Zambia 74381 Economic Brief Recent Economic Developments and the State of Basic Human Opportunities for Children © 2012 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street NW Washington, DC 20433 USA All rights reserved This report was prepared by the staff of the Africa Region Poverty Reduction and Economic Management. The findings, interpretations, and conclusions expressed herein are those of the authors and do not necessarily reflect the views of the World Bank’s Board of Executive Direc- tors or the countries they represent. The report was designed, edited, and typeset by Communications Development Incorporated, Washington, DC. Contents Foreword   v Acknowledgments   vi Executive summary   vii Section 1  Recent economic developments and medium-term prospects    1 Recent developments in the global and Sub-Saharan economies    1 Recent economic developments in Zambia    2 Section 2  The state of basic human opportunities for children in Zambia    11 Inequality in Zambia   11 From inequality to equity    13 Summary of main findings and conclusion    27 Annex A  How is the Human Opportunity Index calculated? A simple example    31 Annex B  Three key properties of the Human Opportunity Index    32 Annex C  Estimating the Human Opportunity Index from household survey data    33 Annex D  Shapley Decomposition of the D-Index—An example    34 Annex E  Opportunities and circumstances for Zambian children    35 Notes   36 References   38 Boxes 2.1 The concept of equality of opportunity    13 2.2 The Human Opportunity Index    15 2.3 The dissimilarity index   18 2.4 What is the Opportunities Benefit Incidence Analysis?    27 Figures 1.1 Sub-Saharan Africa continues robust growth    1 1.2 Industrial metal prices, January–September 2012    2 1.3 Ratio of gross disposable national income to gross domestic product in Zambia, 2000–12   3 1.4 Net savings and “genuine” savings, 2000–10    5 1.5 Recent comovement in domestic revenue (2003 prices) and copper prices, 2004–12   6 iii Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n 1.6 International reserves and current account balance, 2003–12    6 2.1 Consumption growth incidence curves for urban and rural areas, 2006–10    11 2.2 Inequality in Zambia and the world    12 2.3 Coverage and the Human Opportunity Index for school attendance of children ages 10–14 in Zambia and selected Latin American and Sub-Saharan countries    16 2.4 Coverage and the Human Opportunity Index for finishing primary school on time in Zambia and selected Latin American and Sub-Saharan countries    16 2.5 Coverage and the Human Opportunity Index for access to safe water on site in Zambia and selected Latin American and Sub-Saharan countries    17 2.6 Coverage and the Human Opportunity Index for access to improved sanitation in Zambia and selected Latin American and Sub-Saharan countries    17 2.7 Coverage and the Human Opportunity Index for access to electricity in Zambia and selected Latin American and Sub-Saharan countries    17 2.8 Coverage and the Human Opportunity Index for some key opportunities in Zambia, 2010   19 2.9 Change in the Human Opportunity Index and decomposition of changes, 2006–10   20 2.10 Average annual change in the Human Opportunity Index for Sub-Saharan countries   21 2.11 Contribution of circumstances to overall inequality of opportunity, 2010    23 2.12 A snapshot of the vulnerability profile    24 2.13 Relationship between poverty and human opportunity    25 2.14 Poverty and inequality of opportunity (rank correlation)    26 2.15 Distribution of gross and net unitary benefits from public education, by quintiles of probability   28 Tables 1.1 GDP growth, by main sectors, 2003–12    3 1.2 GDP growth for comparator countries, 2003–12    4 1.3 Summary of central government finances, 2008–12    4 1.4 Proposed use of the sovereign bond proceeds    5 1.5 Selected balance of payments indicators, 2008–12    7 1.6 GDP growth projections, by main sectors, 2012–14    8 D1 D-Index based on circumstance set    34 E1 Opportunities and their definitions    36 E2 Circumstances and their definitions    36 iv Foreword With this first Zambia Economic Brief, the gender, ethnicity, place of birth, and family World Bank is launching a series of short eco- origins should not determine people’s eco- nomic updates that will be produced twice a nomic, social, and political success. A person year. Each brief will include two sections: the should not have fewer opportunities in life Bank’s assessment of recent economic devel- just because she is a girl or born in a rural opments and outlook in the short to medium area. This is the core principle behind the term; and its analysis on a specific develop- concept of equality of opportunity—and it ment topic or theme. We expect these briefs is also the framework adopted in this report to support evidence-based policy debate in for Zambia. The analysis shows the extent the country, strengthen public demand for to which basic opportunities for education, good policies and outcomes, and inform gov- health, and infrastructure services in Zam- ernment policies and programs. bia are influenced by circumstances in which In this brief, section 1 reports on Zambia’s children find themselves. We hope that continued robust growth, fiscal outcomes, this new approach will stimulate debate on and capital inflows at the end of 2012 against inequalities in access to basic services, and the backdrop of performance of other Sub- help the government in its quest to attain a Saharan and global economies. It stresses the better Zambia for all. need for the government to spend borrowed The second brief to be produced in June and own resources more prudently. While 2013 will cover the specific development acknowledging continuing strong prospects topic of jobs. for growth, the analysis highlights consider- able downside risks emanating from global uncertainties. Section 2 covers basic human opportuni- ties for children in Zambia. The opportu- Kundhavi Kadiresan nities approach is premised on the notion Country Director for Zambia that predetermined circumstances such as The World Bank v Acknowledgments The first Zambia Economic Brief has been Anwar Ravat, Barbara Senkwe-Kazimbaya, prepared by the Poverty Reduction and Eco- David Sislen, Katia Herrera Sosa, and Rose- nomic Management Unit of the World Bank mary Sunkutu. Ambar Narayan provided Country Office in Lusaka. The team was guidance to the work on human opportu- led by Praveen Kumar and included Allen nities. Jumbe Ngoma and Stevan Jackson Dennis, Asumani Guloba, Kutemba Kam- organized dissemination events. Kundhavi bole, and Sailesh Tiwari (PRMPR). Sailesh Kadiresan, Country Director for Zambia, and Tiwari led the work on the state of human John Panzer, Sector Manager of the AFTP1 opportunities, the focus topic of this issue, unit, provided overall guidance and advice. as a guest co-author; he was assisted by Ale- The team benefited greatly from consulta- jandro Suarez. Sandeep Mahajan and Gayle tions with officers in the Ministry of Finance Martin provided peer review comments. of Zambia. Valuable comments were received at various The report was edited and laid out by stages from Kate Bridges, Colins Chansa, a team at Communications Development Nalini Kumar, Rama Laxminarayanan, John Incorporated, led by Bruce Ross-Larson. Makumba, Patricio Marquez, Brian Mtonya, vi Executive summary Zambia’s economy continued its strong increasing and thus increasing exposure to growth in 2012, estimated at 7.3 percent, just copper price changes. above the 6.8 percent in 2011. Zambia has Capital inflows to Zambia have remained grown faster than most of its peers, including resilient despite global slowdown, with net both mineral-producing and nonmineral- foreign direct investment (FDI) and portfolio producing countries. investments growing steadily, from $350 mil- Growth has been broad-based, led by lion in 2009 to an estimated $991 million in strong performance in agriculture, manu- 2012. FDI inflows continue to be directed facturing, and services, even as mining out- mainly at the mining sector, with manu- put contracted for the second year despite facturing, communications, and financial a new mine. The decline in global copper institutions also contributing to recent FDI prices contributed to the slight contraction growth. Gross international reserves are in mining. Agricultural growth has been at a level of about 3  months of prospective accompanied by increased diversification of imports. The government aims, rightly, to production, with a higher share of nonmaize further increase the reserve coverage, mainly crops—such as wheat, barley, sorghum, and to guard against terms of trade shocks. soybean—partly in response to uncertainty about the government’s maize-buying pro- Prospects for strong growth but gram. Construction growth accelerated considerable downside risks in recent years in response to increased Medium-term prospects for Zambia’s eco- demand coming from rising urban incomes nomic growth remain strong but are subject and a marked pick up in investment in min- to considerable downside risks emanating ing and roads. Growth in transport services is from global uncertainties. GDP growth is also a response to strong growth in demand expected to average more than 7  percent from other sectors of the economy. over the 2013–14 forecast horizon, making Zambia’s recent fiscal policy has been Zambia one of fastest growing Sub-Saharan expansionary. After grants, the country’s economies. Underpinning the projected average overall balance over 2010–12 was growth over the medium term are favor- –3.5 percent of GDP. This policy is sustain- able external and domestic developments. able from a debt sustainability point of view. Despite the ongoing slowdown in China’s However, it needs to be matched to the coun- economy (Zambia’s largest trading part- try’s capacity to prepare and implement high- ner), international metal and mineral prices return projects and spend more efficiently. (including that of copper) are expected to There is also a need to develop fiscal buffers remain fairly high over the medium term. to smoothen spending across shocks. The last As a result, the strong FDI flows to the min- point is gaining importance, with the share ing sector of recent years are expected to of mining revenues in government finances continue. vii Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n Agriculture prospects are good assum- on reducing poverty, particularly in the lat- ing that government policy in the sector will ter half of the decade—for two main reasons. improve, with the Farmer Input Support Pro- First, the growth—driven by industries such gramme becoming more efficient, which is as mining, construction, financial services, expected to enhance agricultural productiv- and tourism—did very little to create new ity. They will also be helped by the proposed jobs and expand economic opportunities scaling up of extension services, irrigation, beyond the small part of the country’s labor and research. Accompanying the mining and force already employed in these industries. agricultural growth is growth in transport, Second, the urban-centered growth also storage, and construction. failed to generate enough spillovers for the Aggregate demand is expected to continue roughly two-thirds of Zambian people who supporting growth. Fiscal policy is expected to live in rural areas, depend on agriculture, remain expansionary, contributing to higher and have seen their incomes stagnate over growth prospects. In addition, recent real the last decade. wage increases (for both the public and private Such growth led to an increasing dispar- sectors) will support household consumption, ity between rural and urban areas. Annual the largest contributor to aggregate demand. consumption growth of different wealth Household domestic spending should also quintiles was less than 1 percent for most of benefit from a stable macroeconomic environ- the rural population between 2006 and 2010, ment (average inflation expected at 6–7 per- whereas growth in urban areas was about 2 cent), lower real interest rates, and increased percent for the first four quintiles and much access to consumer credit. higher for the wealthiest quintile. There are Domestic investment is also expected to also sharp disparities among the country’s continue to support growth. The latest Bank nine provinces. In fact, inequality between of Zambia quarterly survey of business opin- rural and urban areas and among provinces ion and expectations forecasts that business accounts for more than three-quarters of the confidence will remain high through 2013, observed inequality in the country. with firms investing in construction and The concentration of economic growth machinery. in particular sectors and regions over a sus- The expected pickup in the global eco- tained period also manifests itself in Zam- nomic environment, as well as in some of bia’s persistently high inequality. The Gini Zambia’s major trading partners such as the coefficient for consumption in Zambia was European Union and South Africa, should 52 in 2010 and has generally remained above increase demand for Zambia’s exports— 50 for much of the last decade, making Zam- both traditional (such as copper) and non- bia one of the most unequal countries not traditional (such as sugar and beef)—as only in Sub-Saharan Africa but also in the well as tourism services. Demand should be world. The richest quintile of the population supported by growth in Sub-Saharan Africa, in Zambia accounts for 57 percent of con- expected at 6.2  percent in 2013, excluding sumption, compared with 4 percent for the South Africa. poorest quintile. Although Zambia’s growth prospects are fairly robust over the forecast horizon, signifi- Basic human opportunities for children cant headwinds could lead to weaker-than- in Zambia envisaged real GDP growth over 2013–14. Analyzing the opportunities for children The source of risks lies abroad in continued in Zambia can better illuminate the nature tensions in global financial markets, slower and causes of inequality of outcomes among growth in emerging economies, and increases adults. Opportunities for children can also in oil prices—and at home in persistent per- be reliable predictors of economic mobility ceptions of an uncertain policy environment across generations and over time. If access to and adverse weather-related shocks. economic opportunities (such as jobs, credit, land, wages, and financial assets) is corre- Poverty and inequality lated with the circumstances of an individual Despite Zambia’s robust economic growth (such as parental socioeconomic status and over 2000–10, there was very little progress location of residence), as it appears to be in viii Zambia, then it reinforces the link between opportunities across the regions in Zambia children’s circumstances and their opportu- closely mirrors the spatial distribution of nities in life, in turn perpetuating economic poverty, implying that poorer regions have inequality between groups with different lower access to opportunities. In addition, circumstances over generations. For policy the poorest regions in Zambia are also the makers interested in closing gaps in out- regions where children’s circumstances play comes between subgroups of the population the strongest role in determining their access in Zambia, a comprehensive analysis of these to these opportunities. This depresses pros- opportunities can provide a useful blueprint pects of economic mobility in the poorer on which efforts to level the playing field can provinces and may result in the further be based. entrenchment of poverty and the propaga- One of the primary findings here is that tion of the spatial inequalities observed in opportunities for children in Zambia vary Zambia today. widely across different types of goods and Current patterns of public spending, at services. But in general none of the opportu- least in the education sector, are distributed nities is close to universal in their coverage. such that they benefit children fairly uni- Some, such as school attendance for children formly, irrespective of their circumstances. under age 16, are available to most Zambian But this seeming progressivity is achieved not children in the age bracket. But still roughly necessarily because larger shares of public 20 percent of children remain excluded. expenditures