DI RECTIONS I N DEVELOP M ENT Can the Inf uenc6P6tmcy?; Participatory. Poverty As,nies in the Developing*Wor{dc . l CAROLINE M. RO-B_B- < ,908 L Feb. 1998 5r DIRECTIONS IN DEVELOPMENT Can the Poor Influence Policy? Participatory Poverty Assessments in the Developing World Caroline M. Robb The World Bank Washington, D.C. (© 1999 The International Bank for Reconstruction and Development/ THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433 All rights reserved Manufactured in the United States of America First printing February 1999 The material in this publication is copyrighted. The World Bank encourages dissemination of its work and will normally grant permission promptly Permission to photocopy items for intemal or personal use, for the internal or personal use of specific clients, or for educational classroom use is granted by the World Bank, provided that the appropriate fee is paid directly to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, U.S.A., telephone 978-750-8400, fax 978-750-4470. Please contact the Copyright Clearance Center before photocopying items. For permission to reprint individual articles or chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978-750-4470. All other queries on rights and licenses should be addressed to the Office of the Publisher, World Bank, at the address above or faxed to 202-522-2422. Library of Congress Cataloging-in-Publication Data Robb, Caroline M., 1963-. Can the poor influence policy?: participatory poverty assessments in the developing world / Caroline M. Robb. p. cm. - (Directions in development) Includes bibliographical references. ISBN 0-8213-4144-8 1. Poor-Research-Developing countries. 2. Poor-Political activity. 3. Poverty-Research. 4. Poor-Attitudes. I. Title. II. Series: Directions in development (Washington, D.C.) HC59.72.P6 R6 1998 362.5'09172'4-dc2l 97-45176 CIP Contents Foreword v Acknowledgments viii Abbreviations and Acronyms x Summary xii 1. A Status Report 1 1.1 Introduction 1 1.2 What Is a Participatory Poverty Assessment? 4 1.3 How Are Participatory Poverty Assessments Conducted? 6 1.4 What Is the Current Status of PPAs? 15 2. Impact of the PPA 19 2.1 Deepening the Understanding of Poverty 22 2.2 Attitudes and Policy Change 26 2.3 Strengthening the Capacity to Deliver Policy 29 2.4 Determinants of the Level of Impact 31 3. Emerging Good Practice 37 3.1 At the World Bank: Initial Steps and Follow-up 37 3.2 At the Country Level: Listening to Policymakers and Opening Up Dialogue 47 3.3 At the Community Level: Including the Poor 55 The Challenge for Participatory Poverty Assessments 66 Bibliography 67 Annexes 73 Annex 1. Methodology 74 Annex 2. Impact 87 Annex 3. Poverty Assessments, Completed and Scheduled 101 Annex 4. Examples of PRA Exercises in Thailand (1998) and Zambia (1996) 103 Annex 5. Country Case Examples 104 Annex 6. Methodology of This Review 128 iii iv CAN THE POOR INFLUENCE POLICY? Tables, Figures, and Boxes Table la. Methodologies Used 6 Table lb. Time Spent in the Field 7 Table ic. Number of Communities Assessed 7 Table ld. Agency Conducting the Fieldwork 7 Table le. Cost 7 Table If. Year of Fieldwork 8 Table 2. Comparison of Participatory Methodologies 12 Table 3. Range of PPA Impacts 20 Table 4. Summary of PPA Impacts 32 Table 5. Summary of Impact Variables 33 Table Al. PPA Outputs and Applications 113 Figure 1. Good Practice Research Cycle 10 Figure 2. PPA Impact by Quality of Research Team 33 Figure 3. PPA Impact by Quality of PPA Manager and Level of Bank Management Support 34 Figure 4. PPA Impact by Link to Poverty Assessment 35 Figure 5. PPA Impact by World Bank and Government Ownership 35 Figure 6. PPA Impact by Level of Policymaker Involvement 36 Figure 7. Information Filters and Biases: Case Study of the PPA in Zambia 63 Box 1. Background to the World Bank's Participatory Poverty Assessment 4 Box 2. Participatory Poverty Monitoring in Zambia 9 Box 3. World Bank Participatory Poverty Assessments: Status Report 16 Box 4. Examp[es of Planned PPAs 17 Box 5. Enriching the Diagnosis of the Nature of Poverty 22 Box 6. Examples of Explanations Provided by PPAs 24 Box 7. Twelve Hallmarks of Good PPAs 38 Box 8. What Is Poverty Reduction? 39 Box 9. Factors for the World Bank to Consider to Increase the Impact of PPAs 45 Box 10. Handing Over the Document Does Not Equal Ownership 51 Box 11. Participation Is More Than Holding Workshops 52 Box 12. Where Is the Entry Point for Participatory Research Data? Considering the Institutional Framework 54 Box 13. Factors to Consider at the National Level to Increase the Impact of PPAs 54 Box 14. Factors to Consider at the Community Level to Increase Impact of the PPA 64 Foreword An understanding of the nature and causes of poverty lies at the heart A of designing economic and social strategies for development. Much of the analytic work on poverty critical to such an understanding has treated the poor as an object of inquiry: Empirical investigations have been conducted to explain outcomes for the poor in terms of their charac- teristics, the environment in which they live, and the policies of govern- ments and other agents toward them. This tradition of work has been critical to deepening our comprehension of poverty and of the options to alleviate it. There is another tradition of inquiry, however-one that seeks to understand the experience and causes of poverty from the perspective of the poor themselves. Investigations of the poor within this broad tra- dition include, for example, the work of anthropologists and others who have undertaken intensive studies of villages or poor urban areas spanning decades. In the context of development endeavors, a rela- tively recent component of this tradition involves the use of participa- tory techniques. (Although these techniques have often been linked to specific projects, they increasingly have been associated with broader diagnostic investigations of the nature and causes of poverty and of the potential for policy to make a difference.) A variety of techniques have been developed to support this participatory process. All have the aim of giving the poor a voice, a voice that is not distorted by the mind-set of the investigators. Typically, the techniques also have the objective of capturing the perspective of the poor in a way that can be communi- cated to decisionmakers in government and development agencies. Both aspects are important for the ultimate objective of empowering the poor. Poverty studies have become of critical importance to the World Bank in the past decade, since the reaffirmation of poverty reduction as its core purpose. Particularly in the wake of the 1990 World Development Report (WDR) on Poverty, the Bank has become one of the major agents v vi CAN THE POOR INFLUENCE POLICY? and supporters of the study of poverty, through both a series of country- specific poverty assessments and a wide range of other research. Within this experience, the Bank is probably best known for its use of tradi- tional household surveys, especially multipurpose surveys (such as the Living Standards Measurement Surveys) that use questionnaires to document a range of dimensions of household well-being. Indeed, the World Bank has sometimes been characterized as working exclusively with a consumption- or income-based definition of poverty. This has never been true (for example, the WDR 1990 placed considerable emphasis on the lack of health and education as dimensions, as well as causes, of poverty). However, it is true that most poverty assessments have identified the poor in terms of a poverty line, based on a country- specific assessment of the minimum consumption required to meet basic nutritional standards and to effectively participate in a society. The Bank is less well known for its increasing use of participatory techniques in both project and diagnostic work. The present study sur- veys one part of this trend: the use of participatory techniques in poverty assessment work. As Caroline M. Robb shows, their use rose significantly in the mid-1990s and has become common in poverty assessments conducted over the past three years or so. These partici- patory poverty assessments have already yielded rich results, some- times confirming and sometimes contradicting the conclusions of more traditional questionnaire-based national household surveys. They con- firm that the poor themselves see poverty as having many dimen- sions-including lack of material resources and ill health, but also including a vulnerability to adverse economic developments or, in some communities, to physical violence. The assessments provide insight into the nature of coping mechanisms, particularly the role of local networks (or social capital), and have the potential to provide telling information on the effectiveness-or ineffectiveness-of public and private institutions. This participatory work can, and should, also play a role in the design and ongoing evaluation of interventions. Participatory poverty work is expected to be of growing importance to the World Bank in diagnostic, policy, and project work. We already see this in some of the early assessments of the social aspects of the East Asian economic crisis. And while the 1990 World Development Report made limited use of the participatory tradition, one of the major stud- ies in the lead-up to the next WDR on poverty and development (which will be released in September 2000) combines new studies and a syn- thesis of participatory poverty analyses to present the perspective of the poor on the nature of poverty, trends in various dimensions of poverty, and the utility of formal and informal institutions that address the causes and conditions of poverty. Finally, we need to emphasize FOREWORD vii that traditional household surveys and participatory poverty work are fundamentally complements, not substitutes-and certainly not rivals. They mutually inform each other, to everyone's benefit. Recent Living Standards Measurements Surveys increasingly make use of subjective assessments of poverty, while other new studies make use of participa- tory and questionnaire-based approaches in a structured, complemen- tary way. Developing powerful and effective diagnoses of the causes of poverty, and appropriate treatments to reduce poverty, requires both well-designed quantitative investigation and giving a genuine voice to poor people. Gloria Davis Michael Walton Director, Social Development Director, Poverty Reduction and Chief Economist, Human Development Acknowledgments T'his book arose from discussions, meetings, and workshops with peo- 1 ple from a wide variety of organizations and communities, to all of whom I am very grateful. I particularly thank Robert Chambers (IDS) for his unfailing inspiration, encouragement, and direction. In addition, sub- stantial contributions were made by Kimberly Chung (Brown Univer- sity), John Gaventa (IDS), Jeremy Holland (University of Swansea), Andrew Norton (DFID), Ben Osuga (Swinga, Uganda), and Dan Owen (London School of Economics). Additional comments were provided by Nancy Alexander (Bread for the World Institute), Elizabeth Gomart (Con- sultant), Richard Holloway (Private Agencies Collaborating Together), Ramesh Singh (Action Aid, Vietnam), and Joachim Theis (Save the Chil- dren, Vietnam). I also thank the local research teams in case study countries who undertook the fieldwork and the civil society institutions, government agencies, and community members who gave their time so freely to help in this study: In Zambia, Clare Barkworth and Cosmas Mambo (Social Recovery Project, Zambia), Peggy Chibuye (World Bank, Resi- dent Mission), Silverio Chamuka, Helen Muchimba, Hope Kasese, Fan- well Kondolo, Mulasikwanda Liswaniso, Kwibisa Liywalii, John Milimo, Eddie Mwanza, Malako Nabanda, and Lizzie Peme (Participa- tory Assessment Group), and Fred Mutesa and Stephen Muyakwa (University of Zambia); in Costa Rica, Carmen Camacho (UNICEF), Betsy Murray (World Bank, Resident Mission), and Pablo Sauma (Min- istry of Economic Planning); in Pakistan, the Association for Develop- ment of Human Resources, Muhammad Ahsan Ashraf, Asif Farooki, and Parvez Tahir; and in Mozambique, Yussuf Adam (Universidade Eduardo Mondlane). At the Bank, Michael Walton, Gloria Davis, Ishrat Hussain, and Aubrey Williams sponsored the research and publication and guided the research. I particularly thank Michael Walton for providing me with detailed comments on draft copies of this book and for his leader- viii ACKNOWLEDGMENTS ix ship and valuable insights, which were instrumental to the production of this book. Soniya Carvalho, John Clark, Nora Dudwick, James Edger- ton, Paul Francis, and Jenny Rietbergen-McCracken provided substan- tial contributions at all stages of the research. The peer reviewers were Jeanine Braithwaite, Jesko Hentschel, and Peter Lanjouw. Comments on earlier drafts of the book were provided by Lionel Demery, Paula Donnelly-Roark, Kene Ezemenari, Christopher Gibbs, Bruce Harris, Eugene Henkel, Jack van Holst-Pelleken, Emmanual Jimenez, Steen Jorgensen, Mary Judd, Valerie Kozel, Alexandar Marc, Tim Marchant, David Marsden, Caroline Moser, Deepa Narayan, Adega Ouma, Vale- ria Pena, Gill Perkins, Nadine Poupart, Jacomina de Regt, Claude Salem, Larry Salmen, Roger Sullivan, Maurizio Tovo, Tosca Van Vijfeijken, and Frederick Wherry. Hank Chase, Connie Eysenck, Dan Kagan, Alan Kahan, Nicki Marrian, and Don Reisman, of the Bank's Office of the Publisher, contributed editorial, design, and production expertise. Finally, I thank Andrew Steer for his overall guidance and support. Abbreviations and Acronyms ADB Asian Development Bank AMREF African Medical and Research Foundation (Kenya) ASAFE Association pour la Promotion de la Femme Entrepreneur BA Beneficiary assessment CAS Country assistance strategy (World Bank) CII Composite impact index CEDEP Centre for Development of People (Ghana) CEM Country economic memorandum (World Bank) CEP/UEM Centro de Estudos da Populacdo, Universidade Eduardo Mondlane (Mozambique) CFA Communaute Financiere Africaine COD Country Operations Department (World Bank) CSO Central Statistics Office DANIDA Danish Agency for International Development DFID Department for International Development ENVSP Environmental and Social Policy Department (World Bank) EU European Union FAO Food and Agriculture Organization GTZ German Technical Cooperation HIES Household Income and Expenditure Survey IBRD International Bank for Reconstruction and Development IDF Institutional Development Fund (World Bank) IDS Institute of Development Studies LCMS Living conditions monitoring survey LIL Learning and innovation loan M&E Monitoring and evaluation NDS National development strategy (Swaziland) NGO Nongovernmental organization x ABBREVIATIONS AND ACRONYMS xi PA Poverty assessment PAG Participatory assessment group PAID Pan African Institute for Development PER Public expenditure review PIR Poverty and inequality report PME Participatory monitoring and evaluation PPA Participatory poverty assessment PPM Participatory poverty monitoring PPR Participatory policy research PRA Participatory rural appraisal PRMPO Poverty Reduction and Economic Management, Poverty Division PROINDER Programa de Iniciativas de Desarrollo Rural PSA Programa Social Agropecuario RDP Reconstruction and Development Program (South Africa) RRA Rapid rural appraisal SARAR Self-esteem, associative strength, resourcefulness, action planning, and responsibility SDS Secretarie Desarrollo Social (government poverty agency, Mexico) Sida Swedish International Development Authority SIEMPRO Sistema de Informaci6n, Monitoreo y Evaluaci6n de Programas Sociales SSI Semistructured interview TDRI Thailand Development Research Institute UNDP United Nations Development Programme UNICEF United Nations Children's Fund WMS Welfare Monitoring Survey (Kenya) Summary Participatory poverty assessments are showing the World Bank and other outside observers of poverty that we are not the only poverty experts. Poor people have a long-overlooked capacity to contribute to the analysis ofpoverty-and without their insights we know only part of the reality of poverty, its causes, and the survival strategies of the poor. ow can the poor, so removed from the powerful, influence national Hpolcy? For many years, poverty assessments have used income and consumption indicators, education levels, and health status to determine levels of poverty. Such data are derived from household surveys. Recently, poverty assessments have also begun using a new tool called a participa- tory poverty assessment (PPA) to sharpen the diagnosis of poverty. PPAs use participatory research methods to understand poverty from the per- spective of the poor. The method elicits both quantitative and qualitative data on broader indicators of poverty such as vulnerability, physical and social isolation, powerlessness, insecurity, and self-respect. As a result, a poverty assessment that uses the PPA research method gives the poor, marginalized, and excluded a voice in policymaking. PPAs are responding to the challenge of inclusion by directly pre- senting the views of the poor to policymakers, in country and in the World Bank. Although participatory approaches have been used by social scientists in project work for some time, their use for policy analy- sis is new. This new way to influence policy has been developed by the Bank in partnership with governments, nongovernmental organiza- tions (NGOs), academic institutions, and other donors. After five years and 43 PPAs, many lessons are emerging that broaden our under- standing of both poverty and the policy process. POVERTY: PPAs have consistently shown that poor people emphasize different dimensions of poverty than those typically used in policy analysis, including income and consumption levels, health, and educa- xii SUMMARY xiii tion status. The poor also emphasize such aspects as vulnerability, physical and social isolation, lack of security and self-respect, power- lessness, and lack of dignity. POLICY: Experience with PPAs indicates that where there is a broad policy dialogue on poverty that includes different civil society groups, the constituency for reform is widened, ownership is increased, and the resulting policy is more likely to be implemented. What Is a PPA? To strengthen the link between the Bank's assistance strategy and the country's own efforts to reduce poverty, the Bank is committed to com- pleting country-specific analyses of poverty in the form of poverty assessments. The core elements of such assessments are data on the income, consumption, education levels, and health status of the target group, usually based on the results of household surveys. In the past five years, 45 percent of the Bank's completed poverty assessments have also included a PPA. PPAs use participatory research methods to understand poverty from the perspective of the poor by focusing on their realities, needs, and priorities. Instead of a predetermined set of questions as used in household surveys, PPAs use a variety of flexible methods that com- bine both visual (mapping, matrices, diagrams) and verbal (open- ended interviews, discussion groups) techniques, with the objective of better defining the experience of individuals, groups, households, and communities. The principle of a PPA is to ensure that the intended beneficiaries have some control over the research process. Instead of information being extracted from an interviewee, communities share their knowl- edge and are involved in analyzing the results. The assumption is that poor people have expertise and should be part of the decisionmaking process. Experience from past PPAs has shown that the poor have the capacity to appraise, analyze, plan, and act to a far greater extent than has heretofore been acknowledged. Impact of PPAs Over the past few years, the percentage of PPAs in poverty assessments has increased. One-fifth of the Bank's poverty assessments completed in fiscal year 1994 included a PPA. By fiscal 95, this figure had risen to one-third, and in fiscal 96, fiscal 97, and fiscal 98, half the poverty assess- ments included a PPA. Out of the 43 PPAs completed to date, 28 were xiv CAN THE POOR INFLUENCE POLICY? in Africa, 6 in Latin America, 5 in Eastern Europe, and 4 in Asia. These PPAs have entailed a wide variety of approaches and have had a vari- ety of outcomes and impacts. This book proposes a threefold classifica- tion of PPAs based on their varying impacts-those that deepen our understanding of poverty, that influence policy, and that strengthen policy delivery. 1) Deepening our understanding of poverty PPAs are deepening our understanding of poverty by enabling the poor to highlight dimensions of poverty, explain the processes of impoverishment, and rank their priorities. The policy dialogue has been dominated by income and consumption measures and health and education status derived from traditional household surveys. PPAs are adding to this analysis by providing other insights on the nature of poverty from the point of view of the poor. VULNERABILITY: Vulnerable groups are not always identified in house- hold surveys. Neither is the fact that their access to productive resources might be constrained by political, cultural, and social factors. In Arme- nia, single pensioners were consistently ranked by the communities as the poorest-not because they had the least income but because they were isolated and socially excluded. In Togo, the PPA drew attention to vulnerable groups such as displaced people and domestic child labor. ASPECTS OF GENDER: In Tanzania, men identified transportation, farm- ing, and drunkenness as the three most important problems, whereas women identified food shortages, lack of clean water, and illness. CRIME AND VIOLENCE: Some PPAs have been able to highlight the rela- tionship between poverty and illegal activities. In contrast, household surveys often are not able to access such information because of the respondent's reluctance to answer questions from an interviewer she or he does not trust. PPAs have been able to access data on such sensitive topics as child prostitution (Zambia), drugs (Jamaica), and domestic violence (Mexico). The PPA in Ecuador found that street crime and vio- lence restrict women's ability to work away from home and that women and the elderly are reluctant to use public transport, particularly at night, because of safety concerns. SEASONALITY: Many of the PPAs, such as those in South Africa, Zam- bia, Ghana, and Togo, included a seasonality analysis that highlighted great differences in poverty, vulnerability, and coping strategies over the year. SUMMARY xv PPAs have helped in the interpretation of results from traditional household surveys. For example, the PPA in Mexico found that some women in Mexico City are unwilling to leave their houses and go to work. Because they do not have tenancy rights they are afraid that their houses might become occupied. In addition, the PPAs have made it clear that the poor can analyze the causes of their vulnerability and rank their priorities. As a result of the poor's involvement, the PPAs in Ghana, Mali, and Nigeria identified physical isolation and a lack of access to water as major problems. PPAs generally work with information at various levels-from indi- viduals, households, and communities-and study issues of gender, ethnicity, age, and the relationships and differences among various community groups. Some PPAs have focused on individual case stud- ies of people, providing insights into the dynamics of poverty and sur- vival strategies. At the household level, the focus on intrahousehold dynamics can reveal both the unequal allocation of resources among household members and the impact of power relations on the poverty of women, men, and children within the household. Most PPAs also adopt a community perspective to highlight the diversity of social or cultural groups and their wide-ranging coping strategies. 2) Influencing policy Evaluating the extent to which PPAs have influenced policy involves consideration of two main issues: first, has policy changed? and, sec- ond, have policymakers shifted their focus toward a more pro-poor approach? Although causality is usually difficult to establish, there are many examples of how PPAs have influenced policy at the country level and within the Bank, such as the following: * Zambia: The PPA identified the fact that school fees were to be paid at a time of year that caused maximum economic stress for house- holds. The Ministry of Education is now preparing a new regulation to change the timing of school fees. * Ghana: The PPA influenced the composition of the Bank's country program by shifting the emphasis to rural infrastructure and to the quality and accessibility of education and health care. 3) Strengthening policy implementation Finally, a participatory process can help build the capacity of institu- tions to implement a policy more effectively by creating incentives (political or otherwise) and by generating a new institutional alignment to achieve effective, sustainable poverty reduction. To move toward strengthening policy implementation, the PPA needs to be designed to xvi CAN THE POOR INFLUENCE POLICY? * Use participatory techniques to diagnose both the policy environ- ment and the ability and willingness of institutions to deliver the evolving policy; * Build the capacity of institutions to use participatory methods in the formulation and implementation of the policy; and * Initiate appropriate partnerships and linkages among and within formal and informal networks and institutions. PPAs have the potential to increase dialogue and negotiation on poverty at the policy level; increase ownership and commitment to pol- icy delivery on the part of different civil society groups; and strengthen links between communities and policymakers. Over the longer term, this process could chalfenge existing power relations. Although it has not been possible to fully assess the impact of the PPAs, most appear to have achieved the objective of data collection and analysis. Some have achieved the objective of capacity building, but only a few have affected the formulation and implementation of policy, which is necessary if they are to have a wider impact. It is important, at this stage, not to overstate what PPAs have delivered or can deliver. But, the approach does have the potential to affect communities by involving local people in the definition and analysis, including causes, of their own poverty; by helping people shift from passively being dependent to actively seeking ways to reduce their poverty; and by involving communities in policy formulation and delivery as opposed to them being merely acted upon. Emerging Good Practice There is no single model for this type of work. The best approach is often determined by the context. However, this book suggests some minimum standards and good practice for participatory policy research that aims to affect policy change. Emerging good practice at the World Bank includes wide ownership of the PPA across departments, as well as a team approach, the integra- tion and balancing of various sector interests, a commitment to poverty reduction, and management support. At the country level, the potential impact of PPAs on policy change is influenced by the degree of govern- ment support for the exercise, and more generally the level of owner- ship and commitment of in-country stakeholders, which affects the credibility of the analysis. At the community level, the quality, credibil- ity, and effectiveness of the PPA relate to when it is performed (before, during, or after the household survey), the methods used, the length of SUMMARY xvii time allocated for fieldwork, the skills of the researchers, and the degree of institutional linkage established through the fieldwork process. Ethical questions are raised in this new field of influencing policy through dialogue with the poor. In the past, participatory methodolo- gies were widely used at the project level, where there was immediate follow-up and action at the community level. Many practitioners are now questioning the process, principles, and ethics of working directly with communities for policy research where there may be no direct fol- low-up at the community level-the result being more data extraction than community action. All survey work, but especially PPAs, should discuss with participating groups the terms of the relationship. A basic principle is that the results of the PPAs should be shared with all the participating communities. Looking Ahead Diagnosis: Poverty analysis There has been a tendency to see a dichotomy between traditional household surveys, which are quantitative and objective, and PPAs, which are qualitative and subjective. In practice, however, these divi- sions are not as clear and are often misleading, since subjective ques- tions are increasingly being used in traditional surveys and many PPAs contain quantified information and analysis. The objective of a comprehensive poverty analysis, therefore, should be to conduct participatory research and household surveys interac- tively, so that they enhance each other. If a PPA is conducted after the household survey, the results will either explain, challenge, reinforce, or shed new light on household survey data. The results of the house- hold survey can also, of course, explain, challenge, or reinforce the PPA. In Armenia, for example, the PPA was conducted after the survey work and was able to illuminate areas not covered in the survey, such as reciprocity and kinship networks and the impacts of crime. If the PPA is conducted before the household survey, the PPA results could assist in generating hypotheses, shaping the design of the house- hold survey, and developing survey questions appropriate for the respondents. Ideally, this should be an ongoing process whereby both PPAs and household surveys are conducted periodically and feed into each other. The results of past PPAs indicate that when they are used in conjunction with household surveys, the final assessment is a much fuller analysis of the varying dimensions of poverty, and the policy rec- ommendations are more relevant and informed. xviii CAN THE POOR INFLUENCE POLICY? Evaluation: Involving the poor in measuring success PPAs have shown that poor people have the capacity to contribute to the debate on poverty. The question is, therefore, who should deter- mine indicators of success? In the past, such indicators have been defined by those outside the community. Whose values and whose reality count (see Chambers 1997 and Gaventa 1998) are key issues. Emerging from these questions is the further question of who deter- mines reality. To understand how projects and policies affect people's lives, investigations now focus on ways in which the poor can measure and assess outcomes (using indicators and values that make sense to them) and analyze causality. These approaches are increasingly being incorporated into World Bank projects. PPAs are highlighting the potentially powerful role the poor can play in analyzing poverty, developing interventions for its reduction, and assessing the impact of projects and policies. The challenge for the Bank and the rest of the development community is to effectively inte- grate the perspectives and values of the poor into the process of policy and project formulation and implementation. 1 A Status Report 1.1 Introduction Participatory poverty assessments (PPAs) are broadening our under- standing of both poverty and the policy process. The limitations of quantitative measurements of well-being have long been recognized, and there is a rich tradition of anthropological and sociological work that uses a range of techniques to achieve an in-depth understanding of poverty for project work. In this tradition, PPAs use a systematic participatory research process that directly involves the poor in defining the nature of poverty, with the objective of influencing policy. This process usually addresses both traditional concerns such as lack of income and public ser- vices and other dimensions such as vulnerability, isolation, lack of secu- rity and self-respect, and powerlessness. PPAs are also highlighting the fact that policy change involves more than writing statements of intent in a policy document. It requires an understanding of the unpredictable situation within which agenda set- ting, formulation, and implementation continuously overlap and pol- icy choices are made as outcomes of social processes. It also requires an understanding of how a broad-based dialogue with different people in society, including the poor, can help ensure that a policy will be imple- mented and sustained. PPAs have demonstrated the value of * Participatory policy research in the form of participatory problem identification, which includes the poor in the analysis of their own livelihoods using both qualitative and quantitative information; and * Participation in policymaking, which involves linking the informa- tion from participatory research into a broad policy dialogue among a cross-section of stakeholders, leading to increased awareness, atti- tude shifts, and changes in policy and the policy delivery frame- work. 1 2 CAN THE POOR INFLUENCE POLICY? PPAs are part of a trend within and beyond the Bank that is chal- lenging personal, professional, and institutional norms: On a personal level, the new approach is to learn from and listen to others; on a pro- fessional level, to appreciate that we are not the only experts and that many others can contribute to the debate on poverty and development; and on an institutional level, to change organizational culture, meth- ods, and values from top-down practices to adaptable approaches that embrace risk-taking and error. Context In the 1980s, the Bank's poverty-reduction objectives were often over- shadowed by the focus on economic adjustment to achieve macroeco- nomic stability and structural change as foundations for long-term growth. Toward the end of the decade, however, the Bank and other development agencies began to act to mitigate the consequences of eco- nomic and structural adjustment for the poor. For example, the Social Dimensions of Adjustment program, funded by several multilateral and bilateral agencies, was launched in November 1987 in response to their concern about the position of the poor in the structural adjust- ment process in Africa. The program included a strong focus on strengthening national information systems, though with little use of participatory research. The World Development Report 1990 (World Bank 1990), which focused on the issue of poverty, proposed a strategy for achieving more effective poverty reduction. That report was followed in 1991 by a pol- icy paper, Assistatice Strategies to Reduce Poverty (World Bank 1991), which laid out how the findings of the World Development Report could be used to strengthen poverty reduction efforts. The policy paper rec- ommended that a poverty assessment be conducted for each country, with the objective of analyzing the nature and causes of poverty and developing a strategy for poverty reduction. In the World Bank's process, the poverty assessment, which is done routinely for each coun- try, feeds into the country assistance strategy, which lays out the Bank's program of support for a country in relation to its development objec- tives and structural conditions (see World Bank 1992). Poverty assessments use a variety of sources to diagnose the struc- tural causes of poverty. Typically, a national household income or expenditure survey, or a multipurpose living standards measurement survey, is undertaken to provide basic information on the patterns of poverty. The early poverty assessments made little use of participatory techniques, and although they did employ a multidimensional concept of poverty, their principal criterion for defining who is poor was gen- A STATUS REPORT 3 erally consumption or income. This approach, however, has changed over the past decade, with increasing attention being paid to informa- tion from participatory research sources. Such information is generally used to complement, enhance, modify, or interpret conclusions derived from household survey analyses and other quantitative sources. Outside the Bank, there was also a growing realization of the impor- tance of including the poor in diagnosis and policy work. A variety of sources led to this shift from projects to policy dialogue. PPAs devel- oped in response to the broadening thinking on the multidimensional character of poverty associated with such publications as the Bulletin on Vulnerability (Institute of Development Studies 1989) and Putting the Last First (Chambers 1983). In the Bank, there was also ongoing project (as opposed to policy) work on understanding poverty and well-being through beneficiary assessments, participatory rural appraisals, devel- opmental anthropology approaches, and similar methods. The European donors (including Denmark, Norway, Sweden, and the United Kingdom, which support the PPAs through trust funds and operational funding) began to emphasize the social dimensions of poverty and provided funding for many of the Bank's PPAs in Africa.' In the Bank, development of the PPA was initially based on a series of papers by Clark (1992), Norton and Francis (1992), Salmen (1992a and b), and Clark and Salmen (1993). In addition, the Bank's Participation Learning Group (see World Bank 1994c) created a more receptive insti- tutional environment for participatory approaches in both project and policy work. Participatory approaches have been used by other institutions but not as extensively as by the World Bank. For example, as far back as 1990, the Gambian government and the United Nations Development Programme (UNDP) formulated the Strategy for Poverty Alleviation through a process of dialogue with a cross-section of groups in society, including poor communities throughout the country.2 The strategy provided an institutional framework whereby the poor could express their views on poverty.3 And in Bangladesh, the UNDP undertook a national participatory poverty study (UNDP 1996).4 The World Bank has adopted participatory research techniques on a broad basis in a variety of geographical regions and with a range of partners. This experience has enabled the Bank to understand the diverse causes and conditions of poverty and the processes that affect policy change. Annexes 1 and 25 analyze the methodologies and impacts of participatory assessments on a country-specific basis. The objective is to learn from the organizations that have been our partners in this exercise and to reflect on the process. 4 CAN THE POOR INFLUENCE POLICY? 