22819 August 2001 Rural Strategy Background Paper #5 WNhat Has Changed Regarding Rural Poverty Since Vision to Action? Harold Alderman FILE COPY Rural Development Strategy Background Paper #5 What Has Changed Regarding Rural Poverty Since Vision to Action? Harold Alderman 2 3 The World Bank Rural Development Family Work in Progress First printing: September 2001 The International Bank for Reconstruction and Development/the World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. This report is one in the series of background studies prepared for the 2001 update of the World Bank's Rural Development Strategy. This series was created to disseminate findings of work in progress and to encourage the exchange of ideas among Bank staff and all others interested in development issues. This paper carries the name of the author and should be used and cited accordingly. The findings, interpretations, and conclusions are the author's own and should not attributed to the World Bank, its Board of Directors, its management, or any member countries. This paper has been reviewed for publication by the Rural Development Strategy Background Paper Series Editorial Committee: Robert Thompson (Chair), Jock Anderson, Shawki Barghouti, Csaba Csaki, Cees de Haan, Gershon Feder, Sushma Ganguly, and Kees Van Der Meer. ii Contents Ac r onyns .................................... iv Acknowledgements .................................... iv Foreword ....................................v 1. Introduction ....................................1 2. New Approaches to the Eradication of Poverty ....................................9 3. Operationalizing the WDR in Rural Areas ................................... 13 4. Conclusion ................................... 27 References .......................................... 29 iii Acronyms AAA Economic and Sectoral Studies at the World Bank, formerly termed ESW CAS Country Assistance Strategy CDD Community Driven Development DHS Demographic and Health Surveys DIS Direct Income Support ECA Europe and Central Asia ESSD Environmentally and Socially Sustainable Development ESW Economic Sector Work GATT General Agreement on Tariffs and Trade (now WTO) GNP Gross National Product HIPC Heavily Indebted Poor Countries IDA Intemational Development Agency IMF International Monetary Fund NAFTA North American Free Trade Agreement NARS National Agricultural Research Systems OECD Organization for Economic Cooperation and Development OP/BP World Bank's Operational Policy/Bank Procedure PAD Poverty Assessment Document PETI Federally Funded Social Protection Program in rural Brazil PROCAMPO Mexico's Farmn Modernization Program PRSP Poverty Reduction Strategy Papers PRSC Poverty Reduction Support Credits WDR World Development Report Acknowledgements The author would like to thank Robert Thompson and Jock Anderson for their comments and Lynn Brown and Louise Cord for their helpful suggestions. The preparation of this study has been supported by the Netherlands Ministry of Foreign Affairs through the Bank-Netherlands Partnership Program. iv Foreword Poverty reduction is the overarching objective of the World Bank, and with 75 percent of the world's poor living in rural areas, rural development is a key element in achieving progress in this objective. At President Wolfensohn's request, the rural family has prepared a revised rural development strategy, Reaching the Rural Poor. This has been done in close cooperation with the regions and the other sectoral units active in the rural space. The objectives of the new strategy are to revitalize the World Bank's activities in the rural areas by: (a) adjusting the strategic framework; and (b) formulating a program of concrete and implementable actions. The new rural development strategy addresses a rural situation which is different from the past, and a rural population which confronts many new problems, especially the challenges and opportunities facing the poor with regard to globalization. The new vision and articulation of a development strategy builds upon the strengths of past efforts as well as incorporates new ideas from other models. In this context, our priorities are geared to fulfill World Bank poverty reduction objectives in the rural sector. We are convinced that the following critical components of a rural development strategy will contribute most to accelerated growth in rural economies and, consequently, to measurable poverty reduction: crafting efficient and pro-poor policies and institutions; facilitating broad-based rural economic growth; improving access to, and management of natural, physical, and human assets; and reducing risk and vulnerability for the rural poor. A number of studies on both global and regional issues, as well as a broad portfolio analysis were commissioned to support the development of the new strategy. These studies provided a rich foundation for both the regional action plans and the corporate strategy. This study is one of the selected number of background papers which have been published in the Rural Development Strategy Background Paper Series to provide Bank staff and others with a more in-depth look at some of the issues surrounding rural development, beyond what is covered by the strategy document itself. This paper, and others in the series are available on line at: www.worldbank.org/ruralstrategy. Additional information on obtaining other papers from this series can also be found at the end of this report. Robert L. Thompson Director of Rural Development The World Bank v Introduction The strategy for rural development voiced in Vision to Action is grounded in the overall mandate of the Bank to reduce poverty. Too few years have intervened since that strategy was outlined to be able to clearly delineate alterations in the rate at which rural poverty is changing. Yet there have been appreciable changes in how we perceive poverty and, thus, in how we should design strategies to reduce it. This note discusses these changes by summarizing three interconnected facets: i) The persistence of rural poverty; ii) new approaches to the eradication of poverty; and iii) the implication of these approaches for rural operations. The first section, then, covers both trends in the number of the rural poor as defined by income poverty as well as a discussion of a broader set of indicators of deprivation. The second section focuses on two significant recent changes in the way that the Bank approaches its poverty reduction mandate. The first of these is the reappraisal of the causes of poverty and strategies for addressing these causes that is conveyed in the 2000/1 World Development Report. The second important change in the approach is the role that is envisioned for Poverty Reduction Strategy Papers (PRSP) produced by client countries. Both the WDR restatement of the overall poverty reduction strategy and the PRSP process will have a bearing on the rural strategy. These implications are discussed in the third section of this note. The Persistence of Rural Poverty Trends in Income Poverty. Vision to Action claimed that more than 1.3 billion people were compelled to live on less than one dollar a day. The current picture remains similar. Despite a 20% decline in the poverty rate between 1990 and 1998 (from 29 to 24%), the 1.2 Billion persons living below the dollar a day poverty line' is little different than at the beginning of the past decade. It also remains true that the majority of these live in rural areas (Box 1). Moreover, there will be more poor people in rural areas than in cities for at least a generation. The rural share of the total number of poor households declines with urbanization - indeed, it declines at a slightly faster rate than the rural share of population declines, due to selective migration. Still, with current trends, the rural share of the global number of poor will not fall below 50% before 2035 (Ravallion, 2000). Even in highly urbanized Latin America, where there are three times as many urban than rural residents, the number of people living on a dollar a day or less in urban areas did not exceed the number in rural areas until 1996. This reflects the fact that the rates of extreme poverty in rural areas are vastly higher than in urban locales. Moreover, rural towns often have higher poverty rates than more metropolitan areas. This has been observed, for example, for Brazil in which the smaller rural marketing towns rather than Rio de Janeiro and Sao Paulo have most of the urban poor (Ferreira, Lanjouw and Neri, 2000). Regional growth trajectories observed earlier in the decade have largely held. The absolute number of poor continues to increase in Africa as income per capita stagnates. In Europe and Central Asia, which had by far the greatest percentage increase in the number of poor (from a i Actually 1.08 dollars at 1993 purchasing power parity. See Ravallion and Chen 2001. 2 What Has Changed Regarding Rural Poverty low base) during the 1990s, the increase may have leveled off, but poverty has not reverted to pre-transition levels. In partial contrast, the countries in Asia where the largest number of rural poor resides, China and India, have maintained rapid GNP growth rates. However, the link between this growth and poverty reduction for the rural poor in Asia may have attenuated in recent years. Box 1: What Share of the Poor Live in Rural Space? Action on a poverty strategy is influenced by knowledge of where the poor are. To determine the rural share of poverty on a global scale, however, takes extrapolation from the limited number of data sets that report this breakdown. One simple method is to regress poverty shares on urbanization for those countries where the data are available and apply the results to global estimates. Ravallion has done so using 39 country observations. These results imply that 80% of the worldpopulation consuming less that $1.08 per day live in rural areas. But is $1.08 the best point at which to make this calculation? Clearly, using this rate facilitates inter-country comparisons. But, consistent with the view that each country needs to take ownership of its poverty strategy, each country should determine the poverty line it uses. In general these increase with GNP - the $1.08 a day line is based on projecting this pattern back to the poorest country in the data set. Reweighting the shares of rural poverty by the shares of poor at each country's own poverty line - giving less weight to India and China - still indicates that 75% of the poor live in rural areas. Thus, the result that poverty is predominately a rural phenomenon is robust to alternative definitions of poverty. There will be more poor people in rural areas than in cities for at least a generation. The rural share of the total number of poor households declines with urbanization - indeed, it declines at a slightly faster rate than the rural share of population declines due to selective migration. Still, on current trends, the rural share of the global number of poor will not fall below 50% before 2035. Source: Ravallion, Martin. 