are being targeted to children Only about half of children start and finish with the lower probabilities of having access primary school on time, with sizable inequal- to these opportunities, but because the pri- ities attributable to circumstances children vate costs borne by families of children from have no control over. And the quality of edu- the top end of the distribution are signifi- cation is low. Of Zambian children in grade cantly higher. Given that roughly a fifth of 6 who took the regional standardized tests children who should be in school are not cur- designed to monitor educational quality, 44 rently enrolled in Zambia, there seems to be percent had proficiency below basic reading more the government could do in realigning skills, and 67 percent did not possess even expenditures. the basic numeracy skills. Realigning expenditures in education The picture is bleaker for opportunities and health to prioritize the poor and the in health and infrastructure, where cover- underserved and ensuring that the amount age is lower and there are large variations spent is effective will enhance the desired in the inequalities based on children’s cir- results even with the same resource enve- cumstances. Full immunization, exclusive lope. The government already allocates the breastfeeding until the child is 6 months old, bulk of spending in education on primary and growing up through childhood with- school enrollment, for example. Yet as the out being scarred by chronic malnutrition opportunity-benef it incidence analysis are some opportunities critical for a healthy shows, there is significant room to target start to a productive life. But roughly half children whose circumstances make their of Zambian children are deprived of these likelihoods of having access to this oppor- opportunities. Coverage of opportunities on tunity lower. This has to go with necessary access to clean drinking water, adequate sani- changes to the institutional setup of service tation, and electricity is also extremely low, delivery to provide proper incentives to the with coverage depending substantially on service provider and voice and accountabil- the socioeconomic status of the household in ity to the end user, critical to ensuring that which the child is born and the location of quality of the services delivered­ — in educa- residence. tion or health—is of minimum acceptable The analysis also reveals household’s standards. socioeconomic status, urban-rural residence, With the largest driver of unequal access and the province of birth as the strongest to opportunities in Zambia as the province drivers of inequality across broad oppor- of birth, investments in the lagging regions tunities. The regional aspect is particu- need to be scaled up, particularly for infra- larly interesting because the distribution of structure services such as drinking water, ix Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n sanitation and electricity where access is primary school are much more likely to not highly concentrated in certain urban pockets finish primary school themselves, to suffer within particular provinces. from acute malnutrition, or to have access to Finally, policy makers also need to recog- sanitation facilities. The presence of multiple nize that children of certain circumstances deprivations points to the need for policy are vulnerable to deprivations in multiple programs in different sectors (health and dimensions simultaneously. For example, education, for example) to closely coordi- Zambian children living in rural areas and nate to achieve better efficiency and the best with household heads who did not finish results. x Section 1 Recent economic developments and medium-term prospects Recent developments in the global and But global trade is lagging behind the Sub-Saharan economies recovery. In the three months to August, After several months of deceleration in the import demand was still contracting in Ger- first half of 2012, the global recovery started many, Japan, and the United States. Given picking up again, with global industrial pro- that recoveries in global trade tend to lag duction rising at a seasonally adjusted annu- behind those of industrial production, global alized rate of 0.9 percent in the three months trade should rebound by the fourth quarter to August. Activity remains strongest in devel- if the recovery in industrial production is sus- oping countries, with a 5.4 percent annual- tained. Sub-Saharan data through June show ized pace of growth in the three months that exports contracted, largely in response to August (4.5 percent excluding China). to a slowdown in large developing economies Growth in Sub-Saharan Africa remains (such as China), an important destination for robust amid the turbulence in high-income Sub-Saharan Africa’s exports of metals and countries. Indeed, a third of the region’s minerals. countries (including Zambia) will be grow- In line with the recent growth in global ing at or above 6 percent in 2012, with some industrial production, base metals strength- of the fastest growing buoyed by new mineral ened in September, after several months of exports (iron ore in Sierra Leone, oil and decline. The World Bank’s metal and miner- uranium in Niger), a return to peace (Côte als index was up 5.8 percent in September, d’Ivoire), and robust growth in the nonmin- with strong price increases in such key base eral sector (Ethiopia; figure 1.1). metals as copper (7.6 percent), aluminum Figure Sub-Saharan Africa continues robust growth 1.1 8 Sub-Saharan Africa (excluding South Africa) 6 4 Percent 2 Developing countries (excluding China and India) 0 World –2 –4 2005 2006 2007 2008 2009 2010 2011 2012 Source: World Bank 2012a. 1 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n (11.9 percent), zinc (10.5 percent), and nickel year despite a new mine (Luashya-Muliashi; (9.9 percent). Some of the increases could table 1.1). A combination of prolonged trade be due to speculative investments follow- union negotiations and the decline in global ing recent monetary stimulus measures in prices contributed to the slight contraction. G3 economies (the United States, European Agricultural growth has been accompanied Union, and Japan), but the increases more by increased diversification of production, likely reflect the moderate strengthening in with a higher share of nonmaize crops— global economic activity (figure 1.2). such as wheat, barley, sorghum, and soy- The financial market has continued to bean—partly in response to uncertainty show signs of stabilization, with the gains about the government’s maize-buying pro- observed through August sustained in Sep- gram. Construction growth accelerated tember. The reduced uncertainty in finan- in recent years in response to increased cial markets and increased liquidity boosted demand coming from rising urban incomes gross capital flows to developing countries and a marked pick up in investment in min- by $40 billion in September, with bond flows ing and roads. Growth in transport ser- posting a record $32 billion. Unprecedented vices is also a response to strong growth in investor demand allowed frontier-market demand from other sectors of the economy. borrowers to tap the international bond Telecommunication services continue to market, with Zambia issuing an inaugural grow, due both to an expanding customer international bond, successfully raising $750 base and rising share of services in the con- million. Although foreign direct investment sumption bundle. (FDI) data are lagging, FDI flows, which are Looking at Zambia’s largely foreign- less sensitive to changing market sentiments owned and -operated mining enterprises, it than short-term capital flows, likely picked is meaningful to monitor gross national dis- up in the third quarter. In Sub-Saharan posable income (GNI)—which excludes pay- Africa FDI inflows are likely to remain resil- ments to foreign labor and capital employed ient, coming in at around $31 billion in 2012 in Zambia­ — as a measure of contribution to (similar to the $32.5 billion in 2011; World national wealth. Figure 1.3 shows Zambia’s Bank 2012a). GNI as a ratio of GDP. This ratio has fluctu- ated in the low 90s during normal economic Recent economic developments in times and shrunk considerably in uncertain Zambia times. Over 2009–11 it averaged about 93 Zambia’s economy continued its strong percent, so about 7 percentage points of GDP growth in 2012, at 7.3 percent, just above the were not available as national disposable 6.8 percent in 2011. Growth has been broad- income. based, led by strong performance in agri- Zambia is among Sub-Saharan Africa’s culture, manufacturing, and services, even fastest growing countries, with projected as mining output contracted for the second growth of 7.3 percent in 2012, well above Figure Industrial metal prices, January–September 2012 1.2 110 Aluminum Real metal prince index Zinc (April 2012 = 100) 100 Copper 90 World Bank Metal and Mineral Price Index 80 Jan. Feb. Mar. Apr. May Jun. July Aug. Sep. 2012 2012 2012 2012 2012 2012 2012 2012 2012 Source: World Bank staff estimates. 2 Table GDP growth by main sectors, 2003–12 1.1 (constant price = 1994; percentage change from previous year, unless otherwise indicated) Contribution to GDP growth 2012 2012 (preliminary; Indicator  2003–08 2009 2010 2011 (preliminary) percentage points) Primary sector 3.8 12.4 10.2 2.2 4.1 0.9 Agriculture, forestry, and fishing 2.3 7.2 6.6 8.0 7.1 0.9 Mining and quarrying 6.5 20.3 15.2 –5.2 –0.3 0.0 Secondary sector 9.1 6.2 6.5 8.5 11.2 2.7 Manufacturing 4.3 2.2 4.2 8.0 11.2 1.0 Electricity, gas, and water 2.4 6.8 7.4 8.2 2.3 0.1 Construction 17.7 9.5 8.1 8.9 13.0 1.6 Tertiary sectora 5.9 3.9 6.6 7.8 6.8 3.7 Wholesale and retail 3.4 2.3 4.2 7.5 8.9 0.7 Restaurants, bars, and hotels 9.3 –13.4 10.2 7.9 2.1 0.1 Transport, storage, and communications 13.2 7.6 14.9 13.7 11.3 1.2 Financial institutions and insurance 4.5 5.2 6.0 4.9 6.0 0.4 Real estate and business services 3.4 2.8 3.0 2.9 2.9 0.2 GDP 5.7 6.4 7.6 6.8 7.3 7.3 GDP, minus mining 5.6 5.2 6.8 8.2 8.0 GNI 4.7 13.8 1.9 7.2 6.1 Memorandum items GDP at current market prices (billions of kwacha) 36,364 64,616 77,667 93,345 105,256 GNI at market prices (billions of kwacha) 32,858 62,503 71,128 85,760 97,344 Nominal GDP per capita (U.S. dollars) 743 990 1,221 1,414 1,445 Nominal GNI per capita (U.S. dollars) 671 958 1,118 1,299 1,336 Note: Blank cells are not applicable. a. Includes community, social, and personal services and others. Source: Zambian authorities, IMF, and World Bank staff estimates. Ratio of gross national disposable income to gross domestic product in Zambia, Figure 2000 –12 1.3 1.00 0.95 GNI/GDP ratio 0.90 0.85 0.80 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: World Bank staff estimates. the region’s 4.8 percent, at almost the same is conducive to steady growth in regional pace as in 2011. Excluding South Africa, the demand for its exports. region is expected to grow at 6 percent in 2012. In the recent past Zambia has grown Central government finances—growing faster than all its peers, including both public investment mineral-producing and nonmineral-pro- In recent years the share of mining revenue ducing countries (table 1.2). Moreover, it is in central government finances in Zam- surrounded mostly by countries on a steady bia has grown, while the share of foreign and moderately high growth path, which grants declined (table 1.3). Over 2008–11 3 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n Table GDP growth for comparator countries, 2003–12 1.2 (percentage change from previous year) Country and region 2003–08 2009 2010 2011 2012f Zambia 5.7 6.4 7.6 6.8 7.3 Angola 15.0 2.4 3.4 3.4 8.1 Botswana 4.5 –4.9 7.0 5.1 5.5 Congo, Dem. Rep. 6.1 2.8 7.2 7.0 7.2 Ghana 6.3 4.7 6.6 14.4 7.2 Kenya 4.8 2.6 5.8 4.4 5.0 Malawi 5.9 7.6 7.1 5.0 4.5 Mozambique 7.5 6.3 6.8 7.1 6.7 Rwanda 7.5 4.1 7.2 8.6 7.4 Tanzania 7.2 6.0 7.0 6.4 8.0 Uganda 7.9 7.2 5.2 6.7 4.0 Zimbabwe –9.1 5.7 9.0 5.0 2.5 Sub-Saharan Africa 5.8 2.0 5.0 4.7 4.8 f = forecast. Source: World Bank Global Economic Prospects. mining revenues averaged around 2.6 per- improve mining tax compliance while keep- cent of GDP, a sharp increase from the pre- ing the regime predictable. To that end, the vious range of 1.0–1.4 percent over 2003–08. 2013 budget proposes changing the treat- Despite a decline in mining output, mining ment of capital expenditure for tax purposes revenues are expected to be 3.8 percent of and better monitoring minerals being pro- GDP in 2012, partly due to an increase in roy- duced and exported. Mining revenues are alty rates. The government is making efforts expected to continue growing in the medium to further rationalize the fiscal regime and term, as carryover losses of mining companies Table Summary of central government finances, 2008 –12 1.3 (percent of GDP) 2012 2008 2009 2010 2011 (preliminary) Revenue 23.0 18.9 19.6 21.7 21.5 Tax 17.6 14.6 16.4 19.3 17.9 Nontax 1.3 1.4 1.4 1.6 2.0 Grants 4.1 2.9 1.8 0.8 1.6 Expenditure 23.9 21.3 22.6 23.9 25.8 Recurrent 20.4 17.9 19.4 19.7 19.8 Out of which interest payments 1.7 1.6 1.8 1.2 1.8 Out of which employee compensation 8.2 8.2 8.1 7.9 9.0 Out of which FISP 0.9 1.2 2.4 1.0 0.5 Out of which FRA 0.8 0.9 0.8 1.8 1.2 Capital 3.5 3.4 3.2 4.2 6.1 Overall balance (including grants)a –1.5 –2.4 –3.1 –3.9 –4.3 Financing 1.5 2.4 3.1 3.9 4.3 External (net) 0.5 –0.1 0.3 1.7 3.8 Domestic (net) 1.1 2.5 2.7 2.3 0.5 Memorandum items Primary balance 0.2 –0.8 –1.3 –2.7 –2.5 Mining revenues 1.9 1.0 1.9 5.5 3.8 Stock of net external debt 9.4 10.0 9.1 10.2 13.3 Stock of net domestic debt 10.5 12.1 12.9 9.9 10.5 FISP = Farmer Input Support Programme; FRA = Food Reserve Agency. a. Reflects carryover budgetary releases not included in expenditures. Source: Ministry of Finance and IMF estimates. 4 are exhausted and tax compliance is strength- bond are proposed to finance infrastructure ened. Nonmining revenues could also grow projects (table 1.4). given their fairly low current levels.1 Infrastructure spending allows using Zambia’s stock of public debt is fairly low, the resource windfall for building national allowing plenty of headroom to borrow. At wealth, but Zambia could do better on that the end of 2011 Zambia’s gross public debt score. National net savings, conventionally a was about 28 percent of GDP, 11.6 percent- measure of change in national wealth, needs age points of it external debt.2 Most of the to be adjusted for mineral depletion. As an external debt is concessional and from multi- approximation, we could subtract the deple- lateral institutions. But between July 2011 tion of mineral wealth and add investment in and September 2012 the country borrowed human capital (Ley 2011). On this measure of about $1.25 billion from non­ concessional “genuine” savings, Zambia has not been per- sources alone, 3 including the recent issue of forming as well as its neighbors, Botswana and a $750 million sovereign bond. These bor- Namibia (figure 1.4). In fact, despite the com- rowings are not expected to affect Zambia’s modity price boom, Zambia’s average adjusted debt outlook as long as the proceeds are used savings was negative over 2000–10, indicating for high-return projects. There are some con- a net depletion of “genuine” national wealth. cerns on that count, as discussed later. The ramping up of resources for public Zambia is using its growing resource enve- investment also needs to be matched with the lope mostly to ramp up public investment. selection of high-return projects and efficient As a result, its capital budget has risen from implementation. That would not only support 3.2 percent of GDP in 2010 to 6.1 percent in the current growth trajectory but also develop 2012. The 2013 capital budget is planned at a long-run competitiveness and productive similar level, with the bulk of the resources to capacity. It would also keep the debt outlook be spent on transport and energy infrastruc- on a sustainable long-term path. The 2013 ture. Most of the proceeds of the sovereign Budget Address acknowledges that project Table Proposed use of the sovereign bond proceeds 1.4 Billions of kwacha Millions of U.S. dollars Percent of GDP Energy (generation and transmission) 1,275 255.0 1.2 Transport (road and rail) 2,150 430.0 2.0 Rehabilitation of central hospitals 145 29.0 0.1 Access to finance (SME credit line) 100 20.0 0.1 Fees and