1.2 What Is a Participatory Poverty Assessment? A PPA is typically one of many inputs into a poverty assessment (see Box 1). Unlike household surveys, which collect statistical data on the extent of poverty through standardized methods and rules, PPAs focus on processes and explanations of poverty as defined by individuals and communities within an evolving, flexible, and open framework. PPAs are sometimes referred to as qualitative surveys. This name can be confusing because there is a qualitative dimension to traditional survey work, and many PPAs contain quantified information and analysis. The terms "objective" for household surveys and "subjective" for PPAs may also be inaccurate. In household surveys, for example, interviewers and analysts will interpret informants' answers subjec- Box 1. Background to the World Bank's Participatory Poverty Assessment As a result of the World Development Report 1990 on poverty and the 1991 policy paper, Assistance Strategies to Reduce Poverty, the Bank is commit- ted to carrying out complete country-specific analyses of poverty in the form of poverty assessments. As of July 1998, 99 poverty assessments had been completed (see Annex 3). A majority (55 percent) of these were based on statistical assessments without participatory surveys. Each poverty assessment draws a poverty line based on the level of income or con- sumption associated with the minimum acceptable level of nutrition and other necessities of everyday life. People are considered poor if their income falls below this line (World Bank 1991). Poverty assessments gen- erally include an analysis of the depth and severity of poverty and are increasingly using multiple poverty lines. To date, 43 poverty assessments have included a PPA, which provides new dimensions in the analysis of poverty. Policy-focused research using participatory methods is undertaken to understand poverty from the perspective of the poor by focusing on their realities, needs, and priori- ties. Definitions of poverty, therefore, have moved beyond the conven- tional consumption and income indicators to broader issues such as vulnerability, physical and social isolation, powerlessDess, insecurity, and self-respect. The PPAs form part of the povelty assessment, which combines qualitative and quantitative data to achieve a better analysis of poverty. The inclusion of other stakeholders at different levels in the country is required to link the information from the PPAs to policyrmaking. In many countries, this inclusion has led to the creation of partnerships among the Bank, government, and civil society with the objective of reducing poverty. A STATUS REPORT 5 tively. The use of these terms can create the appearance of a dichotomy, while in the best poverty analysis the two merge into one integrated analysis (e.g., the World Bank's poverty assessments for Armenia and Zambia). Traditional survey data can be used to count, compare, and predict. The strength of the PPA is not in counting but rather in under- standing hidden dimensions of poverty and analyzing causality and processes by which people fall into and get out of poverty. Participatory research is undertaken by facilitators using a diverse set of participatory tools determined by the research agenda and local context. Enabling the poor to participate leads to a reversal in the rela- tionship between the community and the outsider that is implicit in tra- ditional surveys. Facilitators of participatory research need different skills and behavior, including listening to and respecting the expertise of participants, building trust, handing over control, and allowing the community to define the poverty issues that matter. The poor are viewed as participants or partners in the research process, data are shared with them, and the analysis of research results takes place within the community. The poor thus have more control over the research process, and their capacity to appraise, analyze, plan, and act is recog- nized. The extent and quality of participation have, however, varied exten- sively. Some PPAs have been criticized for limited participation, espe- cially when interviews were done quickly (less than two weeks of field research in some countries) and the results were not fed back to the communities. In other PPAs, the quality of the participation has been questioned. Although participatory research methods may have been used, some research teams adopted a dominant role, undermining par- ticipation and resulting more in data extraction. For example, the man- ager of the PPA in Ecuador judged that genuine participation was limited and renamed it the Rural Qualitative Survey. Secondary stakeholders (i.e., those beyond the community) have also participated in PPAs. Such stakeholders can include, for example, other donors (bilaterals, UNICEF); national and international NGOs (Save the Children, Oxfam); academic institutions; religious groups and leaders; different levels of government; and local leaders. Even some poverty assessments that did not include direct consultations with the poor were participatory in the sense that they consulted a cross-section of secondary stakeholders (e.g., Malawi). Although PPAs and anthropological research have some similari- ties, there are three main distinctions. First, PPAs provide a perspective from a cross-section of communities in different areas of a country, whereas anthropological research usually analyzes one or two com- munities in depth. Second, PPAs tend to focus on messages for policy. 6 CAN THE POOR INFLUENCE POLICY? Third, PPAs provide a rapid overview of the current situation, which is quickly presented to the policymakers. Anthropological research usu- ally takes longer and focuses more deeply on processes within com- munities, often without a policy focus. In summary, PPAs have been used to provide clearer insight into the perceptions of the poor on the key issues related to poverty reduction (Norton and Stephens 1995). They are contributing to a greater under- standing of the processes by which people fall into and get out of poverty, the complex coping and survival strategies adopted by the poor, and the major priorities and solutions identified by the poor, all within a local or regional context. By combining the PPA with the house- hold survey information, the final poverty assessment is able to more fully analyze the various dimensions of poverty and make more informed and appropriate policy recommendations. 1.3 How Are Participatory Poverty Assessments Conducted? Factors that influence the approach and consequent outcome of PPAs include political context, support, and commitment, both in country and within the Bank; relations between the Bank and the governments; and levels of expertise. Thus, there is a wide range of experiences among the PPAs undertaken to date (see Annex 1 for details of the tim- ing, research teams, institutions involved, and methods used). Tables la-if below summarize the experiences of some of the PPAs. The methodologies in Table la are described in detail starting on page 11. PPA experiences at a glance Table la. Methodologies Used Methodologies used Number of PPAs* % Rapid rural appraisal 13 27 Participatory rural appraisal 15 29 SARAR** 2 8 Beneficiary assessment 9 19 Semistructured interviews and focus groups 8 17 * The numbers add up to more than 43 because some PPAs used more than one method. ** Self-esteem, associative strength, resourcefulness, action planning, and responsibility. A STATUS REPORT 7 Table lb. Time Spent in the Field Length of time in thefield Number of PPAs % 1-2 weeks 3 8 2-4 weeks 8 20 1-2 months 3 8 2-4 months 15 37 4-8 months 11 27 * Where data are available. Table lc. Number of Communities Assessed Number ofcommunities involved Number ofPPAs* % 1-9 6 26 10-24 6 26 25-49 7 30 50-74 2 9 75-100 2 9 * Where data are available. Table ld. Agency Conducting the Fieldwork Agency conducting thefieldwork Number of PPAs* % Local NGO** 8 18 International NGO 8 18 Academic institution 18 42 Government agency 5 11 Independent consultants and firms 5 11 * The numbers add up to more than 43 because some PPAs used more than one type of agency. ** Nongovernmental organization. Table le. Cost Cost (In thousand $) Number of PPAs* % 4 2 10 5-24 2 10 25-49 3 15 50-99 9 45 100-150 4 20 * Where data are available. 8 CAN THE POOR INFLUENCE POLICY? Table If. Year of Fieldwork Yearfieldwork was conducted Number of PPAs* % 1993 7 14 1994 9 18 1995 11 22 1996 5 10 1997 5 10 Ongoing and planned 13 26 * Where data are available. The discussion below focuses on three main issues to be considered when conducting participatory policy research: sequencing and dura- tion, research teams, and methodologies. 1.3.1 Sequencing and duration Some PPAs have been conducted before the household survey and oth- ers afterward. Each data set can inform the other, so the sequencing will be determined by the context in country. If the PPA comes first, its results can help focus the research agenda for the quantitative survey and generate hypotheses. In Armenia, for example, the results of the PPA were used in designing the survey. When PPAs have been con- ducted after the survey, they have been used to explain the results. For example, in Mali the household survey showed what seemed to be a disproportionate amount of money spent on clothing. The PPA found that clothing and cloth were considered investment items as well as sta- tus symbols. Conversely, the results of quantitative surveys can be used to identify the poorest geographical areas on which participatory research should focus. Emerging good practice suggests that the ideal situation is to have an iterative process, as is being developed in Zam- bia (see Box 2 and Figure 1). In Uganda, Togo, Benin, and Mali, short and rapid surveys were undertaken for three to four weeks.6 Methods were based on rapid rural appraisals (RRAs), so feedback to communities was limited. In Togo, time constraints were placed on the field workers by the World Bank's internal deadlines. Some results were, consequently, not disag- gregated by gender and the final report was not written in a way that could be easily understood by policymakers. In some more recent PPAs, such as in Cameroon, the lack of time for community-level analysis meant that some results were too generic. A balance needs to be achieved between quick fieldwork (which leads to less costly and more timely policy messages) and longer, more expensive fieldwork such as household surveys (which can cost up to A STATUS REPORT 9 Box 2. Participatory Poverty Monitoring in Zambia Background Using the same approach developed in the PPA, participatory poverty moni- toring (PPM) has been undertaken in Zambia on a yearly basis (1995, 1996, and 1997) since the completion of the first PPA. The monitoring was con- ducted by the Participatory Assessment Group (PAG), the NGO involved in the PPA. The objective was to monitor changes in poverty over time. Overall, it is evident that the PPA and the PPMs have made a considerable impact and contributed in a meaningful manner to the national policy agenda on poverty. The critical interest in the PPMs and their continuing contribution to policy dialogue lie in their empirical observation and elucidation of trends and changes in livelihood conditions in Zambia. Two areas of PAG's work will require continual reinforcement. Method- ological skills need regular refreshing and upgrading through periodic train- ing. The methodological approach requires repeated investigation of key policy areas using similar research techniques. What is required for succes- sive PPMs to have additive value is consistent innovation in the use of research methods by the research team. The second area that needs continual attention is the dissemination of findings, which involves identifying more precisely the clients for different types of PPM outputs and tailoring specific recommendations to those clients. An improved dissemination strategy is a priority. Initiatives might include local dissemination workshops, condensed reports for NGOs and other local institutions, and networking with other agencies and research institutes. Linkages and impacts The PPMs are not simply a tool for enriching the understanding of poverty in Zambia. They are also an important means of improving participatory plan- ning in the provinces and districts by closing the information loops at those levels. PAG's efforts (in dialogue, participation, and feedback) have been increasingly concentrated at the decentralized level and are well suited to ongoing decentralization efforts. Link with the living conditions monitoring survey There is still much informal discussion about linking the monitoring systems of the PPM and the living conditions monitoring survey (LCMS). The latest proposal suggests a quarterly meeting of a technical committee (comprising PAG and LCMS), with a rotating chair informing each institution of the other's ongoing and planned work. As far as harmonizing work programs, one problem identified was the dif- ference in project cycles of the LCMS survey (at least one-and-a-half years) and the shorter cycle of the PPM. The timing of survey cycles appears to be the only major hurdle to partnership, since PAG and the Living Conditions Monitoring Unit are housed in the same complex at the Central Statistics Office, making it feasible, at least in practical terms, to harmonize their work programs. Source: Based on a note prepared by D. Owen forfield researchfor this study. 10 CAN THE POOR INFLUENCE POLICY? Figure 1. Good Practice Research Cycle First PPA conduded PPA results create hypotheses for testing through household survey Next PPA explains results of and influence survey design. household survey, tests robustness of-survey results, generates new hypotheses for Household survey testing, and influences the >ondurved design of the next survey. conducted d 1( 1 ~~~~~~~~~~Results of household survey test Net PPA conduded | robustness of PPA results and F~~ ~~ influence research agenda and selection of next PPA sites._ $1 million and take up to three years). PPA research teams have spent from one day to one week in a given community and have visited from 4 to 98 communities. Urban areas are more complex, and thus more time and flexibility are needed, since it is difficult to predict the nature of participation. Total time in the field for a PPA has ranged from one week to eight months, depending on the sample size and the number of research teams. 1.3.2 Research teams In Eastern Europe, most of the research was conducted by individuals from local universities. In other countries, NGOs undertook the field research (e.g., Centre for Development of People (CEDEP) in Ghana, CARE in Cameroon, African Medical and Research Foundation (AMREF) in Kenya, Red Cross in Lesotho, Save the Children in Mali). International agencies have also been involved in the research process (UNDP in Togo, UNICEF in Lesotho). In South Africa, a local consult- ing company worked alongside a cross-section of NGOs, whereas in Mozambique and Zambia, local universities were involved. In Latin America, the community-level research was conducted by a cross- section of NGOs, universities, and government departments (e.g., the government poverty agency in Mexico). Some PPAs have used teams experienced in participatory research, as in Zambia, where the research team was given additional training in A STATUS REPORT 11 participatory rural appraisal (PRA) methods for the PPA exercise. Other PPAs have used local teams trained to conduct the research, or have tapped into the country's NGO and consulting firm networks (South Africa). In Ghana, the team was composed of a cross-section of indi- viduals from NGOs, government line ministries, and academia. 1.3.3 Methodologies There is a widening debate about the most appropriate methods to use when conducting participatory policy research. Below is a brief descrip- tion of the main methodologies used7 (see Table 2). In reality, these methodologies are complementary and can be used together. Refer- ences are given for more in-depth information. What is a beneficiary assessment? Many of the early PPAs were undertaken using a methodology called beneficiary assessment (BA), originally developed by the Bank in the early 1980s for use in the urban slums of Latin America. It was one of the methodologies that pioneered the inclusion of the voice of the poor in Bank operations. BAs draw from consumer research, traditional qualitative social science research, anthropological participant obser- vation (observing people and interacting with them in their environ- ments), conversational interviews, focus group interviews, institutional assessments, and investigative journalism. A BA is designed in consultation with policymakers and others who will use the information. Teams of researchers collect information in selected communities through focus groups and individual interviews. A semistructured interview guide is drafted before the research begins. Information is collected mainly through dialogue between beneficia- ries and researchers. The researchers then analyze the collected infor- mation-unlike in a PRA, where some of the analysis is done at the community level (see Salmen 1995a and 1995b for more details). What are rapid and participatory rural appraisals? Many PPAs have used the RRA methodology, which emerged in the 1970s. Its purpose was to develop an approach that would enable out- siders to learn about rural conditions and people's realities quickly and cost-effectively. In the mid-1980s, RRAs evolved into the PRA approach, which placed greater emphasis on community participation. RRAs and PRAs use tools such as mapping; diagrams of changes, trends, and linkages; matrices; and scoring. They also use group ani- mation and exercises to facilitate information sharing, analysis, and action among stakeholders. The information is thereby made visible, Table 2. Comparison of Participatory Methodologies Rapid Rural Participatory Beneficiary Participatory Monitoring Participatory Appraisals Rural Appraisals Assessments and Evaluation Policy Researcht (RRAs) (PRAs) (BAs) (PME) (PPR) When 1970 Late 1980s 1980 1990 1990 Where Universities NGOs World Bank NGOs NGOs, universities, World Bank, governments, donors Objective Data collection Community Data collection for Understanding impact Data collection to for projects empowerment project managers influence policy Focus Project Project Project Project Policy Main actors Outsiders Local people Outsiders Local people Local people and outsiders Key techniques Visuals Visuals Conversational Combination of methods; Combination of methods; interviews e.g., RRA, PRA, BA, SARAR e.g., RRA, PRA, BA, SARAR Outcomes Plans, projects, Sustainable local Better-informed Assessment of Better-informed policymakers publications action and project mangers project process institutions Main innovation Methods Behavior Listening to the Local people's contribution Linking local people to the people to determining indicators national policy dialogue of success Key resource Local people's Local people's Local people's Local people's Local people's knowledge for earlier overlooked knowledge capabilities knowledge perceptions on impact a better understanding of the problem and local people's capability to analyze policy impact Notes: NGO = nongovernmental organization SARAR = self-esteem, associative strength, resourcefulness, action planning, and responsibility Source: Adapted from Chambers (1997). A STATUS REPORT 13 which often creates ownership. The power of the PRA is frequentlv in "group-visual synergy" (Chambers 1997), with analysis being locally led. The main differences between BA and PRA is that PRA combines both verbal and visual techniques and emphasizes community-level analysis, whereas the BA emphasizes verbal techniques and most of the analysis is done by the interviewer. The PRA is also a set of principles that includes follow-up actions, embracing error, showing respect, being willing to unlearn assump- tions and conditioned responses (reversals in learning), using methods or processes only if they make sense in the context (optimal ignorance), compensating for biases, and triangulating data. As Chambers (1997) has noted, "PRA stresses changes in the behavior and attitudes of out- siders, to become not teachers but facilitators, not lecturers but listen- ers and learners." (Also see International Institute of Environment and Development 1991-1998.) What is SARAR? Self-esteem, associative strength, resourcefulness, action planning, and responsibility (SARAR) uses visual aids to stimulate discussions. These visuals are prepared in advance by the researchers (unlike the PRA, in which the visuals are created by the communities to express issues and concerns). The main objectives of the SARAR are to build local capac- ity to plan for community development or to raise awareness of health and sanitation issues. SARAR builds on local knowledge and strength- ens local capacity through a variety of participatory methods. It has also been used by development agencies to increase participation and joint decisionmaking, although it is not often used in PPAs (see Srini- vasan 1990). Use of methodologies There are many different participatory traditions from around the world: some provide the philosophy for participation, others provide the tools, and some provide both. PRA is one of the few that provides a broad philosophy in addition to distinctive tools. The selection of methodologies and tools depends on the context of the PPA (e.g., capac- ity of in-country institutions, PPA manager's knowledge of different methods, government approval, availability of skilled trainers, time available). The tools and approaches can be very different, and all have advan- tages and drawbacks. For example, PRA enables some of the analysis to take place at the community level, leading to greater ownership of the results. A researcher from Zambia, where a PRA was undertaken, stated that community ownership meant that "problems would be 14 CAN THE POOR INFLUENCE POLICY? thought about long after my departure."8 To further promote this own- ership in Zambia, charts and papers created by local people were left with the community. PRA places more emphasis on community-level interviewing, while BA concentrates on households or individuals (Norton and Stephens 1995) and involves less community ownership and control over the analysis and results. Some have argued that the visual tools of PEA might not be suitable for all cultures. Although this statement might be true to some extent, the skill and sensitivity of the facilitator and the understanding that he or she has of the community usually determine the extent to which visual tools will be appropriate. PRAs have been conducted effectively in a diverse range of cultures in more than 100 countries.9 How these methodologies relate to policy work These methodologies were not originally designed to influence pol- icy-they were developed specifically for communities and project work. BAs were traditionally used to seek the views of beneficiaries on the impact of projects and to feed this information back to project man- agers in an attempt to influence project design. SARARs and PRAs were used at the community level to develop community action plans with the wider objective of empowerment. In the 1990s, participatory methods have been used to achieve the broader objective of influencing policy. Sector assessments have used participatory research to influence policy in the following areas: health and education in Zambia (work done by the NGO, Participatory Assess- ment Group; Milimo 1996); urban poverty and violence in Jamaica (Moser and Holland 1996); and wetlands management in India and Pakistan (Gujja, Pimbert, and Shah 1996). Whereas PPAs attempt to influence the broader policy framework, sector assessments attempt to influence specific policies. In this new field of influencing policy through dialogue with the poor, ethical questions are being raised about the possible exploitation involved in using the poor to gain access to information without any benefit to them. When participatory methodologies were widely used at the project level, they comprised tools for gaining information and a set of principles such as action follow-up, empowerment, and capacity- building in the community. When participatory methodologies are used for policy work, however, these principles have often not been fol- lowed. It is suggested that when undertaking participatory research for policy work, the term participatory policy research (PPR) might be more appropriate. The debate has evolved because many PRA practitioners have questioned the process, principles, and ethics of working directly with communities for policy research when there is no direct follow-up A STATUS REPORT 15 at the community level-the result being more data extraction than community action. PPR uses tools from various methodologies but with a different overall objective: the creation of policy messages with communities contributing to the analysis, as opposed to direct action, community empowerment, and capacity-building. PPR, therefore, is not generally a tool for empowerment (Chambers 1997), and while its research value is great, its value at the community level should not be overstated. For policy, the participatory research is meant to be imperfect, rapid, and restricted, and the principle of imme- diate action may not be feasible because the focus is on trends, not proj- ect identification. PPR is a way to inform policy rather than empower local people. In an attempt to respond to the principle of follow-up action, however, some PPAs have linked the information with action- oriented institutions. For example, in Argentina and Brazil the field work has been linked with the work of country NGOs and government line ministries. As a result, the potential now exists for moving from information sharing to continuous dialogue with various stakeholders, including those at the community level. 1.4 What Is the Current Status of PPAs? As of July 1998, 43 PPAs had been undertaken at the World Bank. The fraction of poverty assessments including a PPA has risen from one- fifth in fiscal 94 to one-third in fiscal 95 and one-half in fiscal 96, fiscal 97, and fiscal 98.10 Of the PPAs completed to date, 28 are in Africa, 6 in Latin America, 5 in Eastern Europe, and 4 in Asia. Box 3 shows the dis- tribution of the various participatory methodologies employed, by region. Box 4 details some of the PPAs planned by the Bank and other orga- nizations. 16 CAN THE POOR INFLUENCE POLICY? Box 3. World Bank Participatory Poverty Assessments: Status Report AFRICA EASTERN EUROPE Benin RRA Albania Various Burkina Faso PRA Armenia Various Burundi PRA Azerbaijan Various Cameroon BA Moldova Various Central African Republic RRA Ukraine Various Chad RRA Djibouti PRA Equatorial Guinea RRA Eritrea RRA LATIN AMERICA Ethiopia PRA Argentina BA Gabon RRA Brazil BA Ghana PRA Costa Rica BA Guinea RRA Ecuador PRA Kenya PRA/SARAR Guatemala BA Lesotho PRA Mexico BA Madagascar BA Mali RRA Mauritius RRA Mozambique PRA ASIA Niger RRA India PRA Nigeria PRA Mongolia Various Rwanda RRA Pakistan PRA/Various South Africa PRA Papua New Swaziland PRA/BA Guinea Various Tanzania PRA/SARAR Togo RRA Uganda RRA Zambia PRA/BA Notes: RRA = rapid rural appraisal PRA = participatory rural appraisal BA = beneficiary assessment SARAR = self-esteem, associative strength, resourcefulness, action planning, and responsibility Various = Variety of qualitative research methods were used including open- ended interviews,focus groups, and semistructured interviews A STATUS REPORT 17 Box 4. Examples of Planned PPAs Where Description WMen Organization Argentina Urban poverty To be determined. World Bank assessment. Brazil Urban poverty To be determined. World Bank assessment. Egypt PPA. Ongoing. DFID India In Uttar Pradesh Ongoing. World Bank and Bihar. India Socioeconomic survey of Commenced in DANIDA 40 villages. November 1998. India Urban poverty Commenced in DFID assessment. January 1999. Indonesia Urban assessment of the To be determined. World Bank impacts of the financial crisis. Nepal Situation analysis. Commencing in early Action Aid 1999. Pakistan PPA. Ongoing. DFID Philippines Poverty assessment. Discussions on terms of World Bank reference started in September 1998. Date for participatory analysis to be determined. Sri Lanka Poverty study. Pilot in 20 Pilot commenced World Bank communities in various November 1998. agricultural zones. Completion March 1999. National study will start in January 1999. Thailand Assessment of the Commenced in World Bank social impacts of the September 1998. with TDRI financial crisis. Fifty communities. Thailand, Social impacts of the Commenced at end ADB Philippines, financial crisis. of 1998. Indonesia Uganda PPA. Ongoing. DFID/UNDP/ World Bank United Poverty study. Commenced in Oxfam UK Kingdom September 1998. Vietnam Ongoing assessment to Preparation has World Bank with be completed in seven commenced. Three NCOs (Action provinces. provinces will be com- Aid, Save the pleted by March 1999. Children) Notes: DFID = Department for International Development PPA = participatory poverty assessment LINDP = United Nation's Development Programme TDRI = Thailand Development Research Institute DANIDA = Danish Agencyfor International Development NGO = nongovernmental organization ADB = Asian Development Bank 18 CAN THE POOR INFLUENCE POLICY? Notes 1. The Department for International Development (DFID) - United King- dom seconded a Social Development Advisor (Andrew Norton) to the Bank to work on the development of participatory poverty assessments in Africa from 1992 to 1994. 2. See Robb 1995 and Government of the Gambia, 1994 for more details on this case. 3. In the Gambia, Action Aid and a nongovernmental organization (NGO) coordinating body assisted in organizing the participatory research on poverty using participatory rural appraisal techniques. In addition, a local team con- ducted research to gain an understanding of the informal networks within communities and throughout the country. Initially the policy environment was constrained, with the govemment unwilling to discuss poverty openly. As the dialogue gradually developed, more stakeholders were included until enough policy space was created to put poverty and related issues, such as decentral- ization and gender inequalities, on the political agenda. This process of con- sultation led to increased donor coordination and created an opportunity for the government and NGOs to redefine their heretofore controversial relation- ship. 4. Holland and Blackburn (1998) state that in the poverty study for Bangladesh, new issues were put on the policy agenda, such as the problem that demands for increasingly high dowry payments led to daughters being a burden to their parents and that wives were divorced or abused if the dowry was not paid. Furthermore, if daughters were educated and did not find a job, the demand for a dowry could increase. As a result, some parents were not sending their daughters to school. The study found that throughout Bangladesh, a priority for the poor was the enforcement of antidowry laws. 5. Annex 1 analyzes the various methodological and organizational issues associated with each of the PPAs. Annex 2 focuses on the value added of the PPAs and the impact on the Bank's and borrower's country programs and poli- cies. 6. Although there were many limitations to these early PPAs, they are sig- nificant for having been the first Bank studies to use participatory research methods in poverty analysis. 7. The information in this section comes from a variety of sources but is based mainly on Rietbergen-McCracken and Narayan (1997). 8. Malako Nabanda, Participatory Assessment Group (NGO), Zambia. 9. Including developed countries: For example, PRA is now widespread throughout the United Kingdom. See Inglis and Guy (1996). 10. See Annex 3 for a detailed breakdown of all poverty assessments com- pleted by the Bank to date. 2 Impact of the PPA I ncluding the poor in policy dialogue has great potential for creating bet- ter poverty reduction policies. The original rationale of the participa- tory poverty assessments (PPAs) was to influence the policy dialogue by collecting information on the poor's perceptions of poverty. Most PPAs have achieved this objective to some degree, but with substantial varia- tion in the level of impact. The PPAs with the greatest impact tended to be those that implicitly or explicitly had more ambitious objectives. It is use- ful to assess impact in relation to three objectives: *D eepening the understanding of poverty: Through incorporation of the results of participatory techniques into diagnosis of the nature and causes of poverty. * Influencing attitudes and policy: Through the use of the PPA process within a broader participatory process that engages policymakers. * Strengthening the policy delivery framework: Through creating a new institutional alignment that increases policy impact for effective, sustainable poverty reduction. Although the principal objectives of PPAs have been to diagnose the causes and nature of poverty and to influence policy, some PPAs have been successful in fostering dialogue with and building the capacity of credible poverty reduction institutions, which then create links between traditional and formal institutions. These links have created room for a more coordinated approach to poverty reduction among various stake- holders, including donors, with ongoing and increasing interaction between policy change and stakeholder dialogue. This process is long, slow, and continuous and requires the redefinition of stakeholder rela- tionships, including relationships with the World Bank. Ideally, gov- ernments should lead the process, or lead in partnership with other 19 20 CAN THE POOR INFLUENCE POLICY? institutions, and development partners should offer support and advice. This policy change and institutional strengthening at all levels is part of a wider process of establishing linkages between the poor and those in power. This section uses case examples to explore the diverse array of observed impacts of the PPAs in the context of the three categories mentioned above: the extent to which PPAs have deepened the under- standing of poverty; their impact on attitudes and consequent policy change; and the extent to which frameworks for policy implementation have been strengthened. These impacts are summarized in Table 3, which also links the various levels of impact to the different PPA approaches and the required shifts in the thinking of policymakers. Table 3. Range of PPA Impacts Required shift Typical approach in the thinking PPA impact and inputs Potential outcomes of policymakers 1. Deepening the * Rapid appraisals in * Changes in policy * Poverty viewed as a understanding of the field (e.g., 3 documents reflecting multidimensional poverty weeks). Information views of the poor. phenomenon, the extraction. Limited - The information character of which is Participatory feedback and action might be accurate defined by the com- data at the community and interesting but munity. incorporated in level. there is limited room * Policymakers under- the analysis of * Policymakers not for government own- stand the value of poverty. necessarily included ership or for chang- participatory in the process. ing attitudes of research and of * Prescriptive and policymakers. including the percep- often top-down in v The poor are given a tions of the poor. nature. voice but there is * Isolated exercises limited commitment with limited impact rsrm the top to on the wider devel- poor'shoncrn opment process. poor's concers agenda. * Issues such as power, decentralization, and gender are considered but not always included in ongoing debate. * Policymakers might feel threatened. 2. Influencing * Feedback and * Attitudinal shifts of * Policymakers are policy follow-up to field key stakeholders are seen as partners who appraisals (e.g., the reflected in policy should be included Policies realigned poor validate infor- changes. Policies are from the beginning toward poverty on mation). Ownership refocused toward of the planning a long-term basis of the information at poverty. process. [MPACT OF THE PP4 21 Required shift Typical approach in the thinking PPA impact and inputs Potential outcomes of policymakers by focusing on the community level. * More politically sen- * World Bank is seen changing Development of sitive issues such as as one of many attitudes. action plans and power, decentraliza- stakeholders. follow-up. Longer tion, and gender are * Policy change is process (e.g., 1 year). put on the agenda viewed as part of a -* Government involve- for continuous wider social process. ment from the begin- negotiaton. ning. Administrators * Govermnent and those who ownership and com- implement polcy are mitment are high. included in the debate. Redefinition of the relationships among stakeholders. Emphasis on build- ing partnerships and trust. Increasing coordination and conflict resolution through consensus building. 3. Linking * A continuous process * Strengthening the * Policy agenda formulation of cross-checking policy delivery setting, decision- with policy and dialogue. framework by build- making, and implementation * Identification of cred- ing the capacity of implementation ible institutions for appropriate institu- are interrelated Policy delivery tions, both formal processes. Recogni- framework capacity building, and traditional. tion that policy strengthened. tonships between * Those who imple- change does not formal and informal ment policy are not automatically mean institutions. Aware- just included in the policy implementa- ness of traditional debate but their a and tween management prac- capacity is also discourse and tices. increased. outcomes. * Organizational * Cross-stakeholder development and ownership and * Paricipaton IS seen institutional change. commnitment. on or a componenta * Ongoing participa- * Increasing trans- It is viewed as an tory monitoring and parency and approach within evaluation of accountability. which an overall poverty. * Building instituions framewvork is created at micro level for more effechve contributes to policy formulation decentralization. and implementation. Beginning to * The process of policy challenge existing change is part of a power relations (con- wider process of trol by elites, pat- estabhshing linkages ronage, exclusion of between the poor the poor). and those in power 22 CAN THE POOR INFLUENCE POLICY? 2.1 Deepening the Understanding of Poverty PPAs are deepening our understanding of poverty and contributing to a more in-depth analysis of this complex problem. PPAs are beginning to provide insights into * Dimensions of poverty, * The causes and dynamics of poverty, * Priorities of the poor, and * Different levels of analysis. 2.1.1 Dimensions of poverty It is well known that poverty has many dimensions beyond income and consumption. However, policy dialogue has focused primarily on income and consumption measures of poverty, while other dimensions highlighted in the PPAs have been underemphasized in the policy debate (see Box 5). Box 5. Enriching the Diagnosis of the Nature of Poverty Seasonality In many PPAs, seasonality analysis highlighted great differences in poverty, vul- nerability, and coping strategies throughout the year. In South Africa, for exam- ple, the PPA revealed that payment of school fees coincided with a season of financial stress resulting from a high incidence of sickness and hard work com- bined with shortages of money and food. The household survey in Tanzania concluded that 22 percent of the poor had access to safe water from protected sources, indoor plumbing, standpipes, and covered wells with hand pumps. But the survey overlooked the seasonal dimension of access to safe water and there- fore overestimated the access. The PPA, which collected information from the same viUlages, revealed that in two-thirds of the villages thought to have access to safe water, water was actually a major problem. In the dry season, as water tables fell, people were forced to walk further for water or switch to unsafe alter- natives such as uncovered dug wells, ponds, streams, and rivers. Gender Some PPA data enable analysts to understand gender dimensions of poverty that seem to be masked in survey data. The Zambian PPA was able to distin- guish different kinds of female-headed households. "Womxen without sup- IMPACT OF THE PPA 23 port," as opposed to female-headed households, were identified as the poorer group. In Tanzania, men identified transportation, farming, and drunkenness as the three top-ranked problems, whereas women identified food shortages, water, and health problems. In northern Mexico, the PPA found that it was eas- ier for women than men to obtain jobs. This situation challenged the traditional gender roles as many men found themselves out of work. Conflict within the household had become a major issue. Similarly, in Mexico City, the urban poor, especially the men, felt excluded from job opportunities. Some men were turn- ing to alcohol. Women were left with the double burden of earning income and rearing children, which put pressure on traditional gender roles and fueled the increase in domestic violence. PPAs also try to capture informal sector and non- remunerated activities, which in many cases fill a major part of women's time. Vulnerable groups Vulnerable groups are not always identified in household surveys. Neither is the fact that their access to productive resources might be constrained by political, cultural, and social factors. In Zambia, the PPA highlighted the fact that children were increasingly going into prostitution and that child-headed households were becoming more common. In Togo, the PPA drew attention to vulnerable groups such as displaced people and domestic child laborers. Social exclusion and ethnicity In Guatemala, the PPA highlighted that alcoholism was a major problem for men in the indigenous areas. This had not previously been acknowledged. Most of these men were unemployed or underemployed and felt excluded from the limited opportunities for employment. In Armenia; single pensioners were consistently ranked by communities as the poorest-not because they had the least income but because they were isolated and socially excluded. In East- ern Europe, the PPAs' analysis of social connections revealed that the poor tend to be connected horizontally, that is, within their own networks, for survival and to reduce vulnerability. As a result, poor households tend to remain excluded. In contrast, the better-off households tend to be connected both hor- izontally and vertically, that is, to better-off networks, which enabled them to improve their situation. Illegal activities, crime, and violence Household surveys often are not able to access information on illegal activi- ties because of the reluctance of the respondent to answer questions from an enumerator she or he does not trust. PPAs, however, have been able to shed light on the relationship between poverty and illegal activities. For example, the PPAs in Zambia and Jamaica revealed that prostitution, crime, and vio- lence were major concerns among the poor. People were feeling increasingly scared, unsafe, and insecure as community coherence was threatened because of violence. In some communities, women and the elderly were reluctant to use public transportation, particularly at night, because of safety concerns. 