The World Bank . . 26392 February 2003 Designing Investments in Agricultural Research for Enhanced Poverty Impacts - - y iuii'ii rl-"T-A Derek Byerlee_ ~1 - i Gary Alex P ~~~~~~~ ~~~~~ ~~~~~ '~~~~~~~~~~~~~;IijIi~~~~~~~~~~~ _ Agriculture & Rural Development Working Paper 6 The World Bank Work in Progress for Public Discusion Designing Investments in Agricultural Research for Enhanced Poverty Impacts Derek Byerlee and Gary Alex , A R D ,t3 ^ AGRICULTURE Sustainable Agricultural Systems, AND RAL Knowledge and Institutions (SASKI) DEELPMN First printing: February 2003 ŠThe International Bank for Reconstruction and Development Agriculture & Rural Development Department 1818 H Street, N.W. Washington, DC 20433 The findings, interpretations, and conclusions are the author's own and should not be attributed to the World Bank, its management, its Board of Executive Directors, or the countries they represent. Some of the numbers quoted are estimates or approximations, and may be revised at a later stage. This publication can be freely distributed, but should be appropriately cited to the authors. Byerlee, Derek and Gary Alex. 2003. Designing Investments in Agricultural Research for Enhanced Poverty Impacts. Agriculture & Rural Development Working Paper 6. Washington, D.C.: World Bank Sustainable Agricultural Systems, Knowledge and Institutions (SASKI) is a thematic group that promotes sustainable agricultural production, processing and marketing systems through gender-responsive knowledge institutions with the Agriculture and Rural Development Family of the World Bank. Agriculture is defined to include crops, livestock, and aquaculture. Cover artwork: Column 1 top-stock photo; Column 1 bottom-Ray Witlin (1974) Women in Ethiopia carrying water in jugs on their heads in Jijiga, Ethiopia; Column 2 top-Curt Carnemark (1993) A man using donkeys to plow the field in Yemen; Column 2 bottom-Curt Carnemark (1996) An agricultural field in Kenya (cropped section); Column 3 top-Curt Carnemark (1993) A market selling tomatos in Guatemala; Column 3 bottom-Yosef Hadar (date unknown) Turkey; Column 4 top- Curt Carnemark (1992) A view of crops and vegetation in Tunisia; Column 4 bottom-; Column 5 top-Yosef Hadar (date unknown) Brazil; Column 5 bottom-Yosef Hadar (1984) A man throwing grains in the air in Brazil (cropped section); Column 6-Curt Carnemark (1994) Agricultural landscape in Morocco MA059S12.tif; Back cover-Yosef Hadar (date unknown) Brazil. All photos courtesy of the World Bank photo library. Contents Foreword iv Executive Summary v 1. How Does Research Link to Poverty Reduction 1 2. Key Issues for Research and Poverty Reduction 4 Broad-based Growth Strategies Have Sharply Reduced Poverty 4 Future Strategies Require Poverty Targets 5 Overcoming Institutional Weaknesses in Research Systems 6 Addressing Diversity Within the Sector 7 3. Good Practice for Enhancing Poverty Reduction Impacts of Investments in Research Systems 10 Analyzing the Structure and Nature of Poverty 10 Selecting a Broad Strategy 11 Setting Specific Priorities for Investments 12 Strengthening the Institutional Base 15 Promoting Inclusive Participation 17 Linking to Innovation Uptake Pathways 18 Putting in Place an M&E System 19 References 21 Foreword Agricultural growth in developing countries is follow-up monitoring and evaluation. These in- critical to the Bank's efforts to reduce poverty. vestments must also address long standing in- Agricultural growth in turn requires investment stitutional weaknesses in research systems, such in agricultural research, which has traditionally as inadequate funding, low salaries for scien- provided high returns on the investment. Future tists, and poor links with user groups. investments in agricultural research, however, This series of SASKI Good Practice Notes require a more explicit linkage of the investment aims to disseminate innovative ideas, experi- with poverty reduction strategies. This SASKI ence, and recommendations that may assist Good Practice Note discusses various strategies World Bank staff, leaders in borrower countries, that aim at broad-based productivity growth, and Bank partners to prepare and implement as well as ways of carefully targeting invest- lending programs. The SASKI Thematic Team ments to enhance poverty impacts. Enhancing welcomes feedback and contributions to the series. pro-poor impacts of investments in agricultural This note benefited from inputs and com- research will require: defining the nature and ments by Julio Berdegue, German Escobar, causes of poverty, establishing appropriate Matthew McMahon, and James Ryan. Partici- strategies, setting priorities, strengthening in- pants at a 2002 SASKI seminar, where the pa- stitutions, involving poor farmers at all stages, per was discussed, also contributed many promoting uptake of new technologies, and helpful comments and suggestions. Marie-Helene Collion Chair, SASKI Thematic Group iv Executive Summary Considerable empirical evidence indicates that unfavorable production