transaction costs (actual) 7 1.4 0.0 Discount premium (actual) 73 14.6 0.1 Total 3,750 750.0 3.6 Source: Ministry of Finance 2012. Figure Net savings and “genuine” savings, 2000 –10 1.4 Net savings Genuine savings 40 20 Percent of GNI 0 –20 –40 Botswana Namibia Angola Ghana Sudan Zambia Côte d’Ivoire Guinea Mozambique Source: World Bank 2012a. 5 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n appraisal is weak and promises to “institution- in government accounts. A slower expan- alize a rigorous appraisal system for screening sion would also support the development of investment projects in order to ensure that fiscal buffers to smoothen spending across borrowed funds are only applied to infrastruc- shocks. The last point is gaining importance, ture projects that directly and demonstrably with the share of mining revenues in govern- contribute to the nation’s economic growth” ment finances increasing and thus increasing (Ministry of Finance 2012). The government exposure to copper price changes. Recent should also extend such appraisal to projects revenue trends show that domestic revenues financed from domestic revenues, given the move with copper prices (figure 1.5). opportunity cost of these resources. This is especially relevant for road projects financed External sector largely from domestic resources. For the first time since 2008 the external cur- Zambia’s recent fiscal policy has been rent account is estimated to end in a deficit expansionary. After grants, the country’s of about $674 million in 2012 (figure  1.6). average overall balance over 2010–12 was While declining copper exports were more –3.5 percent of GDP. This policy is sustain- than made up by rising nontraditional able from a debt sustainability point of view. exports, imports grew fast due to the growth However, slower expansion of expenditure in mining investments, intermediate inputs, would allow time to strengthen the country’s and consumer durables. capacity to prepare and appraise projects and Capital inflows to Zambia have remained spend more efficiently. Currently, there are resilient despite global slowdown, with net large unspent balances from budget releases FDI and portfolio investments growing Figure Recent comovement in domestic revenue (2003 prices) and copper prices, 2004 –12 1.5 Percentage change from previous year 100 80 Copper prices 60 40 20 Domestic revenue 0 –20 –40 2004 2005 2006 2007 2008 2009 2010 2011 Source: World Bank staff estimates. Figure International reserves and current account balance, 2003–12 1.6 Current account balance International reserves 1,250 4 Millions of U.S. dollars 750 3 Months of imports 250 2 0 –250 1 –750 0 2003–08 2009 2010 2011 2012 Source: Zambian authorities and IMF estimates. 6 Table Selected balance of payments indicators, 2008 –12 1.5 (millions of U.S. dollars, unless otherwise indicated) 2011 2012 2008 2009 2010 (preliminary) (projected) Current account –1,050 538 1,144 289 –674 Trade balance 404 906 2,704 2,277 1,351 Exports 4,959 4,319 7,414 8,731 9,445 Out of which copper 3,684 3,179 5,768 6,663 6,461 Out of which nontraditional exports 876 900 1,190 1,608 2,379 Imports –4,554 –3,413 –4,710 –6,454 –8,094 Services (net) –615 –465 –629 –808 –825 Income (net) –1,399 –419 –1,363 –1,562 –1,514 Current transfers (net) 560 516 432 382 314 Capital and financial account 1,046 –155 –1,152 –539 1,438 Capital account 230 237 150 151 223 Financial account 816 –392 –1,302 –690 1,215 Out of which FDI and portfolio investments 933 350 708 889 991 Overall balance 13 540 83 208 764 Financing: change in NIR (minus indicates an increase) –13 –540 –83 –208 –764 Memorandum items Current account (percent of GDP) –5.6 6.1 8.0 1.5 –3.4 Gross international reserves 976 1,758 1,897 2,167 2,590 in months of prospective imports cover 2.8 3.7 3.0 2.8 3.0 FDI = foreign direct investment; NIR = net international reserves. Note: Overall balance includes errors and omissions. Source: Zambian authorities and IMF staff estimates. steadily, from $350 million in 2009 to an esti- mineral prices (including that of copper) mated $991 million in 2012 (table 1.5). FDI are expected to remain fairly high over the inflows continue to be directed mainly at medium term. Premised on the expected soft the mining sector, with manufacturing, com- landing of China, copper prices are expected munications, and financial institutions also to recover from $7,900 per metric ton in contributing to recent FDI growth. Despite 2012 to $8,500 in 2013 before falling back the current account deficit, and aided by the to $8,000 in 2014. As a result, the strong FDI sovereign bond, gross international reserves flows to the mining sector of recent years are grew about $282 million in 2012, staying at expected to continue, supporting the expan- about 3 months of prospective imports. The sion of existing mines (Kansanshi, Lumwana, government aims to further increase the and Konkola) and the construction of new reserve coverage, mainly to guard against mines (such as the First Quantum Trident terms of trade shocks. mine and smelter project). Agriculture prospects are good assum- Medium-term growth prospects in Zambia ing that government policy in the sector Medium-term prospects for Zambia’s eco- improves, with the Farmer Input Support nomic growth remain strong but are subject Programme becoming more efficient, which to considerable downside risks emanating is expected to enhance agricultural produc- from global uncertainties. GDP growth is tivity. This will also be helped by scaling up expected to average more than 7  percent extension services, irrigation, and research. over the 2013–14 forecast horizon, making Although the construction of new mines Zambia one of fastest growing Sub-Saha- will require increases in imports, the net ran economies (table  1.6). Underpinning export contribution to GDP is expected to the projected growth over the medium remain positive. Significant FDI inflows and term are favorable external and domestic high copper prices over the forecast horizon developments. Despite the ongoing slow- are also expected to provide support to the down in China’s economy (Zambia’s largest kwacha, thus helping to maintain macroeco- trading partner), international metal and nomic stability. 7 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n Table GDP growth projections, by main sectors, 2012–14 1.6 (constant price = 1994; percentage change from previous year, unless otherwise indicated) 2012 2013 2014 (preliminary) (projected) (projected) Primary sector 4.1 9.2 11.8 Agriculture, forestry, and fishing 7.1 7.3 7.4 Mining and quarrying –0.3 14.3 18.0 Secondary sector 11.2 19.0 18.5 Manufacturing 11.2 6.7 6.1 Electricity, gas, and water 2.3 4.0 4.5 Construction 13.0 12.0 11.0 Tertiary sector 6.8 7.5 7.5 Wholesale and retail 8.9 4.8 4.9 Restaurants, bars, and hotels 2.1 5.0 4.0 Transport, storage, and communications 11.3 10.5 10.6 Financial institutions and insurance 6.0 4.9 6.0 Real estate and business services 2.9 3.0 3.0 GDP 7.3 7.8 8.1 GDP, minus mining 8.0 7.2 7.2 Memorandum item GDP at current market prices (billions of kwacha) 105,256 120,149 136,525 Source: Zambian authorities and World Bank staff estimates. Aggregate demand is expected to con- The expected pickup in the global eco- tinue supporting growth. Fiscal policy is nomic environment, as well as in some of expected to remain expansionary, contrib- Zambia’s major trading partners such as the uting to higher growth prospects. In addi- European Union and South Africa, should tion, recent real wage increases (for both increase demand for Zambia’s exports— the public and private sectors) will support both traditional (such as copper) and non- household consumption, the largest con- traditional (such as sugar and beef)—as tributor to aggregate demand. Household well as tourism services. Demand should be domestic spending should also benefit supported by growth in Sub-Saharan Africa, from a stable macroeconomic environment expected at 6.2  percent in 2013, excluding (average inflation expected at 6–7 percent), South Africa. lower real interest rates, and increased access to consumer credit. Indeed, with ris- Risks to Zambia’s economic growth ing incomes, the increase in personal loans Despite Zambia’s fairly robust growth pros- is expected to continue over 2013–14 (the pects over the forecast horizon, significant share of personal loans in total loans has headwinds could lead to weaker-than-envis- increased from 40 percent in 2010 to 50 per- aged real GDP growth over 2013–14. cent in 2012). Ongoing strong policy inter vention Domestic investment is also expected in Europe (the European Central Bank’s to continue to support growth. The latest bond buying program, plans for a fiscal and Bank of Zambia quarterly survey of business banking union) will keep tensions in global opinion and expectations forecasts that busi- financial markets at bay. But this is not guar- ness confidence will remain high through anteed, as financial markets remain suscep- 2013, with firms investing in construction tible to changing market sentiments. If the and machinery. Indeed, the opening of new Eurozone crisis worsens significantly, shut- mines is expected to have backward linkages ting out some of the large economies from to the services sector, with real estate activ- international capital markets, GDP growth ity around the mining receiving a boost. in Zambia could fall by some 3.3 percentage The multifacility economic zone should help points. And there are risks emanating from boost both domestic and foreign investment other countries. For instance, steeper than to the nonextractive sectors. planned fiscal consolidation in the United 8 States could unravel sentiments in global of consumer prices (inflation) and business financial markets. That could weaken global costs. Although global demand is expected to demand with implications for industrial pro- pick up, the main driver of a potential hike duction, thus reducing demand for industrial in oil prices would be supply side–related metals such as copper. concerns (such as further escalation of ten- Weaker demand from high-income coun- sions in the Middle East). Indeed, if global tries such as China, which accounts for more crude prices were to climb back to the highs than half of Zambian exports, remains perti- of about $130 per barrel in mid-2011 from a nent for Zambia. Indeed, China has already projected $107 per barrel, Zambia’s current shown signs of slowing growth, with GDP account balance and GDP growth could dete- growth falling from 9.2  percent in 2011 to riorate by a further 0.7 and 0.2 percentage 7.7  percent in 2012. Despite these lower points, respectively. growth rates, Chinese demand will remain On the domestic front an uncertain pol- healthy for Zambia’s exports. But if China icy environment could lead to weaker than is unable to land softly, and much weaker expected investments, thus reducing GDP growth rates persist over the forecast hori- growth. The latest Bank of Zambia busi- zon, this could significantly dampen its ness survey noted that some investors have demand for metals, thus leading to a much adopted a wait-and-see attitude and consid- steeper fall in prices including that of copper ered uncertain government policy an imped- and the concomitant decrease in mining sec- iment to continuing investor confidence. tor activity. And significantly weaker copper Another domestic risk is that, given the low prices could lead to a steep depreciation of levels of irrigation, particularly among small- the kwacha, higher imported inflation, and scale farmers, agricultural output is most macroeconomic instability. susceptible to adverse weather conditions, Another commodity-related headwind potentially reducing the sector’s contribu- from the global economy is the risk of soaring tion to GDP growth. Indeed, in the 2011/12 oil prices, given that crude oil (and fertilizer) farming season poor rainfall in some of the accounts for some 15  percent of Zambia’s maize-growing provinces contributed to the import bill and remains an important driver 6.7 percent decline in Zambia’s maize output. 9 Section 2 The state of basic human opportunities for children in Zambia Inequality in Zambia of the rural population between 2006 and Despite Zambia’s robust economic growth 2010, whereas growth in urban areas was over 2000–10, there was very little progress about 2 percent for the first 80 percent and on reducing poverty, particularly in the lat- much higher for the wealthiest 20  percent ter half of the decade—for two main reasons. (figure 2.1). First, the growth—driven by industries such The concentration of economic growth as mining, construction, financial services, in particular sectors and regions over a sus- and tourism—did very little to create new jobs tained period also manifests itself in Zam- and expand economic opportunities beyond bia’s persistently high inequality. The Gini the small part of the country’s labor force coefficient for consumption in Zambia was 52 already employed in these industries. Second, in 2010 and has generally remained above 50 the urban-centered growth also failed to gen- for much of the last decade, making Zambia erate enough spillovers for the roughly two- one of the most unequal countries not only thirds of Zambian people who live in rural in Sub-Saharan Africa but also in the world areas, depend on agriculture, and have seen (figure 2.2). The richest quintile of the popu- their incomes stagnate over the last decade. lation in Zambia accounts for 57 percent of Such growth led to an increasing dispar- consumption, compared with 4  percent for ity between rural and urban areas. Annual the poorest quintile (World Bank 2012c). consumption growth of different wealth Inequality in Zambia also has substan- percentiles was less than 1 percent for most tial geographic dimensions: there are sharp Figure Consumption growth incidence curves for urban and rural areas, 2006 –10 2.1 6 Annual growth rate (percent) 5 Urban 4 3 2 Rural 1 0 –1 1 10 20 30 40 50 60 70 80 90 99 Expenditure percentile Note: Data for 2006 are imputed values and not directly measured values. Source: World Bank 2012c. 11 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n Figure Inequality in Zambia and the world 2.2 Zambia South Africa Sub-Saharan Africa Other countries 80 60 Gini coef cient 40 20 0 5 6 7 8 9 10 11 12 Log of GDP per capita, PPP (constant 2005 international $) Zambia Sub-Saharan Africa Other countries 60 1998 2006 1993 Gini coef cient 50 2004 1996 2003 40 30 5.0 5.5 6.0 6.5 7.0 7.5 8.0 Log of GDP per capita, PPP (constant 2005 international $) Note: The Gini coefficient is a measure of inequality, with a score of 100 indicating perfect inequality, and a score of 0, perfect equality. Source: Word Bank 2011. disparities between urban and rural areas nor pro-poor. 3 The benefits of the agricul- and among the country’s nine provinces. tural input subsidy program are somewhat In fact, inequality between rural and urban progressive but suffer from targeting prob- areas and among provinces accounts for lems: the largest benefits accrue to farmers more than three-quarters of the observed in the middle-income quintiles. inequality in the country (World Bank This section is premised on the notion 2012c).1 This suggests that the rural areas that policies to boost the human capital as well as entire regions may be lagging and development of children and youth promote generally not benefiting from the growth in growth and reduce inequality in the long other parts of the country. run. The link between human capital and Zambia also displays a staggering aggre- economic growth has been well established gate wage inequality. In 2010 the Gini coeffi- in the rich cross-country empirical literature. cient on labor market income was 88 (World For example, Barro (2001) finds growth to Bank 2012c).2 Wage inequality in the labor be positively related to the average years of market is explained partly by gender and school attainment of adult males at the sec- rural or urban location; education contrib- ondary and higher levels. While “quantity” of utes substantially to wage inequality in urban schooling is important, the quality of school- areas. The government’s expenditure policy ing as measured by internationally compa- is not very effective in reducing inequality in rable test scores is even more so—scores on income or access to services. Social