2000. On the Urbanization of Poverty. World Bank. Processed. However, HIV/AIDS threatens to slow the improvement in welfare observed in recent years. A decade ago AIDS was a concern of regional specialists. Even Vision to Action has no mention of the problem. Yet, in 2000 alone, 3 million deaths were attributed to this disease. This is more than the annual rate of deaths among all combatants in the First World War - the so-called Lost Generation. The number of individuals living with HIV/AIDS has more than doubled to 36.1 million since 1991; 41% of these are female. Moreover, this is not just an issue for Africa. For example, the number of HIV positive individuals is greater in India than in any other country except South Africa. Furthermore, while originally perceived (or, perhaps, misperceived) as primarily an urban issue, HIV/AIDS has become as much, if not more, a concern in rural areas as migration, transport, and refugee dislocation facilitates its spread through large populations. In addition, urban residents often return to villages to receive care when ill or send their children to relatives in rural areas. The economic costs of this illness are massive; one estimate calculates that GNP in South Africa will be 17% less in 2010 due to this disease. Moreover, the concentration of mortality on young adults and the legacy of orphans threaten to undennine the social fabric of affected communities. Inter-generational transmission of knowledge as well as intra-village mutual support systems may be severely strained by the magnitude of the challenge. Since Vision to Action? 3 Broadening the Concept of Poverty. Poverty is multi-faceted; the 2000/1 WDR reiterated that, in addition to material deprivation, poverty is characterized by low achievements in health and education. Moreover, the report stressed that the poor also see their plight manifested in vulnerability to economic shocks and in their marginal role in decision making (World Bank, 2000). Rural residents are often even more deprived in terms of health and education than they are in terms of income poverty since access to facilities that deliver services is often limited. The quality of the available choices of service providers is often limited as well. With less access to effective services, the consequences of being income poor differ in rural areas. For example, at the same $1 a day income, a rural child has a third higher chance of dying before its 5th birthday than an urban child (Box 2). Box 2: Rural Poverty Differs from Urban Poverty Two of the most effective measures of poverty are poverty rates and poverty gaps. The former indicates the percentage of a population with expenditures below a poverty line while the latter reports the amount that the expenditures of the poor fall below the poverty line. Still, poverty manifests in other ways as well. Health, for example, reflects access to public and private health services and sanitation infrastructure as well as an individual's income. Rural populations often have less access to these resources and, thus, have worse health outcomes at a given level of expenditures. For example, in nine of eleven countries studied, a rural child in a family with a dollar a day per capita expenditures has a higher chance of dying before its fifth birthday than an urban child with the same household expenditures. The risk of dying at a dollar a day-Under-five 160 - 140- C> GO7 * 1 |1111I11li DRural ~80 - oc, 86o 111 _ _ -iE _ 11 Xllil 0111 ; A z~~~~~ Urban 60 60 *m 40- 20- 0- Brazil Cote Ghana Guyana Indonesia Morocco Nepal Pakistan South Tajikistan Vietnam 1996 d'lvoire 1987/88 1992/93 1993 1990/91 1996 1991 Africa 1999 1998 1988 1993 Source: Alderman and Wagstaff, 2001 Education. For example, while 80% of urban males and 77% of females attend grades 1 through 5 in urban India, the corresponding figures for rural areas are 68 and 56. By grades 9 and 10 the urban net enrollments drop to 41% and 40% while the rural ones drop to 26 and 17. The gender gap in India - generally larger in rural than urban areas - is also a function of wealth. The difference between male and female enrollments for children aged 6-14 in India is only 2.5 percentage points for the rich and 34 percentage points for the poor. 4 What Has Changed Regarding Rural Poverty Another indicator that presents a different dimension of deprivation than income poverty is the nutritional status of children. While this indicator reflects, in part, progress in health care delivery it reveals consequences not shown by trends in infant and child mortality since survivors will have reduced capacity later in life. The prevalence of stunting in children under five in the developing countries was 47.1% in 1980. By 2000 this had dropped to 32.5% (ACC/SCN 2000). Yet, as with other dimensions of poverty, this decline is driven largely by progress in Asia; globally 182 million children in the developing world remain stunted. Indeed, the number of stunted children in Africa continues to rise. In some countries - for example, Burundi, Mali, and Zimbabwe - even the rates of malnutrition are rising. Both cross-country and household data confirm that sustained per capita income growth will go part way towards the 1996 World Food Summit goal of halving child malnutrition rates by 2015. Yet, by itself, income growth is surely unlikely to meet the needs of the coming generation of children. Trends in malnutrition at the country level, however, convey a different picture than trends in income poverty since programs that improve health care delivery or the knowledge and ability of caregivers can contribute to reductions in malnutrition. A related measure at the core of a rural strategy - that of the number of individuals with limited access to calories - is driven both by changes in purchasing power as well as movements in cereal prices. In keeping with poverty reduction in reductions in South and East Asia, the share classified as hungry has declined faster in Asia than in Africa (FAO, 2000). In addition to the factors that increase demand, the past decade has seen a continuation of a long- term trend of increasing food availability in the aggregate. Between 1982 and 1997 cereal production per capita in developing countries increased by 9%. Moreover, driven in part by an even larger increase in production in developed countries, real wheat, rice, and maize prices declined by nearly 30% in that period. The price decline also reflects a complex pattern of producer and trade subsidies in many high income countries. Nevertheless, the global trend in prices has stimulated grain trade and benefited many countries that have not been able to maintain production growth in line with population growth. Projections for food availability point toward a continuation of this underlying trend (Rosegrant and Paisner, 2000). However, extrapolations of past trends need to assess whether the forces that account for the past will continue to drive the global food system. Absent the forces that account for the growth in supply in the past two decades and world grain prices will rise driven by increased consumption of meat as well as cereals as incomes rise. While improvement in agricultural technology and livestock efficiency as well as in water control have contributed to supply increases in the past two decades that exceeded the rate of growth in demand, there is nothing automatic that ensures these improvements will continue. Many of the investments that account for these improvements are public goods that require analysis and advocacy. Moreover, projections of reduction in malnutrition are particularly sensitive to whether yields continue to improve. Thus, declines in malnutrition are indirectly tied to investments in research and irrigation. For example, projections show progress in the battle against food insecurity with worldwide cereal production increasing 33% by 2020 in the baseline case, and 29% under more pessimistic assumptions. However, the 20% decline in the number of malnourished children in Since Vision to Action? 5 the baseline reverses under the less favorable assumptions; the same set of assumptions that support a projection of global production increasing by 29% imply that malnutrition increases by 7%. Malnutrition rates and food security in general are manifestations of the fact that rural livelihoods are comparatively vulnerable. Many rural livelihoods are directly or indirectly dependent on weather. Thus, they are especially vulnerable to income shocks and they respond to shortfalls by drawing down physical assets including livestock and by failing to invest in human capital. Often this depletion is irreversible, either because productive physical and natural assets are sold or because investments in a child's nutrition or education are curtailed. For example, Jensen (2000) reports a decline in the nutritional status of children in Cote d'Ivoire in drought years. Similarly, school attendance in rural India, as well as in rural Brazil, declines in times of low rainfall (Jacoby and Skoufias, 1997; Neri et al. 2000). The rural poor also share the economy's vulnerability to financial shocks. While agricultural producers with a marketed surplus may benefit from devaluation, many of the rural poor are net consumers. Moreover, financial shocks influence remittance pattems and may prompt migration of unemployed urban workers back to rural areas. The 1997 financial crisis in South Asia illustrated this vulnerability to economy wide shocks. In the Philippines and Indonesia, the rural poor had concurrent financial and weather-related shocks. Poverty rates in Indonesia increased 7.5 percentage points between 1996 and 1999 in rural areas (to 20.5%) as well as 5.8 percentage points (to 9.6%) in urban areas (Suryahadi et al. 2000). These rates reflect the impact of both the drought and the financial crisis as well as a partial recovery - which is continuing. Moreover, the measured poverty levels are inclusive of protection furnished by safety nets that helped maintain rice consumption (Box 3). Since vulnerability is a measure of one's ability to insure consumption in the face of an income shock, income itself is not a fair measure of vulnerability. The degree that consumption tracks income changes does indicate how well a household has been able to protect its welfare, but this measure by itself does not indicate whether this protection was obtained by an irreversible asset draw down or by incurring crippling debt. Similarly, the vulnerability to the next shock depends on how one's savings and social networks have been depleted in the previous setback. Thus, there is no single indicator that can summarize vulnerability. Nevertheless, some correlates of the ability of a household to protect its well being, such as access to credit markets or to urban labor markets as well as