 2000 61.2 25.4 46.3 2000a 25.9 7.6 63.1 26.8 Mexico 1996 52.4 26.5 37.1 2002 34.8 11.4 20.3 2002a 4.5 1.2 20.4 6.5 Moldova 2001 64.1 58.0 62.4 2002 67.2 42.6 48.5 2001a 22.0 5.8 63.7 25.1 Mongolia 1995 33.1 38.5 36.3 1998 32.6 39.4 35.6 1998a 27.0 8.1 74.9 30.6 Morocco 1990­91 18.0 7.6 13.1 1998­99 27.2 12.0 19.0 1999a <2 <0.5 14.3 3.1 Mozambique 1996­97 71.3 62.0 69.4 .. .. .. 1996a 37.9 12.0 78.4 36.8 Namibia .. .. .. .. .. .. 1993b 34.9 14.0 55.8 30.4 Nepal 1995­96 43.3 21.6 41.8 2003­04 34.6 9.6 30.9 2003­04a 24.1 5.4 68.5 26.8 Nicaragua 1993 76.1 31.9 50.3 1998 68.5 30.5 47.9 2001a 45.1 16.7 79.9 41.2 Niger 1989­93 66.0 52.0 63.0 .. .. .. 1995a 60.6 34.0 85.8 54.6 Nigeria 1985 49.5 31.7 43.0 1992­93 36.4 30.4 34.1 2003a 70.8 34.5 92.4 59.5 Pakistan 1993 33.4 17.2 28.6 1998­99 35.9 24.2 32.6 2002a 17.0 3.1 73.6 26.1 Panama 1997 64.9 15.3 37.3 .. .. .. 2002b 6.5 2.3 17.1 6.9 Papua New Guinea 1996 41.3 16.1 37.5 .. .. .. .. .. .. .. Paraguay 1991 28.5 19.7 21.8 .. .. .. 2002b 16.4 7.4 33.2 16.2 Peru 1994 67.0 46.1 53.5 1997 64.7 40.4 49.0 2002b 12.5 4.4 31.8 13.4 Philippines 1994 53.1 28.0 40.6 1997 50.7 21.5 36.8 2000a 15.5 3.0 47.5 17.8 Poland 1993 .. .. 23.8 .. .. .. 2002a <2 <0.5 <2 <0.5 Portugal .. .. .. .. .. .. 1994b <2 <0.5 <2 <0.5 Romania 1994 27.9 20.4 21.5 .. .. .. 2003a <2 0.5 12.9 3.0 Russian Federation 1994 .. .. 30.9 .. .. .. 2002a <2 <0.5 12.1 3.1 Rwanda 1993 .. .. 51.2 1999­00 65.7 14.3 60.3 1999­00a 51.7 20.0 83.7 45.5 Senegal 1992 40.4 23.7 33.4 .. .. .. 1995a 22.3 5.7 63.0 25.2 Sierra Leone 1989 .. .. 82.8 2003­04 79.0 56.4 70.2 1989a 57.0 39.5 74.5 51.8 Slovak Republic .. .. .. .. .. .. 1996b <2 <0.5 2.9 0.8 Slovenia .. .. .. .. .. .. 1998a <2 <0.5 <2 <0.5 South Africa .. .. .. .. .. .. 2000a 10.7 1.7 34.1 12.6 Sri Lanka 1990­91 22.0 15.0 20.0 1995­96 27.0 15.0 25.0 2002a 5.6 0.8 41.6 11.9 Tajikistan .. .. .. .. .. .. 2003a 7.4 1.3 42.8 13.0 Tanzania 1991 40.8 31.2 38.6 2000­01 38.7 29.5 35.7 2000­01a 57.8 20.7 89.9 49.3 Thailand 1990 .. .. 18.0 1992 15.5 10.2 13.1 2002a <2 <0.5 25.1 6.2 Togo 1987­89 .. .. 32.3 .. .. .. .. .. .. .. Trinidad and Tobago 1992 20.0 24.0 21.0 1992b 4.0 1.0 20.0 6.3 Tunisia 1990 13.1 3.5 7.4 1995 13.9 3.6 7.6 2000a <2 <0.5 6.6 1.3 Turkey 1994 .. .. 28.3 2002 34.5 21.9 27.0 2003a 3.4 0.8 18.7 5.7 Uganda 1999­00 37.4 9.6 33.8 2002­03 41.7 12.2 37.7 .. .. .. .. Ukraine 2000 34.9 31.5 2003 28.4 19.5 2003b <2 <0.5 4.9 0.9 Uruguay 1994 .. 20.2 .. 1998 .. 24.7 .. 2003b <2 <0.5 5.7 1.6 Uzbekistan 2000 30.5 22.5 27.5 .. .. .. Venezuela, RB 1989 .. .. 31.3 .. .. .. 2000b 8.3 2.8 27.6 10.2 Vietnam 1998 45.5 9.2 37.4 2002 35.6 6.6 28.9 Yemen, Rep. 1998 45.0 30.8 41.8 .. .. .. 1998a 10.2 2.3 45.2 15.0 Zambia 1996 82.8 46.0 69.2 1998 83.1 56.0 72.9 2002­03a 75.8 36.4 94.1 62.2 Zimbabwe 1990­91 35.8 3.4 25.8 1995­96 48.0 7.9 34.9 1995­96a 56.1 24.2 83.0 48.2 a. Expenditure base. b. Income base. 291 Table 3. Millennium Development Goals: eradicating poverty and improving lives Achieve Eradicate universal Combat HIV/ extreme poverty primary Promote gender Reduce child AIDS and other and hunger education equality mortality Improve maternal health diseases Maternal mortality Prevalence ratio per HIV of child Gender parity 100,000 live Births attended prevalence Percentage malnutrition Primary ratio in primary Under-five births by skilled health % of share of % of children completion and secondary mortality rate per Modeled staff population poorest quintile under 5 rate (%) school 1,000 estimates % of total ages 15­49 in national consumption or 1989­ 2000­ 1990­ 2000­ Survey year income 94a 04a 1991 2004 1991 2004 1990 2004 2000 94a 04a 2005 Albania 2002b 9.1 .. 14 .. 99 96 97 45 19 55 .. 98 .. Algeria 1995b 7.0 9 10 79 94 83 99 69 40 140 77 96 0.1 Angola .. 20 31 35 .. .. .. 260 260 1,700 .. 45 3.7 Argentina 2003c,d 3.2 2 .. .. 100 .. 103 29 18 82 96 99 0.6 Armenia 2003b 8.5 .. 3 .. 107 .. 103 60 32 55 .. 97 0.1 Australia 1994d 5.9 .. .. .. 98 101 98 10 6 8 100 .. 0.1 Austria 2000d 8.6 .. .. .. .. 95 96 10 5 4 100 .. 0.3 Azerbaijan 2002b 12.2 .. 7 .. 96 100 97 105 90 94 .. 84 0.1 Bangladesh 2000b 9.0 68 48 .. 76 .. 106 149 77 380 10 13 <0.1 Belarus 2002b 8.5 .. .. 95 101 .. 100 17 11 35 .. 100 0.3 Belgium 2000d 8.5 .. .. 79 .. 101 98 10 5 10 .. .. 0.3 Benin 2003b 7.4 .. 23 21 49 50 71 185 152 850 .. 66 1.8 Bolivia 2002d 1.5 15 8 .. 100 .. 98 125 69 420 47 67 0.1 Bosnia and Herzegovina 2001b 9.5 .. 4 .. .. .. .. 22 15 31 97 100 <0.1 Brazil 2003d 2.6 7 .. .. 109 .. 103 60 34 260 72 96 0.5 Bulgaria 2003b 8.7 .. .. 85 98 99 96 19 15 32 .. 99 <0.1 Burkina Faso 2003b 6.9 33 38 21 29 62 76 210 192 1,000 42 38 2.0 Burundi 1998b 5.1 .. 45 46 33 82 82 190 190 1,000 .. 25 3.3 Cambodia 1997b 6.9 .. 45 .. 82 73 87 115 141 450 .. 32 1.6 Cameroon 2001b 5.6 15 18 56 63 83 81 139 149 730 58 62 5.4 Canada 2000d 7.2 .. .. .. .. 99 100 8 6 6 .. 98 0.3 Central African Republic 1993b 2.0 .. 24 27 .. 60 .. 168 193 1,100 .. 44 10.7 Chad .. .. 37 18 29 41 58 203 200 1,100 .. 14 3.5 Chile 2000d 3.3 1 1 .. 95 100 98 21 8 31 100 100 0.3 China 2001b 4.7 17 8 103 .. 87 100 49 31 56 .. 96 0.1 Hong Kong, China 1996d 5.3 .. .. 102 111 103 95 .. .. .. .. .. .. Colombia 2003d 2.5 10 7 70 94 108 104 36 21 130 82 86 0.6 Congo, Dem. Rep. .. .. 31 46 .. .. .. 205 205 990 .. 61 3.2 Congo, Rep. .. .. .. 54 66 85 90 110 108 510 .. .. 5.3 Costa Rica 2001d 3.9 2 .. 79 92 101 102 18 13 43 98 98 0.3 Côte d'Ivoire 2002b 5.2 24 17 43 43 65 68 157 194 690 45 68 7.1 Croatia 2001b 8.3 1 .. .. 91 102 101 12 7 8 .. 100 <0.1 Czech Republic 1996d 10.3 1 .. .. 104 98 100 13 4 9 .. 100 0.1 Denmark 1997d 8.3 .. .. 98 99 101 102 9 5 5 .. .. 0.2 Dominican Republic 2003d 3.9 10 5 .. 91 .. 105 65 32 150 93 98 1.1 Ecuador 1998b 3.3 .. 12 .. 101 .. 100 57 26 130 .. .. 0.3 Egypt, Arab Rep. 1999­2000b 8.6 10 9 .. 95 81 95 104 36 84 41 69 <0.1 El Salvador 2002d 2.7 11 10 41 86 102 98 60 28 150 51 92 0.9 Eritrea .. 41 40 .. 43 .. 71 147 82 630 .. 28 2.4 Ethiopia 1999­2000b 9.1 48 47 .. 55 68 73 204 166 850 .. 6 .. Finland 2000d 9.6 .. .. 97 100 109 102 7 4 6 100 100 0.1 France 1995d 7.2 .. .. 104 99 102 100 9 5 17 99 .. 0.4 Georgia 2003b 5.6 .. .. .. 91 98 99 47 45 32 .. .. 0.2 Germany 2000d 8.5 .. .. .. 96 .. 99 9 5 8 .. .. 0.1 Ghana 1998­99b 5.6 27 22 63 72 79 91 122 112 540 44 47 2.3 Greece 2000d 6.7 .. .. .. .. 99 100 11 5 9 .. .. 0.2 Guatemala 2002d 2.9 .. 23 .. 70 .. 91 82 45 240 .. 41 0.9 Guinea 1994b 6.4 27 33 17 49 46 72 240 155 740 31 56 1.5 Haiti 2001d 2.4 27 17 27 .. 95 .. 150 117 680 23 24 3.8 Honduras 2003d 3.4 18 17 65 79 108 107 59 41 110 45 56 1.5 Hungary 2002b 9.5 .. .. 93 95 100 99 17 8 16 .. 100 0.1 India 1999­2000b 8.9 53 .. .. 89 70 88 123 85 540 34 43 0.9 Indonesia 2002b 8.4 .. 28 91 101 93 99 91 38 230 37 72 0.1 Iran, Islamic Rep. 1998b 5.1 .. .. 91 95 85 100 72 38 76 .. 90 0.2 Ireland 2000d 7.4 .. .. .. 101 104 102 9 6 5 .. 100 0.2 Israel 2001d 5.7 .. .. .. 104 105 100 12 6 17 .. .. .. Italy 2000d 6.5 .. .. 104 101 100 99 9 5 5 .. .. 0.5 Jamaica 2000b 6.7 5 4 90 84 102 101 20 20 87 79 97 1.5 Japan 1993d 10.6 .. .. 101 .. 101 100 6 4 10 100 .. <0.1 Jordan 2002­03b 6.7 6 4 73 97 101 101 40 27 41 87 100 .. Kazakhstan 2003b 7.4 .. .. .. 110 102 98 63 73 210 .. .. 0.1 Kenya 1997b 6.0 23 20 .. 91 94 94 97 120 1,000 45 42 6.1 Korea, Rep. 1998d 7.9 .. .. 98 105 99 100 9 6 20 98 .. <0.1 Kuwait .. .. .. .. 91 97 104 16 12 5 .. .. .. Kyrgyz Republic 2003b 8.9 .. 7 .. 93 .. 101 80 68 110 .. 99 0.1 Lao PDR 2002b 8.1 40 40 .. 74 75 84 163 83 650 .. 19 0.1 Latvia 2003b 6.6 .. .. .. 92 100 99 18 12 42 .. .. 0.8 Lebanon .. .. .. .. 94 .. 102 37 31 150 .. .. 0.1 Lithuania 2003b 6.8 .. .. .. 98 .. 99 13 8 13 .. 100 0.2 Macedonia, FYR 2003b 6.1 .. .. .. 96 99 99 38 14 23 .. 99 <0.1 Madagascar 2001b 4.9 45 42 33 45 98 .. 168 123 550 57 51 0.5 Malawi 1997b 4.9 28 22 28 59 81 99 241 175 1,800 55 61 14.1 Malaysia 1997d 4.4 22 11 91 91 101 106 22 12 41 .. 97 0.5 Mali 1994b 4.6 .. 33 11 44 59 74 250 219 1,200 .. 41 1.7 Mauritania 2000b 6.2 48 32 33 43 67 96 133 125 1,000 40 57 0.7 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 292 Table 3. Millennium Development Goals: eradicating poverty and improving lives (continued) Achieve Eradicate universal Combat HIV/ extreme poverty primary Promote gender Reduce child AIDS and other and hunger education equality mortality Improve maternal health diseases Maternal mortality Prevalence ratio per HIV of child Gender parity 100,000 live Births attended prevalence Percentage malnutrition Primary ratio in primary Under-five births by skilled health % of share of % of children completion and secondary mortality rate per Modeled staff population poorest quintile under 5 rate (%) school 1,000 estimates % of total ages 15­49 in national consumption or 1989­ 2000­ 1990­ 2000­ Survey year income 94a 04a 1991 2004 1991 2004 1990 2004 2000 94a 04a 2005 Mexico 2002b 4.3 17 .. 86 99 98 102 46 28 83 .. 95 0.3 Moldova 2003b 7.8 .. .. .. 91 105 102 40 28 36 .. .. 1.1 Mongolia 1998b 5.6 12 13 .. 96 109 108 108 52 110 .. 99 <0.1 Morocco 1998­99b 6.5 10 10 47 75 70 88 89 43 220 31 63 0.1 Mozambique 1996­97b 6.5 .. 24 .. 30 72 82 235 152 1,000 .. 48 16.1 Namibia 1993d 1.4 26 24 .. 81 108 104 86 63 300 68 76 19.6 Nepal 2003­04b 6.0 .. 48 .. 75 59 85 145 76 740 7 15 0.5 Netherlands 1999d 7.6 .. .. .. 100 97 98 9 6 16 .. .. 0.2 New Zealand 1997d 6.4 .. .. 100 .. 101 105 11 7 7 95 .. 0.1 Nicaragua 2001b 5.6 11 10 44 73 109 103 68 38 230 .. 67 0.2 Niger 1995b 2.6 43 40 17 25 57 71 320 259 1,600 15 16 1.1 Nigeria 2003b 5.1 39 29 .. 75 79 84 230 197 800 31 35 3.9 Norway 2000d 9.6 .. .. 100 101 102 101 9 4 16 .. .. 0.1 Oman .. 24 .. .. 91 89 98 32 13 87 .. 95 .. Pakistan 2002b 9.3 40 38 .. .. .. 73 130 101 500 19 23 0.1 Panama 2002d 2.5 6 .. .. 97 .. 101 34 24 160 86 93 0.9 Papua New Guinea 1996b 4.5 .. .. 47 54 80 87 101 93 300 .. 41 1.8 Paraguay 2002d 2.2 4 5 71 91 99 98 41 24 170 67 77 0.4 Peru 2002d 3.2 11 7 .. 100 96 100 80 29 410 .. 59 0.6 Philippines 2000b 5.4 30 28 .. 97 100 102 62 34 200 53 60 <0.1 Poland 2002b 7.5 .. .. 98 100 101 100 18 8 13 .. 100 0.1 Portugal 1997d 5.8 .. .. 95 .. 103 102 14 5 5 98 100 0.4 Romania 2003b 8.1 6 3 .. 93 99 100 31 20 49 99 99 <0.1 Russian Federation 2002b 6.1 4 6 .. .. 104 99 29 21 67 .. 99 1.1 Rwanda 1983­85b .. 29 24 33 37 96 100 173 203 1,400 26 31 3.1 Saudi Arabia .. 15 .. 56 62 84 92 44 27 23 .. .. .. Senegal 1995b 6.4 22 23 .. 45 69 90 148 137 690 47 58 0.9 Serbia and Montenegro .. .. 2 .. 96 .. 101 28 15 11 .. 93 0.2 Sierra Leone 1989b .. 29 27 .. .. 67 74 302 283 2,000 .. 42 1.6 Singapore 1998d 5.0 .. 3 .. .. 95 .. 8 3 30 .. .. 0.3 Slovak Republic 1996d 8.8 .. .. .. 99 .. 101 14 9 3 .. 99 <0.1 Slovenia 1998­99d 9.1 .. .. .. 114 .. 100 10 4 17 100 100 <0.1 South Africa 2000b 3.5 .. .. 75 96 104 101 60 67 230 .. .. 18.8 Spain 2000d 7.0 .. .. .. .. 104 102 9 5 4 .. .. 0.6 Sri Lanka 1999­2000b 8.3 38 30 97 .. 102 102 32 14 92 94 96 <0.1 Sudan .. 34 41 41 49 78 88 120 91 590 86 87 1.6 Sweden 2000d 9.1 .. .. 96 .. 102 102 7 4 2 .. .. 0.2 Switzerland 2000d 7.6 .. .. 53 97 97 96 9 5 7 .. .. 0.4 Syrian Arab Republic .. 12 7 89 107 85 94 44 16 160 77 .. .. Tajikistan 2003b 7.9 .. .. .. 92 .. 89 119 93 100 .. 71 0.1 Tanzania 2000­01b 7.3 29 .. 61 54 97 .. 161 126 1,500 44 46 6.5 Thailand 2002b 6.3 19 .. .. .. 95 98 37 21 44 .. 99 1.4 Togo .. .. .. 35 66 59 73 152 140 570 .. 61 3.2 Tunisia 2000b 6.0 .. 4 74 97 86 102 52 25 120 .. 90 0.1 Turkey 2003b 5.3 10 4 90 88 81 86 82 32 70 76 83 .. Turkmenistan 1998b 6.1 .. 12 .. .. .. .. 97 103 31 .. 97 <0.1 Uganda 1999b 5.9 23 23 .. 57 82 97 160 138 880 38 39 6.7 Ukraine 2003b 9.2 .. 1 92 100 .. 99 26 18 35 .. 100 1.4 United Kingdom 1999d 6.1 .. .. .. .. 98 102 10 6 13 .. .. .. United States 2000d 5.4 1 2 .. .. 100 99 11 8 17 99 .. 0.6 Uruguay 2003c,d 5.0 4 .. 94 91 .. 106 25 17 27 .. .. 0.5 Uzbekistan 2000b 9.2 .. 8 .. 97 94 98 79 69 24 .. 96 0.2 Venezuela, RB 2000d 4.7 5 4 43 89 105 103 27 19 96 .. 94 0.7 Vietnam 2002b 7.5 45 28 .. 101 .. 94 53 23 130 .. 90 0.5 West Bank and Gaza .. .. .. .. 98 .. 103 .. .. .. .. 97 .. Yemen, Rep. 1998b 7.4 39 46 .. 62 .. 63 142 111 570 16 27 .. Zambia 2002­03b 6.1 25 23 .. 66 .. 93 180 182 750 51 43 17.0 Zimbabwe 1995b 4.6 16 .. 99 80 92 96 80 129 1,100 69 .. 20.1 World ..w 25w,e ..w ..w 86w 93w 95w 79w 410w 43w 62w 1.0w Low income .. 39e 64f 78g 73 85 147 122 684 32 41 1.7 Middle income .. 11e 91f 96g 91 99 58 39 150 .. 87 0.6 Lower middle income .. 12e 93f 97g 89 99 62 42 163 .. 86 0.3 Upper middle income .. 7e 87f 95g 98 98 41 28 91 .. 95 2.2 Low & middle income .. 26e 80f 87g 84 92 103 86 450 40 60 1.1 East Asia & Pacific 19 15e 97f 98g 89 99 59 37 117 .. 86 0.2 Europe & Central Asia .. 5e 92f 94g 97 96 49 34 58 .. 94 0.7 Latin America & Carib. .. 7e 84f 96g .. 102 54 31 194 77 88 0.6 Middle East & N. Africa .. 13e 75f 88g 82 92 81 55 183 46 72 0.1 South Asia 53 45e 73f 87g 70 85 129 92 564 30 37 0.7 Sub-Saharan Africa .. 29e 51f 61g 80 83 185 168 921 43 42 6.2 High income .. 3e .. .. 100 99 11 7 14 .. 99 0.4 a. Data are for the most recent year available. b. Refers to expenditure shares by percentiles of population, ranked by per capita expenditure. c. Urban data. d. Refers to income shares by percentiles of population, ranked by per capita income. e. Data are for 1995­2004. f. Data are for 1989­94. g. Data are for 2000­05. 293 Table 4. Economic activity External General balance Household gov't. final Gross of goods GDP Agricultural final cons. cons. capital and implicit Gross domestic product productivity Value added as % of GDP expenditure expenditure formation services deflator Agricultural value Millions of Avg. annual added per worker Avg. annual dollars % growth 2000 $ Agriculture Industry Services % of GDP % of GDP % of GDP % of GDP % growth 2005 2000­05 1992­94 2002­04 2005 2005 2005 2005 2005 2005 2005 2000­05 Albania 8,379 5.3 916 1,469 25 20 55 88 9 25 ­23 4.0 Algeria 102,257 5.1 1,743 1,983 8 62 29 39 7 32 22 7.8 Angola 28,038 9.1 99 168 8 66 26 73 ..a 13 15 80.0 Argentina 183,309 2.2 7,335 9,311 10 36 54 63 11 19 7 12.5 Armenia 4,903 12.3 1,464 2,722 21 44 35 73 11 30 ­13 4.2 Australia 700,672 3.3 20,693 27,058 3 26 71 60 18 25 ­3 3.2 Austria 304,527 1.3 12,881 21,083 2 31 67 56 18 22 5 1.7 Azerbaijan 12,561 12.7 922 1,061 12 55 32 59 11 53 ­24 6.0 Bangladesh 59,958 5.3 251 309 21 28 52 77 6 24 ­7 4.3 Belarus 29,566 7.6 1,964 2,612 10 41 49 50 20 30 1 35.8 Belgium 364,735 1.5 27,442 41,536 1 25 73 54 23 20 3 2.0 Benin 4,287 4.0 391 591 32 13 54 78 15 20 ­13 2.9 Bolivia 9,334 3.0 678 749 16 31 53 69 15 12 4 4.8 Bosnia and Herzegovina 9,369 5.1 3,028 5,709 12 28 61 85 23 21 ­29 3.5 Brazil 794,098 2.2 1,839 3,111 10 38 52 58 15 19 8 10.1 Bulgaria 26,648 5.0 2,152 6,635 9 30 60 72 17 28 ­17 4.0 Burkina Faso 5,171 5.1 157 166 31 20 50 83 13 19 ­15 2.7 Burundi 800 2.2 104 79 35 20 45 87 28 12 ­28 8.3 Cambodia 5,391 6.6 276 289 33 29 38 80 5 26 ­11 2.7 Cameroon 16,985 3.8 720 1,111 41 14 45 70 10 20 ­0 2.2 Canada 1,115,192 2.6 29,378 38,509 .. .. .. 56 20 20 4 2.3 Central African Republic 1,369 ­1.4 292 415 54 21 25 .. .. .. .. 2.0 Chad 5,469 14.5 191 225 23 51 26 58 5 17 20 7.6 Chile 115,250 3.0 4,235 3,222 6 47 48 57 12 23 8 5.3 China 2,228,862 9.6 273 373 13 46 41 49 10 39 3 3.2 Hong Kong, China 177,722 4.3 .. .. 0 10 90 59 9 21 12 ­3.6 Colombia 122,309 3.5 3,208 2,971 13 34 53 61 19 19 0 6.6 Congo, Dem. Rep. 6,974 4.4 183 153 46 25 29 87 7 14 ­8 43.7 Congo, Rep. 5,091 3.9 295 337 6 46 48 34 14 24 28 ­0.6 Costa Rica 19,432 4.0 3,364 4,285 8 29 63 77 5 21 ­3 9.5 Cote d'Ivoire 16,055 ­0.5 608 757 22 21 57 71 8 10 10 2.9 Croatia 37,412 4.4 5,189 9,237 8 28 64 57 19 28 ­5 3.3 Czech Republic 122,345 3.5 3,531 4,543 3 39 58 50 22 28 0 2.8 Denmark 254,401 1.5 22,271 37,443 2 25 73 48 27 20 5 1.9 Dominican Republic 28,303 2.1 2,482 4,169 13 27 60 76 7 19 ­2 20.4 Ecuador 36,244 5.0 1,027 1,478 6 28 66 68 8 26 ­2 11.7 Egypt, Arab Rep. 89,336 3.7 1,575 2,007 14 39 47 70 13 17 ­0 5.6 El Salvador 16,974 2.2 1,639 1,618 11 30 60 92 11 15 ­18 2.9 Eritrea 986 3.6 91 56 23 23 55 82 45 20 ­48 15.1 Ethiopia 11,174 4.2 147 144 48 13 39 82 14 26 ­23 4.2 Finland 193,176 2.4 17,815 31,339 3 31 66 53 22 19 6 1.1 France 2,110,185 1.5 24,724 40,521 2 22 76 56 24 20 0 1.7 Georgia 6,395 7.4 2,127 1,442 17 27 56 65 18 27 ­10 6.0 Germany 2,781,900 0.7 13,908 23,616 1 29 70 59 19 17 5 0.9 Ghana 10,695 5.1 301 341 39 25 37 74 15 30 ­19 22.6 Greece 213,698 4.2 8,315 9,303 7 23 70 66 17 26 ­9 3.1 Guatemala 31,683 2.5 2,178 2,275 23 19 58 89 6 18 ­13 7.2 Guinea 2,689 2.9 175 229 26 38 37 85 6 12 ­3 9.4 Haiti 4,245 ­0.5 672 421 28 17 55 91 8 30 ­29 18.0 Honduras 7,976 3.6 992 1,163 13 31 56 72 14 29 ­15 7.3 Hungary 109,154 4.0 2,825 3,986 4 31 65 68 10 23 ­1 6.3 India 785,468 6.9 353 382 19 28 54 61 11 30 ­2 3.8 Indonesia 287,217 4.7 498 564 14 41 45 64 8 23 5 8.2 Iran, Islamic Rep. 196,343 5.8 2,042 2,438 10 44 46 51 14 32 5 18.8 Ireland 196,388 5.0 .. .. 