QO~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~IU > - dow Ei','.'),,ii I ~~~o - . . - .s. ': ;,:0D. ~~mE-4:. cfs :~~'E- IW 4 i : pf) : Xs ages, salaries, through falling wages and the movement of and profits from business activities. In addition, workers into low-paying informal sector jobs changes in interest rates, stock market crashes, falling (Manuelyan Atinc and Walton 1998). real estate prices, and inflation all affect the relative . Cuts in public transfers and serv.ices especially if returns of various types of assets. While the poor do they take place across the board or are not well not generally hold assets, these effects need to be ex- thought-out, may harm those who rely most on amined before ruling out any impact on the poor. public services, usually the poor. Public expenditure cuts, beyond causing declines in * In Chile per capita social spending fell 20 percent labor demand and price effects (for example, when between 1981 and 1986. The poorest 40 percent subsidies are removed), affect cash transfers and in- of families were the hardest hit: their share of per- kind services for the poor. And shocks (such as an sonal income was only 12 percent but they re- exchange rate depreciation or commodity price ceived 50 percent of public spending in health changes) or policies (such as trade reforms, public and education and 20 percent of social security sector price increases, changes in taxes or subsi- payments (Bourguignon and Morrisson 1992). dies) affect relative prices. For example, an exchange * Relative price changes have different impacts on rate devaluation raises the prices of goods that are households depending on their production and traded relative to those of goods that are not. consumption patterns. Inflation erodes the pur- Evidence from several countries indicates that chasing power of wages and transfers that are these changes affect different households in differ- fixed in nominal terms and reduces the relative ent ways: value of fixed-denomination assets. * labor demand shocks have different impacts on * In East Asia the prices of imported goods in- workers with different skills and different levels creased dramatically following exchange rate de- of job security. Low-skill workers and women are valuations. Price increases for food items hurt more likely to lose formal-sector jobs and move households that are net consumers of food- into the informal sector, where earnings decline. especially the urban poor and poor farmers * In both Brazil and Mexico informal sector em- whose food stocks and production were down ployment swelled and earnings declined dur- because of the El Nino-related drought. ing the recessions of the early 1980s. In Brazil . For Brazil Ferreira and Litchfield (1998) suggest employment in the informal sector increased by that inflation appears to be the primary macro- 30 percent during 1981-83, while the ratio of economic force driving the observed increase informal to formal sector wages decreased by in inequality. Regression results show that in- 7 percent. In Mexico informal sector employ- flation reduces the income share of the poor- ment grew 9.5 percent a year between 1983 and est 80 percent of the population and increases 30 POVERTY REDUCTION AND THE WORLI) BANK: PROGRESS IN FISCAL 1998 the share of the top 20 percent, which is exactly of the mid-1980s crisis on poverty were sharp and the change observed between 1981 and 1990. not quickly reversed when growth resumed because Blejer and Guerrero (1990) reach the same con- inequality had also increased. In Mexico the acqui- clusion for the Philippines for the period sition of education was dampened both by lower en- 1980-86.2 rollments of first entrants in primary school and by Households respond to these shocks in various fewer people continuing on to higher levels of school- ways. Consumption and production activities change. ing. This may explain why average years of school- Employment patterns change: children are pulled out ing of household heads were lower in the mid-1990s of school and sent to work, women turn to informal than in 1984 for all deciles up to the seventh and why sector jobs, and individuals and families migrate. inequalities in earnings persisted (Lustig 1995). Savings and other assets are tapped, as are forms of There is also evidence from several countries assistance available through family and communities. that social capital can be damaged during crises. Re- Whole communities as well as households feel the cent tensions and riots in Indonesia are but one ex- impact, as traditional networks and other forms of so- ample. In a different context, there is evidence from cial capital are disrupted by economic hardship, mi- participatory poverty assessments that economic gration, and the emergence of fault lines between hardship has severely eroded family and community ethnic, religious, or racial groups. ties in countries of the former Soviet Union (see Some of these responses have only short-term chapter 3). effects on well-being, but others may have long- term impacts or affect well-being irrevocably. Price Protecting the poor in a crisis effects and the loss of access to public health and ed- ucation may severely limit the ability of the poor to The impacts of a crisis on the poor can be mitigated accumulate human capital, with irreversible effects through appropriate policies. Policies may limit the on long-term productivity. Reducing the nutritional impact on the poor by preventing or dampening intake of babies or pulling children out of school may changes in the key variables that affect household in- have permanent effects on their future learning and come-labor earnings, returns to assets, public ser- income-earning abilities. Malnourishe(d mothers have vices, and prices-or by preventing behavioral low-birthweight children, who in turn are more likely responses by households that may have long-term to suffer from poor health. Distress sales of produc- negative impacts. What is essential is that a concern tive assets may affect the ability of households to re- for the poor be taken into account in the design of sume productive activities after the crisis. Responses policy responses right from the start-from the basic such as these are particularly worrisome because of elements of a stabilization program-and not be rel- their long-term effects, which may well reinforce egated to micro-level interventions set up well after existing inequalities and create persistent poverty the main reform program. (see Ferreira 1995). An agenda for a poverty-focused response to a Evidence from rapid social assessments conducted crisis should include the following actions:4 in Cambodia, Indonesia, the Philippines, and Thai- * Choose macroeconomic stabilization policies that land and surveys in Indonesia, Korea, and Thailand achieve their objectives at the least short- and long- confirm that the poor are indeed responding in ways term cost to the most vulnerable. Macroeconomic that will affect their long-term well-being. Youngsters policy has an important role to play in reigniting are being taken out of secondary school and sent to growth and thereby reducing poverty. If policy is work; health care is delayed or forgone, with in- appropriately countercyclical, it can offset (to creased risks of epidemics of communicable dis- some degree) the worst effects of recession on the eases; malnutrition is spreading; and productive poor. Contractions in aggregate demand, even if assets are being sold (Robb 1998; SMERU 1999).3 harming all equally, are particularly bad for the There is also evidence of reverse migration from the poor and near-poor, because they have no mar- cities to the countryside. gins to absorb the decline in incomes. The ability of the poor to participate in the re- Devaluations generally had a pro-poor im- covery can be severely hampered by these responses. pact in the 1980s in rural Africa because they raised Evidence from Latin America indicates that the effects the competitiveness of the tradable sector. In East FOCUSING ASSISTANCE STRATEGIES ON POVERTY: FACING CRISES 31 Asia, however, the net effect of devaluations- wherever possible-but evidence from Latin possibly because of their magnitude-appears to America suggests that a crisis may not be the have been contractionary, as they led to the col- best time to reform existing subsidies, as resistance lapse of banks and companies that held foreign- from those hurt by the change may stop reforms. denominated debt and to further declines in Other public investments that affect the pro- economic activity. To counter the impact of de- ductivity of the poor should also be protected, es- valuation, expansionary fiscal policies emerged as pecially investments in rural infrastructure and the appropriate macroeconomic, and poverty- the provision of microfinance. And for all pub- alleviation, response. They have become a key el- lic expenditures, especially those in the social sec- ement of internationally-supported adjustment tors, it is also important to assess institutional programs in 1998, notably in Indonesia, Korea, capacity. No amount of public spending will ben- and Thailand. efit the poor if corruption is rampant or if deliv- A fiscal stimulus directed at such labor-inten- ery mechanisms exclude the poor. sive activities as rural roads or other rural em- a Build or reinforce safety nets capable of provid- ployment programs would combine the benefits ing effective insurance before a crisis and boost of growth with those of income support for poor assistance once a crisis hits. During a crisis the groups. If there are intertemporal tradeoffs, for ex- focus should be on supporting or expanding pro- ample, between current consumption and in- grams that deliver the services the poor need to vestment that could boost growth at a later stage, help them cope with the crisis (for example, the potential long-term losses would have to be transfers to buy food), that reach the poor (even weighed against the short- and long-term gains though they may reach some who are not poor), for the poor deriving from a smaller shock. In and that are ready to be scaled up quickly. Pro- choosing a combination of policies, policymakers grams that can be scaled up include public works should avoid those leading to high inflation, as schemes and other workfare programs, which the poor often rely on fixed incomes and are hurt can provide employment for the poorest and re- by fast-rising prices. duce open unemployment. Expanding unem- E Ensure that fiscal adjustment protects spending ployment assistance may be an option where the items important to the poor and redistributes re- institutional infrastructure for such a program is sources and that services areprovided by effective, in place. inclusive institutions. Fiscal policies that protect * Establish mechanisms for monitoring the impact spending on basic education and health care can of the crisis and evaluating responses. Adequate prevent cuts in services that the poor use most and information on the impact of the crisis on vari- protect their ability to build up human capital. ous groups and areas helps in the design of ap- Eliminating unproductive expenditures may help propriate policy responses and signals a need for keep social spending at its pre-crisis levels. In ed- corrections if the desired impacts are not materi- ucation, expenditures on primary schools and on alizing. Information can also play a crucial role nonsalary items that are essential for quality in national and local politics of design and im- should be maintained; targeted subsidies to reduce plementation. At a national level encouragement dropout rates among the poor (for example, feed- of well-informed debate on issues, impacts, and ing programs or scholarships tied to attendance) alternatives is important. At a local level provision should also be increased. Measures should be of public information on intended public action- taken to avoid a rise in child labor. and the rights of potential beneficiaries-can help In health, spending should be maintained for reduce corruption and leakages of program activities with high externalities, such as vacci- benefits. nations and disease vector control, and for health * Set ap interventions thatpreserve the socialfabric care services at the lower levels of the health sys- of societies in crisis and build social capital. There tem. Subsidies on the prices of essential com- is evidence that at least some of the negative so- modities and drugs can help the poor avoid cial impacts of a crisis-rising social tension and malnutrition and illness. Here and elsewhere, the breakdown in family and community ties-may price subsidies should be targeted to the poor persist after the end of the crisis-this was the case 32 POVERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 with violence in urban Latin America (Fajnzylber, * Setting up a safety net that insures the poor Lederman, and Loyaza 1998). It may be possible against risks and protects them when a crisis to mitigate these negative impacts by supporting comes. the build up of local capacity Lhrough NGOs, but not much is known about how to do this. UNDERSTANDING TRANSMISSION MECHANISMS AND DIS- Two observations are relevant in thinking about TRIBUTIONAL IMPLICATIONS OF POLICIES. Formulating bet- how such an agenda would be implemented. First, ter policies to mitigate the impact of crises on the poor institutional capacity constraints must be taken into will require more analytical work and empirical ev- account in the design of policy responses.5 For ex- idence on the distributional impacts of crises and ad- ample, the best safety net program may be far less justment and on what causes households to move out effective than other alternatives if it requires admin- of poverty or fall back into it. These areas constitute istrative capacity well above what is available. Sec- a rich research agenda for the future. ond, action may have to be taken before sufficient On the theoretical side one of the main instru- information is available. There will usually be inad- ments used in analysis has been computable general equate information on likely and actual effects going equilibrium models. These models describe the struc- into a crisis. Nevertheless, those designing a re- ture of an economy on both the production and the sponse should take immediate action based on read- consumption side, distinguishing between socio- ily available analysis and general principles and economic groups. They have been used to generate simultaneously set up a process of information gen- estimates of the impacts of shocks and policy re- eration and diagnosis to inform future action. Often sponses.6 However, the findings are often very sen- there will be the most information on how to move sitive to the assumptions underlying the models, so in the public spending and crisis monitoring domains. their reliability as a guide for policy decisions is not uncontroversial. The politics ofpro-poorpolicy responses On the empirical side, better and more timely in- formation on transmission channels and household To be selected and adequately implemented, policy responses is needed both to calibrate theoretical packages to protect the poor neecd political support. models and to test hypotheses. Such information Crises often present an opportunity to implement should come from participatory studies, which can policies that lead to positive and persistent distribu- provide valuable insights into the dynamics of tional or social benefits, because they create a sense poverty, as well as from standard household surveys. of urgency and a focus on the plight of the poorest, Despite reasonably good overall poverty monitoring and because social unrest may threaten the fabric of systems, the response in East Asia was hampered by society. Whether such opportunity can be seized de- the lack of information on the short-term impact of pends on whether and how well the reforms have been the crisis. Short- and long-term monitoring tools prepared prior to a crisis and on whether political dy- should be developed as part of a long-term devel- namics allow governments or societal actors to sup- opment strategy. port changes against previously entrenched interests. SETTING UP SAFETY NETS. In addition to developing Preparing for crises a better understanding of the distributional impacts of shocks and policy responses, countries should en- Crises, economytvide or localized, will always be with sure that appropriate safety nets are in place to as- us, and responses that protect the poor will be eas- sist households in reducing or mitigating the risks they ier to design if more is understood about the distri- face.7 Earlier crises in Latin America, Asia, and else- butional impacts of crises and if mechanisms to where offer two lessons: protect the poor are built into long-term development * Safety net mechanisms are too often inadequate. strategies. Being prepared for a crisis requires: Coverage is often limited and leakage is high, so * Understanding the mechanisms through which the that assistance falls far short of demand during a impact of a crisis may spread crisis. Moreover, the poor are often unaware of * Understanding the distributional implications of the programs or have too little influence to ob- policies and programs tain what they are entitled to. FOCUSING ASSISTANCE STRATEGIES ON POVERTY: FACING CRISES 33 * It is very difficult to set up effective safety nets dur- * The safety net should be efficient, in the sense that, ing a crisis. Governments are often unprepared, at the margin, money spent on it should be as ef- ill informed, and slow to take action, while both fective in raising the welfare of the poor as money financial and human resources are stretched thin. spent on other programs. Building up the infrastructure and the capacity to Experience in a number of countries helps iden- manage a program takes a long time-and is tify a combination of programs that satisfy these more easily done during good economic times. principles and that have been effective in protecting Without an adequate safety net in place before a the poor during crises. These include programs that crisis hits, the poor are likely to be hurt dispropor- provide employment for those who are able to work, tionately, so setting up safety nets in good times may along with targeted transfers for those who cannot, be the only effective way to protect the poor during or should not, work. There is also a case for special crises. Recent experience also calls for a reexamina- credit programs, though design issues are particularly tion of the distinction between relief programs and reg- important here for efficiency and effective targeting. ular development programs. Before a crisis good * To mitigate the risk of loss of income resulting safety nets may provide the insurance necessary for from the loss of a job, a key element of the sys- households to make riskier choices that increase their tem would be a workfare program that provides productivity and spur growth. During a crisis safety nets work on public infrastructure projects to people may well provide a way of ensuring that crises do not seeking employment. The work would be avail- halt "development"-by protecting the poor from able at all times, but it would expand during being forced to forgo essential household investments crises as demand expands-without the need for in health and education, which are crucial for future administrative decisions. To ensure local rele- growth, or to finance consumption by divesting the vance and ownership, the projects would be pro- physical capital on which their productivity depends. posed by local community groups, with final Thus an effective safety net for the poor should be seen selection by a central agency. The central gov- as a long-term investment. ernment would fund wage costs everywhere (up Several principles should guide in the design of to, say, 15 days a month) and nonwage costs only a safety net: in designated poor areas. Good models for this d The safety net should provide a better way to in- type of program are the Argentinean Trabajar sure the poor against the risk of loss of income. scheme for middle-income countries and the Ma- Other elements of a country's poverty alleviation harashtra Employment Guarantee Scheme in India strategy address economic growth and investment for low-income countries. in human capital; safety nets are meant to pro- A key to success is the wage rate. A low wage vide insurance. The poor frequently rely on in- rate ensures that only those in need apply and that formal insurance mechanisms, including strategies as many people as possible can be employed. A to reduce or diversify risk, but these characteris- low wage would also protect the incentive to tically incur high costs. The key issue for policy take up regular work when it becomes available is not whether publicly provided safety nets dis- again. When the shock is transitory, the wage place private mechanisms, but whether they pro- could be set no higher than, say, 90 percent of vide insurance more efficiently and at a lower cost; the wage rate for unskilled agricultural work pre- if they do, then displacement may be desirable. vailing in normal times. When the shock is per- . The safety net should respondflexibly to the needs manent, setting a wage would require an of the poor and not rely on administrative dis- assessment of the sustainable wage for unskilled cretion. If a safety net does not provide insurance labor after the crisis. This may be difficult; erring when a shock hits (for example, because it does on the side of offering too low a wage may be not reachi the intended beneficiaries or because safer than setting too high a wage, as it is easier access is rationed), then it does not provide ad- to increase the wage than to lower it. At the set equate insurance for the poor. wage, all those who are willing to work should * The safety net should not createperverse incentives find work; if there is not enough work for all, then that would increase unemployment or other forms the program would fail to provide credible of dependency on state support in the long run. insurance. 34 POVERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 A workfare program would protect the poor- getary outlays (Manuelyan Atinc and Walton 1998). est workers-those willing to accept low wages- Amounts of that magnitude would probably be suf- against declines in labor demand and loss of ficient to fund programs such as those described income, thus lessening the harmful impact of de- above. To ensure that resources are available for a clines in labor demand and informal-sector nationwide crisis, a fund could be set up with regu- earnings.8 lar payments to amass resources during normal years. * To mitigate the risk of long-term negative effects of income losses, targeted transfers of cash or Bank assistance to countries in crisis food would be provided to those unable to work, such as pregnant and lactating women or the el- The social dimensions of crises have generally been derly, or those who should not work, such as chil- neglected. This was markedly so during the 1980s dren in school. These programs could be turned debt crisis, when the characteristic response on so- on and off, or expanded, based on indicators of cial dimensions was a social fund or measures to pro- crisis-and the demand for work in the workfare tect social spending-often set up years after the program could be used as arL indicator. A good shock (Jayarajah , Branson, and Sen 1996). In the example of a scheme for keeping children in 1990s the response improved somewhat, but re- school is the Bolsa Escola scheme in Brazil, which mained weak. offers scholarships to families who send all their children to school. Such targeted programs miti- Responding to current crises gate the impacts of increases in the cost of edu- cation and prevent long-term losses in educational During the East Asia crisis the World Bank has worked attainment and earning potential. to help governments implement pro-poor strategies v To assist in the recovery after a crisis, a program from the start. Overall, countries in East Asia re- of carefully targeted microcredit and savings ser- ceived $5.7 billion for five adjustment operations in vices may help the poor reacquire productive fiscal 1998, the largest being two loans to Korea for capital once opportunities appear. During a cri- $3 billion and $2 billion.9 Ninety-three percent of this sis such a program could mitigate the risk of per- lending, including the two loans for Korea as well manent losses in income-earning capacity by as a $300 million loan to Malaysia, was classified as avoiding distress sales of productive assets. How- poverty-focused. These loans helped support pro- ever, often grants rather than loans are needed, poor structural policies and protection of public ex- and care must be taken to ensure that clients un- penditures that benefit the poor and enhance the derstand the nature of the assistance received. quality of social services. Efforts went into improv- While there are concerns about the effectiveness ing the design and financing of social safety nets, in and sustainability of microfinance schemes, mod- particular social funds and social security systems for els of successful schemes exist-for example, the unemployed and the elderly; addressing key so- Grameen and BRAC in Bangladesh (Khandker cial and institutional issues; and setting up poverty 1998), whose ability to help in crises was put to monitoring systems. the test in the aftermath of the 1998 floods. Elsewhere, substantial support went to the coun- These programs would be permanent, dealing tries of the former Soviet Union and Latin America with issues of transient poverty, localized shocks, and for safety net reform and rationalization of social as- poor-area development in normal years and ex- sistance, and to assisting countries hit by natural panding during times of crisis to assist those tem- disasters. porarily in need. The budgetary outlay for such a set of programs may be higher than miost countries cur- CHOOSING PRO-POOR STABILIZATION AND STRUCTURAL rently spend, but possibly lower than the cost of ADJUSTMENT POLICIES. At the start of the crisis, adjust- hastily prepared relief operations set up after a cri- ment programs in Thailand and Korea focused on sis strikes. In Indonesia the cost oi setting up a pub- countering capital outflows and the resulting down- lic employment scheme to transfer enough income ward pressure on the exchange rate by tightening to the poor to restore pre-crisis (1996) consumption monetary policy. Structural reforms addressed weak- levels has been estimated at 3.5-5 percent of bud- nesses in the financial and corporate sectors. In FOCUSING ASSISTANCE STRATEGIES ON POVERTY: FACING CRISES 35 Thailand, and to a lesser extent in Korea, fiscal ad- poor (through free housing and food supple- justment was intended to support the adjustment of ments) and income generation through small the external balance. The programs failed to restore grants. confidence and capital outflows grew considerably * In Indonesia the Structural Adjustment Loan in- larger than expected, aggravating problems in the fi- cludes subsidies on the prices of key goods and nancial and corporate sectors and leading to a col- scholarship funds for 2.6 million needy junior lapse of private demand. In these circumstances secondary school students. staying with the original fiscal targets would have * In Tnhailand the Bank is working closely with the been strongly contractionary. The targets were sub- government to identify public expenditures to stantially revised to permit a more accommodating be protected or expanded under the Second Eco- fiscal stance. Partial evidence suggests that govern- nomic and Financial Adjustment Loan. ments have endeavored to protect social expendi- tures benefiting the poor. Korea's second Structural IMPROVING THE FUNCTIONING OF SAFETY NETS. Safety Adjustment Loan (SAL) supports expansions in ex- net programs are also generally among expenditures penditures accompanied by a higher budget deficit slated for expansion. Most frequently, governments of 5 percent, and a similar expansion is under dis- have expanded public employment programs, fol- cussion in Thailand. lowed by insurance, training, and pension schemes. On the structural reform front both Structural Ad- * In 7Tailand a $300 million loan for a Social In- justment Loans for Korea include significant labor mar- vestment Project will fund job creation for the ket reforms. The $2 billion loan approved in March poor and the unemployed through existing labor- 1998 incorporates measures to increase flexibility intensive government programs that implement while extending coverage of unemployment insur- small-scale community projects and larger munic- ance to employees in small-scale enterprises. The No- ipal projects and expand training for the unem- vember loan carries these reforms further. ployed. It will also support health insurance The design of policies to protect the poor is com- schemes for low-income workers. The loan is ex- plicated by the lack of a full understanding of the im- pected to create roughly 1 million person months pact of policy packages not just on the welfare of the of work and an equivalent amount of training. poor, but on macroeconomic variables themselves. * Pension reform was supported in both Structural The debate on whether bank recapitalization or di- Adjustment Loans in Korea. rect transfers or tax cuts would have a stronger im- * In Malaysia longer-term issues relative to the ad- pact on growth is an example of the difference in equacy of formal safety nets and the governance views on the effectiveness of policies. The distribu- structure of the Employee Provident Fund will be tional impacts of alternative policy packages-how addressed through a Technical Assistance Loan they affect different groups-are even less well and economic and sector work. understood. SETTING UP SYSTEMS TO MONITOR POVERTY. Measures PROTECTING PUMBLIC SPENDING THAT BENEFITS THE POOR. to strengthen or build up systems to provide timely The World Bank has also emphasized the importance information on the impact of the crisis are included of the composition of public expenditures. Expen- in a number of projects: ditures that should be expanded are those that ben- * The Bank is helping to finance a follow-up to the efit the poor. Structural adjustment loans in Indonesia, Indonesia Family Life Survey. This will allow Korea, Malaysia, and Thailand all include components monitoring of the living standards of a subsam- aimed at protecting public spending for health and ple of households that were surveyed in 1993 and education, which has a stronger pro-poor impact 1997 and help assess household-level coping than other types of spending. strategies in response to the crisis. * In .Malaysia a $300 million June 1998 Economic * In Korea the first Structural Adjustment Loan in- and Social Sector Loan supported budgetary cluded measures to improve poverty monitoring. spending for education, health, and rural infra- * In the Philipines a new survey, the Annual Poverty structure and increases in expenditures on pro- Indicators Survey, has been conducted with fund- grams aimed at providing direct support to the ing from the Asia-Europe Meeting (ASEM). 36 POVERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 * The Social Investment Project in Thailand in- * Ongoing work on labor market adjustment, cludes funds to set up a monitoring system to eval- which focuses on policies for unemployment uate the impact of the crisis and of public action and underemployment and on the impact of ac- on the poor, and a national poverty map will be celerated reforms in the state-owned enterprise drawn based on a nationwide systematic partic- sector and the slowdown in aggregate demand, ipatory assessment. which has been exacerbated by the regional crisis RESPONDING TO NATURAL DISASTERS. In disaster relief, * A study examining the impact of the regional cri- the World Bank supported reconstruction and reha- sis on Cambodia and the Lao People's Democra- bilitation of infrastructure and agricultural production tic Republic, which uses rapid social assessments capacity. The past two years were marked by natural of impacts on the poor calamities of enormous proportions-from earthquakes * A regionwide initiative to analyze issues in pen- and the floods and drought related to El Nino in China, sion policy and administration, which includes a to floods in Bangladesh and Hurricane Mitch in Cen- conference in 1999 on governance of pension tral America. Natural disasters may affect the poor funds in EastAsia. more because they tend to live in areas more prone * Assessments of the impact of adjustment pro- to floods and landslides (where land values are lower). grams on vulnerable groups, which is part of a And the poor have a harder time coping because their pilot project on strengthening collaboration be- resources are limited, and they have little to fall back tween the World Bank and the International Mon- on. Their resistance to disease is also often lower be- etary Fund; pilot countries are Cameroon, cause of malnutrition. Labor-intensive reconstruction Ethiopia, Nicaragua, Tajikistan, Vietnam, and efforts provide an opportunity to build local capacity Zimbabwe. and to help the poor find employmnent. Several fiscal 1998 projects were also aimed at * Labor-based construction methods are employed in strengthening safety nets to improve the capacity to the reconstruction of infrastructure damaged by respond to crises: floods in Bank-supported Emergency Reconstruc- * In Argentina a 8284 million loan supported ex- tion Loans in Ecuador, PapuaXNew Guinea, Peru, pansion of the Trabajar project, which provides and Tajikistan. up to 15 days a month of guaranteed employ- * The capacity of local-level public institutions is ment in locally determined small infrastructure being strengthened with support from an Emer- projects for anyone willing to work for a low gency Reconstruction Loan in Cbina's Hebei wage. Province, which was hit by an earthquake, and * In Kazakhstan a $300 million adjustment loan sup- in Peru. In Papua New Guinea private local in- ported pension reforms. A resolution providing stitutions-churches and NGOs-are being used for scheduled increases in the minimum pension to deliver services. adjusted for inflation, together with the estab- lishment of old-age allowances, would help en- Preparingfor thefuture sure sufficient income support for pensioners. Georgia, Hungary, and Moldova received support A number of studies now under way will improve for pension reform through an adjustment loan the understanding of the impact of shocks and pol- as well. icy responses on the poor: * In Russia a $29 million loan, complementing an * Poverty analysis in the Pbilippines based on the earlier adjustment loan, supported a broad pro- results of the 1998 Income and Expenditure Sur- gram of social protection, including the setting up vey and the Annual Poverty Indicators Survey of monitoring systems, data analysis, and reviews * A social and economic study of Malaysia (the first of ongoing programs for unemployment assistance since 1993), which includes an overview of and social protection. About 4 million of the poverty and the social safety net, analysis of how poorest pensioners would receive regular and the poor may be affected by the downturn, and adequate pensions. recommendations for limiting the impact of the Measures to increase preparedness for natural crisis on the poor calamities were also included in a number of projects. FOCUSING ASSISTANCE STRATEGIES ON POVERTY: FACING CRISES 37 Prevention focused on strengthening infrastructure Notes and setting up regular maintenance programs, as well as acquiring emergency equipment. 1. This section builds on Ferreira and Keely (1998). * All five El Nifio-related Emergency Reconstruction 2. It should be noted, however, that in some cases in- Loans included flood prevention and drought flation appears to have had only temporary effects on the mitigation measures, from maintenance of water distribution of income. courses and water drainage system to construc- 3. Enrollment rates in primary education appear not to tion of dams, dikes, and water storage tanks (Bo- have declined. livia, Ecuador, Papua New Guinea, Pent, Uganda). 