safety-net programs can be tabulated. In general, rural populations, particularly those with few physical and natural assets, have limited access to formal safety nets. Moreover the poorest residents in rural may also have limited access to informal safety nets, even when they are present (Morduch 1999). This access is difficult to measure. The effectiveness of informal safety nets to reduce vulnerability is similarly difficult to assess. While informal safety nets are a common hedge against vulnerability used by rural populations, they are less effective in the face of a shock that is shared across a community than they are when the setback affects only a few individuals in a community. Such shared challenges - often called covariant shocks by economists - can be weather- or insect-induced crop losses, or due to price fluctuations, or as a consequence of epidemics and civil strife. The evidence that rural 6 What Has Changed Regarding Rural Poverty households sell assets Box 3. Poverty and Safety Nets during the Economic Crises in Indonesia or suffer increased malnutrition during a After two decades of rapid and steady economic growth and poverty reduction, drought is evidence Indonesia endured a substantial downturn beginning in 1997. Driven in part by that they have limited rapid capital flight and subsequent devaluation, the crisis was compounded by a two-year drought and severe forest fires over much of the country. Given the high or at least costly, inflation, estimates are sensitive to how the poverty line is scaled to account for means to self- and co- the post-crises price environment. Using the consumer price index, rural poverty insure. increased from 13.1% in February 1996 to 20.56% three years later. While the urban increase was proportionally more, it was from a low base of 3.8% to 9.6%. The 2000/1 VWDR also Moreover, the number of rural poor likely increased at a faster pace than the Thei2e000/1 that alsov y poverty rate as unemployed urban workers retumed to their native villages. reiterates that poverty is heterogeneous. It Prior to the fiscal crisis Indonesia's logistics agency BULOG had been one of the advises that this be few agencies in the developing world that had been successful at stabilizing prices borne in mind, for through government procurement and sales. BULOG had traditionally avoided example, when being involved in direct distribution to households. Instead, it had used open considering adjustment market operations, that is, injection of rice from its storage into urban markets to cronramsideg thadstmnt place downward pressure on prices. Nevertheless, in the wake of the 1997 programns that, on devaluation subsidizing rice at well below import prices proved fiscally average, will reduce unsustainable and encouraged smuggling and re-export. In lieu of generalized poverty through price subsidies, Indonesia managed to protect the rural poor with a fairly well enhanced growth yet targeted safety net. A key element of this was the provision of a quota of will leave some sub- subsidized rice to poor households identified on the basis of a simple formula. sectors of an economy Thus, changes in its exchange rate that tumed a policy geared to producer relatively subsidies into one that required consumer subsidies prompted Indonesia to relatively institute targeting to achieve its food pricing objectives. disadvantaged. From the standpoint of rural Sources: Surahadi et al., 2000 Table 8; Tabor and Sawit. 2001 development, this reminder points to the range of responses to market liberalization according to differences in farm size, net sales, access to markets and similar distinctions. Many of the rural poor are net purchasers of food and, thus, price changes will affect them quite differently in that they will influence the welfare of neighbors who are net sellers of food. Similarly, some agricultural commodities are effectively non-tradables outside the community while others are easily transported and are directly influenced by international prices. Furthermore, the participation of the poor in the rural non-farm economy is heterogeneous. For some, non-farm employment is a key to economic mobility, while for other, often uneducated individuals, non-farm employment is a residual form of employment that does little to help them escape poverty. One tangible way that the heterogeneity of poverty is conveyed is through the words of the poor themselves. A set of consultations - assembled for the 2000/1 WDR - documents how the poor see their own lives, stressing their sense of vulnerability and powerlessness and their understanding of the dynamics of impoverishment and of opportunity (World Bank, 2000). This listening has become a central part of the process of designing strategies to assist the poor. Partially allied to the issue of heterogeneity of poverty is the renewed attention to the inequality of income as well as to the levels of poverty. To be sure, a focus on poverty implicitly includes inequality since the link from GNP growth to poverty reduction is mediated by the change in income distribution. Thus, while it is clear that on average growth does lead to a one for one Since Vision to Action? 7 change in the incomes of the poor (Dollar and Kraay, 2001), there is wide variance around this average. This is reason enough for concern; the observation reiterates that policies at all levels are needed to increase the participation of the poor in growth. Moreover, inequality may itself be an impediment to growth. Similarly, inequality may be an obstacle to the inclusion of the poor in decentralized programs (Galasso and Ravallion, 2000), as well as an outcome of poorly designed ones. 2. New Approaches to the Eradication of Poverty Understanding Deprivation: The 1990 WDR elucidated the strategy of poverty reduction that motivated the Bank's investment program during the decade 1990-2000. This strategy rested on three pillars: labor-intensive growth, service provision for human capital development, and the provision of safety nets to those who cannot participate in economic growth. The recent 2000/1 WDR substantively shifts this strategy in a manner that adds to the previous approach. However, this is a re-positioning that broadens but does not reject the previous position. The 2000/1 WDR revisits the market reform agenda of its predecessor by recognizing the complexity of policy design and the benefits of complementary strategies to meet the needs of the poor. Moreover, it expands upon past approaches to poverty reduction to by giving considerable weight to issues of institutions, social structure, and participation. In particular, the 2000/1 WDR views poverty as determined by a deficiency of five types of assets: Human assets, physical assets, natural assets, financial assets and social assets. The importance of the first of these provides the justification of investments in health, nutrition, and education. The Nobel Laureate Robert Fogel attributes much of the economic growth in the past two centuries to improvements in nutrition alone. A vast body of evidence shows the same for increased levels of education. Moreover, focusing on human assets highlights the appreciable costs to an economy when women are excluded from schooling and discriminated against in the workplace (World Bank 2001a). But it is the inclusion of social assets - defined as including the network of contacts and political influence - which most broadens the 2000/1 WDR framework. The returns to human, physical, natural and financial assets depend not only on market performance, but also on the performance of institutions and society. Recognizing this, the 2000/1 WDR outlines a three-part strategy to reduce poverty that seeks to: * Promote opportunity by building up the assets of the poor, implementing reforms for growth and by making markets work better for poor people; * Facilitate empowerment by making institutions more responsive to the poor and by removing barriers that restrict the participation of poor in decision making; and * Enhance security by reducing the risk of natural, financial and health shocks and by enabling households to mitigate their consequences. The PRSP Process. One step in implementing the WDR frarnework is to identify the needs of the poor and a strategy to meet these needs. Thus, in late 1999, the Bank (in conjunction with the IMF) called for the drafting of a PRSP as a condition of HIPC debt relief. Eventually, these PRSPs will replace poverty assessments and will be central to the drafting of country assistance strategies (CAS) for all IDA countries. A key feature of these PRSPs is that to foster a sense of ownership of lending programs among clients, they are to be forged by the countries themselves. 10 What Has Changed Regarding Rural Poverty Moreover, in keeping with the WDR 2000/1 perspective they will include consultations with stakeholders. While there is no single blueprint for a poverty reduction strategy, there are a number of core elements that are likely to be common to all strategies: * Diagnosing the Obstacles to Poverty Reduction and Growth - This entails both a description of who the poor are, and an analysis of the macroeconomic, social, structural and institutional impediments to faster growth and poverty reduction. * Policies and Objectives - In light of a deeper understanding of poverty and its causes, the PRSP can then identify medium and long-term targets for the country's poverty reduction strategy, and set out the macroeconomic, structural and social policies that together comprise a comprehensive strategy for achieving these outcomes. * Monitoring - In order to understand the link between policies and outcomes better, a poverty reduction strategy should include a framework for monitoring progress, and indicators of this progress and mechanisms so that this information can be shared with a country's development partners. * External Assistance - A strategy can also improve the effectiveness and efficiency of donor assistance by identifying the amount of external support-both financial and technical- required in implementing the strategy. It could also assess the potential poverty impact of both higher and lower assistance commitments, including actual savings from debt relief. The strategies should both delineate priorities and list programs (with costs and indicators) to address the priorities. The Bank has drafted an OP/BP (Operational Policy/Bank Procedure in the Operational Manual) on social analysis and poverty analysis that will incorporate many of the themes of the new WDR, including the recognition that poverty is multidimensional and that the impacts of policies vary according to the heterogeneity of poverty. These guidelines will also assist in preparing a PRSP. Additionally, a poverty reduction sourcebook has been compiled to contribute to country capacity for drafting strategies. Nevertheless, these resources may not be widely accessible without direct training. Training requirements include materials to transmit the main messages of the rural strategy (including the technical background studies) to teams working on devising strategies in the client countries as well as the counterpart teams from the Bank. Moreover, a different type of training is needed to alert economnists working on general economic development issues to the contribution of rural investments to overall growth. This second type of cross-sectoral training could also highlight differences in how human and economic development policies affect rural and urban residents. The PRSP process poses a formidable task for rural development as well as offers an opportunity. The multi-sectorial approach that should characterize rural development is difficult to nurture in the best of circumstances. It is likely to be much more difficult than when a team is assembled for the ad hoc task of strategy development. While additional funds are generally available for drafting this task, they are not allocated by sector. Moreover, the timing of strategy preparation is such that there will be comparatively little time for drafting background studies. Thus, the 30% decline of funding for AAA - formerly termed ESW - that has occurred in the past decade (World Bank, 2001b) implies a thinner shelf of relevant material to support the PRSP process. In addition, many of the data that can be used for the initial analysis of poverty are not Since Vision to Action? 