environments, agricul- agricultural growth is a priority for reducing tural innovation can usually lead to sustainable poverty in developing countries. Investment in progress in poverty reduction only as part of a agricultural research is a major source of agri- broad-based effort to diversify livelihood strat- cultural growth, and can provide powerful di- egies of those households. rect and indirect impacts on rural and urban Strong agricultural research systems are poverty. The World Bank's Rural Strategy iden- therefore crucial in addressing agricultural tifies agricultural science and technology as a growth and poverty issues. In many countries, high priority for future support, but this sup- however, these institutions have serious weak- port must be more focused on the Bank's nesses, such as poor incentives to attract qual- overarching objective of poverty reduction. ity scientists, lack of operating budgets, and Several decades of experience provide evi- ineffective links with clients. A first priority dence that a broad-based approach to promot- must be to address these overall institutional ing agricultural growth can have substantial weaknesses and improve efficiency and effec- impacts on poverty reduction where: (a) agri- tiveness of national agricultural research sys- culture is important to the incomes of rural poor; tems. Improvements in research system (b) the agro-ecological base allows significant performance regardless of type of user will potential for productivity growth; (c) land dis- likely provide large benefits to the poor as well. tribution is relatively equitable; and (d) the poor The poverty reduction impacts of investments consume nontradable food staples. Without in agricultural research can be enhanced these preconditions for win-win productivity through the following steps: growth and poverty reduction, agricultural re- search may still have strong poverty reduction Analyzing the nature and causes of impacts, but must be carefully targeted on poor poverty: The starting point in developing op- producers and consumers. tions and priorities for enhancing the poverty- Diversity in farm types and farming systems reduction effects of research investments is the shapes the potential impact of agricultural re- review of poverty assessment surveys and pov- search programs on poverty reduction. Agricul- erty reduction strategic papers (PRSPs) to iden- tural innovation by commercial farmers in tify types of poverty (rural and urban), their favorable environments can lead to poverty re- determinants, and the contexts in which they duction if it generates employment (for example, occur (geographical, farming systems, and live- labor for production of high value commodities) lihood strategies). and/or provides cheaper food. Development Establishing a broad strategy for research of small-scale, market-oriented family farms investments: Based on poverty analyses, offers the greatest potential for poverty reduc- research strategies might differ between tion, if improved technologies are widely and within countries depending on: (a) the in- adopted. For very poor farm households in come level of the country; (b) market and trade v vi Designing Investments in Agricultural Researchfor Enhanced Poverty Impacts policies; (c) agroclimatic potential; (d) rural in- pluralism in carrying out research, and pro- frastructure, service provision, and market ac- mote pro-poor effectiveness. cess; and (e) the labor situation. In general, * Financial sustainability through reforms where most of the poor are in rural areas and that build political support for research es- agriculture is a significant share of their liveli- pecially from poorer groups, and provide hoods (for example, much of Africa and South cofinancing of research on commercial crop Asia), broad-based approaches to stimulate and livestock commodities. agricultural productivity will be the preferred * Decentralized research, especially adaptive strategy, with judicious targeting as appropri- research, to put scientists closer to users and ate. In other situations, more focused pro-poor enhaince user participation in setting the re- approaches might be needed, based on prior- search agenda. ity setting for research resource allocation and * Building capacity of producer organizations institutional innovation appropriate to the to integrate user groups into research. poor. Setting specific priorities for investments: Promote inclusive participation by the Within these broad strategies, resources can be poor: Participatory research is important to allocated to specific priorities important to poor assess relevance of technologies, identify client producers and consumers. Priorities might be needs, build capacity of individuals and local in- set by geographical region (provided there is stitutions, and encourage demand for services. demonstrated agricultural potential), by com- Indusion of women is particularly important modities produced and consumed by the poor because they play a major role in agricultural (especially staple foods and nutrient-enhanced production. Participation by minorities, such as foods), by type of technology (for example, low- tribal and caste groups, is also important to over- external input systems or natural resources con- come past bias and ensure equity for these groups. servation), or type of research (for example, Involvement can occur at various levels, includ- basic research on crop or livestock traits of im- ing governance, program planning, priority set- portance to the poor). ting, research execution, and evaluation. Strengthening the institutional base for Promote uptake of research findings by the research: Investment strategies must promote poor. Research programs need to pro-actively development of research institutions that more plan for promoting uptake of innovations, form- effectively address client needs. Building sus- ing strategic alliances with private firms, NGOs, tainable capacity will generally involve partici- producer groups, and others to make findings patory approaches that empower the poor and available to the poor. rely on: Implement monitoring and evaluation sys- tenis to measure impacts: Monitoring program * Private sector research to free up and refo- impacts, outcomes, and outputs relevant to pov- cus public resources for pro-poor programs erty reduction requires quantitative and quali- by reducing policy and regulatory barriers tative indicators for research investments. to private investment. Beneficiary assessments and participatory * Contractual and competitive mechanisms to evaluations are important in obtaining input improve efficiency, enhance institutional from the poor on program assessments. How Does Research Link to Poverty Reduction? The World Bank is the largest international fin- 1). Investment in agricultural research is one of ancier of agricultural research in developing the major sources of agricultural growth, and countries, with a portfolio of loans totaling hundreds of studies in developing countries about US$1 billion. It is also the largest donor have shown these investments to generate high to the Consultative Group on International payoffs, with returns averaging over 40 percent Agricultural Research (CGIAR). The recently re- (Table 1). vamped Rural Strategy for the Bank, Reaching Investmnent in agricultural research can also the Rural Poor, identifies agricultural science and provide powerful direct and indirect impacts on technology as a priority for future support to rural and urban poverty. Agricultural research the agricultural sector. The Strategy also empha- may directly benefit poor farmers who adopt sizes that all lending operations must be more improved technologies by increasing their in- focused on the Bank's overarching objective of comes and/or reducing production and market- poverty reduction. ing risks (for example, by breeding for pest In this note, the linkages between agricul- resistance). It may also improve management tural research and poverty reduction are sum- of natural resources on which the poor depend marized. Strategies and program design options for many of their livelihood strategies. that can increase the effectiveness of the World Indirect effects may be even larger. Produc- Bank's lending program on poverty reduction tivity growth through research results in are also identified. reduced food prices-a benefit especially impor- Poverty is concentrated in rural areas and tant to the poor in both rural and urban areas as over one-half of the world's extreme poor work they typically spend a high proportion of their in agriculture (IFAD 2001). Empirical evidence income on food. Research can also increase the indicates that agricultural growth is a priority nutritional status of foods important to the poor. for reducing poverty in many situations (Box Employment, migration, and wage effects of 1 2 Designing Investments in Agricultural Research for Enhanced Poverty Impacts Box 1: Empirical Evidence of the Relationship Between Agricultural Growth and Poverty Reduction * A 1 percent increase in agricultural GDP per capita . Wheat prices would have risen 34 percent, and rice led to a 1.6 percent gain in the per capita incomes prices 41 percent, more between 1970 and 1995 of the lowest income fifth of the population in 35 in the absence of intemational agricultural research countries analyzed (Timmer 1997). efforts, adversely affecting poorconsumers of these * A 10 percent increase in crop yields can lead to a staples (Evenson and Gollen 2002). reduction of between 6and 10 percentof people . The average real income of small farmers in living on less than $1/day (Irz and others 2001). southern India rose by 90 percent and that of For African countries, a 10 percent increase in landless laborers by 125 percent between 1973 yields leads to a 9 percent decrease in the num- and 1994, as a result of the "green revolution" ber of those living on less than $1/day. (World Bank 2001). Source: World Bank (2002). increased production and value-added process- compared