transfers science tests have a particularly strong posi- are a very small part of government spending. tive relationship with growth (Barro 2001). The overall incidence of public spending on Several recent studies have shown an impor- education and health is neither progressive tant effect of health on economic growth 12 (see Grimm 2011 for an overview).4 Research From inequality to equity indicates that a healthy population can drive High inequality polarizes political and eco- economic growth and social development but nomic discourse in any country, as the notion is not an automatic by-product of economic of the inequality requiring active policy development. Similarly, at the individual intervention rarely finds consensus. Some level, good health is an important determi- argue that individuals cannot—and indeed nant of economic productivity.5 should not—be rewarded equally irrespec- For inequality, the analysis here focuses tive of their efforts, choices in life (whether on a slightly different variable—the inequality to pursue higher education, for example), of opportunity —than inequality of income or or innate talents. If a person works harder consumption. The notion driving this focus than another, it is only fair that she gets is that the inequality from predetermined suitably compensated; the same for some- circumstances such as gender, ethnicity, one who is more talented than another. But birthplace, or parental wealth in determin- there is much broader agreement that pre- ing people’s economic, social, and political determined circumstances such as gender, success is different from the inequality from ethnicity, birthplace, or family origins should differential efforts, life choices, or innate not determine people’s economic, social, and talents and thus should have a different political success. A person should not have effect on growth in comparison to the latter. fewer opportunities in life just because she is Emerging evidence suggests that the inequal- a girl or just because she is born in a certain ity of opportunity adversely affects growth province. That is the core principle behind and development.6 equality of opportunity, and the framework The analysis delves deeper into how adopted here (box 2.1).7 opportunities for human development dif- Analyzing such opportunities for chil- fer across children born and growing in dren in Zambia can better illuminate the different circumstances. The 2013 Budget nature and causes of inequality of outcomes Speech puts forth a vision for “Delivering observed among adults. Opportunities Inclusive Development and Social Justice,” among children can also be reliable predic- with a focus on expanding opportunities tors of economic mobility across generations for all and ensuring the equitable distri- and over time. If access to economic oppor- bution of the tangible benefits of develop- tunities (such as jobs, credit, land, wages, ment. This analysis puts forth a compelling and financial assets) is correlated with the case for policy to equalize opportunities for circumstances of an individual (such as human development across children in all parental socioeconomic status and location circumstances. of residence), as it appears to be in Zambia, Box The concept of equality of opportunity 2.1 While social scientists and philosophers before the 1970s dealt mostly with the fairness of outcomes, several authors have since delved into fairness of process, equality of resources, and equality of opportunity for welfare. Sen (1979, 2001) has been deeply influential in arguing for an equitable distribution of “capabilities,” which essentially amount to a person’s ability and effort to convert resources into outcomes they can enjoy. Roemer (1998) formalized an equality of opportunity principle, arguing that opportunities should be independent of circumstances and outcomes should depend only on effort and innate ability. Most agree that policy should work to ensure this independence. How does the concept of equality of opportunity translate to measurable objectives for countries? Defining and measuring “opportunities” can be subjective and contextual. But most societies can agree on a basic set of goods and services, such as safe water, adequate sanitation, nutrition, and primary schooling, that conform to a minimalist notion of “opportunities” for citizens. And most societies can agree that equality of opportunities should be a goal to aspire toward when opportunities are defined as access to minimal set of basic goods. The World Bank’s focus on equity is well articulated in World Development Report 2006: Equity and Development (2005), which argues that inequality of opportunity sustains extreme deprivation, results in wasted human potential, and weakens prospects for overall prosperity. Regardless of the choices a society makes about how to universalize opportunities, countries need a systematic way to measure progress toward providing opportunities to all citizens, beyond commonly used measures of overall coverage of goods or services. 13 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n then it reinforces the link between children’s • What circumstances shape the inequal- circumstances and their opportunities in life, ity of opportunity in Zambia, and con- in turn perpetuating economic inequality sequently what are the most vulnerable between groups with different circumstances groups among children? over generations. For policy makers inter- • How has the government’s expenditure ested in closing gaps in outcomes between policy responded to existing inequalities subgroups of the population in Zambia, a of opportunity in education—that is, how comprehensive analysis of these opportuni- progressive are current spending patterns ties can provide a useful blueprint on which from the point of view of opportunities?9 efforts to level the playing field can be based. Opportunities among children are mea- Until recently, framing the debate in such sured here by the HOI, the coverage rate of a terms had not been possible in develop- particular basic service adjusted by how equi- ing countries without an intuitive measure tably the service is distributed among groups of equality of opportunity among children differentiated by circumstances. In discount- (something akin to the Gini) that could be ing inequitable access, the HOI reflects the readily applied to the data available. Tech- extent to which personal circumstances that niques such as the Human Opportunity children cannot be held accountable for Index (HOI) and their application in Latin affect their basic opportunities. So two soci- America and Caribbean over the last decade eties with the same coverage rate for a service have filled some of that gap. This analy- can have different HOIs if people’s access to sis uses some of these techniques to better that opportunity in one is determined to a understand inequality in Zambia. greater extent by their gender, race, family This analysis of opportunities in Zambia background, or other personal circumstance focuses exclusively on opportunities pro- beyond their control and considered an vided to an individual in childhood. For a unjust source of exclusion (box 2.2; Barros child, opportunities are synonymous with and others 2009; Barros, Vega, and Saavedra access to (and use of) basic goods or ser- 2010). vices, such as basic education, health, safe A girl (let’s call her Mwewa) born in a water, and sanitation, while “individual remote village in the province of Luapula to effort” is mostly irrelevant, because the a single, uneducated mother earning 4,000 family, society, or government (and not the kwacha a month and with four other sib- child) are responsible for ensuring whether lings ought to have an equal shot at becom- the child will have access to them. The focus ing a doctor or an engineer as a boy with on children also has implications for public one sibling (let’s call him Melu), born in a policy. Academic research has found inter- two-parent household in Lusaka earning at ventions that equalize opportunities earlier the highest end of the distribution. Mwewa’s in life to be significantly more cost-effective odds should not be shaped before she is even and successful than those attempted later in born, never mind before she has made any life.8 conscientious choice. The HOI provides a unique representation of how equal Mwewa Key questions and Melu’s opportunities are to realize their The analysis poses four key questions on potentials. It also measures progress toward equality of opportunities among Zambian universal coverage of these basic opportu- children: nities and the fairness of allocating access • What is the status of Zambian children’s to those opportunities in a single indicator access to basic opportunities (education, (Barros and others 2009; Barros, Vega, and health, water, sanitation, and so on) in Saavedra 2010). Overall, the HOI brings terms of coverage and distribution among equity to the forefront of policy making with children of different circumstances? an operational measure to track progress. • How has access to these opportunities changed over time in Zambia, and to what Opportunities and circumstances for extent are the changes attributable to Zambian children changes in the scale and equalization of The opportunities considered are those that coverage? enable Zambian children to realize their 14 Box The Human Opportunity Index 2.2 The academic literature offers several ways to measure inequality of opportunity—among them is the Human Opportunity Index (HOI), a scalar measure easily computable from the typical data available in developing countries. Developed by World Bank staff and external researchers (see Barros and others 2009), the HOI is an intuitive measure of a society’s progress toward equitable provision of opportunities for all children. It has been used in two regional reports in Latin America, a study (in draft) for 20 Sub-Saharan countries, and numerous studies conducted (or in progress) for countries around the world. The focus of most studies has been children, with some applications to labor markets. The HOI measures in a single indicator the coverage rate of a particular service, adjusted by how equitably the service is distributed among groups differentiated by circumstance. The construction of the HOI involves aggregating circumstance-specific coverage rates in a scalar measure that increases with overall coverage and decreases with the differences in coverage among groups with different sets of circumstances. The index runs from 0 to 100. A society with universal coverage in a particular opportunity (say, primary school enrollment) would score 100. But a society with an average primary enrollment of 50 percent that is unequally distributed in favor of children of certain circumstances (say, urban children) will have an HOI below 50, with the exact value depending on how unequal enrollment is among children of different circumstances (see annex A). The measure also has a number of desirable and intuitive properties (see annex B). All results are subject to the caveat that the HOI is estimated for a specified list of circumstances, which could change if this list were to change. But the HOI for any opportunity cannot be higher if more circumstances are added to the existing list. Thus if a society wants to measure equality of opportunity with reference to a larger number of circumstances (and groups) than we have considered, the measure of the HOI we provide will serve as an upper bound to the “true” HOI that would take all circumstances into account. To compute the HOI for a certain opportunity for the children of a country, household survey data are essential, with information at the individual (child) and household level. Computing the HOI for a particular opportu- nity when the number of circumstances is large requires an econometric exercise, which involves predicting the HOI from observed access to opportunities and circumstances among children. productive potential directly, by enhanc- child is underweight or stunted. There is ing their human capital, and indirectly, by growing evidence that nutritional depriva- accessing infrastructural amenities that tion in early life has persistent long-term help ensure decent quality of life and facili- effects into adulthood.11 The World Health tate accumulation of human capital (see Organization recommends that children be annex  D). We consider three broad oppor- exclusively breastfed for the first six months tunities: the opportunity to receive adequate of their lives to ensure optimal growth and education, the opportunity to receive a development. And whether a child was breast- healthy start in the first few years of life, and fed exclusively for six months is one of the the opportunity to grow up in a household opportunities relating to a healthy start in under conditions adequate to provide a safe, life. However, breastfeeding alone does not stable, and a stimulating childhood. guarantee children healthy development: evi- For education we consider school atten- dence from across the world shows that most dance for children ages 7–13 and ages 14–16. growth faltering occurs between the ages of Under the education system in Zambia, chil- 6 and 24 months, when children are transi- dren are expected to be enrolled in lower tioning to the family diet and are also most and middle basic education (grades 1–7) vulnerable to diseases. We also use whether between the ages of 7 and 13 and in upper there is at least one bednet in the household basic between the ages of 14 and 16. Since as an essential opportunity for children, the education system is designed for children because bednets are one of the most effective to start formal schooling at age 7, another preventive measures against malaria.12 opportunity considered is to start basic edu- Infrastructure-related services include cation on time. To more reliably measure access to safe water on site, improved the quality of education, data on finishing sanitation, electricity, telecommunica- primary school on time and acquiring basic tions and living in households that are not competency in reading and numeracy by the overcrowded­ — all critical for a safe, stable, time the child is in grade 6 are used.10 and a stimulating childhood. Access to For health we include indicators that cap- safe water, for example, is known to reduce ture whether a child has been fully immu- considerably the risk of waterborne dis- nized by 24 months of age and whether the eases, the leading cause of illness and 15 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n undernourishment in children, in turn children ages 0–16 in the household, whether affecting their education and earning poten- the parents are alive, level of education of the tial.13 Improved sanitation significantly low- household head, gender of the household ers the risk of illness. Access to electricity head, location of the household, and socio- enables nighttime reading, is a healthier economic status of the household measured source of energy than fuel, and opens access by the household’s expenditure. Location to other opportunities, such as sources of is divided into urban and rural on the one information (radio, television, and the Inter- hand and the nine provinces on the other. net). Access to telecommunications enhances the ability to communicate and even the What is the status of Zambian children’s access to critical services such as health access to basic opportunities? care and education. Overcrowding has been It is illustrative to compare Zambian children found to have detrimental effects on child with peers in other developing countries in care, mental health, and social relationships Sub-Saharan Africa and Latin America. How- at home.14 ever, comparing Zambia with countries in The following circumstances are taken other regions is limited because of the sparse into account in calculating the HOI for Zam- coverage of similar analysis outside these two bia: gender of the child, total number of regions. Figures 2.3–2.7 plot the coverage Coverage and the Human Opportunity Index for school attendance of children Figure ages 10 –14 in Zambia and selected Latin American and Sub-Saharan countries 2.3 School attendance (ages 10–14) HOI Coverage 100 75 Percent 50 25 0 Guatemala (2006) Honduras (2006) Nicaragua (2005) Ecuador (2006) El Salvador (2007) Mexico (2008) Colombia (2008) Venezuela, RB (2005) Peru (2008) Costa Rica (2009) Argentina (2008) Brazil (2008) Chile (2006) Nigeria (2008) Mozambique (2003) Congo, Dem. Rep. (2007) Tanzania (2010) Ghana (2008) Malawi (2010) Zimbabwe (2010–11) Namibia (2006–07) Kenya (2008–09) South Africa (2010) Source: Authors’ calculations based on various national household surveys in the Latin American countries and Demographic and Health Surveys for Sub-Saharan countries. Zambia (2010) Coverage and the Human Opportunity Index for finishing primary school on time Figure in Zambia and selected Latin American and Sub-Saharan countries 2.4 Finished primary school on time HOI Coverage 100 75 Percent 50 25 0 Guatemala (2006) Brazil (2008) Nicaragua (2005) El Salvador (2007) Honduras (2006) Costa Rica (2009) Colombia (2008) Venezuela, RB (2005) Peru (2008) Ecuador (2006) Chile (2006) Argentina (2008) Mexico (2008) Mozambique (2003) Congo, Dem. Rep. (2007) Malawi (2010) Ghana (2008) Tanzania (2010) Nigeria (2008) Kenya (2008–09) South Africa (2010) Namibia (2006–07) Zimbabwe (2010–11) Zambia (2010) Source: Authors’ calculations based on various national household surveys in the Latin American countries and Demographic and Health Surveys for Sub-Saharan countries. 