11 disaggregated spatially. It is even more difficult to find location-specific costing of service provision necessary to devise an effective poverty-reduction strategy that fully takes into account the nature of rural poverty and the types of programs and policies which will reduce it. However, in actuality, the PRSP process is not the drafting of a single document, but a process which should identify knowledge gaps that may be filled in the period between successive PRSPs. Indeed, a CAS can assist by including such knowledge generation as part of the support offered to countries. Moreover, it will remain a challenge to put poverty strategies into operations. In the past it was more common to have an analysis of rural poverty reduction in a country assistance strategy than in project appraisal documents (World Bank 2001b). Here again, one facet of the solution is in training, in this case to raise awareness of the role of rural poverty reduction in an overall country strategy among Bank staff involved in both analysis and in institutional design. Also in keeping with the poverty-reduction strategy being designed and supported by the client country, the Bank has indicated that it will shift its lending towards program lending. For example, it has introduced Poverty Reduction Support Credits (PRSCs) which will involve a series of two or three programmatic operations that correspond to the time horizon of the PRSP and CAS. They typically have one tranche and are provided based on up-front completion of a set of priority reform measures and public actions that demonstrate satisfactory progress with the country's social and structural reform agenda in support of its poverty-reduction strategy. PRSCs may focus mainly on economy-wide policy or institutional issues. However, they may also focus on sectoral policies, institutions, and regulatory actions. The emphasis on PRSCs implies support to overall program objectives and policies but less direct project design by Bank staff. The same concern that rural space will not be fully incorporated in poverty reduction strategies carries over to a concern about the design of program lending. Similarly with the concern that the pipeline of AAA is insufficient. 3. Operationalizing the WDR in Rural Areas The strategy of investing in empowernent, of creating opportunities for rural residents, and of reducing vulnerability while stimulating growth, challenges the Bank to reformulate its operations. This section briefly discusses selective themes in keeping with these challenges. Governance may be enhanced by decentralizing responsibilities to local government. The potential for poverty reduction of this decentralization as well as the risks is the first of these themes. A second theme, also related to governance, is how to make good on past policies of including women into development processes. These two topics approach the heterogeneity of poverty from the standpoint of empowerment. Additionally, they feed into the emphasis in the WDR framework on creating opportunities for the poor. In order to increase opportunities, the rural strategy needs to account for institutional realities and distributional impacts of policies and markets across diverse groups. The discussion below considers this need from the perspective of local area effects and also of globalization. Two related themes that complement a strategy on creating opportunities and building assets follow. These cover investing in education in rural communities, as well as community-based approaches to health and food security. Various aspects of vulnerability are folded into the subsequent section on social protection in rural areas. In addition, the note includes a section on monitoring issues. Decentralization and Equity. The Bank's poverty strategy promotes the decentralization of resources, power, and management responsibility to the lowest level decision-making body capable of delivering the services in order to take advantage of local information. Though decentralization is not synonymous with participation, it often increases local involvement in project planning and implementation. Ideally, community-driven programs will allow rural people to design and implement their own approaches to local problems. Such projects build upon experience with social investment funds and community action programs and endeavor to build local capacity for government and to link national programs and funding to community structures. But, in general, the theory of decentralization has more to say on efficiency than equity and the track record of inclusion in decentralization is mixed. Thus one challenge is to ensure that all groups participate in decentralized projects. This requires transparency in decision making and an oversight or review process that does not stifle community initiative. Also, in general, the smaller the community, the more participatory it can be, and the greater the sense of ownership. Thus, one challenge of decentralized rural development is to determine what types of services are shared by a natural community of interest and what are best delivered at higher levels of legal administrations. Often, complex projects require support beyond the community. The more successful experiences in Bank supported projects in Latin America, however, indicate that it is possible to mobilize community participation at the local level while ensuring central technical and financial support; and in establishing joint responsibility, where necessary, across municipalities. These 14 What Has Changed Regarding Rural Poverty projects have been piloted and the initial experiences used to help design regulations to ensure fair administration. Another challenge will be to provide poorly performing areas with the capacity to take advantage of opportunities offered by community-driven development (CDD). While decentralization and CDD can assist in tailoring programs to the specific conditions of intended beneficiaries, this approach can also acerbate local area effects. Thus, as communities are given more say in their own destiny, they need to be provided the tools to ensure that poor areas with limited capacity do not fall behind other regions. This was noted, for example, in the process of decentralization of a number of functions in Italy during which the relatively prosperous northern part of the country was quicker to take advantage of the opportunities afforded by this change of policy (Putnam, Leonardi, and Nanetti (1993). Few rural areas in low-income countries have a sufficient tax base to contribute substantially to the recurrent costs of services. Yet long-term financing is necessary if projects are to be sustainable and if they are not to exacerbate existing regional inequality. In some contexts, communities are provided resources - with or without local co-financing - on a project or sectoral basis. In others, however, a block grant that is fungible over sectors is provided. The latter makes greater demands on local capacity. In either case, to work well, decentralization requires a transparent process of allocation of resources between sectors as well as within sectors. Moreover, it requires a means to equitably allocate between jurisdictions as well as within these communities. Case studies of communities that have been able to target the poor within their jurisdiction (not all can) ranging from Albania, to Bangladesh, and Ethiopia have shown that the allotment of funds from the central government to local authorities may be weakly correlated with the level of poverty due to simplistic per capita allocation formula or urban and ethnic biases.2 In addition, under some formulas, performance-based allocation may penalize those communities with larger obstacles. Thus, in order to take advantage of local governments' assumed access to local information, there must be a corresponding flow of information to the center as well as an incentive to use this information in keeping with national as well as local objectives for poverty reduction and for political inclusion. There are, fortunately, solutions to the problem of information needed for regional allocation. In particular, there are a number of recent improvements in poverty mapping that can provide transparent poverty rankings for small units of local government. For example, precise poverty mapping can be achieved by linking household survey data to census data (Hentschel et al., 2000). This can be precise enough to distinguish among villages as well as address broader allocation between rural and urban. Other data sources linked to geographic information systems can pinpoint gaps in infrastructure and services and assist in equitable funding. Thus, the poverty impact of decentralization hinges on both the fiscal allocation rules in the center and on the institutional structures at the local level. The effectiveness of either is not 2 A similar observation has been made by Tendler (2000) regarding social funds which share many of the attributes of decentralized service delivery. Since Vision to Action? 