4. Note that this agenda does not refer to immediate * Through the Grants Program the World Bank is responses. Faced with rapid capital outflows, a country supporting several Consultative Group for In- cannot change its fiscal or social protection policies ternational Agricultural Research centers around overnight. the world working on agricultural risk mitiga- 5. The extreme cases of post-conflict situations where tion. In Papua iNeuw Guinea the Bank is also sup- the state structure has broken down and there are no porting agricultural research on drought-resistant NGOs or other organizations capable of implementing crops. programs pose a particular challenge. * Several loans (for example, in Bolivia) include 6. The Latin American and the Caribbean Region de- funds for the purchase of bailey bridges, tempo- veloped a general equilibrium model that attempts to go rary power generators, water purification plants, beyond traditional wage rigidity-based dual labor market pumps, and other emergency equipment. models. The model explicitly integrates workers' choice Although natural disasters continue to claim lives, of formal and informal sectors and models segmentation progress is being made. A cyclone that hit Bangladesh as emerging endogenously from efficiency wage effects. in 1996 caused fewer than a hundred deaths; a sim- This makes it possible to incorporate the existence of ilar cyclone that hit roughly the same area in the 1970s both voluntary and involuntary informal workers. The claimed a quarter of a million lives. model was calibrated using Mexican data, and simulations To ensure that countries are better prepared to deal were run on the impact of various shocks and policy with shocks, the Bank is also piloting comprehensive measures on the size of the informal sector and the ratio reviews of policies and programs, as suggested by Pres- of formal to informal wages, and hence distribution; it sug- ident James Wolfensohn in his speech at the 1998 Joint gests possible magnitudes of impact. World Bank-International Monetary Fund Annual 7. This section is based on Ravallion (1999 and forth- Meetings and his proposal for a Comprehensive De- coming). velopment Framework. The reviews would provide an 8. Setting up and implementing workfare programs may opportunity for periodic checks on crisis preparedness. be difficult in countries where community-based organi- Safety nets would figure prominently in the review, zations are weak, or where public works have tradition- to highlight and address the need for reform and ally been hampered by corruption. strengthening before a crisis hits. Work is under way 9. These were followed in fiscal year 1999 by a fur- in a number of pilot countries to define the scope and ther $2 billion loan to Korea and a $1 billion loan to In- process of such reviews. donesia, which were also classified as poverty-focused. CHAPTER 3 Monitoring Poverty Outcomes: Integrating Different Methods in Poverty Assessments World Development Report 1990 Poverty and the Different methodological approaches subsequent 1991 policy paper Assistance Strategies to Reduce Poverty laid the foundation for the World Poverty analysis involves data collection, data analy- Bank to start carrying out poverty assessments in its sis, and strategy formulation.' At the data collection client countries. Depending on the country situation, and analysis stages, the methods applied range from the government's commitment to poverty reduction, large sample surveys analyzed by researchers to and available data, poverty assessments analyze: long-term anthropological village studies. At all three * The poverty profile-the most important eco- stages, but especially in strategy formulation, differ- nomic, demographic, and social characteristics ent degrees of participation can be employed-from of the poor, including gender differences simple dissemination of information to approaches a Trends over time that genuinely involve and empower individuals and * The impact of policies and programs on the poor. communities. While the approaches are sometimes Building on this analysis, poverty assessments ex- characterized as originating in different epistemo- amine strategies to reduce poverty and recommend ways logical traditions-the more quantitative methods to increase the effectiveness of policies and programs. from positivism, the more qualitative and participa- Poverty assessments have relied on various ap- tory methods from relativism-in practice they can proaches to gather and analyze data on poverty. be used together to gain a better understanding of The vast majority make extensive use of data collected poverty and develop strategies to fight it. through sample-based household surveys analyzed by researchers, but an increasing number also rely Contextual and noncontextual data collection on information collected using participatory tech- and analysis methods niques and analyses carried out with and by poor peo- ple in the field. Poverty assessments have also During data collection it is useful to distinguish be- included participation by various stakeholders, from tween types of data and methods of data collection poor people to NGOs to policymakers, shifting, to (Hentschel 1998). Data can be qualitative or quanti- varying degrees, from one-way dissemination of in- tative, while methods are more usefully characterized formation to more meaningful forms of involvement. as contextual and noncontextual. This chapter reviews different methodological ap- * Contextual methods seek to understand human be- proaches to poverty data collection and analysis and havior within the social, cultural, economic, and highlights the advantages of integrating methods. It political environment of a locality, usually a vil- also reviews the advantages and risks of participation, lage, neighborhood or social group. finding that participatory poverty assessments, once * Noncontextual methods abstract from the partic- mainly a way of enriching the description of poverty, ularities of a locality to gauge general trends. now place more emphasis on understanding the This characterization is best seen as describing causes of poverty. The challenges for future poverty the extremes of a continuum, where the methods work lie in better and more widespread use of inte- used range from contextual to noncontextual and the grated approaches and in exploiting participation to data gathered range from qualitative to quantitative, develop better poverty reduction strategies. though both methods can produce both kinds of data. 38 MONITORING POVERTY OUTCOMES: INTEGRATING DIFFERENT METHODS IN POVERTY ASSESSMENTS 39 Household surveys such as the Living Standards Mea- incidence of poverty for such households. But this surement Surveys or the Priority Surveys are closer result was based on the assumption that there (though not exclusively) to the noncontextual end and were no economies of scale in household con- generate mostly quantitative data; Participatory Rural sumption, which means that there are no draw- Appraisals are closer to the contextual end and pro- backs to being part of a small household relative duce mostly qualitative data. Data analysis can also be to a large one. Once this assumption was changed conducted in more or less contextual ways. to account for the presence of economies of Different methods are appropriate to answer dif- scale, survey data confirmed that widow-headed ferent questions. Contextual methods are well suited households, which were smaller than the average, to answer questions on the meaning of poverty, on tended to be poorer as well. priorities, and on motivations: what poverty is, what * Noncontextual studies can be used to assess the are the most pressing needs of the poor, why cer- representativeness of contextual studies. If the tain groups do not use certain services, and so on. characteristics of households covered by contex- Contextual methods allow an understanding of the tual studies fit within the national profile for given experience and meaning of poverty from the per- indicators, the findings from such studies may be spective of the poor, and permit an analysis of psy- more easily generalizable. Similarly, the selec- chological, cultural, and political factors that are not tion of sites for contextual work may be based on easily quantifiable. In addition, they make it possi- the sampling frame used for household surveys. ble to consider local factors. Noncontextual methods * Contextual studies may provide causal linkages are best suited for questions that require general- and hypotheses, which can be tested on a large izations and refer to large and diverse geographical scale using survey data. areas and social groups: how many people in the Integrating methods raises issues of timing and country are below the poverty line, how many chil- sequencing: dren suffer from malnutrition, what relationship ex- * If a participatory poverty assessment is carried out ists in aggregate between household consumption and before a household survey, for example, the as- education, and so on. Noncontextual methods pro- sessment findings can be used to raise new issues, vide results that can be generalized. Since the defi- generate hypotheses, and help formulate or re- nition of poverty is based on the same factors think certain questions. throughout a country, living standards can be com- * If it comes after a household survey, its findings pared across communities. can be used to explore anomalous survey find- ings or to focus on certain types of households. Integrating contextual and noncontextual * When participatory poverty assessments and methods household or other surveys are planned at the same time (even if they are carried out sequen- Contextual and noncontextual methods can be inte- tially), sample sites should be carefully selected- grated in several ways to gain a deeper understand- for example, the participatory poverty assessment ing of poverty: can focus on regions where poverty appears * Contextual studies can clarify and refine survey more persistent or where there are concerns over questions and identify the answer options to be particular vulnerabilities. provided in pre-coded questionnaires. * Deriving a poverty profile from survey data in- The contribution ofparticipation volves a number of assumptions. Contextual stud- ies can confirm the validity of poverty profiles, At all three stages of poverty analysis, but especially highlight inconsistencies, and suggest solutions. in strategy formulation, various degrees of partici- The process can sometimes point to a need to pation can be employed, ranging from information change the methodology used to derive the pro- dissemination (one-way flow of information) to file. For example, contextual studies in South consultation (two-way flow of information) to col- Asia find that widow-headed households are laboration (joint decisionmaking) and empower- among the poorest and most vulnerable; yet sur- ment (transfer of control over decisions and vey data for Nepal did not find an above-average resources). Various stakeholders can participate, 40 POXERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 ranging from poor people to policyrnakers. In data to exist in 45 out of 80 countries for which data on collection contextual methods are often participa- capacity for poverty monitoring were gathered by the tory, whereas noncontextual methods tend to be 1998 census of household surveys (see chapter 1 for nonparticipatory. other results from the 1998 census). More and more Participation in poverty analysis differs from par- poverty assessments rely on both qualitative and quan- ticipation in project evaluation. While participation titative data gathered and analyzed using different in projects ideally aims to maximize the involve- methods and the use of participatory approaches to ment of intended beneficiaries in identifying, de- develop poverty reduction strategies. signing, implementing, and monitoring the project and A review of the experience to date indicates that: often involves collaboration and even empower- * Initially, participatory poverty assessments tended ment, the degree of participation in studying poverty to be stand-alone exercises separate from regu- and formulating strategies is more variable. This is lar poverty assessments, and contextual and non- in part because the link between suclh studies and a contextual methods were not integrated. The concrete policy response is less direct than in pro- main contribution of participatory poverty as- ject work. For the poor participation is limited largely sessments was to broaden and deepen the de- to encouraging them to describe and analyze poverty scription of poverty emerging from household in their own terms but does not include taking ac- survey data. tion. For local policymakers, researchers, and non- * As experience grows, integration between ap- governmental organizations participation may involve proaches is increasing, and the contribution of designing and carrying out the study and discussing contextual approaches and participation is be- the results. ginning to extend to strategy formulation. The advantages of a participatory approach to the formulation of poverty reduction strategies are clear: Initial experience: richer descriptions of poverty those in charge of implementation are more likely to be committed to strategies that emerge from con- The region with the largest number of participatory sultation, collaboration, and empowerment. Moreover, poverty assessments is Sub-Saharan Africa, and sev- the work itself can bring about changes in attitudes, eral reviews of the African experience have been con- as the poor begin to be seen as capable of influencing ducted over time, often in the context of the Special choices about their future rather than as passive re- Program of Assistance. Reviews by Norton and ceivers of services. A dialogue can be created between Stephens (1995) and the Task Force on Poverty in groups in society and partnerships ftor the imple- Sub-Saharan Africa (World Bank 1996b), as well as mentation of policies and the delivery of services. In others, concluded that participatory poverty assess- addition, capacity to carry out participatory poverty ments had enriched poverty profiles in several ways. analysis can be built in the country. But participation Participatory poverty assessments had: carries risks as well: the process can become politi- * Provided a richer, more effective picture of what cized and raise expectations of empowerment that it means to be poor may not be met. * Pointed out dimensions of deprivation other than insufficient consumption or income and poor Recent experience with different health and education-for example, lack of power approaches to poverty analysis and voice, hopelessness and despair, lack of safety and property rights The World Bank's experience with participatory poverty * Highlighted the importance of locally specific assessments is rich and growing. As of July 1998, 49 factors in determining access to services and con- participatory poverty assessments had. been under- straints that the poor faced in availing themselves taken-28 in Africa, 6 in Latin America 11 in Eastern of market-based opportunities Europe, and 2 each in South and East Asia (see Robb . Underscored the vulnerability of the poor to risk 1998). The share of poverty assessments that included and the need for safety nets. participatory components rose from one-fifth in fiscal Comparatively less emphasis was placed on de- 1994 to over half in fiscal 1998.2 Encouragingly, capacity veloping causal explanations of the plight of the to conduct participatory studies of poverty was reported poor. MONITORING POVERTY OUTCOMES: INTEGRATING DIFFERENT METHODS IN POVERTY ASSESSMENTS 41 Eleven participatory poverty assessments were re- able to new shocks or expenses-sudden illness, cently conducted in Eastern Europe and the coun- loss of employment, school expenses. tries of the former Soviet Union, which are part of * Petty trade, small-scale home-based enterprises, the Europe and Central Asia Region at the Bank.3 informal sector employment, domestic work, ca- Poverty has increased dramatically in many of the sual manual labor, migration to cities, and sub- countries of the former Soviet bloc during the tran- sistence gardening have become important sition, and little was known about the extent and char- supports for the new poor, many of whom lack acteristics of poverty in the former Soviet Union. So the education, connections with those in power, it was particularly important to analyze poverty from or access to information and credit that would en- different angles and look at its dynamics, something able them to move into the formal economy. for which participatory poverty assessments are well * Among the poor, poverty has threatened com- suited. munity solidarity and weakened extended family A recent review by Dudwick (1998) highlights the ties, even in societies with strong extended fam- key contributions of participatory poverty assess- ily networks such as Armenia and Georgia. Peo- ments. The profile of poverty in the region was en- ple have drastically curtailed their participation in riched by the following observations: ceremonial and social events such as weddings, * For most people in the region, acute poverty is birthdays, or funerals that previously brought a recent phenomenon. It has affected not only them together and reaffirmed relationships be- those who were economically vulnerable (be- cause they cannot afford the costs. For poor peo- cause of disabilities, for example), but also those ple in the Europe and Central Asia Region, it is a who were highly educated and had stable, even bitter paradox that widescale poverty has simul- prestigious employment. The unexpected nature taneously eroded their social networks and made of impoverishment, which came suddenly with hy- these networks ever more important for finding perinflation, loss of savings, nonpayment of work. Given the degree of unemployment, "con- salaries, and unemployment, has created in the nections" to those with official or unofficial power, new poor a pervasive sense of insecurity, un- bribes, and access to knowledge and initial cap- predictability, and fear for the future. ital-which the poor tend to lack-have become * The socialist regimes claimed to have elimninated essential to finding good employment or starting poverty; hence responsibility for any poverty that successful enterprises. did exist was placed squarely on the individual. In the Europe and Central Asia experience inte- Therefore, despite widespread poverty, many of gration of various methods has been limited so far. the poor view poverty with ambivalence: they as- Often, participatory poverty assessments have served cribe their own family's poverty to unemployment, as background papers for poverty assessments, with late wages, inflation, government corruption, ill- household income and expenditure surveys and ness, and so forth, but are likely to ascribe the other questionnaire-based surveys providing the poverty of others to personal failings such as main basis for analysis. At times, the findings of par- laziness, incompetence, or alcoholism. ticipatory poverty assessments have been compart- * As part of the restructuring of state industrial and mentalized-as colorfully illustrative but analytically agricultural enterprises, social assets-housing, insignificant anecdotes-rather than integrated ana- clinics, kindergartens, community buildings lytically into poverty profiles, discussion of causes, were divested, contributing to the hardship of and policy recommendations. For the most part, par- the poor, who relied more extensively on these ticipatory poverty assessments have been carried services. out in relative isolation from the final poverty * The poorest households in the region have almost assessment. no cushion against shocks or crises. Most of them Both the Africa and the Europe and Central Asia have sold their assets-the jewelry, household experience point to the important role played by par- items, and furniture acquired at marriage or pur- ticipatory poverty assessment in spurring debate on chased during better times; a car, if they had one; poverty. While poverty existed in the former Soviet and even recently privatized apartments. Their de- Union and other socialist countries, it was not dis- pletion of assets renders them ever more vulner- cussed as a social phenomenon-poor people existed 42 POVERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 simply as anomalies of a socioeconomic system that More advanced work: integration of approaches by definition was supposed to exclude gross in- equalities (see Braithwaite 1997). Poor people were As experience with contextual and participatory therefore ignored or excluded from society in vari- methods has grown over the past couple of years, ous ways. Today, poverty has become too wide- their use and acceptance has grown among econo- spread to ignore and is so manifestly connected with mists. Consequently, the degree of integration of the failure of the centrally planned economies that contextual and noncontextual methods has increased, it can no longer be regarded as something outside and so has the use of participatory techniques in the society. Only now are the people confronting the fact development of strategies. that they live in societies no longer based on redis- A recent review of participatory poverty assess- tributive economies and that poverty--or the risk of ments in Sub-Saharan Africa (Booth and others poverty-is intrinsic to post-socialist societies. As at- 1998) found that participatory poverty assessments titudes toward poverty begin to consolidate, the par- continue to enrich the analysis of poverty as described ticipatory poverty assessments can play a crucial above but that the latest work is moving forward in role by bringing these half-formed attitudes into the three important new directions: open where they can be publicly discussed and * Contextual and noncontextual methods are more negotiated. often integrated, and the synergies between meth- The finding that the participatory approach to ods more fully exploited. For the Tanzania par- poverty assessments is valuable in opening up fo- ticipatory poverty assessment, for example, a rums for discussion is in line with the findings of specially designed survey focusing on social cap- an Operations Evaluation Department progress re- ital was coordinated with a larger, traditional view on poverty assessment (OED 1996), which re- household survey, and participatory rural ap- ported that Bank staff considc red poverty praisal techniques were used in some carefully se- assessments that incorporated participatory ele- lected sites. ments to be "especially valuable," in part because * The causes of poverty are explored more sys- they led to fruitful discussions among government tematically, thus yielding insights not just on the and civil society stakeholders on more effective profile of poverty but also on the instruments that ways to reach the poor. can be used to help households escape poverty. In the Europe and Central Asia Region participa- * Country ownership is stronger than in the past. tory poverty assessments contributed to building For example, in Zambia regular poverty moni- capacity among local researchers. Some participatory toring includes both participatory studies and poverty assessments engaged local research firms household surveys: in Mali and other francoph- (such as the Kiev International Institute of Sociology one countries poverty-monitoring "observatoires" for the Ukraine study), training their surveyors to carry have been set up. out and write up qualitative interviews and inviting In Tanzania the integration of methods made it their participation in follow-up workshops to discuss possible to use household-level data to test econo- results. The two participatory poverty studies in Ar- metrically hypotheses on the importance of social cap- menia (1995, 1997) were carried out by anthropolo- ital for household well being that were derived from gists from the Institute of Ethnography and Yerevan focus-group discussions in the same communities. The State University, who made substantial inputs into the analysis showed that, after controlling for other vari- interview guide, choice of sites, and subsequent ables, village-level social capital had a large positive analysis. Both the qualitative methodologies and the impact on household expenditures (Narayan and priority given to respondents as subjects rather than Pritchett 1997). objects of research were new to many social scien- In the Europe and Central Asia Region there are tists, especially in the former Soviet Union, where so- two encouraging examples of integration: ciology and ethnography had historically operated * The two Armenia poverty assessments pulled to- under very stringent ideological limits. Social scien- gether analytically participatory and sample-based tists who worked on the participatory poverty as- studies that were planned and designed separately. sessments expressed great enthusiasm about the The final document uses the participatory findings potential for such an approach. to explain anomalies in the survey findings. MNJONITORING POVERTY OUTCOMES: INTEGRATING DIFFERENT MIETHODS IN POVERTY ASSESSMENTS 43 * For the FIR Mtfacedoniia poverty assessment a the poor's perception of economic mobility was very labor expert from the local Statistical Office, who different from that suggested by a standard, survey- had collaborated extensively with the Bank in the based poverty profile exercise. While the character- analysis of findings from the Statistical Office's istics of better-off and worse-off households were household income and expenditure surveys, similar to those obtained through survey exercises- helped design the participatory poverty assessment for example, wealthier households generally have so that it covered a subsample of the population more agricultural land, more education, higher paid already surveyed. Descriptive material and find- jobs, and better access to basic services-this static ings from the participatory poverty assessment characterization of households has little to do with were incorporated in the poverty assessment and (dynamic) opportunities for economic mobility. Poor helped to fill in critical gaps (on the informal econ- respondents generally had a realistic view of these omy, for example). opportunities that had little to do with emulating the In the South Asia region a recent study of poverty lifestyles of their wealthier, higher-caste neighbors. in Uttar Pradesh and Bihar states in Northern India For example, large landholdings were cited in virtu- illustrates one way to combine the methods in prac- ally every village as one of the main characteristics tice (Kozel and Parker 1998; Bamberger forthcoming). of the wealthy, but land acquisition was rarely cited The research was conducted in two phases: as a feasible means of economic improvement by the * In the first phase contextual research instruments, poor. Upwardly mobile poor households had gen- based on rapid rural appraisal and participatory erally found employment outside the agricultural rural appraisal techniques, were developed and sector and, in many cases, outside the village. Find- applied in 30 villages. The exercises included a ings such as these have the potential to inform much wealth ranking exercise, a social mapping exer- more effective poverty reduction strategies than sta- cise, a social capital inventory, household case his- tic findings. tory interviews, a women's roles-gender issues exercise, and discussions of availability, quality, Challenges for the future and usage of government programs and services. These exercises jointly provided a many-sided pic- While highlighting the potential gains of integrated ture of local poverty as perceived by men and methods, recent experience has also underscored the women, poor and nonpoor. They yielded a rich difficulty of using such approaches. Integrating meth- set of findings that would not have been possi- ods requires assembling a multidisciplinary team ble through analysis of conventional data ob- and giving careful consideration to sequencing. Com- tained through structured household and bining methods requires more time and resources community surveys. than employing either contextual or noncontextual * In the second phase the qualitative findings were methods alone. Thus one of the challenges for fu- analyzed and used to enrich and sharpen the de- ture poverty analysis is to institutionalize integrated sign of the study's household and community work in cost-effective and manageable ways. questionnaires, which were modeled after survey The Bank's emphasis on participation by the poor instruments developed for the Living Standards as subjects and active agents in social development Measurement Study. The field survey was carried rather than its objects calls for far greater reliance on out in 120 villages, including the 30 covered in participatory poverty assessment findings in formu- the qualitative component, with an overall sam- lating policy than is now the case. A first step is mak- pie size of 2,250 households. The findings of the ing the formulation of policy recommendations a qualitative portion of the study are being used to more participatory process. Integration does not stop inform, validate, and add explanatory power to at the analytical stage but extends to the policy for- the survey-based analyses. Joint use of the two mulation stage and to the monitoring of policy out- methods is yielding a rich picture of the situation comes, thus creating a feedback loop from of the poor in rural India and the broad factors information to policy and from policy to information that determine their well being. based on both sample-based and participatory sources This combination of methods yielded a number of data. This represents a second key challenge for of interesting and unexpected findings. For example, future poverty work. 44 POVERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 Notes sessments completed in fiscal 1998 (Chad, Djibouti, India, and Rwanda) included a participatory poverty 1. This follows loosely the steps of empirical investi- assessment. gation identified by Sechrist and Sedani (1995): data col- 3. The eleven countries for which participatory poverty lection, data analysis, data interpretation, and utilization assessments have been conducted are Albania, Armenia, of the information. Azerbaijan, Bulgaria, Georgia, the Kyrgyz Republic, Latvia, 2. As noted in chapter 1, four of the six poverty as- FYR Macedonia, Moldova, Tajikistan, and Ukraine. CHAPTER 4 Rethinking Poverty Reduction: World Development Report 2000/01 on Poverty and Development Poverty amidst plenty is an affront to our basic Contents: challenges and opportunities moral values. World Development Report 2000/01 on poverty and development presents an opportunity World Development Report 2000/01 will center on five to revisit the World Bank's poverty reduction strat- main propositions that reflect current policy con- egy in light of recent development experience and cerns and the analytical literature and that will serve future prospects. This follows the practice, begun to frame a program of analysis and dialogue in the in 1980, of preparing a major report on poverty re- preparation phase of the report: duction-the institution's overarching priority- * Standard of living is multidimensional. In dis- every 10 years. cussing poverty and inequality, therefore, tradi- tional measures based on the levels of income and Context: moving the frontier of consumption need to be accompanied by other knowledge measures that reflect such dimensions as health and education, vulnerability and risk, crime and in looking back over the period up to 1990, it be- violence, integration into the mainstream of so- comes evident that countries and societies that had ciety, and other factors highlighted by the poor made the greatest advances in attacking poverty themselves as being important. Moreover, not shared certain common features. They had pro- only must the measurement of the standard of liv- longed periods of social peace, they experienced sus- ing go below the level of the household to in- tained growth in per capita incomes, and they vestigate the conditions facing women and invested heavily in basic health and education. But children; it must also go above the household level even in the most successful countries there were to take into account community-level considera- pockets of marginalization, and gains for the poor re- tions. mained fragile. * There has been significant progress in poverty re- Since 1990 four factors have been important in ex- duction in the past half century. But this progress plaining successes and failures: has been uneven across countries and across the * Implementation of reform policies in Africa, Latin different dimensions of poverty, and poverty per- America, and Asia sists to an unacceptable degree, despite the * Collapse of social cohesion and the deterioration progress. into civil war in parts of Africa, central Asia, the * The world stands on the brink of extraordinary former Yugoslavia, and elsewhere opportunity for development and poverty re- * The trauma of transition in Eastern Europe and duction over the next 20 years, thanks to the the former Soviet Union global evolution of technology, trade, and polit- * The East Asian crisis. ical systems. As never before, the potential clearly These events have emphasized the importance of so- exists for meeting and even exceeding the vari- cial stability, of risk and vulnerability, and of mov- ous targets around which the international com- ing simultaneously on institutional and market munity is beginning to form a consensus. reforms, in addition to the factors highlighted in the * How broadly the fruits of development are shared, 1990 strategy.1 however, is a complex matter, depending on 45 46 POVERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 economic, social, political, demographic, envi- households cope with risk may determine not just ronmental, institutional, and policy factors. The whether they manage to absorb shocks in the short other side of the coin of extraordinary opportu- term but also whether they can preserve their earn- nity is the increased risk of inequality, vulnera- ing ability in the long term. Thus investing in mech- bility, social exclusion and marginalization of anisms to ensure and protect the poor against some groups, and social dissolution-risks linked risks-whether due to individual, community, or na- to the very same processes as the new opportu- tional shocks-must be as much part of a growth- nites. These risks are present at the international, enhancing strategy as policies that directly expand national, regional and ethnic, community, house- opportunities. Both policies that help the poor build hold, and individual levels. When not kept in up assets such as land, livestock, or savings, and pro- check by open political processes, inequalities and grams that provide income such as well-functioning social tensions lead to civil strife, which can de- public works programs and unemployment insurance stroy decades of progress. schemes are a central part of a strategy to increase The key issue for the early part of the next cen- security for the poor. tury is to bridge this gap between opportunity and Opportunities are created through economic risk. The challenge for policymakers is the design growth, but growth needs to be evenly distributed and implementation of institutions, mechanisms, to benefit the poor. Institutions play an essential role and policies at various levels to harness the po- in ensuring that the poor participate in the benefits tential for poverty reduction, by setting a long-term of growth. Market liberalization produces rewards course that will access global and local opportu- when it occurs within the right regulatory framework, nity and allow broad sharing of the gains from de- but it can contribute to greater inequality and risk velopment, while managing the short-term risks when institutions allow a few to capture its benefits of inequality, vulnerability, marginalization, and and fail to deal with insecurity. Institutional, cul- social dissolution. tural, and political dimensions determine the impact Exploration of these issues startecd in 1998 in a of policies and investments in education, health, and highly participatory process. In terms of broad themes other social services; institutional delivery mechanisms the early thinking is that poverty has to be attacked that fail to reach the poor fail to lead to improvements on three fronts: in their human assets and standards of living, what- • Ensuring empowerment of the poor by increas- ever the budgetary investments in these areas. Thus ing their participation and voice in decisionmak- creating opportunities also means setting up inclu- ing and managing the growth of inequalities sive, responsive institutions. * Providing security against shocks at the household Creating opportunities often goes hand in hand and national levels and protection for those who with reducing gender inequality. Using macro data are unable to share in the benefits of growth on access to education and employment and micro * Creating opportunities by putting in place the data on access to and control of land, labor, and other conditions for sustainable economic expansion, to productive inputs, a recent study found that gender- provide the material basis for poverty reduction. based asset inequality diminishes productivity, out- Voicelessness and powerlessness are intimately put, and growth in Sub-Saharan Africa and that linked to material poverty; participation in local and reducing such inequality would lead to higher growth national decisionmaking not only helps outcomes for and lower poverty (Blackden and Bhanu 1999). the poor directly, but it also helps to improve the qual- The approach taken has clear implications for the ity of development policies and projects. The balance methodological stance of World Development Report of political power is tied to inequality in economic 2000/01. First, the report will have to go beyond stan- power, particularly in the distribution of assets. Re- dard economic analysis and reach out to the insights forms that improve asset distribution may succeed and contributions of other social sciences. This will both in ensuring more voice for the poor and in en- be needed to improve our understanding of the de- hancing social stability. sign and implementation of successful institutions, Harnessing risks, especially for the poor, will be mechanisms, and policies for resolving the tension essential to ensure that opportunities are taken ad- between opportunities for poverty reduction and vantage of. As chapter 2 shows, the way in which the risks associated with these same opportunities. RETHINKING POVERTY REDUCTION: WORLD DEVELOPMEATREPORT2000/01 ON POVERTY AND DEVELOPMENT 47 Second, it will have to complement a macro- and international NGOs through both virtual and actual national-level perspective with a micro-level per- events. A series of conferences and workshops are spective in analysis, bringing in poor peoples' own under way in fiscal 1999, beginning with a discus- experiences of poverty and responses to it. This will sion on the research agenda in the Middle East and be needed to illuminate the nature of risk and vul- North Africa Region at the Mediterranean Develop- nerability, as well as to investigate in detail the ment Forum in Marrakech in September 1998 and processes of sharing in development. seminars at the annual meeting of the Latin Ameri- can and Caribbean Economics Association in Buenos Process: consulting broadly with Aires in October 1998 (box 4.1 summarizes the re- stakeholders around the world sults of these discussions). Workshops have been or will be held on values and norms, social exclusion, World Development Report 2000/01 will go substan- risk and vulnerability, macroeconomic shocks, poverty tially beyond past practice in the extent of the con- and the environment, social capital, and evaluation sultations that will take place during its preparation. of interventions. The process of debating the themes of the report, A first round of virtual consultations with NGOs, gathering diverse experiences, and carrying out re- carried out in September 1998, focused on the draft search at different institutions is not meant simply to work program prepared in May 1997. These con- ensure that the report reflects the thoughts and ex- sultations involved about 80 participants; 52 were pertise of a wide range of development actors. It is from civil society organizations, 40 of them from de- also meant to affect poverty reduction strategies veloping countries. The main insights emerging from around the world by disseminating ideas and the discussions were that poverty is indeed a multi- knowledge, thereby ensuring that the report does not dimensional phenomenon, that institutions do af- remain simply an interesting study but is translated fect policy outcomes, that the poor can be agents of into better strategies for reducing poverty. Pursuing change and must be part of decisionmaking a partnership approach to the exploration of the processes, and that, at the moment, the risks of glob- ends and means of the development process is one alization appear larger than the opportunities. These of the major strategic objectives of the World Bank insights are feeding back into the work of the team. as laid out in the 1998 Joint World Bank-International A second round of virtual consultations took place Monetary Fund Annual Meetings speech of President in the spring of 1999 and focused on a draft outline James Wolfensohn. World Development Report for the report. 2000/01 will be an important instrument in this The consultation process will also directly in- process. A Web site has been set up (http.//www. volve the poor. As a first step a major project has been worldbank.org/poverty/wdrpoverty) to facilitate the dis- launched to explore poor people's perceptions about semination and exchange of ideas. The latest version well-being and their ranking of problems and con- of the outline and regular reports on activities are cerns. It will seek their insights on what affects liveli- posted on the site. hood security and sustainability; trends and changes The report will be produced in time for the Sep- over time related to risk, vulnerability, exclusion, tember 2000 World Bank-International Monetary crime, and conflict; and the importance of institutional Fund Annual Meetings. The synthesis and writing and power relations. The project has two parts. The phase will take place in fiscal 2000, but background first involves a review and synthesis of the results of work, throughout the Bank and outside, began in fis- completed participatory poverty assessments and cal 1999. To launch the process early and allow for other relevant World Bank and external studies that extensive consultations, the staff director and team used participatory methods to explore poor peo- members were named much earlier than is usually ple's perceptions. The second part consists of new the case; although they will come together physically studies using participatory and qualitative research only in the fall of 1999, they are already working as methods, to be undertaken in about 20 countries. The a virtual team. project findings will be discussed and distilled at a The process of consultations leading up to the re- workshop scheduled for September 1999 and will di- port will involve development practitioners, re- rectly feed into the initial drafting phase of World De- searchers, policymakers, and staff of local and velopment Report at the end of 1999. Afterwards, 48 POVERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 BOX 4.1 Find ings of the workshops in Marrakech and Buenos, Aires At a workshop on the proposed content and process for * The political economy of povety reduction is crucial: World Development Report 2000/0 1 held in Marrakech in how to achieveIconsensus and political support for pol- early September 1998 as part of the Second Mediterranean icy interventions. Development Forum, participants emrphasizedithe following * The report ie policymakers with concrete ev- points: ideneonbsprcientvnio. * The report should be forward-looking. * The jreporutneedsyto beinformedt just by the "voices * It should examine governance, corruption, and poverty; of t poor"but also bytheviews of the nonpoor, par- globalization, inequality, and poverty; and war, conflict ticuarly of social elitesthat have a strong influence on resolution, security, and poverty. policy responses. * Ideas, preliminary drafts, and information should be dissem- Durin6g theannual conference of the Latin American and inated in languages other than English (seie http/l nCaribbean EconomicAssociation,alsoin Buenos Aires, a www. worldbank. orglpoverwdrpovertylplan598.htm). panel on W Dvelent Reot2000/01 made these Participants from the Middle East and North Africa would coMMentsi network to carry out research on these themes to feed into * Edution is one of the fewways of redistributing income the report. :i an integrated mart^ ;economy. Achieving the type of In Buenos Aires, a consultation on World Development i refrms necessartoimprove the incomes of the poor re- Report2000/01 with representatives from civil society in the 0 q uiresdificult political changes that few countries have Latin America and Caribbean region (NGOs, triade unions, been able toenforce. business representatives, political leaders, academics) was con- * t is citical that discussions about government interven- ducted during the Forum on Poverty, Inequality, and Vul- tionfcuson thequality ofpublicspending (and on the nerability on October 19-20, 1998. Participants emphasized ; ole of intions) and include solid cost-benefit analyses. the following points: I both events participants a expressedastrong interest * More emphasis is needed on problems of discrimination in remaini ng ivolved in adialoe with the World Devel- and exclusion-race and gender were mrIentioned opment Rrt team throughout the process of writing the explicitly, report. there are plans to work with civil society groups that ization and the new world economy on social mo- have expressed an interest in translating key findings bility, risks, poverty reduction, income inequality, or the draft report itself for direct discussion with poor and social exclusion. In addition to regional back- communities. ground papers, country-level conferences are being Within the Bank the World Development Re- organized in some countries (for instance, in India) port's focus on poverty, as part of the broader re- to link the Bank's program of interventions to the visiting of the development process. is being used strategy proposed in the report. for regionally focused work by each of the re- This process of consulting extensively and draw- gions. In Sub-Saharan Africa the focus is on ana- ing from the expertise of a wide range of develop- lyzing poverty dynamics-one of the major gaps in ment actors, while not aiming for the unachievable our knowledge for this region-both in countries goal of building consensus among all those con- experiencing large increases in poverty (for exam- sulted, is intended to build partnerships that will ple, Nigeria and Zimbabwe) and in countries that form the basis for future programs of action. have achieved progress (for example, Ethiopia and Uganda). This work will culminate in a major stock- Note taking report on poverty for the Special Program of Assistance in Sub-Saharan Africa in 1999. In Eu- 1. A major study by the Operations Evaluation De- rope and Central Asia, the focus is on the rela- partment of the World Bank is evaluating the impact of the tionships between policy reform, institutions, and prescriptions of the 1990 poverty reduction strategy, and inequality and risks. In the Middle East and Xorth will provide valuable input into the World Developmnent Africa the focus will be on the impact of global- Repor. Annexes Annex A. Summaries of Completed Poverty Assessments, Fiscal 1998 This annes summarizes completed fiscal 1998 poverty Building capacity. The lack of access to primary assessments for Bangladesh (update), Chad, Chile, Dji- data on poverty in Bangladesh has been a serious., bouti, India (update), Kazakhstan, and Rwanda (up- long-standing hurdle to more detailed poverty analy- date).1 These summaries were prepared by the World sis. Official poverty estimates have been shrouded in Bank country teams and are based on completed some controversy because independent analysts have poverty assessment reports (see Annex C for full re- never been able to fully replicate the estimates, ex- port titles and document numbers). amine the strengths and weaknesses of the official methodology, or suggest alternative estimates using primary