16 2.7 2.6 2.5 Percent Percent Percent 0 25 50 75 100 0 25 50 75 100 0 25 50 75 100 Nicaragua (2005) El Salvador (2007) Nicaragua (2005) Honduras (2006) Guatemala (2006) Honduras (2006) Guatemala (2006) Honduras (2006) El Salvador (2007) Peru (2008) Nicaragua (2005) Peru (2008) El Salvador (2007) Ecuador (2006) Colombia (2008) Access to electricity Ecuador (2006) Peru (2008) Guatemala (2006) Brazil (2008) Argentina (2008) Ecuador (2006) Mexico (2008) Mexico (2008) Mexico (2008) Access to safe water on site Access to improved sanitation Venezuela, RB (2005) Colombia (2008) Brazil (2008) Costa Rica (2009) Brazil (2008) Venezuela, RB (2005) Chile (2006) Venezuela, RB (2005) Chile (2006) Argentina (2008) Chile (2006) Costa Rica (2009) Colombia (2008) Costa Rica (2009) Argentina (2008) Malawi (2010) Malawi (2010) Nigeria (2008) Mozambique (2003) Tanzania (2010) Malawi (2010) Tanzania (2010) Mozambique (2003) Mozambique (2003) Figure and selected Latin American and Sub-Saharan countries Kenya (2008–09) Congo, Dem. Rep. (2007) Tanzania (2010) Congo, Dem. Rep. (2007) Kenya (2008-09) Congo, Dem. Rep. (2007) Zimbabwe (2010–11) Ghana (2008) Figure Zambia and selected Latin American and Sub-Saharan countries Ghana (2008) Figure Zambia and selected Latin American and Sub-Saharan countries Namibia (2006–07) Nigeria (2008) Kenya (2008–09) Nigeria (2008) Zimbabwe (2010–11) Zimbabwe (2010–11) HOI HOI HOI Ghana (2008) Namibia (2006–07) South Africa (2010) South Africa (2010) South Africa (2010) Namibia (2006–07) Coverage and the Human Opportunity Index for access to safe water on site in Coverage and the Human Opportunity Index for access to electricity in Zambia Zambia (2010) Zambia (2010) Zambia (2010) Coverage Coverage Coverage Coverage and the Human Opportunity Index for access to improved sanitation in Source: Authors’ calculations based on various national household surveys in the Latin American countries and Demographic and Health Surveys for Sub-Saharan countries. Source: Authors’ calculations based on various national household surveys in the Latin American countries and Demographic and Health Surveys for Sub-Saharan countries. Source: Authors’ calculations based on various national household surveys in the Latin American countries and Demographic and Health Surveys for Sub-Saharan countries. 17 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n (distance of dots from the axis) and the HOI water on site (see figure 2.5), Namibia, South (height of the bars) for five opportunities: Africa, and Zimbabwe in access to improved school attendance of children ages 10–14, sanitation (see figure  2.6), and Ghana, whether children finished primary school Namibia, Nigeria, South Africa, and Zimba- on time, access to safe water on site, access to bwe in access to electricity (see figure 2.7). improved sanitation, and access to electricity. Note that, for the five opportunities for Note that the HOI is essentially the coverage the international comparisons, only the low- adjusted for the inequality in coverage based est common set of circumstances (gender, on the circumstances. In other words, the family structure, socioeconomic status of higher the inequality—or the dissimilarity parents, and location) was used to ensure index (box 2.3)—the higher the penalty due comparability. We can now analyze Zam- to this inequality and thus for any given cov- bia more closely. Below, we use not only the erage rate the higher the gap between cov- broader set of circumstances (see table E2 in erage and the HOI. So the distance between annex E) but also an expanded list of oppor- the dot (coverage) and the bar (HOI) mea- tunities (figure 2.8). sures the underlying inequality. This analysis reinforces what was already Among the five opportunities, Zam- apparent through international compari- bia does best in the opportunity related to sons: the state of human opportunities for school attendance for children ages 10–14 Zambian children is fairly dire. Of the 20 (see figure 2.3).15 The coverage and the HOI opportunities analyzed here, roughly three- are above 85 percent. But at best this is on quarters have an HOI below 50 percent. par with some of the poorest Latin American School attendance—an opportunity that countries. Even among the African econo- appears to be fairly well provided through- mies included in the comparison, Zambia out Africa—falls well short of universal in does better than Nigeria and is comparable Zambia. Note, however, that the inequality with Ghana but has a lower HOI than Kenya, of opportunity based on circumstances for Namibia, South Africa, and Zimbabwe. these indicators is fairly low, suggesting that For finishing primary school on time the low coverage and not the inequality of Zambia appears to do better than many Afri- coverage may be the driver of the low HOI. can countries but again lags behind Kenya, The same is not true for other measures of Namibia, South Africa, and Zimbabwe (see education. The average gap between cover- figure 2.4). It is significantly behind most age and the HOI for the remaining opportu- Latin American countries. nities in education—starting primary school Two things become apparent when look- on time, finishing primary school on time, ing at opportunities for access to safe water and obtaining basic competency in reading on site, improved sanitation, and electricity. and numeracy—is roughly 8 percent and the First, the gap between coverage and the HOI average D-Index is close to 15 percent, sug- generally widens not just for Zambia but also gesting a fair degree of correlation between for most other countries, suggesting greater access to these opportunities and individual inequalities in access to these opportunities child circumstances. based on circumstances. Second, most cho- The gap between the HOIs for basic sen African comparator countries lag consid- enrollment and other measures of education erably behind most Latin American countries quality reflects perhaps a deeper malaise in in provision of these opportunities. Even the Zambian public school system. A 2007 within Africa, Zambia appears to be behind Public Expenditure Tracking Study found Namibia and South Africa in access to safe that 20 percent of teachers in Zambia were Box The dissimilarity index 2.3 The dissimilarity index (D-Index) is the measure of inequality of opportunity used in this analysis. Intuitively, the D-Index shows the share of available opportunities that needs to be “reallocated” across circumstance groups in order to achieve equality of opportunity for a given coverage rate. A D-Index above 5 percent is usually considered problematic for policy-making purposes. The D-Index (D ) is also related to the human opportunity index (HOI ) and the coverage (C ) in the following manner: HOI = C (1–D) 18 Coverage and the Human Opportunity Index for some key opportunities in Figure Zambia, 2010 2.8 HOI, 2010 Coverage 100 75 Percent 50 25 0 School attendance (ages 7–13) School attendance (ages 14–16) Start primary school on time Finish primary school on time Full immunization Exclusive breastfeeding No stunting Access to improved sanitation Access to waste disposal services Access to telecommunications Has a television No underweight No overcrowding Basic reading skills or higher Basic numeracy skills or higher Has a mosquito net Access to safe drinking water Access to a near source of water Access to electricity Has a refrigerator Source: Authors’ calculations based on Central Statistical Office (2010). absent on any given school day at the time of water, sanitation, electricity, and telecommu- announced school visits. In other countries nications. The access to these opportunities unannounced school visits—where absentee- is extremely low, and inequalities extremely ism would be expected to be higher—reveal high, as evidenced by a D-Index that var- absence rates from 11 percent in Peru to ies between 15 percent (access to safe water 27 percent in Uganda (World Bank 2012b). on site) to 69 percent (access to electricity). This suggests that Zambia is among countries Only half of Zambian children grow up close with the highest rates of teacher absenteeism. to safe drinking water—defined as having Teacher input is not the only input into the a protected well, a borehole, rainwater, pro- quality of education, but it is an important tected spring, or piped water within a kilome- one. ter of the home—with almost a fifth of these Opportunities for health and nutrition opportunities inequitably allocated (D-Index appear to be universally low in Zambia and of 19.2 percent). Access to electricity is simi- have a high degree of inequality of access larly limited, with only 13 percent of children based on circumstances that children are living in households that use electricity at all born into. The HOI for exclusive breastfeed- for lighting or cooking. And a high D-Index ing and full immunization are 53 and 50 of 69 percent suggests that the access to this percent, respectively. Although the under- opportunity is heavily inf luenced by the weight indicator is fairly high in level of out- child’s circumstances, resulting in a low HOI come, only 48 percent of Zambian children of 4.2 percent. escape chronic malnutrition or stunting. The HOI for mosquito nets is similarly high at 75 How has Zambian children’s access to percent but far from universal, suggesting opportunities changed over time? that many children live in households that The HOIs for Zambia are well short of uni- are systematically unable to protect their versal coverage in many cases and generally young ones against malaria, at least by ade- on par or behind even the poorest Latin quately providing mosquito nets within the American countries, a region marked by household. The D-Index for opportunities high overall inequality. But has been any for health ranges from 2.1 percent for the progress in the HOIs over time? Comparing underweight indicator to 8.4 percent for full the HOI in 2006 with that in 2010, based on immunization, suggesting again a fairly high two Living Conditions Monitoring Surveys degree of circumstance-dependent access. in these years, shows limited progress in The situation is far worse when we consider Zambia. However, there has been astonish- opportunities for access to infrastructure and ing progress in the opportunity related to basic household amenities such as drinking access to telecommunications: the HOI rose 19 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n more than 6 percent a year over 2006–10. telecommunications, an appreciable gain There has also been decent progress on start- was also driven by the equalization effect, ing and finishing primary school on time as with opportunities improving more than well as on access to safe water on site with proportionately for the weaker circumstance the HOIs increasing annually by 2.1, 1.1, 1.9, groups. The composition effect was small and 1.8 percent, respectively. But progress in most cases, highlighting the slow-moving on basic opportunities for school enrollment changes in the relative circumstances of has stalled. A similarly slow pace of growth children. appears to have afflicted the opportunity to How does Zambia’s progress compare escape childhood without being physically with that in other countries in the region? stunted by poor nutrition. Opportunities for Figure  2.10 compares the average annual full immunization and access to electricity percentage point change in the HOI for six have actually been reversed, with the HOI opportunities in Zambia with 15 other Sub- declining at 1.0 percent and 0.1 percent a Saharan countries.16 Despite seemingly low year, respectively. rate of progress, Zambia comes out in the top Changes in the HOI can be decomposed half of the Sub-Saharan countries included into a composition effect, scale effect, and in the analysis on the expansion of educa- equalization effect. A composition effect tion-related opportunities. On child stunting refers to the change in the distribution of cir- the sluggish growth of roughly 0.4 percent- cumstances that can result from broad demo- age points a year still places Zambia sixth graphic changes, economic growth, or social among the comparator countries. progress. A scale effect refers to proportional Zambia appears to be doing poorly rela- or parallel changes in coverage rates for all tive to other Sub-Saharan countries on groups, perhaps as a result of broad-based access to basic sanitation, full immunization, public policy. And an equalization effect sig- and access to electricity, ranking 10th, 15th, nifies a change in the coverage rate for the and 16th. On full immunization the provi- vulnerable groups for a given overall cover- sion of the opportunity actually appears to age, indicating the equity trend in society, be shrinking at roughly a percentage point perhaps as a result of targeted policies. a year. Similarly, access to electricity—an For the largest changes in the HOI opportunity measured by whether house- for Zambia, the gains were driven by the holds are able to use electricity for cooking scale effect (figure  2.9). But for opportu- or lighting—appears to be declining, if at a nities of improved water and particularly slower pace. Progress on access to improved Figure Change in the Human Opportunity Index and decomposition of changes, 2006 –10 2.9 Equalization Scale Composition Total change 8 Percentage point change 6 4 2 0 –2 School attendance (ages 7–13) School attendance (ages 14–16) Start primary school on time Finish primary school on time Full immunization Access to waste disposal services Access to telecommunications No stunting Has a television Access to safe drinking water Access to a near source of water Access to electricity Has a refrigerator Source: Authors’ calculations based on Central Statistical Office (2006, 2010). 