3 41 56 44 15 25 16 3.5 Israel 123,434 1.9 .. .. .. .. .. 59 28 19 ­5 1.3 Italy 1,723,044 0.7 13,672 21,553 3 28 70 60 19 20 1 2.8 Jamaica 9,696 1.5 2,162 1,916 5 33 62 72 14 31 ­17 10.7 Japan 4,505,912 1.3 19,958 26,557 1 31 68 57 18 24 2 ­1.8 Jordan 12,861 5.9 1,810 1,192 2 29 69 91 16 27 ­34 2.6 Kazakhstan 56,088 10.1 1,585 1,420 7 40 54 52 11 28 9 11.7 Kenya 17,977 2.8 301 317 27 18 55 70 11 25 ­6 4.3 Korea, Rep. 787,624 4.6 6,257 9,996 4 41 55 52 13 30 4 2.4 Kuwait 74,658 7.3 .. 13,898 0 53 47 38 21 14 27 6.4 Kyrgyz Republic 2,441 4.0 625 942 34 21 45 82 18 20 ­20 4.7 Lao PDR 2,855 6.2 376 461 46 28 26 .. .. 17 ­4 11.0 Latvia 15,771 7.9 1,624 2,505 4 23 73 64 17 27 ­8 4.8 Lebanon 22,210 4.1 .. .. 7 21 72 87 15 20 ­22 2.5 Lithuania 25,495 7.8 .. 4,363 6 31 63 62 17 25 ­5 1.1 Macedonia, FYR 5,762 1.7 2,104 3,034 12 29 59 77 20 21 ­18 2.1 Madagascar 5,040 2.0 183 174 28 16 56 84 8 22 ­15 11.0 Malawi 2,072 3.4 73 131 35 19 46 95 17 15 ­26 14.7 Malaysia 130,143 4.8 3,918 4,690 9 50 40 43 13 23 21 3.4 Mali 5,098 5.8 205 229 36 24 40 79 10 24 ­13 3.5 Mauritania 1,888 5.0 283 282 17 32 51 .. .. .. .. 8.5 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 294 Table 4. Economic activity (continued) External General balance Household gov't. final Gross of goods GDP Agricultural final cons. cons. capital and implicit Gross domestic product productivity Value added as % of GDP expenditure expenditure formation services deflator Agricultural value Millions of Avg. annual added per worker Avg. annual dollars % growth 2000 $ Agriculture Industry Services % of GDP % of GDP % of GDP % of GDP % growth 2005 2000­05 1992­94 2002­04 2005 2005 2005 2005 2005 2005 2005 2000­05 Mexico 768,438 1.9 2,295 2,727 4 26 70 68 12 22 ­2 7.1 Moldova 2,906 7.0 902 732 21 24 55 97 15 20 ­32 10.6 Mongolia 1,880 5.8 811 661 22 27 51 57 17 37 ­11 11.5 Morocco 51,745 4.2 1,275 1,582 13 31 56 63 20 26 ­9 1.0 Mozambique 6,630 8.6 98 142 23 30 47 77 11 22 ­10 11.4 Namibia 6,126 4.6 845 1,097 10 32 58 50 23 26 1 5.3 Nepal 7,346 2.6 191 208 40 21 38 76 11 26 ­13 4.1 Netherlands 594,755 0.6 27,857 39,358 2 26 72 49 25 20 5 2.7 New Zealand 109,041 3.8 20,319 27,660 .. .. .. 60 18 23 0 2.4 Nicaragua 4,911 3.0 1,221 1,916 19 30 52 88 11 29 ­28 6.9 Niger 3,405 3.7 165 172 40 17 43 79 12 19 ­9 2.4 Nigeria 98,951 5.9 610 863 24 56 20 41 21 21 18 17.4 Norway 283,920 1.7 23,252 32,779 2 39 59 45 22 19 14 2.4 Oman 24,284 3.0 1,000 1,128 2 56 42 45 23 18 14 1.8 Pakistan 110,732 4.8 603 688 22 25 53 80 8 17 ­5 6.0 Panama 15,467 4.3 2,450 3,570 8 18 75 69 13 20 ­2 1.6 Papua New Guinea 4,731 1.3 451 482 26 45 30 .. .. .. .. 9.4 Paraguay 8,152 1.8 2,165 2,453 27 24 49 72 7 24 ­3 12.3 Peru 78,431 4.2 1,169 1,764 9 33 58 66 10 19 5 2.7 Philippines 98,306 4.5 901 1,021 14 33 53 75 10 16 ­1 5.1 Poland 299,151 3.1 1,510 2,003 5 31 65 63 19 19 ­0 2.3 Portugal 173,085 0.3 4,414 5,735 4 27 70 63 21 23 ­8 3.3 Romania 98,559 5.8 2,312 3,519 10 35 55 74 12 24 ­10 21.8 Russian Federation 763,720 6.2 1,746 2,297 6 38 56 50 16 21 14 16.8 Rwanda 2,131 4.9 183 229 42 20 38 88 13 21 ­22 5.9 Saudi Arabia 309,778 4.2 8,905 14,284 4 59 37 26 23 16 34 6.3 Senegal 8,318 4.9 236 235 17 20 63 76 13 23 ­14 2.0 Serbia and Montenegro 27,059 5.3 .. 1,446 16 32 52 88 18 17 ­22 25.3 Sierra Leone 1,193 13.7 .. .. 46 24 30 90 13 15 ­19 6.8 Singapore 116,764 4.2 28,729 32,267 0 34 66 41 11 19 30 0.5 Slovak Republic 46,412 4.9 .. .. 3 29 67 56 20 29 ­4 4.1 Slovenia 34,030 3.4 12,339 34,447 3 35 62 55 20 25 ­0 5.4 South Africa 240,152 3.7 1,764 2,463 3 31 66 59 20 18 4 6.6 Spain 1,123,691 3.1 12,611 19,132 3 29 67 58 18 28 ­4 4.2 Sri Lanka 23,479 4.2 713 743 17 26 57 77 9 26 ­12 8.7 Sudan 27,699 6.1 384 728 34 30 37 65 17 22 ­4 9.8 Sweden 354,115 2.2 21,654 31,716 2 29 69 48 28 16 8 1.6 Switzerland 365,937 0.9 21,565 22,190 .. .. .. 61 12 20 7 0.9 Syrian Arab Republic 26,320 4.0 2,356 2,977 21 26 53 65 13 20 1 4.7 Tajikistan 2,326 9.7 367 401 22 36 42 95 9 14 ­19 21.3 Tanzaniab 12,111 6.9 242 287 45 18 38 77 14 19 ­9 6.3 Thailand 176,602 5.4 481 599 10 47 44 61 10 31 ­2 2.3 Togo 2,203 2.7 360 409 42 23 35 86 10 18 ­13 1.2 Tunisia 28,683 4.5 2,365 2,415 13 28 59 63 14 25 ­3 2.3 Turkey 363,300 5.2 1,772 1,793 12 24 65 69 13 25 ­7 25.5 Turkmenistan 6,774 .. 1,179 .. 21 45 34 52 14 25 9 .. Uganda 8,712 5.4 192 231 34 21 46 77 14 23 ­13 5.1 Ukraine 81,664 8.0 1,235 1,442 11 34 55 55 18 19 8 10.7 United Kingdom 2,192,553 2.3 23,089 26,897 1 26 73 65 21 17 ­3 2.7 United States 12,455,068 2.8 22,868 36,863 1 22 77 71 16 18 ­5 2.2 Uruguay 16,792 1.0 6,213 7,102 11 29 60 74 11 13 2 11.4 Uzbekistan 13,667 5.3 1,263 1,567 28 29 43 51 16 25 8 29.0 Venezuela, RB 138,857 1.3 4,781 5,899 5 52 44 50 13 21 16 28.6 Vietnam 52,408 7.5 225 294 22 40 38 65 6 36 ­7 5.9 West Bank and Gaza 3,454 ­13.3 .. .. 6 12 82 84 53 3 ­39 10.9 Yemen, Rep. 14,452 5.9 383 511 13 35 52 80 13 17 ­10 6.9 Zambia 7,257 4.7 160 206 19 25 56 70 13 26 ­9 20.4 Zimbabwe 3,364 ­6.1 238 242 22 28 50 68 29 38 ­35 223.4 World 44,384,871t 2.8w 772w 863w 4w 28w 68w 62w 17w 21w 0w Low income 1,391,362 6.0 327 364 22 28 50 65 11 27 ­2 Middle income 8,535,129 5.1 581 726 10 37 53 58 14 26 2 Lower middle income 4,869,491 6.3 451 587 13 41 46 56 13 29 2 Upper middle income 3,665,404 3.5 2,279 2,733 7 32 62 63 14 22 1 Low & middle income 9,926,393 5.3 477 567 12 36 52 59 13 26 2 East Asia & Pacific 3,032,573 8.3 .. .. 13 45 42 52 10 34 3 Europe & Central Asia 2,190,933 5.4 1,652 1,971 8 32 60 61 16 23 1 Latin America & Caribbean 2,455,621 2.3 2,233 2,831 8 32 60 64 13 20 3 Middle East & North Africa 632,570 4.1 1,589 1,978 11 41 48 60 13 26 1 South Asia 995,809 6.4 357 394 19 27 54 64 10 28 ­2 Sub-Saharan Africa 615,216 4.2 293 334 17 32 51 63 17 20 0 High income 34,466,198 2.2 .. .. 2 26 72 62 18 20 0 a. Data on general government final consumption expenditure are not available separately; they are included in household final consumption expenditure. b. Data refer to mainland Tanzania only. 295 Table 5. Trade, aid, and finance Merchandise trade External debt Official Domestic Current Foreign development credit provided Manufactured High technology account direct assistance or Present by banking exports imports exports exports balance investment official aida Total value sector % of total % of merchandise manufactured % of Net migration $ millions $ millions exports exports $ millions $ millions $ per capita $ millions GNI % of GDP thousands 2005 2005 2004 2004 2005 2004 2004 2004 2004 2005 2000­05 Albania 654 2,650 82 1 ­572 426 117 1,549 17 10 ­100 Algeria 44,390 20,040 2 1 .. 882 10 21,987 32 11.0 ­100 Angola 23,120 8,150 .. .. 686 1,444 74 9,521 69 5.4 145 Argentina 40,044 28,692 29 8 3,281 4,084 2 169,247 159 45.4 ­100 Armenia 950 1,768 62 1 ­204 219 84 1,224 50 7.2 ­100 Australia 105,825 125,280 25 14 ­42,084 42,469 .. .. 102.4 500 Austria 123,317 124,749 84 12 3,848 4,022 .. .. 105.6 100 Azerbaijan 4,346 4,202 11 2 167 3,556 21 1,986 23 9.7 ­100 Bangladesh 9,190 13,868 90 0 ­279 449 10 20,344 26 30.1 ­350 Belarus 15,992 16,699 60 3 469 169 5 3,717 20 13.9 ­10 Belgium 329,650 320,363 81 8b 6,563 40,080 .. .. 73.1 67 Benin 620 960 9 2 ­331 60 46 1,916 24c 14.6 99 Bolivia 2,671 2,200 14 9 285 117 85 6,096 38c 52.5 ­100 Bosnia and Herzegovina 2,440 7,199 .. .. ­2,087 613 172 3,202 34 43.6 40 Brazil 118,308 77,576 54 12 14,199 18,166 2 222,026 47 81.0 ­130 Bulgaria 11,725 18,181 62 4 ­3,133 2,005 80 15,661 83 36.8 ­50 Burkina Faso 440 1,230 8 10 .. 35 48 1,967 23c 14.9 100 Burundi 110 280 5 6 ­25 3 48 1,385 15 40.5 192 Cambodia 3,100 3,700 97 0 ­217 131 35 3,377 68 9.3 ­10 Cameroon 2,500 2,450 5 1 .. 0 48 9,496 20c 9.1 13 Canada 359,578 320,105 60 14 25,268 6,284 .. .. 97.0 1,050 Central African Republic 140 165 37 0 .. ­13 26 1,078 75 7.2 ­45 Chad 3,230 850 .. .. .. 478 34 1,701 33c 3.2 271 Chile 39,536 32,542 13 5 703 7,603 3 44,058 57 62.5 30 China 761,999 660,118 91 30 68,659 54,937 1 248,934 15 120.1 ­1,950 Hong Kong, China 292,328d 300,635 97d 32 19,706 34,035 1 .. .. 147.6 300 Colombia 21,187 21,204 38 6 ­1,930 3,052 11 37,732 49 34.4 ­200 Congo, Dem. Rep. 2,190 2,270 .. .. .. 0 33 11,841 36 1.5 ­322 Congo, Rep. 5,000 1,980 .. .. ­3 0 30 5,829 331 3.2 ­14 Costa Rica 7,039 9,798 63 37 ­832 620 3 5,700 36 32.3 84 Côte d'Ivoire 7,180 4,690 20 8 303 175 9 11,739 90 14.4 ­371 Croatia 8,809 18,547 73 13 ­2,541 1,243 27 31,548 110 57.5 100 Czech Republic 78,474 76,863 90 13 ­5,595 4,454 27 45,561 51 33.2 50 Denmark 85,708 76,539 66 20 5,941 ­8,804 .. .. 160.3 61 Dominican Republic 5,854 9,210 .. .. 1,399 645 10 6,965 39 36.7 ­140 Ecuador 9,821 9,609 9 7 ­157 1,160 12 16,868 70 22.0 ­250 Egypt, Arab Rep. 10,344 16,552 31 1 3,922 1,253 20 30,292 32 110.8 ­450 El Salvador 3,383 6,712 60 4 ­612 466 31 7,250 54 41.9 ­38 Eritrea 9 495 .. .. .. 30 61 681 53 32.7 280 Ethiopia 860 4,160 11 0 ­668 545 26 6,574 30c 48.9 ­150 Finland 65,998 58,737 83 21 9,698 3,075 .. .. 68.7 41 France 459,246 495,796 83 19 ­38,781 24,521 .. .. 90.8 300 Georgia 867 2,491 37 38 ­689 499 70 2,082 37 9.8 ­248 Germany 970,688 774,069 84 17 115,519 ­34,903 .. .. 112.3 1,100 Ghana 2,520 5,090 14 4 ­236 139 63 7,035 32c 13.1 12 Greece 17,192 54,031 59 11 ­17,879 1,355 .. .. 78.6 179 Guatemala 3,477 8,810 42 7 ­1,188 155 18 5,532 23 15.2 ­300 Guinea 910 845 25 0 ­162 100 30 3,538 45 3.7 ­299 Haiti 473 1,471 .. .. ­13 7 29 1,225 29 14.9 ­105 Honduras 1,694 4,484 27 2 ­413 293 91 6,332 38 37.4 ­30 Hungary 62,194 65,711 88 29 ­7,962 4,608 30 63,159 76 46.4 50 India 89,843 131,648 73 5 6,853 5,335 1 122,723 18 36.9 ­1,400 Indonesia 86,285 68,736 56 16 3,108 1,023 0 140,649 61 24.0 ­1,000 Iran, Islamic Rep. 58,400 41,561 9 2 .. 500 3 13,622 9 38.6 ­1,379 Ireland 109,525 66,356 86 34 ­3,946 11,040 .. .. 136.9 194 Israel 42,588 46,910 94 19 2,385 1,664 70 .. .. 92.2 158 Italy 366,797 379,696 88 8 ­26,814 16,772 .. .. 87.7 600 Jamaica 1,487 4,560 65 0 ­509 602 29 6,399 89 28.3 ­100 Japan 595,750 516,075 93 24 165,783 7,805 .. .. 99.5 270 Jordan 4,284 10,455 72 5 ­18 620 110 8,175 73 74.0 100 Kazakhstan 27,849 17,353 16 2 ­486 4,104 18 32,310 101 26.7 ­600 Kenya 3,450 6,360 21 3 ­379 46 19 6,826 34 40.2 ­212 Korea, Rep. 284,742 261,028 92 33 27,613 8,189 ­1 .. .. 98.2 ­80 Kuwait 44,016 17,422 .. .. 18,884 ­20 1 .. .. 71.6 240 Kyrgyz Republic 672 1,108 43 2 ­75 77 51 2,100 82 7.1 ­75 Lao PDR 435 605 .. .. .. 17 47 2,056 76 6.3 ­7 Latvia 5,122 8,625 61 5 ­1,959 699 71 12,661 110 44.3 ­12 Lebanon 1,880 9,340 68 2 ­4,805 288 75 22,177 121 75.6 ­35 Lithuania 11,815 15,453 58 5 ­1,771 773 73 9,475 54 25.7 ­20 Macedonia, FYR 2,041 3,228 77 1 ­415 157 122 2,044 39 23.2 ­10 Madagascar 750 1,550 23 1 ­309 45 68 3,462 38c 10.0 0 Malawi 460 1,035 16 2 .. 16 38 3,418 60c 22.1 ­20 Malaysia 140,948 114,607 76 55 14,872 4,624 12 52,145 53 133.7 150 Mali 1,150 1,500 .. .. ­271 180 43 3,316 33c 20.0 ­134 Mauritania 410 740 .. .. .. 300 60 2,297 57c 25.9 30 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 296 Table 5. Trade, aid, and finance (continued) Merchandise trade External debt Official Domestic Current Foreign development credit provided Manufactured High technology account direct assistance or Present by banking exports imports exports exports balance investment official aida Total value sector % of total % of merchandise manufactured % of Net migration $ millions $ millions exports exports $ millions $ millions $ per capita $ millions GNI % of GDP thousands 2005 2005 2004 2004 2005 2004 2004 2004 2004 2005 2000­05 Mexico 213,711 231,670 80 21 ­5,708 17,377 1 138,689 24 34.6 ­2,000 Moldova 1,091 2,312 36 4 ­286 81 28 1,868 75 21.3 ­40 Mongolia 1,045 1,145 38 0 63 93 104 1,517 86 32.0 ­50 Morocco 10,463 20,124 69 10 970 769 24 17,672 39 56.7 ­400 Mozambique 1,790 2,420 3 9 ­607 245 63 4,651 17c 2.1 ­20 Namibia 1,990 2,450 41 3 634 .. 89 .. .. 53.2 ­6 Nepal 830 1,860 74 0 197 0 16 3,354 37 .. ­100 Netherlands 401,333 357,869 70 29 40,187 377 .. .. 166.3 150 New Zealand 21,731 26,224 31 14 ­6,456 2,271 .. .. 121.1 79 Nicaragua 858 2,595 11 6 ­772 250 229 5,145 35 85.6 ­100 Niger 410 920 8 3 ­219 0 40 1,950 26c 6.3 ­10 Nigeria 43,500 15,200 2 2 12,264 1,875 5 35,890 71 15.6 ­170 Norway 103,256 54,907 19 18 49,488 502 .. .. 11.1 58 Oman 17,119 9,000 12 1 443 ­17 22 3,872 18 34.9 ­160 Pakistan 15,942 25,335 85 1 ­817 1,118 9 35,687 35 29.3 ­1,810 Panama 1,080 4,180 10 2 ­818 1,012 12 9,469 94 88.2 8 Papua New Guinea 3,070 1,710 6 39 .. 25 46 2,149 66 10.1 0 Paraguay 1,495 2,880 13 7 20 93 0 3,433 52 17.9 ­25 Peru 17,206 12,502 20 2 1,030 1,816 18 31,296 57 17.4 ­300 Philippines 41,224 46,257 55 64 2,080 469 6 60,550 73 34.3 ­900 Poland 88,940 100,487 81 3 ­4,364 12,613 40 99,190 45 26.6 ­80 Portugal 37,858 60,175 85 9 ­17,007 825 .. .. 150.3 250 Romania 27,730 40,463 82 3 ­6,382 5,440 42 30,034 51 9.7 ­150 Russian Federation 245,255 125,123 21 9 84,249 12,479 9 197,335 46 24.2 400 Rwanda 120 410 10 25 ­6 8 53 1,656 15c 10.7 45 Saudi Arabia 178,755 56,092 12 2 87,132 .. 1 .. .. 56.0 250 Senegal 1,600 3,330 39 7 ­437 70 92 3,938 22c 21.2 ­100 Serbia and Montenegro 5,142 11,558 57 .. .. 966 144 15,882 77 .. ­100 Sierra Leone 150 350 7 31 ­74 26 67 1,723 37c 4.7 438 Singapore 229,620d 200,030 84d 59 27,897 16,032 2 .. .. 106.2 200 Slovak Republic 31,973 35,301 86 5 ­282 1,122 44 22,068 67 31.2 5 Slovenia 18,698 20,141 90 6 ­362 827 31 .. .. 45.9 10 South Africa 51,874 66,500 58e 6 ­10,079 585 14 28,500 17 156.4 50 Spain 186,099 277,597 77 7 ­83,136 16,594 .. .. 125.4 2,025 Sri Lanka 6,275 8,985 74 2 ­648 233 27 10,887 50 31.5 ­160 Sudan 5,150 6,100 2 0 ­2,768 1,511 25 19,332 151 7.6 ­519 Sweden 129,922 110,645 81 17 27,485 ­588 .. .. 105.9 157 Switzerland 125,898 121,156 93 22 49,710 ­797 .. .. 161.2 40 Syrian Arab Republic 6,001 7,754 11 1 210 275 6 21,521 101 11.8 ­30 Tajikistan 909 1,330 .. .. ­19 272 38 896 41 17.4 ­345 Tanzania 1,482 2,659 20 2 ­437 249 46 7,800 22c,f 8.6 ­345 Thailand 110,110 118,191 75 30 ­3,719 1,412 ­0 51,307 35 103.7 ­50 Togo 755 1,050 47 0 ­162 60 10 1,812 83 16.0 ­4 Tunisia 10,494 13,177 78 5 ­555 593 33 18,700 79 71.1 ­20 Turkey 73,275 116,352 85 2 ­15,543 2,733 4 161,595 70 54.6 ­250 Turkmenistan 4,935 3,588 .. .. .. .. 8 .. .. .. ­10 Uganda 870 1,810 15 13 ­267 222 42 4,822 33c 6.8 ­15 Ukraine 34,287 36,141 67 5 2,531 1,715 8 21,652 42 25.0 ­700 United Kingdom 377,856 501,223 77 24 ­57,616 72,561 .. .. 156.3 686 United States 904,289 1,732,706 82 32 ­804,961 106,831 .. .. 269.4 5,800 Uruguay 3,422 3,425 32 2 ­103 311 6 12,376 108 30.4 ­10 Uzbekistan 4,706 3,640 .. .. .. 140 9 5,007 46 .. ­300 Venezuela, RB 56,200 24,933 12 3 25,359 1,518 2 35,570 45 10.8 40 Vietnam 32,233 36,881 53 6 ­926 1,610 22 17,825 39 58.9 ­200 West Bank and Gaza .. .. .. .. .. .. 324 .. .. .. ­40 Yemen, Rep. 4,883 4,328 3 13 1,215 144 12 5,488 37 7.7 ­100 Zambia 1,720 2,750 10 1 .. 334 94 7,279 36 8.0 ­65 Zimbabwe 1,490 2,220 29 1 .. 60 14 4,798 33 49.6 ­50 World 10,392,567t 10,652,542t 77w 20w 664,877s 14w ..s 145.5w ..wg Low income 256,379 310,841 51 4 16,576 15 426,945 32.0 ­4,000 Middle income 2,785,199 2,551,288 64 20 194,808 11 2,328,780 68.4 ­11,987 Lower middle income 1,512,592 1,375,639 68 23 106,037 10 1,140,272 84.8 ­10,086 Upper middle income 1,272,607 1,175,649 61 16 88,771 12 1,188,508 47.0 ­1,901 Low & middle income 3,041,588 2,862,091 64 19 211,385 16 2,755,725 63.3 ­15,987 East Asia & Pacific 1,185,932 1,059,945 80 34 64,563 4 588,888 105.7 ­3,939 Europe & Central Asia 759,841 746,370 57 9 62,212 25 794,943 32.0 ­2,665 Latin America & Carib. 561,873 517,073 56 13 60,843 13 778,970 48.9 ­4,012 Middle East & N. Africa 221,252 182,440 20 3 5,340 35 163,935 47.3 ­2,374 South Asia 123,050 186,039 76 4 7,151 5 193,933 35.5 ­1,680 Sub-Saharan Africa 189,636 170,236 31 4 11,276 36 235,056 74.4 ­1,318 High income 7,351,037 7,790,420 81 20 453,492 .. 