data. Recognizing these problems, in late 1994 Bangladesh World Bank staff undertook a collaborative, capac- ity-building initiative with the Bangladesh Bureau of Counting the poor and making the poor count Statistics (BBS) to help enhance the 1995-96 House- hold Expenditure Survey (HES), train BBS staff, im- Counting the poor. Reducing poverty is the cen- prove basic data analysis, and publish an abstract. This tral development challenge in Bangladesh. This initiative has also led to a series of analytical papers poverty assessment answers several basic questions using the 1995-96 and earlier HES data. This report about counting the poor. Who are the poor? How is part of this process. Work is still underway to numerous are they? Where do they live? What are the mainstream poverty analysis into public policy de- characteristics of poor households? Has poverty de- sign, implementation, and evaluation. The Bank is clined? Has inequality increased? now assisting BBS with the implementation of the 1999-2000 HES. Making the poor count. As importantly, the poverty assessment addresses several questions about how to Important findings and policy conclusions make the poor count in the choice, design, and im- plementation of public policies and programs whose Poverty measurement has been put on a sounder aim is to reduce poverty. These questions are more footing. The BBS has adopted the cost of basic needs difficult: What is the relationship between growth and method for estimating poverty incidence, which is inequality? Is this relationship different for rural and preferable to the official methodology used in the past. urban areas? Does education reduce poverty? How Using primary data from successive rounds of the HES much does the poor benefit from. increasing public between 1983 and 1996, the poverty assessment es- spending on health and educationl? Are households timates the incidence of poverty over time calculated that own more land less poor? Do area characteris- according to the cost of basic needs method. Two sets tics such as rural infrastructure affect the incidence of of poverty lines identify the very poor (lower poverty poverty? How cost-effective are safety net programs? line) and the poor (upper poverty line). Where do microfinance programs Slt within a poverty reduction strategy? Do they reach the poorest? How Poverty has declined in the 1990s, but the re- well do NGO services in education and health com- maining challenges are massive. Both the lower and pete with public and private services? the upper poverty lines indicate a statistically signifi- 50 ANNEX A cant decline in poverty after 1991-92. The incidence contribution of faster industrial and service growth of the very poor declined from 43 percent of the pop- to poverty reduction should be quite high. ulation in 1991-92 to 36 percent in 1995-96; the inci- dence of the poor declined from 59 to 53 percent (see THE GAINS FROM EDUCATION AND OTHER HOUSEHOLD Table A-1). Although poverty has declined in both rural AND REGIONAL CHARACTERISTICS SUGGEST AREAS FOR POL- and urban areas, rural poverty is still higher than ICY EMPHASIS. Apart from broad-based growth, tar- urban poverty. Reducing the poverty of the very poor geted investments in the human and physical capital living in rural areas-still at 40 percent of the rural pop- of the poor can reduce poverty and limit inequality. ulation in 1995-96-remains a massive challenge. Which investments should have priority? The poverty assessment provides some partial answers to this RISING INEQUALITY HAS REDUCED THE RATE OF POVERTY difficult question. Education and land ownership are REDUCTION. The decline in poverty observed in the key determinants of living standards. The gains from 1990s contrasts with the stagnation of poverty in the education are high and have persisted over time. 1980s. Why was overall poverty reduction so slow Higher education has the largest impact in urban or nonexistent over the 1980s? This complex ques- areas. Land ownership matters more in rural areas. tion requires considerable inquiry, particularly since The returns to education, as measured by a house- average GDP growth was roughly around 4 percent hold's per capita consumption, are similar for the and exceeded the declining population growth rate. household head and spouse. Differences in poverty The poverty assessment shows that part of the ex- between geographical areas depend more on dif- planation is rising inequality. Depending on which ferences in area characteristics than on differences in poverty measure is used, one-fifth to one-third of the the characteristics of the households living in those potential poverty reduction from growth may have areas. This finding suggests that investment policies been lost because of higher inequality. The higher aimed at poor areas will reduce poverty. Occupation, inequality associated with growth in Bangladesh too, affects living standards. In rural areas, for ex- does not imply that growth should not be pursued. ample, the gains from switching from the farm to the To the contrary, faster growth is needed if poverty non-farm sector are positive and large for the poor, is to be reduced faster, because the net effect of implying that developing the rural non-farm sector growth on poverty reduction is positive. But in ad- holds considerable potential for poverty reduction. dition to faster growth, efforts to limit rising in- equality are required. Over the period 1991-92 to Public expenditures and safety nets reducepoverty, 1995-96, inequality rose the least with agricultural but their targeting and efficiency must be improved. growth, and as a result the net elasticity of poverty The share of expenditures in the Annual Development with respect to growth was the largest in agriculture. Program devoted to social sector spending has more Assuming these elasticities hold unchanged in the fu- than doubled since the early 1990s and is expected ture, growth in agriculture would tend to reduce to increase further in the years ahead, especially the poverty and limit inequality more than identical share devoted to education and health. The poverty growth in industry and services. Industry and services, assessment reviews the performance of public services however, are likely to grow much faster than agri- in these two areas. The case for substantial public ex- culture as they have done in the past, and the net penditures to education and health is strong on ex- TABLE A.1 Bangladesh: Headcount indices of poverty with the cost of basic needs methods, 1983-84 to 1995-96 (percentage of population below the poverty line) Very poor (lower poverty line) Poor (upper poverty line) 1983-84 1985-86 1988-89 1991-92 1995-96 1983-84 1985-86 1988-89 1991-92 1995-96 National 40.91 33.77 41.32 42.69 35.55 58.50 51.73 57.13 58.84 53.08 Rural 42.62 36.01 44.30 45.95 39.76 59.61 53.14 59.18 61.19 56.65 Urban 28.03 19.90 21.99 23.29 14.32 50.15 42.92 43.88 44.87 35.04 Note: See Annex for the definition of the poverty lines. The figures for 1995-96 are also reported in BBS (1997e). Source: World Bank staff estimates. 51 ANNEX A ternality and equity grounds. While public expendi- Bangladesh.2 Its findings outlined above suggest five tures on health appear to be somewhat better targeted pillars for a poverty reduction strategy-accelerating to the poor than public expenditure on education, there economic growth; promoting education for the poor, is much scope for improvement in increasing the particularly primary education, and particularly for quality of and access to such services. Govemment pro- girls; investing in poor areas to take advantage of grams, such as Food for Work, Vulnerable Group De- strong location effects on poverty reduction; improved velopment, Test Relief, and Rural Maintenance are well targeting of public expenditures and safety nets to reach targeted. A detailed assessment of Food for Education, the poor better; and forming further partnerships with the fastest growing program, shows that it raises pri- NGOs to reach the poorest and not-so-poor in ways mary school attendance and is cost-effective as mea- designed to make a stronger attack on poverty. sured by its long-term impact. But it is not as well Discussions with stakeholders, NGOs, the gov- targeted as the other programs, and improvements in ernment, poverty researchers, and other donors in targeting and internal efficiency would further raise its Bangladesh arising from this report will help to build social returns. Investments in the program's growth will support for an action plan and more detailed policy have to be balanced with the need to improve the over- and institutional changes for faster poverty reduction. all quality of primary education. In line with the capacity-building emphasis of the World Bank's country assistance strategy for Buiilding consensusfor a poverty reduction Bangladesh, these discussions will also help to build strategy consensus on the institutional capacity required to mainstream poverty analysis in policy design and im- This report is part of a long-term process of capacity plementation. BBS will field the next Household Ex- building and mainstreaming of poverty analysis in penditure Survey in 1999. This will provide the TABLE A.2 Bangladesh: Priority poverty indicators 1983-84 1985-86 1988-89 1991-92 1995-96 Poverty Lines Upper Poverty Line Headcounta 58.50 51.73 57.13 58.84 53.08 Lower Poverty Line Headcounta 40.91 33.77 41.32 42.69 35.55 Memorandum Item GDP Per Capita (US$) 147.8 154.5 183.6 210.34 261.56 Short-term Income Indicators Real Wage Index (1970=100) General 90 95 107 107 114 Agriculture 70 80 89 93 97 CPI (1987=1 00) Rural Terms of Trade 74.3 91.3 120.3 148.4 163.6 1960 7970 1980 1990 1996 Social Indicators Share of Public Expenditures for Basic Social Services in GDP 0.8 2.0 Gross Primary Enrollment 47.0 54.0 61.0 79.0 96.0 Female 26.0 35.0 46.0 93.0 Male 66.0 72.0 75.0 100.0 Under-five Mortality (per 1,000) 131.6 95.1 74.6 Immunizationb DPT 0.2 69.0 91.0 Measles 0.3 82.0 96.0 Child Malnutrition 70.1 71.0 68.0 Female-to-Male Life Expectancy 94.7 96.6 98.2 101.6 102.7 Total Fertility Rate, 6.7 6.9 6.1 4.2 3.4 Maternal Mortalityd 3,000 850 Note: In most cases, data correspond to the year mentioned; where unavailable, data for the closest year have been used. a. The upper poverty line is the cut-off for the poor; the lower poverty line is the cut-off for the very poor. b. Percent of children under 12 months. c. Births per woman. d. Per 100,000 births. 52 ANNEX A opportunity and the means to further refine our un- in basic education and health, especially for girls, derstanding of the determinants of poverty and the would further improve the well being and agricul- conditions under which households in rural and tural productivity of producers over the long-term. urban Bangladesh can most easily escape poverty. These positive prospects are founded on the gov- ernment's determination to focus public expenditure on poverty reducing investments (on roads, health, Chad and education), to adopt the needed policies, and on the expected development of petroleum resources. This assessment focuses on rural development be- cause more than four-fifths of the population and an Extent and distribution ofpovertv even larger proportion of the poor live in rural areas. The more specific focus is on agriculture since more The background to poverty is the physical size of the than four-fifths of the economically active population country that may lead to isolation, the sometimes- works in agriculture and because agriculture is the harsh natural environment and the diversity of the most likely source of improvements in incomes that environment. The southern or Soudanian zone is would benefit the poor majority. Finally, it focuses ecologically part of the wet Congo basin; the main on women, who contribute much of agricultural cash crop is cotton, but farmers raise large crops of labor, because they are an important source of im- peanuts, millet and sorghum. The dry Sahel zone lies provements in agricultural incomes. at the middle of Chad; the main activities are cattle The assessment finds a pattern of severe poverty herding and farming of cereals such as millet and ber- in the available indicators, whether of living conditions bere (a form of Sorghum). The Saharan zone lies to (potable water, toilets, housing), food security (nutri- the north and the main economic activities are dates, tion, famine), survival, morbidity, education, or of camel herding and transport. In 1993, nearly half the household assets (such as plows), incomes, and ex- population lived in the Soudan and half lived in the penditures. The pattern is often one of poverty that Sahel, while only 4 percent lived in the Sahara. is spread across the major regions, although several Poverty is defined in economic terms as household aspects such as famine, primary school enrollment and expenditures or revenues that fall below a preset the proportion of women-headed households differ poverty line. The best available source of information across regions. Moreover, women in every region are on the extent and characteristics of poverty when this particularly poor in terms of education, health and land. assessment was prepared were the indicators of liv- The assessment concludes that Chad has excel- ing conditions, health, education and family structure lent prospects for emerging from severe poverty and from the 1993 Census. Information from the Census of sustaining improvements in public well being was reinforced by annual government data on edu- through the efforts of poor agricultural producers of cation and health as well as data on rural conditions livestock and crops. A main avenue of action to from the Ministries of Rural Development, Agriculture raise producer prices would be investment in reha- and Livestock and the Office Nationzal de Developpe- bilitation of roads to reduce transport costs of agri- ment Rural. Fortunately, research has shown that cultural products. This would be reinforced by a these indicators are closely correlated with expendi- further avenue of actions to reduce costs to agricul- ture-based measures of poverty. tural producers, particularly taxes imposed by local However, the paucity of information did limit the and traditional leaders and those imposed illegally on measurement of poverty in Chad. There has not transport. A parallel measure would be to reduce the been a national survey of household expenditures and implicit charge on cotton producers, who receive a there does not seem to have been a national survey fraction of the world price of seed-cotton, through of nutrition. The assessment does, however, make use cautious liberalization of the market. Higher incomes of a 1991 household survey for N'Djamena. More im- would then allow agricultural producers to make portantly, the assessment includes a brief summary some of the investments (in livestock, and agricul- of results from the Department of Statistics and UNDP tural equipment) necessary to increase their pro- survey of household expenditures in four large pre- ductivity, so that further increases in production fectures in 1995/96 that became available during the become possible. Higher and sustained investment final editing. The assessment also draws heavily on 53 ANNEX A the considerable volume of studies of by academi- tonchad. More important, perhaps, cotton is the cians, donors and development consultancies, and on major contributor to farm cash income because it re- field visits. mains the main commercial crop in Chad. Available data show that Chad is very poor rela- A poverty reduction strategy must nevertheless tive to the needs of its residents and relative to its consider other crops and livestock as well. Cotton is neighbors in sub-Saharan Africa. GNP per-capita, for produced only in the cotton zone of the Sudan and instance, was US$180 in 1995 compared with US$490 even in the zone, only about one-third of the planted for sub-Saharan Africa. The mortality of children area is in cotton. Surveys in the zone indicate that under one year in 1995 was 117 per thousand live on average, cotton accounted for about two-fifths of births compared with 92 for sub-Saharan Africa, 29 gross farm revenue in 1995, while sorghum, ground- percent of the population had access to safe water nuts and peanuts contributed almost half of the total. compared with 47 percent for sub- Saharan Africa and Livestock is a dynamic source of growth of exports the illiteracy rate in 1993 was 90 percent, compared and also a store of wealth. with 50 percent for sub-Saharan Africa. With regard to crops, one of the most funda- Poverty is severe in every region if one judges from mental facts is that yields are low, even when com- the indicators of housing quality and access to potable pared with similar parts of Sub-Saharan Africa (Figure water and type of toilet, but there are differences. Parts A-1). The other fundamental fact is that producers re- of the Sahel suffer from repeated famines and school ceive a relatively small share of the revenue from their enrollment is relatively low. Moreover, the Census re- products, as measured by the share of producer ports a striking shortage of men in the most eco- prices in final consumer prices (Figure A-2). This lim- nomically active age groups in parts of the Sahel. In its the incentive and the financial capacity to make the Soudan zone, school enrollnent is relatively high, improvements that would improve yields. but several of the health indicators, in particular sur- vival, are relatively weak. Farmers in the Soudan are Transport costs also very poor in terms of agricultural implements such as plows and carts. The poverty rate in four large pre- Physical isolation is one of the causes of the wide gap fectures was about two-thirds in 1995-96, according between producer prices and final consumer prices. to a preliminary calculation using the recent Depart- Some areas within Chad become enclaves during the ment of Statistics-UNDP survey. This rate was defined rainy season. There are pockets of surplus produc- as the percentage of households with food expendi- tion, where foodstuffs sit in storage because they can- tures that fell below the amount needed to purchase the biologically necessary calories. FIGURE A.1 Cotton fiber (lint) yields in Chad, 1996 Agriculturalproductivity Kg/ha 600 Any strategy to reduce this poverty must relax con- straints to agriculture because crop and livestock 500 production are the primary engines of growth in Chad. They contribute about one-half of national 400 income, when food and agriculture-related industry and services are included. For the most part, agri- 300 cultural cash incomes in Chad are derived from sales of cotton, gum Arabic, peanuts, vegetables and ce- 200 reals. Cotton crops and cotton lint have only a 10 per- 100 cent share in the broad definition of agricultural GDP, yet the importance of cotton cannot be denied: 0 cotton contributed 50 percent to the value of exports Chad Burkina Mali Cam- Senegal Benin Togo and 25 percent to government revenues. The cotton Faso eroon sub-sector helps sustain some 345,000 farm house- Note: Yields of lint, not seed cotton. holds and more than 2,000 salaried workers at Cot- Source: International Cotton Advisory Committee, July-August 1996. 54 ANNEX A FIGURE A.2 drive a wedge between the price retained by farm- Chad: Producer shares of crop prices ers and herders and the price paid by the final buyer. Kg/ha The result is to reduce final consumption of Chadian 600 agricultural products, to discourage production and investment in agriculture, and to harm farmers by low- 500 ering their incomes and consumption. 400 Consumer price The main official taxes on agriculture are the 400 UDEAC tariffs on imports of agricultural inputs such 300 as fertilizers and tools, although these have been ad- Producer price justed downwards over the past several years. There 200 Producer price is also a statistical fee, a complimentary tax on agri- cultural equipment and diesel vehicles and rural in- 100 vestment fund and research taxes. The operations of Cottonchad lead to an implicit 0 l . , . ' 1 1 ' " charge on the small cotton farmers. Cottonchad is a ma- 1995 1996 1995 1996 1995 1996 1995 1996 1995 1996 Berbere Millet Rice Soybeans Corn jority state-owned company that also has a private shareholder, the CFDT, which manages much of the Source: CILSS 1995; SIM 1996; Abt Associates Analysis. company under contract. Cottonchad has the exclusive legal right to buy unprocessed Chadian seed-cotton not be sent to consumption centers and there are from producers, mostly poor small-farmers. It is also pockets of food shortage, where additional food a vertically integrated monopsony that gins and then cannot be delivered for months. For example, a markets the processed cotton lint through its own net- major producer of grains and known as the poten- work. Cottonchad buys at a price fixed in advance of tial "breadbasket" of Chad, the Salamat, has long been the growing season, offering in effect a price guaran- largely unreachable during 6 to 8 months of the year tee and also provides fertilizer and pesticides on credit. because of poor roads and high rainfall. The issue of direct concern for the assessment is The main reason for the persistence of these en- the extent to which Cottonchad transfers resources claves is the low density of Chad's road network com- from the small producers to its owners and managers pared with many other Sub-Saharan countries. The by paying a price for seed-cotton that is below the 1972-82 civil conflict left Chad with only 30 km of competitive market price. A comparison of fluctua- paved roads in 1986. Since then, successive road re- tions in the price of seed cotton and cotton lint sug- habilitation projects have improved the road net- gests that Cottonchad has not passed on most of the work, but much remains to be done: today, Chad still benefit of the devaluation of the CFAF to producers has very few paved roads, moreover, many unpaved (Figure A-3). roads are in poor condition. Despite the central government's efforts, local Several studies in Chad link poor road infra- governments still continue to levy taxes on produc- structure to high transport costs and the high trans- ers and traders of agricultural products, though the port costs to high consumer prices and low producer magnitude is reportedly less than in the past. For ex- prices. The conclusion is that without improved ample, prefecture-level officials generally impose a roads, transport costs will probably not be brought tax on movement of cereals and sometimes other pro- low enough to improve the commercialization of agri- duce from one prefecture or even one market to an- cultural products from enclaved regions and to re- other --in spite of repeated instructions in 1993 and duce food insecurity. again in 1994 and public announcements via radio and press that this illegal procedure would not be tol- Charges on agriculturalproducers erated. Furthermore, local traditional and civil au- thorities also collect informally imposed 'taxes.' Another major constraint is the cumulative charge to Despite the central government's efforts, local agricultural producers. Charges are imposed by the governments still continue to levy taxes on produc- central and local governments, traditional leaders, civil ers and traders of agricultural products, though the servants, and armed groups. Together these charges magnitude is reportedly less than in the past. For ex- 55 ANNEX A FIGURE A.3 culture. In many local traditions, for example, there Chad: Cotton prices, 1983-97 are women's crops as well as women's tasks. More- CFAF per kilo over, the traditional allocation of activities by gender 1200 results in a constraint on time. Rural women are tra- A000 ditionally responsible for household chores and 1000 / \/ maintenance, the preparation of food, transport of fuel 800 wood and water, for much farming, the education of Cotton fiber (international) children and family health care. Women work longer 600 hours than men and have less control of their time. This may crowd out time spent in agriculture and 400 lower their productivity. Limits on women's right to inherit land and to con- 200 Seed cotton (Chad) sume the product of the land may further constrain women's productivity. There is a basic tension in the 1983/ 1985/ 1987/ 1989/ 1991/ 1993/ 1995/ legal system because national law recognizes 84 86 88 90 92 94 96 women's rights but traditional law sometimes denies those same rights. Lack of rights may then discour- age women's investment in land improvement, agri- ample, prefecture-level officials generally impose a cultural equipment and livestock. It may also lead to tax on movement of cereals and sometimes other pro- inefficient decisions about which land to work and duce from one prefecture or even one market to an- what agricultural methods to use. other --in spite of repeated instructions in 1993 and Another constraint is limited access to formal again in 1994 and public announcements via radio credit, which limits working capital and investment. and press that this illegal procedure would not be tol- Rural women have access to informal sources of erated. Furthermore, local traditional and civil au- credit and to credit in kind; to accumulate cash sav- thorities also collect informally imposed 'taxes.' ings, women sometimes organize into informal, ro- Several studies have focused on the role of ille- tating, savings associations. gal road charges, or bribes, on raising transport costs Women's general capacity for productive labor within Chad. In recent years, the Chadian government may be constrained by a combination of recurrent and donors have joined forces in an effort to stamp intestinal disease and malaria, poor nutrition, tightly out illegal payments and harassment of passengers and spaced childbirths and female genital mutilation. Un- transporters at internal road barriers. The President is- fortunately, however, little is known about how sued a decree in June 1993, banning road barriers and women's general health is linked to their produc- illegal searches of passengers and transporters through- tivity in agriculture. The female disadvantage in ed- out the territory and there were subsequent efforts to ucation, as measured by school enrollment and suppress roadblocks. However, illegal charges have literacy rates, more clearly constrains women's pro- two root causes that show no sign of disappearing: ductivity. Several econometric studies document one is that low-paid paramilitary forces find in ille- the link between female education and productiv- gal payment a convenient means to supplement in- ity in agriculture. come. The second is that the victims, transporters, often themselves initiate bribes to avoid paying heav- Education as a constraint to rural incomes ier fines for such violations as lack of proper trans- port documents, unsafe vehicles, desire for quick Education may significantly constrain agricultural release at rain barriers and smuggling. productivity by limiting the ability of both women and men to prevent disease, make use of extension and, Women's contribution to agriculture most importantly, to respond to new technologies, markets and regulatory systems. Evidence from Asian Women contribute much of agricultural labor and studies suggests that a literate farmer has signifi- make some production decisions so that constraints cantly greater productivity than an illiterate one. on women reduce household incornes from agri- Moreover, a survey of econometric evidence suggests 56 ANNEX A that educating rural women increases agricultural lent and the prevalence of HIV is increasing. An ad- productivity by making them more receptive to new ditional concern is female genital mutilation, which technology and increases the chances that other is widespread and which poses a risk to women's women will adopt the technology. health, as its physical and psychological effects on Only half of Chadian children, and a third of girls, girls and women can be traumatic and can affect their enter school. A dropout rate of 20 percent and a rep- reproductive health. etition rate of 35 percent suggest high levels of inef- Almost 60 percent of the local health zones now ficiency in the system. Lack of public funds constrains have a functional health center but only about half the ,he supply of basic education, but there is also a lim- population of these zones really have access to basic ited demand for education outside the towns, partic- health services because of distance, lack of roads, ularly in the Sahelian and Sahara regions, and also for etc., and only about 35 percent of the population has girls. Lack of demand is often a response to the poor access to modern health services. The quality of the quality of education. The primary curriculum, calen- services provided is often mediocre or poor because dar and structure may be poorly adapted to Chadian of insufficient equipment, lack of water, an irregular needs: the school calendar, for example, is based on supply of medicines, and above all, a shortage of that of France and a Monday to Saturday school week, qualified personnel. The government has committed rather than on local agricultural cycles. Curriculum in- itself to a National Health Policy that emphasizes ac- cludes theoretical content that cannot be covered in cess to primary health care and a reduced concentration the time allotted and there is insufficient focus on basic of health sector management. Its budget expendi- literacy and numeracy skills and health. A major qual- tures on health have been increasing, and reached 7.5 ity problem is that schools are often closed or teach- percent of the budget in 1996 compared with 3.9 per- ers are not present since they are sometimes not paid. cent in 1988. Since 1989, external aid has represented Teachers who are not well paid or paid irregularly feel more than three-quarters of all health expenses of the obliged to look for activities to supplement their earn- public sector. Even with external aid, total government ings or even to abandon their posts for more lucra- spending on health only reached a relatively low tive employment. Textbooks in rural schools and US$4.50 per inhabitant in 1996. instructional materials are almost non-existent and there are inadequate desks for students. Rule of law The inefficiencies of the system mean that the op- portunity costs of primary education are high with 9 The rule of law as a constraint to rural development to 10 years needed to complete the primary cycle. is a theme that appears in discussions of education, This appears to constrain demand for the education rights to land, taxation of agricultural producers by of both girls and boys. The opportunity cost of en- local officials and traditional leaders and bribes rolling a girl, for instance, in primary school may be charged to transporters. The weakness of the bank- high since school attendance reduces the time avail- ing and payments system may be related to the rule able for girls to do household chores. If they stay of law, since it is difficult to safely transport cash and home they can fetch firewood and water, prepare appears related to enclavement resulting from the lack meals and baby-sit, which frees their mothers for food of roads and telecommunication. The difficulty of or revenue-generating work. making payments increases the costs of trade in agri- cultural goods and inputs and the weak banking Health system means that there is nearly no formal private credit for agriculture. The health situation in Chad remains precarious de- spite the efforts of the Ministry of Health during the An action plan to reduce poverty past few years. In 1993, according to the Census, life expectancy at birth was only 50 years and mortality The assessment proposes a number of feasible av- rates, especially of pregnant women and infants, enues to poverty reduction, starting with actions to were high. The principal causes of mortality and increase agricultural productivity and competitiveness morbidity are infectious and parasitic diseases. Tu- by reducing the costs of transport and of modern in- berculosis, leprosy and poliomyelitis are still preva- puts. The most important measure to reduce costs 57 ANNEX A could be to upgrade the road network through re- tain what tariffs are actually being imposed on imports habilitation. A program of road rehabilitation and of fertilizers, so that tariffs actually paid may be rela- maintenance that would improve road conditions, that tively high. The available evidence points the need to would open the principal enclaves and connect food reduce informal taxes on transport and taxes imposed surplus regions to markets, would be a win-win by local and traditional leaders. The establishment of proposition because it would reduce agricultural law and order on transport routes and among elements prices at final markets and increase prices to pro- that sometimes collect illegal charges on roads would ducers, while improving food security. contribute to the expansion of agriculture. Specific road construction and rehabilitation pro- It appears that there is considerable potential to jects normally need to be justified by an economic improve agricultural incomes through easing the cost-benefit analysis in which the internal rate of re- constraints on women. Some options are increased turn reaches a minimum threshold. Traditional meth- research into food crops raised by women and prepa- ods of economic cost-benefit analysis often fail to ration of female-focused extension messages. Better demonstrate acceptable rates of return because the access to labor-saving technology for household and road network carries very little traffic on average, so agricultural tasks such as fetching water and wood that the direct economic benefits from reduced ve- would increase productivity by easing time con- hicle operating costs are low. On the cost side, the straints. Another option is to encourage women to absence of suitable building materials (such as lat- form affinity groups to obtain credit or save for eco- erite and gravel) in many areas of Cha(d and the need nomic activities. The most fundamental actions would to haul such materials over long distances, makes con- be to improve basic health and education. struction rather expensive. To assure that schools are open and that enroll- Nevertheless roads provide numerous benefits ment expands the government would need to increase which are not easily captured by traditional cost-ben- funding. The government would probably not be efit methods. It is difficult to measure the extent to able, even with the expected petroleum revenues, to which new or rehabilitated roads will stimulate agri- sustainably fund a satisfactory increase in primary en- cultural trade over the medium-term. Some types of rollment. Community participation will therefore be roads would also provide access for most of the year important to complement public funding and to as- to regions that are isolated by seasonal rains. More- sure that schools operate and that quality improves. over, roads can be justified in terms of food security Increased funding by itself would not sustainably by connecting the capital to famine regions and by con- improve enrollment and educational achievement: necting enclaved grain producing regions to markets. that would require an improvement in the quality of Also, roads help to provide access to basic education education. To this end, the government could reduce and health services that are essential to reduce poverty. the length of the primary educational cycle, and Road projects should therefore be justified through a benefit from informal or multi-class approaches so broad and inclusive approach to estimating benefits. that learning can actually take place in the village. A major step toward reducing rural poverty would Moreover, it could harmonize the school calendar with be to increase the share of the world market cotton seasonal cycles in the demand for agricultural labor. price received by the farmer through cautious lib- Some specific steps to increase girls' enrollment eralization of the cotton market associated with would be to provide separate latrines, assure that new greater investment in community participation for school building lowers distances walked to schools, input supply and cotton marketing. T his should be and establish better security for girls through edu- facilitated by government fulfilling its commitment, cation and supervision of teachers. in the Letter of Development Policy of the SAC II, The government should redevelop its education to lift the legal monopoly of Cottonchiad on the pri- strategy, in consultation with civil society and with mary marketing of seed cotton, ginning, and fiber donors, to assure that it is using its resources effi- and cottonseed marketing. ciently. Such a strategy should propose ways to: (i) Formal taxes do not appear high enough, by them- reduce the length of primary schooling; (ii) select, selves, to significantly discourage recovery of agri- train, recruit, and employ teachers; (iii) increase the culture. A qualification is that, because of difficulty in role of Associations Parents-Eleves in school and communicating the tariff code to agents, it is uncer- teacher management; (iv) improve the primary school 58 ANNEX A curriculum and develop Chadian text books; (v) re- icant decline in poverty. The poverty headcount and duce adult illiteracy with large scale adult literacy in- poverty deficit were halved during this period. While terventions; (vi) improve the progression rate from 41 percent of the population lived in poverty in primary to secondary school; and (vii) estimate the 1987, by 1994 this figure fell to 23 percent. The in- costs of a basic education program of primary and cidence fell from 13 percent in 1987 to 5 percent in adult literacy through 2015. 1994. The report confirms that high economic growth To improve public health, the government could is strongly and positively correlated with declining pursue its policy of decentralizing authority and re- poverty; during the sub-period 1992-1994 when eco- sources in the health system. The government would nomic growth fell, there was also a deceleration in also need to continue increasing resources devoted the rate at which poverty declined. to the sector, although donors are likely to continue Second, for the poorest decile group, the slow- to play an important role. These actions would need down in growth from 11 percent in 1992 to 4 per- to be complemented by training of more personnel cent in 1994, and the accompanying rise in and their posting in the regions. Women suffer rel- unemployment from 4.5 percent to 6 percent actu- atively high rates of maternal mortality and this could ally meant a small increase in this group's poverty. be reduced by training of more qualified personnel, The reduction in unemployment to rates around 4.5 better access to antenatal care and a reduction in ado- percent in 1995, by the same token, is likely to have lescent pregnancy. The first step to reduce the health reduced poverty for this group more recently. consequences of genital mutilation would be an ed- Third, the report confirms that income inequal- ucation campaign among religious and civil leaders. ity in Chile is high by international standards. How- Development of petroleum resources offers Chad ever, sustained high growth has resulted in a a unique opportunity to accelerate the implementation significant reduction in poverty, despite this high of a poverty reduction strategy. The government has level of inequality. Chile's success in reducing head- indicated its determination to use additional resources count poverty during 1987-1994 rivals the perfor- for poverty alleviation and in particular rural develop- mance of countries such as Korea, Indonesia, and ment (including infrastructure), health, and education. China; it should be noted, however, that these other These public resources will be shared increasingly countries are believed to have experienced increases with decentralized entities and the execution of pub- in inequality in the process of high economic growth. lic expenditures will rely increasingly on these decen- Fourth, the reduction in poverty during 1987-1994 tralized entities and on non-governmental institutions. has benefited almost all groups classified as vulner- able at the beginning of the period. While growth ob- viously helped those among the poor who could Chile work, poverty reduction policies have benefited even non-workers: for example, older, poorly educated, Despite rapid economic growth and continued efforts male and female household heads in both rural and by the government to improve social equity, a com- urban areas experienced significant declines in the mon perception in Chile is that some groups have probability of being poor. lagged behind and income distribution has worsened. Fifth, education is an important determinant of Using the best available analytical and statistical tech- labor earnings and hence of household income. Dif- niques to examine the developments since 1987, this ferences in educational attainment account for almost report has sought to examine the validity of these one-third of overall income inequality, and are by far concems. It has also tried to identify plausible policy the largest single explanatory factor. options that the government could follow towards re- The report confirms that the income tax is pro- ducing both poverty and income inequality, while not gressive and the VAT is regressive-the combined ef- compromising the growth potential of the economy. fect of which is a tax structure that is largely inequality-neutral. On the other hand, if public spend- What we have confirmed ing on pensions is excluded, the analysis confirms that total social expenditure has substantial redis- First, the report has confirmed that even within a short tributive impact, most of which is accounted for by period of time (1987-1994), there has been a signif- basic education and health care. 59 ANNEX A Finally, overall, Chile's policy in recent years of Limitations of our statistical analysis growth with equity has been demonstrated as effective in reducing poverty without exacerbating income Besides labor earnings and other income (including inequality. As with virtually any program, there is imputed rents of owner-occupied dwellings), mea- scope for improvement at the margin; this report at- sures of household income in this report include mon- tempts to draw out options for consideration by etary Teamsters such as means-tested old age, Chilean policymakers. disability, and family allowances, and other cash subsidies. Omitted from our income measures are What we have refuted welfare-enhancing improvements such as better ac- cess, lower prices, and higher quality of public util- First, contrary to a popular perception, this report doc- ities, which are believed to have improved more for uments that income inequality has not increased poor than for non-poor households. during 1987-1994; in fact, there is a slight improve- More importantly, this income measure does not ment in the income distribution evidenced by a small include the value of government transfers in kind, decline in the Gini coefficient from 0.55 to 0.53. though public housing, health, and education pro- Second, the report finds that inequality in labor grams, which together were 6.5 percent of GDP and earnings has declined considerably in the last 30 percent of total government expenditures in 1994. decade. While the Gini coefficient for labor earn- Non-availability of household level data for all four ings in Gran Santiago between 1960 and 1987 rose survey years for these transfers made this adjustment from 0.43 to 0.58, it has fallen steadily since then impossible. Since the incidence of public spending on to 0.46 in 1996. these programs is regarded as progressive, this results Third, refuting growing concerns about precari- in a greater underestimation of the real income of ousness of employment and earnings, the report poorer households than of the non-poor. Annex 1 in finds that job and income security has increased Volume II of the report does, nevertheless, report some substantially since 1987. Average expected tenure on of the amputations for public education and health a job has risen from 47 to 55 months between 1987 services by household quintiles made by MIDEPLAN. and 1995, and the average duration of unemployment The findings discussed above, and their policy im- has been halved from 5.5 to 2.8 months. plications, must be considered keeping this caveat in Fourth, allaying concerns that rapid growth has mind: our estimates probably overstate the level of exacerbated regional income disparities, the analy- poverty in Chile and understate its reduction over the sis shows a convergent pattern. All indicators of period examined. This shortcoming in data collection inter-regional dispersion of income show a reduc- will need to be corrected for policymakers to be able tion between 1987 and 1994: real per capita income to attain a finer degree of targeting of social programs of the poorest region (region IX) grew by 10 per- than already achieved, viz., if the goal of "supertar- cent, which is twice the national average for the pe- geting" of public spending is to be achieved. riod. Reassuringly, reductions in poverty have been shared by all regions (except region XI, which has Implications for labor policies only 0.6 percent of the country's population) and these gains have not been restricted to Gran Santi- The labor market has served Chile's poor well in re- ago. While Santiago did better in reducing headcount cent years. Labor earnings have contributed more to poverty ratios, other regions matched its perfor- poverty reduction than non-labor income (mainly mance if indicators that measure the depth of poverty transfers). While household inequality measures have are used instead. not shown an improvement, inequality in labor earn- Finally, less encouragingly, while educational at- ings has fallen significantly since 1987. While these tainment of all income groups increased between developments may not justify inaction, they do call 1987 and 1994, the difference in years of schooling for restraint in changing labor market policies. between the richest and poorest groups has in- Labor legislation that would make dismissals more creased. While the years of schooling for the poor- costly for employers is likely to be counterproduc- est 20 percent increased by 0.8 years, the schooling tive and could, as in other countries, actually increase of the richest 20 percent increased by 1.3 years. unemployment. 