15 assured; Bank lending and sector work can contribute to the capacity to design and fair implement community based programs. Empowerment and Gender. Women are crucial to increasing food security in many countries - costing households as much as 15% of potential income at the same overall level of inputs used in agriculture (Udry, 1996). Keeping this in mind, rural development has emphasized the design of gender-sensitive research and extension services as well as the provision of credit to women. Such agricultural strategies can reduce the toll of resource misallocation. In addition, targeted programs to women can also redress under-investment in food consumption since women typically contribute disproportionately to household food budgets and invest more in child health. The current pillar of empowerment, however, goes beyond these strategies. The objective of reducing social exclusion requires an increase in the participation of women in many aspects of economic and political decision making. Barriers may be removed by including the voices of women in the design as well as in the implementation of projects. In many cases, however, empowerment requires a rethinking of the legal framework for land ownership and for inheritance (World Bank, 2001a, Gopal and Salim, 1998). Taking this issue into the dimension of enhancing security also includes a consideration of the legal framework for maintenance following divorce or the death of a husband. But establishing a legal framework is only a step in having this structure be the basis for de facto rights (World Bank, 2001a, Alderman, 2001). Legal reforms may clash with customary law or may be indifferently enforced. However, a mixture of mediation and legal services can complement other programs to ensure rights. For example, reforms of Mexico's ejido system include mediation to protect the property rights of women. Cadastral surveys such as those being supported by the Bank in a number of countries in ECA also assist the enforcement of land rights under law by providing .documentation which can protect women in the case of divorce or widowhood. In the Kyrgyz Republic, the Bank promoted legal reforms as part of preparation for a land registration project and is using studies of traditional rights as part of the supervision of the loan. Similarly, following a successful program for titling land rights in the names of both spouses in Nicaragua, the Bank has embarked on a consultation process with the indigenous population to clarify communal and collective rights. These and similar approaches go substantially beyond expanding extension and rural services and offer opportunities to remove barriers that exclude women from economic and social opportunities. Poor Areas/Local Effects. While regional inequality needs to be explicitly considered in fiscal decentralization, even centrally administered programs need to consider pockets of local poverty. Regional concentrations of poverty may occur with restricted migration - reflecting legal impediments or barriers imposed by language and ethnic heterogeneity as well as differences in access to schooling. They may also stem from imperfections in capital and labor markets. Even in countries with few obvious impediments to mobility, geographic effects have been documented. For example, in Bangladesh, poor areas not only have different levels of infrastructure and human capital, the retums to such capital differ across communities (Ravallion 16 What Has Changed Regarding Rural Poverty and Wodon, 1999). Bangladesh is hardly unique in having some regions progressing less rapidly than other sections of the country. For example, the northeast in both sub-continental India and tiny Albania lags behind the rest of the country while the northwest in China and Ghana are poor compared to their southern regions. That is, there are some regions where everyone is below average. It some cases the low returns to individual investments due to regional poverty imply externalities that justify increased public investments (Jalan and Ravallion, 1998). In other cases, poor areas have relatively high agricultural or livestock potential that can be tapped with improved market access. In such cases analysis which indicates the increased returns to private investments that would be stimulated by public investments in infrastructure and human capital should also guide resource allocation; the time-tested principle of seeking highest marginal returns should stimulate investments in poor areas. Unfortunately, this needs not always be the case. Making markets work for poor regions may not always generate the fastest path to national growth. In such cases, the alternative of concentrating on investments in regions with higher returns not only risks increasing inequality of income, it risks exacerbating political exclusion. This, in turn, may reduce the likelihood that a share of the increased incomes in prosperous regions can be transferred to the less favored. This underscores the need to consider guidelines to investments when there is a tradeoff between growth and poverty reduction. When, for example, should a regional strategy eschew investments aimed at increasing returns in low-potential areas and concentrate instead on facilitating migration through support to education and to financial infrastructure that make remittances easier? Can policies to assist migration minimize negative externalities in over-crowded cities without shifting the costs of avoiding these externalities to the rural poor? These and similar questions derive from the recognition that an understanding of urban migration (as well as inter-rural migration) is a portion of rural development. These questions also are at the core of the Bank's strategy for poverty reduction in middle-income countries, many of which contain large pockets of poverty which have failed to fully participate in the general development of the country. Globalization. To a degree the concerns that are voiced by the general public over the risk of globalization parallel those of local-area effects within a country. For many of the poorest households in Least Developed Countries the problem of globalization is not that they are being impoverished by it, but that they are in danger of being largely excluded from it. The minuscule 0.4% share of these countries in world trade in 1997 was down by one-half from 1980. Global pattems do not reveal a trade-off of growth and poverty reduction. The data do reveal, however, that volatility affecting developing countries is much greater than for developed countries, especially those dependent on a few commodity exports. Thus, the new challenge that globalization adds to a rural poverty strategy is akin to the challenge of reducing vulnerability. Specific measures to mitigate the impact of the volatility of the globalized economy include the range of safety nets mentioned below. In addition, vulnerability will be reduced if steps to enhance global financial stability are realized. Similarly, measures to rationalize financial stabilization programs, leaming from the 1997 East Asia crises, will contribute to rural welfare Since Vision to Action? 17 overall. In addition, there are policy measures that can be taken in OECD countries that may increase the benefits of low-income countries from globalization by raising the average level of trade as well as reducing the variance. Such measures, along with aid and debt relief, are intimately tied with country poverty-reduction strategies. As such, measures to reduce the risks attendant to taking advantage of new opportunities offered by globalization tie back to the principal of consultation with partners at all levels, which is central to the Bank's current approach to poverty reduction. Human Development: Education. Many reviews of Bank projects that have a poverty focus in their design have found that households with few assets fail to benefit from these projects. The reasons are manifold, but among these is the fact that many poor households contain too few members who are both healthy and educated. For example, it is regularly observed that levels of education influence rates of technological adoption. Similarly, a recent study in Vietnam found that the net marginal benefit of irrigation increases strongly with the education of the household (van de Walle, 2000). Moreover, without high-quality education available to all rural children it will not be possible to make non-farm rural employment, as well as migration to urban areas, an integral part of a rural development strategy. Globally, this experience underlies the emphasis on investments in human capital. It also motivates linking social protection to nutrition and education programs. A well-known example is Mexico's Progressa that requires recipients of cash transfers to regularly attend school and visit health posts. Similarly, a federally funded social protection program in rural Brazil (PETI) targets children employed in high risk occupations and offers a cash transfer if they continue schooling and participate in after-school activities in lieu of labor. Recently, the Bank's legal advisors have ruled that such conditional transfers are investments that can be financed in Bank loans. Since the 1990 WDR, education has been central to the Bank's approach to poverty reduction. The 2001 WDR modifies that strategy mainly by emphasizing the importance of local participation in designing and implementing schooling. Many of the more successful cases that inform this strategy have been implemented in rural areas. For example, enhanced community and parental involvement was a key component of the expansion of education in rural El Salvador (Jimenez and Sawada, 1999). Community-based projects can assist in addressing the challenge of schooling in rural areas. However, access to primary schools is only one facet of a strategy to provide a relevant education for rural children. In much of rural South Asia as well as Africa, primary school enrollment has stagnated, necessitating an approach that both improves the quality of education currently offered as well as addresses the poverty and labor constraints that restrict demand for schooling. In addition, in many environments it is necessary to obtain a secondary education in order to take advantage of good non-farm opportunities. However, secondary schooling remains problematic in many rural areas due to the low population density as well as limited local resources to support school budgets. Human Development. Health, Nutrition, and Food Security. The rural strategy recognizes the challenge of persistent malnutrition. To meet the challenge it is necessary to increase global 18 What Has Changed Regarding Rural Poverty aggregate food availability to match both population growth and the expected increase in direct and indirect food demand stemming from increased purchasing power. Improved food security at the national level is a related component of the challenge of reducing malnutrition. When achieved, this improvement at the aggregate level points to a successful mix of agricultural and trade policies. Similarly, enhanced food security at the household level is a reflection of market reforms and poverty reduction strategies. Finally, a reduction in the number of malnourished individuals residing within food-secure households often may be traced back to