166.6 15,970 a. Regional aggregates include data for economies that are not specified elsewhere. World and income group totals include aid not allocated by country or region. b. Includes Luxembourg. c. Data are from debt sustainability analysis undertaken as part of the Heavily Indebted Poor Countries (HIPC) initiative. d. Includes re-exports. e. Data on total exports and imports refer to South Africa only. Data on export commodity shares refer to the South African Customs Union (Botswana, Lesotho, Namibia, and South Africa). f. GNI refers to mainland Tanzania only. g. World total computed by the UN sums to zero, but because the aggregates shown here refer to World Bank definitions, regional and income group totals do not equal zero. 297 Table 6. Key indicators for other economies Population Gross Life Carbon age Gross national income PPP gross national domestic expectancy Adult Literacy dioxide Population composition (GNI)a income (GNI)b product at birth rate emissions Avg. density annual % people per % Millions per capita Millions of per capita per capita % Male Female % ages per capita Thousands growth sq. km Ages 0­14 of dollars dollars dollars dollars growth Years Years 15 and older metric tons 2005 2000­2005 2005 2005 2005b 2005 2005 2005 2004­2005 2004 2004 2000­2004 2002 Afghanistan .. .. .. .. 6,957 ..d .. .. .. .. .. 28 .. American Samoa 58 1.1c 292 .. .. ..e .. .. .. .. .. .. .. Andorra 66 0.2c 141 .. .. ..f .. .. .. .. .. .. .. Antigua and Barbuda 81 1.0 184 .. 885 10,920 948 11,700 2.6 .. .. .. 4.7 Aruba 101 0.8c 529 .. .. ..f .. .. .. .. .. 97 .. Bahamas, The 323 1.4 32 28 .. ..f .. .. .. 67 74 .. 6.7 Bahrain 727 1.6 1,023 27 10,288 14,370 15,470 21,290 5.3 73 76 87 30.6 Barbados 270 0.3 627 19 .. ..e .. .. .. 73 78 .. 4.6 Belize 292 3.1 13 37 1,021 3,500 1,967 6,740 ­0.2 69 74 .. 3.0 Bermuda 65 0.2 1,293 .. .. ..f .. .. .. .. .. .. 7.7 Bhutan 918 2.6 20 38 799 870 .. .. 3.3 62 65 .. 0.5 Botswana 1,765 0.1 3 38 9,145 5,180 18,090 10,250 4.0 36 35 81 2.3 Brunei 374 2.3c 71 30 .. ..f .. .. .. 75 79 93 17.7 Cape Verde 507 2.4 126 40 947 1,870 3,041g 6,000g 3.0 67 74 .. 0.3 Cayman Islands 45 2.3c 173 .. .. ..f .. .. .. .. .. .. .. Channel Islands 149 0.4 745 16 .. ..f .. .. .. 76 83 .. .. Comoros 600 2.1 269 42 387 640 1,201g 2,000g 0.7 61 65 .. 0.1 Cuba 11,269 0.3 103 19 .. ..h .. .. .. 75 79 100 2.1 Cyprus 835 1.2 90 20 13,633 16,510 18,360g 22,230g .. 77 81 97 8.3 Djibouti 793 2.1 34 42 807 1,020 1,776g 2,240g 1.4 52 54 .. 0.5 Dominica 72 0.2c 96 .. 273 3,790 400 5,560 2.3 .. .. .. 1.7 Equatorial Guinea 504 2.3 18 44 .. ..e 3,731g 7,580g .. 42 43 87 0.4 Estonia 1,345 ­0.4 32 15 12,244 9,100 20,740 15,420 10.1 66 77 100 11.7 Faeroe Islands 48 0.2c 34 .. .. ..f .. .. .. .. .. .. .. Fiji 848 0.9 46 32 2,784 3,280 5,052 5,960 0.9 66 70 .. 1.6 French Polynesia 257 1.7 70 28 .. ..f .. .. .. 71 76 .. 2.9 Gabon 1,384 1.7 5 40 6,930 5,010 8,151 5,890 0.6 54 55 .. 2.6 Gambia, The 1,517 2.8 152 40 442 290 2,913g 1,920g 2.2 55 58 .. 0.2 Greenland 57 0.3 0 .. .. ..f .. .. .. .. .. .. 10.0 Grenada 107 1.0 313 .. 418 3,920 773 7,260 0.2 .. .. .. 2.2 Guam 170 1.8 308 30 .. ..f .. .. .. 73 77 .. 25.4 Guinea-Bissau 1,586 3.0 56 48 283 180 1,110 700 0.5 44 46 .. 0.2 Guyana 751 0.2 4 29 759 1,010 3,178g 4,230g ­2.9 61 67 .. 2.2 Iceland 295 1.0 3 22 13,671 46,320 10,258 34,760 4.5 78 82 .. 7.7 Iraq .. .. .. .. .. ..h .. .. .. .. .. 74 .. Isle of Man 77 0.3 135 .. 2,138 27,770 .. .. .. .. .. .. .. Kiribati 99 1.8 136 .. 137 1,390 .. .. ­0.9 .. .. .. 0.3 Korea, Dem. Rep. 22,488 0.6 187 25 .. ..d .. .. .. 61 67 .. 6.5 Lesotho 1,795 0.1 59 39 1,718 960 6,120 3,410 1.4 35 37 82 .. Liberia 3,283 1.4 34 47 436 130 .. .. 3.9 42 43 .. 0.1 Libya 5,854 2.0 3 30 32,354 5,530 .. .. 1.5 72 77 .. 9.1 Liechtenstein 34 0.6c 215 .. .. ..f .. .. .. .. .. .. .. Luxembourg 457 0.8 174 19 29,976 65,630 29,841 65,340 3.2 75 81 .. 21.3 Macao, China 460 0.7 .. 16 .. ..f .. .. .. 78 82 91 4.0 Maldives 329 2.5 1,097 41 787 2,390 .. .. ­6.0 68 67 96 3.4 Malta 404 0.7 1,263 18 5,491 13,590 7,662 18,960 1.8 77 81 88 7.4 Marshall Islands 63 3.6 352 .. 185 2,930 .. .. 0.2 .. .. .. .. Mauritius 1,248 1.0 615 25 6,560 5,260 15,538 12,450 3.4 69 76 84 2.6 Mayotte 180 4.0c 430 .. .. ..e .. .. .. .. .. .. .. Micronesia, Fed. Sts. 111 0.6 158 39 254 2,300 .. .. ­0.4 67 69 .. .. Monaco 33 0.6c 159 .. .. ..f .. .. .. .. .. .. .. Myanmar 50,519 1.1 77 29 .. ..d .. .. .. 58 64 90 0.2 Northern Mariana Islands 79 1.5c 161 .. .. ..e .. .. .. .. .. .. .. Netherlands Antilles 183 0.8 228 23 .. ..f .. .. .. 73 79 .. 27.8 New Caledonia 234 1.9 13 28 .. ..f .. .. .. 72 78 96 8.2 Palau 20 1.3c 43 .. 154 7,630 .. .. 4.5 .. .. .. .. Puerto Rico 3,911 0.5 441 22 .. ..f .. .. .. 74 82 .. 3.5 Qatar 813 6.2 74 22 .. ..f .. .. .. 72 76 89 53.0 Samoa 185 0.8 65 41 387 2,090 1,199g 6,480g 4.8 67 73 .. 0.8 San Marino 28 0.7c 473 .. .. ..f .. .. .. .. .. .. .. Sao Tome and Principe 157 2.3 163 39 60 390 .. .. 0.7 62 64 .. 0.6 Seychelles 84 0.8 184 .. 701 8,290 1,347g 15,940g ­3.3 .. .. 92 6.4 Solomon Islands 478 2.6 17 41 282 590 898 1,880 1.8 62 63 .. 0.4 Somalia 8,228 3.2 13 44 .. ..d .. .. .. 46 48 .. .. St. Kitts and Nevis 48 1.6 133 .. 394 8,210 600 12,500 2.7 .. .. .. 2.4 St. Lucia 166 1.2 271 29 794 4,800 990 5,980 3.9 72 75 .. 2.4 St. Vincent and the Grenadines 119 0.5 305 29 427 3,590 769 6,460 4.4 69 74 .. 1.6 Suriname 449 0.7 3 30 1,140 2,540 .. .. 4.5 66 73 90 5.1 Swaziland 1,131 1.6 66 41 2,579 2,280 5,870 5,190 0.8 43 42 80 0.9 Timor-Leste 976 4.4 66 41 729 750 .. .. ­3.5 .. .. .. .. Tonga 102 0.4 142 36 224 2,190 823g 8,040g 2.0 71 74 99 1.0 Trinidad and Tobago 1,305 0.3 254 22 13,632 10,440 17,190 13,170 6.4 67 73 .. 31.8 United Arab Emirates 4,533 6.7 54 22 102,693 23,770 104,069 24,090 .. 77 81 .. 25.0 Vanuatu 211 2.0 17 40 338 1,600 670g 3,170g 4.8 67 71 74 0.4 Virgin Islands (U.S.) 115 1.1 329 24 .. ..f .. .. .. 76 81 .. 92.8 a. Calculated using the World Bank Atlas method. b. PPP is purchasing power parity; see Definitions. c. Data are for 2003­2005. d. Estimated to be low income ($875 or less). e. Estimated to be upper middle ($3,466 ­$10,725). f. Estimated to be high income ($10,726 or more). g. The estimate is based on regression; others are extrapolated from the latest International Comparison Program benchmark estimates. h. Estimated to be lower middle income ($876­3,465). 298 Selected world development indicators 299 Technical notes using the least-squares regression method (see the Statistical meth- These technical notes discuss the sources and methods used to ods section below). Because this method takes into account all avail- compile the indicators included in this edition of Selected World able observations during a period, the resulting growth rates reflect Development Indicators. The notes follow the order in which general trends that are not unduly influenced by exceptional values. the indicators appear in the tables. The Selected World Develop- To exclude the effects of inflation, constant price economic indica- ment Indicators uses terminology in line with the 1993 System of tors are used in calculating growth rates. Data in italics are for a National Accounts (SNA). year or period other than that specified in the column heading--up to two years before or after for economic indicators and up to three Sources years for social indicators, because the latter tend to be collected less regularly and change less dramatically over short periods. The data published in the Selected World Development Indicators are taken from World Development Indicators 2006. Where possible, Constant price series however, revisions reported since the closing date of that edition have been incorporated. In addition, newly released estimates of An economy's growth is measured by the increase in value added population and gross national income (GNI) per capita for 2005 produced by the individuals and enterprises operating in that econ- are included in table 1 and table 6. omy. Thus, measuring real growth requires estimates of GDP and The World Bank draws on a variety of sources for the statistics its components valued in constant prices. The World Bank collects published in World Development Indicators. Data on external debt constant price national accounts series in national currencies and for developing countries are reported directly to the World Bank by recorded in the country's original base year. To obtain comparable developing member countries through the Debtor Reporting Sys- series of constant price data, it rescales GDP and value added by tem. Other data are drawn mainly from the United Nations and industrial origin to a common reference year, 2000 in the current its specialized agencies, from the IMF, and from country reports version of World Development Indicators. This process gives rise to a to the World Bank. Bank staff estimates are also used to improve discrepancy between the rescaled GDP and the sum of the rescaled currentness or consistency. For most countries, national accounts components. Because allocating the discrepancy would give rise to estimates are obtained from member governments through World distortions in the growth rate, it is left unallocated. Bank economic missions. In some instances these are adjusted by staff to ensure conformity with international definitions and con- Summary measures cepts. Most social data from national sources are drawn from regu- The summary measures for regions and income groups, presented lar administrative files, special surveys, or periodic censuses. at the end of most tables, are calculated by simple addition when For more detailed notes about the data, please refer to the World they are expressed in levels. Aggregate growth rates and ratios are Bank's World Development Indicators 2006. usually computed as weighted averages. The summary measures for social indicators are weighted by population or subgroups of Data consistency and reliability population, except for infant mortality, which is weighted by the Considerable effort has been made to standardize the data, but full number of births. See the notes on specific indicators for more comparability cannot be assured, and care must be taken in interpret- information. ing the indicators. Many factors affect data availability, comparability, For summary measures that cover many years, calculations are and reliability: statistical systems in many developing economies are based on a uniform group of economies so that the composition of still weak; statistical methods, coverage, practices, and definitions dif- the aggregate does not change over time. Group measures are com- fer widely; and cross-country and intertemporal comparisons involve piled only if the data available for a given year account for at least complex technical and conceptual problems that cannot be unequiv- two-thirds of the full group, as defined for the 2000 benchmark year. ocally resolved. Data coverage may not be complete because of spe- As long as this criterion is met, economies for which data are miss- cial circumstances or for economies experiencing problems (such as ing are assumed to behave like those that provide estimates. Readers those stemming from conflicts) affecting the collection and report- should keep in mind that the summary measures are estimates of ing of data. For these reasons, although the data are drawn from the representative aggregates for each topic and that nothing meaning- sources thought to be most authoritative, they should be construed ful can be deduced about behavior at the country level by working only as indicating trends and characterizing major differences among back from group indicators. In addition, the estimation process may economies rather than offering precise quantitative measures of result in discrepancies between subgroup and overall totals. those differences. Discrepancies in data presented in different edi- tions reflect updates by countries as well as revisions to historical Table 1. Key indicators of development series and changes in methodology. Thus readers are advised not to Population is based on the de facto definition, which counts all res- compare data series between editions or between different editions of idents, regardless of legal status or citizenship, except for refugees World Bank publications. Consistent time series are available from not permanently settled in the country of asylum, who are gener- World Development Indicators 2006 CD-ROM and in WDI Online. ally considered part of the population of the country of origin. Average annual population growth rate is the exponential rate Ratios and growth rates of change for the period (see the Statistical methods section below). For ease of reference, the tables usually show ratios and rates of Population density is midyear population divided by land area. growth rather than the simple underlying values. Values in their Land area is a country's total area excluding areas under inland original form are available from World Development Indicators 2006 bodies of water and coastal waterways. Density is calculated using CD-ROM. Unless otherwise noted, growth rates are computed the most recently available data on land area. 300 WORLD DEVELOPMENT REPORT 2007 Population age composition,ages 0­14 refers to the percentage The Carbon Dioxide Information Analysis Center (CDIAC), of the total population that is ages 0­14. sponsored by the U.S. Department of Energy, calculates annual Gross national income (GNI) is the broadest measure of anthropogenic emissions of CO2. These calculations are derived national income, measures total value added from domestic and from data on fossil fuel consumption, based on the World Energy foreign sources claimed by residents. GNI comprises gross domestic Data Set maintained by the UNSD, and from data on world cement product (GDP) plus net receipts of primary income from foreign manufacturing, based on the Cement Manufacturing Data Set main- sources. Data are converted from national currency to current U.S. tained by the U.S. Bureau of Mines. Each year the CDIAC recalculates dollars using the World Bank Atlas method. This involves using a the entire time series from 1950 to the present, incorporating its most three-year average of exchange rates to smooth the effects of transi- recent findings and the latest corrections to its database. Estimates tory exchange rate fluctuations. (See the section on statistical meth- exclude fuels supplied to ships and aircraft engaged in international ods below for further discussion of the Atlas method.) transportation because of the difficulty of apportioning these fuels GNI per capita is GNI divided by midyear population. It is con- among the countries benefiting from that transport. verted into current U.S. dollars by the Atlas method. The World Bank uses GNI per capita in U.S dollars to classify economies for Table 2. Poverty analytical purposes and to determine borrowing eligibility. The World Bank produced its first global poverty estimates for devel- PPP gross national income, which is GNI converted into inter- oping countries for the World Development Report 1990 using house- national dollars using purchasing power parity (PPP) conversion hold survey data for 22 countries (Ravallion, Datt, and van de Walle factors, is included because nominal exchange rates do not always 1991). Incorporating survey data collected during the last 15 years, reflect international differences in relative prices. At the PPP rate, the database has expanded considerably and now includes 440 sur- one international dollar has the same purchasing power over veys representing almost 100 developing countries. Some 1.1 million domestic GNI that the U.S. dollar has over U.S. GNI. PPP rates randomly sampled households were interviewed in these surveys, allow a standard comparison of real price levels between countries, representing 93 percent of the population of developing countries. just as conventional price indexes allow comparison of real values The surveys asked detailed questions on sources of income and how over time. The PPP conversion factors used here are derived from it was spent and on other household characteristics such as the num- price surveys covering 118 countries conducted by the Interna- ber of people sharing that income. Most interviews were conducted tional Comparison Program. Data for OECD countries data come by staff of government statistics offices. Along with improvements from the most recent round of surveys, completed in 1999; data in data coverage and quality, the underlying methodology has also for other countries are from either the 1996 survey or the 1993 or improved, resulting in better and more comprehensive estimates. earlier round (extrapolated to the 1996 benchmark). Estimates for countries not included in the surveys are derived from statistical Data availability models using available data. Since 1979 there has been