60 ANNEX A An unemployment insurance system is largely ing programs should be scientifically evaluated (with unnecessary given existing severance benefit legis- a control group). lation, the low duration of unemployment, and the existence of means-tested transfers. Steps to introduce Implications for tax and social expenditure an unemployment insurance system should be care- policies fully evaluated keeping in mind the distortions such a system can create. Because introducing dual VAT rates or exemptions While the minimum wage could be increased, on for goods comprising relatively large shares of the the basis of the analysis in Annex 3 it appears that budget of poorer households will raise administra- further increases in the ratio of the minimum wage tive costs, further study of the tax system is warranted to per capita income is likely to result in greater un- despite the regressivity of the current VAT system, if employment and could actually worsen poverty. looked at in isolation. Despite the payoff in terms of inequality reduc- Implicationsfor education and training policies tion that a greater reliance on income taxes would yield, the current strategy of concentrating on clos- Compared with measurement of the effects of im- ing loopholes has the advantage of being equity-en- proved access to schooling (reflected in years of hancing and increasing tax collection efficiency. schooling), the benefits of improved quality of school- To better monitor the incidence of social spend- ing are difficult to measure. Our estimates, however, ing as a whole, the classification of what constitutes indicate that increasing access to primary and sec- social spending should be re-examined. Spending on ondary education yield smaller increases in earnings items such as pensions of public servants, military and relative to both improvements in education quality at police should not be lumped with those with clearly these levels, and increased access to higher education. identified social policy objectives. But even access to higher education by the poor is best Chile's past successes in improving the efficiency facilitated by raising the quality of their primary and of targeting notwithstanding, there is considerable secondary education. These findings suggest that ef- scope for further improvements. This can be done forts to improve education quality in municipal and both by allocating a greater fraction of social spend- private subsidized primary and secondary schools ing to programs that have had better targeting records would help reduce poverty and inequality. Combined (e.g:, primary education and health, and cash bene- with institutional measures to improve the quality of fits, where the poorest 40 percent receive 60 percent instruction, reducing public spending on tertiary ed- of expenditures) and improving targeting efficiency ucation (62 percent of which goes to the richest one- of programs such as housing subsidies, which are es- third of individuals) and allocating it to sentially untargeted. quality-enhancing initiatives at lower levels could be The report finds that despite extensive individual an effective medium-term strategy to reduce inequal- and household-level controls, simply living in poorer ity of education, and hence of earnings. comunas (districts) remained an important factor in The socioeconomic status of vocational-technical stu- accounting for the likelihood of being poor. Strong dents is considerably lower than those enrolling in hu- location or "neighborhood effects" on poverty deserve manistic-scientific education programs, and secondary further study, because they expand the set of potential vocational education has lower economic returns than measures to deal with residual or hard core poverty. general education because it is more expensive. Until These findings and policy implications are based improvements in primary education quality for the on five studies that were commissioned for this re- poor have been achieved, however, measures to in- port. These studies, in turn, analyzed recent changes crease the relevance and quality of secondary vocational in income distribution and poverty, provided a pro- schools are likely to be equity enhancing. file of poverty, isolated critical policies that influence Based on international experience, if designed poverty reduction, examined selected labor market well, such training programs may help carefully se- issues and policies, and evaluated social policies in lected groups find jobs but are not likely to reduce Chile. A brief synopsis of the results of these stud- earnings inequality of employed workers. Because ies, together with a more detailed discussion of these these programs are always expensive, Chile's train- topics are available in Volume I of the main report. 61 ANNEX A The findings and the analytical and statistical tech- Poverty profile niques employed in all five studies are contained in Volume II. To formulate and implement a comprehensive strat- egy to combat poverty, it is essential to identify the poor. The analysis for this poverty profile is based Djibouti on (i) a priority survey-type (EDAM, 1996) com- plemented by (ii) a participatory poverty assessment In Djibouti the problem of poverty reduction remains (PPA, 1996). EDAM excludes three highly impov- significant. The country's social indicators show erished and vulnerable groups - nomads, the home- that the gaps in social well-being are enormous less, and those living in temporary structures. and recent economic performance has been poor. Therefore, the household survey results on poverty, Over the medium term, the government must de- inequality, and standards of living are likely to velop new sources of growth in the economy. This present an over-optimistic picture of the real situ- report analyzes the nature and extend of poverty in ation in Djibouti. Djibouti, and the distinguishing characteristics of the Poverty in Djibouti is high. In 1996 about 45 per- poor. The final goal is to propose a set of measures cent of Djiboutians were living in households with to reduce poverty in the medium to long-term and adult equivalent expenditures below the level nec- minimize the short-term social effects of structural essary to provide basic needs. Ten percent were es- adjustment. timated to live in extreme poverty, i.e. they could not Djibouti's main assets are its strategic location afford to buy the food basket necessary to maintain on the Horn of Africa, its harbor and associated in- a minimum level of caloric consumption. Including frastructure, and its access to relatively rich fishing the homeless and nomads in the analysis would in- waters. However, Djibouti's strategic location is crease the percentage of households living in poverty also the cause of its problems. Djibouti attracts and extreme poverty. populations from other countries in the region be- Poverty is more widespread and deeper in Dji- cause it offers relative peace and stability, and em- bouti's rural areas and in the urban areas outside Dji- ployment opportunities with salaries paid in hard bouti-ville than in the capital city. The incidence of currency. Immigrants and refugees place a tremen- extreme poverty is more than seven times higher in dous burden on already-weak sociaL services and rural areas (45.1 percent) than in Djibouti-ville (6.2 economic security. percept). This is probably because a household liv- Djibouti is a lower-middle-income country, with ing in the capital can take advantage of safety nets average per capita income of US$780 in 1993. How- derived from the presence of a commodity market ever, estimated living standards are distorted by the and services, and job market opportunities not avail- high cost of living and well-paid civil servant com- able in rural areas. munity. Most Djiboutians are at a subsistence level, Refugees, nomads, the homeless and those living and living standards are more comparable to those in temporary structures, and the street children are in Ethiopia than those in non-African countries with highly impoverished and vulnerable groups. Al- similar per capita GDP levels. GDP per capita in 1993 though refugees living in camps benefit from food prices fell from USS1,505 in 1,978 to US$854 in 1991. aid, and free health care and education, they face a The situation further deteriorated in 1991 as the difficult situation and describe themselves as having economy experienced a series of exogenous shocks: lost everything, even their identity. War and poor rain- (i) a sharp inflow of refugees from Somalia and fall have changed the nomad's normal patterns of Ethiopia, (ii) the 1992-94 armed conflict between transhumant behavior. Nomads cope by engaging in the Issas and the Afars, and (iii) increased competi- small-scale border trade, and receiving help from rel- tion from Eritrea in providing. transport services to the atives living in Djibouti-ville. The street children, na- region. Concurrently, the Government of Djibouti tives of Somalia or Ethiopia, live in dire poverty. They began to incur increasing deficits (internal and ex- left their countries because of war or poverty, but have ternal) that led to the signing, in April 1996, of a one few chances to break the cycle of poverty: unable year stand-by arrangement with the IMF designed to to attend school, they cope by taking odd jobs, and stabilize the fiscal situation. eating out of trash cans. 62 ANNEX A What characterizes the poor in Djibouti level in 1996, the net enrollment rate among the poor- est quintile was 53.1 percent, 25.3 percentage points The poor are characterized by absence of purchasing lower than the rate in the top quintile. At the secondary power, low human capital accumulation, and low stan- school level, the gap widens to a five-fold difference dards of living. The poor (and the very-poor) differ from between the poorest and richest quintiles. the rest of the population in socioeconomic and de- Health. Health indicators are below regional stan- mographic characteristics, and employment status. dards. Life expectancy at birth is low at 49 years. The infant mortality rate, at 1 14 deaths per 1,000 live 1) UNEMPLOYMENT AND POVERTY. Unemployment is births, is one of the highest among Middle-Eastern and extremely high and increases the risk of the house- Sub-Saharan countries. Diarrhea and malnutrition to- hold being in poverty. The average unemployment gether constitute the leading cause of death among rate is close to 45 percent, with the most disadvan- children under 5. Respiratory infections (associated taged economic groups having unemployment rates with chronic malnutrition) is the second leading cause higher than 60 percent. The anticipated demobiliza- of mortality. High maternal mortality, estimated at 740 tion of up to 9,000 soldiers will add to the ranks of deaths per 100,000 live births, can be attributed to high the unemployed. fertility rates, anemia caused by malnutrition, and Most households in Djibouti are able to escape the widespread practice of Female Genital Mutilation. poverty because of their earnings from employ- In addition, endemic problems - Tuberculosis, malaria, ment. Fifty-five percent of the very-poor live in cholera, and AIDS- affect the population in general. households where no member is employed. These Malnutrition. Child malnutrition is widespread. results underscore the importance of job creation in Among children under 5 years of age, 26 percent are Djibouti. chronically malnourished (stunted), and 13 percent suffer from acute malnutrition (wasted). Considering 2) THE POOR LACK HUMAN CAPITAL. Social indicators that the first 4-5 years of life are critical to the future in Djibouti are below regional standards. The poor's development of the child, malnourished children welfare is negatively affected by low levels of edu- will be more likely to have low educational attain- cation, poor health status, and malnutrition. ment and achievement. Education. Poverty, illiteracy, and low educational Water and sanitation. Appropriate sanitation fa- attainment are strongly correlated. The literacy rate is cilities and clean water are necessary to reduce the estimated to be approximately 57 percent; but the incidence of diarrheal disease and preserve the en- poor, women, and old people are at a significant dis- vironment. Lack of access to water is clearly corre- advantage in literacy skills. The poor have an illiteracy lated with poverty. Households belonging to the rate of 64 percent, but even in the most affluent 20 per- richest quintile are seven times more likely to be con- cent of Djiboutian households, 37 percent of individ- nected to the ONED (the network that supplies tap uals are illiterate. The gender gap is wide: while 73 water directly to houses) than households belonging percent of the men older than 10 are literate, only 45 to the poorest quintile. A large percentage of poor percent of the women in the same group are literate. households need to purchase water from a water Households with lower levels of education are at truck. This water costs four times the price paid by a higher risk of being poor: a one-year increase in richer households connected to the ONED network. the average education of the household reduces the Sanitation indicators are sub-standard. In Dji- probability of the household being poor by six per- bouti, more than 40 percent of households have no centage points. proper drainage system for used water. The situation Enrollment ratios show that educational attainment is particularly dangerous in the densely populated is not likely to improve substantially in the near future. areas of Djibouti-ville where most of the population Children begin school relatively late. Approximately one lacks proper sanitation services. out of four children starts school after the age of nine and they begin leaving school when they are 14. Less A strategyforpoverty reduction than 50 percent of those aged 14-20 are in school. En- rollment rates are low and children from poor house- This report shows a strong correlation between holds are the least likely to be in school. At the primary poverty and unemployment and between poverty 63 ANNEX A and low investment in human capital: policies aimed (a) reduction in wages paid by the formal (espe- at creating new jobs can have a substantial impact in cially public) sector; and reducing poverty. Investment in economic and social (b) policies to facilitate entry into the labor market and infrastructure (transport and telecommunications, creation of new jobs. water, sanitation, and housing) will have beneficial ef- fects both in the short- and in the long-run. In the 2) INVESTING IN HUMAN CAPITAL. Increasing human short-run these projects will generate employment and capital is a key component of a poverty-reducing and allow some households to escape from poverty and growth-promoting strategy. Progressive liberalization improve their living standards. In the long-run, these of the economy requires higher education levels to projects will play a fundamental role in increasing the support modernization of the productive sectors. In- level of human capital and enhancing productivity. vestment in human capital provides direct economic These investments are essential to enhance the com- benefits and improves the general welfare of indi- petitiveness of Djibouti in the provision of services to viduals and of society. In addition, it generates im- the countries in the region, notably Ethiopia. They also portant cross-effects between sectors. For example, could contribute to transforming Djibouti into a re- higher education levels are associated with reduced gional trade and financial center. female genital mutilation (FGM) and fertility. More ed- A poverty-reduction strategy for Djibouti should: ucated and healthier mothers produce healthier, bet- 1) raise grovw th and generate employment; ter nourished and more educated children. 2) increase investment in human capital; and Education. Pro-poor education policies should 3) develop targeted interventions for the poorest. emphasize two areas of intervention: (a) expanding all children's access to education, par- 1) GROWTH AND EMPLOYMENT. Given Djibouti's pre- ticularly to primary and lower secondary, with spe- sent welfare distribution, the country would have to cial emphasis on the poor and the very-poor; sustain consistently high growth in GDP per capita and to reduce poverty effectively by the year 2015. As- (b) expanding girl's access to secondary education. suming a 1.5 percent per capita growth rate, an es- Given their effect on health and fertility, targeted timated 38 percent of the population would be living subsidies for girls' education will have very high in poverty by the year 2000; and more than twenty social returns. years from now, 20 percent of the population would Improving the quality of and the access to pri- remain in poverty. mary and lower secondary school requires addi- Djibouti is characterized by: (i) low labor force tional funding to the sub-sectors. Given the current participation, (ii) exceptionally high unemployment financial constraints, Djibouti should explore all po- (particularly long-term unemployment), and (iii) a tential sources for improving the education system, dual labor market (with the formal sector paying among these: (i) community-based management; wages well above the informal sector). The high and (iii) enlisting the private sector to mobilize levels of unemployment do not cause a reduction in additional resources. wages because the high public sector wages, to- Health. Poor health status is both a cause and an gether with other labor market restrictions, limit any effect of poverty. In Djibouti the health of children downward adjustment in private sector formal wages, and mothers is among the most serious issues. giving rise to a dual labor market and high unem- Measles, respiratory infections, diarrhea, and FGM ployment rates. cause many deaths that could have been prevented. Given the country's strategic location, level of in- These health problems are amenable to low-cost frastructure, and limited natural resources, the service and effective interventions. sector (with particular focus on the Ethiopian mar- The Djiboutian health system is too centralized and ket) is the main source of potential grovvth. In this too focused on curative care. Currently, health ser- regard, high labor costs (both wages and social se- vices are free and available to everybody. This leads curity payments) represent a major obstacle to the to excess demand compared with available resources. competitiveness of Djibouti's private sector. Any Since households are more willing to pay for cura- strategy to cope with this situation will need to con- tive care than for preventive interventions, a reallo- centrate on two elements: cation of resources from curative to preventive care 64 ANNEX A could significantly improve the efficiency and equity pation, could broaden the provision of sustainable, of the health care system. low-cost housing. The housing policy in rural areas Effective strategies to reduce the incidence of must be considered within a wider framework, in- FGM include: cluding the development of economic activities and (a) incorporating FGM treatment and counseling into improvements in social service delivery, transporta- health programs for women and children, and or- tion, and infrastructure. ganizing information and education programs for Efficiency ofpublic spending. The review of gov- community and religious leaders. ernment expenditures and subsidies on health, ed- (b) enforcing sanctions against individuals perform- ucation, housing, and flour suggest that their present ing the practice. levels and patterns are not conducive to effective Fertility is high in Djibouti across the population. poverty reduction. Most programs are inefficient, The total fertility rate is 5.8 children per woman. The i.e., the share of benefits received by the poor is high fertility rate, young childbearing, and close birth smaller than their proportion of the population. The spacing have detrimental effects on the health of government has already taken steps to correct this women and children. Programs aimed at increasing problem by substantially reducing housing allowances female access to education (together with IEC cam- to public employees. A further increase in efficiency paigns) are the most likely to succeed in (i) promot- could be achieved by: ing the use of birth control techniques and (ii) (a) reallocating resources from university education increasing the ability of couples to space their children and hospitals to primary and secondary education, optimally. health dispensaries, and preventive health; Consumption ofkbat. Djiboutians are major con- (b) promoting policies to increase the participation sumers of khat. Indeed, 55 percent of households of women and the poor in secondary and tertiary have at least one khateur. This practice poses enor- education (e.g., targeted scholarship programs); mous social, economic, and health problems. Two (c) implementing altemative mechanisms to finance ter- key areas of intervention are: tiary education. To promote a service economy, it (a) increasing taxation of khat. Besides contributing is important to have an adequate number of uni- to reducing the demand for khat, such policy versity graduates. However, given the tight budget would generate additional tax revenues that could constraint the government faces and the pro-rich be earmarked for social programs; and focus of the university scholarship program, the (b) organizing IEC campaigns to raise awareness of government should means-test scholarships and the negative effects of using khat. support alternative financing mechanisms; Water and housing conditions. Djibouti is facing a (d) reassessing the baguette subsidy. This subvention serious challenge in trying to provide its population with is inefficient: the poorest 40 percent of popula- potable water. This will require policies on the supply- tion receive only 31 percent of the bread subsidy. and demand-sides. On the supply-side, the introduc- The wealthiest 20 percent of individuals gain 28 tion of new water harvesting methods is extremely im- percent of the subsidy. portant. Regarding the distribution of water, in the There are three main problems with Djibouti's so- urban areas, the strategy should focus on installation cial security system: (its very limited coverage of the and management of standpipes and minimization of population; (ii) a bias toward the middle and upper the loss of the water network. For rural areas, the classes; and (iii) the relative generosity towards those strategy should include a program of digging new who participate. To solve its financial problems, Dji- boreholes and wells, and rehabilitating and maintain- bouti's social security system needs to broaden the ing existing ones. On the demand-side, the government tax base. This requires a system of incentives to in- should introduce an appropriate pricing system. duce contributors to pay contributions on their ac- Housing conditions are poor in Djibouti. A reha- tual earnings, and to start a process of gradual bilitation program would improve basic living stan- "formalization" of the activities that are currently dards and generate employment. Squatter-upgrading outside the scope of taxation. Some possible actions or slum-improvement programs, that include (i) af- are: fordable standards, (ii) cost recovery, (iii) improved (a) linking all social security benefits to the level of institutional capability, and (iv) community partici- contribution; 65 ANNEX A (b) reducing the level of contribution paid by em- India ployers and employees; (c) wage-testing family allowances. Poverty profile 3) INSTRUMENTS TO REACH THE POOREST. Poverty is The 1998 Poverty Assessment builds on the 1997 so widespread in Djibouti that a long-term policy Poverty Report and focuses on using public expen- to address it should be accompanied by short-term diture to expand the poor's access to quality educa- income transfer mechanisms to alleviate the poor's tion and health care. Based on the poverty profile most urgent problems. A poverty reduction program presented in the 1997 Poverty Assessment, poverty in should target (i) inhabitants of the slums of Djibouti- India is declining, but slowly, and remains wide- ville and the homeless, (ii) populations of the dis- spreadx35 percent of the population (37 percent rural tricts outside Djibouti-ville and nomads, and (iii) and 31 urban) lived below the poverty line in 1993- demobilized soldiers. These groups could be 94 (latest year for which household survey data are reached if they participated in labor-intensive con- available). India has the largest concentration of poor struction of public infrastructures. These activities people in the world particularly in rural areas where not only generate employment opportunity for the almost three out of four Indians and 77 percent of the most vulnerable groups but also are consistent Indian poor live. The report also found that stagger- with long-term development objectives. They could ing as the overall numbers remain--240 million rural (i) directly impact Djiboutian standards of living poor and 72 million urban poor--they do not tell the (e.g., improvement of water and sanitation infra- full story. Social indicators of well being--health, ed- structures, construction of a water network, reha- ucation, nutrition describe a country which has made bilitation and construction of schools and substantial gains against widespread deprivation over dispensaries; development of a system for the col- the 50 years of its independence but has not achieved lection of urban waste), (ii) resuLt into higher lev- the momentum needed to bring the great majority of els of human capital, and (iii) increase Djibouti's its poor into the economic mainstream. While India competitiveness as a regional service center for has succeeded over the last four decades in doubling the Horn of Africa (e.g. construction of major roads life expectancy and reducing infant mortality, its death and other infrastructures). rate for infants under five remains one of the highest Income-generating activities would also contribute in the world. Communicable diseases and prenatal and to provide income support to the poor. maternal mortality, which account for 12.5 percent of This Report highlights the potential of the fish- the annual deaths of rural women aged 15 to 45, ing sector. Training young fishermen and providing cause about 470 deaths per 100,000 population in them with financial support can reduce unemploy- India-a rate four times that of China and 2.5 times ment, improve Djibouti's trade balance, and reduce that of the world as a whole. Tuberculosis alone kills malnutrition. 500,000 people a year. Half of all children under five are malnourished and, because their mothers often are Monitoring poverty as well, one-third of all newborn babies are under- weight. India's rates of malnutrition among children The design and implementation of poverty-ori- andwomen are among the highest in the world. And ented social and economic policies in Djibouti are fewer than half the children from poor households are made very difficult by the scarcity of relevant and in school, reducing their prospects of escaping poverty. reliable statistical data. To effectively implement Two-thirds of all women and two fifths of all men re- poverty reduction policies, it is essential to de- main illiterate. Half of all girls aged 5-14 years do not velop a permanent and integrated poverty moni- attend school. Among the most disadvantaged social toring system. The poverty monitoring unit would groups, the ratios are even worse: literacy rates of just be responsible for: 19 percent among scheduled-caste women and 46 per- (a) implementing household surveys; cent for men. (b) monitoring the impact of social and economic poli- The report found that in general gender, literacy, cies on the poor's living standards; landownership, employment status, and caste are (c) disseminating all poverty-related information. closely associated with poverty. Thus, an illiterate rural 66 ANNEX A woman, a member of a scheduled tribe or caste, a of premature death. And because the poor are less person who lives in a landless household or is de- likely to be educated and must often use shared pendent on wage earnings, all face a significantly sources of water and surface water (lakes, streams, higher than average risk of poverty. A 1994 survey and ponds) without adequate sanitation facilities, of rural households indicates that the incidence of they are more frequently exposed to illness and pre- poverty was 68 percent among landless wage-earn- mature death. The study further finds that India's pri- ers, 51 percent for members of scheduled castes and mary health centers failed to deliver the care needed scheduled tribes, and 45 percent for members of to reduce infant mortality, but did not find any sig- household in which no-one was literate. And where nificant correlation between child survival and the poverty is deepest, female literacy is exceptionally availability of public health facilities. low. In general, where social indicators reflect little progress, poverty has also been very slow to decline. Safety nets The incidence of poverty and the poor's access to so- cial services varies considerably from state to state Traditional anti-poverty programs are funneling many and even within states in India. of their benefits to the non-poor. According to data from the 1993-94 National Sample Survey (NSS), 76 Public expenditures percent of the wealthiest rural households, for in- stance, are likely to take advantage of the subsi- Given that the government is pursuing many strate- dized prices for food under the Public Distribution gies to reduce poverty, this report focused on the use System while, at the opposite end of the scale, fewer of public expenditure to help the poor gain better than 70 percent of the poorest households benefit access to public goods and services. In general, pub- from food subsidies. The poor do participate in rural lic subsidies will benefit the poor most when the items public works and to a lesser extent in credit programs subsidized are used disproportionately by the poor (IRDP), but all three schemes (PDS, IRDP, and pub- relative to the non-poor. Benefit incidence analysis lic works) remain loosely targeted. was used to examine the extent to which the poor have benefited from public spending on education, Poverty strategy certain health services, and some of the most im- portant safety net programs (food, public works and Since the bulk of the required investments to build credit scheme programs). Findings indicate that pub- human capital comes from public rather than private lic spending for education and health are not effec- sources in India, and yields many externalities, it is tively reaching the poor. In education, the poor who important that poverty-reduction strategies focus in- do not send their children to school as well as those creased public spending on expanding the poor's ac- who do not keep them there are not benefiting from cess to quality education and health care. The report public spending on equal footing as compared with suggests that targeting government spending to pri- those who are better-off, as reflected in the enormous mary education, reducing communicable diseases, difference in the enrollment and attainment levels of improving water and sanitation, and reducing house- the two groups. Fewer than half the children from hold insecurity through public works programs are poor households enroll, and when they do, only one priority actions to reduce poverty. To avoid increas- in five of them completes the eight-year cycle of basic ing the already large fiscal deficit, increased funding education. In contrast, wealthier households in all for these activities would come from the reduction in areas do send their children (both sexes) to school, costly and untargeted subsidies that are currently the and over 80 percent of them complete grade 8. Poor source of large fiscal imbalances and microeconomic girls are only one-eighth as likely to complete grade distortions. 8 as their female counterparts among the well-to-do. In health, the poor face a disproportionately higher Education. A strong political commitment on the risk than the rich of falling sick, particularly from in- part of the government is needed if the educational sta- fectious diseases. They are more likely to lose their tus of poor children are to improve. Increased spend- children before they reach the age of two. Poor ing on education inputs (teachers, classrooms, members of scheduled castes run even higher risks textbooks, and instructional material) alone will not be 67 ANNEX A enough to improve enrollment and educational at- ative to the non-poor, public interventions in these areas tainment of the poor. Improvements in the quality of will achieve their biggest impact on the poor. The net schooling is also important. This requires fundamen- cost to the government of extending water and sani- tal reforms to change the incentive framework within tation facilities to poor areas may not be very large, since which teachers, school officials, bureaucrats, and politi- willingness to pay for these goods is usually quite cians have been operating. The accumulated experi- high and could cover the extension of the system. Third, ence under the District Primary Education Project, the analyses have shown that health education concern- LokJumbish, and the experiences of states such as Ker- ing basic hygiene, the value of better nutrition, and pre- ala and Himachal Pradesh should help provide the in- ventive care such as public campaigns against tobacco formation necessary to underpin such deeper use and for the use of appropriate measures to avoid educational reforms, particularly in the seven states with contracting HIV-AIDS and other sexually transmitted the highest incidence of school dropouts or non-en- diseases, is an important part of encouraging behav- rolLment. Relevant findings from other theoretical and ioral changes needed for long-term improvements in empirical research provide guidance on reforms that health outcomes. Fourth, public subsidies to hospital are likely to improve outcomes, of which some are al- care can play an important redistributive role as long ready underway in many states in India. These include as referral systems are reformed to ensure that access decentralization of control over the provision of school- is based on need rather than income and social status. ing to local authorities, direct parental involvement, Because the rural poor must often meet the financial competition through school choice, and community in- burden of medical emergencies through debt, distress volvement. Madhya Pradesh, for example, is by far the sale of real assets, or reductions in food or other im- most advanced in decentralizing school management portant consumption items, subsidizing hospital treat- to local institutions such as the Panchayati Raj Institu- ment will help alleviate the burden on the poor. In tions, with a resulting increase in school enroLLment and addition to assistance through publicly managed hos- retention among children of undetprivileged groups. pitals, the poor could also be served by public fi- The process and the potential for bringing edu- nancing of private provision of services in rural areas cation closer to its users and beneficiaries through de- (with an appropriate system of incentives and moni- centralization has generated significant optimism but toring); by a major effort to increase the quality of care also considerable caution. The chief fears reflect through training, changes in incentives, and regulations; concern that decentralization cou:Ld perpetuate, and or by community-based insurance schemes. India's quite possibly increase, regional disparities in school policy makers will need to evaluate the option of sub- quality, and social and economic inequities in access sidizing hospital treatment against other alternatives in to good schooling. Lowered funding levels for edu- order to support appropriate services that will help re- cation also might be a problem if local control serves duce poverty. Further research is needed, here again, as a pretext for diminishing central responsibility. to underpin the necessary reforms to both public and private health systems in India. Health. Effective health programs must comple- ment education in raising the potential productivity of REFORMING ANTI-POVERTY PROGRAMS. The marginal in- labor. Public expenditure on health as currently con- cidence analysis suggests that expanding access of stituted is likely to have only a limited redistributive im- the poor to the public works programs (preferably pact. This report identifies four priority areas of public through reallocation of spending away from less ef- spending that will have the greatest impact in im- fective poverty programs), would benefit them most. proving the health of the poor. Firt, combating com- Priority would be to improve the effectiveness of pub- municable disease and expanding public health lic works by better targeting the genuinely needy and interventions (see below) would deliver substantial making these programs fiscally sustainable. Effective gains from public health spending, particularly for the targeting need not be exclusive targeting, as some poor. Second, improving access to safe water sources level of spill-over to the non-poor is unavoidable if and sanitation facilities and vaccinations would help political support for such programs is to be main- reduce infant and child mortality and thus reduce fer- tained. Such targeting could be done by setting the tiLity and improve maternal health. Because these are wage rate at a level which is no higher than the pre- activities in which the poor are vastly underserved rel- vailing market wage, with willingness to work at this 68 ANNEX A wage rate being the only eligibility criterion. The to problems with the measurement of income in the scheme could also be geographically targeted to FBS. Second, given the proportion of the population poor areas. In addition, early results from the im- that can be considered poor, it should be clear that plementation of the Targeted Public Distribution Sys- the problem of poverty cannot be eliminated by pub- tem (TPDS) in UP and Bihar indicate that the poor lic transfer programs. Indeed, the social safety net has seem to be benefiting more than the non-poor from shrunk substantially since independence under the the retargeted subsidies on foodgrains. If this find- combined impact of falling revenues and the need to ing is generalized across India, then targeted food- maintain