successful nutritional interventions as well as to programs to improve the status of women. Thus, meeting the challenge of persistent malnutrition with a strategy of sustainable intensification of production is a key pillar of maintaining food security, but it is not a guarantee that malnutrition will be appreciably reduced. Increases in household income will improve access to food and thus address another dimension of food security. Ensuring equitable intra- household allocation reduces gender and age discrimination. Addressing vulnerability will reduce the fluctuations in this access. Nevertheless, just as food and agricultural policy goes astray if it seeks food self-sufficiency in the name of food security, nutrition policy misses the mark when it equates food access with improvements in nutrition. Since the nutrition of a child reflects the interplay of food consumption, health status, and child care, promising avenues for nutritional improvements include a range of services that address whichever of these factors is most limiting. For example, one of the most innovative recent interventions in nutrition, Atencion Integral a la Niinez Comunitario in Honduras, eschews food transfers and concentrates on regular weighing of children as an entry point in informnation sharing among the community. Similarly, analysis of malnutrition in Ethiopia shows that changes in community knowledge on nutrition or female education could have as much or more impact on nutrition as plausible changes in income or food prices (Box 4). The Bank supports similar community nutrition programs in Asia and Africa, again stressing demand-driven information sharing and simple referrals to basic services, including immunizations and control of diarrheal disease. Common elements of such programs include decentralization, the provision of a mix of services and subsidies, learning by doing with piloting and subsequent scaling up, attention to supervision of service providers at the local level and a focus on prevention. These programs often are examples of the oft sought but rarely sighted cross-sectoral approaches, with the lead agency frequently determined by Bank and country capacity. For example, three very similar community nutrition projects in Benin, Madagascar, and Senegal, are listed in Bank documents as rural development, health, and social protection loans, respectively. Many of the world's poor have inadequate intakes of key vitamins and minerals, including vitamin A, iron, and iodine. As with other nutritional deficiencies income growth is part of the long-run solution. Additionally, in recent years much progress has been made in designing programs that can accelerate this process. In addition to supplementation campaigns, links with agro-processing may allow new technologies in single and double food fortification to become mainstream. Moreover, there is an increased potential to address micro-nutrient deficiencies within the context of agricultural research. For example, agronomists have used conventional Since Vision to Action? 19 plant breeding to raise the total bio-availability of iron and zinc in rice. Breeding for nutritional content may prove more cost effective than fortification or supplementation in many situations since a successful technology can be replicated and adapted with little recurrent cost. Moreover, the strategy combines breeding for human nutrition with breeding for plant nutrition and may also increase yields in soils with limited available iron or zinc. Box 4: Role of Knowledge in Addressing Malnutrition There is a clear relationship between income growth and poverty reduction. Nevertheless, by itself plausible income growth will leave many of the next generation's children malnourished. While this is a global issue, the relative roles of income growth and non-income factors can be illustrated with the example of Ethiopia, which has the highest rates of malnutrition in Sub Saharan Africa. Projections based on household survey data indicate that income growth is important for alleviating child stunting, though other interventions can have as large an impact on malnutrition. For example, bringing one female adult per household up to the primary education level would reduce the stunting prevalence of pre-school children by a similar percentage as 15 years of 2.5% per adult equivalent income growth. Enhancing awareness of growth faltering in communities, through increasing the community's ability to rightly diagnose stunted and non stunted children can have similar effects as bringing one female adult per household to the primary education level. Thus a combination of growth and specific nutrition programs will be most effective. Ethiopia: Decline in prevalence of pre-school child stunting improved community nutritional knowledge on 25% cereal price dedine . . ..... universal -emale primary II 'education 2.5% income growth over 15 years 0 2 4 6 8 10 12 percentage improvement Source: Christiaensen, Luc and Harold Alderman, 2001. Child Malnutrition in Ethiopia: The Role of Income, Education and Nutrition Education Programs. World Bank. Processed. The well-publicized recent inclusion of genes to produce beta-carotene into rice also demonstrates the application of new techniques in bioengineering to address a need of low- income producers and consumers. It also illustrates the extension of the strategy of fostering linkages between private and publicly funded research (both international and NARSs). The challenge of creating global public goods in agriculture to benefit the poor when many of the newest techniques and processes for research are patented adds to the complexity of this task. Social Protection in the Rural Strategy. Social assistance serves three different functions: i) it transfers income to individuals with few assets and thus, helps the chronically poor, ii) it serves as a safety net, augmenting private and community-based insurance and, thus, assists in risk 20 What Has Changed Regarding Rural Poverty management and iii) it may assist the transition to a more efficient policy framework by providing compensation to groups who are negatively affected by reforms. Vision to Action acknowledged the importance of the third function to a rural strategy. It recognized that, in order to have effective agricultural policy, it must assure that the poor and vulnerable are protected during the period of transition. However, few such transitional programs for rural areas had been articulated. One relevant example of a safety net program designed to assist the transition to a more favorable policy environment is Mexico's Procampo program (Box 5). This program provided cash transfers to producers in lieu of subsidies and to offset the loss of a protected market with the opening of free trade under the NAFTA and GATT agreements. While the amount delivered was capped, the program was largely untargeted - compensation increased with the size of past plantings up to 100 hectares. Thus, the program does not meet one guideline commonly used for publicly provided safety nets. Nevertheless, it not only made for a more favorable political climate, it provided a stimulus to local production, possibly due to its relaxing a credit constraint (de Janvry and Sadoulet). Similarly, Turkey is piloting a direct income support (DIS) program to partially compensate farmers for the removal of price supports and marketing subsidies. As with Procampo, the DIS program is not targeted to the poor, though it is capped at 20 hectares. In principle, both Procampo and DIS have an exit strategy - the former is scheduled to end after 15 years and the latter will eventually be merged with the European Union Common Agricultural Policy. The safety nets in Mexico and Turkey cost less than the producer and consumer subsidies they replace. However, a reform may be an improvement, even if it is fiscally neutral, if it removes economic distortions. Albania's targeted social assistance program illustrates this feature of safety net programs introduced in tandem with price policy reforms. Unlike Procampo and DIS, Albania social assistance transfers are of the same magnitude as the producer and consumer subsidies that were phased out. However, in all three cases the shift from price subsidies to cash grants rationalized price signals. Whether or not they are linked to policy reforms, targeted transfers serve the first function of a safety net listed above. Unfortunately, to date, safety nets in general and food security in particular have been neglected in PAD analysis (World Bank, 200 lb). Moreover, such programs face special challenges in rural areas due to both the difficulty in finding targeting criteria that can be easily collected as well as the difficulty of administering programs in communities with low population densities and undeveloped infrastructure. Many of the most successful programs in rural areas, such as Mexico's Progressa and South Africa's Old Age Pension, are administered in middle income countries. There is a particular challenge to design such safety nets for low-income countries. The poorest countries with the greatest need for poverty programs also have the greatest need to be selective to avoid compromising macro-economic stability or investment in human and physical capital (Smith and Subbarao, 2000). To illustrate, consider a program in Mozambique that is of the scale of South Africa's largely rural Old Age Pension, that is, one that transfers 1.5% of GNP to roughly 20% of the households. In an economy as relatively well off as South Africa and with Since Vision to Action? 