considerable expansion in the number PPP GNI per capita is PPP GNI divided by midyear population. of countries that field such surveys, the frequency of the surveys, Gross domestic product (GDP) per capita growth is based on and the quality of their data. The number of data sets rose dramati- GDP measured in constant prices. Growth in GDP is considered cally from a mere 13 between 1979 and 1981 to 100 between 1997 a broad measure of the growth of an economy. GDP in constant and 1999. The drop to 41 available surveys after 1999 reflects the lag prices can be estimated by measuring the total quantity of goods between the time data are collected and the time they become avail- and services produced in a period, valuing them at an agreed set of able for analysis, not a reduction in data collection. Data coverage is base year prices, and subtracting the cost of intermediate inputs, improving in all regions, but Sub-Saharan Africa continues to lag, also in constant prices. See the section on statistical methods for with only 28 of 48 countries having at least one data set available. details of the least-squares growth rate. A complete overview of data availability by year and country can be Life expectancy at birth is the number of years a newborn obtained at http://iresearch.worldbank.org/povcalnet/. infant would live if patterns of mortality prevailing at its birth were to stay the same throughout its life. Data are presented for males Data quality and females separately. The problems of estimating poverty and comparing poverty rates Adult literacy rate is the percentage of persons ages 15 and do not end with data availability. Several other issues, some related above who can, with understanding, read and write a short, simple to data quality, also arise in measuring household living standards statement about everyday life. In practice, literacy is difficult to from survey data. One relates to the choice of income or consump- measure. To estimate literacy using such a definition requires census tion as a welfare indicator. Income is generally more difficult to or survey measurements under controlled conditions. Many coun- measure accurately, and consumption comes closer to the notion of tries estimate the number of literate people from self-reported data. standard of living. And income can vary over time even if the stan- Some use educational attainment data as a proxy but apply differ- dard of living does not. But consumption data are not always avail- ent lengths of school attendance or level of completion. Because able. Another issue is that household surveys can differ widely, for definition and methodologies of data collection differ across coun- example, in the number of consumer goods they identify. And even tries, data need to be used with caution. similar surveys may not be strictly comparable because of differ- Carbon dioxide emissions (CO2) measures those emissions ences in timing or the quality and training of survey enumerators. stemming from the burning of fossil fuels and the manufacture of Comparisons of countries at different levels of development cement. These include carbon dioxide produced during consump- also pose a potential problem because of differences in the relative tion of solid, liquid, and gas fuels and from gas flaring. Carbon importance of consumption of nonmarket goods. The local mar- dioxide per capita is CO2 divided by the mid-year population. ket value of all consumption in kind (including own production, Selected world development indicators 301 particularly important in underdeveloped rural economies) should between urban and rural poverty lines found in practice reflects only be included in total consumption expenditure. Similarly, imputed differences in the cost of living. In some countries the urban poverty profit from the production of nonmarket goods should be included line in common use has a higher real value--meaning that it allows in income. This is not always done, though such omissions were a the purchase of more commodities for consumption--than does the far bigger problem in surveys before the 1980s. Most survey data rural poverty line. Sometimes the difference has been so large as to now include valuations for consumption or income from own pro- imply that the incidence of poverty is greater in urban than in rural duction. Nonetheless, valuation methods vary. For example, some areas, even though the reverse is found when adjustments are made surveys use the price in the nearest market, while others use the only for differences in the cost of living. As with international com- average farmgate selling price. parisons, when the real value of the poverty line varies it is not clear Whenever possible, the table uses consumption data for decid- how meaningful such urban-rural comparisons are. ing who is poor and income surveys only when consumption data By combining all this information, a team in the World Bank's are unavailable. In recent editions of World Development Indicators, Development Research Group calculates the number of people there has been a change in how income surveys are used. In the past, living below various international poverty lines, as well as other average household income was adjusted to accord with consumption poverty and inequality measures that are published in World Devel- and income data from national accounts. But in testing this approach opment Indicators. The database is updated annually as new sur- using data for some 20 countries for which income and consumption vey data become available, and a major reassessment of progress expenditure data were available from the same surveys, income was against poverty is made about every three years. found to yield not only a higher mean than consumption but also higher inequality. When poverty measures based on consumption Do it yourself: PovcalNet and income were compared, these two effects roughly cancelled each Recently, this research team developed PovcalNet, an interactive other out: statistically, there was no significant difference. Recent edi- Web-based computational tool that allows users to replicate the cal- tions of World Development Indicators use income data to estimate culations by the World Bank's researchers in estimating the extent of poverty directly, without adjusting average income measures. absolute poverty in the world. PovcalNet is self-contained and pow- ered by reliable built-in software that performs the relevant calcula- International poverty lines tions from a primary database. The underlying software can also be International comparisons of poverty estimates entail both concep- downloaded from the site and used with distributional data of vari- tual and practical problems. Countries have different definitions of ous formats. The PovcalNet primary database consists of distribu- poverty, and consistent comparisons across countries can be diffi- tional data calculated directly from household survey data. Detailed cult. Local poverty lines tend to have higher purchasing power in information for each of these is also available from the site. rich countries, where more generous standards are used, than in Estimation from distributional data requires an interpolation poor countries. Is it reasonable to treat two people with the same method. The method chosen was Lorenz curves with flexible func- standard of living--in terms of their command over commodities-- tional forms, which have proved reliable in past work. The Lorenz differently because one happens to live in a better-off country? curve can be graphed as the cumulative percentages of total con- Poverty measures based on an international poverty line attempt sumption or income against the cumulative number of people, to hold the real value of the poverty line constant across countries, starting with the poorest individual. The empirical Lorenz curves as is done when making comparisons over time. The commonly estimated by PovcalNet are weighted by household size, so they are used $1 a day standard, measured in 1985 international prices and based on percentiles of population, not households. adjusted to local currency using purchasing power parities (PPPs), PovcalNet also allows users to calculate poverty measures under was chosen for the World Bank's World Development Report 1990 different assumptions. For example, instead of $1 a day, users can because it is typical of the poverty lines in low-income countries. specify a different poverty line, say $1.50 or $3. Users can also spec- PPP exchange rates, such as those from the Penn World Tables ify different PPP rates and aggregate the estimates using alternative or the World Bank, are used because they take into account the country groupings (for example, UN country groupings or group- local prices of goods and services not traded internationally. But ings based on average incomes) or a selected set of individual coun- PPP rates were designed for comparing aggregates from national tries. PovcalNet is available online at http://iresearch.worldbank. accounts, not for making international poverty comparisons. As a org/povcalnet/. result, there is no certainty that an international poverty line mea- sures the same degree of need or deprivation across countries. Notes on the 2002 estimates Early editions of World Development Indicators used PPPs from Survey year is the year in which the underlying data were collected. the Penn World Tables. Recent editions use 1993 consumption PPP Rural poverty rate is the percentage of the rural population liv- estimates produced by the World Bank. Recalculated in 1993 PPP ing below the national rural poverty line. terms, the original international poverty line of $1 a day in 1985 Urban poverty rate is the percentage of the urban population PPP terms is now about $1.08 a day. Any revisions in the PPP of a living below the national urban poverty line. country to incorporate better price indexes can produce dramati- National poverty rate is the percentage of the population living cally different poverty lines in local currency. below the national poverty line. National estimates are based on Issues also arise when comparing poverty measures within coun- population-weighted subgroup estimates from household surveys. tries.For example,the cost of living is typically higher in urban than in Population below $1 a day and population below $2 a day are rural areas. One reason is that food staples tend to be more expensive the percentages of the population living on less than $1.08 a day and in urban areas. So the urban monetary poverty line should be higher $2.15 a day at 1993 international prices. As a result of revisions in than the rural poverty line. But it is not always clear that the difference PPP exchange rates, poverty rates for individual countries cannot 302 WORLD DEVELOPMENT REPORT 2007 be compared with poverty rates reported in earlier editions of World rates that make use of all available information in a transparent Development Indicators. way, a methodology that fits a regression line to the relationship Poverty gap is the mean shortfall from the poverty line (count- between mortality rates and their reference dates using weighted ing the nonpoor as having zero shortfall), expressed as a percentage least squares was developed and adopted by both UNICEF and the of the poverty line. This measure reflects the depth of poverty as World Bank. well as its incidence. Maternal mortality ratio is the number of women who die from pregnancy-related causes during pregnancy and childbirth Table 3. Millennium Development Goals: eradicating per 100,000 live births. The values are modeled estimates based on poverty and improving lives an exercise carried out by the WHO and UNICEF. In this exercise, Percentage share of the poorest quintile in national consumption maternal mortality was estimated with a regression model using or income is the share of total consumption or income that accrues information on fertility, birth attendants, and HIV prevalence. This to the lowest 20 percent of the population. cannot be assumed to provide an accurate estimate of maternal Prevalence of child malnutrition is the percentage of children mortality in any country in the table. under five whose weight for age is less than minus two standard Births attended by skilled health staff are the percentage of deviations from the median for the international reference popula- deliveries attended by personnel trained to give the necessary super- tion ages 0­59 months. The reference population, adopted by the vision, care, and advice to women during pregnancy, labor, and the World Health Organization in 1983, is based on children from the postpartum period; to conduct deliveries on their own; and to care United States, who are assumed to be well nourished. Estimates of for newborns. The share of births attended by skilled health staff child malnutrition are from national survey data. The proportion is an indicator of a health system's ability to provide adequate care of children who are underweight is the most common indicator of for pregnant women. Good antenatal and postnatal care improves malnutrition. Being underweight, even mildly, increases the risk of maternal health and reduces maternal and infant mortality. But data death and inhibits cognitive development in children. Moreover, it may not reflect such improvements because health information sys- perpetuates the problem from one generation to the next, as mal- tems are often weak, material deaths are underreported, and rates of nourished women are more likely to have low-birth-weight babies. maternal mortality are difficult to measure. Primary completion rate is the percentage of students com- Prevalence of HIV is the percentage of people ages 15­49 who pleting the last year of primary school. It is calculated by taking the are infected with HIV. Adult HIV prevalence rates reflect the rate of total number of students in the last grade of primary school, minus HIV infection in each country's population. Low national preva- the number of repeaters in that grade, divided by the total number lence rates can be very misleading, however. They often disguise of children of official graduation age. The primary completion rate serious epidemics that are initially concentrated in certain localities reflects the primary cycle as defined by the International Standard or among specific population groups and threaten to spill over into Classification of Education (ISCED), ranging from three or four the wider population. In many parts of the developing world most years of primary education (in a very small number of countries) to new infections occur in young adults, with young women especially five or six years (in most countries) and seven (in a small number vulnerable. The estimates of HIV prevalence are based on extrapo- of countries). Because curricula and standards for school comple- lations from data collected through surveys and from surveillance tion vary across countries, a high rate of primary completion does of small, nonrepresentative groups. not necessarily mean high levels of student learning. Gender parity ratio in primary and secondary school is the Table 4. Economic activity ratio of the female gross enrollment rate in primary and secondary Gross domestic product is gross value added, at purchasers' prices, school to the male gross enrollment rate. by all resident producers in the economy plus any taxes and minus Eliminating gender disparities in education would help to any subsidies not included in the value of the products. It is calcu- increase the status and capabilities of women. This indicator is an lated without deducting for depreciation of fabricated assets or for imperfect measure of the relative accessibility of schooling for girls. depletion or degradation of natural resources. Value added is the net With a target date of 2005, this is the first of the targets to fall due. output of an industry after adding up all outputs and subtracting School enrollment data are reported to the UNESCO Institute for intermediate inputs. The industrial origin of value added is deter- Statistics by national education authorities. Primary education pro- mined by the ISIC revision 3. The World Bank conventionally mea- vides children with basic reading, writing, and mathematics skills sures GDP in U.S. dollars and applies the average official exchange along with an elementary understanding of such subjects as history, rate reported by the IMF for the year shown. An alternative conver- geography, natural science, social science, art, and music. Secondary sion factor is applied if the official exchange rate is judged to diverge education completes the provision of basic education that began at by an exceptionally large margin from the rate effectively applied to the primary level, and aims at laying foundations for lifelong learn- transactions in foreign currencies and traded products. ing and human development, by offering more subject- or skill-ori- Gross domestic product average annual growth rate is calcu- ented instruction using more specialized teachers. lated from constant price GDP data in local currency. Under-five mortality rate is the probability that a newborn Agricultural productivity refers to the ratio of agricultural baby will die before reaching age five if subject to current age-spe- value added, measured in constant 2000 U.S. dollars, to the number cific mortality rates. The probability is expressed as a rate per 1,000. of workers in agriculture. The main sources of mortality date are vital registration systems Value added is the net output of an industry after adding up all and direct or indirect estimates based on sample surveys or cen- out-puts and subtracting intermediate inputs. The industrial origin suses. To produce harmonized estimates of under-five mortality of value added is determined by the ISIC revision 3. Selected world development indicators 303 Agriculture value added corresponds to ISIC divisions 1­5 and (basic manufactures), 7 (machinery and transport equipment), and includes forestry and fishing. 