24 CAN THE POOR INFLUENCE POLICY? PPAs also portray the reality of poor people's lives. In Equatorial Guinea the results of the PPA highlighted the feelings of hopelessness and despair many people felt after years of declining well-being and repression. Suicide-not generally considered an issue in Africa-was mentioned as a problem. The results of this PPA were described by one Bank economist as "terrifying." In the postconflict countries of Burundi, Rwanda, and Uganda it was not possible to undertake household surveys. In these countries, the PPA proved to be a very useful tool in providing initial data on poverty and conflict impacts. 2.1.2 Explanatory Power of the PPAs Many PPAs provide insights into the dynamics and processes of impov- erishment (see Box 6). 2.1.3 Priorities of the poor Although some problems highlighted by the PPAs were already known, the PPAs have made it clear that the poor have the capacity to analyze Box 6. Examples of Explanations Provided by PPAs Why some women were not working in Mexico City: The PPA in Mexico found that some women in Mexico City were unwilling to leave their houses and go to work. Because they did not have tenancy rights, they were afraid that their houses might become occupied in their absence. Why the poor spend a "disproportionate" amount of money on clothing in Mali: The results of the quantitative survey showed that a disproportionate amount of money was spent on clothing. The PPA explained that clothing and cloth are investment items in addition to being status symbols and therefore play an insurance role. Why migrants with money still lack access to land in Zambia: The PPA explained that the social status of certain groups sometimes determines their economic status. Migrant groups might lack access to high-value land not because they lack money but because they lack entitlement in the view of local social institutions that determine land ownership. Lack of social status therefore prevents migrants from gaining title because social institutions actively prevent transfers of land. Why people were not using health facilities in Kenya and Pakistan: The PPAs explained that communities were discouraged from using health facili- ties because health staff were often rude and condescending. IMPACT OF THE PPA 25 the causes of their vulnerability and rank their priorities. In PPAs carried out in Ghana, Mali, and Nigeria, for example, the poor said that phys- ical isolation and lack of access to water were problems. In Costa Rica, the PPA highlighted linkages between home ownership and status in society. 2.1.4 Beyond the household unit: Different levels of analysis Household surveys often interview only the head of household (usu- ally a man). PPAs typically gather information on intrahousehold issues from more than one perspective, and also explore interhousehold and community-level social issues in addition to gathering household data. PPAs have focused on individual case studies of people, providing insights into the dynamics of poverty and survival strategies; intra- household dynamics, revealing both the unequal allocation of resources among household members and the impact of power relations on the poverty of women, men, and children within the household; inter- household dynamics, illustrating, for example, the fact that female- headed households might rely on interhousehold transfers; household-level information; and a community perspective highlight- ing the diversity of social or cultural groups and their wide-ranging coping strategies. Local people's understanding of their poverty can be increased if the PPA-especially if it includes a PRA-involves the community in the analysis. In Zambia, one participant in the yearly participatory poverty monitoring stated that the research had enabled the people in the com- munity to get together to discuss their problems and reflect on their sit- uation, that of their neighbors, and the community as a whole. Owen (1997) adds that by using PRAs, the PPA in Mozambique encouraged communities to become conscious of their life conditions, opportuni- ties, strengths, and limitations. This, he says, is particularly important where government has limited capacity to assist people in many areas of the country. In summary, PPAs are deepening the understanding and providing a dynamic picture of poverty. For example, all of the following insights have emerged from Zambia's PPA: child-headed households, child labor, crime, violence, and prostitution as coping strategies; increased feelings of insecurity and lack of safety as an outcome of these strate- gies; seasonal fluctuations in sickness, rates of work, and access to food as triggers of greater vulnerability; and the impact of these new dimen- sions on people's behavior as individuals, as household members, and as part of a community. 26 CAN THE POOR INFLUENCE POLICY? 2.2 Attitudes and Policy Change Formulation of more appropriate and poverty-focused policies can be constrained when Bank staff and government officials involved in the policy dialogue have different attitudes. Some governments have little immediate political incentive to help the poor because the poor are often not organized, have a weak voice, and are difficult to reach. And most Bank staff have little direct experience with poverty. Some PPAs have helped to change the attitudes of both Bank staff and senior government policymakers, thereby contributing to policy formulation. It is rarely possible to establish clear causality between the PPA and policy change because policymaking is part of a wider social process. In addition, it is usually difficult to separate the impact of the PPA from that of the poverty assessment. However, some indicative evidence is presented below. 2.2.1 World Bank Insights arising from the PPAs are contributing to the broader debates within the World Bank on how to measure and monitor poverty, inte- grate social dimensions into policy and project work, and increase the impact of the Bank's operations by adopting participatory approaches. There is a growing realization of the value of integrating quantitative and qualitative data in the analysis of poverty, in order to produce better measurement, better analysis, and, through more appropriate policy rec- ommendations, better action (Carvalho and White 1997; Chung 1997). Influencing the World Bank lending program In certain instances, PPAs have successfully contributed to a shift of policy emphasis. In Nigeria, for example, the World Bank had been focusing on health and education, yet the PPA highlighted that the poor viewed water and roads as the priorities. There is now a greater focus on water and roads. In Ghana, the PPA contributed to a shift of empha- sis within the Bank to rural infrastructure and the quality and accessi- bility of education and health care (see Norton 1996), which was subsequently followed by the preparation of the Village Infrastructure Project. In Ecuador, the PPA highlighted the fact that women were reluctant to work away from home because of street crime and violence. The poverty assessment identified the provision of street lighting and guarded public buses in the evening as effective ways to address this problem. In Zambia, the World Bank's Social Fund supported some of the priorities identified by the communities in the PPA, and a health IMPACT OF THE PPA 27 project now includes cost recovery conditions as identified in the PPA. In Niger, the PPA influenced the design of the proposed Infrastructure Project to be more poverty focused and include pilot rural operations. And in Burundi, the Bank is designing a community-based poverty project, which will use the recommendations of the PPA. In other cases, PPA impact has been less evident. The poverty assess- ment in Kenya reflected the major findings of the PPA, but the results have not been extensively incorporated into other country reports. In Costa Rica, delays in the analysis and dissemination of the findings have limited the impact of the PPA. Rapid assessments using PPA approaches are now being employed by the Bank to understand the social impacts of the financial crisis in East Asia. Initial surveys have been undertaken in Thailand, the Philip- pines, Indonesia, Cambodia, and Laos.' The objective was to consult with a cross-section of organizations including community groups, local and international NGO networks, academic institutions, labor unions, professional associations, other donors, and government departments to determine shifting patterns of vulnerability. Focus groups, rapid assessment techniques, and participatory exercises were used. These initial assessments have contributed to creating the frame- work to begin a dialogue with governments and other donors and jointly formulate strategies for action. In all these countries, there is a time lag in obtaining reliable statisti- cal data. The advantage of the initial rapid assessments has been to quickly produce a series of hypotheses about the potential impacts of the financial crisis on the poor. These data now form the basis for fur- ther ongoing problem identification with the objective of providing a baseline and defining the issues for more detailed, systematic, and rep- resentative participatory surveys. This next step has been taken in Thai- land, where a national participatory assessment, using the PPA approach, is being designed as part of the Bank's Social Investment Project. The objectives of this assessment are to first, increase the Bank's and the country's understanding of the shifting patterns of vulnerabil- ity as the impacts of the crisis deepen; second, inform Bank and gov- ernment policymakers and therefore influence policy; and third, strengthen the capacity in country to undertake participatory surveys and to analyze results, particularly by consolidating the results of par- ticipatory and traditional surveys. Links to World Bank policy documents The results of some PPAs have been reflected in Bank policy docu- ments. An example is the Bank's Taking Action for Poverty Reduction in Sub-Saharan Africa (World Bank 1996i), the product of the Bank's Africa 28 CAN THE POOR INFLUENCE POLICY? Region Poverty Task Force that resulted from extensive dialogue with development partners. The task force was established to assess the Bank's operations in the Africa Region, and the report is now being used as a basis for the Bank's strategy in Africa. In addition, a series of PPAs in Zambia, Mali, Ghana, and Nigeria identified both physical iso- lation and lack of access to water as major concerns. As a consequence, it was recognized that rural water and roads infrastructure had been neglected areas of investment for the Bank. The report recommended that these should be priority areas in the future. In Gabon, the results of the PPA influenced the Bank's decisions to undertake a Public Expen- diture Review in the health and education sectors. Links to country assistance strategies (CASs) are difficult to deter- mine at this stage. However, two examples where the CAS was clearly influenced by the PPA and the poverty assessment are Armenia and Zambia. In Armenia, the PPA highlighted the importance people place on health and education. The CAS emphasizes protecting access to these sectors. In Zambia, the PPA highlighted the limited access the poor have to public services. The CAS has made this a central theme. In Niger, the value added of the participatory process for the poverty assessment and the PPA was recognized and will now be adopted for the Niger CAS. In Ethiopia, the results of the quantitative survey were delayed because of data problems. The CAS for Ethiopia therefore drew extensively on the data from the PPA. In Rwanda, the PPA results fed into the CAS, the agriculture strategy note, and an agricultural learning and innovation loan. The importance of using poverty assessment to focus the CAS on the Bank's overarching objective of poverty reduction is now widely rec- ognized in the Bank. Thus, in order to more effectively focus the CAS and country programs on poverty, the Bank is shifting from mandatory one-time poverty assessments to long-term strategic poverty monitor- ing that combines periodic household surveys with periodic participa- tory research. 2.2.2 National level Attitudinal change starts with appreciating the value of how the poor perceive their situation. In Tanzania, the government was initially cau- tious about the PPA exercise but became more receptive when the PPA highlighted the capacity of local people to analyze their own problems. Policymakers began to understand the value of including the poor in the policy dialogue. Similarly, in Benin the PPA strengthened the inter- est of the Ministry of Planning in consulting the poor through a partic- ipatory assessment. IMPACT OF THE PPA 29 In Zambia, the government was influenced by the priorities expressed by the poor in a ranking exercise. The Ministry of Health has been using the results of the PPA and the poverty assessment to develop policy. In other countries, the poverty assessment and PPA have opened up the policy debate, enabling discussions of once highly sensitive issues. In Swaziland, the workshop convened in February 1997 to dis- cuss the results of the PPA was the first government-sponsored work- shop on national poverty. Key insights from the PPA on such issues as women's rights, land tenure, and the role of traditional authority were given higher priority in the policy agenda as a result of this workshop and the dialogue surrounding the PPA. In Lesotho, three key themes emerged from the PPA that were not identified in the household sur- veys: alcoholism, political factors, and corruption. Through the gov- ernment's action plan, these issues were placed on the policy agenda. As the process has developed in Lesotho, government ownership has increased, and the topic of corruption has now appeared in speeches and policy discussion documents. Opening up the policy debate can be a conflict-ridden process. The results of both the PPA and poverty assessment were a shock to people in Cameroon inside and outside the government, as poverty had not previously been acknowledged as a serious problem. Ownership had not been developed among key policymakers during the PPA process, because the central government was not strongly committed to poverty reduction. As a consequence, there was little acceptance or use of the results. The poverty assessment and PPA did seem to have an impact at the local government level, where officials expressed a great deal of interest in replicating the methodology of the PPA. 2.3 Strengthening the Capacity to Deliver Policy Tlhe impact of a PPA on strengthening the capacity to deliver poverty- focused policies can be assessed by identifying new institutional align- ments and partnerships that arise as a result of the PPA. Increased dialogue and consequent partnerships can also contribute to widening the constituency for reform, increasing ownership, and strengthening the commitment to poverty reduction. 2.3.1 World Bank The extent to which PPAs have had ari impact on the Bank's capacity to fulfill its poverty reduction mandate is difficult to determine at this early stage. As stated above, the links between the PPA and poverty assessment, the lending program, and the CAS are evident in only a 30 CAN THE POOR INFLUENCE POLICY? few countries. The Bank is now developing an interdisciplinary approach to the diagnosis of poverty and the analysis of how all types of institutions affect the poor. This approach will lead to a better under- standing of the problem of poverty and increase the Bank's capacity to work with the relevant institutions. 2.3.2 National level In some countries the process of compiling the PPA has helped to cre- ate a dialogue and partnerships for policy delivery. One of the strengths of the Mozambique PPA was the diversity of the involved institu- tions (university, government, NGOs) and researchers. The multi- institutional approach has strengthened relationships among the various participating institutions (Owen 1997). In Argentina, increased coordination between government agencies and programs has been developed. In addition, dissemination of the results of the PPA has val- idated the methodology and contributed to the development of an inte- grated (qualitative and quantitative) approach to monitoring and evaluating social programs undertaken by different organizations. In other countries, the PPAs have increased the capacity of certain civil society institutions as well as government. In Cameroon, the Bank manager of the PPA stated in an internal communication to team mem- bers that "involving local institutions and holding workshops with both government and civil society are mechanisms for expanding own- ership of the poverty problem and in-country capacity to analyze and address it." In some cases, the researchers and intermediary institu- tions that undertook the PPA were empowered by the process. In South Africa, for example, the local researchers later adopted an activist role. In Ghana, the capacity of local organizations to undertake credible par- ticipatory research has been developed, with the local NGO, Centre for Development of People, benefiting from extensive training and institu- tional linkages created by.the PPA process. Tlhrough dialogue at the community level, communities that are no longer passive recipients of a policy might become more committed to policy delivery. In some communities, PRAs resulted in local people identifying their priorities, which were later followed up in the form of projects supported by various agencies. In South Africa, for example, the PRA work became a catalyst for communities to initiate a project to benefit the poor. The impact of PPAs has been limited where follow-up has not been extensive, leading many to question the value to the poor of such work. Indeed, at the workshop in Mozambique organized for this study, participants explained that many communities had "respon- dent fatigue-fadiga dos informantes." The workshop concluded that IMPACT OF THE PPA 31 many communities, especially those accessible from major cities, are the subject of excessive research, and "agreed.. .that before initiating any study, a review be undertaken of existing data and material per- taining to the area" (Owen 1997). 2.4 Determinants of the Level of Impact 2.4.1 Methodology Of the 43 PPAs reviewed for this study, 21 (the ones with sufficient data) were analyzed in more detail to quantify the level of impact and take the first step in exploring the effects of a variety of possible explanatory variables.2 From the data in annexes I and 2, a list of key impact variables was identified. For each variable, a rating of high, medium, low, or zero was assigned to each country PPA on the basis of desk work, discussions with participants, and field research in the five countries the author visited for this study. The ratings are largely sub- jective; they are not objectively measured indices of a PPA's success. In the future, more empirical research will be required. The results of the analysis are summarized in Table 4. The analysis of the 21 PPAs suggests a significant influence on the diagnosis of poverty in 71 percent of the cases examined. Twenty-nine percent of the PPAs had a significant impact on policy formulation, both in the Bank and in country, while in only 24 percent of the cases did the PPA have a significant effect on the country's capacity to deliver some policies. The analysis involved classifying the PPAs in a 3X3 matrix based on the composite impact index (CII) and a variety of possible explanatory variables. The CII used the ratings assigned to four key impact vari- ables. These variables and the weights assigned to them were as follows: DUP-deepening the understanding of poverty (1) IWP-influencing World Bank policy (2) ICP-influencing country policy (2) ICD-increasing capacity to deliver policy (4) The four rating levels were assigned a score as follows: none = 0; low = 1; medium = 2; high = 3. For each of the 21 country PPAs, the CII was calculated as: CII = 1 (DUP rating) + 2 (IWP rating) + 2 (ICP rating) + 4 (ICD rating) 32 CAN THE POOR INFLUENCE POLICY? The maximum attainable score was therefore 27, calculated as: CII = 1(3) + 2(3) + 2(3) + 4(3) The next step was to define a series of independent variables that had potential to explain the level of PPA impact as measured by the CII. For each independent variable, subjective ratings of low, medium, and high were assigned for each country PPA. These results are set out in Table 5. Pairs of impact variables were then chosen and charted them against the CII. Only high and low ratings were included to highlight the more marked differences observed among the PPAs. 2.4.2 Main findings 1) When the CII was plotted against the subjective estimates of the quality of the PPA research team, there was a clear positive correlation (see Figure 2). PPAs judged to have high-quality teams averaged more than 15 out of a possible 27 on the CII. Those judged to have medium- quality teams averaged 7, and those with low-quality research teams averaged only 2. 2) The relationship among the CII, the quality of the PPA manager, and Bank management support revealed a high level of interaction between the latter two variables (see Figure 3). Where both Bank man- agement support and PPA manager quality were high, average CII was Table 4. Summary of PPA Impacts % of PPAs with Number of PPAs % of PPAs with low, significant (out of2l analyzed) medium, or high impact impact* Impact variables None Low Medium High Low Medium High 1. Deepening the ... 6 10 5 28 48 24 71 understanding of poverty 2. Influencing policy 8 7 4 2 33 19 10 29 at the World Bank Influencing policy at 9 6 4 2 28 19 10 29 the national level 3. Increasing country's 13 3 3 2 14 14 10 24 capacity to implement policy * significant impact = medium or high rating. IMPACT OF THE PPA 33 Table 5. Summary of Impact Variables Number of PPAs out of2l Low Medium High Impact variable None extent extent extent World Bank 1. Ownership within the World Bank 0 10 6 5 (by staff, departments) 2. Bank management support 0 11 5 5 3. Links to poverty assessment 2 8 5 6 4. Links to CAS 0 19 0 2 5. Team work 7 5 6 3 Country level 1. Involvement of policymakers 0 7 6 8 2. Involvement of other stakeholders 0 8 9 4 3. Ownership by government 0 8 6 7 4. Dissemination at the national level 0 9 4 8 Community level 1. Skills of researchers 0 7 8 6 2. Dissemination to communities 13 6 2 0 3. Length of time in field 0 4 9 8 4. Cost 0 7 7 7 5. Follow-up; action with communities 9 6 6 0 Figure 2. PPA Impact by Quality of Research Team 20 s 15- E 1 0 - U*I 0 0- 1 0) 0 Low Medium High Quality of Research Team 34 CAN THE POOR INFLUENCE POLICY? Figure 3. PPA Impact by Quality of PPA Manager and Level of Bank Management Support x 5/ U E I0~ 0 E / v M - J 7t Quality of PPA u Loo Manager High LO Low Bank Management Support high (17 out of a possible 27). However, even high-quality PPA man- agers were unlikely to produce high-impact PPAs without strong Bank management support-the PPAs in that category scored only 7 out of a possible 27. There were no PPAs with high Bank management support and low manager quality; hence the zero CII score in the lower left cor- ner of the chart above. 3) When the CII was plotted against links to the poverty assessment there was a clear positive correlation (see Figure 4). PPAs judged to have a greater link to the poverty assessment averaged a CII of more than 19 out of a possible 27. Those judged to have medium links aver- aged 5, and those with limited links averaged only 2. To have a signifi- cant impact, PPAs need to be linked to the poverty assessment. 4) The relationship among CII, ownership by the government, and ownership within the World Bank revealed a high level of interaction between the latter two explanatory variables (see Figure 5). Where there was a high degree of ownership in the Bank and by government, the CII reached 20 out of a possible 27. Ownership by both the Bank and government is important to achieve a high-impact PPA. There was no instance of a PPA with high World Bank ownership and low govern- ment ownership; hence the zero CII score in the top right corner of Fig- ure 5. IMPACT OF THE PPA 35 Figure 4. PPA Impact by Link to Poverty Assessment 20 18- 1 6 x 14- -: 12 - U 7 0- 10 E .2 8 0 -6- -I E 6 0 2- 0 LOW Medium High Link to Poverty Assessment Figure 5. PPA Impact by World Bank and Government Ownership 25 20 u 5 ~~~~~~~~~~~~World Bank 0 ~~~~~~~~~~Ownership Hih Low Government Ownership 36 CAN THE POOR INFLUENCE POLICY? 5) When the CII was plotted against the extent to which policymak- ers were involved, there was a clear positive correlation (see Figure 6). Where there was a high level of policymaker involvement, the CII was more than 16 out of a possible 27. Those PPAs judged to have medium involvement averaged 7 and those with limited involvement averaged only 3. The level of PPA impact depends to some extent on the level of policymaker involvement. The following chapter builds on this analysis of the key variables in more detail, analyzing case examples to elucidate recommendations for good practice. Figure 6. PPA Impact by Level of Policymaker Involvement 20 *~15 0 E -;5 _i 0 0- Low Medium High Level of Policymaker Involvement Notes 1. See Robb (1998) for a summary of these initial surveys. 2. Section 2.4 was compiled with assistance from James Edgerton of the World Bank's Social Development Department. 3 Emerging Good Practice Mhis chapter identifies good practices that should be considered when Tundertaking participatory policy research for policy change. Emerg- ing good practice builds on the diverse impacts of key variables discussed in the previous chapter. It is divided into three main areas in which issues are similar and linked: first, issues to be considered from an institutional perspective within the World Bank'; second, good practice when manag- ing a PPA in country, at the national level, including how to open up the dialogue in participatory policymaking; and third, emerging good prac- tice in conducting participatory research with the poor at the community level, and the principles behind this method of data collection. There is no unconditional good practice in this type of work because the best approach will be determined by the context. However, Box 7 below gives some suggestions for good practice and minimum standards that have emerged from experience with the Bank's first 43 PPAs. These issues are then discussed in more detail throughout the chapter. 3.1 At the World Bank: Initial Steps and Follow-up This section is divided into five main parts: 1) Professional input and commitment 2) Who owns the PPA in the Bank? 3) Management support and follow-up 4) PPA design 5) Linking to the Bank's country assistance strategies 3.1.1 Professional input and commitment The first step in initiating a PPA and poverty assessment is to ask, Will the outcome drive policy reform within the country and in the work of the Bank? Whether PPA and poverty assessment will move from an 37 38 CAN THE POOR INFLUENCE POLICY? Box 7. Twelve Hallmarks of Good PPAs World Bank * PPA research agenda is designed with the country team members, leading to broader ownership and understanding of results within the Bank. * Results of the PPA are combined with household survey data and inte- grated into the poverty assessment and the country assistance strat- egy. * Bank managers and staff observe the PPAbeing conducted in the com- munity so as to better understand the strengths and weaknesses of the data. * PPA is designed as an ongoing process, undertaken intermittently, to build up poverty-monitoring data. Country level * Government support is secured from the beginning. * Key policymakers and administrators are included in designing, plan- ning, and implementing the PPA and analyzing the results. * Timing and extent of involvement of other stakeholders (NGOs, line ministries, unions, religious groups, local social science institutes) are attuned to the social and political environment. * In-country capacity to conduct ongoing PPAs is strengthened so PPAs can feed into the policy dialogue. Community level * Local research teams are trained to conduct high-quality participatory research, with an understanding of both the principles and techniques. * Communities are involved in analysis of the data. * Results of the PPA are disseminated to the communities involved in the policy research and to agencies that can follow up at the commu- nity level with action and projects. * Results are presented in a clear and concise manner. academic exercise to influencing policy depends on the extent to which the Bank and, more specifically, the sponsoring Country Department is committed to poverty reduction. Although there is no one approach to poverty reduction, and the definition of poverty is broadening (see Box 8), PPAs have yet to be as generally accepted as traditional household surveys. However, the Bank is now moving toward undertaking ongo- ing poverty analysis and monitoring as opposed to one-time poverty assessments, and is recognizing the importance of including the poor in this analysis. Experience has shown that the PPA manager needs to have a diverse set of skills, including technical methodological skills and skills in man- aging a participatory policy dialogue. Operating at a policy level and EMERGING GOOD PRACTICE 39 Box 8. What Is Poverty Reduction? The Bank is broadening its view of poverty reduction, as reflected in the cormments below by economists and sociologists affiliated with the Bank. * "Commitment to poverty reduction is dependent upon the government's pub- lic expenditure priorities. An example may be the targeting of clean waterfor the poor, which would consequently improve their health and thus increase their productivity." * "[Poverty reduction is] increasing income and general assets to a level where the poor are less vulnerable to risks andfalling below a certain level." * "Poverty reduction is giving people greater control and the means to deter- mine their lives." * "Poverty reduction starts with the poor's perceptions of their own poverty in a process of sharing strategies, priorities, and solutions of various stake- holders." opening up the policy dialogue in country often means that conflict will occur. Conflict is not always negative-from it, greater under- standing of the problems of the diverse groups involved can evolve. An understanding of people and their motivations, as well as sensitivity, tact, and diplomacy, are required when opening up a policy dialogue. This is never a smooth process: it is unpredictable, and no matter how skillful the PPA manager, the process might not go according to plan. 3.1.2 Who owns the PPA in the Bank? Ownership within the World Bank, across departments, emerges as a key issue when considering the impact of PPA exercises on World Bank policy and projects. For the PPAs in Pakistan and Cameroon, for exam- ple, there was limited ownership and understanding of the process. Consequently, the results were not reflected to a significant extent in other World Bank documents. In Cameroon, in particular, changes in the team managing the country program occurred while the poverty assessment was being prepared. Within the country department, the PPA results had limited credibility with those who were not part of the process. Additionally, keeping poverty issues on the agenda proved difficult when the CFA (Communaut6 Financi&e Africaine) was deval- ued as the emphasis shifted to macroeconomic issues. To achieve greater policy relevance and broader ownership, a broader team approach is important. In Armenia, for example, the man- ager of the poverty assessment had in-depth country knowledge, built up respect among key policymakers and within the country's academic community, and encouraged a team approach within the Bank. In addi- 40 CAN THE POOR INFLUENCE POLICY? tion, the PPA manager worked closely with those managing the house- hold surveys and the country department's macroeconomist to estab- lish a research agenda for the PPA. As a result, the process had the following outcomes: first, the results of the PPA were reflected in the poverty assessment; second, the country program and the CAS inte- grated the results of the poverty assessment; and third, the poverty assessment was well received and used by policymakers in Armenia. 3.1.3 Management support andfollow-up Limited management support and follow-up within the Bank have sometimes led to lost opportunities. In Madagascar, for example, there was a high degree of in-country support because key policymakers were included from the beginning. With changes in the Bank manage- ment, however, there was a delay in follow-up of more than a year and a half, and the commitment and interest of the government conse- quently weakened. In Equatorial Guinea, the information was contro- versial and the Bank was reluctant to continue the process. Appropriate follow-up measures are sometimes difficult to identify because the outcomes of the PPA and poverty assessment consultations are not always reported accurately. In one country, many NGOs and high-ranking government officials openly opposed the results of the poverty assessment. In addition, many felt unhappy that their views were sought during the consultation but then not included in the final poverty assessment. Yet within the Bank this poverty assessment was considered technically sound and successful. A recommendation for good practice is to monitor not just the outcome of the policy dialogue or the poverty assessment but also the process and outcomes of the par- ticipation and consultation. For example, CASs, poverty assessments, and PPAs could document who was consulted and how, and the major lessons learned from consulting each of the key stakeholders. Decentralizing the management of the PPA to resident missions might be appropriate in some countries because it is difficult to coor- dinate the PPA process from Washington. The manager of the Tanza- nia PPA, for example, suggested the need to strengthen that resident mission to enable it to undertake frequent PPAs and contribute to a broader poverty assessment. Teams could be located in the field, and people skilled in the analysis of poverty could be located within the mission. To increase the capacity of the resident mission, training in participatory policy research could be conducted and tool kits pro- vided. Where appropriate, the NGO officers and social scientists (recently recruited in many resident missions) could assist in such poverty-focused work. EMERGING GOOD PRACTICE 41 3.1.4 PPA design Some PPAs can be strengthened by the use of sampling methods. For example, the selection of PPA sites could be informed by traditional household survey data. In Kenya, the Welfare Monitoring Survey (WMS), based on a nationally representative sample of some 12,000 households, was used to identify the poorest districts in each of six provinces. These households became the focus for the PPA. Within each of these poorest districts, two WMS clusters (roughly equivalent to a village) were randomly selected for the PPA, and the WMS survey enu- merators most familiar with those clusters were then attached to the PPA teams to serve as guides. The PPA was conducted in a subsample of clusters used for the WMS. Trust and understanding should be developed among those who use different approaches to defining research agendas and collecting and analyzing data with the aim of influencing policy The integration of data sets will evolve through this trust building. Both survey and participatory assessment practitioners need to understand the limita- tions of various data sets, appreciate the biases in their own research methods, and know when alternative methods can compensate for some of these limitations. In an attempt to better understand the various approaches to poverty analysis, the local NGO research team (Participatory Assessment Group) in Zambia is currently undertaking participatory poverty mon- itoring exercises and combining the results with those of the household survey carried out by the Central Statistics Office. In other countries, policymakers have visited research teams in the communities. In Costa Rica, for example, a senior official from the Ministry of Economic Plan- ning was involved with the research and consequently was better able to appreciate the value and limitations of the PPA. And in Armenia, where the manager of the poverty assessment built trust among those managing the household survey, the Bank's PPA team, and govern- ment policymakers, the resulting integrated analysis of poverty was widely used both by the Bank and by government policymakers. Another PPA design issue is the need to ensure that the results are shared with various stakeholders. Dissemination of results should be part of the PPA planning and budget, but in most PPAs this has not been the case. If the information gathered is not fed back to the com- munities, the participatory nature of the work is incomplete. There are several important reasons to feed back information from the PPA: to validate the information; to continue the process of constructing a dia- logue with communities so that semipermanent linkages are created; to show respect for the partnership with the community by sharing the 42 CAN THE POOR INFLUENCE POLICY? information; to continuously reevaluate the relationship of the PPA facilitator (e.g., the Bank) with the various stakeholders, especially the poor; to increase the credibility of the information and thereby enhance the potential of the PPAs to influence policy formulation and delivery; to facilitate a process whereby the poor monitor and evaluate the impact of the PPA; and to encourage action at the community level. In designing a dissemination strategy, the demands of the various stakeholders should be considered and key stakeholders should be involved. Where governments have not been involved, results have been mixed. In Cameroon, for example, the results of the PPA were published without full government support and the government felt threatened by the results. Thus, the impact of the PPA was greatly reduced. To build a political base for policy change, effective use of the media as part of a communications strategy can help to increase under- standing between the government and the public. Different documents might be needed to meet various stakeholder demands. For example, those at the community level might be more interested in detailed site reports of their community, whereas line min- istries might want a country-level document. To increase awareness and to disseminate the broad results and policy recommendations of the PPA, countries such as Zambia, Guatemala, and Lesotho have pro- duced clear, well-written, short summary documents that have con- tributed to widespread ownership and understanding. Another suggestion is for the Bank to produce a separate document on the PPA results in addition to integrating these results into the overall poverty assessment. This could give the PPA managers more autonomy and accountability. The design stage should include consideration of how the data will be presented. In Ghana, for example, the information from the PPA was relatively complex and extensive, making incorporation into other World Bank reports time-consuming. In other PPAs, it might be appro- priate to use the visual diagrams from the PRA exercises (see Annex 4) in the final report as a means of conveying information. One reason that PPAs have not resulted in more action has been the lack of specificity in the presentation of results. Wherever possible, therefore, proposals should be presented in matrix form, detailing the following: * Actions that could be taken immediately; * Actions that require policy change; * Cost requirements; * Whether a short or long time is required for results; * Administrative order or legislation required; and EMERGING GOOD PRACTICE 43 Which ministry, donor, or NGO could take responsibility for carry- ing out the action. To follow up on such proposals, the PPA should include a monitor- ing component. 