21 the skew in its income distribution, this transfer increases household incomes of recipients by 50% (Case and Deaton, 1998). In contrast, a transfer of 1.5% of Mozambique's substantially lower GNP would only add 15% to the incomes of its beneficiaries. Box 5. Rural Cash Transfer Programs Safety nets provide important tools to assist governments in meeting the needs of the poor, to help them to cope with risk and to expand economic opportunities. However their use has been comparatively limited in rural areas given high transactions costs and the importance of the informal economy (which has made it difficult to collect beneficiary contributions and income data for targeting). Traditionally, rural safety net programs have focused on public works and social funds, which rely upon decentralized administrative structures and community or self-targeting. Also, in some countries governments have relied heavily upon price supports to transfer resources to the rural, mainly farm sector. However, recent experience suggests that cash transfer programs can provide effective safety nets to the rural poor and vulnerable. A review of six rural cash (or quasi-cash) transfer programs suggests that they can reduce poverty, facilitate long-run economic growth, and have relatively low administrative costs. They also can be less distorting, more poverty targeted and less costly than price supports. Two examples from this review look at programs explicitly designed to enable policy reforms. Mexico's Farm Modernization Program (PROCAMPO) was introduced in 1993 to compensate producers for the elimination of price subsidies and trade liberalization associated with the NAFTA and the GATT. * Thus it is not an anti-poverty program per se, but one that aims to protect a group made vulnerable by reforms. * It provides a payment/hectare that is targeted to land users accorded to historical acreage in staple crops. Maximum of 100 ha per farmer. Relies upon network of agricultural extension offices to distribute the check. * Fiscally sustainable because replacing an expensive subsidy program. Estimated that the 1997 costs of PROCAMPO payments is about 10% less than the adjusted value of market supports paid in 1994. * Without PROCAMPO during the 1996 pesos crisis, poverty in the rural sector would have been 5.1% higher. Evidence points to a PROCAMPO income multiplier whereby for each dollar transferred to a Mexican farm household in the ejido or social sector, household income rises by 1.5 to 2.6 dollars. Turkey's Direct Income Support Program (DIS) * Like PROCAMPO, DIS partially compensates farmers for phase-out of subsidies. * DIS is less costly than the subsidy program it is slated to replace. The cost of the income support program in 2002 is estimated at US$1.9 billion, about 40% of the pre-reform agricultural support program. * Targeted to farmers with a constant payment/ha up to a maximum of 20 ha. There are no production or land-use requirements. * Checks are deposited directly into producers' accounts with state agricultural bank. Source: Cord 2001 The necessary selectivity might place particular emphasis on the second motive for safety nets mentioned above. While it may be difficult for a low-income country to provide regular income transfers in a manner that treats all poor households with similar characteristics in the same manner, it may as be possible to provide transitory assistance to prevent destitution in the face of an income fluctuation. Many such disasters have a low probability of occurring in any given area yet high costs when they occur and, thus, are unsuited to private insurance. It is, however, a challenge to design publicly funded programs to stabilize income or consumption without creating inducements that encourage risk taking and which have a clearly defined exit strategy. 22 What Has Changed Regarding Rural Poverty To date, most government crop insurance programs have failed (Anderson, 2001), falling prey to political expediency and the inability to gather timely and accurate information on effort and output. Proposed innovative approaches to rainfall-based insurance that are less plagued by asymmetric information are currently being tested. Cost-effective income stabilization programs, however, are not completely unprecedented. For example, subsidies to livestock transport in Kenya have successfully prevented price collapse during a drought and, thus, have served as insurance for pastoralists. In Bangladesh, access to micro-finance stabilizes consumption as well as raises income. In other countries, investments in forestation reduce the risk of flooding. Turning this short list into a longer list is one goal of rural development efforts as well as the Bank's overall Social Protection strategy. Similarly, the Bank is assisting in disaster early warning and has recently redrafted the OP/BP for emergency reconstruction to clarify guidelines. Previously, the Bank's comparative advantage in long-term lending for development and the necessary procedures for project preparation and supervision had not seemed compatible with response to disasters. Moreover, lending for disaster preparedness does not readily translate to standard calculations of rates of returns to investments. Following major efforts to respond to floods following the 1997 El Nifno and Hurricane Mitch as well as to assist earthquake victims in Turkey and India, however, the distinction between development and disaster response seems artificial. A second way that low-income countries may deliver assistance to households with an income loss is by supporting informal support programs that build upon traditional rural community structures. While these frequently fail in times of shared hardship (Morduch, 1999, Alderman, 2001), this shortcoming may be reduced with support from a central government. Moreover, individual shocks such as illness are often as large or larger a source of income variability as covariate shocks. This is particularly the case for resource poor farmers. Surveys in Mexico, for example, indicate that health risks are no less important than agricultural risks to such farmer households (World Bank, 2000c). An approach that supports community structures may allow access to the norms under which a community allocates assistance through defined grants to local organizations, ensuring both fairness and incentives for communities to monitor targeting and exit criteria. However, this type of program must still be subject to oversight to protect minorities, as mentioned above. Assistance to micro-credit programs might also be included in this category of support. In addition to the judicious provision of subsidies to operating costs (but not interest rates), the insurance function of micro-credit may be enhanced with reinsurance in the face of large covariant shocks, such as the floods which devastated Bangladesh in 1998. A third approach to smoothing consumption in times of economic crisis is by offering employment in public works programs. Such programs generally allow self-selection by offering wages slightly below market rates. Thus, low-income individuals can seek employment in accord with their need and to exit when other opportunities arise (Subbarao et al., 1997). The value to the poor of such programs is not measured only in terms of income support but also in terms of the benefits they receive from the infrastructure created. Therefore, programs can be enhanced with technical assistance in drafting projects, preferably with a shelf of such projects Since Vision to Action? 23 available to be drawn upon in the face of drought or other crisis. Since not every poor household has an able-bodied member who may participate, public works can also be enhanced with complementary programs targeted to households unable to take up employment opportunities, as well as removing a barrier to participation of caregivers by the provision of creches. An issue that is related to that of disaster management is that of reconstruction after conflicts. As with many types of disasters rural areas may be particularly vulnerable to devastation in such conflicts as isolated communities often are targets in insurgencies. The Bank's overall approach to post-conflict recovery, coordinated by the Post-Conflict Unit within ESSD, recommends flexibility of procedures and, in acknowledgment of reduced capacity in recovering countries, also suggests that fewer projects are recommended3. Priorities include capacity building, infrastructure rehabilitation and demobilization of ex-combatants. The issue of capacity is so central that - depending on local conditions - the Post Conflict Unit recommends continuing ESW and non-lending assistance even while other projects are suspended. The spatial, ethnic, and gender concerns that run throughout the rural sector strategy are also crucial in the design of transitional support strategies. However, the task of reconstruction generally exceeds the finding that the Bank can provide. Thus, external partnerships, always part of country strategies, play a key role in post-conflict programs. Monitoring. If we view poverty as multi-dimensional, how do we track progress in poverty reduction? A number of proposals for an index of human development have been proposed over the past two decades. These offer a weighted combination of measures of welfare such as income, life expectancy, education, gender equality etc. While these indices highlight the varied nature of poverty, the attempts at aggregation mask more than they reveal. For one thing, they make it hard to draw attention to countries that perform unexpectedly well (or poorly) in one dimension. Moreover, the weights for aggregation are necessarily arbitrary and implicitly define a value of one outcome in terms of another. For example, if one has national index that includes GNP ranking and life expectancy one can determine the amount of increase in income that will leave the index unaffected when life expectancy decreases. This gives an implicit price to life expectancy, but one that most proponents of indices would reject. Moreover, with some commonly used indices the implicit value of reducing life expectancy is rather small for low- income countries and increases rapidly at higher levels of income (Ravallion, 1997). Thus, it is far preferable to track a set of indicators and to use the independent movement of the indicators to infer on the processes affecting overall welfare (Okidegbe, 2001). For example, while any plausible tracking of poverty will include a measure of income or consumption poverty, separate measures of health can indicate which countries or regions have been able to implement policies that accelerate improvements over that which would be expected with observed growth. Many of the key indicators that summarize progress in rural welfare - expenditure poverty, life expectancy, school enrollment, child mortality, and malnutrition rates - can be derived from sample surveys. For example, many countries regularly conduct income and expenditure surveys. Similarly, demographic and health surveys (DHS) provide a basis to track trends in welfare. 3 This is covered in OP/BP2.30. See also Koester, 2001. 