20 Figure Average annual change in the Human Opportunity Index for Sub-Saharan countries 2.10 School attendance (ages 10–14) Equalization Scale Composition Total change 4 Percentage point change 3 2 1 0 –1 Nigeria Zimbabwe Kenya Malawi Namibia Ghana Tanzania Cameroon Zambia Uganda Mozambique Mali Rwanda Ethiopia Niger Madagascar Finish primary school on time Equalization Scale Composition Total change 4 Percentage point change 3 2 1 0 –1 Nigeria Ghana Uganda Rwanda Mozambique Mali Kenya Cameroon Zambia Ethiopia Malawi Zimbabwe Tanzania Namibia Niger Madagascar No stunting Equalization Scale Composition Total change 5 4 Percentage point change 3 2 1 0 –1 –2 –3 Namibia Zimbabwe Rwanda Cameroon Mali Ghana Kenya Mozambique Tanzania Zambia Malawi Uganda Ethiopia Nigeria Niger Madagascar (continued) sanitation, however, appears extremely slow, to each of the circumstances. For example, with rates that imply a percentage point for the opportunity to have the basic numer- improvement every three years. acy skills or higher by the time the child is assessed in the sixth grade, the overall What circumstances shape inequality of D-Index is 14 percent of which almost 42 per- opportunity? cent is explained by socioeconomic status­— Figure 2.11 decomposes the contribution of measured by the household’s per capita each circumstance to the overall D-Index. expenditure. Together with urban-rural loca- The connected blue dots correspond to the tion, region, and education of the household total D-Index for each opportunity and the head, the household’s socioeconomic sta- colored bars denote the shares attributable tus explains roughly 80 percent of the total 21 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n Figure Average annual change in the Human Opportunity Index for Sub-Saharan countries 2.10 Full immunization Equalization Scale Composition Total change 5 4 Percentage point change 3 2 1 0 –1 –2 –3 Zimbabwe Zambia Uganda Namibia Tanzania Nigeria Kenya Ethiopia Rwanda Malawi Mali Ghana Cameroon Mozambique Niger Madagascar Access to improved sanitation Equalization Scale Composition Total change 4 Percentage point change 3 2 1 0 –1 Nigeria Tanzania Kenya Ghana Rwanda Namibia Zambia Uganda Cameroon Malawi Zimbabwe Mali Senegal Mozambique Ethiopia Niger Madagascar Access to electricity Equalization Scale Composition Total change 2 Percentage point change 1 0 –1 Zambia Uganda Mozambique Tanzania Nigeria Rwanda Malawi Cameroon Kenya Ethiopia Zimbabwe Mali Namibia Senegal Ghana Madagascar Niger Note: The surveys are for different years across countries. The average years are 1997 and 2007 and the average period between two survey years for any one country is nine years. For Zambia, the survey years are 1996 and 2007. Average annual change in HOI (in percentage points) is considered to make the changes comparable across countries and independent of the base period HOI. Source: Authors’ calculations based on Demographic and Health Surveys for African countries. Source: Authors’ calculations based on Demographic and Health Surveys. inequality in the opportunity related to basic and the province they live in appear to be numeracy skills. Note also that the gender of some of the strongest drivers of inequal- the child emerges as a nontrivial contributor ity across a broad range of opportunities in of the inequality—particularly for education Zambia. The role of regions is particularly opportunities of finishing primary school on prominent for health- and infrastructure- time and attaining basic reading and numer- related amenities such as exclusive breast- acy skills. feeding, full immunization, stunting, access The socioeconomic status of households, to safe water on site, and access to improved whether they reside in rural or urban areas, sanitation. Urban-rural residence explains 22 Figure Contribution of circumstances to overall inequality of opportunity, 2010 2.11 Socioeconomic status Urban-rural Region Gender of the household head Education of the household head Orphan status Household composition Gender D-Index 100 Marginal contributions (percent) 75 50 25 0 School attendance (7–13 years) School attendance (14–16 years) Start primary school on time Finish primary school on time Basic reading skills or higher Basic numeracy skills or higher Exclusive breastfeeding Full immunization No underweight No stunting Has a mosquito net Access to improved drinking water Access to a near source of water Access to improved sanitation Access to electricity No living in overcrowding Access to waste disposal services Access to telecommunications Has a refrigerator Has a television Source: Authors’ calculations based on Central Statistical Office (2010). the largest share of inequality in the access to finish primary school on time (probabilities improved sanitation, expected with amenities of completion in the bottom 20 percent), such as flush toilets being urban phenomena. almost all live in rural areas; about 70  per- However, the fact that other circumstances cent are males; only 10  percent are led by like education of the household head, region heads with educational attainments beyond of residence, and household’s socioeconomic secondary level; only 1 percent are in a house- status also contribute to the inequality sug- hold in the top expenditure quintile in the gests that there may be significant inequali- country. Similar patterns emerge for mostly ties within urban and rural areas as well. all opportunities presented here: children Finally, the methodology underlying the living in urban areas, with household heads analysis so far offers another way of looking educated beyond the basic primary level, and at the importance of circumstances on the with expenditure levels placing them in the inequality of opportunities in each of these higher echelons of Zambia’s socioeconomic opportunities. For every opportunity a cir- fabric appear to have better probabilities of cumstance-based profile of vulnerability for accessing a broad range of opportunities.17 children can be constructed based on their predicted probabilities of accessing particu- Geography of exclusion lar opportunities. These profiles allow us to Inequality in Zambia has sharp regional identify who the underserved are, what their dimensions. There are significant disparities characteristics are, and how this compares between provinces such as Luapula (relatively with those who have better than average isolated and lagging behind the rest of the access. country) and Lusaka (benefiting from rapid Figure 2.12 presents a snapshot of the vul- growth in construction, transportation, and nerability profiles for selected opportunities. various service sectors as well as the over- For each opportunity the comparisons are whelming presence of the public sector). The between children in the lowest quintile (bot- difference in poverty rates between these two tom 20 percent) based on their chances of provinces is one way of summarizing the dis- accessing a service and those in the highest parity: about 80 percent of the population quintile (top 20 percent). in Luapula is poor, compared with 34.3 per- Take the opportunity of finishing primary cent in Lusaka. The north-eastern provinces school on time. Of children least likely to together with the western province are the 23 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n Figure A snapshot of the vulnerability profile 2.12 School attendance (ages 7–13) No stunting Bottom quintile Top quintile Bottom quintile Top quintile 100 75 Percent 50 25 0 Lives in Is male Household Household Lives in Is male Household Household urban areas head in the top urban areas head in the top educated expenditure educated expenditure beyond quintile beyond quintile secondary secondary Finish primary school on time Access to safe water on site Bottom quintile Top quintile Bottom quintile Top quintile 100 75 Percent 50 25 0 Lives in Is male Household Household Lives in Is male Household Household urban areas head in the top urban areas head in the top educated expenditure educated expenditure beyond quintile beyond quintile secondary secondary Full immunization Access to electricity Bottom quintile Top quintile Bottom quintile Top quintile 100 75 Percent 50 25 0 Lives in Is male Household Household Lives in Is male Household Household urban areas head in the top urban areas head in the top educated expenditure educated expenditure beyond quintile beyond quintile secondary secondary Source: Authors’ calculations based on Central Statistical Office (2010). poorest in the country, while the more cen- able to provide some basic opportunities to trally located provinces, particularly Lusaka children who live there? Figure  2.13 juxta­ and Copperbelt, are the better off ones. poses the HOIs for selected opportunities What is the geographic distribution of against the underlying poverty rate for each opportunities in Zambia? Does it broadly of the provinces in Zambia. The darker mimic the concentration of economic activi- shade indicates higher poverty (in the map ties? Or are even the most excluded provinces showing poverty), while darker shades in the 24 Figure Relationship between poverty and human opportunity 2.13 Poverty headcount ratio (percent) Finish primary school on time (HOI) Lower than 51 20–30 51–60 31–40 61–70 41–50 71 or higher Northern 51–60 Northern 72.5 61–70 41.7 71–80 Luapula Luapula 80.2 39.8 Northwestern Copperbelt Northwestern Copperbelt 68.4 39.7 36.6 75.8 Eastern Eastern 74.9 22.8 Central Central 54.7 54.9 Lusaka Lusaka 34.3 67.0 Western Western 74.1 46.8 Southern Southern 66.0 55.9 Access to improved sanitation (HOI) Access to telecommunications (HOI) 0–10 20–30 11–20 31–40 21–30 41–50 31–40 Northern 51–60 Northern 41–50 7 61–70 27 71–80 81–90 Luapula Luapula 7 25 Northwestern Copperbelt Northwestern Copperbelt 5 40 26 66 Eastern Eastern 5 30 Central Central 16 49 Lusaka Lusaka Western 44 Western 81 3 21 Southern Southern 14 44 Note: The map for poverty headcount ratio is obtained from World Bank (2012c). The other charts plot the Human Opportunity Index (HOI) for the respective opportunities calculated for each region. The lighter the shade of the color the lower the poverty rate and the HOI for all the maps. For the HOI the higher the number the better the access to the opportunities. Source: Authors’ calculations based on Central Statistical Office (2010). HOI maps denote better opportunities. It regions are poor in the first place. But the becomes obvious from the maps that oppor- original question can be nuanced a bit to ask tunities are better provided in regions that the following: Are the poorer regions also the are already relatively more affluent. regions with higher inequality of opportuni- Given the close relationship between ties? Or, in other words, do circumstances actual coverage of these opportunities that children have no control over matter and the HOI, that poorer areas have worse more in poorer provinces compared with the opportunities is hardly saying anything relatively better off ones? new, because access to services is commonly Evidence suggests that this is precisely so poorer in less well-off regions in Zambia and for Zambia. Figure  2.14 plots the inequal- many other parts of the world. Some may ity of opportunity rank for the nine prov- even argue that the general exclusion and inces against their respective poverty rank poorer penetration of some of these basic for four opportunities. The results show a services is the very reason some of these strong (rank) association between poverty 25 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n Figure Poverty and inequality of opportunity (rank correlation) 2.14 Finish primary school on time Access to safe water on site 9 Eastern 9 Northern Luapula Inequality of opportunity rank Northwestern Western Western Northern Northwestern 6 6 Southern Copperbelt Luapula Central 3 Central 3 Southern Lusaka Lusaka Copperbelt Eastern 0 0 0 3 6 9 0 3 6 9 Poverty rank Poverty rank Access to improved sanitation Access to telecommunications 9 Western 9 Western Inequality of opportunity rank Northwestern Eastern Northwestern Northern 6 Northern 6 Luapula Southern Eastern Luapula Southern 3 Central 3 Central Copperbelt Copperbelt Lusaka Lusaka 0 0 0 3 6 9 0 3 6 9 Poverty rank Poverty rank Source: The poverty ranks of the provinces are based on 2010 headcount poverty rates obtained from the most recent poverty assessment, while the opportunity ranks are based on authors’ calculations. and inequality of opportunity. The implica- opportunities in the education sector? The tions are stark. First, the juxtaposition of the focus is on the indicators related to school map on regional poverty with several maps enrollment of children ages 9–13 and ages on the distribution of opportunities reveals 14–16. The tool is a slight variant of the ben- that poorer regions have worse opportunities efit incidence analyses—the Opportunities for children. Second, and perhaps the more Benefit Incidence Analysis (box  2.4). If the debilitating finding, the poorer regions are circumstances that children are born into also the regions with higher inequality of already make them likely to have access to a opportunity. Since inequality of opportunity particular opportunity, then to what extent is in many ways a direct measure of the extent are public expenditures relevant to those to which one’s chances in life are determined opportunities redressing or reinforcing these by circumstances one has little control over, likelihoods? An education system backed by this implies that the prospects of economic a progressive expenditure regime would mobility in Zambia are lower precisely in the channel a greater share of the resources to provinces where one might think they are children with circumstances that make them needed most. This in turn has serious conse- least likely to attend school. This analysis quences for the entrenchment of poverty in helps answer the question of how progres- certain regions and propagation of regional sive or regressive the education expendi- inequality countrywide. tures in Zambia are from the perspective of opportunities. Government expenditures: redressing or Zambia spends 3.6  percent of GDP on reinforcing inequality of opportunities? education, low compared with most Sub- How does the Zambian government’s expen- Saharan countries (de Kemp and Ndakala diture policy address existing inequalities of 2012). Public expenditure rose significantly 26 Box What is the Opportunities Benefit Incidence Analysis? 2.4 The Opportunities Benefit Incidence Analysis presents an incidence analysis of public expenditure on any particular government service—such as education—along the distribution of opportunities, instead of the distribution of income or expenditure as typical of benefit incidence analyses. This yields two main advantages over the traditional benefit incidence analyses. First, it facilitates the analysis of the allocation of public resources to education against a concept of vulnerability related directly to education as opposed to being mediated through variables such as income or expenditure. Second, it accommodates analysis along multiple dimensions—all the circumstances that determine the access to education-related opportunities, for example— as opposed to the unidimensional analysis that characterizes traditional benefit incidence analyses. between 2006 and 2009, but in 2010 alloca- universalize school attendance. Yet note that tion to education declined 18 percent due the progressivity is achieved not because to lower domestic and external allocations of larger public transfers toward children to the sector. The largest share of education whose circumstances worsen their chances spending goes to the primary education level. of attending school but because of larger The Opportunities Benefit Incidence Analy- out-of-pocket contributions at the top of the sis requires expenditure data on education distribution. broken down by subsector and administrative The bottom panels in figure 2.15 further data on the number of children enrolled at crystallize this point. For each quintile of each level for all nine provinces. The analysis the probability of attending school, the com- involves comparing three broad categories of parison here is between the share of children education spending per beneficiary: the mon- in that group—always 20 percent—and the etized benefits accruing to children attending share of the total public expenditure accru- public schools at the specific age bracket (the ing to that group. Looking at the figures, “gross unitary public transfer”), the private broadly the distribution appears uniform, out-of-pocket household spending incurred suggesting very little distortion. But it could to put children in school (the “unitary private be argued, especially with enrollment in spending”), and the difference between the basic education far from universal, that the two (the “net unitary benefit”). pattern of spending need not necessarily The top panels of figure  2.15 plot the be equitable to ensure better equity of out- gross and net unitary benefits of Zambian comes. In other words there is perhaps fur- government expenditure on education for ther space for the government to realign children in basic education for children clas- expenditure in education so that a larger sified into five groups based on their likeli- share is directed toward children with cir- hood of attending school at each level. If the cumstances that make their likelihood of relevant metric is the gross unitary benefit, attending lower. then the spending on education for both age categories seems more or less uniform: Summary of main findings and there are no indications that public spend- conclusion ing accrues more to children whose circum- There is broad consensus that granting stances already make them more or less likely access to basic goods and services to every to attend school at this age. But children with individual is fundamental in building a just circumstances that make them more likely to society and fostering economic and social attend also live in households that can afford development. Basic opportunities refer to out-of-pocket expenses to send their chil- goods and services necessary to give children dren to school. So if one looks at net unitary a decent start in life, such as primary school- benefit­— the difference between what the ing, nutrition, safe water, and adequate sani- public spends on any particular quintile and tation. Whether a person is born a boy or a what the households privately spend—if any- girl, in Lusaka or Luapula, or to an educated thing, the public expenditures appear to be and well-off parent or otherwise should not skewed toward children whose circumstances be relevant to reaching his or her full poten- make them less likely to attend schools. This tial: ideally, only the person’s effort, innate is somewhat consistent with the spending talent, choices in life, and, to an extent, sheer pattern of a country that may be looking to luck would be the influencing forces. 