3.1.5 Link to country strategies To better reach the poor, the results of PPAs and poverty assessments should be closely linked to the World Bank's CASs. Their impact on CASs have been weak because of broad constraints on adopting par- ticipatory approaches in both projects and policy work. These con- straints include * Accountability: In some areas, it is not always possible to assess the quality and extent of participation. Stakeholder analysis and a plan for including stakeholders in the evolving dialogue are not always clearly presented. Thus, it is difficult to track the process and com- pare the level and quality of actual participation with the level and quality of planned participation. * Support: Some participatory activities are limited because of the lack of time and funding. In both project and policy work, it is some- times difficult and time consuming to obtain funding to include a wider cross-section of stakeholders. Trust funds (grants given by bilateral aid agencies) are available but can be difficult to access for policy work. Core Bank funding is often not available, and many governments are still unwilling to borrow money for such activities. Pressure to conform to ever-tightening deadlines often undermines broader participation and consequent ownership and commitment. * Evidence: A few people in the Bank and some government officials question the cost benefit of participation and believe that an intel- lectual case for participation needs to be made. In 1996, the Operations Evaluation Department surveyed the man- agers of completed and ongoing poverty assessments (see World Bank 1996j). Only 46 percent of those who answered the question, "What influences the impact of poverty assessments?" believed that the poor should participate in the design and preparation of such assessments. However, this is an evolving situation and in many sectors throughout the Bank, support for participation is growing rapidly, with some recent significant advocacy from senior managers. There is now a move to increase the poverty focus of operations by overcoming barriers such as 44 CAN THE POOR INFLUENCE POLICY? * Strategic issues: weak links between the PPA and the poverty assess- ment, between the poverty assessment and the CAS, and between the CAS and operations; * Lending: emphasis on loans approved rather than on poverty reduc- tion goals; and * Impact: focus on input and disbursement indicators rather than on laying the foundation for assessing impacts on the poor. Another Bank report, Taking Actionfor Poverty Reduction in Sub-Saha- ran Africa (World Bank 1996i), notes that "Poverty reduction is rarely a central or motivating theme for the business plan or country assistance strategies, although responsiveness on this issue has recently improved" (p. 15). The report contends that CASs are too general to address poverty and that much of the poverty focus of projects is lost by the time the lending program is implemented. The report further states that CASs usually do not make poverty reduction a core objec- tive of economic development programs, that poverty reduction is inci- dental to macroeconomic stability or lending, and that the link between the reform agenda and poverty reduction usually is not made. The report adds that past CASs have lacked a "strong strategic vision on poverty reduction and clear monitorable actions for reducing poverty" and argues that This shortcoming at the operation level is often rooted in: (a) a lack of information on poverty, (b) inadequate analysis, (c) a dis- interested attitude toward poverty reduction, and (d) Bank man- agement's willingness to compromise on poverty reduction to maintain good country relations and to be satisfied with lending operations that address aggregate growth with little attention to the distribution of growth. ...Operational decisions, therefore, tend to be based more on sector interests than on poverty reduc- tion jwhereas poverty is] a multisector issue requiring an inte- grated strategy (p. 20). The report calls for the Bank to revamp its strategy to include respon- siveness to the needs of the poor which in turn, requires a better under- standing of poverty-precisely what the PPA can deliver, in conjunction with household surveys. There is now a demand for better poverty analysis to help both the Bank and governments focus their projects and policies more effec- tively. To achieve this, PPAs should aim to become a building block and not just an adjunct to CASs and policy framework papers. Experience from past PPAs shows that this linkage is greater where the research EMERGING GOOD PRACTICE 45 agendas for the PPA and the poverty assessment have been developed with those working in country departments and on CASs. This coop- eration can be time consuming and requires more preparation, but the payoff is a greater impact on the CAS. Another report, Poverty Reduction and the World Bank (1997c), details how many of the CASs have become distinctly more poverty focused, particularly those for Sub-Saharan Africa, ever since the May 1996 directive from senior management to put poverty reduction at the cen- ter of the country assistance strategies. Other recent developments include more poverty-focused guidelines for CASs and the rewriting of the Operational Directive on poverty. The reports note that all CASs that are written a year or two after a poverty assessment incorporate the main findings of the assessment, although some do so more com- prehensively than others. To further strengthen this linkage, CASs for a limited number of pilot countries-about one per region-will focus on incorporating poverty concerns. Best-practice approaches will be shared across coun- tries and regions. In summary, although there is scope for further improvement, CASs have increased their focus on poverty issues and the Bank is making an effort to incorporate into such strategies the find- ings of poverty assessments. A summary of this section is provided in Box 9. Box 9. Factors for the World Bank to Consider to Increase the Impact of PPAs Professional input and commitment . Promote poverty reduction as a clear cornmitment. The extent to which country departments and country directors at the Bank are committed, to poverty reduction will affect the impact of the PPA and poverty assessment. Where this comrmitment is not clear, operations will tend to be more biased toward sector interests rather than poverty reduc- tion. . Measure performance of country directors by the poverty focus of the CAS and pipeline projects. * Develop skills to conduct poverty assessments and PPAs. Challenge individual behavior, approaches, and motivations. * Bank staff should observe participatory research in target commurti- ties to understand the strengths and weaknesses of PPAs. * Increase information on PPAs through the Knowledge Management System. (continued) 46 CAN THE POOR INFLUENCE POLICY? Box 9. (continued) Poverty analysis * Develop trust and understanding between those who manage data collection for the various approaches (surveys and participatory research) and those who are doing the poverty analysis. * Promote a team approach within the Bank and include different disci- plines to enhance the understanding of the various dimensions of poverty. Ownership in the Bank * Establish broad ownership within the Bank for greater policy rele- vance. * Create the research agenda for both the poverty assessment and PPA with others working in country departments. Management support and follow-up * Increase the capacity of resident missions to supervise poverty analy- sis. * Undertake continuous participatory poverty monitoring, as in Zam- bia (see Box 2), to build up time sequence data. * Monitor not just the outcome of the policy dialogue (the PPA, poverty assessment, and CAS) but also the process of participation and con- sultation. Also monitor the follow-up of the PPA and poverty assess- ment recommendations. * Document in the final PPA, poverty assessments, and CAS who was consulted, what the major lessons were from consulting each of the stakeholders, and what changes were made as a result of the consul- tation process. PPA design * Link participatory research with household surveys where appropri- ate and build an iterative process whereby traditional surveys and participatory research inform each other on an ongoing basis. * Design PPAs with dissemination strategies. * Produce clear, well-laid-out reports. Produce different reports for dif- ferent audiences. Detail the process of consultation in each report. * Use the media to promote communication between the government and the public and increase the political base for policy change. Link the PPA and poverty assessment to the CAS * Ensure that PPA and poverty assessment are building blocks for the CAS. * Work to ensure that the poverty assessment drives policy reform, both in country and in the work of the Bank. * Identify, in the CAS, clear, monitorable actions for reducing poverty. * Build on existing social knowledge in the country. EMERGING GOOD PRACTICE 47 3.2 At the Country Level: Listening to Policymakers and Opening Up Dialogue Using the PPA examples, this section looks at the major issues to be considered when working with institutions in country. It is divided into the following parts: 1) Starting point-understanding the political environment 2) Creating a conducive policy environment 3) Who controls the research agenda and outcome? 4) Strengthening the policy delivery framework 3.2.1 Starting point-understanding the political environment Participatory policymaking involves linking information from com- munities into a broader policy dialogue that includes a cross-section of stakeholders. In moving from community-level research results to pol- icy analysis, issues surrounding policy change should be considered. For example, policy formulation is an inherently political process. Rules, legislation, traditions, networks, ethnic alliances, patronage, political allegiances, and bureaucratic structures all interact to form a complex and fluctuating policy environment. Key questions, therefore, include what factors affect policymakers decisions to create, sustain, alter, or reverse polices; what are the legal complexities of policy change; and what influence does individual survival in an institution, institutional survival in a government, and the maintenance of a regime within a country have on policy choice? A further complexity of the policymaking process is the relationship between policy formulation and implementation. Policymaking and implementation are not disconnected but are part of ongoing interre- lated processes of change (Grindle and Thomas 1991). But while some policymakers might be willing to incorporate certain issues in the pol- icy agenda as statements of intent, they might be less willing to imple- ment the resulting policies because of the political dimensions of implementation (see Wildavsky 1979; Moser 1993; and Wuyts et al. 1992). It is within this dynamic that the World Bank is trying to influence policy and therefore needs to understand the often hidden influences on policy decisions, including the many institutional, formal, personal, and informal networks that can either help or hinder implementation. For example, in some of the countries where PPAs have been under- taken, poverty has not been high on the political agenda. Limited polit- ical support, or a lack of trust between the government and the World Bank, has led to a lack of support in country for some PPAs. In 48 CAN THE POOR INFLUENCE POLICY? Cameroon, there was a perceived lack of support from the central gov- ernment, in part because some key policymakers felt excluded from the PPA dialogue. Although the field work was considered to be good qual- ity and the results relevant, the government was not willing to embrace the findings of the PPA or to include in the political agenda controver- sial issues emerging from the PPA. In general, open political environments provide greater opportuni- ties for building consensus in regard to poverty issues. For example, in Costa Rica, where there is a tradition of bringing marginal groups into the political sphere, the government was eager to better understand poverty from the perspective of the poor and welcomed the PPA. Sim- ilarly, in Argentina, the government requested assistance from the World Bank to undertake participatory research. As a result, a strong level of commitment and coordination existed between the Bank and the government in the preparation of the poverty assessment and the PPA. In contrast, in Mali, because of the sensitivity of the poverty issue, the PPA had to be renamed the Living Conditions Survey and open dia- logue on poverty was constrained. In countries where poverty is highly sensitive, however, not all pol- icymakers will be opponents. Individuals respond to a great many fac- tors including bureaucratic structures, political stability and support, technical advice, and international actors (see Grindle and Thomas 1991). Some might support the PPA if they perceive it to be for the good of their society, since not all policymakers are just rent seekers. It is good practice to identify and include those who support the idea of the PPA at the beginning of the dialogue and gradually build up broad- based support. Such good practice requires that Bank teams have an in- depth country knowledge of policymakers and that they develop relationships with and understanding of the key players. 3.2.2 Creating an environment conducive to poverty dialogue Without government support, or even with limited support, the impact of the PPA is lessened. Because the ultimate objective is to influence policy rather than just produce technically sound documents, the value of conducting a PPA with little government support should be ques- tioned. With limited support, a key issue will be What happens when the research results run counter to the government's interest? Thus, dialogue is needed to build trust and understanding between the Bank and the government before the PPA is undertaken. Generating a more open cli- mate can help ensure that the government is less threatened by the PPA results and that the PPA thus will have greater impact. EMERGING GOOD PRACTICE 49 The participatory process will vary greatly from country to country, and the inclusion of different stakeholders within the PPA and poverty assessment should be attuned to the country's overall political, social, economic, and institutional environment. In this kind of highly con- text-specific work, it is not possible to provide a blueprint; personal judgment is required. In some countries it might be appropriate to include a cross-section of stakeholders rather than targeting only a few policymakers. In South Africa, for example, the unexpected closure of the South African Reconstruction and Development Office meant that the initial strategy of focusing on one particular department was ren- dered inadequate (see May and Attwood 1996). Maintaining a receptive attitude is not easy in a dynamic environ- ment, where unexpected conflict often occurs and agendas and people change. Continuous follow-up and dialogue with various stakeholders is therefore recommended. This approach requires a shift from top- down prescription to a more flexible process approach, with local dia- logue being maintained in country. The challenge for many PPAs, and for the Bank's wider country programs, is to maintain the new part- nerships created through such dialogue. 3.2.3 Who owns and controls the research agenda and outcome of the PPA? At the national level, ownership and commitment of stakeholders has varied among the PPAs. The Bank's experience has shown that the involvement of key policymakers from the beginning enhances owner- ship and commitment. Where appropriate, the following measures can help to increase policy impact: - Involve policymakers in the early planning of the PPA. * Bring key policymakers into the field to participate in the PPA. * When sharing a report with government policymakers, include local communities who contributed their analysis. * After the results are presented, convene workshops with policy- makers and local people. * Negotiate high-level commitment to follow up the PPA and monitor the implementation of key recommendations. In Argentina and Zambia, key government officials were included from the beginning and often led the process. As NGOs and other stake- holders were gradually included, the room for dialogue on poverty increased. This approach led to greater understanding and trust between the government and the NGOs. In South Africa, stakeholder 50 CAN THE POOR fNFLUENCE POLICY? involvement from the beginning was a time-consuming but important step in a complex process of dialogue, with a high level of ownership and commitment evident. In contrast, in Togo and Cameroon, key pol- icymakers were not included early in the process and, therefore, the PPA's impact has been limited. Similarly, in Lesotho, the government was initially not included and there was limited ownership. Local own- ership was created only when the action plan was formulated by the government with a cross-section of stakeholders. In regard to control, Owen (1996), in his analysis of the PPA in Mozambique, discusses the difficulty of satisfying the demands of mul- tiple stakeholders. He asks, "Whose PPA is this?" Diverse and some- times conflicting demands have the potential to undermine the participatory nature of the PPA, with the institutions that control the process wanting to produce documents according to predetermined deadlines and documents that represent their point of view. Owen fur- ther points out that where control has been relinquished there may be a trade-off between ownership and quality. Box 10 discusses the com- plexity of achieving ownership even where a participatory process has been adopted. Control and ownership of the PPA are also linked with the govern- ment's ability to negotiate with the World Bank. Generally, if donors adopt a top-down approach to assisting in policy formulation, there will be limited ownership and commitment on the part of the govern- ment. Several government officials in Guatemala felt excluded from the PPA process, and relations between the Bank and the university that undertook the PPA were weak and antagonistic. Ownership of and commitment to the PPA results were, therefore, limited until the uni- versity published an independent document on poverty in the country, without any World Bank input. Although the information from PPAs might be relevant and result in changes to policy documents, without ownership there will be no long- term shifts in attitude. It is recommended that for greater ownership, the research agenda should not be determined solely in Washington. Those who influence policy in country should be part of the discussion. This process might take much longer than anticipated, so the PPA design should be flexible to accommodate unexpected delays. Delays become more likely as more stakeholders become involved, and it is not always possible to predict how or even if consensus will be achieved (see Box 11). EMERGING GOOD PRACTICE 51 Box 10. Handing Over the Document Does Not Equal Ownership Zambia: There was extensive dialogue with a cross-section of stakehold- ers in the Zambian poverty assessment and PPA and, as a consequence, there was a strong and widely shared feeling of ownership of the process and the action plan. The Zambians drafted the recommendations section of the poverty assessment. However, in discussions with the local research team in Zambia,* one government official asked about the PPA: What is there on this document's cover to show that it is owned by the gov- ernment? There is no coat of arms or government logo, no preface by any government official. An NGO representative added The World Bank calls a national workshop at Mulungushi International Conference Center, introduces the poverty assessment, and hands over the ownership of the poverty assessment to the Permanent Secretary chairing the workshop. Just like that and the Bank thinks it has resolved the own- ership issue. It had been clearly stated and widely understood from the beginning that the poverty assessment was a Bank document. Although one objective is government ownership, it might not be appropriate to expect some gov- emments to feel ownership of documents that were initiated in Wash- ington and carry the World Bank logo. Some governments might not even want ownership, but might want the document to remain identified as a Bank document in order to promote an independent assessment. However, in other cases it might be appropriate for the Bank and the gov- ernment to publish a joint document. South Africa: The PPA included key policymakers from the beginning and ownership gradually developed among high level stakeholders. For example, the cabinet met twice to discuss the PPA. The first meeting took two hours and was chaired by Thabo Mbeki, the Deputy President of South Africa. *For this study, a local research team was contracted to review the process and impact of the PPA. For afull report, see Mutesa and Muyakwa (1997). 52 CAN THE POOR INFLUENCE POLICY? Box 11. Participation Is More Than Holding Workshops Pakistan: The poverty assessment was the first economic-sector work in Pakistan to be widely disseminated and discussed. The workshops were followed by many positive press reports and increased awareness of poverty issues. The process helped encourage the government to form a group specifically to look at poverty issues. There was a general feeling that the poverty assessment was a good analysis but that it was too narrow because it used only the consumption measurement of poverty. How to measure poverty was the subject of extensive debate. Government officials and NGOs felt that the main mes- sage from the assessment was that poverty in Pakistan had declined. This was disputed by some Pakistani economists, who stated that different measurements would produce different results, and by NGOs that had extensive countrywide experience. Stakeholder views had been expressed in various workshops for the poverty assessment but many felt these views had not been adequately reflected in the final document. As a consequence, some challenged its objectives. One senior government official had attended many work- shops but felt that his extensive participation during the workshops and written comments had not been considered. The question was raised about which institution controlled the research agenda and outcome of the poverty assessment. The main message from this experience is that participating in work- shops is not the end of a process of participation. A final consensus might not be feasible, so differing views should be reflected in the final docu- ment. Furthermore, if people's views are not included, that should be explained. A recommendation is that PPA and poverty assessment man- agers should know how to organize workshops and do appropriate follow-up, including incorporating the views of all participants in the research results where possible, or at least the main themes emerging from the research. The quality and follow-up of workshops will affect both the impact of the PPA and the relationship among participating stakeholders. 3.2.4 Strengthening the policy delivery framework Policy change is not just about writing a new policy document-it is also about implementing that policy. To link policy formulation to implementation, good practice is to focus on * Increasing in-country capacity for ongoing research; EMERGING GOOD PRACTICE 53 * Creating channels for ongoing dialogue among a cross-section of stakeholders; * Opening up a process of continual negotiation on the political agenda, in which the views of the poor are taken into account; and * Maintaining partnerships. In most countries, it will be important to build a constituency for reform beyond the government because societies are becoming increas- ingly pluralistic and change often depends on a variety of partnerships. The role of other international donors, which have the power to influ- ence national policy, should also be considered. The United Nations Development Programme (UNDP) is currently undertaking poverty analysis in some countries using participatory methods. In Togo, the UNDP was a partner in the PPA exercise, and its resident mission con- tinues to promote participatory analysis. In Ecuador, UNICEF used PPA methodologies to evaluate the impact of its program. Some PPAs have been carried out in partnership with institutions that specialize in social research (universities, networks of social scien- tists, etc.). Such partnerships help to increase the capacity of such insti- tutions while avoiding the duplication of research and helping to ensure that PPAs become part of the body of social knowledge. The process of policy implementation often alters intended policies. It is, therefore, important to understand the linkages between intention (policy) and outcome (implementation), and identify and include those who will implement policy in the policy dialogue. Administrators at the central and local levels must be included in the PPA. To increase understanding of the various research approaches, it is also crucial to include statisticians from line ministries. For example, in Kenya, the Central Bureau of Statistics assisted in coordinating the PPA. Because governments and donors have traditionally focused on sec- tors as opposed to cross-cutting themes, it might be difficult to place participatory research results within one institution (see Box 12). A rec- ommendation is to identify an institution in country where such data could be analyzed, coordinated, and disseminated. Many countries have collected great quantities of participatory data but lack follow-up and coordination. Finding an entry point for participatory research results might encourage more continuous research by a cross-section of institutions, thus contributing to broadening the policy dialogue and eventually to increased government and Bank commitment to poverty alleviation. See Box 13 for a summary of this section. 54 CAN THE POOR INFLUENCE POLICY? Box 12. Where Is the Entry Point for Participatory Research Data? Considering the Institutional Framework In Lusaka, Zambia, the institutional implications of trying to more effec- tively integrate poverty issues and participatory research data were debated at a workshop convened for this study. Within the government, there has, until recently, been no common entry point for poverty issues,* so it was difficult to integrate participatory research results into the national planning process. In addition, policymakers tend to be sector biased and not to focus on cross-cutting themes. As a result, the work- shop proposed that there be a focal institution for participatorv research data, similar to the Central Statistics Office. There were two suggestions- a Policy Analysis Unit in the Cabinet and the soon-to-be-established Human Development Unit in the Ministry of Finance. This focal institu- tion would coordinate and disseminate participatory research results and encourage networking among organizations that carry out participatory research. 'The entry point in Zambia is now the Ministry of Finance and Economic Planning. Box 13. Factors to Consider at the National Level to Increase the Impact of PPAs Understand the political environment * Undertake the PPA only after potential political implications have been thought through. * Use the institutional, formal, personal, and informal structures and networks and understand the impact they have on policymakers. This requires Bank teams to have an in-depth country knowledge. Create a conducive policy environment if possible * Question the value of conducting a PPA where there is limited gov- ernment support. * Build dialogue to create a more open climate so that governments feel less threatened by the resulting data. * Maintain a policy dialogue through continuous follow-up with vari- ous stakeholders. * Use personal judgment and attune stakeholder involvement to the overall political, social, economic, and institutional environment in country There is no blueprint approach to the timing of stakeholder inclusion in the policy dialogue. EMERG1NG GOOD PRACTICE 55 Promote ownership * Include key policymakers from the beginning. Develop relationships with and understanding of the key players. * Consider publishing joint documents (World Bank and government) where appropriate. * Know how to organize workshops with appropriate follow-up. Work- shops are not the end of a process of participation. Final consensus might not be achieved so the documents should reflect the differing views. If people's views are not included, that should be explained. The quality and follow-up of workshops will affect the impact of the PPA and the relationship among participating stakeholders. Strengthen the policy deliveryframework * Identify a credible institution where participatory research could be analyzed, coordinated, and disseminated. Investigate provincial capacities. * Work with institutions (universities, networks of social scientists, etc.) already undertaking social research to ensure that research is not duplicated and the PPA becomes part of the body of social knowledge. 3.3 At the Community Level: Including the Poor This section analyzes how to undertake participatory research at the community level, focusing on good practice to achieve credibility and legitimacy of the PPA. The section is divided as follows: 1) Research teams 2) Management of research teams 3) Research process 4) Analysis and synthesis 3.3.1 Research teams COMPOSITION: The composition of the research team working at the community level is usually context specific. In general, men and women should be equally represented, and familiarity with local culture, espe- cially a knowledge of local languages, is essential. In Zambia, for exam- ple, the research team comprised one manager (male), and five male and four female facilitators of mixed ages and ethnicity. This team then split into mixed gender groups of three to four researchers and spent two to three days in each community. In Tanzania, 35 researchers split into teams of five or six and worked in six different provinces. 56 CAN THE POOR INFLUENCE POLICY? PREPARATION: Teams should be well prepared before going to research sites. PPA experience has shown that even where teams are experienced in participatory methods, at least two weeks of training are required to discuss the complexities of undertaking national-level policy analysis; match participatory tools with the research agenda; decide on methods of recording and reporting; create an initial frame- work for analysis of results; build up a team spirit; and discuss atti- tudes and behavior. Compromising on training time leads to poor-quality research. Teams should also be aware of major policies linked to the research agenda before going to communities. SKILLS: The skill and role of facilitators become increasingly impor- tant to achieving credibility when participatory exercises are extended from the project level to the national level for large PPAs. The speed of scaling up, often to fit with donor agendas, has often led to compro- mises on the quality of research. If the facilitation of participatory meth- ods is poor, data could be biased, vulnerable groups excluded, and outcomes inaccurately analyzed. This bad practice has hurt the credi- bility of participatory methods. Good-quality work requires a combi- nation of factors, including a good attitude, technical skills, and experience on the part of the facilitator. In Mexico, it was difficult to find a suitable national consultant who was not politically affiliated to coordinate the PPA. In addition, con- trolling the process of gathering information proved problematic because the teams attempted to follow their own agenda. In Togo, the teams in the field had limited skills to analyze the results. In Mozam- bique, in an internal evaluation of the preliminary research phase, it was concluded that teams were too unfamiliar with the communities to develop trust, and some were not able to apply the methods effectively. The major question now emerging is how to integrate the diverse data sets into a comprehensive analysis of poverty. Some have also argued that integration could be relevant at the data collection stage (see Ravallion 1996; Chung 1997). Integrating quantitative and qualitative research using the same teams has implications for the types of skills required by research teams. Whereas questionnaire surveys require enumerators, participatory research requiresfacilitators who have a com- pletely different set of skills, behaviors, and attitudes. Therefore, although it might not be feasible to expect a team of enumerators to conduct credible participatory research, different teams could be used for different research techniques (for example, PPAs in India [Uttar Pradesh and Bihar] and Kenya). EMERGING GOOD PRACTICE 57 3.3.2 Management of research teams A key issue for good-quality participatory research that is emerging from this study is how to effectively manage research teams. Two key concerns require further investigation: DIVERSE TEAM STRUCTURE: Most PPA research teams have been selected to represent the major groups in society. In Tajikistan, where participatory research was undertaken to support a World Bank poverty alleviation program, team members were selected to reflect the composition of Tajik society. The team consisted of men and women of all ages (college students, middle-aged people, elders) and education levels (from village schoolteachers to doctors and academics), from rural and urban areas, and from all major ethnic groups. The objective was to design a team that was not biased toward any one subgroup, especially the more educated urban elite. During debriefing sessions and informal discussions with field- workers, the research manager was able to gather a great deal of infor- mation as long as she did not show preferential treatment toward any group. This meant breaking some social rules in Tajikistan by making room for the less-educated rural woman to voice her opinion. How- ever, it also meant creating opportunities for the elder male to represent the team in meetings with local officials. The manager stated, "While on the whole this choice had positive results for the team, participation practitioners need to be aware that this minisociety is not necessarily easy to manage" (Elizabeth Gomart, personal communication). Certain members of the team tried to control the discussions based on their societal role. In Tajikistan, social hierarchies are designed along education, age, and gender lines. There is also a hierarchy among regional ethnic groups and among castes within some groups. The manager noted that although she was able to supervise and effectively manage the debriefing sessions, the dominant people were able to take over the report writing, which was done in separate groups. PSYCHOLOGICAL TOLL OF POVERTY RESEARCH: Another challenge is man- aging the psychological toll of poverty research. PPAs, which are based on the premise of seeing poverty from the point of view of the poor, might expose fieldworkers to some degree of trauma for which they are not prepared. In Tajikistan, although the fieldworkers had been involved in surveys and other studies, most of them lived in the capi- tal and had little information about the depth of poverty in the regions. The manager stated that as fieldwork progressed into its second week, 58 CAN THE POOR INFLUENCE POLICY? some fieldworkers broke down as they described their day's work. In Equatorial Guinea, as well, the poverty was more severe than expected and in this case also, fieldworkers broke down during debriefing ses- sions. The outcome is often that fieldworkers feel depleted emotionally and physically, which could affect the quality of their analysis. 