fiscal discipline. Strong and sustained growth grain subsidies could supplement public works will therefore be key to poverty reduction in Kaza- programs to reduce household food insecurity. khstan. Third, given that it is unrealistic to expect a significant expansion of social protection expendi- Statistical system tures in the near future, assisting the poor will hinge on making existing social programs more effective. This There is a clear need to improve the quality and time- should include maintaining an adequate income floor liness of the statistical database necessary for poverty for pensioners, increasing the coverage of unem- measurement, monitoring, and analysis in India. This ployment benefit, and reducing the leakage of child includes not only the data necessary for direct poverty allowances and other social assistance to the non-poor. measurement (national sample surveys, annual con- sumer expenditure surveys) but also related databases Developments in the macroeconomy and the and statistical information available at the sectoral level labor markets (e.g. education and health statistics). The World Bank has set up a program of technical assistance with the Although poverty was not unknown in Soviet times, Department of Statistics, that has laid the groundwork Kazakhstan was one of the least poor central Asian for a series of lending activities designed to modern- republics. However, the sharp contraction in out- ize and strengthen overall statistical capacity in the coun- put since independence has contributed signifi- try. For example, work is currently ongoing in India cantly to an increase in poverty. In cumulative to identify ways to improve the timeliness and quality terms, the decline in output in Kazakhstan since of information collected under the National Sample Sur- 1991 has been of the order of 40 percent. Although veys. In addition, plans are being developed to intro- employment has fallen, the burden of adjustment duce a quarterly labor force surveys, which is crucial has fallen on the price of labor. Wages have fallen in tracking changes in employment patterns in poor by over 50 percent in real terms, with serious im- regions and in key sectors in which the poor are em- plications for household income and consumption. ployed. There are also plans to improve the current sys- Open unemployment remains low relative to the tem of monitoring consumer prices and wage rate scale of output decline. Unemployment (ILO defi- levels. Many of these initiatives are likely to be sup- nitions) stood at around 4.5 percent at the end of ported under the Statistical Systems Modernization 1994, rising to around 6 percent in 1996. Workers Project, which is scheduled to go to the Board in 2000. who are the most severely affected by unemploy- ment appear to be new labor force entrants, those with low education or vocational and technical Kazakhstan skills and, to a lesser extent, women. At independence, Kazakhstan inherited an econ- This report examines the impact of the transition on omy deeply dependent on Soviet supply and trade living standards in Kazakhstan, assesses the scope for networks. The demise of central planning, the loss burgeoning growth to reduce poverty, and makes of established patterns of supply and trade, and the recommendations on how to strengthen the social loss of transfers from Moscow following indepen- safety net. There are three principal findings. First, over dence all contributed to large falls in output. The need a third of the population lived below a 'subsistence to bring inflation under control squeezed the econ- minimum' living standard in 1996. This is significantly omy further. Stabilization efforts were however suc- lower than estimates of poverty based on Family Bud- cessful in reducing inflation which fell from over 1800 get Survey (FBS) income distribution data, and is due percent in 1994 to 29 percent in 1996 and 11 percent 69 ANNEX A in 1997. The aftermath of separation from the Soviet many children or many dependents (whether young Union appeared to have been finally thrown off in or old) tend to be poorer, as do households with un- 1996 when, after declining continuously for five employed members. There is also a distinct correla- years, the economy began to grow again although tion between low levels of education, and technical at a modest pace. Economic growth during 1997 is and vocational training, and poverty. Pensioners are estimated at 2 percent. also found to be at risk of poverty, although they are Another feature of macroeconomic decline has relatively well-protected from extreme (bottom quin- been a fall in the government's ability to mobilize tax tile) poverty. revenues. As a result, tax revenue as a share of GDP The analysis of household expenditures points to fell from 39 percent in 1992 to 24 percent in 1996. Poor the emergence of own-production as a very impor- revenue performance has contributed to dispropor- tant source of food consumption in Kazakhstan. tionate declines in the social safety net, in public in- Over a third of all food consumed on average in July vestment, and the funding of public services. In an 1996 was derived from home-production. In addition effort to keep the budget deficit uncder control, the gov- to growing their own food, individuals engage in ernment has accumulated large wage and cash ben- other informal activity, including working in second efit arrears. At the end of 1996, arrears in wages stood jobs and running their own businesses. There are at 6 percent and arrears in pensions at 2 percent of large, private inter-household transfers in Kazakhstan GDP. Pension arrears were fully cleared in 1997 at the -mostly from family and other relatives-which are cost of considerable belt-tightening in other areas. another important source of income. The growth of informal incomes does not come as a surprise given The profile of the poor in Kazakhstan the sharp fall in main wages and formal social sup- port. But equally, informal incomes significantly This report makes use of the Kazakhstan Living Stan- compromise the ability of the government to effec- dards Survey (KLSS), a representative household sur- tively means-test available social assistance. vey conducted in July 1996, to exarmine the extent and main correlates of poverty in Kazakhstan. Households The role of the social safety net in Kazakhstan whose per capita consumption is less that the Gov- ernment's 'subsistence minimum' (prozhitochnyi min- Kazakhstan's social safety net has undergone major imum) are considered to be poor.3 Nearly 35 percent transformation since 1991. The need to maintain fis- of the population was found to be living in poverty in cal discipline in the context of a sharp contraction July 1996. While by no means low, the estimate is sig- in revenues meant that spending on the social safety nificandy lower than poverty estimates based on income net as a share of GDP fell by over a third from 11.2 distributions generated by the Family Budget Survey (64 percent in 1992 to 6.6 percent in 1996. Although new percent in 1995 and over 80 percent in 1996), largely forms of social protection have been introduced- due to problems with the measurernent of income. In- such as unemployment assistance-there has been come data are often subject to a under-reporting, par- a general decline in program beneficiaries assisted ticularly in the case of income from. own business and by a tightening of eligibility criteria. Real benefit lev- informal activity. While there are differences in rural and els have declined substantially, with the exception of urban poverty rates, the more striking difference in Kaza- means-tested child allowances. Although they have khstan is by region. The poverty rate in the south of increased in real terms in the last two years, pensions the country (69 percent) is nearly twice the national av- remain at less than 50 percent of 1993 levels. erage, while that in the north (9 percent) is many times Pensions dominate social protection expendi- lower.4 In terms of number of poor, nearly two out of tures (over 75 percent), while child allowances come three poor people live in the south or east of the coun- a distant second. The housing allowance scheme, try. This has the strong implication that reducing poverty which is currently quite small, is expected to expand in Kazakhstan is largely a question of improving the op- in the coming years. Despite the general contraction portunities for people in these two regions. of the system, there remain a multiplicity of small ben- The poverty profile highlights correlates of poverty efits which are complex to administer. that are similar to that observed in other parts of the The 'thinning' of the social safety net is evident former Soviet Union (FSU). Large households with from the fact that over 60 percent of poor households 70 ANNEX A in Kazakhstan receive no public transfers at all. This public investment program, and judicious handling is partly related to pure 'gaps' in the social safety net, of likely future natural resource revenues, (vi) fur- and partly due to non-payment of benefits to eligi- ther restructuring and improvement in the quality of ble households due to resource limitations. In the low- health and education services, and (vii) improved gov- est quintile, pensions and other transfers contribute ernance through the development of a professional, to about 20 percent of consumption on average. well-remunerated civil service. Among existing social assistance programs, tar- While growth is a necessary condition for the re- geting of benefits towards the lowest quintiles is duction of poverty in Kazakhstan, it is not sufficient. generally poor. In the case of the means-tested child Whether poverty is reduced by growth is a function allowance scheme, over 50 percent of the program of how income growth is distributed in society, and expenditure leaks to the non-poor (defined as the top the opportunities offered by growth to the poor to 60 percent). Better targeting outcomes have been improve their income earning opportunities. The re- achieved in other countries such as Uzbekistan, Al- structuring that has yet to take place (notably in bania and Chile. Social assistance which is targeted agriculture, but also mining, transportation, utilities, using categorical criteria generally does far worse in and public administration) is likely at least in the short terms of leakage than the means-tested child al- run to swell the ranks of the unemployed. The evi- lowance program as the categories in use are inher- dence from the KLSS on the returns to education sug- ited from Soviet times and do not appear to gests that those with low education or mismatched correspond to the present realities. skills may find it difficult to find work. The social safety net will be critical for these and other groups Reducing poverty in Kazakhstan: growth and who are not well-positioned to benefit from growth. the social safety net While important for maintaining the living stan- dards of the poor, severe constraints have been im- Given the scale of output decline, and the rise in posed on the social safety net by poor revenue poverty, it should be clear that broad-based growth performance and the need to keep the budget deficit that both raises wages and creates employment will under control. Moreover, the transition from a pub- be key to reducing poverty in Kazakhstan. Growth lic pay-as-you-go pension system to a private fully is critical not just for raising incomes but also for pro- funded system starting in 1998 will add a significant viding the resources for a fiscally sustainable safety burden to the budget for many years to come due net to help those who might be temporarily or chron- to the reduction in payroll taxes contributed to the ically disadvantaged. public system.6 A national poverty reduction strategy In general, the prospects for a sustained increase should therefore aim to make social programs more in output over the medium term in Kazakhstan are effective within existing fiscal constraints. This report good, both because it is one of the stronger re- makes five main recommendations in this respect: formers among the FSU countries, and because it is well-endowed with natural resources.5 The World Ensure the adequacy ofpensions under thepen- Bank estimates that the economy will grow modestly sion reform. From a poverty perspective, the strength in the short-run (2-3 percent), picking up speed of the existing pension system in Kazakhstan is that (4-6 percent) early in the next century. However, pensioners do not face increased risks of extreme there are several critical areas that need attention if (bottom quintile) poverty. The 1998 reform provides growth in Kazakhstan is to be sustained: (i) further for a minimum pension of 2,400 Tenge which is development of the legal and regulatory framework about 70 percent of the 'subsistence minimum' and for private sector development, (ii) stronger regula- in keeping with World Bank recommendations. How- tion and oversight of the financial sector, (iii) deep- ever, a key concern is whether the minimum pen- ening of capital markets in conjunction with the sion will be maintained in real terms. A second issue careful management of pension reform, (iv) sound is whether state pensions will be paid on time. It will management of the process of enterprise restructur- be important for the Government to maintain the min- ing to encourage output and productivity growth, (v) imum pension in real terms to protect the living improved management of public resources, includ- standards of poor pensioners, and to keep to its ing an expansion in the level and the quality of the commitment to prevent the emergence of new pen- 71 ANNEX A sion arrears. A medium-term issue under the re- STRENGTHEN THE SYSTEM OF UNEMPLOYMENT COMPEN- formed system is pension adequacy for low-income SATION. Although not directed at the poor, unem- workers, those with short work histories, and women, ployment benefit in Kazakhstan is relatively all of whom typically receive lower pensions from well-targeted. Attention should be paid to increasing systems which rely on personal saving. There are a coverage, including through better provision of in- number of options here for improving outcomes, in- formation to the unemployed and, possibly, through cluding: higher contribution rates, contributions from raising compensation levels. At the same time, rules other social insurance funds to cover periods of tem- governing receipt-especially with respect to job re- porary inability to work (owing to sickness, mater- fusals-should be tightened so as not to escalate costs. nity or unemployment), joint treatment of contributions, equalization of retirement age for men CONSIDER ASSISTING THE UNEMPLOYED THROUGH RE- and women, and the use of standard life tables when GIONALLY TARGETED TEMPORARY EMPLOYMENT SCHEMES. issuing annuities. The Government should develop Given the administrative costs and complexity of the capacity to analyze likely pensionis for workers improving the targeting of individual programs, the with non-standard work-history and other charac- scope for exploiting the self-targeting potential of pub- teristics and develop options to ensure adequate re- lic works programs deserves closer attention. placement rates. Monitoring progress with poverty reduction Improve the targeting of social assistance. Tar- geting of the main programs of social assistance in Whatever improvements are introduced, it will be es- Kazakhstan (e.g. child allowances) has largely relied sential to monitor changes in poverty and targeting on formal income-testing. But given the structure of effectiveness. This report recommends developing the incomes in Kazakhstan, alternative approaches clearly Family Budget Survey into an appropriate instru- need to be developed. One way to imnprove target- ment for such a monitoring exercise. Improvements ing at low cost would be to use additional house- to the FBS being initiated under the World Bank's hold characteristics that are well-correlated with Technical Assistance Loan should be sustained. The poverty, such as the presence of young children or Government should maintain a commitment to track- unemployed members. More sophisticated proxy ing the impact of growth and policy changes on liv- means-testing could also be considered, with atten- ing standards and developing the capacity to tion both to administrative capacity and costs. At- undertake analyses of alternative reform options. tention should also be paid to features of the housing allowance scheme that tend to exclude rural house- Future agenda holds (who are equally needy) and to replacing di- rected subsidies in utilities with targeted cash transfers Finally, this report does not address three types of to those who are in genuine need. policies that will need to complement the strength- ening of the social safety net to improve living stan- RATIONALIZE SOCIAL ASSISTANCE PROGRAMS. In addition dards in Kazakhstan: to main programs which are means-tested, there a (i) Labor market policies designed to facilitate labor number of small programs (e.g. other child allowances, market restructuring, encourage mobility, and local social assistance programs) which are categori- equip workers for changed economic conditions; cally targeted at pre-identified "vulnerable" groups. (ii) Policies aimed at reducing regional imbalances; Data from the KLSS show that these programs largely (iii) Health and education sector policies to maximize benefit the middle income groups. These resources the poverty reduction impact of reduced expen- should be redirected towards those wlho are truly in ditures. need. A simple way to do this would be to continue The first of these is to be the subject of a subse- the existing trend of merging existing programs with quent study to be initiated during 1998. A start has means-tested ones. Where this is not feasible, it would been made on the second, through the World Bank's be worth considering making benefits subject to an in- study of the Kzyl-Orda oblast which is affected by come test, which although it has its limitations, would the Aral Sea crisis. Further work, particularly per- be an improvement over present criteria. taining to the role of the republican government in 72 ANNEX A reducing imbalances, and role of regional develop- scribing the principal changes in the poverty profile, nment policies, is clearly warranted. Restructuring the the report is divided into a section that addresses the health and education systems so as to maximize immediate constraints to recovering incomes and quality and efficiency from reduced resources will be meeting basic social needs, followed by a section on an ongoing challenge over the medium term. A the medium-term prospects for poverty reduction. Bank-sponsored health reform program, which is expected to be approved by March 1999, should be Poverty profile followed closely for lessons for the reform of the med- ical care system throughout the country. Although household budget survey data are not avail- able it is possible to estimate the changes in income poverty levels since 1994 through information on Rwanda changes in the population, GDP, and targeted inter- ventions such as food aid. It is clear that the events The last World Bank poverty assessment for Rwanda of 1994 caused a catastrophic rise in poverty in was based on research completed in 1993 and painted Rwanda, with approximately 70 percent of all house- a picture of declining incomes and contracting eco- holds falling under the poverty line in 1997 com- nomic opportunities for the poor. In April-June 1994, pared to 53 percent in 1993. The depth of poverty has up to one million people were killed from a total pop- also increased, meaning that not only have poor ulation of 7.6 million in a series of genocidal mas- households increased in number, they have also got sacres. In the aftermath of the genocide, two million poorer. Furthermore, there has not been a steady re- people fled to neighboring countries and up to one covery in the poverty situation since 1994: after a million people who had been pushed into exile by small improvement in 1995-96, poverty levels rose previous purges returned to their country. In eco- again in 1997 following the influx of population from nomic terms, Rwanda saw GDP fall by 50 percent in the Democratic Republic of Congo (DRC) in late 1996. 1994, and output has still not regained pre-war lev- The geographical profile of poverty has also changed, els. Rwanda remains one of the poorest countries in with the poorest prefectures before the war remain- the world, with per capita GDP under US$220 in 1997. ing poor in income terms but possessing better access The effects of this degree of social and economic to social services than other areas of the country. The turmoil on the poor are likely to have been substan- characteristics of poor households have also changed tial. Although numerous studies have been conducted dramatically: poor households are now more likely to on reconstruction needs, very little information is be female headed or to lack able-bodied labor. This available on the situation of poorer households since labor constraint poses a formidable challenge to im- the genocide. The impact of the events on the poor proving economic opportunities for these households. is important not only because social protection has In the social sectors, there has been a severe de- been, and continues to be, a goal of reconstruction terioration in health indicators, with infant mortality efforts since the war, but also because poverty levels rising from 8.7 to 13.1 percent and maternal mortal- are linked to political stability. Rising poverty un- ity almost quadrupling since 1990. Other social in- doubtedly played some role in exacerbating social ten- dicators-primary education, access to water and sions leading up to the genocide, and reducing sanitation-have not declined as quickly as income poverty is critical in its aftermath not only as a goal poverty has increased, primarily due to strong gov- in itself but also as a means to improve the prospects ernment and donor support for the social sectors. for social and political stability. Having shown great However, while enrolment rates have not greatly commitment to reconstruction since the wvar, the gov- decreased, there is a serious problem in the quality ernment is now in the process of launching an am- of basic education. Only 32 percent of primary school bitious growth strategy. It is important that this strategy teachers possess appropriate qualifications to teach. benefit all sections of society, including the poorest. A limited survey of schools undertaken for this study The study aims to provide an update on the sit- showed that less than 30 percent of basic teaching uation of poorer households three years after the materials are available to teachers, and children pos- genocide, and to place on the table for debate a se- sess less than 25 percent of the minimum package ries of strategies for poverty reduction. After de- of school materials. This indicates that many children 73 ANNEX A attend school without a teacher who is equipped to or imprisonment and restrictions on mobility; (iii) lack teach, or the basic materials to follow classes or of access to inputs; (iv) increased insecurity of land record their new knowledge. access, and; (v) weak transport and market links. What are the implications of this changed poverty Recommendations to address these constraints in- profile for poverty reduction strategies? First, the clude: (i) an increase in public investment in agricul- sharp increase in income poverty relative to social in- ture, oriented towards quick impact projects to rebuild dicators, combined with the dependence of the ma- rural capital. These may include direct transfers, agri- jority of the poor on agriculture, indicates that the key cultural credit, and rural public works. However, labor to short-term poverty reduction is a rise in agricultural intensive public works should be avoided in regions incomes. This should in itself help reverse the decline that suffer from strong labor constraints, and may not in social indicators, as nutrition improves and house- be able to address the poorest labor deficit households, holds generate more savings for education, but this who will need a more flexible type of income sup- is likely to be more effective if combined with actions port; (ii) rationalizing restrictions on mobility, reduc- to improve the quality and accessibility of social ser- ing the cost of documentation and increasing the vices. Second, a medium-term perspective on poverty speed of processing to enable people to move between reduction must be growth-oriented, but will need to communes for work and trade; (iii) accelerating the ensure that the sources of growth are in sectors where allocation of land that is unoccupied (marshland) or the poor can benefit, and that the poor are equipped under temporary occupation by returning exiles. to access new economic opportunities. Both the Rwanda has also become highly dependent on short- and medium-term strategies must take into ac- food aid since 1994. This risks distorting the market, count the changes in the characteristics of poor house- in particular due to weak seasonal management of holds following the genocide, in particular gender distribution. The food aid program also suffers from divisions and lack of able-bodied labor. targeting leakage, since food for work programs are often inaccessible to the poorest households due to Actions to reduce poverty in the short term their lack of able-bodied labor and need for imme- diate payment. It is recommended that a substantial Based on the poverty profile, it is r ecommended propornion of food aid be monetized as per current that short-term poverty reduction initiatives focus government proposals-with counterpart funds used on the following objectives: (i) enabling poorer for agricultural investment. Food security will also be households to increase their agricultural output and improved by the establishment of public information market their produce at fair prices; (ii) improving the services to disseminate information on crop price quality of primary education and lowering the costs movements, to improve the negotiating power of for the poor, and; (iii) improving access of poor small producers, provide early warning of food households to basic medical care. deficits and manage the importation of food aid. ADDRESSING INNMIEDIATE CONSTRAINTS IN AGRICULTURE. IMPROVING ACCESS TO BASIC EDUCATION AND HEALTH Agricultural productivity was already in decline before CARE. The principal constraints in primary education the war, and now suffers from both structural prob- occur in two areas. On the demand side, low real in- lems and conjectural constraints stemming from the come levels and labor constraints in poor families genocide and population movements. Output has re- have increased both the direct and opportunity costs covered to only 78 percent of pre-war levels (with a of schooling. On the supply side, as discussed above, higher population) and, amongst poorer households, the quality of education-teacher training, availabil- agricultural incomes are frequently insufficient to ity of materials-is a more binding constraint than the cover consumption needs and generate seasonal sav- physical availability of infrastructure. Recom- ings. The immediate constraints to raising agricultural mendations to improve educational opportunities incomes include: (i) the destruction of rural capital, in for poor children in the short term should therefore particular livestock, which both reduces yields and focus on improving quality rather than targeting im- makes households more vulnerable to seasonal inse- mediate increases in enrolment, and include: (i) ex- curity and sudden income shocks; (ii) labor con- panded teacher training programs; (ii) investment in straints, due to the loss of family labor through death teacher materials; (iii) measures to subsidize the cost 74 ANNEX A of school books, and targeted subsidies for the poor- have declined dramatically since the war. Of this re- est households. duced pie, the share of the social sectors in current In health as in education, the major constraint is expenditure has decreased from 35 percent in the mid- the cost of healthcare in relation to the appallingly eighties to around 22 percent in 1996-97. This is par- low levels of monetary income available to poor ticularly destructive as needs have increased-due to households. The cost of care is likely to rise in the reduced income levels households are unable to near future, as the Ministry of Health has recently in- spend the same on health and education as was troduced a policy of 110 percent cost recovery on spent before the war-so higher allocations are nec- medicines. Evaluation of the impact of this policy on essary to overcome this poverty effect. Current ex- poorer households is a top priority, and transitional penditures on health in 1997 were only US$0.6 per measures to extend subsidization, possibly targeted capita, and on education only US$5.62. to the poorest households, should be considered if Both military expenditures and the debt burden utilization rates fall. Other quick impact interventions constrain allocations to the education and health in health efforts to increase the availability of skilled sectors. The former should be addressed by gov- personnel and the reorganization of pharmaceutical ernment in keeping to demobilization targets as far distribution channels to avoid ruptures in the supply as possible given the security situation, and the lat- of basic medicines. ter through rapid negotiations to treat Rwanda as a special case for accelerated debt reduction under the Medium-termframeworkforpoverty reduction HIPC framework, together with external assistance for debt service in the transitional period. The gov- Rwanda will reach the limits of recovery-based growth ernment has already raised allocations on the social fairly quickly, as the remaining factors of production sectors by 30 percent between 1997 and 1998, and that have been idled since the war are brought back assistance in relieving the debt burden should enable into activity. Growth is expected to return to its long- further rapid progress. In order to achieve pre-war term trend between 2001 and 2003. At this point, current expenditure targets of US$11.00 per capita on poverty reduction will rest less on ensuring that the education and US$4.50 per capita on health by 2003, poor are able to recover their economic activities and the total share of health and education in the current incomes, and more on the prospects for rapid and sus- budget would need to rise to 40 percent. Increased tainable growth. To assist in meeting the overall goal investment spending on the social sectors will also of medium-term equitable growth, poverty reduction be necessary in the medium term to finance an ex- objectives should: (i) promote growth in sectors that panded infrastructure program, in particular for sec- will directly benefit the poorest; (ii) ensure that legal ondary and technical education. and institutional frameworks for economic activities favor access by poorer households; (iii) equip poor AGRICULTURE. Medium-term growth in agriculture households with the skills and knowledge necessary is constrained by low productivity, lack of viable in- to access new economic opportunities, and; (iv) pro- vestment opportunities, lack of tradable land rights and vide the social and economic infrastructure necessary adverse incentives for cash crop production. Ad- to support the movement of poorer households into dressing these constraints will require: (i) investment non-agricultural employment, in particular through in fertilizer promotion and the cultivation of higher- improvements in urban infrastructure and services. value-added crops; (ii) careful consultative design of a land reform program to transfer tradable rights MACROECONOMIC ENVIRONMENT. The government has while protecting access of women and the poor; (iii) achieved strong progress in macroeconomic stabi- liberalization of the tea sector, with short-term action lization since 1994, and has implemented a number prior to liberalization to raise producer prices, and; of structural reforms. Prioritizing those structural re- (iv) suppression of the variable tax on coffee. forms that will improve economic opportunities and incomes for the poor remains vital. But the principal PRIVATE SECTOR DEVELOPMENT AND EMPLOYMENT CRE- macroeconomic constraint for poverty reduction in the ATION. Rwanda has strong export potential for selected medium term is the lack of public resources available commodities-both traditional exports such as tea, for investment in human capital. Government revenues coffee and hides and skins, and non-traditional such 75 ANNEX A as fruit and vegetables, cut flowers, pyrethrum and and that funds for AIDS prevention and assistance to textiles. However, many of these, vvhile carrying victims and carers be increased. high potential producer shares, have low employment potential. Thus growth in these export sectors Will URBANIZATION. Urban infrastructure in Rwanda is assist rural poverty reduction but is unlikely to help entirely inadequate to meet even the needs of the ex- the landless poor. Semi-processing of local and re- isting population, and cannot absorb rapid urban pop- gional materials provides better potential for non-agri- ulation growth. The principal constraint is sanitation: cultural employment creation, as does the promotion the Prefecture of Kigali estimate that they currently have of small and medium enterprises for manufacturing capacity to treat only 3 percent of the liquid waste pro- and services. Constraints to private sector develop- duced in the city. Land property rights and registration ment and job creation include: (i) weak domestic pro- procedures also discourage investment in housing. duction logistics; (ii) barriers to international trade, Commercial infrastructure-roads and markets-is lim- including the level of trade taxes and fees, lengthy ited in coverage and badly degraded. In effect, in the freight immobilization and heavy documentation past Rwanda has survived an almost total lack of urban procedures; (iii) strict registration requirements for in- planning due to the small size of its urban population. formal sector businesses; (iv) lack of access to But the country is now at a crossroads: if urban de- medium-term investment credit, in particular for velopment is to be supported as part of the growth and small, medium and micro-enterprises; (v) regulatory poverty reduction strategy, this must go hand-in-hand barriers for women entrepreneurs. with adequate planning and management of urban in- frastructure. It is, therefore, an urgent priority to develop EDUCATION AND HEALTH. Lack of a literate and and implement an effective urban planning frame- skilled workforce will both constrain investment and work; to expand investment in sanitation infrastructure exacerbate inequality, by producing high wage dif- in Kigali, with the aim of extending sanitation services ferentials between skilled and unskilled workers. to all areas of the city, including the poor peripheral Expansion of primary school enrolment is therefore areas; and to improve and expand market infrastruc- vital in the medium term and will require higher in- ture in both Kigali and secondary centers. vestment in infrastructure to maintain the quality of learning. But the key medium-term constraint in ed- THE POLITICAL CONTEXT FOR POVERTY REDUCTION STRATF- ucation is in the secondary school systern, which, with GIES. The Government of National Unity established gross enrolment at only 4.6 percent, provides a bot- in July 1994 has faced a formidable series of chal- tleneck preventing the majority of the poor from in- lenges. These include continued security threats from creasing their skills beyond the most mninimal level. former soldiers and militia housed in camps along the Reforming the boarding school system Lo lower costs border with Zaire during 1995 and 1996, the need to to parents, expanding secondary and technical school ensure adequate legal process against the 130,000 pris- infrastructure, and improving the curriculum are pri- oners suspected of genocide crimes currently con- orities for medium-term poverty reduction. tained in Rwandese jails, the need to accommodate The principal challenges for the health system in over 800,000 returning Rwandan citizens who had left the medium term are; (i) containment of population the country between 1959 and 1994, and the final re- growth, which at 2.9 percent annually constricts the turn and reintegration of 1.3 million people to Rwanda potential for per capita income growth and means from Tanzania and the Democratic Republic of Congo that many poor women are trapped irL too onerous in December 1996. In addition, the new government a cycle of child-bearing and child-care; (ii) prevent- was faced wvith an economy in shambles, a civil ser- ing further increases in HIV incidence, which is es- vice that had lost three quarters of its qualified staff, timated to have risen sharply over the last 10 years and widespread destruction of social and economic to around 11 percent of the population. This is still infrastructure. Under these difficult circumstances, the below the rates in neighboring countries, but is ris- government has shown great commitment to recon- ing fast; (iii) decreasing external dependence in the struction and has succeeded in stabilizing the econ- finding of the health system. It is recommended that omy and reconstituting basic services. more vigorous efforts be made to publicize contra- However, the climate remains very fragile: the in- ceptive methods and the benefits of spacing children, creased insurgency following the influx of returnees 76 ANNEX A to the north-west of the country continues to pose spreadsheets are being prepared to allow analysts and pol- serious security problems; processing the cases of icymakers to simulate poverty measures based on chosen genocide suspects is proceeding slowly; and a high household characteristics and to explore the impact of poli- level of mistrust remains between different sections cies that change these characteristics. of society. Sustainable poverty reduction will rely not 3. The subsistence minimum is based on a consump- only on the developmental interventions outlined in tion basket, 70 percent of which is food and 30 percent this report, but on a successful transition to political non-food goods. The food basket is derived from official and military stability and the development of a more nutritional norms. In dollar terms, it is equivalent to around representative and accountable model of governance. US$70 per capita per month at purchasing power parity (and * around US$40 at the market rate of exchange). Although INSTITUTIONAL FRAMEWORK TO IMPLEMENT THE POVERTY more generous than many poverty lines, the subsistence REDUCTION STRATEGY. Poverty reduction is a multisec- minimum does have the advantage of being recognizable. toral issue, and requires strong coordination to avoid 4. The regions are defined as: NorF-Kostanai, Kok- duplication of efforts or contradictory policies. It is shetau, Pavlodar, and North Kazakhstan oblasts. Center- also important that the government has the capacity Zhezkazgan, Karaganda, Akmola and Torgai oblasts; to monitor monetary and social indicators of poverty West-Manghystau, Atyrau, Aktyubinsk, West Kazakhstan; on an ongoing basis, in order to be able to adjust de- East-Semi Palatinsk, East Kazakhstan, Taldykorgan and Al- velopment strategies as new priorities emerge. It is maty oblasts and Almaty city; Soritb-Kzyl-Orda, South recommended that a coordination structure for Kazakhstan and Zhambyl oblasts. In April 1997, the num- poverty reduction be established, chaired by the ber of oblasts was reduced from nineteen to fourteen. Ministry of Finance and Economic Planning, and 5. Natural resources can be a mixed blessing. The supported by a small technical unit. competitiveness of non-oil sectors of the economy has been found to be a problem in almost every country that has experienced a rapid expansion in foreign exchange earn- Notes ings from petroleum exports. In a recent study covering growth in ninety-seven developing countries during the pe- 1. All but Kazakhstan are IDA borrowers. riod 1970-89, Sachs and Warner (NBER Working Paper, 2. In addition to this report, two other tools have been 1995) found that countries with high export earnings based developed to facilitate the use of the poverty assessment. on natural resource wealth grew more slowly than those First, a World Bank Internet web site is under construction without such natural resources. to give greater access to the poverty assessment and to link 6. In the first year of the reform, the payroll tax it with its background work. The web site will be linked funding the public system would fall from 25 percent to to the World Bank's public web site. Second, easy-to-use 15 percent. 77 ANNEXB. POVERTYASSESSMENTS, COMPLETED AND SCHEDULED, FISCAL 1989-2001 Completed Scheduled FY89-97 FY98 FY99 FY2000 FY2001 Total East Asia and the Pacific 12 0 0 5 0 17 Europe and Central Asia 1 1 1 6 6 4 28 Latin America and the 24 1 4 3 1 33 Caribbean Middle East and North Africa 5 0 1 0 2 8 South Asia 7 2 1 3 0 13 Sub-Saharan Africa 3 5 3 0 3 0 41 Total 94a 7b 12c 20d 7e 140 a. Includes 83 first-round poverty assessments and IL updates. b. Includes 3 first-round poverty assessments and 4 updates. c. Includes 7 first-round poverty assessments and 5 updates. d. Includes 7 first-round poverty assessments and 13 updates. e. Includes 4 first-round poverty assessments and 3 updates. f. Includes 104 first-round poverty assessments and 36 updates. Note: Poverty notes are not considered full-fledged poverty assessments because they do not provide complete coverage of the topics specified in the World Bank's Operational Directive on Poverty. Some country teams have chosen to prepare a poverty note instead of a full poverty assessment for reasons that include political constraints and data or resource limitations. While poverty notes do not contain the same level of comprehensive analysis as a full poverty assessment, they serve as a springboard for action and further analysis. Poverty notes have been completed for the following countries: Burkina Faso (fiscal 1997) and Central African Republic (fiscal 1998). 