27 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n Distribution of gross and net unitary benefits from public education, by Figure quintiles of probability 2.15 Public Households Net (public – households) Total education spending for children ages Total education spending for children ages 7–13, by probability of attending school 14–16, by probability of attending school 300 200 Billions of kwacha 100 0 –100 –200 Bottom Fourth Third Second Top Bottom Fourth Third Second Top quintile quintile quintile quintile quintile quintile quintile quintile quintile quintile (least (most (least (most probable) probable) probable) probable) Share of public spending on education Share of public spending on education for children ages 7–13, by probability of for children ages 14–16, by probability of attending school attending school 25 20 15 Percent 10 5 0 Bottom Fourth Third Second Top Bottom Fourth Third Second Top quintile quintile quintile quintile quintile quintile quintile quintile quintile quintile (least (most (least (most probable) probable) probable) probable) Source: Authors’ calculations based on Central Statistical Office (2010). One of the primary findings here is that The picture is bleaker for opportunities opportunities for children in Zambia vary in health and infrastructure where cover- widely across different types of goods and age is lower and there are large variations in services. But in general none of the opportu- the inequalities based on children’s circum- nities is close to universal in their coverage. stances. Full immunization, exclusive breast- Some, such as school attendance for children feeding until the child is 6 months old, and under age 16, are available to most Zambian growing up through childhood without being children in the age bracket. But still roughly scarred by chronic malnutrition are some 20 percent of children remain excluded. opportunities critical for a healthy start to a Only about half of children start and finish productive life. But roughly half of Zambia’s primary school on time, with sizable inequal- children are deprived of these opportunities. ities attributable to circumstances children Coverage of opportunities on access to clean have no control over. And the quality of edu- drinking water, adequate sanitation, and elec- cation is low. Of Zambian children in grade tricity is also extremely low, with coverage 6 who took the regional standardized tests depending substantially on the socioeconomic designed to monitor educational quality, 44 status of the household in which the child is percent had proficiency below basic reading born and the location of residence. skills, and 67 percent did not possess even In international comparisons Zambia’s the basic numeracy skills. best performance is on school enrollment, 28 but even that is barely on par with some of the are not currently enrolled in Zambia, there poorest Latin American countries. Compari- seems to be more the government could do sons with Sub-Saharan countries are slightly in realigning expenditures. more favorable, but that is perhaps more infor- While the attempt here has been to lay mative about the overall state of opportunities out a comprehensive diagnostic description in the region than about Zambia. Trends show of the distribution of basic opportunities in variable rates of improvements with astound- Zambia, this discussion should ideally also ing progress on the access to telecommunica- speak to what this implies for policy. What tions, modest but commendable progress on kind of sectoral policies in education, health, starting and finishing primary school on time drinking water, and other economic services and having access to drinking water, sluggish can be used to enhance opportunities for improvements in the opportunities for access all children in the country? Are there trad- to improved sanitation and the opportunity to eoffs between policies that aim to improve escape acute malnutrition, and actual rever- population-level access versus policies that sals in the opportunities for full immuniza- target those that are underserved on account tion and access to electricity. of their circumstances? If yes, where should The analysis also reveals household’s policy makers place a greater emphasis? socioeconomic status, urban-rural residence, The analysis demonstrates the HOI to and the province of birth as the strongest be a powerful measure to track a society’s drivers of inequality across broad oppor- progress on equitable distribution of basic tunities. The regional aspect is particu- opportunities. Together with a robust data- larly interesting because the distribution of gathering and monitoring and evaluation opportunities across the regions in Zambia system, it can help improve the targeting closely mirrors the spatial distribution of and efficacy of social policy. Making use of poverty, implying that poorer regions have it, several Latin American countries, includ- lower access to opportunities. In addition, a ing Brazil and Peru, have already confronted close positive relationship is found between their inherent inequalities with well-targeted poverty and inequality of opportunity in social policies, with positive initial results. Zambia­ — the poorest regions in Zambia are If equity is a legitimate social goal, then the also the regions where children’s circum- evidence presented here on the inextricable stances play the strongest role in determin- link between circumstances and opportuni- ing their access to these opportunities. This ties suggests the possibility for similar social depresses prospects of economic mobility policy interventions in Zambia. in the poorer provinces and may result in With the largest driver of unequal access the further entrenchment of poverty and to opportunities in Zambia as the province the propagation of the spatial inequalities of birth, investments in the lagging regions observed in Zambia today. need to be scaled up, particularly for infra- The analysis also covers the role of the structure services such as drinking water, Zambian government, particularly through sanitation, and electricity where access is its expenditure policy in addressing the highly concentrated in certain urban pock- observed inequality of opportunity in the ets within particular provinces. But in addi- education sector. Current patterns of spend- tion to these resources, there may be lessons ing, at least in the education sector, are dis- that provinces can learn from each other. tributed such that they benefit children Consider the Eastern and Western provinces. fairly uniformly, irrespective of their cir- The level of poverty in these regions is com- cumstances. But this seeming progressivity parable, but significantly more children in is achieved not necessarily because larger the Western province are finishing primary shares of public expenditures are being tar- school on time than are children in the East- geted to children with the lower probabilities ern province. But the opposite is true for of having access to these opportunities, but improved access to drinking water, where because the private costs borne by families of children in the Eastern province have much children from the top end of the distribution better access to drinking water. are significantly higher. Given that roughly For many other opportunities, par- a fifth of children who should be in school ticularly those for education and health, 29 Z a m b i a E c o n o m i c b r i e f — Re c e n t E c o n o m i c D e v e l o p m e n t s a n d t h e S t a t e o f B a s i c H u m a n O p p o r t u n i t i e s f o r C h i l d r e n resources are often neither necessary nor and accountability to the end user, criti- enough to scale up the opportunities, cal to ensuring that quality of the services despite Zambia’s spending much less on edu- delivered­— in education or health—is of cation and health than other Sub-Saharan minimum acceptable standards. economies. Realigning expenditures to pri- Finally, policy makers also need to recog- oritize the poor and the underserved and nize that children of certain circumstances ensuring that the amount spent is effective are vulnerable to deprivations in multiple will enhance the desired results even with dimensions simultaneously. For example, the same resource envelope. The govern- Zambian children living in rural areas and ment already allocates the bulk of spending with household heads who did not finish in education on primary school enrollment, primary school are much more likely to not for example. Yet as the opportunity-benefit finish primary school themselves, to suffer incidence analysis shows, there is signifi- from acute malnutrition, or to have poor cant room to target children whose circum- access to sanitation facilities. The presence stances make their likelihoods of having of multiple deprivations points to the need access to this opportunity lower. This has to for policies in different sectors (health and go with necessary tweaks to the institutional education, for example) to closely coordi- setup of service delivery to provide proper nate to achieve better efficiency and the best incentives to the service provider and voice results. 30 Annex A How is the Human Opportunity Index calculated? A simple example Consider two societies A and B, in which half total number of opportunities that need to the population lives in rural areas and the be redistributed to ensure equitable access other half in urban areas. Now consider a based on the equality of opportunity prin- basic opportunity such as access to primary ciple. For society A this will constitute “reallo- education. Say half of children go to school cating” 25 percent of total enrollments from in both the societies. Looking at the overall urban children to rural children, with a pen- coverage, both these societies will appear alty of 25 percent and an HOI, which is the similarly placed. But suppose that in society coverage minus the penalty, of 25 percent. A no rural child attends a school, while in For society B there is no inequality based on society B half of both rural and urban chil- location, and the penalty is zero. This implies dren attend school. The HOI discounts the that the HOI is 50 percent, or equal to the coverage rate of 50 percent by imposing a coverage. Therefore, society B is more equal “penalty” when access is more unequal based than society A based on equality of opportu- on circumstances such as location. The pen- nity, even though average enrollment rate is alty can be interpreted as the share of the the same in both societies. 31 Annex B Three key properties of the Human Opportunity Index There are three key properties of the HOI. circumstances chosen for analysis. But this First, the HOI is sensitive to scale—if access is mitigated by an additional property that is improves for all groups by, say, a factor of highly desirable given that it is often impos- k (additively or multiplicatively), then the sible to identify all relevant circumstances for HOI changes by the same factor k. Second, it any population and opportunity: the HOI rewards Pareto improvement—if coverage rate will not be higher if more circumstances are improves for one circumstance group without added to the existing set of circumstances in decreasing coverage rates for the remaining the analysis. This implies that the computed groups, the HOI will rise. Third, the measure inequality serves as a lower bound to the will always improve if access changes so that “actual” inequality where all circumstances of the more vulnerable groups (groups with interest could be included in the analysis. coverage rates lower than the overall cover- age rate) have higher access. An important Source: Barros, Molinas Vega, caveat: this measure is sensitive to the set of and Saavedra 2010. 32 Annex C Estimating the Human Opportunity Index from household survey data To construct the HOI, we need to obtain the are estimated using the predicted probability conditional probabilities of access to oppor- ˆ and sampling weights (w): p tunities for each child based on their cir- n ˆ ; C = ∑ wi p cumstances. And to do so, one can estimate i =1 i,n  n C ∑ wi |pi,n – C | a logistic model, linear in the parameters D = 21 ˆ  ; i =1 ß, where the event I corresponds to access- P = C * D ; and ing the opportunity (such as access to clean water), and x the set of circumstances (such HOI = C – P. as gender of the child and education and gender of the household head). We fit the An important caveat to the logistic estima- logistic regression using survey data: tion model is that the list of regressors does not include any interaction terms between P [I = 1|X = (x1, …, xm)] m circumstances (such as between parental Ln = ∑ xk ßk . 1 – P [I = 1|X = (x1, …, xm)] k =1 education and location). Given the number of circumstances we have (all are dummy where x k denotes the row vector of vari- variables), limited sample sizes, and the ables representing the k -dimension of large number of countries and opportunities circumstances. Thus, x  =  (x 1,  …,  xm) and for which these regressions have to be run, b’  =  (b1,  …,  bm) is a corresponding column including interactions would lead to intrac- vector of parameters. From the estimation of table problems in at least some of the cases. this logistic regression one obtains estimates The interaction terms are thus omitted, of the parameters {ßk} to be denoted by {ß ˆ }, even though translating the exact definition k,n where n denotes the sample size. Given the of D-Index to the logistic regression model estimated coefficients, one can obtain for would require including these terms. If the each individual in the sample his/her pre- interactions were included, it would result dicted probability of access to the opportu- in a higher D-Index (and a lower HOI), just nity in consideration: as it would happen if more circumstances were added. This in turn implies that the ˆ estimated D-Index for all countries and ˆ i,n = Exp(xi ßn) . p 1 + Exp(xi ß ˆ ) opportunities is the lower bound of inequal- n ity of opportunities (and the estimated The overall coverage rate (C ), the HOI is the upper bound) for a given set of D-Index  (D), the penalty (P ), and the HOI circumstances. Source: Barros, Molinas Vega, and Saavedra 2010. 