3.3.3 Research process SELECTION OF INSTITUTION: Identifying an appropriate institution to undertake the research can be difficult. Local knowledge of credible, neutral institutions is required. In general, PPAs have been more suc- cessful when the selected institution has some existing capacity to undertake participatory research; for example, a research institute, NGO, or social science network. However, some organizations claim to have experience in participatory research but do not have the capacity to undertake good-quality research, thereby compromising the credi- bility of the PPA. To increase credibility, it might be appropriate to use an existing NGO network, where there is often a wealth of knowledge and skills. The advantages of using these networks, as opposed to training new teams, are as follows: * Many NGOs have already established trust with communities and undertaken participatory research; * The results could be followed up by the NGOs working in the com- munities, thereby ensuring that the research is not purely extractive. The limitation here is that the research results would be biased toward communities where the NGO has already had some impact, and the poorest communities might not be included; * The PPA research could help to strengthen the capacity of existing NGO networks; and * Information could be collected by NGOs over time, and links estab- lished among the NGOs, policymakers, and statistical departments. It should be noted, however, that few NGOs have the skills and capacity to undertake good-quality research on a large scale and that some NGOs may have sector biases. RAISING EXPECTATIONS: The research process in some PPAs has been viewed as exploitative because it takes the community's time, raises expectations, and undermines self-reliance. Facilitators should, there- fore, clearly state the objective of their visit. An example of bad practice is producing community wish lists instead of analyzing the commu- EMERGING GOOD PRACTICE 59 nity's needs. Furthermore, if the agency then funds the priority identi- fied on the wish list without community participation and capacity building, dependence on the outside organization increases, commu- nity self-reliance is undermined, and false expectations are raised. PPA researchers in Pakistan, Mozambique, and Zambia reported that some communities expressed hostility toward the research teams, especially where there was extensive research with limited follow-up. In Armenia and Moldova, communities expressed frustration and anx- iety over being involved with many research exercises with no improvement in their situation. In these countries, the fieldworkers also reacted with frustration and some accused the participants of com- plaining rather than doing and being stuck in old ways. The manager of these assessments suggested that the fieldworkers were reflecting the frustration of the participants. TIME SPENT IN COMMUNITIES: In many communities, it is easier and quicker to interact with the local elite, thereby missing the poorest (who are often less articulate, overworked, and unable to attend meetings) and women (who do not often leave their homes and are used to being excluded). To overcome this limitation, facilitators need to be aware of the power relations in the community and the composition of the com- munity as a whole. Some PPAs have rushed the research process in order to meet deadlines, often leaving out the poorest and those on the periphery. The difficulty of undertaking participatory research in urban areas has been an issue in many PPAs (see Norton 1994) where more time and flexibility are required than in rural areas. For example, Moser and Holland (1996) highlight the issue in Jamaica of confidentiality in wealth ranking and fear of being identified as part of the research because of safety In the urban areas in Zambia, it was difficult to iden- tify social groups, and people were occupied and not willing to partic- ipate. In other urban areas there might be a question of safety for the research teams, especially for women researchers, as was the case in Costa Rica and Zambia. TOOLS: There is a widely held belief that for participatory research to be more accurate, the tools and techniques should be standardized. However, flexibility can be strength, for the approach, tools, and tech- niques will vary depending on the community. But in some circum- stances it is possible to use certain standardized participatory methods on a wide scale to generate numeric information. Beneficiary assess- ments have quantified results based on a sampling frame, as in Costa Rica and Madagascar. The UNDP's PPA in Bangladesh used standard- 60 CAN THE POOR INFLUENCE POLICY? ized methods for focus discussion groups and the identification of pri- orities (see UNDP 1996). The utilization survey conducted by Action Aid in Syndhupalchowk, Nepal, used participatory mapping in more than 130 villages to generate service utilization data.2 In some household questionnaire surveys, questions are preset by outsiders and the respondent is not likely to know the interviewer. PPAs that use the participatory rural appraisal tools (visuals and group analysis) typically elicit more accurate responses when * Institutions conducting the research are known and trusted by the communities; * Group dialogue and analysis encourage people to challenge inaccu- rate responses; * Data are triangulated (checked with informants and data sources) to test for accuracy and to find areas that need probing; * Researchers and local people learn from the process; * Marginal groups are targeted; and * Data are analyzed by the community. Skilled facilitators are needed to conduct this type of participatory research. Where skills have been lacking, the accuracy of PPA data has suffered. During the research process, teams can learn from each other in reg- ular meetings where tools and approaches are reviewed and differ- ences among various social groups discussed. Site reports could be compiled as a result of the meetings and later disseminated to commu- nities. Local officials should be included where appropriate and results of the participatory research at the community level shared with them. 3.3.4 Analysis and synthesis Some PPAs have collected valuable information but not all of it has been useful to policymakers. PPAs should try to achieve "optimal igno- rance" (Chambers 1993) so that information is collected only on issues relevant to policymaking. Careful selection of methods that link to the identified research issues is required. In Mozambique, the PPA pre- sented too much information to the policymaker. This was the result of the lack of coordination between the research agenda and methods applied in the field, as well as the reporting style of the coordinating institution. PPAs have achieved less credibility when the results have been too broad, too obvious, or too complex for policy use. EMERGING GOOD PRACTICE 61 Research teams should begin to analyze information during the research process. However, analysis and synthesis require highly trained teams to ensure that research results are valid. In South Africa, a two-day workshop was convened for the PPA researchers on report writing, and card sorting techniques were used with the communities to analyze the material and determine categories for the reports. Poli- cymakers could be involved at this early analysis stage to better under- stand the process. Quick and early feedback to key individuals could help policymakers understand the preliminary findings and feel some early ownership before the final report is issued. In the process of data collection, analysis, and synthesis, the key question is, Who controls the selection of data used to influence policy? The handling of data is often determined by power relations. In most traditional surveys, control remains in the hands of those outside the community, especially in * Designing the questionnaire, which is inflexible and based on what policymakers want to know; * Asking the questions, with control remaining with the interviewer and respondents often feeling inhibited by power differentials (espe- cially between educated enumerator with paper and pen and illiter- ate respondent; male enumerator and female respondent; urban enumerator and rural respondent). Respondents frequently react to such power relations by telling the enumerators what they want to hear; and • Analyzing the data, which remains outside the community, as does the control of the publication. Once results are accepted, they are then repeated and, consequently, widely believed. Chambers (1997) notes, "To this day, the extent to which survey results are socially and personally constructed remains under-researched, under-reported and under-recognized" (p. 95). PPAs have generally been able to give more control to those who are being studied. Even though the analysis of the PPA is controlled by the community to some extent, local people often lose control when the information is translated into macro policy messages and the results are aggregated.3 Thus, although complex and detailed community- level information is valuable for project design, inaccuracies sometimes arise from extrapolating to the national level for policymaking. Some PPAs have been able to overcome this limitation. In Zambia, for exam- ple, although the number of communities in the PPA was small, they 62 CAN THE POOR INFLUENCE POLICY? had certain uniform characteristics such as timing of school fees in December and January, which was a time of stress for rural areas all over the country. The simple message on school fees, therefore, remained accurate on the national level. In general, however, different types of conditions and whether they can be generalized for policy pur- poses need to be identified. In both PPAs and traditional surveys, bias emerges through the inter- pretation of answers and, most critically, the analysis of results. In par- ticipatory research, changing the relationship between those from outside the community who are undertaking the research and the com- munity members is not an easy process. In a few cases the outsider facilitator remained dominant and community members tended to say what they thought the facilitator wanted to hear. PRA visual exercises can help to reduce such distortions by opening up the discussion and analysis. But some distortions might still exist, because the process of compiling PPA results involves many stages of information filtering (see Figure 7). When one is synthesizing data, it is not always possible to directly compare different data sources. In any case, comparing results should be avoided because it can set up a false dichotomy. However, compar- isons might be useful as a starting point in the analysis. For example, in the Kenya PPA (see Narayan and Nyamwaya, 1996), the results of the Poverty Profiles from the Welfare Monitoring Survey 1992 were com- pared with the PPA. The report concluded that in three of the five dis- tricts the results of the two approaches were almost identical, with a similar percentage of people falling below the conventional poverty line. And in Tanzania, the PPA report noted the similarity of results from two separate surveys: 50.3 percent were identified as poor and very poor by the PPA, while 49.7 percent fell below the poverty line in the Human Resources Development Survey (HRD) (Narayan 1997). It should be noted, however, that these data sets are not directly comparable and that such comparisons have value only as an indicator for further investigation, for the poor might be defined differently in each case. See Box 14 for a summary of this entire section. EMERGING GOOD PRACTICE 63 Figure 7. Information Filters and Biases: Case Study of the PPA in Zambia REALITY Community Selection of community members and methods Research team's analysis of site reports Implementing agency's clean-up of site reports Incorporation of site report analyses into PPA by World Bank staff in Washington Areas of information filtering L and potential biases Incorporation of PPA and potential biases information into the final poverty PssLIsmenC Poverty assessment used by policymakers POLICY 64 CAN THE POOR INFLUENCE POLICY? Box 14. Factors to Consider at the Community Level to Increase Impact of the PPA Research teams * Develop trust between the research teams and communities. * Be aware of bad practice in participatory rural appraisals (PRAs). Facilitators need experience, skills in applying the tools, and the abil- ity to hand over control. * Training of teams takes at least two weeks to discuss the complexities of undertaking national-level policy analysis; match participatory tools with the research agenda; decide on methods of recording and reporting; create an initial framework for analysis of results; build up a team spirit; and discuss attitudes and behavior. Experience has shown that compromising on training time leads to poor-quality research. * Teams should be aware of major policies linked to research agenda before going to communities. Management of research teams * Be aware of the difficulties in managing diverse research teams that often represent different ages, genders, and ethnic groups. * Research teams working with poor communities may experience some degree of trauma for which they are not prepared. Managers should understand that this outcome is possible. This is an emerging issue and more training is required for both field researchers and managers to find ways in which such outcomes can be better managed. Research process * Share information with communities on an ongoing basis. * Do not undermine community self-reliance. * Be aware of respondent fatigue and raising expectations. Many com- munities-especially those accessible from major cities-are the sub- ject of excessive research. * Before initiating any study, review existing data and material pertain- ing to the area. * Identify credible, not just experienced, institutions to undertake research. Use existing NGO networks where appropriate to promote follow-up. * Allow for more flexibility in urban than in rural areas. * Link results of PPA with other institutions for follow-up. * Write clear site reports to disseminate to communities. * Recognize the limitations of the PPA. Participatory poverty research is not a methodology for empowerment. EMERGING GOOD PRACTICE 65 Methodologies * Adapt the methodologies to the research agenda. * For greater community-level analysis and ownership, use PRA. Be aware of the dangers of rapidly scaling-up PRA methods, which can undermine the quality of the research. * Avoid biases-triangulate data. * Quantify and record the number of people involved in the participa- tory research. Analysis and synthesis * Understand the difficulties of drawing macro conclusions from micro analysis. * Present clear policy messages-do not present everything. Notes 1. Some of the issues highlighted here may be appropriate only for the World Bank. However, the author hopes that other institutions will find the Bank's experience useful. 2. See Mukherjee (1995) for other participatory methods being used to gen- erate commensurate data. 3. See Attwood (1996) for a case study of this issue in South Africa. The Challenge for Participatory Poverty Assessments The moral imperative for giving the poor a voice in the poverty debate Tis self evident. The bonus is that engaging with the poor also leads to better technical diagnosis of the problem, and better design and imple- mentation of the solution. Through PPAs, the poor deepen our under- standing of poverty and can influence policymaking. This new approach challenges traditional power relations and calls for a variety of partner- ships that require trust, openness, and integrity. Both poverty and policy change are inherently linked to the political process in any country. But when undertaken in an environment of increased trust, PPAs can present opportunities for a more open dia- logue and greater understanding between the powerless and those in power. Such dialogue is the beginning of wisdom-and the beginning of a journey that can lead from hopelessness to opportunity. 66 Bibliography Attwood, H. 1996. South African Participatory Poverty Assessment Process: Were the Voices of the Poor Heard? Paper for the PRA and Policy Workshop, Institute of Developmental Studies, Sussex (May 13-14). Beall, J., N. Kanji, F. Faruqi, C. Mohammed Hussain, and M. Mirani. 1994. Social Safety Nets and Social Networks: Their Role in Poverty Alleviation in Pak- istan, Volume 1, Executive Summary and Report. Commissioned by the Overseas Development Administration, United Kingdom. Carvalho, S., and H. White. 1997. Combining the Quantitative and Qualitative Approaches to Poverty Measurement and Analysis: The Practice and the Potential. World Bank Technical Paper No. 366. Washington, D.C. Chambers, R. 1983. Putting the Last First. London: Longman. 1993. Challenging the Professions. Frontiers for Rural Development. Lon- don: Intermediate Technology Publications. 1997. WNhose Reality Counts. Putting the First Last. London: Intermediate Technology Publications. Chung, K. 1997. Combining Quantitative and Qualitative Techniques for Improved Policy Analysis, Draft paper for the Policy Research Department, World Bank. Washington, D.C. Clark, J. 1992. Participatory Country Poverty Assessments. Draft note. Washing- ton, D.C.: World Bank. Clark, J., and L. Salmen. 1993. Participatory Poverty Assessments: Incorporating Poor People's Perspectives into Poverty Assessment Work. Washington, D.C.: World Bank. Clay, E. J., and B. B. Schaffer. 1984. Roomfor Maneuver. London: Heinmann. Conyers, D. 1982. An Introduction to Social Planning in the Third World. Chich- ester, England: Wiley. Dogbe, T. 1996. The One Who Rides the Donkey Does Not Know the Ground Is Hot. Paper for the PRA and Policy Workshop, Institute of Development Stud- ies, Sussex (May 13-14). Gaventa, J. 1998. "Poverty, Participation, and Social Exclusion in North and South." IDS Bulletin 29(1) (January). Gomart, E., Personal Communication. July 1997, Washington, D.C. Govemment of The Gambia. 1994. Strategy for Poverty Alleviation. Geneva. 67 68 CAN THE POOR INFLUENCE POLICY? Government of Lesotho. 1996. Pathway Out of Poverty. An Action Planfor Lesotho. Lesotho. Grindle, M.S., and J.W. Thomas. 1991. Public Choices and Policy Change. The Polit- ical Economy of Reform in Developing Countries. Baltimore and London: The Johns Hopkins University Press. Guba, E., and Y. Lincoln. 1985. Effective Evaluation. California: Jossey-Bass Publishers. 1996. "Competing Paradigms in Qualitative Research." In N. Denzin and Y. Lincoln. Handbook of Qualitative Research. London: Sage Publica- tions. Gujja, B., M. Pimbert, and M. Shah. 1996. Village Voices Challenging Wetland Management Polices: PRA experiences from Pakistan and India. Paper for the PRA and Policy Workshop, Institute of Development Studies, Sussex (May 13-14). Holland, J., and J. Blackburn, eds. 1998, Whose Voice? Participatory Research and Policy Change. Rugby, England: Intermediate Technology Group. Inglis, A., and S. Guy. 1996. Scottish Forest Policy "U-turn"`-Was PRA in Laggan Behind It? Paper for the PRA and Policy Workshop, Institute of Devel- opment Studies, Sussex (May 13-14). Institute of Development Studies. 1989. Bulletin on Vulnerability. Sussex. Institute of Development Studies. 1994. Poverty Assessments and Public Expendi- ture: A Studyfor the SPA Working Group on Poverty and Social Policy. Sussex. International Institute of Environment and Development. 1991-1998. RRA Notes (now PLA Notes). London: Sustainable Agriculture Programme. Johda, N.S. 1988. "Poverty Debate in India: A Minority View." Economic and Political Weekly. India (November). Marshall, C., and G. Rossman. 1995. Designing Qualitative Research. London: Sage Publications. May, J., and H. Attwood. 1996. Kicking Down the Doors and Lighting Fires: the South African PPA. Paper for the PPA and Policy Workshop, Institute of Development Studies, Sussex (May 13-14). Milimo, J. 1995, Coping with Cost Sharing in Zambia. Stockholm: Swedish Inter- national Development Authority (Sida). 1996. A Note on the Use of PRA Approaches and Methods and Their Impacts on Policy and Practice in Zambia. Paper for the PRA and Policy Work- shop. Institute of Development Studies, Sussex (May 13-14). Milimo, J., A. Norton, and D. Owen. 1998. "The Impact of PRA Approaches and Methods on Policy and Practice: The Zambia PPA." In J. Holland and J. Blackburn, eds. 1998, Whose Voice? Participatory Research and Policy Change. Rugby, England: Intermediate Technology Group, pp. 103-111. Moser, C. 1993. Gender Planning and Development: Theory Practice and Training. London: Routledge. Moser, C., and J. Holland. 1996. A Participatory Study of Urban Poverty and Vio- lence in Jamaica: Analysis of Research Results. Urban Development Divi- sion, World Bank. Washington, D.C. BIBLIOGRAPHY 69 Mukherjee, N. 1995. Participatory Rural Appraisals and Questionnaire Surveys: Comparative Field Experience and Methodological Innovations. New Delhi: Concept Publishing Company. Mutesa, F., and S. Muyakwa. 1997. Participatory Poverty Assessment Review: Reportfor the World Bank, Lusaka. Washington, D.C.: World Bank. Narayan, D. 1997. Voices of the Poor. Poverty and Social Capital in Tanzania.Wash- ington, D.C.: World Bank. Narayan, D., and D. Nyamwaya. 1996. Learningfrom the Poor: A Participatory Poverty Assessment of Kenya. Environment Department Paper No. 034. Washington, D.C.: World Bank. Norton, A. 1994. "Observations on Urban Applications of PRA Methods from Ghana and Zambia: Participatory Poverty Assessments." PLA Notes 21: 55-6. 1996. The Ghana Participatory Poverty Assessment: Some Notes on the Process and Lessons Learned. Paper for the PRA and Policy Workshop, Institute of Development Studies, Sussex, (May 13-14). Norton, A., and P. Francis. 1992. Participatory Poverty Assessments in Ghana. Draft discussion note and proposal. Washington, D.C.: World Bank. Norton, A., and T. Stephens. 1995. Participation in Poverty Assessments. Environ- ment Department, World Bank, Washington, D.C. Owen, D. 1994. Mozambique Participatory Poverty Assessment Discussion Paper and Proposal. July Washington, D.C.: World Bank. 1996. The Mozambique Participatory Poverty Assessment-Lessonsfrom the Process, Paper for the PRA and Policy Workshop. Institute of Develop- ment Studies, Sussex (May 13-14). 1997. Mozambique Participatory Poverty Assessment. Stakeholder Workshop Report. Washington, D.C.: World Bank. Parker, B. 1994. Pakistan Poverty Assessment: Human Resource Development: A Social Analysis of Constraints. Paper prepared for the Pakistan Poverty Assessment. Participatory Assessment Group. 1995. Participatory Poverty Monitoring in Zam- bia. Lusaka. Ravallion, M. 1996. "How Can Method Substitute for Data? Five Experiments in Poverty Analysis." The World Bank Observer 11(2): 199-221. Rietbergen-McCracken, J., and D. Narayan. 1997. Participatory Tools and Tech- niques: A Resource Kit for Participation and Social Assessments. Washing- ton, D.C.: World Bank. Robb, C. 1995. Moving Towards a Process Approach to Policy Formulation in The Gambia. Paper presented at the World Bank/DFID Conference, "Finance Against Poverty: Challenges and Advances in Banking with the Poor." University of Reading (March 27-28). 1997a. "PPAs: A Review of the World Bank's Experience. " In J. Holland and J. Blackburn, eds., 1997, Whose Voice? Participatory Research and Pol- icy Change. Rugby: Intermediate Technology Group. 1997b. Increasing the Impact of Participatory Research: Lusaka, Zambia. Summary of a workshop convened in Lusaka as part of the PPA Review. 70 CAN THE POOR INFLUENCE POLICY? - 1998. Social Impacts of the East Asian Crisis: Perceptions from Poor Com- munities. Paper prepared for the East Asian Crisis Workshop. Institute of Development Studies, Sussex (July 13-14). Robb, C., and C. Zhang. 1998. Social Aspects of the Crisis: Perceptions of Poor Com- munities in Thailand. Washington, D.C.: World Bank. Salmen, L. 1992a. Participatory Poverty Assessment. Draft note. Washington, D.C.: World Bank. 1992b. Reducing Poverty: An Institutional Perspective, Poverty and Social Policy Series, Paper No. 1. World Bank, Washington, D.C. . 1995a. Participatory Poverty Assessment. Incorporating Poor People's Per- spectives into Poverty Assessment Work. Social Assessment Series, Paper No. 024. World Bank, Washington, D.C. 1995b. Beneficiary Assessment: An Approach Described, Environment Department Paper 023. World Bank, Washington, D.C. 1995c. "Listening to the People." In Finance and Development 32(2): 44-48, World Bank and International Monetary Fund (quarterly). Shah, M., and G. Nikhama. 1996. Listening to Young Voices: Participatory Appraisals on Adolescent Sexual and Reproductive Health in Chawama Com- pound, CARE International, Zambia. Lusaka. Srinivasan, L. 1990. Tools for Community Participation: A Manual for Training Trainers in Participatory Techniques. New York: PROWWESS/UNDP. Stone, L., and J. Gabriel Campbell. 1984. "The Use and Misuse of Surveys in International Development: An Experiment from Nepal." Society for Applied Anthropology 43(1). UEM (Universidade Eduardo Mondlane). 1996. A Pobreza em Mozambique: Um Estudo Particpativo. Relat6rio da Primeira Fase. CEP/95/003 - Doc 22b. Centro de Estudos da Populan,co (May). 1997. Summary Report. Centro de Estudos da Populacao (March). United Nations Development Programme (UNDP). 1996. UNDP's 1996 Report in Human Development in Bangladesh. A Pro-Poor Agenda. Volume 3: Poor People's Perspectives. Dhaka, Bangladesh. United Nations Development Programme (UNDP). 1997. Prospects for Sustain- able Human Development in Zambia: More Choices for Our People. University of Cape Town. 1995. Key Indicators of Poverty in South Africa. Ministry in the Office of the President, Reconstruction and Development Pro- gram. South Africa. Wildavsky A. 1979. The Art and Craft of Policy Analysis: Speaking Truth to Power. Boston: Little, Brown and Co. World Bank. 1990. World Development Report 1990. Washington, D.C. 1991. Assistance Strategies to Reduce Poverty. Washington, D.C. 1992. Poverty Reduction Handbook. Washington, D.C. 1993a. Tanzania. A Poverty Profile. Report No. 12298-TA. Washington, D.C. (December). 1993b. Mali Assessment of Living Conditions. Report No. 11842-MLI Washington, D.C. (June). 1994a. Rwanda. Poverty Reduction and Sustainable Growth. Report No. 12465-RW. Washington, D.C. (May). BIBLIOGRAPHY 71 1994b. Benin. Toward a Poverty Alleviation Strategy. Report No. 12706- BEN. Washington, D.C. (August). 1994c. Zambia Poverty Assessment: Volume 5: Participatory Poverty Assess- ment. Report No. 12985-ZA. Washington, D.C. (November). 1994d. The World Bank and Participation. Operations Policy Department. Washington, D.C. 1995a. Kenya Poverty Assessment. Report No. 13152-KE. Washington, D.C. (March). 1995b. Cameroon. Diversity, Growth, and Poverty Reduction. Report No. 13167-CM. Washington, D.C. (April). 1995c. Guatemala. An Assessment of Poverty. Report No. 12312-GU. Wash- ington, D.C. (April). 1995d. Brazil Poverty Assessment. Report No. 14323-BR. Washington, D.C. (June). 1995e. Ghana Poverty Past, Present and Future. Report No. 14504-GH. Washington, D.C. (une). 1995f. Uganda. The Challenge of Growth and Poverty Reduction. Report No. 14313-UG. Washington, D.C. (June). 1995g. Lesotho Poverty Assessment. Report No. 13171-LSO. Washington, D.C. (August). 1995h. Reducing Poverty in Zambia: Gettingfrom Ideas to Action. Wash- ington, D.C. 1995i. Pakistan Poverty Assessment. Report 14397-PAK. Washington, D.C. (September). 1996a. Malawi Human Resources and Poverty. Profile and Priorities for Action. Report No. 15437-MAI. Washington, D.C. (March). 1996b. Nigeria Poverty in the Midst of Plenty. The Challenge of Growth with Inclusion. A World Bank Poverty Assessment. Report No. 14733-UNI. Washington, D.C. (May). 1996c. Armenia Confronting Poverty Issues. Report 15693-AM. Washing- ton, D.C. (June). 1996d. Mexico: Poverty Reduction. The Unfinished Agenda. Report No. 15692-ME. Washington, D.C. (June). 1996e. Togo. Overcoming the Crisis, Overcoming Poverty. A World Bank Poverty Assessment. Report No. 15526-TO. Washington, D.C. (June). 1996f. Niger Poverty Assessment. A Resilient People in a Harsh Environ- ment. Report No. 15344-NIR. Washington, D.C. (June). 1996g. Azerbaijan Poverty Assessment. Report No. 15601-AZ. Washing- ton, D.C. (June). 1996h. Madagascar Poverty Assessment, Volumes I and II, Report No. 14044-MAG. Washington, D.C. (June). 1996i. Taking Actionfor Poverty Reduction in Sub-Saharan Africa: Report of an African Regional Task Force. Human Resources and Poverty Division, Technical Department, Africa Region. Report No. 15575-AFR. Wash- ington, D.C. 1996j. Poverty Assessment: A Progress Review, Operations Evaluation Department, Report No. 15881. Washington, D.C. 72 CAN THE POOR INFLUENCE POLICY? - 1996k. Mozambique Participatory Poverty Assessment Phase I Rural Sum- mary. Maputo (une). . 1997a. Costa Rica. Identifying the Social Needs of the Poor: An Update. Report No. 15449-CR. Washington, D.C. (May). . 1997b. Swaziland: Poverty Assessment by the Poor. Draft (February). .1997c. Poverty Reduction and the World Bank. Progress in Fiscal 1996 and 1997. Washington, D.C. Wuyts, M., M. Mackintosh, and T. Hewitt. 1992. Development Policy and Public Action. London: Oxford University Press. Annexes Annex 1. Methodology Annex 2. Impact Annex 3. Poverty Assessments, Completed and Scheduled Annex 4. Examples of PRA Exercises in Thailand (1998) and Zambia (1996) Annex 5. Country Case Examples Annex 6. Methodology of This Review Annex 1. Methodology Country, Timing, and Cost Context Bank Context In-country Institutions Involved Methodology Level of Participation AFRICA Benin Manager of the PPA also Linmited permission sought A unit in the Ministry of Plan- RRA: Twenty-three villages and some Communities: Information sharing. Partial PPA completed; responsible for overall from central goverrnent, ning assisted with coordina- urban communities were covered in five three weeks field work PA. Outside consultant which was supportive of ion. Several NGOs were of the regions (the sixth had already been Others: The government was coop- in 1993. assisted in the PPA. Van- the approach. Local gov- consulted. extensively covered); RRAs involved crative and receptive. Discussions ous divisions in the emment extensively semistructured interviews, children's with NGOs and government dur- Cost: $40,000 Bank consulted at all involved. Stable political drawings. ing management workshop. stages. environment. Burundi Manager of the PPA also Government strongly sup- The study was coordinated PRA: A lst of criteria was agreed upon Some participation within the Bank, Oct-Dec 1997 responsible for overall ported the PPA as part of by a Poverty Committee con- with government for selecting the com- with country economist reviewing PA. The PA and PPA the wider PA, although vened by the Ministry of Plan- munities, including degree of impact of topic lists and preliminary results. Cost: $130,000 were requested and some were skeptcal about ning, and the study brought the conflict, proportion of the community strongly sponsored by the lack of statistical signif- together UNDP and the Bank displaced, socioeconomic status, degree Others: Strong participation from the country director. An icance of results. New lend- as the two main partners for of isolation from roads and markets, government, with senior officials external consultant ing in Burundi had been poverty reduction. At the sug- access to social infrastructure, and agro- attending PRA training course, for trained at IDS Sussex, suspended following the gestion of the Bank, the com- chmabc zone. Ten communities were cov- example. Very actve participation provided training to the 1996 coup, and the new mittee was widened to ered. Results presented in summary at a from UNDP local consultants who government believed that include other key ministries in technical workshop before the report was carried out the PPA. a favorable assessment in poverty reduction issues. written. the PA could lead to new lending. Cameroon Human Resources was Debt-distressed country. CARE-Cameroon with sup- BA: In-country four-day technical work- Communites: Information sharing. Mar-Sept 1994 the managing division. No longer an IBRD coun- port from CARE-Canada pro- shop followed by national-level confer- Manager for PA also try. CFA devaluation. Some vided a technical advisor for ence for one day A technical workshop Information sharing with selected Cost: $150,000 managed the PPA and key policymakers reticent the PPA and carried out two was organized in Kribi and a national institutions and with the govem- was part of Africa's to support the PA and PPA of the five regional conference in Yaounde in November ment at different levels. Technical Department. processes. assessments. University of 1994. At these workshops, broad-based ENVSP assisted in the Yaounde, ASAFE, and PAID discussions of the PA and the views of the PPA. Involvement in carried out the other three poor and some key NGOs were used to and ownership of PA regional assessments. redefine key priorites for the poverty by country and sector reduction strategy. BA used in six regions departments were lim- and included 1,559 households at about ited. The country econo- 30 sites, as well as 150 interviews with mist focused on issues key informants-local government offi- surrounding the CFA cials, community leaders, service devaluation, with coun- providers, and church and women's try department priorities groups. Fifty percent of those interviewed shifting to development were women. Range of participatory of new lending and teclusiques used in regiontal assessments. resumption of adjust- ment support as opposed to poverty. Country department was restructured and management team changed during the course of the PA. Equatorial Guinea A Poverty Note was Government involvement FAO. Local government offi- RRA: Fifteen villages and two urban Communities: Information sharing. Two weeks fieldwork written rather than a was minimal. cials interviewed. communities in the capital and two in Limited dialogue with key stake- in 1995. full-fledged PA. COD another city. holders. cn requested that fieldwork Italian trust funds be conducted because of a lack of reliable data. Ethiopia The PPA was intended The PPA was jointly coor- Freelance Ethiopian consul- PRA: Six rural and four urban sites with a Good participation within the March-April 1997 to complement quantita- dinated by the Ministry of tants were employed. The mix of socioeconomic levels, different Bank, with the country economist tive analysis performed Planning and the Bank. PPA teams collaborated agroclimatic zones, and different levels of and resident mission strongly sup- Cost: $100,000 in preparation for a full Central govemment orga- closely with a PRA-based isolation from roads and markets. Full porting the PPA. PA and CAS. The task nized approvals to enter study on women being con- use of tools: wealth ranking, causal dia- manager for the PPA rural villages, without ducted at the same time by grams, pie charts, timelines, seasonal cal- Poor participation by in-country was not the task man- which it would not have the government's Women's endars, daily calendars, Venn diagrams. NGOs. ager for the full assess- been possible to conduct Affairs Office. Teams were Urban teams developed new tools to use ment. The PPA was the fieldwork. trained by an external consul- for analysis of unemployment. financed by the Dutch tant from IDS Sussex. Trust Fund for Poverty. (Table continues on the following page.) Annex 1. (continued) Country, Timing, and Cost Context Bank Context In-country Institutions Involved Methodology Level of Participation Gabon A typical IBRD country The government welcomed Freelance Gabonese consul- RRA: Team of five people. Participant Communities: Information sharing. Fieldwork: four weeks with a gross domestic the Bank's initiative to tants (including students and observations, case studies, individual and during May-June 1995. product per capita of carry out the poverty a university professor) recom- group interviews; four out of nine regions Others: Donors, government, and, Results included in more than US$4,000 and assessment (including the mended by UNDP, the Plan- covered; 325 qualitative interviews con- to a lesser extent, civil society March 1997 poverty extremely unequal PPA), which was viewed as ning Ministry, and the ducted (80 in Libreville, 140 in small involved through the Interminister- assessment. income distnbution. a means to (1) collect infor- Employment Office. Very cities, and 105 in mural areas). ial Committee. Bank's exposure is lim- mation on poverty; (2) weak in-country NGO Cost: $49,000, of which ited. Initially, limited obtain technical policy rec- capacity (both national and about $19,000 was resources allocated to ommendations from the international). spent on local costs the poverty assessment. Bank; and (3) possibly send (mainly consultant fees The PPA was cofinanced a signal to the donor com- and travel) and was by the French (interna- munity that reduction of financed by the Chent tional consultant) and poverty will require better- Consultant Fund: and the Client Consultation adapted assistance and cr about $30,000 ($19,000 Fund (local survey team closer donor involvement. of which financed by and computer special- The government set up an the French trust fund) ist). The Task Manager Interministerial Technical was spent on the inter- (TM) or the poverty Committee (about 40 mem- national Consultant assessment was also TM bers) to review each ver- who initiated the sur- of the PPA. The Gabon sion of the assessment, and vey and helped ana- PA is a flagship partici- also provided a vehide and lyze results. pation project. driver for the PPA team. Ghana Clear lines of communi- Stable political Teams from academic institu- PRA: Three phases, 15 urban and rural Communities: Information sharing. May-June 1993, cation established environment. Government tons, Ministry of Local Gov- communities. Focus groups and PRAs. April-May 1994, Nov between the PA manager support initially limited erunent and Rural Others: Key policymakers not 1994. Conducted after and technical but now very strong. Development, NGOs, and involved extensively until they quantitative survey department. intematonal aid agencies gained a greater understanding of (Oct 1991-Sept 1992). (especially UNICEF). the PPA. Cost: DFID.UK funded phases I and 2, $50,000. Phase 3 (social service assess) funded by UNICEF, $50,000. Kenya Manager of the PA was Relatively economically AMREF (Regional NGO) and SARAR, PRA, and household question- Communities: Information sharing. Preparation: Feb 1994 initially cautious. The stable for Africa. Govern- the DFIYD UK Filnal document naires aimed at community groups and Fieldwork: March 1994 managers of the PA and ment centralized. Central published by UNICEF/DFID schoolchildren. Seven districts were Others: Local government was Write-up: the PPA were involved Bureau of Statistics and and AMREF selected using information from the cen- more involved than central govern- April and May 1994. in drafting the Terms of Ministry of Planning sor cluster samples. The poorest commu- ment. Final document pub- Reference and preparing involved. nities were then selected: 35 villages and lished in 1995. the PPA. The manager of urban areas in Nairobi; 514 households the PPA coordinated interviewed. Teams spent three days in Cost: $100,000 most of work in-country. each village. Lesotho The new government is UNICEF, Red Cro)ss, NGO PRA: The original PA had no action plan. Communities: Information sharing. Two qualitative sur- open to the inclusion of from Zambia, co4ncil of At a three-day workshop the veys conducted in 1991 stakeholders in the analysis NGOs, and local government. govermnent, NGOs and World Bank Others: The action plan received and 1993. of poverty. There is a repre- A private consulting firm, agreed to draft the action plan. Partici- extensive support from a cross- sentative body of NGOs Sechaba, undertook the PRA. pant observation, case studies, individual section of the stakeholders. which is supported by the and group interviews in rural and urban government but its capac- areas. ity is limited. Madagascar The manager of the PA The PPA was supported by Steering Committee BA: 2,600 qualitative interviews Communities: Informaton sharing. N Eight months was committed to the the Minister of the Econ- composed of key line conducted. Periodic progress reviews N commencing Nov 1993 approach and worked omy and Planning and ministries, parliamnentarians, with UNDP and government committees. Others: Key policymakers have closely with the man- Communications and Cul- NGOs, a national consultancy Four regions. Focus groups of 6-12 peo- been fully involved in a process of ager of the PPA. How- ture but this support is firm, and the university. A ple. Participant observation involved resi- consultation fron the beginning. ever, change in fragmented. local consulting firm for two dence in selected sites for two to three management of the divi- regions, and two groups of weeks. Institutional assessments. sion means follow-up academics for the other two has not been extensive. regions. Several Malagasy consultants and One Canadian consultant coordinated the activities. (Table continues on the following page.) Annex 1. (continued) Country, Timing, and Cost Context Bank Context In-comtnfry Institutions Involved Methodology Level of Participation Mali The managers of the PA Because of the sensitivity Save the Children, CARE, BA conducted in Bamako. RRA in three Communities: Information sharing. Three weeks field- and PPA were able to surrounding poverty, the local university, rural radio. rural regions, semistructured interview- work for Bamako for communicate clearly An PA assessment was ing, and children's drawings. Others: Initially limited participa- BAs and three weeks intemational consultant renamed the Assessment of tion of key stakeholders. of RRA in rural areas also assisted. Living Conditons. Prelimi- (1992-93). nary results of household survey were used. Funded mainly by UNtDP. Mozambique Freestanding Govemment very support- Poverty Alleviation Unit, PRA. Conmmunities: Information sharing. Ongoing. First phase document-not linked ive of the process of collect- established by the World July 1995. to a PA. Manager of the ing qualitative information. Bank, and the university Others: One of the main objectives PPA located in Moputo Govemment has under- undertook the PRA surveys. of the PPA has been to involve a had been involved in the taken its own PA and the NGOs were extensively wide range of stakeholders from Zambia PPA. Continued PPA will feed into it. involved, especially with the beginning. Bank support now problem ranking and prioriti- unclear. zation. Niger The manager of PA The country is plagued A national sociologist super- RRA: Informal interviews, open question- Communities: Information sharing. One month in April worked closely with the with political instability, vised the urban phase. The naires, and focus groups. 