78 Annex C ANNEX C. POVERTYASSESSMENTS COMPLETED, FISCAL 1989-1998 Country Report Title Report No. EAST ASIA AND THE PACIFIC China Strategies for Reducing Poverty in the 1990s 10409-CHA Fiji Restoring Growth in a Changing Global Environment 13862-FIJ Indonesia Poverty Assessment and Strategy Report 8034-IND Indonesia (update) Public Expenditures, Prices and the Poor 11293-IND Lao PDR Social Development Assessment and Strategy 13992-LA Malaysia Growth, Poverty Alleviation and Improved Income Distribution in 8667-MA Malaysia Mongolia Poverty in a Transition Economy 15723-MOG Philippines The Philippines: The Challenge of Poverty 7144-PH Philippines (update) An Opening for Sustained Growth 11061-PH Philippines (update) A Strategy to Fight Poverty 14933-PH Thailand Growth, Poverty, and Income Distribution: An Economic Report 15689-TH Vietnam Poverty Assessment and Strategy 13442-VN EUROPE AND CENTRAL ASIA Albania Growing Out of Poverty 15698-ALB Armenia Confronting Poverty Issues 15693-AM Azerbaijan Poverty Assessment 15601-AZ Belarus An Assessment of Poverty and Prospects for Improved Living 15380-BY Standards Estonia Living Standards During the Transition 15647-EE Hungary Poverty and Social Transfers 14658-HU Kazakhstan Living Standards During the Transition 17520-KZ Kyrgyz Republic Poverty Assessment and Strategy 14380-KG Poland Poverty in Poland 13051-POL Romania Poverty and Social Policy 16462-RO Russia Poverty in Russia: An Assessment 14110-RU Ukraine Poverty in Ukraine 15602-UA LATIN AMERICA AND THE CARIBBEAN Argentina Argentina's Poor: A Profile 13318-AR Bolivia Poverty Report 8643-BO Bolivia (update) Poverty Equity & Income: Selected Policies for Expanding Earning 15272-BO Oppor. for the Poor (2 volumes) Brazil Brazil: A Poverty Assessment (2 volumes) 14323-BR Chile Social Development Progress in Chile: Achievement and 8550-CH Challenges Chile (update) Poverty and Income Distribution in a High-Growth Economy: 1987- 16377-CH 1995 (2 volumes) Colombia Poverty Assessment Report (2 volumes) 12673-CO Costa Rica Public Sector Social Spending 8519-CR Costa Rica (update) Identifying the Social Needs of the Poor: An Update 15449-CR Dominican Republic Growth with Equity: An Agenda for Reform 13619-DO Ecuador A Social Sector Strategy for the Nineties 8935-EC Ecuador (update) Poverty Report 14533-EC El Salvador The Challenge of Poverty Alleviation 12315-ES Guatemala An Assessment of Poverty 12313-GU Guyana Strategies for Reducing Poverty 12861-GUA 79 Annex C Honduras Country Economic Memorandum/Poverty Assessment 13317-HO Jamaica A Strategy for Growth and Poverty Reduction 12702-JM Mexico Mexico in Transition: Towards a New Role for the Public Sector 8770-ME Nicaragua Poverty Assessment 14038-NI Paraguay Public Expenditure Review -- the Social Sectors 10193-PA Paraguay (update) Poverty and the Social Sectors in Paraguay: A Poverty Assessment 12293-PA Peru Poverty Assessment & Social Sector Policies & Programs for the 11191 -PE Poor Trinidad and Tobago Poverty and Unemployment in an Oil Based Economy 14382-TR Uruguay Poverty Assessment: Public Social Expenditures and Their Impact 9663-UR on the Income Distribution Venezuela Fromr Generalized Subsidies to Targeted Programs 9114-VE MIDDLE EAST AND NORTH AFRICA Egypt, Arab Republic of Alleviating Poverty During Structural Adjustment 9838-EGT Jordan Poverty Assessment 12675-JO Morocco Povertv, Adjustment, and Growth 119 18-MOR Tunisia Poverty Alleviation: Preserving Progress while Preparing for the 13993-TUN Future (2 volumes) Yemen, Republic of Poverty Assessment 15158-YEM SOUTH ASIA Bangladesh Bangladesh Poverty and Public Expenditures: An Evaluation of the 7946-BD Impact of Selected Government Progs. Bangladesh From Counting the Poor to Making the Poor Count 17534-BD India Poverty, Employment and Social Services 7617-IN India (update) Achievements and Challenges in Reducing Poverty 16483-IN India (update) Reducing Poverty in India 17881 -IN Nepal Relieving Poverty in a Resource-Scarce Economy 8635-NEP Pakistan A Profile of Poverty 8848-PAK Pakistan (update) Poverty Assessment 14397-PAK Sri Lanka Poverty Assessment 13431-CE SUB-SAHARAN AFRICA Benin Toward a Poverty Alleviation Strategy for Benin 12706-BEN Cameroon Diversity, Growth, and Poverty Reduction 13167-CM Cape Verde Poverty in Cape Verde: A Summary Assessment and a Strategy for 13126-CV its Alleviation Chad Poverty Assessment: Constraints to Rural Development 16567-CD Comoros Poverty and Growth in a Traditional Small Island Economy 13401-COM Congo Poverty Assessment 16043-COB Cote d'Ivoire Poverty in Cote d'Ivoire: A Framework for Action 15640-IVC Dj ibouti Crossroads of the Hom of Africa Poverty Assessment 16543 -DJI Eritrea Poverty Assessment 15595-ER Ethiopia Toward Poverty Alleviation and a Social Action Program 11306-ET Gabon Poverty in a Rent-Based Economy 16333-GA The Gambia An Assessment of Poverty 11941-GM Ghana 2000 and Beyond: Setting the Stage for Accelerated Growth and 11486-GH Poverty Reduction Ghana (update) Ghana: Poverty Past, Present and Future 14504-GH Guinea A Socioeconomic Assessment of Well-Being and Poverty 16465-GUT Guinea-Bissau Poverty Assessment and Social Sector Strategy Review 13155-GUB Kenya Poverty Assessment 13152-KE Lesotho Poverty Assessment 13171-LSO Madagascar Poverty Assessment 14044-MAG Malawi Growth Through Poverty Reduction 8140-MAI Malawi (update) Human Resources and Poverty: Profile and Priorities for Action 15437-MAI 80 Annex C Country Report Title Report No. Mali Assessment of Living Conditions 11842-MLI Mauritania Poverty Assessment 12182-MAU Mauritius CEM: Sharpening the Competitive Edge 13215-MAS Mozambique Poverty Reduction Framework Paper* None Namibia Poverty Alleviation with Sustainable Growth 9510-NAM Niger A Resilient People in a Harsh Environment: Niger Poverty 15344-NIR Assessment Nigeria Poverty in the Midst of Plenty: The Challenge of Growth with 14733-UNI Inclusion Rwanda Poverty Reduction and Sustainable Growth 12465-RW Rwanda (update) Poverty Note: Rebuilding an Equitable Society - Poverty and 17792-RW Poverty Reduction After the Genocide Senegal An Assessment of Living Conditions (2 volumes) 12517-SE Seychelles Poverty in Paradise 12423-SEY Sierra Leone Policies for Sustained Economic Growth and Poverty Alleviation 11371-SL Tanzania The Challenge of Reforms: Growth, Incomes and Welfare 14982-TA Togo Overcoming the Crisis, Overcoming Poverty: A World Bank 15526-TO Poverty Assessment Uganda Growing Out of Poverty 11380-UG Zambia Poverty Assessment 12985-ZA Zimbabwe Achieving Shared Growth: Country Economic Memorandum (2 13540-ZIM volumes) * Document was prepared for consultative group meeting. 81 Annex E ANNEX D. PROGRAM OF TARGETED INTERVENTIONS, FISCAL 1998 Objectives of Projects in the Program of Targeted compared with $4.1 billion, or about 29 percent in Interventions (PTI), Fiscal 1998 fiscal 1997. Of the 240 Bank-approved investment projects in fiscal 1998, 101 were PTI operations. In fiscal 1992, the Bank developed and began to track Table D-1 presents the Bank's fimding for PTI the Program of Targeted Interventions (PTI) to projects over the last seven years. Between fiscal provide a measure of targeted poverty lending. A 1992 and fiscal 1998, the amount and share of IBRD project is included in the PTI if it has a specific and IDA lending to projects in the PTI have mechanism for targeting the poor and/or if the increased significantly. The only exception in the proportion of poor people among its beneficiaries is trend was in fiscal 1997, when there was a decline in significantly larger than the proportion of the poor in both the amount of overall Bank lending and the the total population. share of finds for PTI projects. In contrast, the big jump in the share of World Bank PTI lending in Targeted projects are intended to reach specific groups fiscal 1998 was due to the fact that there were of poor people, including those in less developed or relatively more PTI projects and relatively larger low-income rural and urban areas, those lacking amounts of PTI lending for projects in the education, access to land, migrants, refugees, and other transient social protection, environment, and water supply population groups, unskilled workers, the disabled sectors as compared with the previous two years. and other vulnerable groups, certain ethnic The amount of PTI lending also increased minorities, and groups with certain demographic significantly in the South Asia region. For IDA characteristics such as gender, age (children under five countries, PTI lending was 54 percent of fiscal 1998 years of age), or health risks (diseases correlated with investment lending, which is the same level as in poverty such as schistosomiasis, stunting, or fiscal 1997. With the exception of fiscal 1996, when cholera). the share of IDA PTI lending was unusually high because of a greater number of large PTI loans During fiscal 1998, approximately US$6.7 billion or approved that year, the share of IDA PTI lending has about 40 percent of World Bank investment lending remained at around 54 percent over the last four years. was channeled for direct poverty-targeted projects, Table D-1. Program of Targeted Interventions (PTIs), fiscal 1992-1998 World Bank (IBRD and IDA) PTI lending Millions of dollars 3,836 4,674 4,441 5,437 5,408 4,090 6,733 Percentage of investment lendingb 25 27 25 32 32 29 40 Total number of projects in the PTI 57 72 63 75 79 77 101 Totalnumberofprojects 187 214 197 208 222 203 240 IDA PTI lending Millions of dollars 1,812 2,137 1,853 2,423 3,246 1,874 3,267 Percentage of IDA investment lendingt 44 41 43 54 63 54 54 Total number of IDA-funded projects in the 34 44 35 46 5C 36 55 PTIc Note: A project is included in the PTI if it has a specific mechanism for targeting the poor and/or if the proportion of poor people among its beneficiaries is significantly larger than the proportion of the poor in the total population. a. Fiscal 1992 figures differ from those in linplementing the World Bank's Strategy to Reduce Poverty (World Bank, 1993a) because they include seven projects that were added to the PTI after the earlier report went to press. b. Investment lending includes all lending except for adjustment, debt and debt-service reduction operations, and emergency recovery loans, which are distinct from regular investment operations (see Annex F). c. The number of IDA-funded projects in the PTI excludes joint IBRD/IDA projects, which are counted only once, as IBRD projects. There was one such PTI project in fiscal 1992, two in fiscal 1995, one in fiscal 1996, one in fiscal 1997, and four in fiscal 1998. 82 Annex D The majority of the PTI projects in fiscal 1998 was new agricultural technologies and investments in broadly targeted, as has been the case in the previous villages infrastructure (e.g., roads and processing fiscal years. Broad targeting generally expands the facilities) under the Bengkulu Regional Development provision of services, such as basic education and Project (IBRD) will help boost farm incomes of primary health care, to provide access to as many villagers living in the second poorest province in people as possible. However, it is the poor who Sumatra. benefit most from the extension of basic services since the better-off generally already have access to these Developing human capital. Forty PTI operations in services. A smaller number of PTI projects were fiscal 1998 will develop human capital, including narrowly targeted. Narrow targeting is used in increasing access to or improving the quality of countries where poverty is relatively isolated, primary health care, nutrition, and basic education. geographically concentrated, where the incidence and The education projects will improve female literacy, location of poverty is well understood, and where the develop skills in the informal sector, support training govemment has the capacity to implement targeted of primary and secondary school teachers, provide programs. Many PTI operations combined broad and school materials, and construct educational facilities. narrow targeting and were mostly concemed with Health and population projects will provide packages agriculture, education, population, health, and of basic health and family planning services, and nutrition (PHN), and social protection. support disease prevention and immunization programs. The following are the main issues addressed by projects in the Program of Targeted Interventions in A Basic Health Services Project (IDA) in China will fiscal 1998: provide improved health care for an estimated 45 million people in poor counties through improved Raising agricultural productivity. Twenty-four management of health resources and the upgrading of projects in fiscal 1998 aim to increase agricultural rural health facilities. Key interventions will benefit productivity. The focus of these projects was on women, children, and minority populations who may building and rehabilitating irrigation and rural roads, otherwise have limited access to health services, and supporting on- and off-firm income generation, will include support to matemal and child health increasing access to rural finance, promoting care, tuberculosis treatment, immunization of agroprocessing and farm product marketing, newbom children for hepatitis B, cataract surgery, and providing agricultural support services, managing the deworming of school children. In Madagascar, the environment, natural resources, and resettlement, and Community Nutrition II Project (IDA) aims to strengthening agricultural institutions. improve the nutritional status of children under three years of age (nearly half of whom are malnourished), The projects funded in fiscal 1998 illustrate the pregnant and lactating women, and school-age different methods used to increase productivity. The children. It will support a community nutrition Egypt East Delta Agricultural Services Project (IDA) program that includes rehabilitating malnourished will provide agriculture extension services, tertiary children and training health administrators. An irrigation and drainage works, and a line of credit Information, Education and Communication program through the Principal Bank for Development and to increase health awareness among local Agriculture Credit to finance the purchase of communities will be initiated. agricultural inputs. Measures will be taken to facilitate women's access to credit, including a In the education sector, the Ethiopia Education Sector program of home visits to assist in the preparation of Development Program (IDA) is a comprehensive credit applications and the establishment of four new initiative to restructure the education system. Its village banks. The Land Management III Project in objective is to expand the access of primary and Brazil (IBRD) aims to increase agricultural vocational education to the rural poor, girls, and productivity in areas highly susceptible to soil dropouts. Among the activities to be supported are erosion and with the highest concentration of small- the construction of more schools, with particular scale farmers and poverty through the promotion of attention given to rural and undeserved areas, the sound natural resource management practices and elimination of school fees for grades 1-10, the conservation techniques. incorporation of gender-sensitive materials and local languages into the curriculum, and training for The Southem Govemorates Rural Development teachers. The El Salvador Education Reform Project Project (IDA) in Yemen, will help very poor landless (IBRD) will expand access to preschool and basic families get established in new farming operations by education, and will launch an innovative Initial providing them with secure titles for land, training Education program with parent participation in rural and access to credit. In Indonesia, the introduction of and marginal urban areas to benefit children ages 0-3 83 Annex D years. In addition, the implementation of alternative economic adjustment or to provide assistance to and accelerated education modalities will help meet those unable to work. The Russia Social Protection the needs of overage children and curb repetition. The Implementation Loan (IBRD) aims to lay a solid project focuses on promoting gender equity in foundation for building a pension and welfare system classrooms, and boosting girls' self-esteem as a that meets the needs of a market economy, and means of addressing the alarming increase in teen includes monitoring and evaluation systems to track pregnancy. The Second General Education Project unemployment and progress in social protection (IDA) in Sri Lanka will target deprived schools in programs. Initiatives under these programs will rural and urban slum areas in providing facilities and ensure higher and more regular pensions for about 4 quality inputs that will bridge the gap between poor million of the poorest pensioners, guaranteed and affluent schools. minimum unemployment benefits, as well as provide effective child allowances, and better sickness and Urban development, water supply and sanitation. In maternity benefits. In addition to improving the the urban development and the water supply and quality, effectiveness, and coverage of services sanitation sectors, 11 projects in fiscal 1998 will provided in selected primary health centers (located improve living conditions of poor people by mostly in areas where the demographic and increasing access to basic infrastructure. These socioeconomic status of the population are poor) the projects will construct and rehabilitate water, Andhra Pradesh Economic Restructuring Project sewerage, flood control, and drainage systems, as (IDA/IBRD) in India will support the Integrated well as community irrigation schemes. The Algeria Child Development Services (ICDS) program to Low-Income Housing Project (IBRD) will benefit improve the nutrition and health of poor pre-school about 50,000 households in non-serviced, low- children and women, particularly pregnant and income areas, including urban slum residents. The lactating mothers and adolescent girls from deprived project provides for affordable subsidized housing and households. ICDS centers are mainly located in rural land titles. Site development for low-income families areas where over 40 percent of the population are will offer basic infrastructure services, including defined as being below the poverty line. In Comoros, roads, safe drinking water, sewerage, drainage, and under the Social Fund Project (IDA), resources will electricity. In Mongolia, the ULanbaatar Services be allocated to demand-driven initiatives developed Improvements Project (IDA) will address the by the community to improve access to basic services inequities in services between the ger (low-income and create employment through labor-intensive urban settlements) and formal hiousing areas by works. Results of the beneficiary assessment played extending the water supply network, implementing a an important role in the design of the project. The water leakage and waste recluction program, Angola Post-Conflict Social Recovery Project (IDA) upgrading infrastructure, including roads, footpaths, will test in a number of target areas a program to and drainage in ger areas. The Water Supply, support war-affected rural communities in their efforts Sanitation and Health Project in U:zbekistan (IBRD) to reintegrate displaced people and revitalize the will provide reliable and safe water supply to about economy. Subprojects will seek to reestablish 1.12 million inhabitants in rural areas and 0.41 household livelihood and rehabilitate basic million in urban areas. A community-based approach community-level social and economic infrastructure will be used to promote hygiene education and such as water supply, roads, schools, health posts, improve sanitation facilities that will reduce the rural markets, and irrigation works in accordance with incidence of water-bome diseases, particularly the needs and priorities expressed by the diarrheal diseases, which are more prevalent among communities. In Jordan, the Training and children. Technical assistance will help build Employment Support Project (IBRD) will test the institutional capacity in organizations involved in demand for and the effectiveness and efficiency of providing water and sewerage services. employer-based training targeted at the unemployed poor. Social Protection. Eleven operations in fiscal 1998 support measures to protect the vulnerable through Other PTI areas. The PTI also includes projects in social funds or some type of safety net measure (e.g., areas such as transport, environment, and regional retraining schemes, nutrition programs, development. unemployment benefits, and the re-form of the social protection system). Social action funds generally The Transport Infrastructure Rehabilitation Project support community-level microprojects aimed at (IDA) in Niger will help develop a rural roads increasing access to basic education and health, safe maintenance program to increase access to markets water, irrigation facilities, and rural roads. Safety net and basic services and reduce high transport costs for measures are typically short-term measures designed poor people living in outlying areas. In Ecuador, the to protect vulnerable groups during periods of Indigenous and Afro-Ecuadorian People Development 84 Annex D Project (IBRD) focuses on increasing the income and leaders. In an effort to decentralize decision-making quality of life of about 815,000 minority people by and strengthen local govemment and community providing increased access to land and credit. institutions, the Kecamatan Development Project Subprojects include investments in natural resources (IBRD) in Indonesia will give block grants to 725 of management and community-based economic and the poorest subdistricts to finance economic, social social infrastructure. Poor people living in communes and infrastructure activities in line with self- surrounding nature reserves and forestlands will determined local priorities. benefit from the Vietnam Forest Protection and Rural Protection Project (IDA). In addition to protecting Table D-2 shows the share of PTI lending to total and managing existing forests, activities will also investment lending by sectors over the last seven include land allocation, social and agricultural years. Three-year averages have been calculated to support services and rural infrastructure, all designed smooth out year-to-year fluctuations, which are to increase income, employment, and market normal occurrences. opportunities for the poor living in remote communities. A Learning and Innovation Loan in As indicated in Table D-2, a large majority of both Colombia, the Magdalena Medio Regional IDA and IBRD projects in the education, PHN, and Development Project (IBRD), addresses the links social sectors belongs in the PTI. For total World between poverty and violence in rural areas. The Bank lending the average share of PTI lending in the project will help build the operational capacity of the education and water supply and sanitation sectors has Consortium and the Citizen's Network to implement declined over the period fiscal 1992-1998. The share the Program for Development and Peace. It will also of PTI lending in PHN and urban development has finance specific subprojects for education, health, and remained steady, and PTI lending in the social and the environment. agriculture sectors has increased (see Table D-2). Over the period fiscal 1992-1998, the average share of Communitv Participation and Decentralization. IDA PTI lending in total investment lending has Many PTI projects in fiscal 1998 involved local increased significantly in agriculture, education, the organizations and communities in their preparation social sector, and urban development. It has and implementation, as well as invested in building increased moderately in the PHN sector, and has capacity. Local participation ensures that priorities fluctuated in the water supply and sanitation sector, and concerns of beneficiaries are taken into account. It with a noticeable decline in the last three years. also fosters ownership of activities and develops capacity for self-management. For example, the Mali Table D-3 shows the regional variation in the Bank's Grassroots Initiatives to Fight Hunger and Poverty funding for PTI projects. Table D-4 briefly describes Project (IDA) supports a Community Development why specific fiscal 1998 projects were included in the component that starts out with a participatory needs PTI. assessment, which forms the basis for beneficiary communities to play a central role in the decision- Although projects in the PTI obviously seek to making process throughout the project cycle. Among benefit the poor, this does not mean that non-PTI its activities is the setting up of a Community investment projects do not also contain provisions Management committee, with emphasis on including that help to reduce poverty. For example, the women, to lead and coordinate programs, together Armenia Municipal Development Project (IDA), with local capacity building measures to strengthen though not poverty-targeted, will make emergency community-development skills. An Information, short-term improvements in the water supply system Education, and Communication program will be to ameliorate the drinking water supply to Yerevan, tailored to community needs. The Bolivia thereby improving living conditions also for the most Participatory Rural Investment Project (IDA) will deprived population. By improving the quality of provide technical assistance to municipalities, urban transport services in the Metropolitan area, the indigenous districts, rural communities and NGOs in Brazil Rio de Janeiro Mass Transit Project (IBRD) participatory planning and implementation of will benefit the urban poor by facilitating access and subprojects and non-formal training of municipal making it affordable. councils, vigilance committees, and community 85 Annex D Table D-2. Program of Targeted Interventions (PTIs) by sector, fiscal 1992-1998 Sector FY92-94 FY93-95 FY94-96 FY95-9 7 FY96-98 PTI Lending as Percentage of Total Investment Lending' (IDA and IBRD) Agriculture 27 31 42 49 47 Education 72 67 63 56 62 Population, health, and nutrition (PHN) 79 84 76 78 78 Social protection 82 95 87 88 90 Urban development 37 37 27 28 39 Water supply and sanitation 44 38 31 24 35 Other sectors 6 7 8 10 10 IDA PTI Lending as Percentage of IDA Investment Lendinga Agriculture 46 44 54 62 61 Education 76 84 92 89 88 Population, health, and nutrition (PHN) 87 89 94 94 96 Social protection 72 88 93 95 98 Urban development 29 25 34 47 66 Water supply and sanitation 57 48 59 45 27 Other sectors 12 11 12 17 16 Note: Figures differ from those in earlier reports because of recent sector reclassification of operations. a. Investment lending includes all lending except for adjustment, debt and debt-service reduction operations, and emergency recovery loans, which are different from regular investment operations in objective and format. Table D-3. Program of Targeted Interventions (PTIs) lending by region, fiscal 1998 Sub- East Europe Latin Americ, Middle South Saharan Asia and and and the East and Asia Africa the Central Caribbean North Total Lending Pacific Asia Affrica Total PTI Lending Millions of dollars 789.7 979.8 551.0 1,910.7 470.2 2,031.9 6,733.3 Percentage of investment lending 39 25 25 45 60 56 4U Total number of PTI projects 25 14 11 27 10 14 101 IDA PTI Lending Millions of dollars 789.7 370.1 72.4 1,61.8 142.2 1,730.6 3,266.8 Percentage of all PTI lending 100 38 13 8 30 85 49 Percentage of IDA investment lending 40 52 18 66 85 68 54 Total number of IDA-funded projects in the PTIa 25 4 6 3 4 13 55 a. The number of IDA-funded projects in the PTI excludes joint IBRD/IDA projects, which are counted only once, as IBRD projects. There was one such PTI project in fiscal 1992, two in fiscal 1995, one in fiscal 1996, one in fiscal 1997, and four in fiscal 1998. 86 Annex D Table D-4. Program of Targeted Interventions, fiscal 1998 Country and project Reason(s)for inclusion' East Asia and the Pac- - China The project will provide improved health care for an estimated 45 million people in Basic Health Services Project 97 designated poor counties in ten provinces. This will be achieved by improving (IDA: US$85 m.) the allocation and management of health resources, upgrading rural health facilities, enhancing the quality and effectiveness of health services and programs, and increasing risk sharing and affordability of essential health care for the poor. Improvements in basic health care will particularly benefit women, children, and minority groups. [b] Forestry Development in Poor Areas The development of forest resources in poor areas of central and westem China will Project be supported on a sustainable and participatory basis. Beneficiaries comprise a large (IBRD: US$ 100 m., share of ethnic minorities and women. Project activities are expected to a) increase IDA: US$ 100 m.) per capita income for about 2 million poor people (about 450,000 households), lifting them above the poverty line by project closing; b) increase the supply of forest products used as inputs for construction and agro-processing; c) increase the value- added in processing of forestry products; d) improve environmental management of ecologically fragile mountain watersheds; and e) develop and disseminate agricultural technologies. [b] Guangxi Urban Environment Project This environment project - the second implemented in this poor westem region, will (IBRD: US$72 m., IDA: US$20 m.) target the urban poor for the first time by supporting improvements in environmental sanitation. Beneficiaries include women and ethnic minorities, who make up a majority in the Guangxi. [b] Tarim Basin II Project The project aims to a) increase the incomes of about 200,000 poor households (IBRD: US$90 m., IDA: US$60 m.) through irrigated agriculture development, b) promote sustainable techniques to upgrade marginal land, including major water storage, transfer, irrigation and flood control in the Tarim Basin, and c) restore and preserve the "green corridor" in the lower reaches of the Tarim River. Minority groups, including Uygur, Hui, Mongols, and Kirghiz are the overwhelming majority of the project's target population, and poor farmers and herdsmen are direct project beneficiaries. [a,b] Indonesia Poverty in 220 villages of Bengkulu - the second poorest province in Sumatra - will Bengkulu Regional Development be reduced. The adoption of new agricultural technologies and investments in village Project infrastructure will help boost farm incomes and reduce the burden on women. [b] (IBRD: US$20.5 m.) Coral Reef Rehabilitation and Protecting, rehabilitating and achieving sustainable use of coral reefs and associated Management Project ecosystems in Indonesia are major objectives of this project. The welfare of coastal (IBRD: US$6.9 m.) communities will improve through public awareness campaigns, surveillance and enforcement, and community-based management. Two pilot sites will adopt community-based fisheries management to restore the productive potential of coral reefs and assist poor fishermen in finding altemnative income opportunities to minimize destructive activities (illegal fishing practices, use of cyanide, large-scale explosives and overfishing). [b] A project must meet at least one of two criteria to be included in the Program of Targeted Interventions (PTI). Criterion [a] refers to projects with a specific targeting mechanism for reaching the poor and criterion [b] refers to projects for which the proportion of the poor among the project's beneficiaries is significantly larger than the proportion of the poor in the total population. The applicable criteria for each project are indicated in square brackets. 87 Annex D Country and project Reason(s) for inclusion' Kecamatan Development Project This project aims to raise rural incomes, strengthen kecamatan (subdistrict) and (IBRD: US$225 m.) village development and community institutions, and build public infrastructure using labor-intensive methods. About 25 million people living in rural areas of 725 kecamatans selected from among the poorest in the country will be targeted. [b] Maluku Regional Development About 50,000 households or 520,000 people including isolated, vulnerable people, Project women and ethnic minority groups in the rural and coastal areas of Maluku, will (IBRD: US$16.3 m.) benefit from increased agricultural productivity, reduced losses due to pests, diseases and lack of storage, and improved agricultural processing and market access. This will be achieved through investments in crops, processing facilities, small-scale infrastructure, and the provision of greater access to credit. [b] Mongolia The extension of the water distribution network to the ger areas, low-income urban Ulaanbaatar Services. Improvement settlements, will improve the lives of poor people. The project includes building four (IDA: US$16.7 m.) reservoirs to store water, and 12km of pipelines; implementing a water leakage and waste reduction program, and undertaking a ger area infrastructure-upgrading program. [b] Philippines About 1.8 million families in four regions, half with incomes below the poverty line, Community-Based Resources are expected to benefit from natural resource management projects aimed at reducing Management Project poverty and environmental degradation. The projects will be designed and (IBRD: US$50 m.) implemented by the local population, and are expected to particularly benefit upland farmers and fishermen. [b] Early Childhood Development This project will improve early childhood development policies and programs in Program health, nutrition, cognitive development and primary education for 10 to 1 1 million (IBRD: US$19 m.) pre-school children, with emphasis on 3 to 4 million children living below the poverty line. Actions will include an expanded program of immunization, management and treatment of childhood illnesses, malnutrition prevention, curriculum development and service delivery enhancement. [b] SZOPAD Social Fund Project Thie project aims to increase the access of about 10 million people in the most (IBRD: US$10 m.) conflict-affected areas of the Special Zone for Peace and Development (SZOPAD) to basic economic and social infrastructure, public services, and employment opportunities. [b] Vietnam The project aims to diversify and intensify agricultural production by developing the Agricultural Diversification Project sniallholder rubber sector, upgrading livestock production, and intensifying crop (IDA: US$66.9 m.) production in the poor central highland and coastal provinces of Vietnam. Ethnic minorities (ethnic Kinh), smallholders and subsistence-shifting cultivators in one of Vietnam's poorest regions will be targeted. [a,b] Forest Protection and Rural Villagers in 40 communes and towns, many of whom are living below the poverty Development Project line, will benefit from actions to increase food security, improve literacy, health and (IDA: US$21.5 m.) sanitation, and expand employment opportunities. Numerous measures are incorporated to safeguard the interests of ethnic minorities and other vulnerable groups, including support for community-based activities with representation from indigenous ethnic minorities and women's groups. [b] Albania The most common illnesses at the primary health care level, including acute Health System Recovery and respiratory infections, diarrheal diseases, and reproductive health problems will be Development Project addressed through the provision of better access and service delivery to people living (IDA: US$17 m.) in urban Tirana and the surrounding rural districts. [b] 88 Annex D Country and project Reason(s) for inclusion' Georgia The project will support the government's ongoing Cultural Heritage Initiative Cultural Heritage Project (CHI) and will provide investment funding for the rehabilitation and restoration of (IDA: US$4.5 m.) four pilot sites where a large number of very poor people live. It will also test approaches to revive the once dynamic tourism industry. An Emergency Rehabilitation Fund for emergency repairs at cultural sites or cultural artifacts will generate local jobs to help reduce poverty. [b] Municipal Development and This project will support the initial phase of Georgia's decentralization process, Decentralization Project which aims to gradually shift responsibility and accountability to municipalities for (IDA: US$20.9 m.) maintaining and improving urban services. Among the benefits accruing to the urban population in cities with inadequate services and environmental risks are reduced exposure to water-borne diseases, and increased employment opportunities for low- skilled and unskilled workers. [b] Social Investment Fund Project About 400 communities in the poorest and mountainous areas will be targeted to (IDA: US$20 m.) receive small-scale infrastructure improvements in water supply, sanitation, solid waste management, local health clinics, primary schools and feeder roads. The monitoring and evaluation component will improve the capability to analyze poverty levels and trends, with poverty data disaggregated by gender. Laborers are expected to benefit from an estimated 494,000 employee labor days generated from activities in microprojects. [b] Macedonia, FYR About 15,000 primary students in rural and poverty-stricken areas will benefit from Education Rehabilitation Project the rehabilitation of about 50 schools in danger of collapse or serious physical (IDA: US$5 m.) deterioration, teacher training and textbook reform, curriculum development and institutional development of government agencies. [b] Moldova This project will develop and test a cooperative rural banking system that aims to Rural Finance Project provide effective financial services to an estimated 10,000 small private farmers and (IDA: US$5 m.) rural entrepreneurs. Savings and Credit Associations (SCA) will be established, training will be provided to members, and a rural credit line aimed at providing small loans to finance investments for productive purposes such as seasonal agricultural inputs, spare parts for equipment, small works, and other small investments, will be financed. [b] Romania The project will support the government's Department of Child Protection (DCP) in Child Welfare Reform Project reforming the child-care and child protection system. This will be achieved by (IBRD: US$5 m.) reducing the number of children in institutions, improving the quality of care for institutionalized children, and developing altemative care systems. It will target street children in Bucharest who are particularly at risk to poverty and social deprivation. Support for community-based child welfare services include family counseling, shelter for young single mothers and day care centers for children of families who are at risk. [a,b] School Rehabilitation Project By rehabilitating rural schools and ensuring the provision of adequate facilities, this (IBRD: US$70 m.) project will improve the quality and expand the coverage of education in the poorest communities in Romania. About 200,000 students will benefit from the rehabilitation, upgrading and furnishing of nearly 900 pre-university school buildings (100 kindergarten, 700 primary and 100 general secondary), which are in imminent danger of collapse. [b] 89 Annex D Country and project Reason(s) for inclusion' Russia As a complement to the social reforms proposed under the Social Protection Social Protection Implementation Adjustment Loan, this project supports a broad program of social protection, Loan including setting up monitoring systems, data analysis, and reviews of on going (IBRD: US$28.6 m.) programs for unemployment assistance and social protection. About four million of the poorest pensioners will be targeted to receive regular and adequate pensions. Together, the programs will improve the lives of more than 15 million people over the medium and longer term, providing Russia with a social safety net that will be wvell targeted towards vulnerable groups. [a,b] Turkey To help attain universal coverage and improve the quality of basic education, a new Basic Education Project expanded eight-year basic education cycle will be implemented. Regions and groups (IBRD: US$300 m.) with the lowest school attendance and educational performance in basic education will be targeted, with special emphasis on girls in the East and Southeast and children from poor households. Free textbooks and school uniforms will be given to children from poor households; teachers will be recruited and trained; schools will be upgraded; and parental and community participation will be encouraged. [a,b] Uzbekistan About 50,000 inhabitants in rural and peri-urban areas will benefit from the use of Water Supply, Sanitation and Health safe and reliable water through improved sanitation facilities. This will also decrease Project the incidence of water-bome diseases, particularly diarrheal diseases, which are more (IBRD: US$75 m.) prevalent among children. Technical assistance will be provided for the institutional building of six organizations involved in providing water and sewerage services. [b] Argentina The project will increase the productive capacity of about 30,000 poor rural Small Farmer Development Project households by improving production-oriented infrastructure, diversifying production (IBRD: US$75 i.) activities, and integrating beneficiaries into local organizations. A rural Investment Fumd will be established to finance subprojects with specific mechanisms to target poor households, indigenous groups, women and rural youth. [b] Third Social Protection Project This project will support the third phase of the TRABAJAR safety net program by (IBRD: US$284 m.) providing temporary financial assistance to an estimated 300,000 poor workers. S,ubprojects will finance the upgrading and restoration of schools, health facilities, sanitation facilities, and small roads, as well as create employment for low-skilled workers. [b] Bolivia This project will complement the existing education reform program, which seeks to Education Quality and Equity increase the years of schooling completed, particularly at the primary level, improve Strengthening Project the quality of schooling, and develop approaches to reduce gender disparities in (IDA: US$75 m.) selected municipalities. Target beneficiaries are boys and girls from impoverished rural and urban marginal areas, of whom 70-75 percent are indigenous. The rural girls component is expected to benefit approximately 16,000 girls (80-90 percent are indigenous) in the priority municipalities. [b] Participatory Rural Investment Project The inhabitants of the 200 poorest municipalities with the lowest index of satisfied (IDA: US$62.8 m.) basic needs, as well as indigenous districts, will benefit from a) improved access to basic services and markets; b) investments in diversifying and intensifying rural production systems; c) improved