8 (miscellaneous manufactured goods), excluding division 68. Industry value added comprises mining, manufacturing, con- High technology exports are products with high R&D inten- struction, electricity, water, and gas (ISIC divisions 10­45). sity. They include high-technology products such as in aerospace, Services value added correspond to ISIC divisions 50­99. computers, pharmaceuticals, scientific instruments, and electrical Household final consumption expenditure is the market value machinery. of all goods and services, including durable products (such as cars, Current account balance is the sum of net exports of goods and washing machines, and home computers) purchased by households. It services, net income, and net current transfers. excludes purchases of dwellings but includes imputed rent for owner- Foreign direct investment is net inflows of investment to occupied dwellings. It also includes payments and fees to governments acquire a lasting management interest (10 percent or more of vot- to obtain permits and licenses. Here, household consumption expen- ing stock) in an enterprise operating in an economy other than diture includes the expenditures of nonprofit institutions serving that of the investor. It is the sum of equity capital, reinvestment of households, even when reported separately by the country. In practice, earnings, other long-term capital, and short-term capital, as shown household consumption expenditure may include any statistical dis- in the balance of payments. Data on the current account balance, crepancy in the use of resources relative to the supply of resources. private capital flows, and foreign direct investment are drawn from General government final consumption expenditure includes the IMF's Balance of Payments Statistics Yearbook and International all government current expenditures for purchases of goods and Financial Statistics. services (including compensation of employees). It also includes Official development assistance or official aid from the high- most expenditures on national defense and security, but excludes income members of the OECD are the main source of official government military expenditures that are part of government external finance for developing countries, but official development capital formation. assistance (ODA) is also disbursed by some important donor coun- Gross capital formation consists of outlays on additions to the tries that are not members of OECD's Development Assistance fixed assets of the economy plus net changes in the level of invento- Committee (DAC). DAC has three criteria for ODA: it is under- ries and valuables. Fixed assets include land improvements (fences, taken by the official sector; it promotes economic development ditches, drains, etc.); plant, machinery, and equipment purchases; or welfare as a main objective; and it is provided on concessional and the construction of buildings, roads, railways, and the like, terms, with a grant element of at least 25 percent on loans. including commercial and industrial buildings, offices, schools, hos- Official development assistance comprises grants and loans, net of pitals, and private dwellings. Inventories are stocks of goods held by repayments, that meet the DAC definition of ODA and are made to firms to meet temporary or unexpected fluctuations in production countries and territories in part I of the DAC list of aid recipients. or sales, along with "work in progress." According to the 1993 SNA, Official aid comprises grants and ODA-like loans, net of repayments, net acquisitions of valuables are also considered capital formation. to countries and territories in part II of the DAC list of aid recipients. External balance of goods and services is exports of goods and Bilateral grants are transfers in money or in kind for which no repay- services less imports of goods and services. Trade in goods and ser- ment is required. Bilateral loans are loans extended by governments vices comprise all transactions between residents of a country and or official agencies that have a grant element of at least 25 percent and the rest of the world involving a change in ownership of general for which repayment is required in convertible currencies or in kind. merchandise, goods sent for processing and repairs, nonmonetary Total external debt is debt owed to nonresidents repayable in gold, and services. foreign currency, goods, or services. It is the sum of public, publicly The GDP implicit deflator reflects changes in prices for all final guaranteed, and private nonguaranteed long-term debt, use of IMF demand categories, such as government consumption, capital forma- credit, and short-term debt. Short-term debt includes all debt hav- tion, and international trade, as well as the main component, private ing an original maturity of one year or less and interest in arrears final consumption. It is derived as the ratio of current to constant on long-term debt. price GDP. The GDP deflator may also be calculated explicitly as Present value of debt is the sum of short-term external debt a Paasche price index in which the weights are the current period plus the discounted sum of total debt service payments due on quantities of output. public, publicly guaranteed, and private nonguaranteed long-term National accounts indicators for most developing countries are external debt over the life of existing loans. collected from national statistical organizations and central banks The main sources of external debt information are reports to the by visiting and resident World Bank missions. Data for high-income World Bank through its Debtor Reporting System from member economies are from the OECD. countries that have received World Bank loans. Additional informa- tion has been drawn from the files of the World Bank and the IMF. Table 5. Trade, aid, and finance Summary tables of the external debt of developing countries are pub- Merchandise exports show the free on board (FOB) value of goods lished annually in the World Bank's Global Development Finance. provided to the rest of the world, valued in U.S. dollars. Domestic credit provided by banking sector includes all credit Merchandise imports show the CIF value of goods (the cost of to various sectors on a gross basis, with the exception of credit to the the goods including insurance and freight) purchased from the rest central government, which is net. The banking sector includes mon- of the world valued in U.S. dollars. Data on merchandise trade come etary authorities, deposit money banks, and other banking institu- from the World Trade Organization (WTO) in its annual report. tions for which data are available (including institutions that do not Manufactured exports comprise the commodities in Standard accept transferable deposits but do incur liabilities such as time and Industrial Trade Classification (SITC) sections 5 (chemicals), 6 savings deposits). Examples of other banking institutions include 304 WORLD DEVELOPMENT REPORT 2007 savings and mortgage loan institutions and building and loan asso- factor. The purpose of the Atlas conversion factor is to reduce ciations. Data are from the IMF's International Financial Statistics. the impact of exchange rate fluctuations in the cross-country Net migration is the net average annual number of migrants comparison of national incomes. The Atlas conversion factor for during the period, that is, the annual number of immigrants less any year is the average of a country's exchange rate (or alterna- the annual number of emigrants, including both citizens and non- tive conversion factor) for that year and its exchange rates for the citizens. Data shown in the table are five-year estimates. Data are two preceding years, adjusted for the difference between the rate from the United Nations Population Division's World Population of inflation in the country and that in Japan, the United King- Prospects: The 2004 Revision. dom, the United States, and the Euro Zone. A country's inflation rate is measured by the change in its GDP deflator. The inflation Table 6. Key indicators for other economies rate for Japan, the United Kingdom, the United States, and the See technical notes for table 1, Key indicators of development. Euro Area, representing international inflation, is measured by the change in the SDR deflator. (Special drawing rights, or SDRs, Statistical methods are the IMF's unit of account.) The SDR deflator is calculated This section describes the calculation of the least-squares growth as a weighted average of these countries' GDP deflators in SDR rate, the exponential (endpoint) growth rate, and the World Bank's terms, the weights being the amount of each country's currency Atlas method for calculating the conversion factor used to estimate in one SDR unit. Weights vary over time because both the com- GNI and GNI per capita in U.S. dollars. position of the SDR and the relative exchange rates for each cur- rency change. The SDR deflator is calculated in SDR terms first Least-squares growth rate and then converted to U.S. dollars using the SDR to dollar Atlas Least-squares growth rates are used wherever there is a sufficiently conversion factor. The Atlas conversion factor is then applied to a long time series to permit a reliable calculation. No growth rate is country's GNI. The resulting GNI in U.S. dollars is divided by the calculated if more than half the observations in a period are missing. midyear population to derive GNI per capita. The least-squares growth rate, r, is estimated by fitting a linear When official exchange rates are deemed to be unreliable or regression trendline to the logarithmic annual values of the vari- unrepresentative of the effective exchange rate during a period, an able in the relevant period. The regression equation takes the form alternative estimate of the exchange rate is used in the Atlas for- ln Xt = a + bt, mula (see below). The following formulas describe the calculation of the Atlas which is equivalent to the logarithmic transformation of the com- conversion factor for year t : pound growth equation, S$ S$ Xt = Xo (1 + r)t. p p p p * et =1[ et t t / + t t / + ] S$ et-1 S$ et In this equation, X is the variable, t is time, and a = log Xo and 3 -2 p p t-2 p p t-2 t-1 t-1 b = ln (1 + r ) are the parameters to be estimated. If b* is the least- squares estimate of b, the average annual growth rate, r, is obtained as and the calculation of GNI per capita in U.S. dollars for year t : [exp(b* )­1] and is multiplied by 100 to express it as a percentage. $ * The calculated growth rate is an average rate that is representa- Yt = (Yt/Nt)/et , tive of the available observations over the entire period. It does not where et* is the Atlas conversion factor (national currency to necessarily match the actual growth rate between any two periods. the U.S. dollar) for year t, et is the average annual exchange rate (national currency to the U.S. dollar) for year t, pt is the GDP defla- Exponential growth rate tor for year t, ptS$is the SDR deflator in U.S. dollar terms for year t, The growth rate between two points in time for certain demo- Yt is the Atlas GNI per capita in U.S. dollars in year t, Yt is current $ graphic data, notably labor force and population, is calculated from GNI (local currency) for year t, and Nt is the midyear population the equation for year t. r = ln (pn /p1)/n, Alternative conversion factors where pn and p1 are the last and first observations in the period, n is The World Bank systematically assesses the appropriateness of offi- the number of years in the period, and ln is the natural logarithm cial exchange rates as conversion factors. An alternative conversion operator. This growth rate is based on a model of continuous, factor is used when the official exchange rate is judged to diverge by exponential growth between two points in time. It does not take an exceptionally large margin from the rate effectively applied to into account the intermediate values of the series. Note also that the domestic transactions of foreign currencies and traded products. exponential growth rate does not correspond to the annual rate of This applies to only a small number of countries, as shown in the change measured at a one-year interval, which is given by primary data documentation table in World Development Indicators (pn ­ pn )/pn . 2006. Alternative conversion factors are used in the Atlas methodol- ­1 ­1 ogy and elsewhere in the Selected World Development Indicators as single-year conversion factors. World Bank Atlas method In calculating GNI and GNI per capita in U.S. dollars for certain operational purposes, the World Bank uses the Atlas conversion Index In the index, b refers to box, f refers to figure, and t refers to table. Bakassi,Boys, Nigeria, 176 Bangladesh citizenship, exercising, 169, 171t ABC (Abstain, Be faithful, use Condoms) campaign, 132f decision-making capabilities of young people, 15­16, 15f, 54f abortion, 139 demographics and policy priorities, 212 Abrindo Espaços, Brazil, 142­43b early marriage of girls, 145b, 156, 157b Abstain, Be faithful, use Condoms (ABC) campaign, 132f ideal spouse, concept of, 148b accountability life skills, importance of teaching, 83 evaluation, via. See monitoring and evaluation microcredit programs, 57b, 58, 155b government policies and reforms, 214, 215 migration and mobility, 195, 196, 201 rehabilitation of youth and need for, 20­21, 63 nutritional issues, 148, 154 Afghanistan, 4, 177, 210b out-of-school youth, programs for, 20, 90b African Virtual University (AVU), 205b overage enrollment in primary education, 71 agency of young people. See capabilities, focus on part-time schooling, demand for, 93 AGETIP program, 116 sexual and reproductive health, 149, 150.154 AIDS. See HIV/AIDS subsidies for unmarried girls in school, 18, 58, 87 Albania work and education, combining, 86 citizenship, exercising, 171t working opportunities for young women, improving, 51b decision-making capabilities of young people, 15f, 54f behavioral and thinking skills. See life skills higher education, regional network of student organizations behavioral economics and risk-taking behavior, 58b on, 220 Belgium, 109 ICT, global flow of, 203 Benin, 149 informal sector, starting work in, 102 Better Life Options program, India, 16­17, 155­56 migration and mobility, 194, 196f bilateral youth programs, 209b Alcatraz Project, Venezuela, 181b birth control, 139, 149­50, 150f alcohol consumption. See substance abuse Bolivia, 85, 169, 201 Algeria, 17, 83, 207 Bolsa Familia, Brazil, 142b anemia, 149b, 149f, 150­51, 152 Bosnia and Herzegovina, 101b, 183 Annan, Kofi, 183 Botswana, 131, 175 antenatal care, 150­51, 150f boys. See men and boys antiretroviral therapy (ART) for HIV/AIDS, 138­39, 140 Brazil apprenticeships, 111­12. See also training programs access