to other public investments. This is ing are important to the poor, who often depend shown by recent IFPRI studies in India and on agricultural wage labor. Agricultural growth China comparing the reduction in the number also provides important multipliers to overall of poor people per unit of investment for differ- economic growth through demand for produc- ent types of public investments (Box 2). In these tion inputs and increased purchasing power of studies, investment in agricultural research, in- producers. Finally, the development of human frastructure, and education provided the high- and social capital through demand-driven est poverty reduction effects. Translated into US agricultural research systems is important to dollars, the cost of removing one person from empowering the poor to improve their access poverty is about $180 in India and about $380 not only to technology, but also to assets and in China. These figures are quite comparable to markets. an international study that estimated a cost of For these reasons, poverty impacts of agri- $179 per person removed from poverty for Asia cultural research investment are often high (Thirtle, Lin and Piesse 2002). Since the rate of Table 1: Estimated Rates of Return to Investment in Agricultural Research Region Number of estimates Median rate of return Africa 188 34 Asia 222 50 Latin America 262 43 Middle East/North Africa 11 36 All developing countries 683 43 All developed countries 990 46 All 1,772 44 Source: Alston and others (2000). How Does Research Link to Poverty Reduction? 3 Box 2: Impact of Past Agricultural Research on Poverty Reduction Substantial evidence indicates that the poverty re- spent on agricultural research reduces the number duction impact of agricultural research investment of poor by 90 persons. Likewise in China the im- is high compared to other public investments. In pact of Yuan 100,000 on poverty reduction is higher India agricultural research had the highest produc- than for all other investment except education (Fan, tivity impacts and the second highest poverty re- Zhang, and Zhang 2000; Fan, Hazell and Thorat, duction effect after rural roads. One million Rupees 1999). Number of persons removed from poverty for a given public investment c 140 > 120 = India 8 80 > = | * ~~~~~~~~~~China 80 0, CL o20 0- Ag R&D Roads Education Irrigation Rural Dev return on these investments is high (Table 1), ogy use, access to assets, infrastructure, markets investment in agricultural research in these situ- and institutions, and other factors. Therefore, ations is clearly a win-win proposition. for any specific situation, an analysis of the po- The strength of the link between agricultural tential for research to contribute to poverty re- growth and poverty is, however, conditioned duction must start with a review and assessment by agro-ecological conditions, level of technol- of the broader sectoral context for investment. 2 Key Issues for Research and Poverty Reduction The potential for research investments to reduce of the poor. Several decades of experience pro- poverty depends on many factors, including the vide three major conclusions that justify broad- structure of farm types and markets, the strat- based research investments for poverty egy adopted for public financing of agricultural reduction in many circumstances: research, and the strength and effectiveness of research institutions, especially their ability to * Except where farm size is strongly bimodal, reach the poor. In all cases, investment in re- small-scale farmers have usually adopted search is only one instrument for poverty reduc- improved technologies with a short lag time tion, and must always be designed as part of a relative to larger farmers wider national strategy for poverty reduction. * Increasing agricultural productivity through yield-enhancing technology has enhanced BROAD-BASED GROWTH STRATEGIES HAVE rural employment and wages with direct SHARPLY REDUCED POVERTY benefits to the poorest households * Indirect benefits of technical change for poor A key issue in financing research investments consumers (both rural and urban) have of- for poverty reduction is whether to invest for ten been larger than the direct effects on pro- broad-based productivity growth that provides ducers, especially where rapid technical both direct and indirect benefits to the poor (for change lowers prices of food staples that both producers and consumers) or to focus "pro- constitute a large portion of the expendi- poor" investments directly on the livelihoods tures of poor households.1 1 Globally increased cereal yields-mainly due to technological innovation-contributed to a decline in real prices for cereals of 28-30 percent between 1982 and 1997 (Rosegrant and others 2001). 