natural resource management and environmental protection; d) opportunities in off-farm income from ethno- and eco-tourism; e) employment generation from subprojects (e.g., roads, bridges, irrigation systems, storage facilities); and f) social capital development initiatives at the prefectoral, municipal and community levels. A Small Farmer Development Fund will finance small artisanal equipment for women's groups. [b] 90 Annex D Country and project Reason(s) for inclusion' Brazil The project aims to improve access to stable water supplies for potable, municipal, Bahia Water Resources Management industrial and irrigation uses in poor river basin areas. Urban and rural water users of Project Bahia will benefit from increased supply of safe drinking water, abatement of water- (IBRD: US$51 m.) vectored health problems, and infrastructure activities. [b] Federal Water Resources Management This project aims to promote rational use and participatory management of water Project (PROAGUA) resources in Brazil with emphasis in the Northeast region. Infrastructure investments (IBRD: US$198 m.) include the construction of water storage and conveyance infrastructure to provide reliable water supply, particularly to those populations living in rural areas and in small towns. Some of the 1.3 million beneficiaries will also benefit from new employment opportunities and increased agricultural production. [b] School Improvement Project Brazil's FUNDESCOLA I project will finance activities covering about 1.3 million (FUNDESCOLA I) students aged seven to fourteen in approximately 4,200 schools. It will also reinforce (IBRD: US$62.5 m.) the government's goal of ensuring that children from the poorest regions successfully complete the basic eight-grade education. Major components include: raising schools to minimum operational standards (educating and certifying teachers, supplying basic furniture and equipment, and financing school improvement investments) and strengthening educational management and administration at all levels, with emphasis on fostering community participation. [b] Land Management III Project By promoting improved land management and soil conservation practices, this (IBRD: US$55 m.) project will increase agricultural productivity in areas with the highest susceptibility to soil erosion and highest concentration of small-scale farmers and poverty. Approximately 4.5 million ha of farmland and 90,000 farmers (of which 49,500 are small farmers) will be covered. Main activities include extension services, strengthening institutional capacity of farmers' associations and promoting rational use of inputs, crop rotation and diversification. [b] Rural Poverty Alleviation Project The project will provide basic social and economic infrastructure and employment Maranhao and income-generating opportunities for the rural poor. Among its objectives are (IBRD: US$80 m.) decentralizing resource allocation and decision-making to local levels and deepening the existing social capital. It will lay the groundwork for participatory institutions and processes at the municipal and community levels. Small-scale subprojects identified by local groups as priority investments include infrastructure projects (rural electricity, local roads, and water supply), social projects (health-related improvements, daycare centers, and schools), and productive projects (communal tractors and minor irrigation schemes). Established criteria will ensure that project activities reach poor rural dwellers and vulnerable groups. [b] NE Rural Poverty Alleviation Project This project will assist the State Government of Paraiba in reducing rural poverty by Paraiba providing basic social and economic infrastructure, employment and income- (IBRD: US$60 m.) generating opportunities for about 1.1 million poor people in isolated settlements of less than 7,500 people. The project will provide matching grants to rural community associations to finance small-scale subprojects identified by these groups as priority investments. Projects will be operated and maintained by the beneficiaries themselves, and will include improving local roads, setting up day care centers, establishing communal irrigation schemes, and providing agricultural equipment. [a,b] Water Sector Modernization II Project Improved water supply, sewage collection and treatment services will especially (IBRD: US$150 m.) benefit the poor population in six states and 24 municipalities. [b] 91 Annex D Country and project Reason(s) for inclusion' Colombia This project seeks to improve the quality of education for 5-13 year olds. This will Antioquia Basic Education Project be achieved by a) expanding access and increasing school retention of preschool and (IBRD: US$40 m.) lower secondary students living in poor rural and urban areas, as well as those in at- risk-of-violence-communities; b) increasing the capacity of schools, communities, municipalities, and the education department to effectively provide better services. Approximately 1,200 schools and 220,000 students will benefit from the programs. Ib] Magdalena Medio Regional The operational capacity of the Consortium for the Program for Development and Development Project Peace of the Magdalena Medio Region will be developed. This will enhance the (IBRD: US$5 m.) ability of its citizens network and other partners to work together towards poverty reduction objectives under a peaceful coexistence in the second most violent region in the country. The program provides funding for outreach activities (workshops, training, and study tours) and uses a participatory methodology involving a continuous process of social analyses in which communities play a central role. Roughly 500,000 people with incomes below the poverty line will benefit. [b] Education Project (Municipality of Financing for public services to 167 schools will be increased, primarily benefiting Pasto) 95,000 children aged 5-15. The project provides for educational materials, (IBRD: US$7.2 m.) infrastructure rehabilitation, and funding for subprojects, and includes the active participation of families and local communities in Pasto, where 40 percent of the population lives below the poverty line. [b] Peasant Enterprise Zones for Peace The project develops a replicable methodology for the establishment and operation of Project a Peasant Enterprise Zone for areas affected by violence and illicit activities and (IBRD: US$5 m.) supports prototype actions to protect rainforest reserves and nearby indigenous territories. Project beneficiaries include about 250,000 families, including many landless and low-income peasant families displaced by violence. The project will finance the issuance of land titles and recording in land registry, capacity building focusing on rural finance and micro-enterprise development. The specific needs of women, youths and indigenous people will be addressed. [b] Dominican Republic The health status of about 3.8 million people, nearly 48 percent of the country's total Provincial Health Services Project population will be improved. The project will help expand essential health care (IBRD: US$30 m.) services to the poorest groups and strengthen the capacity of provincial health organizations to deliver quality health care. Project interventions are expected to iraprove infant and child survival, especially for children under five years of age, and matemal health, particularly for poor pregnant and lactating women, and reduce the prevalence and severity of the most common diseases. [a,b] Ecuador The proposed technical assistance project aims to improve the availability, Agricultural Census and Information consistency, validity, and timeliness of agricultural information in order to facilitate System Technical Assistance Project the decision-making process in both the public and private sectors. Improved (IBRD: US$20 m.) information would lead to better targeting of the govemment's rural poverty alleviation programs. [b] Health Services Modemization This project complements the Social Development Project: Health and Nutrition (MODERSA) Project (fiscal 1993), and will support the improvement of health care for the poor, (IBRD: US$45 m.) particularly mothers and children under five years of age. This will be achieved by expanding access to essential health services (prenatal care, nutrition surveillance, tetanus vaccination, treatment of STDs, etc), and improving the quality and efficiency of ambulatory and hospital services, as well as developing a new health care organization. The potential beneficiary population is estimated at about 2 million or 20 percent of the total population in the country. [a,b] 92 Annex D Country and project Reason(s) for inclusion' Indigenous and Afro-Ecuadorian The project is expected to increase the income and quality of life of about 815,000 Peoples Development Project poor indigenous and Afro-Ecuadorian people by providing increased access to land (IBRD: US$25 m.) resources and financing for investment subprojects. Project activities in 288 parroquias, which have been identified as predominantly indigenous or Afro- Ecuadorian based on an Indigenous Poverty Map, will likely be implemented by local membership organizations and communities. [a,b] El Salvador The project will support the expansion of preschool and basic education to children 4 Education Reform to 15 years old, and will launch a pilot program in Initial Education to benefit (IBRD: US$88 m.) children ages 0-3 years old. These programs will be aimed at rural and marginal urban populations, with emphasis on the poorest areas. Project activities address gender, equity in classrooms, and teen pregnancy. [a,b] Secondary Education Project Improved quality of and greater access to secondary school will be achieved by (IBRD: US$58 m.) constructing and expanding schools, providing distance education in rural areas, and reducing the direct and opportunity costs of education for low-income students through a targeted scholarship program. Special measures will be taken to target individuals from low-income households and rural areas, eliminate gender stereotypes in books, introduce career guidance programs and include sex education in the curriculum. [a,b] Mexico This project supports the Phase I of the Education Development Program, which Basic Education Project seeks to improve the quality of initial and basic education and non-traditional (IBRD: US$115 m.) modalities of post-primary education. Project activities include a) improving the school supervision system; b) establishing and implementing school-based management strategies; and c) designing and piloting new education modalities to better meet the needs of migrant, indigenous children, and those children living in marginal urban areas. The project will target children in rural and marginalized urban areas with the highest incidence of poverty. [a] Higher Education Financing To help the government promote greater equity and quality in higher education, the (IBRD: US$180.2 m.) project will provide loans to a minimum of 25,600 academically qualified but financially needy students. Approximately 70 percent of the beneficiaries will come from middle-low to low-income families. The students are selected using established criteria such as academic background, financial need, area of study, and socioeconomic background. [a,b] Rural Development in Marginal Areas This project is the first phase of a program to improve smallholders' productivity and (IBRD: US$47 m.) food security in six marginal areas, which are among the poorest in the country. About 40,000 poor, small farm households will benefit directly from increased agricultural production and diversification through the enhancement of processing and marketing activities, technical assistance and training, and improved access to local markets. The project will promote the participation of indigenous people who comprise 67 percent of the target population, as well as women who play a key role in agricultural production and family nutrition. [b] Nicaragua This project is the first phase of a program to improve the efficiency, effectiveness and Health Sector Modernization Project equity of the Nicaraguan health system by a) strengthening first-level care and (IDA: US$24 m.) nutrition, b) modernizing the hospital network, c) developing the institutional capacity of the Ministry of Health, and d) supporting social security reform. A pilot program will test the expansion of a network of women's centers to combat the high maternal and perinatal mortality associated with inadequate access to health care services, lack of education and poverty. [a,b] 93 Annex D Country and project Reason(s)for inclusion' Paraguay Water supply and sanitation services will be provided to 330 poor communities, Fourth Rural Water Supply and including 35 of the poorest indigenous communities. The expansion of these services Sanitation Project will help reduce poverty, increase productivity, and improve health status, (IBRD: US$40 m.) particularly of women and children. [b] Venezuela (General health and living conditions, particularly in poor areas, will be improved Environmental Management and through support for the modernization and institutional capacity building of Cartography Project environmental management at the national, state, and municipal levels. Major (IBRD: USS28 m.) benefits include improved air quality, reduced water pollution from organic wastes and toxic chemicals, increased access to public information, and awareness of long- term health consequences stemming from broad-based campaigns and community participation. [b] Algeria Akbout 50,000 households in non-serviced, low-income areas, including urban slum Low Income Housing Project residents, will benefit from this project through the provision of affordable subsidized (IBRD: US$150 m.) housing and land titles. [b] Egypt Support services will be financed to facilitate the resettlement of 26,000 poor farm East Delta Agricultural Services families on recently developed areas in the East Delta and to increase their Project agricultural production. [b] (IDA: US$15 m.) Health Sector Reform Program This project will help implement the first phase of the govemment's long-term (IDA: US$90 m.) Health Sector Reform Program (HSRP) intended to improve the population's health status. The primary health care package and universal coverage initiatives will especially benefit women, children and other disadvantaged groups. [b] Jordan Approximately 290,000 people in 27 poor urban settlements and 1.3 million people Community Infrastructure Project in approximately 300 municipalities and villages will benefit from the provision of (IBRD: US$30 m.) safe access roads, street paving and lighting, drainage systems, potable water supply and sanitation services, and social infrastructure such as schools, health centers, and community and youth centers. Target beneficiaries are the poor living in underdeveloped urban squatter settlements and refugee camps, and low-income municipalities and villages. [b] Training and Employment Support This project will pilot test the demand for and the effectiveness and efficiency of Project employer-based training. It is supported by a Training Fund which is targeted at the (IBRD: US$5 m.) unemployed poor and provides for in-service training for up to six months. Jordanian job seekers will be encouraged to accept jobs currently performed by non-Jordanians, and women will be offered jobs in non-traditional occupations. [a,b] Lebanon Students from lower income families, women, and workers in need of training will Vocational and Technical Education benefit from a project designed to increase their access to vocational and technical (VTE) Project training, improve its relevance, responsiveness, and quality, and make it affordable (IBRD: US$63 m.) for economically disadvantaged groups. [a,b] Morocco The standard of living of about 4 million people living in 90 small and medium- Second Municipal Finance Project sized communes located in Morocco's underdeveloped areas is expected to improve (IBRD: US$70 m.) thr-ough the provision of basic infrastructure services (e.g., water supply and sewerage, solid waste, wholesale markets and roads). [b] Rural Water Supply and Sanitation Safe water and sanitation will be provided for approximately 1.3 million rural Project inhabitants in the poorest provinces, thereby improving their health status and (IBRD: US$ 10 m.) productivity. The construction and rehabilitation of public water points will help reduce the burden of rural girls who are traditionally involved in fetching water over long distances. [b] 94 Annex D Country and project Reason(s) for inclusion' Yemen Land allocation and titling, irrigation, farmer services (training, extension services, The Southern Govemorates Rural and provision of credit for equipment and inputs), and off-farm income opportunities, Development Project (livestock and fishing development), as well as support for microenterprises, are (IDA: US$24.7 m.) expected to supplement the means of economic livelihood for about 8,750 very poor rural individuals and their families. About 1,950 very poor landless families will be established in new farming operations, including the provision of land and title. [b] Seeds and Agricultural Services This project seeks to increase food security and agricultural productivity for about Project 116,000 rural households, the poorest segment of the Yemeni rural population. It (IDA: US$12.5 m.) will set up a sustainable system for improving and expanding supply marketing channels for seeds of adapted varieties, fertilizers, and other inputs. [b] Bangladesh The program will increase access to essential services for the poor, lower maternal Health and Population Program mortality and morbidity, and further improve child health and family planning. It is Project expected that 25 million married women and approximately 16 million children (IDA: US$250 m.) below the age of five will benefit from better health, nutrition, hospitals and public health structures. [b] Primary Education Development This project supports the government's objectives to improve the quality and Project management of the school system and to provide universal coverage and equitable (IDA: US$150 m.) access to primary schooling. The latter will be achieved by providing additional classrooms and repairing and rehabilitating schools in the under-served areas. Students from the poorest households will be provided with stationery. [b] Silk Development Project The first phase of a long-term program to assist small-scale silk producers, 80 percent (IDA: US$11.4 m.) of whom are poor rural women, will be supported. Job creation, improved technology, institutional and policy environment will foster the sustainable development of the silk sector. The project will also empower poor rural women to become financially independent by providing training and extension services. [a,b] Bhutan Children in the poorest rural regions will benefit from a program aimed at providing Second Education Project full access to basic education. Teacher training, curriculum development, reading (IDA: US$13.7 m.) programs, and construction and rehabilitation of schools will be financed. The provision of boarding facilities in project schools for grades VII and VIII and the appointment of female teachers will further encourage girls' enrollment. [b] India Socially disadvantaged groups and the rural poor will benefit from a) greater access to Andhra Pradesh Economic primary education, particularly for girls, scheduled castes, and scheduled tribe Restructuring Project children; b) better primary health care focusing on young children and pregnant and (IBRD: US$301.3 m., lactating mothers; c) increased access to 250 nutrition centers and services for about IDA: US$241.9 m.) 2.1 million direct beneficiaries; and d) upgraded infrastructure, including rural roads, irrigation schemes, and agricultural services. [a,b] Third District Primary Education An estimated 5.5 million primary school children in 17 educationally disadvantaged Project (DPEP III) districts of Bihar will benefit from better access to schools and quality of instruction. (IDA: US$152 m.) The project supports infrastructure investments, teacher training, and improving state and district institutional capacity. Early childhood education centers will be established for children aged 0-6, literacy campaigns targeted at women will be launched, and grants to schools and communities will be provided to purchase supplies. Special attention will be given to the most socially disadvantaged groups including girls, scheduled castes, scheduled tribe children, working children, and children with learning disabilities. [a,b] 95 Annex D Country and project Reason(s) for inclusion' Orissa Health Systems Development The allocation and use of health resources will be made more efficient through policy Project and institutional development, with the objectives of improving the quality, (IDA: US$76.4 m.) effectiveness and coverage of health services. The project will target the poor, and women, especially those living in remote rural areas. It will also implement a strategy to address the specific needs of tribal populations. [a,b] Uttar Pradesh Forestry Project Poor general and scheduled caste villagers and tribal groups living in 1,000 forest (IDA: US$52.9 m.) fringe village communities and 230 communities in the plains will be the major beneficiaries of a project that is designed to improve access to forest products and grazing, as well as provide funds for improved agriculture and alternative income generating activities. [a,b] Uttar Pradesh Second Basic Education This project will increase access to elementary education, improve student learning, Project and enhance the govemment's management capacity in 10 districts. New schools will (IDA: US$59.4 m.) be constructed, additional teachers will be trained, learning materials will be distributed, and a new improvement fund will be established. [b] Women and Child Welfare Project The quality and cost-effectiveness of the Integrated Child Development Services (IDA: US$300 m.) (ICDS) program in the states of Kerala, Maharastra, Rajasthan, Tamil Nadu, and Uttar Pradesh will be supported in order to improve the nutrition and health status of pre-school children and women. Poor children aged 0-6 years old will be targeted, with particular emphasis on preventing malnutrition for those under 3 years of age. Child care practices and awareness of health and nutritional needs of poor women and adolescent girls will also be addressed. [a,b] Nepal Infrastructure development, along with associated institutional and operational Irrigation Sector Project measures, will ensure reliable and equitable distribution of water. Productivity, job (IDA: US$79.8 m.) opportunities and farm incomes are expected to increase for about 132,200 small and marginal farmers whose landholdings are less than 1.0 ha and 53,200 medium farmers whose landholdings are 1.0 to 3.0 ha. [b] Pakistan Over one million children in kindergarten to grade 8, living in poor and underserved Northem Education Project areas, will benefit from better trained teachers and the rehabilitation and (IDA: US$22.8 m.) reconstruction of schools. Specific actions will address gender concems, and community and NGO participation will be encouraged. [b] Second Social Action Project (SAPP Following up on the first SAPP launched in 1992, this project seeks to further reform II) the country's weak social services. It aims to improve the quality of and access to (IDA: US$250 m.) elementary education, primary health care, and rural water supply and sanitation, especially for the urban and rural poor. An important element is building government capacity for planning, monitoring and implementing social service programs, as well as increasing its transparency and governance. [b] Sri Lanka This project will target deprived schools in rural and urban slum areas and will Second General Education Project provide facilities and quality inputs to bridge the gap between poor and affluent (IDA: US$70.3 m.) schools. Over one million students will benefit from the installation of 2,000 new libraries, provision of books, instructional equipment and supplies, curriculum development, and adequate resources for school expenditures. [b] A4b~J~*r~m AfPW '- Angola This project will test a program designed to help war-affected rural communities Post-Conflict Social Recovery Project reintegrate displaced people and revitalize community-level economic and social (IDA: US$5 m.) activities. In line with the needs and priorities expressed by the communities, subprojects under the project's Municipio Recovery Program, to be funded on a grant basis, will seek to reestablish household livelihood and rehabilitate basic community-level social and economic infrastructure. [b] 96 Annex D Country and project Reason(s) for inclusion' Benin In addition to strengthening the capacity of village communities to design, prepare Borgou Region Pilot Rural Support and implement village development plans, the project will seek to meet the Project immediate needs of about 450,000 people living in Borgou. The project will (IDA: US$4 m.) promote increased access to services and basic productive and social infrastructure (e.g., storage facilities, rural roads and tracks, water points, wells). Women's groups and Fulani herders will be able to voice their concems in decision-making processes and investment opportunities. [a,b] Social Fund Project The project will finance subprojects that will help increase the poor's access to social (IDA: US$16.7 m.) services, financial services, and income-eaming opportunities. The target population will include poor rural and peri-urban communities, with a emphasis on women and unemployed youth. Local communities will be involved in the design and implementation of subprojects. [a,b] Burkina Faso Smallholder production will be increased through the provision of essential services Second National Agricultural Services such as fertilizer, farm chemicals, and improved seeds. The agricultural research Development Project system will be reorganized to better meet farmer's needs, and the link between 'IDA: US$41.3 m.) agricultural research and extension will also be strengthened. The project will help empower rural women by sponsoring gender-specific initiatives and literacy programs to ensure the effectiveness of extension programs. [b] Chad Using a community-based approach that includes community awareness campaigns, Household Energy Project this project aims to improve the living conditions of poor rural and urban households (IDA: US$5.3 m.) by providing an economical and sustainable supply of energy. [b] Comoros Increasing access to quality health services and controlling the exposure to malaria Health Project will improve family welfare and lower mortality and morbidity rates, particularly (IDA: US$8.4 m.) among women and children. Three of the four health facilities to be rehabilitated are on islands where poverty is most prevalent. [b] Social Fund Project Employment will be created and access to basic social services will be improved for (IDA: US$11.5 m.) the country's poorer communities. Funding will be available for demand-driven initiatives developed by local communities to improve essential services. Other subprojects will support the rehabilitation of feeder roads, bridges, markets, and the protection of the environment. [b] Cote d 'Ivoire Given the govemment's goal for universal primary education by 2010, children not Education and Training Support already in school, those at risk (of repetition or drop out) and girls will benefit from Project greater access to better quality schooling. About 40,000 additional places for primary (IDA: US$53.3 m.) school children, provision of textbooks, better student-teacher ratio, and reduced gender-bias in the curriculum will be supported. Altemative ways of mitigating the cost of schooling for parents will be piloted. [b] Ethiopia Small farmers with holdings of less than two hectares will benefit from a more Agricultural Research and Training responsive agricultural research program that will address their priority problems. Project Women farmworkers will also benefit from improved post-harvest technology and (IDA: US$60 m.) other technologies that will improve efficiency and reduce the drudgery of their farming activities. [b] Education Sector Development The restructuring of the education system supported by this project includes Program (ESDP) incorporating the use of local langtuages in school curriculum, expanding and (IDA: US$100 m.) improving the quality of primary and vocational education, especially in rural and undeserved areas, and eliminating school fees to reduce the financial burden on parents. Girls will benefit from gender-sensitive curricula and books, more female teachers and gender-appropriate facilities. [b] 97 Annex D Country and project Reason(s) for inclusion' Gambia About 84 percent of the Gambian population living in rural and peri-urban areas, Participatory Health, Population and particularly women of reproductive age and children under 5, will be targeted. This Nutrition Project project will improve the quality of reproductive health services (including STI/HIV (IDA: US$18 m.) prevention and control) and infant and child health services, as well as implement a family health program. The community will be involved, particularly in the micronutrient initiative to improve the nutritional status of women and children. [a,b] Guinea This project will help transform three existing financial institutions operating in rural Microfinance Project areas: Credit Rural, Credit Mutuel, and Le Programme Integre pour le (IDA: US$5 m.) Developpement de I'Entreprise (PRIDE), from donor-funded development projects into sustainable financial institutions. The project will help facilitate the provision of credit to poor farmers and artisans and will establish a Microfinance Support and Monitoring Unit (MSMU) to provide short-term technical advisory support and supervise the institutions to ensure their effective management. [b] Guinea-Bissau The health status and well-being of the population, particularly those of women and National Health Development children in poor areas, will be improved by programs to increase the coverage and Program quality of health services. Resources will be shifted to rural areas to ensure that a (IDA: US$11.7 m.) minimum package of health care including the provision of pharmaceuticals, medical supplies and health staff reach the poor. Family planning and AIDS prevention programs are also supported. [b] Madagascar Leaming outcomes and access to primary and secondary education will be improved Education Sector Development in the poorest rural areas through the distribution of textbooks and teaching guides, Project the provision of adequate infrastructure, furmiture and equipment, and staff (IDA: US$65 m.) development. A fund will be set up for the development of higher education, and local communities will be encouraged to participate. [b] Community Nutrition II Project The nutritional status of about 1.4 million direct beneficiaries, including 425,000 (IDA: US$27.6 m.) malnourished children and about 700,000 pregnant and lactating women will be improved. A school nutrition program will provide iron/folate supplements, anti- helminth treatment, and promote good hygiene practices. Information, Education and Communication (IEC) and community mobilization programs together with other community activities, will help ensure the long-term sustainability of nutritional outcomes. [a,b] Rural Water Supply and Sanitation Better hygiene and sanitation will benefit about 300,000 people in rural areas, Project contributing to the reduction in morbidity and mortality due to water-related (IDA: US$17.3 m.) diseases. Women and children will particularly benefit from time saved in the collection of water. [b] Mali To reduce poverty in the most vulnerable communities in Mopti and other regions Project to Support Grassroots this project will finance sub-projects that have the highest priority according to the Initiatives to Fight Hunger and communities, such as the construction of tubewells and classrooms. Emphasis will Poverty be given to including women in the village management committee, the key (IDA: US$21.5 m.) decision-making body at the community level, and also to giving them preference in the selection of village animation agents. [b] Mauritania Although the overall objective is to improve the health status of the general Health Sector Investment Project population, this project focuses on providing accessible and quality health services to (IDA: US$24 m.) the rural poor, women and children. [b] Niger A strategy for rural road maintenance, including the regraveling of 13 earth roads, Transport Infrastructure Rehabilitation will be developed to help ease the hardships faced by the poor. [b] Project (IDA: US$28 m.) 98 Annex D Country and project Reason (s) for inclusion' Senegal This program aims to reduce morbidity and mortality, particularly among poor Integrated Health Sector Development women and children below the age of 5 living in peri-urban slums and rural areas, Program increase the use of contraceptive methods nationwide, and create more cost-efficient (IDA: US$50 m.) and better managed public health facilities. Project activities include the construction and rehabilitation of health posts in densely populated poor areas and sparsely populated rural areas, the introduction of a new program to control endemic diseases, and the formulation of a strategy to help vulnerable groups (e.g., drug addicts and prostitutes). [a,b] Urban Development and Low-income urban neighborhoods, which currently face inadequate levels of services Decentralization Program and serious environmental risks, will benefit from reduced costs of and better access (IDA: US$75 m.) to municipal services and basic infrastructure. Employment will be generated by investments in public works activities (road, drainage, school and health facilities, markets, etc.). [b] Tanzania The project supports a govemment initiative to implement education sector reforms Human Resources Development Pilot aimed at raising enrollments and the quality of primary and secondary education and Project building capacity at the district and community levels to carry out these initiatives. (IDA: US$20.9 m.) The Community Education Fund will also support parental and community involvement in monitoring school performance. The Girl's Secondary Education Support Fund will give scholarships to girls from poor families who would otherwise not be able to attend secondary school. [a,b] Togo Agricultural productivity will be increased through an extensive program aimed at National Agricultural Services achieving food security and increasing exports. It includes improved cultivation Support Project techniques, implementation of priority research programs (crop and animal (IDA: US$26.2 m.) husbandry, improved planting materials, soil and water conservation, post-harvest storage and processing), increased credit access for smallholders and women producers, and the establishment of a Village Development Fund to support community-based infrastructure. [a,b] Uganda Large scale investment in early childhood development for children under six will Nutrition and Early Childhood enhance school readiness, the provision of child care arrangements will provide a safe Development Project environment for preschoolers to allow mothers to work, and the promotion of child (IDA: US$34 m.) care practices will help families fostering orphaned children (one in four households) whose parents have died of AIDS. [a,b] Zimbabwe The project aims to strengthen the capacity of poor communities to participate in Community Action Project local resource allocation, gain access to social and economic services, and manage (IDA: US$60 m.) natural resource endowments. The target population in subprojects are poorer communities in rural areas, with particular attention given to vulnerable groups (e.g., women, children, landless, elderly, unemployed). The project will operate in 26 (out of 57) rural districts with the lowest basic needs indicators according to the district ranking exercise developed for the project. [a,b] 99 ANNEXE. POVERTY-FOCUSED SALS AND SECALs, FISCAL 1998 Poverty-focused Adjustment Operations 1998. Table E-4 briefly describes the poverty-focused adjustment operations for fiscal 1998. Adjustment operations support macroeconomic and sectoral policies that promote growth and efficient Although the share in the number of operations resource allocation, which are essential for reducing classified as poverty-focused decreased between fiscal poverty. Some adjustment operations that contain 1997 and fiscal 1998, the total amount of Bank specific poverty reduction measures are referred to as lending for poverty-focused operations increased "poverty-focused operations." significantly during the same period. Poverty-focused operations as a share of all adjustment operations An adjustment operation is considered to be poverty- decreased from 60 percent in fiscal 1997 to 43 percent focused if it meets at least one of the three criteria in in fiscal 1998. However, funds going to poverty- supporting govemment efforts to: (i) reallocate public focused operations jumped from US$2.65 billion in expenditures in favor of the poor; (ii) eliminate fiscal 1997 to US$7.24 billion, owing to two very distortions and regulations that disadvantage the poor large loans to Korea totaling US$5 billion. For IDA and limit their income-generating opportunities; countries only, the volume of poverty-focused and/or (iii) support safety nets that protect the most adjustment lending decreased slightly from US$689 vulnerable members of the population.2 million in fiscal 1997 to US$630 million in fiscal 1998, but the share of poverty-focused lending in The Bank approved 37 adjustment operations in total IDA adjustment lending fell dramatically from fiscal 1998. Forty-three percent (16 of 37) of these 73 percent to 47 percent (reflecting a big jump in adjustment operations were classified as poverty- non-poverty-focused lending, from US$259 million focused. Table E-1 shows the number and amount of in fiscal 1997 to US$724 million in fiscal 1998). The lending for Structural Adjustment Loans (SALs), decline in the share of operations classified as Sector Adjustment Loans (SECALs), and poverty-focused for both IBRD and IDA reflects an Rehabilitation Import Loans (RILs). Table E-3 increase in fiscal 1998 in the number of loans for presents the number and amount of lending for institutional, private sector, and capital markets poverty-focused adjustment operations for fiscal 1992- development, and balance of payment support that did not include specific poverty-targeted components. Table E- 1. Poverty-focused adjustment operations, Almost half of the non-poverty-focused operations fiscal 1998 were in Latin America, including four loans to Mk:t : : ': '9 , gdifferent provinces in Argentina to reform and restore : 0 f ; i ff 0 t: fif 9 0 Total IDA IRD their government institutions. SAL lending (S mil.) 6,315 c10 5,805 Number of SALs 12 6 6 The regional spread of poverty-focused operations depends on regional priorities, which vary from year SECAL lending (S mi.) 910 110 800 to year. Poverty-focused lending in fiscal 1998 was Number of SECALs 3 2 1 concentrated in the East Asia and Pacific region, where an unprecedented amount of US$5.3 billion- RIL lending (S mi.) 1 0 10 0 74 percent of total poverty-focused adjustment Number of RILs I I 0 lending -was approved to finance loans in Korea and Malaysia in response to the crisis. The Europe and Central Asia region received US$1.6 billion for Total SAL/SECAL,URIL: nine operations, and the Africa region received Lending (S mil,) 7,235 630 6,605 US$360 million for four operations. Lending to the Number of loans 1 6 9 7 ECA and Africa regions mirrors the pattern of last fiscal year in that the bulk of fiscal 1997 funds (US$2.1 billion) went to the Europe and Central Asia region while Sub-Saharan Africa received the remainder (US$565 million). Nearly all the poverty- 2 In the past, support to poverty monitoring or the development of focused adjustment loans in fiscal 1998 included poverty policies were also included as criteria. These have been safety nets measures, and all the poverty-focused dropped. This does not affect comparability with past figures, operations in Africa emphasized reallocating however, because no operation was ever classified as poverty- expenditures toward the social sectors. The following focused on either of these two criteria alone. 