to secondary and tertiary education, 73 Arab Republic of Egypt. See Egypt, Arab Republic of achievement levels in secondary education, 69 Argentina actions taken by young people in, 225b financial crisis, effect on young people of, 51 after-school programs and conditional cash transfers, 87 informal sector, starting work in, 102 citizenship, exercising, 15, 53b, 167­72, 169b Joven(es) programs, 117b demographics of, 43f migration and mobility, 193 dependency decline in developing countries, 35f secondary and tertiary education, access to, 73 donors, role of, 210b self-employment and entrepreneurship, 114 equity issues in, 142­43b technical and vocational specialization, 113 favelas, 143b, 168, 169b, 187 work, second chance programs providing, 115­16 gangs, 176 Armenia, 39, 146, 201, 212 government policies and reforms, 216, 221b ART (antiretroviral therapy) for HIV/AIDS, 138­39, 140 HIV/AIDS treatment, 139 Australia, 18, 172 ICT, global flow of, 206b AVU (African Virtual University), 205b individual learning accounts, 88 305 306 WORLD DEVELOPMENT REPORT 2007 mortality rates and poverty levels, 125 Joven(es) programs, 117b nutrition in early childhood, 48 life transitions for young people in, 41­42, 42f overage enrollment in primary education, 71 labor demand fluctuations for adult vs. young males, 103f rural youth, opportunities for, 110 minimum wage laws, 108 self-employment and entrepreneurship, 114 ministries of youth, role of, 215b youth voice in government policies and reforms, 221b restrictive employment laws, effect on young people of, 105 Bulgaria, 220 secondary and tertiary education, 73, 74, 79b Burkina Faso, 79, 102, 125, 157 self-employment and entrepreneurship, 114 technical and vocational specialization, 113 teenage mothers, flexible school policies for, 157 Cambodia Un Techo Para Chile, 25 anemia and antenatal care, 150­51 youth departments, 219b citizenship, exercising, 172 China ICT, global flow of, 204 citizenship, exercising, 162, 163, 167, 173 life skills, importance of teaching, 83 demographics and policy priorities, 212 restrictive employment laws, effect on young people of, 106 dependency decline in developing countries, 35f, 36 secondary and tertiary education, evaluation of, 74 diet and exercise, health problems related to, 128 work and education, combining, 85 economic restrictions on alcohol and tobacco access, 136 Cameroon education's relevance to work in, 76 health education, 131, 135b ICT, global flow of, 201, 204t migration and mobility, 194, 201 migration and mobility, 189, 197 nutrition in early childhood, 48 migration of youth of, 14, 110­11 teenage mothers, flexible school policies for, 157 technical and vocational specialization, 113 Canada, 162, 189b, 192, 198 citizenship, exercising, 9­10, 10f, 160­84 capabilities, focus on, 2, 15­19, 15f, 53­59. See also knowledge base advantages of youth participation, 167 of young people; life skills capabilities, focus on, 172­78 action to be taken by young people, 225­28b community service, 171­72, 171t citizenship, exercising, 172­78 consequences in later adulthood, 164­67 donors, role of, 209­12b defining citizenship, 161b family and parenthood, 153­56, 155b, 156f in democracies, 32f, 165­66, 168 government policies and reforms, youth voice in, 214­17, 219­23 disadvantage, civil unrest, and violence, effects of, 166, 169b, 186b health issues, 129­35, 130b, 135b education and schools, impact of, 173­74 ICT, global flow of, 207­8 encouraging participation in future, 22­23 income and resources, control of and access to, 17­19, 56­58, environmentalism, youth sensitivity to, 167f 84­88 gangs, 175­76 life transitions, control of young people over, 15­16, 15f, 54b, 54f gender disparities in, 163, 165t, 170 policy formation, youth voice in, 214­17 indicators of youth development, 222b secondary and tertiary education, encouraging better decisions legal identity, need for, 173 about, 80­88, 92t migration and mobility, 195­96 Sierra Leone, rebuilding lives and institutions in, 186b military service, 170­71, 171t working life, start of, 92t national service, 171­72, 171t capacity building for government policies and reforms, 218­21 negative aspects of. See civil unrest; crime capital punishment for young people, 178 opportunities, focus on, 14­15, 52­53, 53b, 167­72, 173b cardiovascular disease, 124, 128 policy summary, 183­84, 184t career and school guidance counselors, 84, 89b, 92t rising/declining rates of, 161­64, 162­63f, 163­65t, 168 cell phones. See information and communication technology second chances, focus on, 178­83 (ICT), global flow of voting by young people, 32f, 168­69, 169f Chad, 146, 212 by women and girls, 163, 164, 165, 165t, 170, 174, 175 child care, 10f, 71, 155 working life, start of, 174­75 child labor, 7, 37, 50­51, 87, 96­98, 97f youth councils, 169­71 children living on streets as substance abusers, 127b youth development and youth action programs, 175 children of young people. See family and parenthood civil unrest Chile effects on development of, 166 citizenship, exercising, 168, 169f, 172 rehabilitative programs for young combatants, 181­83, 182b government policies and reforms, 216 in Sierra Leone, 185­86b health education, 130 social inequities and, 166, 173 ICT, global flow of, 207 youth participation in, 166, 173b, 177­78, 185­86b Index 307 Colombia taking advantage of, 4­5, 5f, 35­36 citizenship, exercising, 175 violence and, 166b civil unrest, rehabilitative programs for young combatants women and girls, education and participation in labor market involved in, 181 of, 86, 104 credit-based courses, 74 working life, effect on start of, 104, 120­21b crime, juvenile, 180 youth opportunities affected by, 33­36 employer training, 112 Denmark, 198 equivalence programs in, 91 departments and ministries of youth, role of, 215b, 219b government policies and reforms, effects of, 108 Desa Siaga, Indonesia, 154 gun control, 177 development. See economic growth and development Joven(es) programs, 117b developmental psychology of risk-taking behavior, 58b nutritional deficiencies, 148 diabetes, 124, 128 school vouchers program, 87 diet. See nutrition and health community service, 171­72, 171t direct job creation schemes, 109 conditional cash transfers, 86­87, 142b, 152­53, 157b disabled youth, employment of, 115b, 204 condom use, 125­26, 125f, 132­33f, 135, 140, 151­52 discrimination. See equity and young people; women and girls Congo, Democratic Republic of, 178, 181 distance education, 204­5, 205b contraception, 139, 149­50, 150f domestic violence, 145b coordination of policies, 22, 23b, 143b, 214, 215­21, 220b, 224 Dominican Republic corruption consulation of youth in project preparation, 210b democratic citizenship reducing, 166 education, informing young people about advantages of, 16, in higher education, 25, 78b 17b, 56 cost sharing in secondary and tertiary education, 80 family and parenthood, 146 Costa Rica, 108, 189 ICT, global flow of, 207 Côte d'Ivoire, 173b, 189 Joven(es) programs, 117b credit and loans. See also financing and funding donors, role of, 209­12b income and resources, young peoples' control of and access to, dowry payments and domestic violence, 145b 18, 57b drug abuse. See substance abuse income-contingent loans, 87­88, 88b microcredit, 57b, 58, 155b for secondary and tertiary education, 18, 57b, 87­88, 88b early child development crime child care and pre-school programs, 10f, 71, 155 ages of people most likely to commit, 20 parenting skills, 151, 155 combating, 177 East Asia financial crisis of 1997-1998, 50­51 effects of, 166 economic analysis of risk-taking behavior, 58b gangs, 175­76, 196 economic growth and development, 29­31, 38­39b private sector interventions, 181b citizenship, exercising, 164­67 punishment for, 178­81, 179f family and parenthood, 144­47 rehabilitative programs, 20­21, 180­81b health issues, 123­24 restorative justice, 178­81 HIV/AIDS second chances, focus on, 20­21, 178­83, 180­81b effects of growth on, 94b status crimes, youth arrested for, 178 effects on growth of, 29­30, 30b, 123­24 Croatia, 220 ICT, policies enhancing development impact of global flow of, cross-sectoral effects of youth programs, 22, 23b, 187 206­8 Czech Republic, 109, 196 migration and mobility as development issue, 191­92, 191t, 192f, 196­201 policy priorities and, 212­16 death penalty for young people, 178 working life, start of, 106­7, 107b decision-making capabilities of young people. See capabilities, economic incentives focus on family and parenthood, 152­53 democracies, exercising citizenship in, 32f, 165­66, 168 health issues, 136­37, 137b Democratic Republic of Congo, 178, 181 secondary and tertiary education, 18, 81­82, 81f demographics, 26­27 Ecuador, 71, 155 degree of urgency for policy reform and, 212 education, 22t, 68­93. See also primary education; secondary and dependency decline in developing countries, 34­36, 35f tertiary education family and parenthood, 146­47, 146f actions taken by young people, 226b regional variations, 33f, 43­44, 44f citizenship and, 173­74 308 WORLD DEVELOPMENT REPORT 2007 education (continued) restrictive employment laws, effect on young people of, 105 coordination of health and education policies, 220b European Union (EU), migration to, 198. See also member countries cross-sectoral effects of school uniform program in Kenya, 23b evaluation. See monitoring and evaluation family and parenthood affected by, 147b exercise, lack of, 124, 128f, 129f gender and, 65­66 experience, returns to, 102f health and, 129­35, 130b, 220b experimentation, youth as time of, 126 ICT, global flow of, 204­5, 205b indicators of youth development in, 222b literacy, 37f, 70f, 91, 182f, 209f family and parenthood, 9f, 144­59. See also marriage migration and mobility, 194 abortion, 139 nutrition, 154­55 age at birth of first child, 144, 145, 146f policy summary, 92t, 93 birth control, 139, 149­50, 150f quality of basic education, improving, 6b, 11­12, 16, 48­49, 48f, capabilities, focus on, 153­56, 155b, 156f 71­72, 73 couples, reproductive health services for, 154 remedial programs, 89­90 demographics of, 146­47, 146f sex education, 149, 153­54 domestic violence, 145b Sierra Leone, rebuilding lives and institutions in, 185b early childbearing teenage mothers, 156­57 effects of, 28­29, 39b, 86, 145b of women and girls. See women and girls preventing, 153­54 work and education, combining, 85­86, 86f, 97 second chances for young mothers, focus on, 156­57 Egypt, Arab Republic of economic growth and poverty reduction, 144­47 apprenticeships, 112 economic incentives, 152­53 dependency ratio decline in developing countries, 35f education and, 147b diet and exercise, health problems related to, 128 government investment in, 147 high unemployment rates for young people in, 99 health issues, 144, 147­55 (See also nutrition) Internet use in, 201 high unemployment and low marriage rates, relationship nutritional deficiencies, 148, 149 between, 40b, 42, 147 social barriers to work opportunities for women in, 106 HIV/AIDS, 148 status crimes, youth arrested for, 178 indicators of youth development, 222b street children as substance abusers, 127b knowledge base of young people regarding, 149­51, 153­56 El Salvador, 194, 195, 196, 215b, 219b life skills for women and girls, 155­56, 156f emergency contraception, 139, 140 maternal mortality, 145­46 emigration. See migration and mobility maternal schooling, positive effects of, 80 employment. See working life, start of nutritional issues. See nutrition and health Employment protection legislation (EPL), effect on young people opportunities, focus on, 151­53 of, 13, 51­52, 104­9, 108b, 121b ORT (oral rehydration therapy), parental knowledge of, 151 Entra 21, 92, 116­17 parenting skills, 151, 155 entrepreneurship. See self-employment and entrepreneurship policy summary, 157­59, 158t environmentalism, youth sensitivity to, 167f poor preparation for, 147­51 EPL (employment protection legislation), effect on young people reproductive health services, 138­40, 149­51, 154­55 of, 13, 51­52, 104­9, 108b, 121b second chances, focus on, 156­57 equity and young people favelas, Brazil, 143b, 168, 169b, 187 Brazil, addressing disparities in, 142­43b fertility rates. See demographics civil unrest and social inequities, 166, 173 financing and funding economic equity, 28­29 conditional cash transfers, 86­87, 142b, 152­53, 157b equality of opportunity, 84­88 income and resources, young peoples' control of and access to, gangs and gang formation, 175, 176 18, 57b gender equity, 28­29, 38­39b, 65­66 income-contingent loans, 87­88, 88b in health issues, 122, 125­26 microcredit, 57b, 58, 155b minority youth, unemployment rates for, 120b secondary and tertiary education, 78­80, 79b, 84­88 Estonia, 3, 107 secondary and tertiary education, expanding access to, 49­50 Ethiopia five life transitions of young people. See life transitions for young citizenship, exercising, 171t people decision-making capabilities of young people, 15f, 54f France demographics and policy priorities, 212 apprenticeships, 111 Internet use in, 201 citizenship, exercising, 163, 171 life skills, importance of teaching, 83 countercultural language of North African immigrants to, 173 migration and mobility, 196 government policies and reforms, effects of, 108b Index 309 migration and mobility, 194, 198 Guatemala, 48, 86, 155 minority youth, unemployment rates among, 120b guidance counselors, 84, 89b, 92t Guinea, 145b gun control, 177 gangs, 175­76, 196 gender issues and youth opportunity, 65­66, 216b. See also men and boys; women and girls Haiti, 40­42, 41f, 149, 195, 206 geographic mobility. See migration and mobility harassment and violence in schools, 77b Georgia, 78b, 93, 102 health issues, 8f, 122­41. See also specific conditions and habits, Germany e.g. AIDS/HIV, smoking and tobacco use community and national service, 171 access to health services, 135­36, 138­39 coordination within donor agencies on youth issues, 209­12b attitudes and preferences, changing, 132­34 dual system in, 76, 111 capabilities, focus on, 129­35, 130b, 135b migration and mobility, 192, 198 changes in, 32 political violence in, 173b, 177 civil unrest, rebuilding lives and institutions following, 185b Ghana coordination of health and education policies, 220b apprenticeships, 112 economic incentives, 136­37, 137b citizenship, exercising, 171 education and, 129­35, 130b, 220b health education, 131 equity in, 122, 125­26 ICT, global flow of, 202 expenditures on health care, 124 literacy program, 91 family and parenthood, 144, 147­55 minimum wage laws, 108 ICT, global flow of, 205­6, 206b quality of basic education, 6b indicators of youth development, 222b reproductive health services, access to, 138 information campaigns, 129­35, 130­31b, 132­33f, 134t, rising demand for secondary and tertiary education, 31f 135b, 143b Giddings State School, Texas, 180b knowledge base and decision-making capabilities, 16­17, 39, girls. See women and girls 55­56, 125­26, 129­35 globalization, 32­33, 187, 188. See also information and commu- long-term effects of risky behavior, 122, 126 nication technology (ICT), global flow of; migration and migration and mobility, 195, 200­201 mobility nutrition. See nutrition and health government jobs, 109 opportunities, focus on, 135­37, 137b government policy towards youth, 211­21. See also policy foci; policy summary, 140­41, 141t policy summaries poverty reduction and economic growth affected by, 123­24 accountability, 214, 215 primary threats to young people, 3 capacity building, 218­21 public intervention, 124­25 civil unrest, rebuilding lives and institutions in countries reproductive health services, 138­40, 149­51, 154­55, 205­6, 206b wrecked by, 185­86b restricting access to health risks, 136­37, 137b coherent framework, developing, 215­21 second chances, focus on, 20, 137­40 coordination of, 22, 23b, 143b, 214, 215­21, 220b, 224 treatment and prevention, relationship between, 21, 63, 138­40 economic growth and poverty reduction, effects of, 95b higher education. See secondary and tertiary education family and parenthood, 147 HIV/AIDS gender issues, addressing, 216b ABC (Abstain, Be faithful, use Condoms) campaign, 132f health issues, 124­25 access to testing and treatment, 135, 138­39, 152 ICT, global flow of, 17, 206­8 age of infection, 20 migration and mobility, 196­201 ART (antiretroviral therapy), 138­39, 140 ministries and departments of youth, role of, 215b, 219b condom use, 125­26, 125f, 