1994. manager of the PPA. which has major rural phase received support Others: Involvement of the gov- consequences for achieving from NGOs and several ernment and NGOs has been economic gmwth and regional projects funded by increasing. poverty reduction. The FAO and GT2. PPA was undertaken right after the CFA devaluaton. One of its intentions was to capture the preliminary impact of the devaluation on the poor. Nigeria The manager of the PA The govemment was not DFID, Ministry of Planning, PRA: Focused discussion groups; 2,000 Communities: Information sharing. Late 1993 and early approached DFID for initally supportive. As the NGOs, and UNICEF. No local people in 98 rural and urban locations. 1994. Three months in technical assistance in process developed, how- NGOs were involved in the Others: Local NGOs initially the field. the form of an economist ever, the support increased. PPA work but since the gov- excluded. Govemment included but DFID sent a social The govemment now runs eimnent has taken over, local from the beginning and became scientist. After seeing the its own poverty analysis NGOs are now involved. more involved as the value of the value of the qualitative program. qualitative information became information, the country apparent. team became fully sup- portive of the PPA process. Rwanda The PPA was inibated to The PRA had previously The PPA was carried out in PRA: Ten rural and two urban comrnuni- Bank: Primarily information Oct-Dec 1997 complement quantita- never been used in partnership with Reseau des ties were selected based on degree of sharing. Cost: $150,000 tive survey analysis. Rwanda. In the wake of the Femmes, a women's NGO impact of conflict, proportion of the com- Manager of the PA was genocide, many Rwandans specializing in rural develop- munity displaced, socioeconomic status, Other: Very active participation by also manager of the PPA, were skeptical that com- ment. A local representative of degree of isolation from roads and mar- the govenmment. Government steer- and was based in the munities would be this NGO was present in each kets, access to social infrastructure, popu- ing committee selected the commu- resident mission on a prepared to talk to of the communities, thus lation density, settlement pattem, and nities, amended the question list, short-term assignment outsiders. Govemment was improving trust and speeding agroclimatic zone. Results were discussed seconded a govenmment official to for duration of the PA. involved from the begin- up the process. The team also at a large workshop in Kigali to which participate in the teams, and hotly The PA and PPA were ning through the PA steer- collaborated with UNDP, community leaders from PPA sites were debated results. An official from the strongly supported by ing committee which UNICEF, and FAO to develop invited, together with national and inter- Ministry of Planning was seconded country director, included four government questions. national NGOs, government full-time to work as a counterpart ministries. Developing a departments, and donors. manager of the PA. Also very good regional balance in the participation from Rwandan civil teams was difficult because society, with good attendance at of previous conflict, and meetings to debate terms of refer- team travel was ence of study and results. interrupted frequently because of the conflict. (Table continues on thefollowing page.) Annex 1. (continued) Country, Timing, and Cost Context Bank Context In-country Institutions Involved Methodology Level of Participation South Africa The PPA was initiated to Govemment involvement Worked with a private-sector PRA: Not roving teams, Regionally tar- Communities: In some cases the Ongoing. PPA work- complement the house- sought from the begmning. development research consul- geted. The three poorest provinces were PRA work became a catalyst for shop convened in Feb hold survey, completed Initially distant but now tancy and NGOs. The consul- selected, representing 62 percent of the conmsitTnents to initiate a project to 1995. Final document in Aug 1994. One person very involved and commit- tancy established a manage- poor. The household survey was used to benefit the poor. July/August 1997 manages both the PPA ted through the RDP, ment committee comprising a identify the poorest provinces. Thirty to (bkely to be published and PA. which subsequently closed cross-section of stakeholders fifty conmunrities were involved. The Others: Broad initial consultative by the government). down. In parallel to the selected during the initial approach was to build upon the existing workshop. The PPA process so far PPA, at the Bank's initia- workshop. The govenmment network of NGOs rather than create a has stressed the importance of con- tive, the government, the was represented through the parallel system. PRA training was pro- tinuously including a cross-section Bank, and UNDP are col- Reconstruction and Develop- vided. The existing network had already of stakeholders. Very strong gov- laborating on the PA, now ment Program Office. established trust in many of the commu- emient ownership of the PIR, called the PIR. The PIR was Cofunded by DFID. nities. which incorporated the findings of approved by the Cabinet. the PPA. Interuninisterial Commit- co South Africa may borrow tee on Poverty and Inequality set from the Bank for the first up to oversee the PIR. time snce the 1960s. Swaziland Current lending Poverty debate in govem- University of Swaziland car- PRA and BA: 600 households, 100 focus Communities: Information sharing. 1995. Fieldwork com- program confined to ment was slow to be inib- ried out the PPA. UNDP coor- groups in 63 coummunities throughout pleted. Report forth- urban sector project, ated. PPA and HIlES dinated admirnistration. Swaziland. Focus discussion groups, Others: The government and NGOs coming. under implementation, originally conceived to Support from UFID for the PRAs, and interviews. became increasingly involved. Cost: $99,500 exclud- and proposed education contribute to UNDP national workshop. ing Bank cost. Trust project. PPA not initiated Human Disparities Analy- funded. as part of a Bank PA but sis, which has in effect carried out in tandem become subsumed in with HIES, undertaken process of preparing NOS. by the CSO with support The PPA results have fed of IDF, from which a into this process. The poverty profile is being Poverty Note will be drawn. Bank will now geared specifically to the follow up with Poverty NDS. Note integrating the results of the two exercises. Tanzania Manager of the PA was Governrment was coopera- University of Tanzania (but SARAR, PRA: A team of 36 people visited Communities: Information sharing Preparation: interested in and aware tve and fully involved at capacity linited). 85 villages over 40 days. 6,000 people with no immediate follow-up. Feb/March 1995; of the work being car- the district level. were involved. Fieldwork: May ned out in Tanzania. Others: Govemment was coopera- tive and attended policy Cost: $100,000 workshops, which were coordinated with the Bank's social sector review and CEM preparation. Togo The PA was completed Social unrest prevailed UNDP Fifteen unemployed RRA: Semistructured interviews; infor- Communities: Information sharing. Two weeks Nov 1994; alongside the environ- from 1992 to 1993. Before graduates were trained. One mation sheets; children's drawings one week Feb 1995 mental assessment. COD the PPA, the government team of five and a second depicting poverty. Covered all rural Others: Donors' participation more (seven teams working was very supportive and and the UNDP had already team of ten led by a Dutch regions plus the capital. Forty villages extensive than government's. at the same time). committed to the begun a policy debate consultant. covering five regions and urban neigh- Discussion of results during a series approach. Lines of com- about poverty. borhoods in Lome. of regional workshops with NGOs Systematic Cbent Con- munication were clearly and govermnent sultation Fund established. The Resi- dent Mission was coop- erative. co Uganda The PPA was conducted Civil war in certain areas. Ministry of Planrnig, UNDP, RRA, pictorial drawings. The PPA was Communities: Information sharing. One week 1992. One of with the PA. Government willing to and the University. conducted only in areas where quantita- the first PPAs. accept that poverty exists. bve informabon did not exist: that is, in Others: Involvement of other insti- tfhe war zones. tutions limited because of time con- Cost: $3,000 straints. Zambia The PA and PPA man- Government gradually Nine-person interdisciplinary BA and PRA: Interview guide for semi- Communities: Moved beyond Research Sept-Nov agers worked closely included through the Sys- team of researchers. The team structured interviews with individuals information sharing-the poor 1993 throughout the process. tematic Client Consultation later formed an NGO called and groups. Ten research sites over a vari- were consulted on an ongoing The PA manager had approach. the PAG. ety of communities (urban and rural). basis. PAG returned to the commu- Funded by Sida, supported qualitative nifies on a yearly basis to assess the $100,000 techniques in a previous changes in their welfare/poverty. Bank project in Zambia (Social Recovery Project) Others: Extensive stakeholder con- and promoted the sultation. Zambians drafted the rec- BA/PRA approach in ommendations secions of the PA. the Bank. (Table continues on thefollowing page.) Annex 1. (continued) Country, Timing, and Cost Context Bank Context In-country Institutions Involved Methodology Level of Participation LATIN AMERICA AND CARIBBEAN Argentina Before PPA was under- The government requested Ministry of Social Welfare BA: Conversafional interviews and par- Communities: Information sharing. PPA started in Oct taken, time had been the assistance of the Bank through the direct tial observation. The initial PPA was of 1995. Estimated dura- spent building an under- in conducting qualitative involvement of the minister, limited scope and involved only a few Others: Ministry and Minister of hon six months. PA standing between the research. Good coordina- NGOs. SIEMPRO under the rural areas. The objective of the PPA was Social Welfare fully involved. A unit already completed. technical team and the hion among government Ministry of Social Welfare. to test methodologies and develop nshi- has been established within the COD team, which agencies. The initial activi- PSA and PROINDER under tutional support. In fact, after the initial ministry to monitor poverty and engendered a positive ties were carried out dur- the Secretariat of Agriculture. exercise, PPAs for two provinces (Salta social programs. A seminar has attitude toward the PPA ing the preparation of the and Missions) have been planned. been held with high-level govem- from the outset. Some Social Protection Project ment officials. Strong interest has questions were raised by (Ln AR-35495), particularly already been expressed by other the COD on whether the Component C: Technical departments. NGOs will be Go infornation would be Assistance for the Improve- involved in the execution work. "sound bite"-focused. ment of Social Information The dialogue between the govem- (SIEMPRO). ment and the NGOs has gradually increased. Brazil Focus on education and Strong interest by govem- Ministry of Education and BA Communities: Information sharing. Nine months employment. ment in the qualitative Ministry of Planning approach. Others: Local, state, and federal government Costa Rica Coordinated with the PA The govemment was very No NGOs were involved. The BA in four regions; 262 interviews Communities: Information sharing. 1995, two months manager who lives in supportive of the process. govemment wants to include fieldwork Honduras. Lines of com- Senior officials from the them extensively at the dis- Others: Government was munication between the Ministry of Economic Plan- semination phase. extensively involved from the Cost: $36,500 managers of the PA and ning were involved from beginning. PPA were, therefore, the beginning. often unclear. Ecuador The manager of the PA The government neither UNICEF cofinanced the PRA: seven villages and one urban com- Communities: The PPA was called a Preparation: April 1994 had no access to funds supported nor objected to pmcess. Two NGOs were munity. SSJs and workshops. Rural Qualitative Survey as it was Fieldwork: May 1994 from the Bank and had the PA or the PPA. involved in both the rural and felt that the process was not partici- Meetings with stake- to raise the funding. As urban areas. Government patory but more information shar- holders: Oct 1995 such he was unable to institutions were not exten- ing. NGOs went back to share the recruit consultants from sively involved at any level. results of the studies with several Cost: $70,000 the Bank's technical communities. Dutch Tnist Fund department. From the beginning, the manager Others: Participabon of was able to clearly government institutions was mini- define the information mal. The NGOs were extensively he considered to be involved. Interest of nonparticipat- relevant. ing NGOs was very high. Guatemala The PPA was The government would Liaised with the university BA. Two three-person teams. Average of Communities: Information sharing. Phase 1 in early 1993 undertaken without have liked to be more but relations between the 15 days in every municipality; 223 inter- Phase 2: 3 months, extensive consultation involved in the decision- Bank and the university have views; 22 focus groups. Participatory Others: The govemment produced Nov 1993-Jan 1994 with the Country making process of the PPA. not been strong. UNDP and mapping too sensitve to undertake and its own publicahon using the Department. UNICEF initially supportve. subject to misinterpretation. Research results of the BA. teams could not stay overnight in some LQ communities for security reasons. Mexico Clear lines of communi- Major devaluation. Strong SDS (Government Poverty BA: Four teams interviewed 722 people Communities: Information sharing. Interviews conducted cation established initial support lessened as Agency) actively participated in four areas (two urban, two rural). in Feb and March 1995. between the PPA advisor other priorities took over in the fieldwork, All consul- Qualitative research and conversational Others: UNDP and UNICEF cofi- Conducted with the and the PA manager. tants hired were from NGOs. interviews. nanced. The capacity of the SDS to PA. However, communica- conduct qualitative assessments tions with the supervisor increased. undertaking the PPA in the field were difficult. (Table continues on thefollowing page.) Annex 1. (continued) Country, Timitng, and Cost Context Bank Context In-country Institutions Involved Methodology Level of Participation EUROPE AND CENTRAL ASIA Albania The report vvas managed Individual field researchers Various: Individual interviews with Communities: Information sharing. The Green Cover and written by a senior contracted. No institutions households; interviews with key infor- report was dated Aug economist and was nar- invotved in the research. mants includirng academics, expat and 1996; mnost compo- rowly focused. One part local staff of agencies implementing land nents were carried out was devoted to rural- privatization, and microcredit. Jan-July 1996. urban mnigrants squat- ting on the outskirts of Cost about $50,000, the capital and one other including World Bank city. The other part was time and travel. In- focused on the beneficia- country research com- ries of the agriculture ponent ranged from microcredit program. $13,000-25,000 (paid for by UNDP). Armenia Wanted to coordinate Ministry of Economy In the PA, Armenian Assem- Seven hundred seniistructured Communities: Informaton sharing. Started in 1994. Com- the PA with the Social bly of Armenia. Most other interviews with individuaLs from poor pleted June 1995. Investment Fund. Good NGOs were involved in emer- and medium-income households and Others: The PPA manager and the relations with COD gency aid. Church was also with local officials, medical personnel, field researchers (professional Fieldwork in Oct 1994 Senior management involved. In the PPA, the uni- teachers, and aid workers. anthropologists and social scien- -March 1995. support. versity was a formal organizer fists) presented field research find- and contractor for the qualita- ings at several workshops to local tive research. NGOs, government officials in Yerevan, international NGOs. Their input was incorporated into the final report. Azerbaijan SORGU Institute attached to Seventeen interviewers; mainly sociolo- Fieldwork conducted the Baku institute of Sociology gists and education personnel with previ- from Aug 1995-lan and Political Science. NGOs ous experience of quantitative and 1996. and government assisted with qualitative fieldwork. Semistructured selection of sites. interviews with groups of five and eight people. Results combined with commu- nity surveys conducted in 91 populaton points throughout the country in parallel with a natonal household survey in Nov-Dec 1995. Georgia PPA designed to comple- Govemment informed, on In-country research was part Various: Semistructured, in-depth Communities: Series of stakeholder Draft PPA completed ment other poverty sur- board, otherwise not of a project financed by the household interviews. Sermistr-uctured workshops will be convened and April 1997. veys (income and involved. UNDP. Project maniaged by interviews witll "expert" informants- will include the poor, to feed back expenditure, etc.) and local social scientist and one aid workers from local and international the preliminary findings, elicit com- contribute to the CAS. deputy. NGOs and donor organizations; head ments and critiques, all of which doctors; school directors and teachers; will be incorporated into the final and officials. report and recommendations. Moldova The manager of the PA Local NGO formed by and Various: Semistructured, in-depth Communities: Information sharing. PPA began in June made considerable effort working under the auspices of household interviews. Semistructured 1996 and completed in to coordinate other proj- an American NGO. interviews with "expert" informants- June 1997. ects with the PPA aid workers from local and intemational (including the Social NGOs and donor organizations; head Investment Fund, an doctors; school directors and teachers; cc agriculture sector social officials. cr assessment, micro- finance, etc.) in terms of selecting regions, highlighting issues, and trying to gather comple- mentary data rather than repeat previous research. Ukraine PA consisted of several World Bank manager and a Various: Semistructured, in-depth Communities: Information sharing. Completed 1996. components, one of U.S. anthropologist contracted household interviews. Semistructured which was the PPA. with a Kiev-based sociological interviews with "expert" informants- Others: Results were presented in research institute and some aid workers from local and intemational several workshops for academics, individual researchers, to con- NGOs and donor organizatons; head NGO representatves, and govem- duct the interviews through- doctors; school directors and teachers; ment officials (and Ukraine office out the country. and officials, etc. World Bank staff) upon completion of the field research; their input was incorporated into the final docu- ment. (Table continues on the following page.) Annex 1. (conti-nued) Country Tinting, and Cost Context Bank Context In-country Institutions Involved Methodology Level of Participation SouTH ASIA Pakistan The manager of the PA The government did not The Federal Bureau of Stalis- PRA: Local consultants were recruited. Communities: Information sharing. Feb-July 1994. was given limited time support or oppose the PPA. tics was involved in selecting Roving teams were used. Fieldwork for two to complete the PA. However, some the communities. Others: Workshops were held with months. Many felt that the PPA government officials and a wide cross-section of stakeholders information was not NGOs disagreed with the for the PA. adequately represented PA's conclusions. Although in the PA. The PPA was there was consultation, conducted after the some stakeholders felt their household survey analy- views were not considered sis was completed. The and that the ongoing supervisor of the PPA national poverty debate was an outside consul- was not represented in the tant. The Human final PA. Resources Division and COD managed the PPA. Notes: See list on pp. x-xi, for definitions of abbreviations and acronyms. The column describing the levels of participation has limited value and is only indicative. To be more accurate, a multiple stakeholder analysis of participation using the stakeholders' own indicators would be required. The diversity of experiences of the PPAs has been affected by many factors, including the context in the World Bank and in country. This is detailed in the table, as are the methodologies employed to elicit the views of the poor. Annex 2. Impact Impact on Impact on Impact on Country PPA Highlights the World Bank the Borrower Other Institutions Lessons Learned AFRICA Benin Children's drawings were This was one of the first The PPAs increased the interest used to understand their PPAs in the Bank, and its of the Ministry of Planning in perceptions of poverty. results initiated the ongo- conducting qualitative ing dialogue on the use of assessments. qualitative and quantta- tive information. Those working on the PAstated that the PPA made the PA more interesting and readable. Burundi The PPA results stressed Ongoing: The country Ongoing: The govenmment is UNDP has used the PPA * The PPA was useful in a postconflict situation as a the vicious cycle of team has recommended currently reviewing the poverty results to feed into its own rapid way to gauge the principal poverty issues 00 hunger, health problems, that the Bank undertake a note, which includes the results poverty reduction work. when quantitative data were not available. and low agricultural out- new commnunity-based of the PPA, and intends to * A full assessment needs to be made of the skills put. The new phenome- poverty project, the develop its own poverty reduc- and experience available within the country to non of child-headed design of which will use ton strategy. conduct the PPA. Where the teams have previ- households resulting the recommendations of ously done neither PRA nor poverty work, two from deaths in the conflict the PPA. weeks of training is insufficient, and they may was highlighted during need external technical assistance to analyze the PPA. In urban areas, results. This PPA did not produce the depth of the PPA extracted the sto- analysis expected, primarily because the tearns ryline of how the infor- had too sharp a learning curve during the work mal sector had been fieldwork, affected by the crisis and * PRA teams consisting entirely of economists (one embargo. of the Burundi teams) tend to be weak: They focus too much on extracting a number and do not properly document all the qualitative infor- mation a community is giving. (Table continues on thefollowing page.) Annex 2. (continued) Impact on Impact on Impact on Country PPA Highlights the World Bank the Borrower Other Institutions Lessons Learned Cameroon The emphasis given by Although The results of the PA and PPA NGOs and other institutions * Working with NGOs in preparing the PA and the poor to problems of macrocconomic manage- were a shock to Cameroonians involved in the PPAs under- PPA provided a highly cost-effective means of hunger, nutribon, and ment and debt issues pre- both inside and outside govem- stood the value of the tapping into expertise and capacity. high food expenditures dominate in the country ment, as poverty had not previ- approach and appreciated the * The effectiveness of the exercise depends on the justified and amplified dialogue, some effort was ously been acknowledged as a opportunity to engage in dia- willingness and commitment of government to the focus on addressing made to integrate a serious problem. Ownership logue on poverty issues with engage in dialogue with civil society and on its food insecurity in the poverty reduction strat- was not developed among key the government, the Bank, determination to tackle the poverty problem poverty reduction strat- egy into the CAS, build- policymakers, as the central and other donors. identified. This commitment was largely absent egy. The PPA also high- ing on the results of the goverusnent was not strongly inCameroon and the results of the PPA were lighted problems of PA/PPA. Key elements committed to poverty reduction published without extensive government isolation (transport sys- are support for small- or to building on the results of support. There was, therefore, limited leaming tem) and govemance scale food production, the PA/PPA process. Some local and shifting of attitudes. (decentralization). It pro- processing, and market- government officials did * The composition of the team involved in thc Bank cos vided key insights into ing, and measures to develop a keen interest in the affects the way the information is managed, dis- 00 the gender dimensions of enhance the status of PEA and in replicating its seminated, and analyzed. There was limited own- poverty, confirming the women, including land methodologies elsewhere. ership in the country and sector departments and disproportionate work- and legal reform, rural the PPA was managed in the Technical Depart- load of women, and the infrastructure, and girls' ment. fact that changing gender education. Some viewed * The PPA was a valuable instrument for bringing roles bring new opportu- the PPA as having limited the concems of the poor into the dialogue. nities and new burdens. credibility, with some * The PPA provided critical new insights (gover- information being too nance, isolation) and reinforced the priority of generic. Interpretation of tackling food insecurity and poor infrastructure. the data in the Bank was * Ensuring gender balance in the PPA yielded key limited because of lack of insights into the dynamics of poverty. time. * Involving local institutions and holding workshops with both govermnent and civil soci- ety are mechanisms for expanding ownership of the poverty problem and in-country capacity to analyze and address it. Equatorial The PPA was considered It is too early to assess the UNDP Assistant Resident * The information may be accurate but if the insti- Guinea sensitive and was rewrit- impact on the government, Representative in country and totional frameworks of the borrower and the ten in the Bank. which has not yet seen the the Executive Director of the Bank make them unable to embrace the results, rewritten Green Cover version. Bank have requested a meet- the impact will be limited. ing to discuss the findings of the PPA. Ethiopia Provided ideas on the Because data problems Very limited, except for the part None, because the Bank and * Institutional links with government are vital to causes of recent increases delayed the results of of the PPA results that came government have not yet ensure that the PPA results are used. in agricultural production quantitative surveys, the through in the CAS. The govern- released the report. * It would be useful to have a standard publication found in survey data. Dif- PPA results were exten- ment department that acted as a vehicle for PPAs. In this case, the PPA was ferentiated winners and sively drawn upon for the counterpart for the PPA has little intended as part of a PA and would have been losers among rural com- CAS. Results also fed into clout and did not widely pubhshed as an annex. But the PA has been munites. Raised the issue the upcoming food secu- disseminate or debate the delayed two years because of data problems. of the inappropriate hm- rity project and the social- results. In this type of situation it would be useful to pub- ing of the school calendar sector note. lish the PPA separately, since it often contains and payment of school information that would be useful to local organi- fees for poor families. zations but is only valuable within a limited time Showed the importance window. of seasonal poverty in * This PPA showed the value of including at least urban areas. one amthropologist as a team member Gabon Household data exist for The PPA shed light on the The PA, incorporating the * Despite the delicate character of poverty in oO Libreville and Port-Gentil inefficiency of public results of the PPA, was Gabon, the participatory process of the PA gener- (50 percent of the popula- spending in the soeial sec- discussed with the ated significant ownership and the Interminister- tion) only The PPA com- tors. To follow up on rec- government's Interministerial ial Technical Committee provided more detailed plemented the ommendations, PER is Technical Committee, which and constructive comments than did the Bank. quantitative information being carried out in the received it well and provided for these two cities and health and education sec- detailed and constructive com- provided key qualitabve tors. Depending on the ments. The government recently information for small PER recommendations, requested the Bank's assistance cities and rural areas. the Bank might envisage to improve the transparency and Quantification of the projects in these sectors. efficiency in public spending in qualitative results permit- the health sector A poverty sem- ted the definibon by zone inar was held in June 1997 with of clear priorities of the financing from the government, poor the Bank, and UNDP. The objec- tives were to disseminate the results of the PA, to define action plans for the health and educa- tion sectors, and to build capac- (Table continues on the following page.) Annex 2. (continued) Impact on Impact on Impact on Country PPA Highliglhts the World Bank the Borrower Other Institution7s Lessons Learned ity to collect and analyz.e statis- tical data, in collaboration with other donors. Ghana The PPA complemented The information from the The informaton from the PPA Other insttutions were rInitially, key stakeholders were reluctant to the quantitative informa- PPA is relatively complex and PA has been analyzed in a already involved in promot- become involved. However, a process approach tion and provided further and extensive, thus mak- U1NICEF report, later disserni- ing a dialogue on poverty. It is was adopted whereby the PPA and PA were information on such lag incorporation of its nated at a national conference thus difficult to assess the viewed as a means of initiating dialogue and problems as the problem analysis into other Bank attended by key government impact of the PPA alone on not an end in itself. For such an approach, Bank of female-headed house- reports often time con- policymakers. An ongoing other institutions. The formu- follow-up is vital. holds in the north. The suming and difficult. process of dialogue has now lation of a poverty policy importance of rural infra- However, the CEM-an developed between the Bank through joint donor action structure and the quality influential Bank instru- and the government regarding and the Consultative Group and access of education ment-had a poverty poverty. The government was meeting in Paris is now being us and health were focus which, in part, was initially not receptive to the developed. highlighted. influenced by the results results of the PA and PPA. How- of the PA. ever, the government's interest was underscored by the initial PPA results in the CEM. The capacity of the Statistics Depart- ment was strengthened through the PA (published poverty pro- file, training, increased dialogue with line ministries). Kenya The information in the The PA does reflect the Some in govemment were ini- Capacity in country to con- * Sequencing of the PPA and quantitaive analysis PPA was used to design a major findings of the dally skeptical and not willing duct qualitative assessments is important. The PPA was able to influence the more effective and PPA. Some argued that to become involved directly. The has increased. design of the household surveys. focused quantitative the PA and PPA could benefits of adopting the * More time was required to develop dialogue with questionnaire. The PPA have been more exten- approach were not clear to key stakeholders. focused on issues such as sively incorporated into them. However, after the first social capital, coping other country reports. PPA analysis and dissemration strategies, female-headed workshops, the government ini- households, and the use bated a scond round with the of services including NGOs. This is being funded by water. It resulted in the the DFID. recogrntion that rural water was a problem. It highlighted the fact that people defined female- headed households differently Lesotho Some key themes Initially limited ownership by PA widely used by donors * Initially there was limited ownership in country. emerged from the PPA govemment. Some in govem- and other agendes in country. Ownership increased only when responsibility that were not highlighted ment felt that the draft PA was for the action plan was handed over to a cross- in the quantitative sur- not a clear policy document. But section of stakeholders. A workshop was held in veys: for example, alco- as govermment ownership February 1997 that solidified local ownership of holism and poltical increased, such issues as corrup- the action plan. factors such as injustice tion and the role of local govem- and corruption. These ment appeared in speeches and issues fed into the policy documents. level through the acton plan. Madagascar The PA information put Impact upon Bank docu- Govemment comnnitment and Impact on other key institu- * Because the initial approach was not just a one- such issues as access to ments has been limited to ownership of the poverty prob- tons in country that were time intervention of information gathering, but social services and secu- date. lem vary. Those who were involved in the PPA has been part of a process of building up dialogue at differ- rity on the agenda for dis- involved in the PA are now high. ent levels with various stakeholders, government cussion. more committed. Government ownership has developed. However, because officals have visited the Bank Bank follow-up has not been extensive, it has not on several occasions to follow been possible to consolidate all the advances in up the results of the PPA. How- policy dialogue. ever, follow-up by the Bank has not been extensive. (Table continues on the following page.) Annex 2. (continued) Impact on Impact on Impact on Country PPA Highlights the World Bank the Borrower Other Institfutions Lessons Learned Mali The information from the Project on grassroots ini- Was a first step in puttinlg qualitative survey tiatives was identified poverty on the political agenda explained some of the and is under preparation. as a cross-cutting issue in itself. perceived anomalies from The PPA was one of the the quanttative survey. first in Africa and its For example, the dispro- methodology was repli- portionate amount of cated in the Bank's PAs in money spent on clothing Niger, Chad, and Benin. was explained by the fact that clothing is also an investment for "social insurance." Mozambique The PPA generally Too early to assess policy The PPA process was success- Other stakeholders have been * There was a trade-off between local ownership sounds out local commu- impact. fully intemalized in the included through widespread and quality control. nities using an approach Ministry of Planning. Fligh dissemination of the PPA * For increased impact, the PPA reports should be fhat is more flexible and degree of local ownerslhip. material from the beginning. written in a more concise manner more open to defining Working groups in sector min- A real strength of the PPA has * In the first phase there was an overcrowding of issues according to the istries have used information on been its multidisciplinary the research agenda and the interview guide was poor's own concerns. It specifc sector issues (such as approach, in terms of back- too broad. Careful matching of the research issues encourages and is based health, water, livestock). Min- ground, type of institution to methods of investigation is required. on their direct participa- istry of Social Action aiid other (university, govemment, * A major problem has been the demands of mult- tion and embraces direct institutions have nominated NGO), and type of researcher ple stakeholders. The World Bank had its own observation as a key com- staff for PRA training and sec- ("insiders" and "outsiders"). internal deadlines and the Bank PPA manager, ponent of the research onded staff to participate as This multiinstitutional located in Maputo, also had to be responsive to method, members of the field teams, approach has also strength- the needs of the other stakeholders. ened relationships among the * PRA represents a significant and useful method- participating institutions. Col- ological approach to encourage communities to laborating NGOs (partners in be more conscious of their life conditions, oppor- fieldwork in Phase II) have tunities, strengths, and limitations. This is partic- benefited directiy while Non- ularly important in the context of a govemment participating NGOs have without capacity to help in many areas of the used field-site data for country. improved targeting, and poverty mapping data for longer-term planning. Niger Some key elements of the The manager of the PA, The government formed an As a result of the Bank's PPA * The primary lesson leamed is that the perception poverty profile (based on since the publication of Interministerial Committee on and PA process, UNDP and of poverty issues by major stakeholders is a key statistical data) were con- the Gray Cover, has suc- Poverty and has been actively EU have now participated in element in understanding poverty and will be firmed by the PPA (for ceeded in influencing the involved in the PA process, hav- regional workshops for the taken into account in future projects, programs, example, food insecurity, design of the proposed ing written its own PA with Niger CAS 1997 in an attempt and policies dealing with poverty reduction. low enrollment,) and Infrastructure Project to UNDP. A Round Table on to design their own assistance some other elements were include pilot mural opera- Poverty has been planned but strategies, with poverty as a added (causality for low tions. The CAS will use has not yet been held, largely central focal point. NGOs are enrollment nonuse of participatory techniques because of pohtical uncertainty now major participants in the health services) and will such as regional in the country. poverty dialogue. be incorporated into the workshops and consulta- new survey design. dion, which have been recognized as very valu- able based on the experi- ence of the PA. .n Nigeria The PPA highlighted that The Bank refocused its In-country, the PPA has initiated NGOs are now being increas- * While funding has been shrinking, the PPA the poor viewed water program toward water an ongoing debate about ingly more accepted as part of process facilitated increased donor coordination. and roads as