100 Annex E three areas represented the main objectives addressed regulatory framework. In the social sectors, regulatory by poverty-focused SALs, SECALs, and RILs in barriers to the provision of health and education in fiscal 1998 (see Table E-2, which indicates the outlying provinces will be removed; the regulatory poverty-focused objectives of each operation). framework for the import and distribution of essential and generic drugs will be revised; and regulatory Public expenditure reforms barriers that artificially inflate the price of textbooks will be eliminated. The govemment will facilitate The majority of poverty-focused operations (10 of 16) women's access to appropriate technologies at a low will reform public expenditures to benefit the poor in cost, as well as eliminate obstacles resulting from different ways. The Armenia Second Structural lack of access to land and credit. In Korea, the Bank Adjustment Credit (IDA) will improve the access to approved two loans in fiscal 1998 to help overcome and quality of education and health services for the immediate liquidity problems resulting from the East poor by supporting the government's reallocation of Asia economic crisis and to lay the foundation for a spending away from higher education towards lower medium term program of financial and real sector levels of education, as well as outlining plans to reforms. An unprecedented amount of US$3.0 billion increase funding for primary health care. Despite a was provided under the Economic Reconstruction tight fiscal program, the share of allocations for both Loan (IBRD) and another US$2.0 billion was added the education and health sectors was set at no less under the Structural Adjustment Loan (IBRD). than 9.3 percent of total current expenditures, and was Among many initiatives, the first operation will slated for an increase in the next budget cycle. In support measures to improve labor market flexibility: Guinea, the Public Expenditures Management bottlenecks to a well-functioning labor market will be Adjustment Credit (IDA) will implement a new addressed, including a rigid employment security approach to managing public expenditures in order to program that makes it difficult to shed workers in achieve financial sustainability and improve public declining industries, the lack of a legal framework for sector effectiveness in four priority sectors. Public private provision of temporary manpower leasing expenditures will be reoriented toward education, services, restrictions on the scope of private job health, road maintenance, and support for rural placement services, and strict job security regulations development, which will benefit the majority of the that deter employers from hiring new workers. The poor. A decentralized, integrated community-based second operation will also expand protection for approach to service delivery will be adopted, and a workers working in small firms (fewer than five set of sector-specific performance indicators will be employees) by extending the coverage of the labor monitored to assess the impact of policies on standards act, which grants workers important rights. economic performance and poverty reduction. The In addition, regulations that impose high costs on Education Sector Adjustment Operation (IDA) in women's employment will be reviewed in order to Uganda seeks to better expand universal primary reduce gender inequalities in the labor market. education, as well as to ensure better access for the poor, girls, and those living in rural areas by Targeted programs and safety nets increasing resources to primary education and promoting more efficient use of these resources. All but one poverty-focused operations in fiscal 1998 Primary education spending is expected to double in will support targeted programs and safety nets. The real terms between fiscal 1995-1996 and fiscal 1998- Albania Rehabilitation Credit (IDA) will strengthen 1999, with allocations to primary education the social safety net and support measures to promote increasing from 13 percent of total government budget employment following the collapse of pyramid in fiscal 1996/1997 to 18 percent in fiscal 1999/2000. schemes that left many households impoverished. To Measures to promote the efficient use of resources provide adequate and timely payment of social include double shift teaching to increase efficiency in assurance benefits, the govemment will pay the full teacher and classroom utilization, as well as multi- amount budgeted for social assistance to local grade teaching to provide education services to authorities in the form of block grants and will children in remote areas. A system for monitoring the introduce a 10 percent increase in social assistance to accountability of public funds allocated to districts poor families in the next budget cycle. Resources will and schools will be implemented. be earmarked for improving welfare institutions that provide care to orphans and the elderly, as well as for Addressing distortions that especially affect the poor employment programs and services that help the unemployed obtain jobs. The Post-Conflict Three poverty-focused operations approved in fiscal Rehabilitation Credit (IDA) in Tajikistan will 1998 addressed distortions that especially affect the strengthen the funding of the social safety net for poor. The Third Structural Adjustment Credit (IDA) children and pensioners through the payment of in Cameroon will support reforming the incentive and arrears, particularly to offset the war's adverse impact 101 Annex E on the poor and to help increase purchasing power in addition to increasing expenditures on major anti- towns and villages. Under the Public Sector poverty programs (Fund for Food, small-scale loans Adjustment Loan (IBRD) in Htngary, low-income to the poorest in rural areas for income-generation, pensioners will be protected by new features of the improving water supply in the poorest rural areas), public pension scheme, and a, social assistance safety net programs will be expanded to cover those program will be set up for the elderly poor who do who have become vulnerable as a result of the recent not qualify for a minimum pension. The Economic crisis. Programs include urban micro-credit programs Recovery and Social Sector Loan in Malaysia (IBRD) for hawkers, traders, and small entrepreneurs, and will help strengthen the social safety net to mitigate provision of special skills training through public and social tensions due to the economic crisis. In private entities to retrain retrenched workers. 102 Annex E Table E- 2. Poverty-focused components of adjustment operations, fiscal 1998 Objectives' Reforming Tranche public Addressing Targeted release expenditurest distortionsc programs conditionse Fiscal 1998 Poverty-focused adjustment operations Structural Adjustment Loans Albania: Rehabilitation Credit* 4 4 Armenia: Second Structural Adjustment Credit* a i Azerbaijan - Structural Adjustment Credit* 4 Cameroon: Third Structural Adjustment Credit* 4 4 4 Georgia: Structural Adjustment Credit* 4 4 Guinea: Public Expenditure Management Adjustment Credit* 4 4 Hungary: Public Sector Adjustment Loan A Kazakhstan: Pension Reform Adjustment Loan 4 4 Korea: Economic Reconstruction Loan 4 4 Korea: Structural Adjustment Loan 4 4 4 Malaysia: Economic Recovery and Social Sector Loan N 4 Moldova: Second Structural Adjustment Credit and Loan 4 4 Sectoral Adjustment Loans Cape Verde: Economic Reforms Support Operation Credit* 4 4 Russia: Second Coal Sector Adjustment Loan 4 V Uganda: Education Sector Adjustment Operation* N 4 4 Rehabilitation Import Loans Tajikistan: Post-Conflict Rehabilitation Credit N 4 Total 10 3 15 8 * IDA credit a. Nine operations in fiscal 1998 also support the collection of data on poverty and the monitoring of the impact of adjustment on the poor: the Azerbaijan Structural Adjustment Credit, the Cape Verde Economic Reforms Support Operation Credit, the Georgia Structural Adjustment Credit, the Guinea Public Expenditure Management Adjustment Credit, the Korea Economic Reconstruction Loan, the Korea Structural Adjustment Loan, the Malaysia Economic Recovery and Social Sector Loan, the Russia Second Coal Sector Adjustment Loan, and the Uganda Education Sector Adjustment Operation. b. Supports the reallocation of public expenditures towards physical infrastructure and basic social services for the poor. c. Focuses specifically on reducing distortions that especially disadvantage the poor. d. Supports programs that provide safety nets or that target specific groups. e. Contains tranche release conditions related to poverty-focused measures. Source: World Bank. 103 Annex E Table E- 3. Poverty-focused adjustment lending, fiscal 1992-1998 Lending FY92 FY93 FY94 FY95 FY96 FY97 FY96S Poverty-focused adjustment lending (US$ millions) 2,838 1,165 1,665 1,648 2,227 2,649 7,235 Total adjustment lending 5,847 5,253 2,868 5,324 4,509 5,086 11,289 Percentage of adjustment lending 49 22 58 3 1 49 52 64 Total number of poverty-focused adjustment operations 18 6 2C 14 17 1 I 1 6 Total number of adjustment operations 32 23 28 3C 3 C 3 C 37 Percentage of poverty-focused adjustment operations 56 26 7 1 47 57 6 C 43 IDA Poverty-focused adjustment lending (US$ millions) 1,168 645 875 598 1,027 689 63C IDA adjustment lending 2,152 1,423 1,998 1,069 1,679 948 1,354 Percentage of IDA adjustment lending 54 45 44 56 61 73 47 Total number of IDA poverty-focused adjustment 1 C 3 11 9 13 9 9 operations Total number of IDA adjustment operations 1 6 8 18 1 5 1 9 11 17 Percentage of IDA poverty-focused adjustment 63 38 61 6C 68 82 53 operations Note: Adjustment operations include SALs, SECALs, RILs, and DRLs. For joint IBRD/IDA operations, the amount of lending is split between IBRD and IDA as stipulated in the loan/credit document, but it is counted only once, as an IBRD operation. 104 Annex E Table E- 4. Poverty-focused SALs and SECALs, fiscal 1998 Loan or credit Poverty-related objectives East Asia and the Paciftc Korea In response to the recent East-Asia economic crisis, this loan supports government Economic Reconstruction Loan efforts to deal with the immediate liquidity crisis, as well as lay the foundation for (IBRD: US$3000 m.) a medium-term program of broader financial and real sector reforms. Labor market impediments to industrial restructuring will be removed and a program of measures to improve labor market flexibility will be prepared. This will promote redeployment and deregulate private provision of job placement services and manpower leasing services. Measures to ease the social costs of adjustment will include reviewing the adequacy of fiscal expenditures on social safety nets (including unemployment insurance benefits), assessing long-term financial viability of the social security programs, and reporting on the level and distribution of poverty to prepare an appropriate mitigation plan. Structural Adjustment Loan This loan builds on the work initiated under the Economic Reconstruction Loan (IBRD: US$2000 m.) and is part of an intemationally coordinated support package for Korea. In addition to helping Korea overcome the immediate liquidity problems and regain investor confidence, the SAL will support a program of wide-ranging and deep reforms aimed at bringing about a major restructuring of the incentive and institutional framework to underpin financial stability and improve efficiency and competitiveness. The loan supports the following programs that will especially benefit disadvantaged groups: a) relaxing legal restriction to improve labor market flexibility and facilitate redeployment; b) increasing worker protection by extending coverage of the Labor Standard Act, which grants important rights to workers in firms with fewer than five employees; c) reducing gender inequalities in the labor market by reviewing regulations that impose high costs on women's employment; d) improving targeting of anti-poverty programs and maintaining the level of real expenditures per beneficiary; e) expanding in real terms the budget allocations to anti-poverty programs; f) improving poverty monitoring; and g) reforming the pension system to provide better security in income support for the elderly, and (h) extending unemployment insurance coverage to firms with 5-10 workers. Malaysia This operation supports a preemptive reform program to minimize the downturn in Economic Recovery and Social economic activity, including the strengthening of social safety nets in order to Sector Loan soften the negative impact of the crisis on lower-income and vulnerable groups. (IBRD: US$300 m.) The budget shares for education and health will be protected. To help the hard-core poor and near-poor, public expenditures will be increased to support major anti- poverty programs, including a food program, a rural and urban microcredit program, and a rural water supply program. Assistance will also be provided to small and medium industries to improve productivity and competitiveness. Europe and Central Asia Albania Financial support will help reform the banking sector, strengthen the social safety Rehabilitation Credit net, and promote employment. To provide adequate and timely payment of social (IDA: US$25 m.) assistance benefits, the government will pay the full amount budgeted for social assistance to local authorities in the form of block grants and will introduce an increase of ten percent for social assistance for poor families in the next budget cycle. Support will also be given to improve welfare institutions for the elderly and orphans. Employment programs and services will be strengthened to help the unemployed obtain stable jobs. 105 Annex E Loan or credit Poverty-related objectives Armenia The credit will support the govemment's program to consolidate macroeconomic Second Structural Adjustment stability and to lay the foundations for sustained private-sector-led growth over the Credit medium term, as well as measures to improve the targeting of social benefits to (IDA: US$60 m.) vulnerable populations and strengthen the social safety net (including reforms in the public pension system and in the health and education sectors). To ensure that the poor have adequate access to quality education and health services, the government is reallocating spending away from higher education towards general education, and is outlining plans to increase funding for primary health care. The share of appropriations for both the education and health sectors was set at no less than 9.3 percent of total current expenditures and was slated for an increase in the next budget cycle. Azerbaijan The credit will support the govemment's effort to liberalize trade, demonopolize Structural Adjustment Credit and privatize enterprises, reform the banking system, and protect vulnerable groups (IDA: US$70 m.) from utility price increases. The targeting of subsidies under the cash compensation scheme and communal services (electricity, gas, and water) for pensioners and poor households will be improved. Georgia In a(ldition to supporting programs to strengthen economic recovery, the credit Structural Adjustment Credit will also support measures to improve the targeting of social benefits. Reforms in (IDA: US$60 m.) health, education, and pension schemes will ensure equitable access to essential services and protect the most vulnerable. Conditions for second tranche release call for the govemment to increase the share of total public expenditures allocated to health and education in the 1998 budget, adopt a hospital masterplan to determine which hospitals should be privatized, and enact the Law on Private Pensions to address the sustainability and provision of adequate income for the elderly. Hungary This operation will support the implementation of a comprehensive reform of the Public Sector Adjustment Loan pension system. A social assistance program will be set up for the elderly poor (IBRD: US$150 m.) who do not qualify for a minimum pension. Low-income pensioners will be protected by new features of the public scheme. Kazakhstan Support for pension reforms in Kazakhstan include establishing laws, regulations Pension Reform Adjustment and institutions to develop a sound private pension system and strengthen the Loan management of existing schemes. A resolution providing for scheduled increases in (IBRD: US$300 m.) the minimum pension adjusted for inflation, together with the establishment of old-age allowances, will help ensure sufficient income support for pensioners. A study will be undertaken to measure the adequacy of pensions for different groups of employees and recommend actions to address identified retirement income deficiencies. Moldova Critical reforms will be undertaken to ensure macroeconomic stability, including Second Structural Adjustment land reform, enterprise privatization, energy sector reforms, pension and social Credit and Loan protection reforms. These reforms include measures to offset the negative impact on (IBRD: US$55 m., the most vulnerable groups, such as ensuring that arrears to the population are paid IDA: US$45 m.) promptly, avoiding delays in the payment of pensions, and targeting compensation to the poor and unemployed to mitigate the impact of energy price increases. The introduction of a pension formula affecting contributions and benefit levels will ensu-e minimum benefit levels to the poorest. Before the third tranche release, target measures to make energy affordable to the poor and vulnerable groups must be in place. 106 Annex E Loan or credit Poverty-related objectives Russia This loan extends the Bank's support for restructuring a critical sector of the Second Coal Sector Adjustment Russian economy that is still dependent on government subsidies. Because the Loan transition to a more efficient industry can impose hardships on employees of coal (IBRD: US$800 m.) enterprises, this operation also supports social safety net programs that will develop the institutional capacity to deliver social benefits in a transparent and equitable manner. Those most affected by the restructuring -redundant coal workers, their families and communities - will be targeted. Before the first floating tranche is released, a report on the social impact of the coal-restructuring program (based on predetermined social indicators) will have to be issued. Tajikistan The proposed credit will help ease the foreign exchange shortage so that critical Post-Conflict Rehabilitation imports can be acquired to restore production, employment and consumption in Credit Tajikistan's post-conflict economy. Strengthening the funding of the social safety (IDA: US$10 m.) net for these groups through payment of arrears will partially offset the war's adverse impact on the poor and help increase purchasing power in towns and villages. The government is committed to protect social spending and has allocated over a quarter of the 1998 total expenditures to education, health, and safety net programs. Sub-Shahan Africa Cameroon This adjustment program will provide an enabling environment for growth and Third Structural Adjustment support initiatives to create jobs and reduce poverty. Sector specific strategies Credit (transport, agro-industries, financial, forestry and social) will address inefficiencies, (IDA: US$180 m.) regulatory barriers, and legislation that inhibit growth and prevent the poor from accessing basic services. Priority will be given to facilitating access and improving the quality of services for the poor, women, children, and the elderly living in underserved areas. Among the key actions proposed are the support for labor- intensive programs to generate employment opportunities and the increase of capital and recurrent expenditures for improving programs in primary health and education. Cape Verde Balance of payments will be financed to support the government's adjustment Economic Reforms Support program to promote private sector development by accelerating privatization, Operation Credit further liberalizing the economy and maintaining a stable macroeconomic (IDA: US$30 m.) environment. The credit will support the recycling and retraining of redundant workers, as well as job creation for the unemployed. It will also improve budget allocations for social sectors and welfare programs in order to reduce poverty. To monitor the operation's impact on growth and poverty reduction, a number of indicators such as life expectancy, and pre-natal mortality will be monitored within the framework of quantitative targets for the year 2000. Prior to the second tranche release, there must be satisfactory progress in the implementation of the three-year rolling Public Investment Program (PIP), a public financial management system overseeing budget allocations to social sectors and investments in infrastructure. Guinea In order to help Guinea ensure financial sustainability and improve public service Public Expenditure Management effectiveness in sectors that benefit the majority of the poor the credit will reorient Adjustment Credit (PEMAC) public expenditures by increasing non-wage recurrent budget allocation to four (IDA: US$70 m.) priority sectors from 25 percent to 30 percent. Priority areas include primary education, particularly for girls, primary health care, particularly in remote areas, rural development, especially in food security, and the maintenance of rural infrastructure. Conditions for second tranche release call for the disbursement of non-wage recurrent resources of at least GNF 3.8 billion for education, GNF 2 billion each for public health and road maintenance, and GNF 635 million for rural development in the first half of 1998, recruitment of at least 1,600 additional primary school teachers, and expenditure of GNF 5,000 per primary school student on pedagogical materials. 107 Annex E Loan or credit Poverty-related objectives Uganda The oudgetary support required for the government to implement its Universal Education Sector Adjustment Primary Education Policy will be provided, including programs that will make Operation more efficient use of resources, improve the quality of teaching, expand the (IDA: US$80 m.) provision of physical infrastructure, and strengthen sector management. In addition, recent actions to curb the inequity of education subsidies will allow a greater share of government expenditures to be allocated to primary education. Improvement and expansion of the educational system will especially benefit children from the poorest families, girls, and those who traditionally have not been able -o attend school. A third tranche release condition includes the implementation of a monitoring system for the accountability of public funds allocated to districts and schools, as well as provision of evidence that randomly selected schools have received shipment of instructional materials. 108 Annex F ANNEXF. POVERTY-FOCUSED ERLs, FISCAL 1998 Emergency Recovery Loans (ERLs) are designed to apply the poverty criteria for investment lending to provide immediate assistance after major calamities ERLs. By nature, ERLs in fact resemble adjustment such as wars, civil disturbances or natural disasters. more than investment operations; thus, the poverty They are usually identified, prepared, approved and criteria established for adjustment operations also completed within a short period, normally three apply to ERLs. The Bank approved nine Emergency years. They are processed and often disbursed Recovery Loans in fiscal 1998, of which eight were quickly. Although considered to be projects and, by poverty-focused. Five out of eight operations were in convention, included in the Bank's investment direct response to the El Niffo event, and financed the lending total, ERLs are different from regular reconstruction of public infrastructure, investments in investment operations in objective and format. Some preventive measures, and the resettlement of mostly ERLs provide critical balance of payments support poor populations from flood-prone areas. Table F-i and help lay the foundation for implementing presents the number and amount of lending for economic reforms while others support programs to poverty-focused ERLs for fiscal 1992-1998. Table F-2 reform social sector expenditures and safety nets. By describes the poverty-focused ERLs for fiscal 1998. the very concept of their design, it is not feasible to Table F-1. Poverty-focused ERL lending, fiscal 1992-1998 Lending FY92 FY93 FY94 FY95 FY96 FY97 FY98 Poverty-focused ERL lending (US$ millions) 16 285 13 0 1oo 98 316 Total ERL lending 316 840.2 387.1 240 110 170.6 516 Percentage of total ERL lending 5 34 3 0 9 1 57 6 1 Total number of poverty-focused ERLs I 1 1 0 1 5 8 Total number of ERLs 3 8 3 4 3 8 9 Percentage of poverty-focused ERLs 33 1 3 33 0 33 63 89 IDA Poverty-focused ERL lending (US$ 16 0 1 3 0 0 98 101 millions) IDA ERL lending 316 142 259 90 10 171 101 Percentage of IDA ERL lending 5 0 5 0 0 57 100 Total number of IDA poverty-focused ERLs 1 0 1 0 0 5 5 Total number of IDA ERLs 3 3 2 2 2 8 5 Percentage of IDA poverty-focused ERLs 33 0 50 0 0 63 100 109 Annex F Table F-2. Poverty-focused ERLs, fiscal 1998 Loan or credit Poverty-related objectives China Assistance will be provided to an estimated 250,000 earthquake victims, most of Hebei Earthquake Rehabilitation whom are poor, to repair or reconstruct their homes, schools, drinking water facilities Project and health centers. The construction of grain storage facilities and the installation of (IDA: US$28 m.) wells and small sprinkler irrigation systems will help restore productive economic activities. Papua New Guinea This operation will mitigate the hardships suffered by smallholders hardest hit by the El Nino Drought Response Project drought by improving the supply of rural potable water to the neediest areas and (IBRD: US$5 m.) developing infrastructure maintenance programs with full participation of rural communities. Employment will be generated through labor-intensive rural public works projects to help protect the living standards of villagers. Assistance to women's groups in health awareness, education, and management of village water will be provided Bosnia and Herzegovina This operation will help post-war reconstruction efforts to rebuild infrastructure Emergency Natural Gas System facilities and restore safe and reliable gas supply to people living in Sarajevo and Reconstruction Project other towns, many of whom are very poor and living in temporary shelters. This (IDA: US$10 m.) will include replacing home made heating and cooking devices of low-income families with simple wall heaters and plate cookers, thereby ensuring safety and reducing environmental pollution. Tajikistan In support of the government's Peace Agreement, this project will help reintegrate Post-Conflict Emergency communities that were cut off and marginalized by the war into the national Reconstruction Project economy. This will be accomplished through targeted programs for reconstruLcting (IDA: US$10 m.) physical and social infrastructure and emergency support to jump-start agriculture production. War-damaged roads, bridges, schools, and health care facilities will be reconstructed. Bolivia It is estimated that about 600,000 persons were affected by the drought, and another El Nino Emergency Assistance 350,000 person by severe flood conditions as a result of the recent El Nifho weather Project phenomenon. Widespread damage to infrastructure and losses in agricultural (IDA: US$25 m.) production has exacerbated the socio-economic situation in a country that already has very high levels of poverty. This project will reconstruct public infrastructure facilities (e.g., roads, health posts, schools, irrigation systems); regenerate productive capacity in drought areas (e.g., drilling, wells, rehabilitation of irrigation systems, procurement and distribution of seeds); finance flood-preventive works (e.g., building dikes, cleaning storm drains) and drought-mitigation works (installing pumps, building dams); and strengthening institutions to plan and manage emergency response. Ecuador Assistance will be provided to people who were adversely affected by the El Ni-no El Nino Emergency Recovery weather phenomenon. Financing for subprojects will help rebuild public Project infrastructure damaged by recent floods, invest in flood preventive measures, and (IBRD: US$60 m.) support the resettlement of people from high-risk flood areas. Special assistance will be provided to poor families, especially when the head of those households are women or ethnic minorities. Reconstruction subprojects will be designed to have high labor content in order to provide more employment opportunities for rural laborers. The capability of the government to forecast and respond to future natural disasters will be strengthened. 110 Annex F Loan or credit Poverty-related objectives Peru This operation will help minimize the loss of human life and damage to economic El Nifio Emergency Assistance infrastructure resulting from the El Nin'o weather phenomenon. It will be Project implemented in three phases: a) a prevention phase (flood protection and drought (IBRD: US$150 m.) mitigation), b) an emergency phase (emergency provisions during flood disasters), and c) a reconstruction phase (reconstruction of public infrastructure, resettlement of families in flood prone areas, and labor-intensive rural works program). Sub-Saharan Africa Uganda Investments to restore key road and bridge infrastructure facilities severely damaged El Nino Emergency Road Repair by floods from the El Nifio weather pattern will reestablish the flow of goods and Project services countrywide. Project activities involving the Atiak-Moyo road of Northern (IDA: US$28 m.) Uganda will significantly improve accessibility to one of the more remote and poverty-stricken regions of the country. 111 ANNEX G. ANNUAL LENDING TO SELECTED SECTORS, FISCAL 1998 While all World Bank lending is intended to help 1995 to 21 percent between fiscal 1996-1998. reduce poverty, lending to certain sectors supports the efforts of governments to reduce poverty more Within the human capital development sector, there directly. The human capital development, agriculture, has been a steady increase in the volume of lending water supply and sanitation sectors contain a higher allocated to population, health, and nutrition, and percentage of projects with elements that directly social protection. Although the volume of lending to target the poor. As shown in Figure G-1, average the education sector has fluctuated slightly, the lending for human capital development increased overall trend shows about a three-fold increase more than seven-fold in dollar terms between the between fiscal 1981-1998. early 1980s and fiscal 1998. The share of total Bank lending to these sectors increased from an average of 5 Between fiscal 1993-1995 and fiscal 1996-1998, the percent during fiscal 1981-1983 to an average of 21 share of lending to the agricultural and rural percent during fiscal 1996-1998 (for IDA, the share development and water and sanitation sectors dropped increased from 8 percent to 31 percent) (see Table G- from an average of 13 percent to 12 percent and from 1). In recent years, the share of lending to the human an average of 5 percent to 3 percent respectively. capital development sectors increased from an average Table G-2 provides data for these selected sectors on of 17 percent of total lending betveen fiscal 1993- an annual basis for the last seven fiscal years. Figure G-1. Trends in lending for human capital development, fiscal 1981-1998 6000 - Human Capital Development Education 5000 - Pop/Health/Nutrition -x- Social Protection 4939 4000 3 2 3000 / ~~~~2014 2000 - 1951 / 1689 *1761 1000 671 615~~~28 5 6 0 '- ---.X->O - '---- '1 FY81-83 FY84-86 FY87-89 FY90-92 FY93-95 FY96-98 Fiscal Year 112 Annex G Table G-1. Average lending to selected sectors, fiscal 1981-1998 FY81-83 FY84-86 FY87-89 FY90-92 FY93-95 FY96-98 World Bank (IBRD and IDA) lending (US$ millions) Human capital development 670.7 1,110.6 1,085.2 2,935.8 3,753.1 4,938.9 Education 615.0 826.1 765.2 1,688.6 2,014.2 1,950.8 Population, health, and nutrition 55.6 284.5 303.0 1,024.3 1,071.8 1,761.4 Social protection a n/a n/a 17.0 222.9 667.1 1,226.7 Agriculture 3,583.9 3,942.0 3,649.6 3,210.9 2,861.0 2,780.9 Water supply and sanitation 595.6 686.1 758.6 965.3 1,080.2 615.2 Total 13,261.3 15,408.5 19,420.5 21,697.6 22,351.2 23,031.0 As share of total (percent) Human capital development 5 7 6 14 17 21 Agriculture 27 26 19 15 13 12 Water supply and sanitation 4 4 4 4 5 3 IDA lending (US$ millions) Human capital development 267.5 450.6 471.9 1,464.1 1,737.6 1,958.6 Education 240.6 298.1 308.2 748.3 787.0 747.2 Population, health, and nutrition 40.5 152.5 146.7 545.2 570.5 877.2 Social protection a n/a n/a 17.0 170.6 380.1 334.2 Agriculture 1,242.1 1,281.8 1,362.3 1,435.8 1,318.7 1,026.0 Water supply and sanitation 107.4 110.5 175.1 321.3 269.3 165.8 Total 3,169.7 3,247.7 4,292.7 6,121.7 6,337.6 6,331.2 As share of total (percent) Human Capital Development 8 14 11 24 27 31 Agriculture 39 39 32 23 21 16 Water supply and sanitation 3 3 4 5 4 3 Note: The data are for average annual lending during the three-year period indicated. The World Bank's fiscal year runs from July I of the previous year to June 30 of the year indicated. Three-year moving averages have been reported since the first Progress Report on Poverty in fiscal 1992 to smooth out year-to-year fluctuations. Because of a recent sector reclassification of projects, some numbers may differ from those reported in earlier tables. Note that these sectors do not account for all poverty-focused lending; projects in such sectors as urban development and transport also have components designed to help reduce poverty. a. Social protection lending includes employment, social assistance, social insurance, and social investment funds. 113 Annex G Table G-2. Annual lending to selected sectors, fiscal 1992-1998 Bank Lending (US$ millions) FY92 FY93 FY94 FY95 FY96 FY97 FY9S World Bank (IBRD and IDA) lending (US$ millions) Human capital development 2,715.4 3,814.3 3,241.C 4,204.1 5,053.6 3,327.5 6,435.7 Education 1,666.5 1,871.2 2,119.2 2,052.2 1,705.7 1,017.4 3,129.3 Population, health and 922.1 1,228.6 885.7 1,101.2 2,353.4 939.9 1,990.9 nutrition Social protection a 126.8 714.5 236.1 1,050.7 994.5 1,370.2 1,315.5 Agriculture 3,209.9 2,825.8 3,551.8 2,205.3 2,078.9 3,546.5 2,717.4 Water supply and sanitation 786.4 1,283.9 975.2 981.5 609.8 682.8 552.9 Total 21,705.7 23,695.9 20,836.C 22,521.8 21,352.2 19,146.7 28,594.0 As share of total (percent) Human capital development 13 16 16 19 24 17 23 Agriculture 15 12 17 10 10 19 10 Water supply and sanitation 4 5 5 4 3 4 2 IDA lending (US$ millions) Human capital development 1,297.8 2,139.5 1,245.1 1,828.2 2,197.6 1,015.7 2,662.5 Education 561.3 970.2 619.3 771.6 784.9 255.1 1,201.5 Population, health and 615.1 541.8 519.7 649.9 858.2 694.1 1,079.4 nutrition Social protection a 121.4 627.5 106.1 406.7 554.5 66.5 381.6 Agriculture 1,219.2 1,084.0 1,520.0 1,352.0 1,105.1 735.9 1,236.9 Water supply and sanitation 297.4 395.4 103.2 309.2 80.7 302.4 114.3 Total 6,549.7 6,751.4 6,592.1 5,669.2 6,864.1 4,621.8 7,507.8 As share of total (percent) Human capital development 20 32 19 32 32 22 35 Agriculture 19 16 23 24 16 16 16 Water supply and sanitation 5 6 2 5 1 7 2 Note: The World Bank's fiscal year runs :fom July I of the previous year to June 30 of the year indicated. These sectors do not account for all poverty-focused lending; projects in such sectors as urban development and transport also have components designed to help reduce poverty. See World Bank Annual Report 1998 for further details on sector lending. Because of a recent sector reclassification of projects, some numbers may differ from those reported in earlier tables. a. Social protection lending includes employment, social assistance, social insurance, and social investment funds. 114 References Ahuja, Vinod, B. Bidani, F. Ferreira, and M. Walton. Dudwick, Nora. 1998. "Participatory Poverty Assessments 1997. Everyone's Miracle? Revisiting Poverty and In- in ECA: A Review." World Bank, Environmentally and equality in East Asia. Washington, D.C.: World Bank. Socially Sustainable Development Sector Unit, Wash- Bamberger, Michael, ed. Forthcoming. Integrating Quan- ington, D.C. titative and Qualitative Research In Development Pro- Ezemenari, Kene, Anders Rudqvist, and K. Subbarao. jects. A Directions in Development book. Washington, 1998. "Impact Evaluation: A Note on Concepts and D.C.: World Bank. Methods." World Bank, Poverty Division, Washington, Blejer, Mario, and Isabel Guerrero. 1990. "The Impact of D.C. Macroeconomic Policies on Income Distribution: An Fajnzylber, Pablo, Daniel Lederman, and Norman Loyaza. Empirical Study of the Philippines." Review of Eco- 1998. "What Causes Violent Crime?" World Bank, Of- nomics and Statistics 72 (August): 414-23. fice of the Chief Economist, Latin American and Booth, David, Jeremy Holland, Jesko Hentschel, Peter Caribbean region, Washington, D.C. Lanjouw, and Alicia Herbert. 1998. Participatory and Ferreira, Francisco. 1995. The Dynamic Impact of Expen- Combined Methods in African Poverty Assessments: diture Reduction on Income Distribution in a Dual Renewing the Agenda. London: Department for Inter- Economy. Economics Department, London School of national Development. Economics, London. Bourguignon, Francois, and Christian Morrisson. 1992. Ferreira, Francisco, and Louise Keely. 1998. "The World Adjustment and Equity in Developing Countries. De- Bank and the Process of Structural Adjustment: The velopment Centre Studies. Paris: Organisation for Eco- 1980s and 1990s." World Bank, Poverty Division, Wash- nomic Co-operation and Development. ington, D.C. Blackden, C. Mark, and Chitra Bhanu. 1999. "Gender, Growth, Ferreira, Francisco, and Julie Litchfield. 1998. "Education and Poverty Reduction: 1998 Status Report on Poverty for or Inflation? The Roles of Structural Factors and Macro- the Special Program of Assistance for Africa." World Bank economic Instability in Explaining Brazilian Inequal- Technical Paper 428. Washington, D.C.: World Bank. ity in the 1980s." DARP 41. London School of Economics Braithwaite, Jeanine. 1998. "Poverty in Russia during the and STICERD, London. Transition: An Overview." InJeni Klugman, ed., Poverty Hicks, Norman. 1998. "Measuring the Poverty Impact of in Russia: Public Policy and Private Responses. Eco- Projects in LAC." Latin America and Caribbean Qual- nomic Development Institute Development Study. ity Enhancement Series. World Bank, Latin America and Washington, D.C.: World Bank. Caribbean Region, Washington, D.C. Demery, Lionel, and Michael Walton. 1998. ArePoverty Re- Hentschel, Jesko. 1998. "Distinguishing between Types of duction and Other 21st Century Social Goals Attain- Data and Methods of Collecting Them." Policy Re- able? Washington, D.C.: World Bank. search Working Paper 1914. World Bank, Poverty Re- DAC/OECD (Development Assistance Committee of the Or- duction and Economic Management Network, Poverty ganisation for Economic Co-operation and Develop- Division, Washington, D.C. ment). 1996. Shaping the 21st Century. Paris: Jayarajah, Carl, William Branson, and Binayak Sen. 1996. Organisation for Economic Co-operation and De- Social Dimensions of Adjustment: World Bank Expe- velopment. [http://www.oecd/org/dac/Indicators/ rience, 1980-93. Operations Evaluations Study. Wash- htm/backgrd.htm]. ington, D.C.: World Bank. 115 116 POVERTY REDUCTION AND THE WORLD BANK: PROGRESS IN FISCAL 1998 Khandker, Shahidur R. 1998. Fighting Poverty with Mi- - Forthcoming. Can the Poor Influence Policy? Par- crocredit: Experience in Bangladesh. New York: Ox- ticipatoiy Assessments in the Developing World. Wash- ford University Press. ington, D.C.: World Bank. Kozel, Valerie, and Barbara Parker. 1998. "Poverty in Rural Salmen Lawrence F. 1995. "Participatory Poverty Assess- India: The Contribution of Qualitative Research in ments: Incorporating Poor People's Perspectives in Poverty Analysis." World Bank, Poverty Reduction and Poverty Assessment Work." Environment Department Economic Management Sector Ulnit, Washington, D.C. Paper 024. World Bank, Environment Department, Lustig, Nora, ed. 1995. Coping with A usterity: Poverty and Washington, D.C. Inequality in latinAmerica. Washington, D.C.: Brook- Sechrist, L., and S. Sedani. 1995. "Quantitative and Qual- ings Institutions. itative Methods: Is There an Alternative?" Evaluation Manuelvan Atinc, Tamar, and Michael Walton. 1998. So- and Program Planning 18: 77-87. cial Consequences of the East Asian Financial Crisis. SMERU (Social Monitoring and Early Response Unit). 1999. Washington D.C.: World Bank. [http://www.world "Social Impact of the Indonesia Crisis: New Data and bank.org/poverty]. Policy Implications." Jakarta, Indonesia. [http://vww. Narayan, Deepa, and Lant Pritchett. 1997. "Cents and Socia- smeru.or.id]. bility: Household Income and Social Capital in Rural Subbarao, K., K. Ezemenari. J. Randa, and G. Rubio. 1998. Tanzania." Policy Research Working Paper 1796. World "Impact Evaluation in FY98 Bank Projects: A Review.' Bank, Social Development and Development Research World Bank, Poverty Division, Washington, D.C. Group, Poverty and Human Resources, Washington, D.C. World Bank. 1990. World Development Report 1990: Poverty. Norton, Andrew, and Thomas Stephens. 1995. "Participa- New York: Oxford University Press tion in Poverty Assessments." Environment Depart- . 1991. Assistance Strategies to Reduce Poverty. World ment Papers 020. WVorld Bank, Environment Bank Policy Paper. Washington, D.C.: World Bank. Department, Washington, D.C. . 1993. Poverty Reduclion and the World Bank: Operations Evaluation Department. 1994. "An Overview Progress in Fiscal 1992. Washington, D.C.: World Bank. of Monitoring and Evaluation in the World Bank." . 1994. Poverty Reduction and the World Bank: \XTorld Bank, Washington, D.C. Progress in Fiscal 1993. Washington, D.C.: World Bank. - 1996. "Poverty Assessment: A Progress Review. Re- . 1995a. Poverty Reduction and the World Bank: port 15881." World Bank, Washington, D.C. Progress in Fiscal 1994. Washington, D.C.: World Bank. Ravallion, Martin. 1999. "Protecting the Poor in a Crisis- . 1995b. World Development Report 1995. Workers in and Beyond." PREMnote 12. WVorld Bank, Poverty Re- an Integrating World. New York: Oxford University Press. duction and Economic Management Network, Safety . 1996a. Poverty Reduction and the World Bank: Nets Thematic Group, Washington, D.C. Progress and Challenges in the 1990s. Washington, - Forthcoming. "Appraising Workfare." World Bank D.C.: World Bank. Research Observer . 1996b. TakingActionforPovertyReduction inSub- Ravallion, Martin, and Shaohua Chen. .1996. "What Can New Saharan Africa: Report of an Africa Region Task Force. Survey Data Tell Us about Recent Changes in Living Human Resources and Poverty Division, Africa Tech- Standards in Developing and Transitional Economies?" nical Department. Washington, D.C.: World Bank. World Bank, Policy Research Department, Development . 1997. "Strategic Compact." World Bank, Wash- Research Group, Washington, D.C. ington, D.C. [http://www.worldbank.org/'html/ Ravallion, Martin, and Jyotsna Jalan. 1998a. "Are There Dy- extdr/backgrd/ibrd/comsum.htm]. namic Gains from Poor Area Development Programs?" . 1998a. "Africa Region: Review of Impact Evalua- Journal of Public Economics 67 (1): 65-85. tion, FY94 and FY98 Projects." World Bank, Africa Re- - 1998b. "Transfer Benefits to the Poor from Work- gion, Washington, D.C. fare: A Matching Estimate for Argentina." World Bank, . 1998b. Assessing Aid: What Works, What Doesn 't, Development Research Group, Washington, D.C. and Why. Policy Research Report, Washington, D.C.: Robb, Caroline M. 1998. "Social Impacts of the East Asia World Bank. [http://www.worldbank.org/research/aid/i. Crisis: Perceptions from Poor Communities." Paper . 1998c. Global Economic Prospects and Developing prepared for the East Asian Crisis Workshop, Institute Countries 1998/99: Beyond Financial Crisis. Washing- of Development Studies, University of Sussex, Sussex, ton, D.C.: World Bank. [http://www.worldbank.org/ United Kingdom. prospects/gep98-99/appxl/index.htm]. REFERENCES 117 . 1998d. "Monitoring Poverty Impacts: Review of Se- World Bank. [http://www.worldbank.org/html/extdr/ lected Bank Projects in South Asia." World Bank, South pov_red/default.htm]. Asia Region, Washington, D.C. . 1998f. World Development Indicators 1998. Wash- . 1998e. Poverty Reduction in the World Bank. ington, D.C.: World Bank. Ihttp://www.worldbank. Progress in Fiscal 1996 and 1997. Washington, D.C.: org/wdi]. THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433 USA Telephone: 202-477-1234 Facsimile: 202-477-6391 Telex: MCI 64145 WORLI)BANK MCI 248423 WORLDBANK World Wide Web: http://www.worldbank.org/ E-mail: books@vworldbank.org ISBN 0-8213-4384-X