24 What Has Changed Regarding Rural Poverty Unfortunately, as often as not, the data from these are reported in an aggregated form. A published report may, for example, indicate the difference in rural and urban child mortality or primary school enrolments, and, elsewhere report these statistics by poverty group. Yet it may not be possible to use these reports to ascertain other patterns of interest, such as whether the gaps between the rich and the poor or between males and females is greater in rural areas than in cities. Thus, there is often need to return to the primary data for poverty monitoring. Various networks and regions at the Bank are attempting to maintain databases for these purposes. With additional support these resources can be documented to assist in analysis and monitoring to track spatial dimensions of poverty. Surveys, however, are not generally performed on an annual basis in any given country. While surveys undertaken, say, every five years may be sufficient for most purposes, it is difficult to use such data for indicating risk and vulnerability. Often poverty measures are interpolated between surveys using national accounts, or food balance sheets. The Bank is assisting in efforts to improve the methodologies needed for this inference. Additionally, it is improving methodologies for other means of gauging rural and urban welfare, including participatory assessments utilized by the WDR and in PRSP poverty diagnoses. Similarly, the Bank has supported Living Standards Surveys since the middle of the 1980s. Recently the experience gained with such surveys over 15 years has been assessed (Grosh and Glewwe, 2000). This re- examination makes recommendations for improving modules for collecting information on topics not generally covered in income and expenditure or DHS surveys.4 This includes, among others, modules on agriculture and household enterprise, and sections on environmental issues, migration, credit use, and saving. Measures such as income poverty or anthropometric status and school enrollment that can be obtained from such surveys reveal the interplay of various household community and national factors that affect welfare. However, they can not generally be used to evaluate the impact of specific polities or Bank projects. Among the reasons for this limitation is the difficulty of establishing a causal link from the investment or policy change and the outcomes being monitored. It is common practice to include monitoring and evaluation in Bank loans and networks often offer additional resources for evaluating innovative programs. In practice, however, monitoring is more often in regards to inputs and disbursements and, while evaluation undertaken within lending operations often provides insight on processes, it generally sheds less light on impacts. The need for impact evaluation is heightened with the entry of the Bank into new initiatives to empower excluded groups. While there are some good practices that inform current programs, the knowledge base in many areas remains thin. This points to new emphasis in the rural research program on issues such as the development impact of community-based rural organizations in general and of Bank support to these organizations in particular. Other areas of research that can support the rural strategy include: how rural non-farm development is stimulated, how it can be made to include individuals with limited assets, how to design effective insurance programs that reach the rural poor and how to design safety-net programs to prevent destitution in the face of income shortfalls. Indeed, the Bank's ability to deliver programs in the last of these areas will be enhanced by research that improves the measurement of vulnerability. 4 Draft questionnaire modules on various subjects are also included in that volume. Since Vision to Action? 25 In addition, continuing research on the delivery of public services (including health and education), on cost-effective ways to finance and manage infrastructure and on market-assisted land redistribution will improve programs in these core areas of the rural strategy. Such research, as well as more focused monitoring, will inform both the managers of specific projects as well as create public goods for a wider development community. This can complement an additional source of information on performance, the on-going review of the poverty focus of the rural portfolio (Proctor et al, 2001). This review provides a detailed assessment of the degree to which various dimension of poverty were addressed in the underlying analysis and subsequent design of various projects and offers a methodology for tracking improvements in project formulation in the future. Including poverty in the conceptualization of a project addresses one key step in rural development; ensuring that the implementation achieves the objectives and monitors the actions. 4. Conclusion Though papers generally include a section on conclusions, a strategy is a beginning. Getting a rural strategy right is not the last word in poverty reduction; getting it wrong may be. References ACC/SCN. 2000. The Fourth Report on the World Nutrition Situation. ACC Subcommittee on Nutrition (Geneva) and the International Food Policy Research Institute (Washington D.C.). Alderman, Harold. 2001. Implications of Private Safety Nets for Public Policy: Case Studies of Mozambique and Zimbabwe. World Bank. Africa Region Social Protection Strategy Working Paper. Alderman, Harold, Simon Appleton, Lawrence Haddad, Lina Song, and Yisehac Yohannes. 2000. Reducing Child Malnutrition: How Far Does Income Growth Take Us? World Bank. processed. Alderman, Harold, and Adam Wagstaff. 2001. "Life and death on a dollar a day: Does it matter where you live?" World Bank. Processed. Anderson, Jock R. 2001. Risk Management in Rural Development Background Study for Updating Vision to Action (VtoA). Processed. Case, Anne and Angus Deaton. 1998. Large Cash Transfers to the Elderly in South Africa. Economic Journal. 108(3): 1330-1361. Christiaensen, L. and H. Alderman. 2001. Child Malnutrition in Ethiopia: The Role of Income, Education and Nutrition Programs. World Bank. Processed. Cord, Louise. 2001. Rural Safety Nets. Lessons from Experience. World Bank. Background Study for Updating Vision to Action (VtoA). Processed. Dollar, David and Art Kraay. 2001. Growth is Good for the Poor. World Bank Policy Research Working Paper #2587 FAO. 2000. The State of Food Insecurity in the World, 2000. Rome. Ferreira, Francisco, Peter Lanjouw and Marcelo Neri. 2000. A New Poverty Profile For Brazil Using Ppv, Pnad And Census Data World Bank. Processed. Fogel, Robert. 1994. Economic Growth, Population Theory, and Physiology: the Bearing of Long-Term Processes on the Making of Economic Policy. American Economic Review 84 (3) 369-95. Galasso, Emanuela and Martin Ravallion. 2000. Distributional Outcomes of a Decentralized Welfare Program. World Bank Policy Research Working Paper #23169. Gopal, Gita and Maryam Salim. 1998. Gender and Law. Eastern Africa Speaks. World Bank. 30 What Has Changed Regarding Rural Poverty Grosh, Margaret and Paul Glewwe (eds.). 2000. Designing Household Survey Questionnaires for Developing Countries: Lessons from Ten Years of LSMS Experience. Washington: Oxford University Press. Hentschel, J., J. Lanjouw, P. Lanjouw and J. Poggi. 2000. Combining Survey Data with Census Data to Construct Spatially Disaggregated Poverty Maps: A Case Study of Ecuador, World Bank Economic Review. 14(1): 147-166. Jacoby, Hanan and Emmanuel Skoufias. 1997. Risk, financial Markets, and Human Capital in a Developing Country. Review of Economic Studies. 64:311 -335. Jensen, Robert. 2000. Agricultural Volatility and Investments in Children. American Economic Review. 90(2): 399-404. Jalan, Jyotsna and Martin Ravallion. 1998. Geographic Poverty Traps? World Bank. Processed. Jimenez, Emmanuel and Yasuyuki Sawada. 1999. Do Community Managed Schools Work? An Evaluation of El Salvador's EDUCO Program. World Bank Economic Review. 13(3):415- 442. Koster, Markus. 2001. Towards an Operational Framework on Conflict and Development in the Africa Region. World Bank. Post-Conflict Unit. Processed. Morduch, Jonathan. 1999. Between the Market and the State: Can Informal Insurance Patch the Safety Net? World Bank Research Observer. Neri, Marcelo Cortes, Emily Gustafsson-Wright, Guilherme Sedlacek, Daniela Ribeiro da Costa, and Alexandre Pinto. 2000. "Microeconomic Instability and the Children's Human Capital Accumulation: The Effects of Idiosyncratic Shocks to Father's Income on Child Labor, School Drop-Outs and Repetition Rates in Brazil." World Bank Background Paper, May. Washington, D.C. Okidegbe, Nwanze. 2001. Rural Poverty Trends. World Bank. Background Study for Updating Vision to Action (VtoA). Processed. Proctor, et al. 2001. In-depth Review of the World Bank's Rural Portfolio. Component III: Poverty focus of the Portfolio. World Bank, processed. Putnam, Robert, Robert Leonardi, and Raffaella Nanetti. 1993. Making Democracy Work. Princeton: Princeton University Press. Ravallion, Martin. 2001. On the Urbanization of Poverty. World Bank. Processed. Ravallion, Martin 1997. Good and Bad Growth: The Human Development Reports. World Development 25(5):631-638 Since Vision to Action? 31 Ravallion, Martin and Shaohua Chen. 2000. How Did the World's Poorest Fare in the 1990's? World Bank Policy Research Working Paper #2409. Ravallion, Martin and Quentin Wodon. 1999. Poor Areas, or Only Poor People? Journal of Regional Science. 39(4), 689-71 1. Smith, William and Kalanidhi Subbarao, 2000. What Role for Safety Net Transfers in Very Low Income Countries World Bank. Safety Nets Primer Forthcoming. Subbarao, Kalanidhi. et. al. 1997. Safety Net Programs and Poverty Reduction: Lessons from Cross-Country Experience. The World Bank. Suryahadi, Asep, Sudamo Sumarto, Yusuf Suharso, and Lant Pritchett. 2000. The Evolution of Poverty during the Crisis in Indonesia, 1996-99. World Bank Policy Research Working Paper #2435. Tabor, Steven and M. Hussein Sawit. 2001. Social Protection via Rice: the OPK Subsidy Program. Journal of Developing Economies (Tokyo) Forthcoming. Tendler, Judith. 2000. Why are Social funds so Popular? In Yusuf, Shahid, Weiping Wu and Simon Everett (eds.). Local Dynamics in the Era of globalization. Oxford University Press. Udry, Chris. 1996. Gender, Agricultural Production, and the Theory of the Household. The Journal of Political Economy, Vol. 104, No. 5. (Oct.);1010-1046 van de Walle, Dominique. 2000. Are Returns to Investment Lower for the Poor? Human and Physical Capital Interactions in Rural Vietnam. World Bank. Processed. World Bank. 2000. World Development Report 2000/2001. Attacking Poverty. New York: Oxford University Press. World Bank. 2001 a. Engendering Development Through Gender Equality in Rights, Resources and Voice. New York: Oxford University Press. World Bank. 2001b. Sharpening the Focus on Rural Poverty. Operations Evaluation Department. Processed. World Bank, 2001 c. Mexico. Crop Insurance. Main Report. Latin America and the Caribbean Regional Office Other Background Papers in this Series This study is the fifth in a series of background papers published by the Rural Development Department of the World Bank in the preparation of the Bank's new rural development strategy. For additional information on this, or forthcoming papers in the series, please contact Mr. Alan Zuschlag at (202) 458-5591. Rural Development Strategy Background Paper #1: Long Term Prospects for Agriculture and the Resource Base - August 2001 Rural Development Strategy Background Paper #2: The Role of Agriculture in Economic Development - August 2001 Rural Development Strategy Background Paper #3: Rural Poverty: Trends and Measurements - August 2001 Rural Development Strategy Background Paper #4: Rural Non-Farm Activities and Rural Development: From Experience Towards Strategy - August 2001 Th. World Bank Rural Development Department The World Bank 1 81 8 H Street, N.W., Room MC5-724 Washington, D.C. 20433 website: http://www.worldbank.org