33 Annex D Shapley Decomposition of the D-Index—An example In country A we want to calculate the contri- difference in the D-Index before and after bution of income to the inequality in access income is added. Finally, we average the mar- to a basic opportunity. The circumstances ginal contributions over all combinations. In considered are the gender of the house- a set with three circumstances (Income, I; hold head, the gender of the child, and the gender of child, G ; and gender of household household income, and the opportunity is head, H ), there are six different sequences defined as having electricity in the house- in which income can enter {(C ,H,I ) (H,C ,I ) hold. The total D-Index is obtained using (C ,I,H ) (H,I,C ) (I,C ,H ) (I,H,C )}. Neverthe- all circumstance variables and equals 3.48 less, since in the regression model two sets percent (table D1). The D-Index using only of covariates with the same circumstances income as a circumstance equals 3.24 per- and different order generate the same result, cent and the index without circumstances there are only four different values for the (only a constant in the logistic regression) marginal contribution of income. In the equals 0. example below income contributes to 63 per- To obtain the marginal addition to the cent of the D-Index. D-Index of income (DI), we estimated the D-Index with all possible sequences of cir- DI = 2/6[D(I,C ,H ) – D(C ,H )] + cumstance variables where income can be 1/ [D(I,C ) – D(C )] + 6 added. In each situation we calculate the 1/ [D(I,H ) – D(H )] + 6 marginal contribution of income as the 2/ [D(I ) – 0 6 Table D-Index based on circumstance set D1 Circumstance set Contribution to the D-Index D-Index Gender head U gender child U income D(gender head U gender child U income) 3.48 Income D(income) 3.24 Combinations of circumstance sets where income is added Income U gender child D(income)–D(constant) 3.24 Income U gender head D(income)–D(constant) 3.24 Gender child U income D(gender child U income)–D(gender child) 2.40 Gender head U income D(gender head U income)–D(gender head 1.29 Gender child U gender head U income D(gender child U gender head U income)–D(gender child U gender head) 1.50 Gender head U gender child U income D(gender head U gender child U income)–D(gender head U gender child) 1.50 Average contribution of income 2.20 Share contribution of income (%) 63 Source: Hoyos and Narayan 2011. 34 Annex E Opportunities and circumstances for Zambian children Table Opportunities and their definitions E1 Opportunities Description Education School attendance (ages 7–13) Currently attending educational institution for children ages 7–13 School attendance (ages 14–16) Currently attending educational institution for children ages 14–16 Start primary school on time Attending grade 1 of primary for children age 7 Finish primary school on time Have reached primary (grade 7) education completed for children age 14 Basic reading skills or higher Classified in the 3 or higher level of competence in reading (SACMEQ) Basic numeracy skills or higher Classified in the 3 or higher level of competence in numeracy (SACMEQ) Health Exclusive breastfeeding Only received breast milk and nothing else, for children below 6 months Full immunization Immunization against polio, BCG, DPT, and measles for children ages 12–23 months No underweight Over –2 in the Z-Score for weight for age. Children ages 3–59 months No stunting Over –2 in the Z-Score for height for age. Children ages 3–59 months Has a mosquito net At least one mosquito net in the household. Children ages 0–16 Housing Access to improved drinking water Access to a source of drinking water such as protected well, borehole, rainwater, protected spring, or piped water. Children ages 0–16 Access to a near source of water Access to an improve source of water in less than 1 kilometer. Children ages 0–16 Access to improved sanitation Flush toilet, or own or communal pit toilet latrine with slab Access to electricity The household uses electricity for lighting or cooking. Children ages 0–16 No living in overcrowding Ratio persons per room is less or equal to 1.5 Access to waste disposal services Garbage is collected Access to telecommunications At least one cellphone (operating) or land phone (operating) in the household. Children ages 0–16 Has a refrigerator At least one refrigerator in the household. Children ages 0–16 Has a television At least one television in the household. Children ages 0–16 Table Circumstances and their definitions E2 Dimension Circumstances Details Gender Gender Household composition Total number of children ages 0–16 in the household Is the father alive? Orphan status Is the mother alive? Education of the household head Level of education of the household head 4 categories (none, primary, secondary, and tertiary) Gender of the household head Gender of the household head Location Urban-rural province Socioeconomic status Expenditures quintiles 35 Notes Section 1 growth to be important and probably 1. Zambia’s tax revenues minus mining aver- exceeding the reverse effect, namely that aged 14.5 percent of GDP since 2009, com- of income on health. But the debate is paring unfavorably with countries at the not fully settled, due to the difficulties same income. There is room for improve- mentioned above. ment in the domestic value-added tax. 5. There are four channels through which 2. Like many other Sub-Saharan countries, health could contribute to an econ- Zambia received debt relief under the omy and ultimately economic growth: Heavily Indebted Poor Countries Ini- enhanced labor productivity, greater tiative and the Multilateral Debt Relief labor supply, education and training fos- Initiative, and its external debt fell sub- tering higher skills, and more savings for stantially from about 86 percent of GDP investment in physical and intellectual in 2005 to about 9 percent in 2006. capital. 3. Nonconcessional sources are those with 6. Molina and others (2012) finds that less than 35 percent grant element. inequality of opportunity—attributable to circumstances a person is born into— Section 2 in education among children negatively 1. This is based on the decomposition of the affects per capita income. Similarly, Mar- Theil measure of inequality, or GE(1). rero and Rodriguez (2010) find a nega- 2. International comparisons are difficult to tive relationship between the component make as estimates for wage inequality (as of income inequality attributable to cir- opposed to aggregate income inequality) cumstances and economic growth. for countries are not widely available. 7. Perhaps most important for the pro- 3. While the primary and secondary educa- posed work are the contributions of John tion spending is progressive, their pro- E. Roemer, whose Equality of Opportu- gressivity is outweighed by the extreme nity (1998) was the first to formalize an concentration of tertiary education ben- equality of opportunity principle. efits among the wealthiest members of 8. See, for example, Chetty and others Zambian society. (2010) for evidence that early childhood 4. Estimating health effects on income is education has substantial long-term difficult due to problems in measuring impacts, ranging from adult earnings to health and the potential endogeneity retirement savings. Child malnutrition of health (Deaton 2006). Bloom, Can- has also been shown to generate life-long ning, and Sevilla (2004), Weil (2007), learning difficulties, poor health, and and Lorentzen, McMillan, and Waczi- lower productivity and earnings over a arg (2008), using different methods, lifetime (Alderman, Hoddinott, and Kin- have shown health effects on income or sey 2006; Hoddinott and others 2008). 36 9. The data used come mainly from the effective against malaria, are 53 percent 2006 and 2010 rounds of the Living Con- and 29 percent. ditions Measurement Survey conducted 13. The major health threat posed by drink- in Zambia. For opportunities related to ing unsafe water is infectious diarrhea­— the quality of education, data are from the leading cause of mortality for the 2007 round of the Southern and children under age 5 and estimated to Eastern Africa Consortium for Monitor- cause 1.5 million deaths a year (Cabral, ing Educational Quality. And, to make Lucas, and Gordon 2009). In India the cross-country comparisons, the report prevalence and duration of diarrhea uses data from the Demographic and among children under age 5 in rural Health Surveys. areas are significantly lower for families 10. Grade progression requires that children with piped water than for those without it learn the content adequately enough. (Jalan and Ravallion 2003). And in Paki- For Zambia we measure this by the stan private behavioral choices and poli- opportunity to have completed grade 7 cies that affect the health and nutrition (lower and middle basic) on time, when of rural children have important effects the child reaches age 14. A more direct on school enrollment and thus on even- measure of learning in schools would be tual productivity (Alderman and oth- through test scores. We use data from ers 2001). Improved nutrition increases the 2007 round of the Southern and enrollments, especially for girls, thus Eastern Africa Consortium for Monitor- closing a portion of the gender gap. ing Educational Quality to define the 14. See, for example, Gove, Hughes, and Gal- opportunity as having acquired basic le’s (1979) study on Chicago households. competency in reading and numeracy by 15. Note that the age group used for cross- the time the child reaches grade 6. country comparison is different from the 11. Children who experience spells of mal- age groups used for various other enroll- nutrition in early childhood are found ment indicators for Zambia—to ensure to have poorer test scores on cognitive comparability. In addition, wherever assessments, activity level, and attention cross-country comparisons are made, span (Alderman, Hoddinott, and Kin- the data used for Zambia are from the sey 2006). They also tend to start school Demographic and Health Surveys. later and are at a greater risk of dropping 16. These countries were the basis of a out before finishing primary school. broader, regionwide analysis of inequal- A recent study in Guatemala finds that ity of opportunities in Africa. They being stunted at age 6 is tantamount to represent a wide variety of countries of losing four grades of schooling based different sizes, geography, and level of on performance in tests (Maluccio and development. others 2009). The accumulated evidence 17. Keep in mind the fraction of Zambia’s on child malnutrition suggests that chil- overall population that would fall into dren’s learning potential in school and each of these categories while making their productivity in later life is predeter- interpretations. For example, for oppor- mined largely by their health and nutri- tunities, such as access to electricity, tional status before age 2. 85 percent of the children in the top 12. In Zambia malaria is the primary cause quintile live in urban areas, reflecting of morbidity and mortality, with 4.3 mil- a sharp rural-urban divide because the lion reported cases and 50,000 deaths urban population in Zambia is only 35 a year. It also accounts for a quarter of percent of the total. For full immuniza- childhood deaths. Note, however, that tion the share of urban children in the owning a bednet is not the same as actu- top quintile is about 34 percent, very ally using it. Indeed, in Zambia, for all similar to the fraction of the population types of bednets, ownership was 64 per- in the urban areas, suggesting that the cent in 2007 while use was only 33 per- coverage of full immunization may not cent. The same figures for insecticide be significantly different between urban treated bednets, known to be more and rural areas of Zambia. 37 References Alderman, Harold, Jere R. Behrman, Vic- Central Statistical Office of Zambia. 2006. tor Lavy, and Rekha Menon. 2001. “Child “Zambia Living Conditions Monitoring Health and School Enrollment: A Lon- Survey 2006.” Lusaka. gitudinal Analysis.” The Journal of Human ———. 2010. “Zambia Living Conditions Resources 36 (1): 185–205. Monitoring Survey 2010.” Lusaka. Alderman, Harold, John Hoddinott, and Bill Cervelatti, Matteo, and Uwe Sunde. 2009. Kinsey. 2006. “Long-term Consequences of “Life Expectancy and Economic Growth: Early Childhood Malnutrition.” Oxford Eco- The Role of the Demographic Transition.” nomic Papers 58 (3): 450–74. IZA Discussion Papers 4160, Institute for Barro, Robert J. 2001. “Human Capital and the Study of Labor, Bonn, Germany. Growth.” American Economic Review 91 (2): Chetty, Raj, John N. Friedman, Nathaniel Hil- 12–17. ger, Emmanuel Saez, Diane W. Schanzen- de Barros, Ricardo, Francisco H. Ferreira, bach, and Danny Yagan. 2010. “How Does Jose R. Molinas Vega, and Jaime Saavedra. Your Kindergarten Classroom Affect Your 2009. Measuring Inequality of Opportunities in Earnings? Evidence from Project STAR.” Latin American and the Caribbean. Washing- NBER Working Paper 16381, National ton, DC: World Bank. Bureau of Economic Research, Cam- de Barros, Ricardo, Jose R. Molinas Vega, and bridge, MA. Jaime Saavedra. 2010. “Measuring Progress Deaton, Angus. 2006. “Global Patterns of toward Basic Opportunities for All.” Brazil- Income and Health: Facts, Interpretations, ian Review of Econometrics, 30 (2): 335–67. and Policies.” NBER Working Paper 12735, Bloom, David E., David Canning, and Jaypee National Bureau of Economic Research, Sevilla. 2004. “The Effect of Health on Cambridge, MA. Economic Growth: A Production Function Gove, Walter R., Michael Hughes, and Omer Approach.” World Development 31 (1): 1–13. R. Galle. 1979. “Overcrowding in the Bloom, David E., David Canning, and Gün- Home: An Empirical Investigation of its ther Fink. 2009. “Disease and Develop- Possible Pathological Consequences.” ment Revisited.” NBER Working Paper American Sociological Review 44 (February): 15137, National Bureau of Economic 59–80. Research, Cambridge, MA. Grimm, Michael. 2011. “Does Inequality in Cabral, Christie, Patricia Lucas, and Dave Gor- Health Impede Economic Growth?” Oxford don. 2009. “Estimating the Health Impacts Economic Papers 63 (3): 448–74. of Unsafe Drinking Water in Developing Hoddinott, John, John A. Maluccio, Jere R. Country Contexts.” Aquatest Working Behrman, Rafael Flores, and Reynaldo Paper 01, University of Bristol, Water and Martorell. 2008. “The Impact of Nutri- Health Research Centre, Aquatest 2 Proj- tion during Early Childhood on Income, ect, Bristol, UK. Hours Worked, and Wages of Guatemalan 38 Adults.” The Lancet 371 (February): the Study of Economic Inequality, Palma, 411–16. Spain. Hoyos, Alejandro, and Ambar Narayan. 2011. Ministry of Finance. 2012. “2013 Budget “Inequality of Opportunities among Chil- Address.” Delivered to the National Assem- dren: How Much Does Gender Matter?” bly of Zambia, October 12. Background Paper for World Development Molina, Ezequiel, Ambar Narayan, and Jaime Report 2012, World Bank, Washington, DC. Saavedra. 2012. “Opportunity and Devel- Jalan, Jyotsna, and Martin Ravallion. 2003. opment: Assessing the Effect of Equality “Does Piped Water Reduce Diarrhea for of Opportunities in Development Out- Children in Rural India?” Journal of Econo- comes.” World Bank, Washington, DC. metrics 112 (1): 153–73. Roemer, John E. 1998. Equality of Opportunity. de Kemp, Antonie, and Charles Ndakala. Cambridge, MA: Harvard University Press. 2012. Unfinished Business: Making a Differ- Sen, Amartya. 1979. “Utilitarianism and Wel- ence in Basic Education. 2012. The Hague, farism.” The Journal of Philosophy 76 (9): the Netherlands: Ministry of Foreign 464–89. Affairs of the Netherlands. www.minbuza. ———. 2001. “Development as Freedom.” nl/binaries/content/assets/minbuza/nl/ Oxford, UK: Oxford University Press. services/evaluatie/making-a-difference-in- Weil, David N. 2007. “Accounting for the Effect basic-education-in-zambia.pdf. of Health on Economic Growth.” Quarterly Ley, Eduardo. 2011. “Exhaustible Resources Journal of Economics 122 (3): 1265–1306. and Fiscal Policy: Copper Mining in Zam- World Bank. 2005. World Development Report bia.” World Bank, Washington, DC. 2006: Equity and Development. Washington, Lorentzen, Peter, John McMillan, and DC. Romain Wacziarg. 2008. “Death and Devel- ———. 2011. Word Development Indicators 2011. opment.” Journal of Economic Growth 13 (2): Washington, DC. 81–124, June. ———. 2012a. Africa’s Pulse: An Analysis of Maluccio, John A., John Hoddinott, Jere R. Issues Shaping Africa’s Economic Future. Vol- Behrman, Reynaldo Martorell, Agnes R. ume 6, October. Washington, DC. Quisumbing, and Aryeh D. Stein. 2009. ———. 2012b. “Zambia Human Development “The Impact of Improving Nutrition dur- Note: Challenges and Opportunities.” ing Early Childhood on Education among World Bank, Africa Region Human Devel- Guatemalan Adults.” Economic Journal 119 opment Network, Washington, DC. (537): 734–63. ———. 2012c. “Zambia Poverty Assessment: Marrero, Gustavo A., and Juan G. Rodriguez. Stagnant Poverty and Inequality in a Natu- 2010. “Inequality of Opportunity and ral Resource-Based Economy.” June, Wash- Growth.” Working Paper 154, Society for ington, DC. 39