132­33f, 135, 140, 151­52 monitoring and evaluation, 214, 215, 221­26, 222b, 223b coordination of health and education policies, 220b policy foci, 2, 10­11, 11f, 22t, 45­64, 45f (See also capabilities, cost of health care expenditures on, 124 focus on; opportunity, focus on; second chances, focus on) economic growth and development Poverty Reduction Strategy Papers (PRSPs), 217b effects of, 94b priorities and country context, variations in, 212­16 effects on, 29­30, 30b, 123­24 restrictive employment laws, effect on young people of, 13, family and parenthood, 148 51­52, 104­9, 108b, 121b government policies and reforms, 219 working life, effects on start of, 107­9, 108b, 121b health education, 130­35, 131b, 132­33f, 134t, 135b youth voice in, 214­17, 219­23, 221b ICT, global flow of, 206b Grameen Bank, Bangladesh, 155b knowledge base of young people regarding, 17, 39, 56 growth. See economic growth and development long-term and interactive effects of secondary education and Grupo Cultural Afro Reggae, Brazil, 187 AIDS prevention investments, 5b 310 WORLD DEVELOPMENT REPORT 2007 HIV/AIDS (continued) teacher absenteeism in, 76 long-term effects of risky behavior, 122 violence and harassment in schools, 77b migration and mobility, 195, 200­201 women and girls, social barriers to continuing education of, 83 older men with high infection rates, young women having sex indicators of youth development, 222b with, 16, 126, 141, 145b, 148, 153­54 individual learning accounts, 88 peer-to-peer learning in Jamaica, 17 Indonesia prevalence among young people, 129f citizenship, exercising, 163, 171 quality of basic education and, 6b civil unrest, youth participation in, 173b substance abuse treatment and, 138 economic growth and young people's work opportunities, 106 treatment and prevention, relationship between, 21, 63, 138­40 employer training, 112 unsafe sex, 124, 125­26 health and education, 130b Hong Kong (China), 89 high unemployment rates for young people in, 99 Horizon Jeunes, Cameroon, 135b ICT, global flow of, 201, 202t human capital maternal care and nutrition, 154 application of model to young people, 47b migration and mobility, 201 defined, 28b minimum wage laws, 108 donors, role of, 209b ORT (oral rehydration therapy), parental knowledge of, 151 economic growth and, 29­31 secondary education, transition rate to, 69f importance of investing in young people as, 2­4 sexual and reproductive health knowledge in, 149, 154 long-term and interactive effects of secondary education and smoking in, 124, 127, 136 AIDS prevention investments, 5b smoking rates for young people in, 39 low-paying and dead-end jobs leading to waste of, 101­3 informal sector, starting work in, 101­2, 112 poverty reduction and, 27­29 information and communication technology (ICT), global flow of, human trafficking, 196, 200­201 187, 188, 201­8 Hungary, 27, 102, 109 access of young people to, 17, 32­33, 32f, 201­3, 202­3f hypertension, 124, 128 capabilities, focus on, 207­8 communal/public access, 17, 202, 203f communication with people abroad, 203f ICT. See information and communication technology (ICT), definition of ICT, 201b global flow of education, 204­5, 205b identity of young people, 172­78. See also capabilities, focus on gender differences in access to, 202 illegal migration, 196f, 200­201 indicators of youth development, 222b ILO (International Labour Organisation) convention 182 on the information campaigns and, 17 Worst Form of Child Labor, 97 policies enhancing development impact of, 17, 206­8 immigration. See migration policy summary, 208t incarceration of young people, 178­81, 179f reproductive health information, obtaining, 205­6, 206b incentives, economic. See economic incentives risk-taking behavior, 207­8 income and resources, control of and access to, 17­19, 56­58, smoking cessation via text messaging, 137b 84­88 SMS (short messaging system) or text messaging, 137b, 201, income-contingent loans, 87­88, 88b 202, 203, 210b India use of new technologies by young people, 201­3, 201t, 202­3f, autonomous migrants, young people as, 55 204t Better Life Options program, 16­17, 155­56 working life, start of, 203­4 citizenship, exercising, 162, 163, 165, 167, 169, 174 youth transitions affected by, 203­6 credit-based courses, 74 information base. See knowledge base of young people early marriage prevention in, 153 information campaigns, 16­17, 55­56 emergency contraception, effects of availability of, 140 citizenship, exercising, 169 exercise and diet, health problems related to, 128 health issues, 129­35, 130­31b, 132­33f, 134t, 135b, 143b family and parenthood issues, 146, 148, 149, 150, 153, 154 International Covenant on Civil and Political Rights, 178 HIV/AIDS in, 152 International Labour Organisation (ILO) convention 182 on the ICT, global flow of, 203­4, 205, 207 Worst Form of Child Labor, 97 maternal schooling, positive effects of, 80 international migration. See migration and mobility migration and mobility, 193, 197 Internet. See information and communication technology (ICT), nutritional issues, 128, 148, 149, 154 global flow of remedial education, 89, 90 Iran, Islamic Republic of, 113, 168, 178, 181 reproductive health services, 150, 154 Iraq, 15­16, 15f, 54f, 171t, 196, 212 rural youth, opportunities for, 110b Ireland, 121b, 198 Index 311 Israel, 90, 171 labor market, entering. See working life, start of IT (information technology). See information and communication Lao People's Democratic Republic, 71 technology (ICT), global flow of Liberia, 183 Italy, 198, 226b life skills, 19, 58­59 making decisions about continuing education and, 82­83 quality of secondary and tertiary education improved by, 71, Jamaica 72b, 75b child care and pre-school programs, payoff of investment in, 12f value set on life skills by young people, 93 crime rates among young people in, 20 for women and girls, 155­56, 156f gangs, 176 life transitions for young people, 5­10, 39­42, 67. See also citizen- health education, 130 ship, exercising; family and parenthood; health issues; migration and mobility, 194 secondary and tertiary education; working life, start of ministries of youth, role of, 215b control over, 15­16, 15f, 54b, 54f parenting skills, 155 ICT, effects of global flow of, 203­6 peer-to-peer learning in, 17 indicators of youth development and, 222b teenage mothers, integrated programs for, 157 migration and mobility affecting, 192­96, 193f Japan, 76, 163 spillover effects, 222­25 job creation schemes, 109 literacy, 37f, 70f, 91, 182f, 209f job training. See training programs loans. See credit and loans; financing and funding jobs generally. See working life, start of Joven(es) programs, 21, 66, 88, 92, 117­18, 117b, 215, 224 Macedonia, former Yugoslav Republic of, 25, 175, 210b, 220, 225b Make a Connection program, 72b kafoolu, 173 Malawi, 225b Kant, Immanuel, 160 Malaysia Kenya citizenship, exercising, 171t actions taken by young people in, 225­28b decision-making capabilities of young people, 15f, 54f apprenticeships, 112 demographics and policy priorities, 212 child care, 155 economic growth and young people's work opportunities, 107 citizenship, exercising, 172 employer training, 112 economic growth and productivity, effect of HIV/AIDS on, 30b migration and mobility, 194, 196 economic incentives in secondary and tertiary education, 81f secondary and tertiary education, access to, 73 employer training, 112 technical and vocational specialization, 113 health and education, 16, 130b, 134, 140­41 working opportunities for young women, improving, 13, 51b ICT, global flow of, 204 Mali, 125, 145b, 149, 150 illicit abortions, complications arising from, 139 marriage. See also family and parenthood migration and mobility, 194 age at, 144­45, 145b model of synergies between education and health policy, 219 childbearing and marriage, relationship between, 145b ORT (oral rehydration therapy), parental knowledge of, 151 domestic violence, 145b primary education, overage enrollment in, 71 early marriage school uniform program, 23b, 136 incidence of, 144­45, 145b secondary and tertiary education, access to, 73 preventing, 153, 156, 157b sex education in, 16, 145b, 153­54 economic and family pressures regarding, 18­19, 147 status crimes, youth arrested for, 178 high unemployment and low marriage rates, relationship support for evaluation of SMS services in, 210 between, 40b, 42, 147 treatment programs and risk-taking behavior, 21, 63, 140 income and resources, control of and access to, 18­19, 57­58 violence and harassment in schools, 77b young women married to older men, 16, 18, 126, 141, 145b, women and girls, social barriers to continuing education of, 65, 148, 153­54 83 mass marketing campaigns. See information campaigns knowledge base of young people, 16­17, 55­56 maternity. See family and parenthood family and parenthood, information pertinent to, 149­51, Mauritius, 113 153­56 media campaigns. See information campaigns health issues, 16­17, 39, 55­56, 125­26, 129­35 men and boys, 65­66. See also family and parenthood; marriage poverty reduction and economic growth, 39 age at birth of first child, 144 working life, starting, 104 Brazil, Program H in, 143b Korea, Republic of, 51, 74, 79b, 113, 117 citizenship, gender disparities in exercising, 163, 165t, 170 Kyrgyz Republic, 201, 202f employment levels for, 101 312 WORLD DEVELOPMENT REPORT 2007 men and boys (continued) gender issues and youth policy, 216b high unemployment and low marriage rates, relationship government policies and reforms, 214, 215, 221­26, 222b, 223b between, 40b, 42, 147 of health education programs, 134­35 Internet, public access to, 202 indicators of youth development, 222b labor demand fluctuation for young vs. adult males, 103f of policies, 214, 215, 221­26 reproductive health, sex, and family planning services for, 151, of secondary and tertiary education systems, 74, 77­78 154, 155b of youth programs, 23 Mexico Morocco, 105, 199b actions taken by young people in, 225b motivation and involvement of students, 80­84 diet and exercise, health problems related to, 128 motor traffic accidents, prevalence of youth deaths caused by, 16, employer training, 112 32, 58­60, 214 equivalence programs in, 91 Mozambique, 113, 186b, 196 family and parenthood, 151, 153 Myanmar, 181 health and education, 130, 130b ICT, global flow of, 205 income and educational opportunities, relationship between, Namibia, 130, 131b, 176 18, 57 national policy. See government policies and reforms integrated education programs for disadvantaged youth, 89b national service, 171­72, 171t migration and mobility, 14b, 192, 193, 194­95, 196, 196f, 197, 199 Nepal Oportunidades program, 18f, 66f, 87, 136, 153, 214, 223b actions taken by young people in, 225b self-employment and entrepreneurship, 114 citizenship, exercising, 165, 170 smoking rates for young people in, 39 ICT, global flow of, 206b microcredit, 57b, 58, 155b life skills programs for women and girls, 155­56 micronutrient deficiencies, 148­49, 152 nutrition in, 48, 148 migration and mobility, 187, 188­201 smoking rates for young people in, 39 autonomy of young people and, 55 Netherlands, 121b, 226b citizenship, exercising, 195­96 New Zealand, 121b, 137b, 190b, 194, 198, 199 as a development issue, 191­92, 191t, 192f, 196­201 Nicaragua economic growth and development, effects of, 94­95b autonomous schools in, 80 education, 194 family and parenthood, 148, 152­53 health issues, 195, 200­201 gangs, 176 indicators of youth development, 222b health education, 131 legal identity, need for, 173 migration and mobility, 189 policy summary, 208t minimum wage laws, 108 policies to enhance development impact of, 196­201 ministries of youth, role of, 215b reasons for, 189­91 secondary education, transition rate to, 69f receiving countries, concerns of, 197­99 self-employment and entrepreneurship, 113­14, 113t remittances, 192, 194 youth departments, 219b restrictions on allowing citizens to migrate, 200 Niger, 74, 194, 212 return home, 192, 196­97, 197f, 199 Nigeria risk-taking, illegal immigration, and human trafficking, 196f, citizenship, exercising, 162, 171 200­201 crime, juvenile, 178 small island states, high rates of emigration from, 190b gangs, 176 temporary worker programs, 197­99 health education, 131, 138 working opportunities for young people and, 14b, 52, 104, 105f, illicit abortions, complications arising from, 139 110­11, 192­93 reproductive health services, access to, 138 military service, 170­71, 171t sexual and reproductive health knowledge in, 149 Millennium Development Goals and young people, 28, 29b, 91 STI prevention and treatment in, 20 minimum wage laws, effects of, 13, 51­52, 104­6, 108­9, 121b nutrition and health ministries and departments of youth, role of, 215b, 219b access to nutrition services, 152 minority youth, unemployment rates for, 120b early childhood nutrition, 48, 71 mobile phones. See information and communication technology education programs, 154­55 (ICT), global flow of maternal, 144, 147­49, 149f, 152 mobility, geographic. See migration and mobility micronutrient deficiencies, 148­49, 152 Moldova, 170, 220 obesity, cardiovascular disease, diabetes and hypertension monitoring and evaluation related to, 124, 128f actions taken by young people, 226b young people, nutritional deficiencies common in, 148­49, donors, role of, 209b 149f Index 313 obesity, 124, 128f, 129f ICT, global flows of, 208t OECD countries migration and mobility, 208t baby boom in, 120­21b working life, start of, 118­19, 118t ministries of youth, role of, 215b political violence. See civil unrest Oportunidades program, Mexico, 18f, 66f, 87, 136, 153, 214, 223b politics, youth interest in. See citizenship, exercising opportunities, focus on, 2, 11­15, 46­53 postprimary education. See secondary and tertiary education basic services, enhancing quality and relevance of, 48­49 poverty reduction and young people, 27­29, 50b citizenship, exercising, 14­15, 52­53, 53b, 167­72, 173b education and income levels, 84­88, 85f, 86f, 88b donors, role of, 209b effects of poverty reduction and economic growth, 94­95b education. See primary education; secondary and tertiary family and parenthood, 144­47 education government policies and reforms, 217b family and parenthood, 151­53 health issues, 123­24 health issues, 135­37, 137b working young people, poverty levels of households with, 103 human capital model applied to young people, 47b Poverty Reduction Strategy Papers (PRSPs), 211, 217b postbasic services, expanding access to, 49­50 pre-school programs, 10f, 71, 155 variations in youth preparation for, 36­39 pregnancy. See family and parenthood working life, start of, 13­14, 50­52, 92t prenatal care, 150­51, 150f oral rehydration therapy (ORT), parental knowledge of, 151 primary education. See also education Otpor youth movement, Serbia, 167, 168b expanding access to, 5­6, 31f, 68 late starts in, 37, 71 quality of basic education, improving, 6b, 11­12, 16, 48­49, 48f, Pakistan, 150, 160, 167, 177, 187, 225b 71­72, 73 Panama, 108, 117b, 176 second chances, providing, 20 Paraguay, 117b prison, young people in, 178­81, 179f parental household, position of women in, 18­19, 147 private firms, youth wage subsidies for, 109 parenthood of young people. See family and parenthood private ownership, young peoples' attitudes towards, 163, 164t parenting skills, 151, 155 Program H, Brazil, 143b peace education programs, 177 ProJovem, Brazil, 142b peer-to-peer learning, 17, 56, 131 prostitution Peru economic pressures leading to, 19 contraceptive use in, 149­50 gang protection, 176 health education, 131 migration and human trafficking, 196 Joven(es) programs, 117b unequal power and risky sex, 126 minimum wage laws, 108 PRSPs (Poverty Reduction Strategy Papers), 211, 217b ORT (oral rehydration therapy), parental knowledge of, 151 public employment services, 117 secondary and tertiary education, access to, 73 public intervention. See government policies and reforms self-employment and entrepreneurship, 113­14, 113t public-private partnerships teenage mothers, flexible school policies for, 157 family and parenthood services, 151­52 Philippines secondary and tertiary education, 73­74, 80 actions taken by young people in, 225b public sector jobs, 109 citizenship, exercising, 169, 170 public works programs for disadvantaged youth, 115­16 crime, juvenile, 176, 180 diet and exercise, health problems related to, 128 equivalence programs in, 91 Qatar, 99 ICT, global flow of, 204 quality of basic