4 Key Issues for Research and Poverty Reduction 5 The broad-based approach can be the major from poverty. The low elasticity for Latin strategy for research investments where: agri- America reflects both the low share of GNP from culture is important to the incomes of rural poor; agriculture, as well as the bimodal agrarian the agro-ecological base allows significant po- structure in which commercial farms have re- tential for productivity growth; land distribu- ceived most of the gains in agricultural produc- tion is relatively equitable; the poor consume tivity. In Latin America, agricultural research non-tradeable food staples; and reasonably well may still have strong poverty reduction impacts, functioning markets exist for agricultural inputs but must be carefully targeted on poor produc- and outputs. In such situations, productivity ers and consumers. growth in the dominant small-scale farming sec- tor will likely have large poverty reduction ef- fects. In these situations, efforts to shift research FUTURE STRATEGIES REQUIRE POVERTY resources from general productivity growth to TARGETS other social objectives have often resulted in unacceptable productivity trade-offs, ultimately Recent developments suggest increased oppor- threatening interests of the poor. tunities for focusing on pro-poor investments, Nonetheless, the preconditions for win-win even in areas that have previously benefited productivity growth and poverty reduction (for from broad-based approaches. example, unimodal agriculture, and reasonably well-functioning input and output markets) are * Rural incomes are becoming more diversi- often not present. Estimated elasticities of pov- fied, with a declining share of income from erty reduction with respect to agricultural pro- agriculture, especially from food staples ductivity increases in different regions are * Increasing private investment in research for shown in Table 2. These are relatively high in more commercial agriculture should free Africa and Asia, where large numbers of poor public resources to focus on poorer farmers people depend on agriculture. For example, an and marginal areas increase in yields of 1 percent in Africa reduces * Liberalization of trade has integrated mar- the number of poor people by 0.72 percent. kets, reducing effects of productivity growth Based on the estimated effect of research invest- on domestic food prices and opening new ments on yields, an investment of $179 in agri- opportunities for high-value crops in spe- cultural research would remove one person cific locations Table 2: Effect of a 1 Percent Increase in Yields on Reduction in the Number of People Living on Less than $1 Per Day Elasticity of Research cost Number in numberofpoor perperson Percentin poverty to yield removed from Region poverty (millions) changes poverty EastAsia 15 278 0.48 $179 South Asia 40 522 0.48 $179 Sub-Saharan Africa 46 291 0.72 $144 Latin America 16 78 0.10 $11,397 Source: Thirtle, Lin and Piesse (2002). 6 Designing Investments in Agricultural Researchfor Enhanced Poverty Impacts * Food producers are subject to competitive national commitment to agricultural research pressures that can squeeze many farmers- and to addressing fiscal and governance prob- especially poor farmers-out of markets or lems of public research institutions. At the same further marginalize them, unless they de- time, the private sector has little incentive to fill velop new knowledge and skills the gap in research funding, especially in de- * There is growing pressure to focus on less- veloping technologies for poor farmers. A first favored regions and environments that have priority must be to address these institutional been bypassed by past research investments weaknesses to enhance the overall efficiency and * Livelihood shocks, due to natural disasters, effectiveness of research programs. Institutional conflict and market risks, are becoming innovations to improve research system perfor- more common and increasing the vulner- mance regardless of type of user will likely pro- ability of the poor. vide large benefits to the poor as well (Byerlee 2000). Analysis of the potential for research invest- The development of a national constituency ments to contribute to poverty reduction must for research is critical to building effective and recognize this complexity and the possible sustainable research programs. Providing a tradeoffs between targeting general productiv- voice to the poor in the operation of those pro- ity growth versus direct poverty reduction. Spe- grams is also important to enhance responsive- cific strategies will depend largely on poverty ness to issues of importance to them. Enhancing diagnosis relating to livelihood strategies of the client influence over, and support for, public poor, their resource constraints, and market sector research underlies many current reforms environments for agricultural products, but (decentralization, participatory planning and some targeting of pro-poor investment will implementation, user co-financing, better moni- likely be appropriate in most situations. toring and evaluation, and client responsive- ness). These reforms promote development of the "demand-side" of the market for research OVERCOMING INsTnTUTIoNAL WEAKNESSES services and