priorities. In and roads. Targeting pub- poverty and gender issues. The the development process. addition, the weakness of lic expenditures in health, PPA process initiated the gov- the coping mechanisms education, and water was erunent's increasing interest was highlighted. Strategy indicated to be important and involvement in the work of needs to be focused on in alleviating the suffer- the NGOs. pattem of growth, as bot- ing of the poor tom 20 percent of the population has become worse off despite an over- all poverty decline. (Table continues on thefollowing page.) Annex 2. (continued) Impact on Impact on Impact on Country PPA Highlights the World Bank the Borrower Other Institutions Lessons Learned Rwanda The PPA highlighted the The PPA was a central Govemrnent was initially some- The results of the PA and PPA * PRA can be an extremely useful tool, even in the labor constraint in the input to the PA, which what skeptical about the PPA, were widely disseminated worst of postconflict situations (and there can agricultural sector since was used as a base docu- but has increasingly become and debated in Rwanda, hardly be less trust than in postconflict Rwanda). the genocide, the national ment for the Consultative interested in and supportive of although the concrete impact e Putfing together a team of people based on their debate had previously Group meetings. PPA the results. The results of the on other institutions is still to commitment as well as their skills can immensely assumed a continuing results also fed into the PPA were very high profile in be seen. improve the quality of results-in this case the labor surplus economy CAS, the agriculture Rwanda, in part because of the team showed a remarkable commitment to pro- It also highlighted restric- strategy note, and the controversy about labor ducing a good analysis. tions on labor mobility, agricultural LIL. constraints and the trade-off * The influence of the PPA can be greatly increased the increasingly female between economic costs and by carefully choosing partners to conduct it and face of poverty, problems security benefits in imposing gaining the involvement of high-profile govem- of rising costs of health mobility restrictions. The gov- ment or civil society individuals. These informal care, migration pattems, ermnent is reviewing the PA networks disseminated and lent credibihty to the usL and information on and will make a decision about results. changes in community whether it wishes to publish a * For the PRA training, we had allocated two networks and social rela- joint PA and poverty reduction weeks. This was sufficient to transfer the tools but tionships. strategy. not the substance of all the questions. It would be useful to increase the length of the training and carry out fieldwork in several rounds, so that teams can analyze results from the first communi- ties immediately Trust with communities is increased and the process moves more quickly where the team includes a member from the community. South Africa The PPA highlighted the Too early to assess. The PPA induded keypolicy- Too early to assess. * Stakeholder involvement from the beginning was various dynamics of the makers from the beginning and an important step. Although the initial decisionmaking process, ownership gradually developed stakeholder workshop was time consuming and coping strategies, season- among high-level, influential problematic to convene, many advantages ality, intrahousehold gen- stakeholders. The Cabinet met became apparent as the process evolved. The der relations, and the twice to discuss the PPA. The workshop identified the most appropriate constrained access to ser- first meeting took two hours approach and methodology. As a result, the PPA vices. As an example of and was chaired by Thabo was both rapid and efficient. the stresses caused by Mbeki, the Deputy President of * Management of the process by the local consult- seasonality, the problems South Africa. ing firm was transparent and effective. of paying for school fees * The unexpected closure of the South African at a time when income Reconstruction and Development Office rendered was short was also high- the initial strategy of focusing on one particular lighted through the PPA. department inadequate. Tanzania Both the PA and the PPA The finandal-sector The government was initially * More time and resources are required to promote estimated that the num- reform is using the same cautious but became more a longer process, which would lead to a greater ber of poor in the rural methodology. The infor- receptive as it understood the understanding of poverty and its links to policy. areas was approximately mation from the PPA is value of the approach. The PPA * Teams could be located at the field level. Coordi- 50 percent of the popula- reflected in the PA. highlighted the capacity of the nation by one person in Washington proved diffi- lion. The PPA highlighted poor to analyze their own cult. The Resident Mission in Tanzania could be that a larger proportion of problems. strengthened to take the iniiatfive. People skilled these poor households are in the analysis of poverty could be located withiin female headed. Whereas the Resident Mission. To increase the capacity of the PA focused on con- the Resident Mission, training in best practice of sumption and expendi- gathering qualitative information could be con- ture, the PPA used criteria ducted and tool kits provided. as defined by the poor, C.n such as feelings of power- lessness and hopeless- ness. Many problems were gender specific; the women identified food, water, and health as their main problems, whereas men identified transport, farning, and drunken- ness. Togo Aftention drawn to gener- There was limited impact Greater ownership of proposed Other donors such as the * It was difficult for researchers to analyze and ally ignored vulnerable on the CAS because the strategy. The PA was "more UNDP are also promoting the organize information in the field within the lim- groups: displaced people, PPA was completed after- interesting" and therefore more use of qualitative techniques. ited time frame. Therefore, some analysis might and domestic child labor. ward. The PA was written readable. The PPA assisted in building not have been acciurate and it was not written in a by the PPA manager and dialogue between the Bank way easy to understand. thus the qualitative infor- and other donors. * The data should have been disaggregated by mation was incorporated. gender. (Table continues on the following page.) Annex 2. (continued) Impact on Impact on Impact on Country PPA Highlights the World Bank the Borrower Other Institutions Lessons Learned Uganda Knowledge about areas The Ugandan PPA was - Men and women were consulted but the informa- of the country where no one of the first in the tion was not disaggregated. information was avail- Bank and it initiated a A lot can be learned quickly able because of the dvil Bankwide discussions on war. the value of qualitatve data. Zambia Information detailed and The PA includes a The government was influenced The NGO, PAG, has devel- - Feeding information back to the communites comprehensive. Disag- detailed action plan that by the priorities expressed by oped into an effective policy- and promoting ongoing dialogue thould be part gregated by gender incorporates some of the the poor in the ranking oriented institution. The of the design of the PPA. Information from the where appropriate. Such recommendations of the exercises. The Ministry of capacity of the NGO has been PPA could then be used to develop action plans. issues as school fees and PPA. Specific elements Health has been using the built. However, it is now * Including key stakeholders from the beginning the tiaing of their pay- that influenced the action results of the PPA and the PA in dependent on government enhanced long-term ownership. ment were highlighted. plan included emphasis developing pohcy In the Min- and donors for sustainability uz on rural infrastructure istry of Education, a new policy and its capacity requires fur- investments and urban is in preparation with reference ther strengthening. services. The poverty to the timing of school fees. Pos- profile, especially com- itive feedback has been received miunity-based identifica- from the communities in the tion of the ultrapoor; PPA on the fumctioning of the coping strategies; safety emergency safety net during the nets; and targeted inter- southem Africa drought of 1992. ventions were also influ- enced by the PPA. The Bank's Health Project contains conditions of cost recovery based on the PPA anld supported by the second Social Recovery project. LATIN AMEIUCA AND CARIBBEAN Argentina Identification of eligibility PA has been completed. Increased coordination between With NGOs: Only a few * Issues of ownership in the Bank context are rele- and targeting criteria for There is great potential government agencies and pro- NGOs have been able to meet vant. From the beginning the PPA was planned beneficiaries of social pro- for the results to be inte- grams. Dissemination of the the technical qualifications and prepared using an inclusive, consultative grams. Development of grated into other Bank results has validated the required by SIEMPRO to approach within the Bank. To be effective and impact indicators to mon- programs because of the methodology and contributed to carry out the PPA. It is have multplier effects, PPAs have to be linked to itor social programs. team ownership within the development of an expected that the higher stan- broader operabons or sector work. the Bank. integrated (qualitative/quantita- dards set by SIEMPRO would * Results have to be translated into operational rec- tive) approach to monitoring have a positive impact on the ommendations for ongoing operations (M&E and evaluating social programs. NGO community as they methodologies and indicators, eligibility and tar- Some government programs are would have to professionalize geting criteria, etc.). modifying their M&E indicators their services. At the same as SIEMPRO has developed pro- time, SIEMPRO is carrying gram-specific indicators, out a training program for government officials and planning to develop a more structured training program (a master program). Outside '-I the country: SIEMPRO experi- ence on monitoring and eval- uation of social programs is being disseminated to other countries. Costa Rica The PPA highlighted the Delay in the analysis and The government was eager to Too early to assess. * A clear dissemination strategy should be defined linkages between home dissemination of findings disseminate the results but it as part of the PPA's design. ownership and status in has meant that the impact took nine months for the Bank sodety. Family was within the Bank has been to grant permission. viewed as the most limited to date. important institubon, and in times of stress people rely on their families for support. (Table continues on the following page.) Annex 2. (continued) Impact on Impact on Impact on Country PPA Highlights the World Bank the Borrower Other Institutions Lessons Learned Ecuador Quality of information is The PA iiformation has Although the government was The NGOs in country have * There were advantages of the qualitative assess- good. The results fed been strongly reflected in not included in the process of increased their capacity to ment preceding the quantitative assessment. The directly into the type of the CAS. Several sector the PPA and the PA, the results conduct qualitative surveys. qualitative information was used in the design of questons analyzed in the divisions have started of the PA have affected the UNICEF used the PPA the quantitative survey quantitative survey, sector studies as prepara- country's perceptions of its pri- methodology to evaluate the * The PA manager should be closely involved in tion for operations based orities. Such issues as access to impact of its program. the whole process.A greater understanding of the on the PA results. secondary schools and off-farm qualitative research techniques from the begin- rural markets, previously not ning would have enhanced the results. It is pm- part of the poverty debate in posed that the preparation of the teams involved Ecuador, were placed on the in the PPA and PA be clearly thought through for agenda. The PPA work has initi- each team and each country. ated dialogue between different * The results should be analyzed by someone who groups and the Bank. has an understanding of the country and its cul- us ture in order to put the poverty into the country Cu context. Guatemala The findings of the PPA have recently been pub- lished in a book, and fol- low-up studies are underway on such issues as gender, problems of indigenous peoples, and rural-urban dichotomies. Mexico The quality of the PPA The results are still being The results are still being Too early to assess. e It was difficult to find a suitable national consul- was nsixed. The informa- assimilated. assimilated. tant who was not politically affiliated. tion was not ranked ade- * Controlling the process of gathering PPA infor- quately However, it was mation proved problematic, as the teams gender specific, which attempted to follow their own agenda. added value. The report found that the women of Mexico City are unwill- ing to leave their houses and go to work. Bccause they do not have tenancy rights they are afraid that their houses may become occupied. In the northern areas it is easier for women to obtain jobs than for men. This chal- lenges the traditional gen- der roles as many men find themselves out of work. Conflict within the household was highlighted as a major issue. EUROPE AND CENTRAL ASIA Armenia The qualitative informa- The PA manager knew The results were disseminated at * If there had been adequate resources and time, s fnbon assisted in the analy- the country well, had a seminar in March 1996. the PA should have been integrated with the us sis of the results of the built up respect among Social Investment Fund. quantitative surveys. The key policymakers and PPA highlighted the great within the country's acad- variety of coping strate- emic community, and gies and the lack of trust encouraged a team for any organization such approach within the as local government, Bank. In the Bank, the NGOs, and community PPA manager and those groups. managing surveys worked closely to estab- lish a research agenda for the PPA. The country department's Macro- economist was also exten- sively involved. The outcome was the follow- ing: first, the results of the (Table con tin ues on the following page.) Annex 2. (continued) Impact on Impact on Impact on Country PPA Highlights the World Bank the Borrower Other Institutions Lessons Learned PPA were reflected in the PA; second, the country program and the recently drafted CAS integrated the results of the PA; and third, the PA was well received in country. SouTH ASIA Pakistan The PPA highlighted the The awareness of some Limited. Limited. Some key stakeholders were consulted during the fact that the poor spend a Bank staff of the informa- preparation of the PA. The Resident Mission helped large proportion of their don contained in the two to organize the workshops. Some felt that although income on health care. PPA studies is limited. the consultations were fairly extensive, the final The poor felt that social document did not reflect the views of the majority. °Z, services were inadequate and there was a lack of accountability to the com- munities. Many income- earning opportunities were lost through ill health. Notes: See list on pp. x-xi for definitions of abbreviations and acronyms. In some cases, it has been too early to assess the impact. In others the impact of the PPA has been difficult to isolate from other factors. Pol- icy change and attitude shifts are part of a complex social process and thus it is often difficult to isolate the impact of the PPA. ANNEXES 101 Annex 3. Poverty Assessments, Completed and Scheduled (by Country), Fiscal 1989-2000 EAST ASIA AND THE PACIFIC LATIN AMERICA AND THE CARIBBEAN (CONTINUED) Completed (12) Paraguay 1992 Philippines* 1989 Peru 1993 Indonesia 1991 Uruguay 1993 Malaysia 1991 El Salvador 1994 China' 1992 Guyana* 1994 Philippines (update) 1993 Jamaica 1994 Indonesia (update) 1994 Paraguay (update) 1994 Fiji 1995 Argentina 1995 Vietnam* 1995 Brazil 1995 Lao People's Democratic Republic' 1996 Colombia 1995 Mongolia' 1996 Dominican Republic 1995 Philppines (update) 1996 Guatemala 1995 Thailand 1997 Honduras* 1995 Scheduled (3) Nicaragua' 1995 Cambodia* 1999 Bohvia (update)* 1996 Pacdfic Island States (PIS)* 1999 Ecuador (update) 1996 Papua New Guinea 1999 Trinidad & Tobago 1996 Scheduled Updates (1) Costa Rica (update) 1997 Philippines 1999 Scheduled (2) EUROPE AND CENITRAL ASIA Haiti' 1999 Panama 1999 Completed (12) Kyrgyz Repubhcl 1995 Scheduled Updates(4) Poland 1995 Mexico 1999 Russia 1995 Uruguay 1999 Armenia* 1996 Venezuela 1999 Belarus 1996 Argentina 2000 Estonia 1996 MIDDLE EAST AND NORTH AFRICA Hungary 1996 Completed (5) Ukraine 1996 Egypt Arab Republic of* 1992 Albania* 1997 Morocco 1994 Azerbaijan' 1997 Jordan 1995 Romania 1997 Tunisia 1996 Kazakhstan 1998 Yemen, Republic of" 1996 Scheduled (12) Scheduled (1) Bulgaria 1999 Algeria 1999 Georgia' 1999 Scheduled Updates (1) FYR Macedonia4 1999 Egypt Arab Republic of' 1999 Moldova 1999 Latvia 1999 SOUTH ASIA Lithuania 1999 Completed (9) Tajikistan4 1999 Bangladesh* 1990 Uzbekistan 1999 India* 1990 Bosnia and Herzegovina' 2000 Nepal* 1991 Croatia 2000 Pakistan* 1991 Turkey 2000 Sri Lanka' 1995 Turkmenistan 2000 Pakistan (update)* 1996 Scheduled Updates (1) India (update)* 1997 PoLand 2000 bangladesh' 19988 Poland 2000 ~~~~~~~India (update)4 1998 LATIN AMERICA AND THE CARIBBEAN Scheduled Updates (2) Completed (24) India' 1999 Bolivia4 1990 Nepal* 1999 Chile 1990 Costa Rica 1991 Ecuador 1991 Mexico 1991 Venezuela 1991 (Table continues on thefollowing page.) 102 CAN THE POOR INFLUENCE POLICY? Annex 3. (continued) SUB-SAHARAN AFRICA Zimbabwe' 1995 Completed (37) Eritrea' 1996 Malawi* 1990 Lesotho* 1996 Mozambique* 1991 Madagascart 1996 Ethiopia* 1993 Malawi (update)* 1996 Ghana' 1993 Niger* 1996 The Gambia* 1993 Nigeria* 1996 Mali* 1993 Tanzania 1996 Namibia 1993 Togo' 1996 Sierra Leonet 1993 Congo 1997 Uganda* 1993 Cote d'lvoiret 1997 Berint 1994 Gabon 1997 Cape Verde' 1994 Guineat 1997 Guinea-Bissau* 1994 Chadt 1998 Rwanda* 1994 Djibouti* 1998 Seychelles 1994 Scheduled (2) Cameroont 1995 South Africa 1999 Comoros> 1995 Angola* 2000 Ghana (update)* 1995 Scheduled Updates (4) Kenyat 1995 Cameroon 1999 Mauritaniat 1995 Tanzaniat 1999 Mauritius 1995 Djiboutit 2000 Senegal* 1995 Mozambique* 2000 Zambiat 1995 Total Number of Assessments: Completed 99 Scheduled 20 Scheduled Updates 13 Grand Total 132 Notes: = Intemational Development Association borrower. FYR = former Yugoslav Republic. Assessments classified as completed are in gray or red cover, except for three assessments that were completed before Operational Directive 4.15: Poverty Reduction was issued. Schedule as of July 16, 1998. In East Asia and the Pacific, no assessment is scheduled for Korea, which receives only linited Bank support, or for Myanmar, because of its inactive status. In Europe and Central Asia, poverty assessments are not yet scheduled for the former Yugoslav Republics (except fiscal R Macedonia and Slovenia), because of emerging political developments; for Cyprus, the Czech Republc, Slovenia, and Portugal, because of high income levels; and for the Slovak Republic. In Latin America and the Caribbean, Suriname is excluded because of its inactive status. Poverty assessments for Antigua and Barbuda, Belize, Dominica, Grenada, St. Kitts and Nevis, St. Lucia, and St. Vincent and the Grenadines will be prepared by the Caribbean Development Bank. Poverty assesssments are not planned for the Bahamas or for Barbados, both of which have graduated from the Bank. In the Middle East and North Afnca, Bahrain, Iraq, Israel, Kuwait Libya, Malta, Oman, Qatar, Saudi Arabia, Syria, and the United Arab Emirates are excluded because of their inactive status. Poverty assessments are not scheduled for Lebanon or Iran because of the uncertain political situation. In South Asia, poverty assessments are not scheduled for Afghanistan because of the unstable political situation; for Bhutan, which receives only limited Bank support; or for Maldives, which is not a major IDA borrower. In Sub-Saharan Africa, Botswana, Liberia, Somalia, Sudan, and Zaire are excluded because of their inactive status. A poverty note was completed for Burkina Faso in fiscal 97. Poverty notes will be prepared for Bunmdi, the Cen- tral African Repubhc, Equatorial Guinea, Sao Tome and Principe, and Swaziland. Table compiled by PRMPO, World Bank, Washington. ANNEXES 103 Annex 4. Examples of PRA Exercises in Thailand (1998) and Zambia (1996) Example of PRA Exercise in Thailand Impact of the Economic Crisis: Khon Kaen Dotr 26 Jonry 199B F.oocioto: Motin Weel, Impc Children Ten_t commu ty se amn eroee. Ooediines, ted t/ ob \ ~~t rli ngsp rc I xh etlZr,SeS \ e h em h liier thinsee dicnoo Source: Robb and Zhang (1998). Example of PRA Exercise in Zambia C auses and Impact of Early Initiation of Sex Among Girls (Analysed by a group of girls, Ckawama Compound) baza mukwsutira K~~~~~~ugl \ 1 bu les) Kugul\ men mak prmises nio2 m we ofmarriage wich t u t u \t r on' reish s-etsl bamuno ~ ~ ~ ~ ~ ~ ~ ~~~~~~ k -\ mid h. or the love oi- tar y - ore X oe otf o t-el f.d -cr of I001 ; > _ ~~~~~~~~~initiation of t6 ihebeu ftr. Ihmr1'men being naughty ~~~~sex camong --us 0. be i/rnpci id, he n rri- w t / < ~~~~~~~ ~~girls i-;nid. h. drcles ~hr te xr too0 .1 . / \ r,odicuir r resull ~~~. . hoso.hi r liws. Th. num.xZr 1 0 /\\tgirl j.ut.ln..l.f1OO)wh0.. ~~wK / \ \ r~~~~~~~af~ htedty hik Ys40 Source: Shah and Nikhama (1996). 104 CAN THE POOR INFLUENCE POLICY? Annex 5. Country Case Examples Costa Rica Background Costa Rica has a per capita income of US$2,590 (1995) annually and thus is at the higher end of the lower-income countries. Its quality-of- life indicators are similar to those of a developed country. However, the key indicators of social well-being are more similar to those of a middle- income country.1 Costa Rica has traditionally had an efficient public social sector and a strong pro-poor political party, and government is actively seeking ways to alleviate poverty and open up the policy dia- logue. A program called the National Plan to Combat Poverty, admin- istered under the Second Vice President, has identified the 17 poorest communities in Costa Rica. Under the plan, pilot studies to analyze poverty have been initiated. Process POLICY DIALOGUE IN THE POVERTY ASSESSMENT: The Bank consulted a wide range of government line ministries as part of the preparation of the poverty assessment and later to share the findings of the report. National workshops were convened with a cross-section of stakehold- ers. By the time the assessment was completed, consensus had been achieved through dialogue, according to the Bank manager. However, some ministries were not widely aware of the report. Officials in the Ministry of Planning and Economic Policy-the implementing agency-as well as in the Second Vice President's Office felt that although the Bank had made an agreement with the previous govern- ment to undertake the poverty assessment, the consultations with the new government had been less extensive. Some officials stated that they thought the assessment was an internal Bank document. PARTICIPATORY RESEARCH PROCESS: National consultants were con- tracted to undertake the PPA. Because of the political commitment to alleviate poverty, high-level government officials supported the PPA from the beginning. Senior advisors from the Planning Ministry were involved and are now committed to incorporating the results into the analysis of poverty. They are in direct contact with the minister and have the ability to influence policy. However, the involvement of other line ministries has so far been limited, to the extent that the PPA was described by one government agency as "the secret study." In addition, ANNEXES 105 there was limited consultation with the NGO community. However, the ministry is now committed to the wide dissemination of what it perceives to be a valid and credible document. The dissemination process should result in wider ownership. There is confusion over the ownership of information contained in the PPA. In the implementing ministry, information was felt to be the property of the World Bank. Nevertheless, government officials were eager to publish the PPA results without waiting for completion of the poverty assessment, since they considered the PPA a valid stand-alone document with clear and implementable policy messages. In addition, they were concerned that the final poverty assessment would not reflect the findings of the PPA. The Bank manager attempted to gain clearance as quickly as possible for publication of the PPA but there were admin- istration delays. Permission to print and disseminate the information was finally gained, nine months after the government's initial request. METHODOLOGY-BENEFICIARY ASSESSMENT: The field work for the PPA was undertaken in December 1994 and lasted one month. Seven sites were selected from the government's National Plan to Combat Poverty, which had identified the poorest areas. A cross-section of rural, peri- urban, and urban communities was selected. The fieldwork included a combination of individual interviews and focus group discussions. A team of researchers was selected from students at the university and recent graduates. Senior government officials assisted in the field work. A consultant from the United States trained the team in interviewing techniques. During the pilot phase in one community, techniques were refined and a manual was written by the research team. The final report was written by a multidisciplinary team. The total cost of the study was US$36,500. Value Added The PPA found that housing is a major priority of the poor (up to one- third of the PPA report focused on housing). Twenty percent of those surveyed felt that housing was a major goal before any other material possession; 20 percent felt that one of their most serious problems was not having a home; and 50 percent of the families felt that their houses were in poor condition, with, for example, poor or incomplete roofing or an earth floor. Other priorities of those interviewed included poor quality of ser- vices in health centers; lack of day care centers in urban areas; and the need for more effective transport services and feeder roads to take their goods to market. Although literacy rates were high (94.6 percent for 106 CAN THE POOR INFLUENCE POLICY? females and 94.4 for males), secondary education was not perceived as a priority in a majority of households in either urban or rural areas. Links to Policy Change The PPA approach is new to Costa Rica, and the director of the study felt that the process had been a learning experience. It was the first study in Costa Rica to undertake a nationwide survey using anthropo- logical techniques. In the past, such studies were confined to small sec- tions of the population and had a sector focus. The lack of sector bias in the PPA enabled people to express priorities instead of focusing on pre- determined sectors. Because senior government officials were involved in the studies at the community level, there was a greater understanding of and com- mitment to the PPA approach within the Ministry of Planning. Ministry officials felt that the PPA approach could have a wider impact in the future. Rather than serving as an add-on to the poverty assessment, the PPA is being treated as a building block to gain a wider understanding of poverty issues. Ministry officials see a need for more participatory studies in the future. Lessons for Increasing Impact 1) Increase ownership Overall, broad ownership of the PPA study was limited despite the fact that government officials were included from the beginning. Ministry officials felt that the delay in approving publication of the findings reduced the credibility of the information in the PPA. The Min- ister of Planning and Economic Policy had already read the PPA and agreed with the conclusions but was reluctant to pass it on to the Vice President and the other ministers before receiving approval from the Bank. 2) Include a wider range of stakeholders The extent to which other stakeholders could have been involved and the timing of their inclusion were subjects of debate. The ministry felt that including a wider range of stakeholders during preparation would complicate the process. The Ministry now plans to undertake a series of workshops at the national and regional levels to disseminate the findings among a wider cross-section of stakeholders. The Association of Latin America NGOs felt that many groups had information and experience that could have been valuable during preparation of the PPA, and that involving a wider range of stakehold- ANNEXES 107 ers would have created broader support for the policy recommenda- tions. For example, the Central American Council of Cooperatives had already undertaken significant work on how poor people have been affected by various social and economic policies. The NGO association also felt that the information in the PPA could have been cross-refer- enced with existing studies to make the conclusions more representa- tive. The Ministry of Planning now intends to involve the NGOs extensively in the ongoing dialogue. 3) Dissemination of the study Impact of the PPA should increase now that the government is able to disseminate the information. Some government agencies feel they can apply the approach effectively in their own work. For example, the Social Welfare Fund is attempting to work directly with local govern- ment, and fund officials stated that the approach could assist district councils in identifying community priorities. In addition, the coordi- nating body for the National Plan to Combat Poverty commented that the PPA would be relevant to their work of realigning the program to meet community needs. Dissemination of the study to communities could help build national ownership and awareness and increase involvement of communities in the poverty debate. However, a ministry official commented that feed- back had already been given to communities during the fieldwork and that communities would be more interested in proposed interventions than in the findings of the PPA. To increase impact of the PPA, it could be disseminated through existing communication structures to broaden the policy debate. Costa Rica already has an effective communications strategy for social issues. Recent campaigns have included awareness of health and domestic violence issues. Through the use of these existing structures, the PPA could become a vehicle for deepening the understanding of poverty. A recommendation for the Costa Rica PPA, and for future PPAs throughout the Bank, is that a dissemination strategy should be part of the PPA design. It should be detailed in the terms of reference and bud- geted from the outset. 4) Management in the Bank Coordination between the poverty assessment and the PPA was logis- tically difficult. The task manager for the poverty assessment lives in Honduras and one consultant lives in Chile. Three others, however, live in Costa Rica and could have been more extensively involved in the PPA. Their involvement would have given them a better under- standing of the participatory approach and would have helped the 108 CAN THE POOR INFLUENCE POLICY? team to more effectively combine the household survey results from the poverty assessment with the results of the PPA. The resident mission felt excluded from the PPA, although the mis- sion had not been established at the time of the PPA fieldwork. The NGO liaison officer had extensive knowledge of the various groups in civil society and believes that he can now assist the government in for- mulating a dissemination strategy for the PPA. 5) Timing The Bank set a deadline for the PPA to be completed by December 1994. This was to correspond with the completion of the poverty assessment, which was later delayed. Because of this deadline, the PPA director felt that the fieldwork had been rushed and that it could have been more extensive (it should have included other poor areas, such as in the north) and more intensive (more time should have been spent at each site). Only marginal costs would have been incurred had the deadline been extended. Mozambique Background The PPA was sponsored by the Poverty Alleviation Unit (Department of Population and Social Development) of the National Directorate of Planning in the Ministry of Planning and Finance, and financed by the Department for International Development (DFID) and the World Bank through the Dutch Trust Fund for Poverty Assessments.2 Process THE POLICY DIALOGUE: The PPA was initiated in late 1994 to corre- spond with the government's preparation of a poverty assessment and was motivated by the need for qualitative insights on poverty at the household and community levels. The objectives of the exercise were to contribute to government policy formulation by the Poverty Allevia- tion Unit in the Ministry of Planning; sharpen the focus on poverty alleviation in donors' work programs; contribute to a broader under- standing of livelihood trends and changes in the country; and enhance the capacity of the Universidade Eduardo Mondlane at Maputo, the Poverty Alleviation Unit, and collaborating agencies to carry out par- ticipatory research. ANNEXES 109 The specific objectives of the Mozambique PPA, as set out in the ini- tial discussion paper,3 were to explore, in poor rural and urban com- munities, the following: * The main concerns, problems, and priorities in people's lives; how these have changed since the peace accord; how they differ accord- ing to gender; and the perceived constraints to addressing poverty problems * Local conceptions of relative well-being; causes of vulnerability and seasonal stress; and the nature and effectiveness of community cop- ing mechanisms, household survival strategies, and other (govern- ment/NGO) safety nets * Perceptions of social service delivery: access, quality, and cost of dif- ferent service providers (public, traditional, NGO) * Access to land: security and conflict in tenure, and situations under which terms of entitlement are changing * Access to infrastructure, markets, and other social and economic ser- vices; and the barriers that limit access to income and participation in markets, employment, and so forth T'he PPA was structured in three phases: Phase I: a preparatory phase to produce preliminary poverty profiles using wealth and problem rankings and priority needs assessments from two districts in each of the country's 10 provinces. Preparation for Phase I began in February 1995 and involved broad consultation with the government and the NGO, donor, and research communities; Phase II: to more closely define the research agenda, with much of the work subcontracted to partner NGOs, which carried out extended livelihood assessments in fieldwork areas and compiled poverty data for five provinces. Fieldwork for Phase II was carried out between Sep- tember and December 1996; and Phase III: a short follow-up in rural sites to capture aspects of season- ality through supplementary fieldwork in selected communities; com- pletion of overall PPA synthesis, documentation, and dissemination. Feedback on progress of the PPA was provided through regular meetings with the Poverty Alleviation Unit and line ministries, donors, NGOs, and the research community. Emerging findings from the PPA were disseminated through the national press and numerous work- shops and seminars within and outside Mozambique, including 110 CAN THE POOR INFLUENCE POLICY? through the Red Cross and the UNDP Poverty Forum. In addition, PPA outcomes were integrated into poverty analysis and participatory methodologies in academic and practical courses at the Universidade Eduardo Mondlane. Information on PPA methodology and materials was also provided to various local and international NGOs and to donors. All PPA documentation has been freely available to the public. PARTICIPATORY RESEARCH PROCESS: The methodology for the PPA was a mix of participatory rural appraisal (PRA) techniques including semi- structured conversational interviewing, direct observation, and also more complex visual research methods such as thematic mapping, sea- sonality diagramming, wealth ranking, institutional mapping, and trend and livelihood analysis. Value Added * A qualitative approach based on direct observation enabled researchers to be more flexible and open to the concerns of the poor and to encourage their direct participation. * A real strength of the PPA approach has been its inclusion of multi- disciplinary researchers and multiple stakeholders. This approach has also strengthened relationships among the participating institu- tions (the university, the government, and NGOs). * The PRA approach enabled communities to become more conscious of their life conditions, opportunities, strengths, and limitations. This is particularly important because the government does not have the capacity to help the poor in many areas of the country. The PPA has made a considerable impact through the participatory process. The participation of a variety of local institutions and stake- holders was encouraged: collaborating NGOs (partners in fieldwork in Phase II) benefited directly, while nonparticipating NGOs have used field data for improved targeting and poverty mapping data for longer- term planning; working groups in sector ministries have used informa- tion on specific sector issues (such as health, water, livestock); and Ministry of Social Action and other institutions nominated staff for PRA training and seconded staff to participate as members of the field teams. Institutional issues The PPA was adopted by the Poverty Alleviation Unit in the Ministry of Planning and Finance and contracted to the Centro de Estudos de Pop- ulafao at the Universidade Eduardo Mondlane (CEP-UEM). The emphasis ANNEXES 111 on local ownership resulted in the PPA process being successfully inter- nalized in the strategic poverty work of the Poverty Alleviation Unit. In institutional terms, Phase II provided for increased emphasis on partnerships, particularly with NGOs under subcontract, to carry out fieldwork or analyze poverty data in their areas of operation. This fea- ture of Phase II allowed for the realization of the capacity-strengthen- ing component of the PPA. Collaborating NGOs benefited from training and from guidance in poverty-sensitive community assessments as a consequence of their participation in the exercise. The PPA began as a World Bank initiative but beyond Phase I the Bank did not play a significant role in the exercise-partly because of the protracted absence of a focal counterpart in the Bank after the depar- ture of the PPA task manager in June 1996 and the consequent reorga- nization of task responsibilities in Washington. A participant in the workshop organized for this study asked whether the PPA was con- tributing to project or policy formulation at the World Bank. The facili- tator responded that the primary client for the PPA was the government of Mozambique. The government's assessment of the PPA The director of the Poverty Alleviation Unit gave a presentation at the workshop in which she underscored the value of the PPA as a source of community-level information on rural livelihood conditions in this postwar period (and given the lack of data because of conflict condi- tions). The PPA has been closely consulted by a number of ministries- Education, Health, Labor, Youth and Culture, Social Action, and Environment-as they formulate development plans. The Poverty Alle- viation Unit has also used the PPA results to evaluate proposed gov- ernment strategies and test the validity of strategic priorities. The PPA has highlighted the heterogeneity of poverty and the complexities inherent in different regions and among different social groups of the poor, and has encouraged the Poverty Alleviation Unit to systemati- cally monitor poverty in selected districts. Despite this interest at the national level, there are limitations to what the PPA can achieve because locally specific descriptive material might not be applicable at the macroeconomic level. PPAs can be valu- able at the microeconomic level, however-especially if conjoined with other survey results-even if they do not directly influence policy. In terms of institutional linkage, the bridge between the Poverty Alleviation Unit and the university was considered to be extremely beneficial, and both parties hope that their collaboration will continue. 