education, 6b, 11­12, 16, 48­49, 48f, 71­72, 73 migration and mobility, 194, 195, 196, 197, 200, 201 nutrition in early childhood, 48 public employment services, 117 rehabilitative programs. See also second chances, focus on support for evaluation of SMS services in, 210 accountability and rehabilitation, 20­21, 63 Poland, 102, 109, 196 civil unrest, young combatants involved in, 181­83, 182b policy foci, 2, 10­11, 11f, 22t, 45­64, 45f. See also capabilities, focus crime, juvenile, 20­21, 180­81b on; opportunity, focus on; second chances, focus on religious identity, 173, 174 policy recommendations, summary. See also government policies remedial education, 89­90 and reforms remittances, 192, 194 citizenship, exercising, 183­84, 184t reproductive health services, 138­40, 149­51, 154­55, 205­6, 206b education, 92t, 93 Republic of Korea, 51, 74, 79b, 113, 117 family and parenthood, 157­59, 158t Republic of Yemen, 10, 65b, 210b health issues, 140­41, 141t República Bolivariana de Venezuela, 102, 108, 117b, 174, 181b 314 WORLD DEVELOPMENT REPORT 2007 resources and income, control of and access to, 17­19, 56­58, 84­88 training programs, 21, 116­18 restorative justice, 178­81 working life, start of, 114­18, 117b, 118t risk-taking behavior, 58b secondary and tertiary education, 5­7, 6f, 67. See also education age range associated with, 8f achievement levels, 69­70, 70f decision-making capabilities of young people and, 17, 19 advantages, informing young people about, 16, 17b, 56 experimentation, youth as time of, 126 barriers to enrollment in, 3f, 68­71, 69f health issues related to, 122 capability of young people to make decisions about, improving, ICT, global flow of, 207­8 80­88, 92t income and resources, control of and access to, 58 career and school guidance counselors, 84, 89b, 92t insufficient knowledge leading to, 125­26 corruption in, 25, 78b migration and mobility, 196f diversification of options, 73­74 neuroscience and, 58, 59b efficiency gains, 80 second chances and, 19, 21, 58­60, 63 enhancing opportunities for, 72­80 unequal power and risky sex, 126 equivalence programs, 90­91 youth tendency towards, 19, 58­60 evaluation and accountability, 74, 77­78 road traffic accidents, prevalence of youth deaths caused by, 16, 32, financing and funding, 78­80, 79b, 84­88 58­60, 214 flexibility, need for, 74 Romania ICT, global flow of, 204­5, 205b abortion, legalization of, 139 incentive mechanisms, 81­82 citizenship, exercising, 171t income and resources, control of and access to, 17­19, 56­57, crime, juvenile, 180 84­88 decision-making capabilities of young people, 15f, 54f knowledge base, improving, 16, 39, 55­56 demographics and policy priorities, 212 long-term and interactive effects of secondary education and ICT, global flow of, 203 AIDS prevention investments, 5b migration and mobility, 196 motivation and involvement of students, 80­84 rural youth, opportunities for, 109­10, 110b out-of-school youth, programs for, 90­92, 90b Russia policy summary, 92t, 93 actions taken by young people in, 225b postbasic services, investment in, 49­50, 49f alcohol sale restrictions, 137b poverty levels and income, 84­88, 85f, 86f, 88b citizenship, exercising, 163, 170, 171 quality of basic education and, 6b, 11­12, 16, 48­49, 48f, 71­72, civil unrest, youth participation in, 173b 73 informal sector, starting work in, 102 quality of, improving, 74­78 migration and mobility, 189, 197 relevance of, 12­13, 48­49, 73, 74­78, 84, 95b minimum wage laws, 108 rising demand for, 31­32, 31f, 68, 70 Rwanda, 173b second chances, providing, 20, 62­63, 71, 89­92, 89b, 90b, 92t stakeholders in school policy, young people as, 83­84 teacher training, 76­77 "Scared Straight" programs, 180 tracking, effects of, 63, 73 school guidance counselors, 84, 89b, 92t variations in access to, 36­39, 38b school uniform programs, 23b violence and harassment in, 77b school vouchers, 87 vocational specialization, 73, 74, 76 schools generally. See education; primary education; secondary young people neither employed nor studying, 100b and tertiary education youth unemployment levels and, 120b second chances, focus on, 2, 19­21, 59­63 self-employment and entrepreneurship accountability, importance of, 20­21, 63 income and resources, young peoples' control of and Brazil, addressing disparities in, 142­43b access to, 19 citizenship, exercising, 178­83 opportunities for young people, improving, 14 coordination with mainstream programs, 20, 62­63 as start of working life, 113­14 criminal youth, 20­21, 178­83, 180­81b Senegal, 69f, 116, 206b donors, role of, 210b Serbia and Montenegro, 115, 220 family and parenthood, 156­57 service programs for young people, 170­72 health issues, 20, 137­40 sex discrimination. See women and girls restorative justice, 178­81 sex education, 149, 153­54 risk-taking behavior and, 19, 21, 58­60, 63 sex workers. See prostitution in secondary and tertiary education, 20, 62­63, 71, 89­92, 89b, sexual abuse and harassment in schools, 77b 90b, 92t sexual attitudes and behavior of men and boys, Brazil's Program H targeting, importance of, 20, 62 addressing, 143b Index 315 sexual risk-taking amongst Vietnamese young people, 94b support for evaluation of SMS services in, 210 sexually transmitted infections (STIs), prevention and treat- working life, starting, 101, 115b, 116 ment of, 20, 131, 133, 135­39, 138f, 151, 152. See also status crimes, youth arrested for, 178 HIV/AIDS STIs (sexually transmitted infections), prevention and treat- short messaging system (SMS) or text messaging, 137b, 201, 202, ment of, 20, 131, 133, 135­39, 138f, 151, 152. See also 203, 210b HIV/AIDS Sierra Leone strategic policy foci, 2, 10­11, 11f, 22t, 45­64, 45f. See also capabili- actions taken by young people in, 225b ties, focus on; opportunity, focus on; second chances, citizenship, exercising, 163 focus on civil unrest, rehabilitative programs for young combatants street children as substance abusers, 127b involved in, 181, 183 Suami Siaga, Indonesia, 154 donors, role of, 210b substance abuse, 126­28, 127f policy priorities for, 213 consequences of, 124 rebuilding lives and institutions in, 185­86b poverty reduction and economic growth, effects on, 123 Singapore, 113, 174, 215b restricting access, 136­37, 137b skills. See also life skills by street children, 127b defined, 71b treatment programs, 137­38 parenting, 151, 155 Sudan, 181 secondary and tertiary education supplementary tutoring, 89­90 achievement levels, 69­70, 70f Sweden, 109, 162, 198, 215b, 218b relevance of, 12­13, 48­49 Switzerland, 192, 198 Slovak Republic, 109 Syrian Arab Republic, 99 Slovenia, 107b, 170 smoking and tobacco use, 127f age at time of taking up, 8, 42 Taiwan (China), 110, 112 cessation programs, 137b Tajikistan, 15f, 54f, 171t, 196 consequences of, 124 Taliban, 177 in developing countries, 39 Tanzania, 73, 130, 139, 204 early intervention, importance of, 124 teacher training, 76­77 economic analysis and, 58b, 59 technical and vocational specialization, 73, 74, 76, 112­13 long-term effects of, 122, 123f technology. See information and communication technology peer pressure as reason for, 58 (ICT), global flow of restricting access, 136­37, 137b temporary worker programs, 197­99 SMS (short messaging system) or text messaging, 137b, 201, 202, tertiary education. See secondary and tertiary education 203, 210b text messaging or short messaging system (SMS), 137b, 201, 202, social marketing campaigns. See information campaigns 203, 210b South Africa Thailand citizenship, exercising, 167, 171, 174 credit-based courses, 74 crime, juvenile, 178, 180 economic growth and young people's work opportunities, 107 diet and exercise, health problems related to, 128 education loans, 18 health and education, 130b, 131 maternal care and nutrition, 154 HIV/AIDS treatment, 139 public employment services, 117 ICT, global flow of, 204 secondary education, transition rate to, 69f migration and mobility, 189b thinking and behavioral skills. See life skills reproductive health services, access to, 138 tobacco. See smoking and tobacco use self-employment and entrepreneurship, 114 Tonga, 190b, 194, 199 teenage mothers, flexible school policies for, 156 tracking, effects of, 63, 73 violence and harassment in schools, 77b traffic accidents, prevalence of youth deaths caused by, 16, 32, vocational specialization in, 74 58­60, 214 South Korea, 51, 74, 79b, 113, 117 trafficking, human, 196, 200­201 Spain, 108b, 189b, 193, 198 training programs, 111­13 spillover effects, 223­26 employer training, 112 Sri Lanka Entra 21, 92, 116­17 actions taken by young people in, 225­28b experience returns and, 102 civil unrest, effects of, 9, 166, 186b formal apprenticeships, 111­12 demography of, 44b in informal sector, 102, 112 migration and mobility, 195 Joven(es) programs, 21, 66, 88, 92, 117­18, 117b, 215, 224 316 WORLD DEVELOPMENT REPORT 2007 training programs (continued) migration to. See migration and mobility opportunities for young people, improving, 13­14 nutritional deficiencies, 149 for out-of-school youth, 91­92 political violence in, 178 second chances, programs providing, 21, 116­18 residential vs. non-residential job training programs, 17, 56 technical and vocational specialization, 73, 74, 76, 112­13 smoking, long-term effects of, 122, 123f training the trainer, 17, 56, 76­77 teenage mothers, flexible school policies for, 156 transitions of young people. See life transitions for young people treatment programs and risk-taking behavior, 140 Tunisia, 73, 74, 99, 105 wage flexibility and youth employment levels, 121b Turkey, 12f, 89, 109, 148 Uruguay, 114, 117 tutoring, 89­90 Venezuela, República Bolivariana de, 102, 108, 117b, 174, 181b Uganda verlan, 173 civil unrest, rehabilitative programs for young combatants Vietnam involved in, 182b child care, 155 condom use in, 125 citizenship, exercising, 162, 163, 173 health education, 130, 132 civil unrest, effects of, 186b HIV infection rates in married women, 126 dependency decline in developing countries, 35f literacy rates, 182f economic growth and poverty reduction in, 94­95b, 106 Ukraine, 102, 189 migration and mobility, 201 Un Techo Para Chile, 25 policy priorities for, 213­16 unemployment rates among young people, 7, 8f, 39, 51 restrictive employment laws, effect on young people of, 106 education levels and, 120b rising levels of education in, 94b failure of young people to enter labor market, 98­101, 99f, road traffic accidents in, 32, 214 100b, 100f, 101b work and education, combining, 86 harm caused by, 120­21b violence. See also civil unrest; crime high unemployment and low marriage rates, relationship demographics and, 166b between, 40b, 42, 147 domestic violence, 145b lasting effects of unemployment on youth, 100­101 schools, violence and harassment in, 77b minority youth, 120b vocational and technical specialization, 73, 74, 76, 112­13 OECD countries, baby boom in, 120­21b voting by young people, 32f, 168­69, 169f, 196 public employment services, 117 vouchers, school, 87 second chance programs, 114­18 Vozes Jovens, Brazil, 221b young people neither employed nor studying, 100b United Kingdom apprenticeships, 111, 112 Weather Underground, 178 citizenship, exercising, 169 West Bank and Gaza, 4, 83 emergency contraception, effects of availability of, 140 women and girls, 65­66, 216b. See also family and parenthood; integrated education programs for disadvantaged youth, 89b marriage; prostitution migration and mobility, 198 age at birth of first child, 144 political violence in, 178 Better Life Options program, India, 16­17, 155­56 wage flexibility and youth employment levels, 121b birth control, 139, 149­50, 150f United Nations Convention on the Rights of the Child, 178 citizenship, exercising, 163, 164, 165, 165t, 170, 174, 175 United States decision-making capabilities of young people, 15­16, 15f actions taken by young people in, 225b domestic violence, 145b apprenticeships, 111 early childbearing, effects of, 28­29, 39b, 86 career academies, 76 early marriage, preventing, 153 childbearing outside of marriage, rates of, 145b education citizenship, exercising, 163, 169, 170, 171, 172, 174 fertility rates and participation in labor market related to, community college system, second chances provided by, 20, 62 86, 104 crime, juvenile, 177, 178, 180b incentive programs, 18, 82 economic restrictions on alcohol and tobacco access, 136 social barriers to, 83 gangs, 176 employment levels for, 101 Head Start program, 71 equity for, 28­29, 38­39b, 65­66 health and education, 130b fertility rates related to education and participation in labor ICT, global flow of, 205, 206 market, 86, 104 integrated education programs for disadvantaged youth, 89b gang protection, 176 Index 317 human trafficking, 196f second chances, focus on, 114­18, 117b, 118t income and resources, control of and access to, 17­19, secondary and tertiary education, relevance of, 12­13, 48­49, 57­58, 57b 73, 74­78, 84, 95b Internet, public access to, 202 self-employment. See self-employment and entrepreneurship life skills for, 155­56, 156f Sierra Leone, rebuilding lives and institutions in, 185­86b migration and mobility, 196, 201f social barriers to, 106 parental household, position in, 18­19, 147 training for. See training programs reproductive health services, access to, 138­40, 149­51 social barriers to, 83, 106 working opportunities for, 13, 51b, 106 Yemen, Republic of, 4, 65b, 210b working life, start of, 7, 7f, 22t, 96­119. See also unemployment young people in developing countries, 1­25. See also more specific rates among young people entries, e.g. equity and young people capabilities, focus on, 92t action to be taken by, 225­28b child labor and starting work too early in life, 7, 37, 50­51, 87, defining young people, 27b 96­98, 97f donors, role of, 209­12b citizenship, exercising, 174­75 economic growth and, 29­31 demand fluctuation for young vs. adult males, 103f evaluation of youth programs, 23 demographics affecting, 104, 120­21b five life transitions for. See life transitions for young people disabled youth, 115b government policies and reforms, voice in, 214­17, 219­23 economic growth, effects of, 106­7, 107f importance of investing in, 2­5 education and work, combining, 85­86, 86f, 97 indicators of youth development, 222b experience returns, 102f participation in future, encouraging, 22­23 government policies and reforms, effects of, 107­9, 108b, 121b policy foci for, 2, 10­11, 11f, 22t, 45­64, 45f (See also capabili- high youth employment rates in countries where youth must ties, focus on; opportunity, focus on; second chances, work, 100b focus on) ICT, global flow of, 203­4 poverty reduction and, 27­29, 50b income and resources for, 57 youth councils, 169­71 indicators of development in, 222b youth departments and ministries, role of, 215b, 219b informal sector, starting work in, 101­2 youth development and youth action programs, 175 Joven(es) programs, 21, 66, 88, 92, 117­18, 117b, 215, 224 Youth Employment Network, 210 lack of access to information and credit, 104 Youth programs in international financial institutions, 209b low-paying and dead-end jobs, 101­3 Yugoslavia, former measuring youth transition to work, 98b Otpor youth movement in, 167, 168b migration and mobility, effects of, 14b, 52, 104, 105f, 110­11, 192­93 opportunities for young people, improving, 13­14, 50­52, Zambia 106­11, 118t citizenship, exercising, 174 policy summary, 118­19, 118t condom use in, 125, 126 preparing youth for, 111­13 employer training, 112 restrictive employment laws, effect on young people of, 13, HIV infection rates among married women and girls, 145b 51­52, 104­9, 108b, 121b quality of basic education, 6b rural youth, opportunities for, 109­10, 110b Zimbabwe, 112, 134, 162, 170, 173b E C O - A U D I T Environmental Benefits Statement The World Bank is committed to · 83 trees saved preserving endangered forests and · 57 million BTUs of natural resources. We printed the energy saved World Development Report 2007 on · 3,865 pounds of 15 percent post-consumer recycled solid waste not paper, processed chlorine free. The generated World Bank has formally agreed to · 30,101 gallons of follow the recommended standards water saved for paper usage set by Green Press Initiative--a nonprofit program · 7,252 pounds less supporting publishers in using fiber greenhouse gases that is not sourced from endangered released into the forests. For more information, visit atmosphere www.greenpressinitiative.org.