112 CAN THE POOR INFLUENCE POLICY? An assessment by NGO partners During the field research, a representative from the NGO Kulima, from Inhambane, suggested that involvement by subcontract in Phase II of the PPA enabled the NGO to achieve greater understanding of com- munities with which they work and learn new methods for community development, especially methods for targeting vulnerable groups. With this experience, Kulima expects to scale up its participatory approach in priority needs assessments and project support. The representative also said that CEP-UEM could have provided more technical support in training and report writing.4 A representative from Concern, an inter- national NGO, noted that its participation in the PPA contributed to internal planning and programming in Nampula province. She also referred to potential conflicts in PPA outcomes, particularly if commu- nity action plans are not consistent with government priorities for a district. Enhancing in-country capacity in participatory methodologies An important feature of Phase II was the development of a PRA partic- ipation network (a rede de PRA). Through this network, PRA approaches and methods have evolved and spread rapidly, but research and process documentation are still sorely lacking. The PRA network aims to facilitate the sharing of experiences and critical reflection. It has suc- cessfully hosted several open meetings attended by representatives of government, donors, the university, and NGOs. Links to Policy Change Although policymakers generally recognize the value of the PPA, many have serious reservations about using qualitative findings from micro- economic level field studies to inform the national policy debate and create macroeconomic level policy. However, certain policy-relevant information is immediately appar- ent from the PPA. First, outputs from wealth-ranking and problem- ranking exercises in the poverty assessments show who the poor are and their priority concerns. Second, the results of aggregated liveli- hood analyses show the multidimensional reality of deprivation. In policy terms, the PPA has contributed to the poverty profiles of the Poverty Alleviation Unit; to sector working groups; to NGO opera- tions and programming; and to policy debates on livelihoods and poverty. It has also given rise to a process of participatory poverty mon- itoring and to an effective network of alliances among local and national NGOs, research institutes, and government agencies. ANNEXES 113 Table Al. PPA Outputs and Applications Relevant outputs of PPA Amenable action Government-Poverty Poverty profile; problem Rural poverty assessment Alleviation Unit ranking; livelihood and institutional analysis Government-provincial- Provincial and district Input to decentralized and district-level offices reports planning initiatives Government-sector Institutional and Sector planning and ministries livelihood analysis; policy debates priority ranking Donors Provincial summary and Review portfolio program synthesis mix Nongovernmental Local field site reports Participatory microprojects organizations Research community Site and summary Contribution to research reports seminars on livelihood changes Source: Reproduced from Phase II Summary Report (UEM 1997). Regarding the PPA's substantive contributions to a general under- standing of poverty in Mozambique, the following were considered key outputs from the work: * Phase I poverty profile outputs were based on wealth ranking in communities and on a comprehensive poverty mapping exercise using available data in Maputo and the provinces.5 As expected, the participatory poverty mapping contributed a more nuanced com- posite profile and challenged the somewhat heterogeneous catego- rizing of better-off south, average center, poor north, which has characterized much of the poverty debate. The PPA, by contrast, found poverty to be highly disbursed throughout the country, dis- trict by district. Furthermore, wealth ranking revealed community members' understanding of community-level stratification (gener- ally defined by four levels of relative well-being). * Phase I and Phase II analysis of the linkage between isolation and poverty highlighted both the negative deprivation-inducing dimen- sions of isolation and positive impacts such as social stability and environmental and natural resource balance.6 * Problem ranking in rural communities provided ample evidence of the reasoning behind long-term survival strategies, most of which were based on physical labor. The site reports showed consensus in the communities on entitlements for social welfare, identification of the most vulnerable (the elderly and the physically incapacitated), 114 CAN THE POOR INFLUENCE POLICY? and identification of those who are capable of working and should not receive formal welfare assistance. * The PRA tools of problem ranking and matrix analysis were designed to evaluate two sets of priorities, one relating directly to livelihood issues and the other to the services needed to sustain those livelihoods (and people's lives). The summary priority needs assessment from the PPA is often presented as follows:7 1. Roads/transport 2. Commercial networks/markets 3. Water 4. Health 5. Education Social services such as water, health, and education were identified as priorities by all communities. That they often were ranked after access, mobility, and infrastructure concerns probably reflects a per- ception that health, education, and water services are unlikely to be extended to inaccessible areas. Women, however, consistently gave health and other social services the higher rankings. Of interest in the problem-ranking exercises was the lack of refer- ence to consumption as a dimension of poverty at the household level, suggesting that household food security is not a common comparator of relative well-being among households. It was also surprising that rural extension ranked very low, suggesting either that extension is not effective or that it is not considered a priority. When probed, respon- dents expressed satisfaction with local technical knowledge. Lessons for Increasing Impact 1) Key issues for PPA design * Community priorities change over time in response to many social, political, and economic factors. It is important to take this into con- sideration in conceptualizing a policy dialogue mediated by PRA- type interlocutor mechanisms with communities. * PRA can be an important tool for facilitating continual dialogue between policymakers and communities, and for defining policies and strategies for implementing poverty alleviation programs. * It is important to fuse material outputs from both qualitative and quantitative research approaches and to couple qualitative and quantitative information on community priorities for action with the global policies and strategies of government and policymakers. * PRA should not be used simply as a diagnostic test to assess poverty but also as a monitoring tool at the community level. It should be ANNEXES 115 exploited to its fullest potential, enabling community members to participate and make decisions at the local level on development programs that affect them. 2) Limitations of PRA Limitations of the PRA method include the potential mismatch between the rapid application of research methods and the gradual and some- times paralyzed pace of development; the problem of transferability and replicability of methods from one village or region to another; the raising of expectations and community research fatigue; and the need for thorough training to ensure quality of facilitation. 3) Weaknesses of the approach Weaknesses of the PPA approach include: * Little standardization of criteria for the selection of community infor- mants, and a continuing tendency-despite efforts at reversal-to interview community leaders and the more visible, articulate, and sociable members of the community; * Difficulty on the part of community members in understanding the point of particular rapid appraisal methods, particularly visualiza- tion exercises such as institutional diagramming; * Limited time in the field and limited time for preparation of field- work; * Difficulty in analyzing participatory research material and drafting a summary report that reflects all interviews and community-level interactions; and * No satisfactory means to address the problems of raising expecta- tions and community fatigue with research teams. 4) Recommendations for future work Future PPA work should: * Clearly explain the research objectives to the community. Researchers should also have a thorough knowledge of the locale and of previous work conducted in the research areas. Fieldwork should not duplicate information available from previous assign- ments. * Elicit insiders' knowledge and experience of how to confront com- munity-level problems (researchers should not rely on the strong opinions of district administrators, for example). * Match the issues under investigation with the right mix of skills in the research team (particularly the gender mix of team). Research teams should also have the skills to use different methods in sequence and to overcome unanticipated obstacles. 116 CAN THE POOR INFLUENCE POLICY? 5) Main conclusions The PPA has shown that * Participatory methods can be useful for generating insights relevant to a poverty reduction strategy and that these local-level insights can be selectively translated to the national policy agenda. * Involving government policymakers in the PPA process will enhance its policy impact. * Systematically involving local NGOs for direct follow-up on community concerns and community-generated action plans is beneficial. * The participatory process is useful as a means of encouraging debate on poverty. * There is no perfect method for poverty assessment, and method- ological approaches and tools still need to be practiced and per- fected. Self-critical reflection will lead to improved poverty assessments and to improved dissemination and learning. * Assessing and alleviating poverty is a long-term effort, and PPAs should be structured with this understanding in mind. 6) Next steps The workshop participants had two main concerns related to follow-up and continuity of the PPA: how to maintain a database of district-level information and how to train teams for research and analysis. The Poverty Alleviation Unit felt that CEP-UEM should play a key role in developing participatory methods for poverty assessment in Mozambique, consolidating the experience gained to date, holding training workshops, and maintaining the link with the government. Workshop participants considered the PRA network that grew out of the exercise an important resource for linking different sources of infor- mation from different institutions. Pakistan Background In 1995, Pakistan had a per capita income of US$460 and a population of 129.7 million. With the population growing at 3 percent per year, Pakistan is one of the world's most populous and fastest growing coun- tries. The gross domestic product growth rate between 1970 and 1991 was 5.5 percent. However, disparities are high-20 percent of the households receive 43.6 percent of the total income while the poorest 20 percent receive only 7.9 percent. Pakistan lags behind other low-income countries with regard to health and education. The infant mortality rate ANNEXES 117 is more than 100 per 1,000 live births; maternal mortality is 270 per 100,000 births; and less than 30 percent of the population is literate. Process POLICY DIALOGUE IN THE POVERTY ASSESSMENT: The poverty assessment was completed in September 1995 after extensive dialogue with the government, NGOs, and other groups in civil society. The resident mis- sion organized workshops and meetings, including a high-level semi- nar in Islamabad and three provincial workshops in Peshwar, Quetta, and Lahore in December 1995, to discuss the results with a cross-sec- tion of stakeholders.This was the first economic-sector work in Pak- istan to be disseminated and discussed so widely The workshops were followed by many positive press reports and increased awareness of poverty issues. The process helped encourage the government to form a group to look specifically at poverty issues. PARTICIPATORY RESEARCH PROCESS: As part of the poverty assessment, participatory studies were carried out after the household survey analysis. The first study, funded by the World Bank and managed by the Human Resources study department and COD, was undertaken by an outside consultant working with local consultants from the Pakistan Institute of Development Economics. The Federal Bureau of Statistics was involved in selecting the communities-10 rural and urban com- munities in Punjab, Balochistan, and the Northwest Frontier Province. The fieldwork lasted for two months (March-April 1994) and was car- ried out by a roving team. The focus was on factors that influence invest- ments made by the poor in education, health, and family planning. The methodology used was open-ended interviews and focus groups. The second study took place during October and November 1993 and was funded by DFID.UK. Its main objective was to study the for- mal and informal safety nets and social networks in Pakistan. PRA methods were used to collect data. The study focused on the poorest segments of Pakistani society and was based on the perceptions of the poor. In addition, the team conducted semistructured interviews with NGOs, research organizations, and government officials at various lev- els. The research was undertaken in both rural and urban areas although there was a bias toward urban areas. Value Added The conclusion of the poverty assessment that drew the most attention was that the incidence of consumption poverty had fallen sharply, from 118 CAN THE POOR INFLUENCE POLICY? 46 percent in 1984/85 to 34 percent in 1990/91. This conclusion was quoted in the World Bank's November 1995 country assistance strategy paper. The report went on to say that a major concern in Pakistan is the low human development indicators. However, the poverty assessment integrated the results of the participatory surveys only to a limited extent. For example, the second report, funded by DFID, detailed the institutional issues related to social safety nets and the roles of govern- ment and the NGOs. That information was not extensively incorpo- rated into the final poverty assessment report. Links to Policy Change The impact of the report, both in Pakistan and within the Bank, has not been significant. Many commented that the poverty assessment was a good piece of analysis but felt that the final report had some limita- tions. Although the report recognized that poverty is multidimensional, some felt that the report could have presented the wider debate in Pak- istan as opposed to focusing on consumption poverty. Although the process of consultation was extensive, some felt that their views were not reflected in the final document. Furthermore, there is currently an extensive and well-documented debate on the measure- ment of poverty in Pakistan. To increase the impact and credibility of the poverty assessment, this debate could have been included in the report. One objective of the poverty assessment was to help reconcile the views of the government and the Bank. But some senior govern- ment officials felt that the poverty assessment did not accomplish this. Focusing on the PPA, the first survey undertaken with the Institute of Development Economics highlighted the fact that the poor spend a large proportion of their income on health but feel that service stan- dards are low and accountability is limited. However, this survey was criticized for having a limited sample size, and the validity of drawing conclusions for policy was questioned. A member of the research team stated that more detailed community-level information could have been gained if PRA methods had been used instead of just focus groups and semistructured interviews. The DFID.UK report was thought to be more credible because it used a larger sample size and included both individual and community views. However, awareness of this report was limited and it was not widely disseminated. Lessons for Increasing Impact 1) Is participation linked to influencing the final outcome? Ideally the poverty assessment is an investment in creating a policy ANNEXES 119 reform process that is a byproduct of consensus building. However, in Pakistan this has not yet occurred. The impact of the PPA might have been lessened because of the limited participatory follow-up. Work- shops and meetings are not an adequate measure of participation if those attending feel that their views have been ignored. Moreover, such an approach could have a negative impact if disappointed participants become less willing to engage in future dialogue. If their views are not included, then the reasons for this should be explained. A process of sharing results before the document is finalized may be of value to ensure that participants' views are represented and that information is not just extracted. 2) If there is a debate, it should be included in the final policy analysis. The objective of the poverty assessment in Pakistan was to contribute to the ongoing poverty debate. But the debate was not clearly reflected in the final report. As a result, many felt that the report represented only the Bank's narrow analysis of poverty. 3) Increasing the quality and credibility of participatory research. To increase the credibility of participatory research, it might be appro- priate in some countries to use the existing NGO networks, which often have a wealth of knowledge and skills. Pakistan has a number of such networks, including Strengthening Participatory Organization and Association for Development of Human Resources. The advantages of using these networks, as opposed to training new teams of people, are as follows: * Many NGOs have already established trust with communities and have undertaken participatory research. * To ensure that research is not purely extractive, the results could be followed up by NGOs working in the communities. The limitation here is that the results of a follow-up survey would be biased toward communities-not necessarily the poorest-where the NGOs have already played a role in development. * The capacity of existing NGO networks could be strengthened by the experience of undertaking countrywide PPA research. * Time-sequencing data could be collected by NGOs and links estab- lished among NGOs, policymakers, and statistical departments. However, some NGOs might have sector biases or limited capacity. To increase the credibility of participatory research, policymakers could join the teams undertaking participatory research in order to 120 CAN THE POOR INFLUENCE POLICY? understand the value and limitations of including the poor; there should be a greater focus on recording, reporting, and analyzing PRA research results to ensure that the information collected reflects the research agenda; and a dissemination strategy should be developed to feed back the results to the communities involved. For example, the DFID report was written in two volumes. The second volume con- tained the results of the surveys and was designed to be disseminated to those who participated. 4) Management in the Bank Limited ownership of the poverty assessment within the Bank appears to be linked to the lack of emphasis the poverty assessment was given as a management priority. Although the assessment took a long time to complete, a team approach was not extensively adopted. Zambia Background COUNTRY CONTEXT: Until 1975, Zambia was one of the most prosper- ous countries in Sub-Saharan Africa. According to Prospectsfor Sustain- able Human Development in Zambia (UNDP 1997), human conditions have worsened since the mid-1980s; people have become poorer and most government services have further declined. The report states that "...[economic] decline for two decades has been accompanied by stag- nation and collapse in people's livelihoods and in available forms of social support. This has been especially severe under structural adjust- ment after 1991" (Summary: page i UNDP 1997). Roughly six million people (two-thirds of the population) are living below the poverty line. Average annual growth in gross domestic product fell from 2.4 percent in the 1970s to 0.7 percent from the 1980s onward. With a gross national product per capita of only $290 in 1992, Zambia is now one of the poor- est countries in the world. The United Nations has estimated that 1.1 million Zambians will die from AIDS by the year 2005 and that Zambia is the fourth worst-affected country in the world after Uganda, Zaire, and Tanzania. In the past, there have been limited opportunities to promote partic- ipatory approaches. During the era of one-party rule, the tradition of self-help was replaced by dependency on the state. However, the capac- ity of the state to provide services was gradually eroded. Also during this period, many aspects of administrative rule were politicized, such as the positions of district governor and provincial secretary. The ANNEXES 121 appointees to these positions were not accountable to the local elec- torate, thus further decreasing the people's expectations. Prolonged economic decline led to political discontent, and with the rise of democratic elections in other countries, multiparty elections took place in Zambia in 1991. The new government has attempted to reform the economy by reducing inflation and the budget deficit. In addition, since 1991 the new government has been attempting to intro- duce a more decentralized administrative structure and promote greater participation and ownership. Donor agencies such as Africare, World Vision, and UNICEF, in conjunction with the government, have been developing participatory ways to include people in the develop- ment of their communities. A social sector Rehabilitation and Mainte- nance Task Force has been established to look into the social service delivery system and accelerate social infrastructure rehabilitation and maintenance. However, poverty continues to grow. The government has yet to for- mulate a national policy on alleviating poverty. One permanent secre- tary stated that there was a lack of national perspective on poverty issues, with members of Parliament being focused only on their own areas. She added that the civil servants and NGOs were aware of poverty issues but that members of Parliament were less aware, and she questioned whether there was a political understanding of the problem even at the highest levels. Because of a lack of exposure and adequate information on the extent and impact of poverty, there is a lack of emphasis on the problem and a consequent lack of political will. Donor and government interventions have thus remained ad hoc and uncoordinated. WORLD BANK CONTEXT: The PPA in Zambia built upon an approach developed by the Southern African Department in the World Bank and on the experiences of the Bank's Social Recovery Project (SRP) in Zam- bia. Before the PPA in Zambia, participatory research had been con- ducted under the SRP using beneficiary assessment (BA) methods such as focus discussion groups and semistructured interviewing. In 1992, when the first BA in Zambia was undertaken, the approach of consult- ing beneficiaries in a systematic way was not widespread throughout the Bank. Within the country department, management support existed and the poverty assessment manager was willing to take the risks involved in supporting a new initiative. A consultant from the division made regular visits to Zambia to assist in the development of the BA and build the capacity of the research team, located at the Rural Devel- opment Studies Bureau, University of Zambia. In 1994, the Southern 122 CAN THE POOR INFLUENCE POLICY? African Department introduced a method called systematic client con- sultation, which promoted continuous dialogue with those affected by World Bank-supported programs and projects. The Task Manager of the SRP also managed the poverty assessment and thus had already gained an understanding of the value of the approach. Therefore, unlike other countries in which PPAs have been conducted, here the Bank had experience in participatory research. Process POLICY DIALOGUE IN THE POVERTY ASSESSMENT: The poverty assessment was based on data from two priority surveys and included studies on the urban, rural, and macroeconomic sectors as well as the PPA. A wide cross-section of stakeholders was consulted throughout the process. The two Bank PPA managers were closely involved in the critiquing and commenting on drafts of the poverty assessment to ensure that the PPA material was satisfactorily integrated. PARTICIPATORY RESEARCH PROCESS: The objectives of the PPA were to * Explore local conceptions of poverty, vulnerability, and relative well- being in poor urban and rural communities in Zambia. * Explore what the poor themselves see as the most effective actions for poverty reduction that can be taken by (1) individuals or fami- lies, (2) communities, (3) government agencies, and (4) other institu- tions. * Investigate local perceptions of key policy changes related to eco- nomic liberalization. * Investigate what people in poor urban and rural communities see as the main concerns and problems in their lives at present and how these have changed over the past 5 to 10 years. The PPA was conceived and designed by the World Bank in Wash- ington and was somewhat less participatory than the poverty assess- ment. However, the preparation for fieldwork included a wider range of institutions. METHODOLOGY: A team of researchers (five women and five men) based at the Rural Development Studies Bureau at the University of Zambia conducted the research work. The team later formed an NGO called the Participatory Assessment Group (PAG). DFID contributed to the cost of training the research team, and Sida supported the in- country costs. Ten research sites were selected, representing a variety of ANNEXES 123 urban and rural communities. BA and PRA tools and techniques were used. An interview guide for semistructured interviews with individu- als and groups was compiled. The researchers prepared site reports fol- lowing each period of fieldwork. These reports were used at a final synthesis workshop to bring together policy insights and information from the exercise. There was a Poverty Assessment Conference in August 1994 at which both the PPA and poverty assessment papers were presented. In 1995, workshops were convened in four provinces to draft provincial plans of action. However, because of a lack of resources the government has not been able to hold such workshops in the remaining five provinces. Furthermore, no additional capacity was created to implement the provincial action plans. Value Added The PPA contributed to a greater understanding of the survival strate- gies of the poor; the impact of sector programs and policies; the devel- opment of both national and provincial-level action plans; and the compilation of baseline data for participatory poverty monitoring. NEW UNDERSTANDING OF POVERTY: The wealth-ranking exercises pro- vided consistent messages on the characteristics of the very poor. Many people interviewed commented on the fact that the poverty assessment was useful in the respect it was the first comprehensive study on poverty in Zambia. One important finding of the PPA was that the term "female-headed household" did not fully capture what the report suggested is better understood as the "feminization of poverty" (see World Bank 1994d, Vol. 1, p. 135).The PPA highlighted the fact that "women without support" was a more appropriate term. This term describes women who have no current relationship with a man and have no adult children who could provide either labor or remittances. Women without support were often ranked as the poor- est by the communities. The priority-ranking exercises provided valuable insights into the cross-sector balance of priorities. Consistent messages were generated from these exercises. Seasonality analysis revealed the dynamic dimen- sions of poverty (see World Bank 1994d, Vol. 1, p. 47) and covered issues such as income and expenditure, health status, and food security. Stress periods such as the hungry season in urban and rural areas were high- lighted through the participatory research and incorporated into the final report (see World Bank 1994d, Vol. 1, p. 52). At the community level, the PPA covered access to services such as health, education, and 124 CAN THE POOR INFLUENCE POLICY? credit. The information was detailed and comprehensive and was dis- aggregated by gender where appropriate. PARTICIPATORY POVERTY MONITORING (PPM): PAG now undertakes yearly PPAs in some of the same communities as well as some new communities, to monitor changing living conditions. The results of the participatory poverty monitoring are used as a complement to house- hold survey data. INSTITUTIONAL CAPAcIrY BUILDING: The PPA has contributed to the cre- ation of an in-country capacity to conduct participatory research on an ongoing basis. PAG was officially registered as an NGO in August 1995. The group originated at the University of Zambia, where members used to undertake research assignments for the university's Rural Develop- ment Studies Bureau. In August 1994, the Rural Development Studies Bureau was phased out and only 3 of the 11 members were retained by the university. PAG now consists of an interdisciplinary and gender-bal- anced team of 12 people-6 men and 6 women from various disciplines. The World Bank, Sida, and the Microprojects Unit of the European Union have continued to increase the capacity of PAG. Since 1992, the members have received training in PRA methods from the Institute of Development Studies and other consultants. PAG continues to do research and PRA training for government ministries and donor agen- cies. Its current program includes BAs, participatory planning, and PPAs. PAG works with government ministries and donor agencies and is conducting a study for Sida on Coping with Cost Sharing in Health and Education. In the future, PAG will work closely with the LCMU in the Department of Statistics. It has recently moved its offices to the Cen- tral Statistical Office with the objective of more closely coordinating its participatory research with traditional household surveys. Links to Policy Change The PPA influenced the poverty action plan recommended in the poverty assessment. The stress on rural roads and water infrastructure and on urban services such as water supply were revealed by the PPA. The poverty profile in the poverty assessment also drew from the PPA on such issues as community-based identification of the ultra-poor, coping strategies, safety nets, and targeted interventions. The govern- ment was also influenced by the priorities expressed by the poor in the ranking exercises. Positive feedback was received from communities involved in the PPA on the functioning of the emergency safety net dur- ing the Southern Africa drought of 1992. ANNEXES 125 In recognition of the value added of the PPA, a permanent secretary stated "Everyone knows that poverty exists in Zambia and people always talk about it. But the PPAs have enabled us to appreciate thefact that there is growing poverty in urban areas. Even high-ranking politicians do not talk about urban poverty. The PPAs are helping us appreciate, therefore, that poverty is a nationwide problem, not just a rural one" (Personal communication). MINISTRY OF HEALTH: The Ministry of Health has been using the results of the PPA and the poverty assessment to develop policy. The National Strategic Health Plan refers specifically to the poverty assess- ment. A policy recommendation from the PPA was that the drought area should be exempt from paying health fees. This was taken up by the Ministry of Health and is now policy. In addition, the PPA high- lighted the fact that the poor were not using health facilities because of the rudeness of health staff. To empower and decrease the frustration of health workers, the Ministry of Health has increased resources allo- cated to rural areas. As a result of the PPA, PAG undertook an evaluation of the Public Welfare Assistance Scheme in 1996. The evaluation recommended that communities should select the beneficiaries of the scheme. Closely con- nected to this evaluation was a further study undertaken by PAG to develop an eligibility profile for those who should receive welfare ben- efits and exemptions from health care costs and education fees. This study was undertaken in collaboration with the ministries of Health, Education, and Community Development. MINISTRY OF EDUCATION: In the Ministry of Education, a new policy is being prepared regarding the timing of school fees, which currently coincide with the period of maximum stress. MINISTRY OF AGRICULTURE: The PPA methodology is being replicated in the Agricultural Sector Investment Project for planning and moni- toring. DONORS AND NGOs: Some of the NGOs interviewed for this study by the local research team felt that "the use of participatory methods in the preparation of the Poverty Assessment by the World Bank encouraged and justified their own use of [qualitative] methods" (Mutesa and Muyakwa 1997, p. 15). The researchers added that some NGOs were surprised at certain results, such as the finding that Copperbelt is a very 126 CAN THE POOR INFLUENCE POLICY? poor province. This information has encouraged them to initiate proj- ects in that province. Lessons for Increasing Impact 1) The strengthening of PAG The sustainability of PAG is a key concern at this stage. PAG has the potential to influence other projects and government policies. It also has the potential to help increase the understanding of poverty by com- bining its participatory work with quantitative surveys. PAG's capac- ity to continue to produce good-quality work is in question, however, because it has a limited capacity to analyze results and write reports. Although PAG has received extensive support from the World Bank's Social Recovery Project, continual follow-up is required to ensure that quality is maintained and management systems are established. 2) Methodology Working with communities requires detailed follow-up on the effec- tiveness of various approaches. For example, Milimo, Norton, and Owen (1998) point out that "In the first PPA one of the field teams held regular meetings to check on recording and reporting, to discuss find- ings and strategies, and to plan the next day's work, while the other field team functioned with less coherence. The difference in the quality and coherence of the outputs and policy insights was very striking" (p. 109). In addition, PAG stated that by staying overnight in the villages, the team developed more trust with the communities. The PAG team recommended the use of PRA tools in future research because such tools can lead to "greater involvement of the communi- ties and more enthusiasm" and "encourage the participation of the women" (Personal communication). For PRA, continual training of field researchers is required to ensure that teams are adhering not only to the methods but also to the principles of such research; that is, by embracing error, showing respect, optimal ignorance, offsetting biases, and triangulation of data. 3) Process issues * The researchers felt that the time frame for the PPA had been too tight, with only four months from research design to analysis. * There were differences in undertaking research in urban and rural settings. Urban communities were more complex and more difficult to organize, with community being difficult to define. Some method- ologies, such as wealth ranking, were inappropriate because neigh- bors were not always aware of each other's wealth or the patterns of ANNEXES 127 social networks. In rural areas the social networks were more visible, being based, in some cases, on kinship and community. * The institutional framework should be studied further. As Milimo, Norton, and Owen (1998) add, "The PPA was much more effective at eliciting priorities at the local level than on outlining the institutional mechanisms by which identified needs and problems could be resolved-a stronger focus on institutional issues would have increased policy impact" (p. 110). * The manager of the poverty assessment stressed the importance of combining the PPA data with other methodologies such as longitu- dinal sociological studies, survey data, econometric modeling, and household behavior models. Notes 1. For example, life expectancy is 75 years; the infant mortality rate is 15 per 1,000 live births. 2. This section is based on a summary of a workshop by D. Owen carried out for this study. 3. See Owen (1994). 4. CEP-UEM did provide Kulima with technical assistance for PRA training in Inhambane. 5. Universidade Eduardo Mondlane (1996). 6. See especially World Bank (1996k) for a preliminary discussion on relative isolation. 7. Problem ranking and priority lists are dependent on context and are vul- nerable to misinterpretation, indirect influence, and poor facilitation. General- izing on the basis of local ranking exercises should be done with utmost caution and the results treated as indicative only. 128 CAN THE POOR INFLUENCE POLICY? Annex 6. Methodology of This Review The first phase of the participatory poverty assessment (PPA) review was a desk study based upon existing PPAs, poverty assessments, and related documents, both within the Bank and outside the Bank. In addi- tion, semistructured interviews were held with a wide cross-section of people in the Bank who undertook the PPA and/or the poverty assess- ment. This first phase resulted in the formulation of a number of hypotheses. A World Bank in-house workshop was convened in Janu- ary 1996 to discuss the results of the desk study and interviews. The results of the first phase and the PPA in-house workshop were then dis- cussed at a workshop at the Institute of Development Studies in Sus- sex, UK, in May 1996. Many of the PPA practitioners, from a cross-section of countries, presented their experiences.' The hypothe- ses were then tested in the following countries during a second phase of field work from 1996 to 1997: Zambia, Costa Rica, Pakistan, Mozam- bique, and Swaziland. A variety of approaches were used, including semistructured interviews, focus groups, and workshops with com- munities, government officials, donors, NGOs, and civil society orga- nizations. In presenting good-practice situations, it has been difficult to repre- sent the perception of all participants in this limited study. Personal interpretation has been inevitable, although an attempt has been made to present multiple perspectives. Much of the work of the PPAs has been innovative and new. The main objective of this study has been to identify examples from which to learn. It is hoped that this study will also be useful to practitioners. The analysis has relied heavily on many ideas from people both within and outside the Bank. Notes 1. The outcome of this workshop is sununarized by Holland and Blackburn (1998). THE WORLD BANK 1818 H Street, N.W Washington, D.C. 20433, U.S.A. TELEPHONE: (202) 477-1234 FACSIMILE: (202) 477-6391 TELEX: MCI 64145 WORLDBANK MCI 248423 WORLDBANK WORLD WIDE WEB: http://www.worldbank.org/ E-MAIL: books@worldbank.org CI. COVER DESIGN: THE MAGAZINE GROUP ISBN 0-8213-4144-8