influence political processes at the IN RESEARCH SYSTEMS national level that determine the level of financ- ing for research and who gets the services fi- Although past investment in agricultural re- nanced. Investments must build a strong search has had important impacts on poverty, constituency for public sector research and a many research systems are constrained by in- constituency that ensures that this public re- stitutional weaknesses that limit their efficiency, search addresses public goods issues and the effectiveness, and sustainability. These prob- needs of the poor. lems are multiplied by growing funding con- Past research programs were often not de- straints. Although most products of agricultural signed with adequate understanding of the research continue to be public goods, public needs and constraints of users, and of the poor funding for agricultural research averages only in particular and, even as research institutions 0.6 percent of agricultural GDP for developing become more user oriented and demand-driven, countries, compared to 2.6 percent for indus- they may not necessarily address the needs of trial countries (Pardey and Beintema 2001). the poor. Priorities tend to be set by wealthier Public research organizations in many coun- farmers who are more politically powerful, tries have serious institutional weaknesses re- more accessible geographically, socially and lin- lated to their incentives to attract quality guistically, and more literate and cognizant of scientists, lack of operating budgets, and lack research potential. A positive development in of effective links with clients. The underlying recent years has been the growth in number problem in all of these constraints is inadequate of private, quasi-private, not-for-profit, and Key Issues for Research and Poverty Reduction 7 community-based organizations that are stepping recovery mechanisms to increase sustain- in to fill gaps in public sector provision of agricul- ability bias services against the poor? tural research services. Many of the new organi- * Are the poor able to participate and are they zations are primarily or exclusively dedicated to involved in meaningful ways in program working with poor rural communities that often governance and evaluation? How can hu- had little voice under old institutional structures. man and social capital development be Key issues to consider when assessing op- maximized to empower the poor in setting portunities for improving agricultural research the research agenda? systems include: * Is sufficient attention given to a pluralistic ADDRESSING DIVERSITY WITHIN THE SECTOR mix of public, private and NGO organiza- tions in funding and delivery of research Diversity in farm types and farming systems services? Which types of farms are typically shapes the potential impact of agricultural re- served by which institutions? search programs on poverty reduction. Farms * Is the service cost-effective in terms of cost vary substantially in resources available and the per beneficiary and impact? Are programs production and marketing environments in financially sustainable and able to retain which the farm operate (Figure 1) (Berdegu6 and high quality human resources? Will cost- Escobar 2001).2 Figure 1: Conceptual Delineation of Three Farming Worlds - Commercial (A), Small-Scale Market-Oriented (B), and Subsistence-Oriented (C) HIGH ASSET POSITION 0r >>> S t11 l - m ° Ow I a. z 0~~~~~~~~ z~~~~~~~~~~ LOW ASSET POSITION 2 To be complete, this figure should include rural landless households that depend on agriculture for wage labor and employment. Often the poorest groups, they benefit directly from employment resulting from increases in farm productivity, as well as from lower food prices. 8 Designing Investments in Agricultural Research for Enhanced Poverty Impacts Agricultural innovation by commercial For very poor farm households in unfavor- farmers (A) in favorable environments can lead able production environments (C), agricultural to poverty reduction if it generates employment innovation can usually lead to sustainable (for example, labor for production of high value progress in poverty reduction only as part of a commodities) and/or provides cheaper food, broad-based effort to diversify livelihood strat- both of which have positive poverty alleviation egies of those households. In some extremely effects. However, much of the research for this harsh environments, there is little potential to sector should be provided by the private sector. improve agricultural productivity and invest- Development of small-scale but market-ori- ment in social and economic infrastructure to ented family farms (B) in areas with reasonable increase options, because off-farm employment or good agricultural potential offers the great- and out-migration will usually provide higher est opportunity for direct poverty reduction payoffs than investment in research. through agricultural innovation. Agricultural The direct and indirect poverty reduction research policies must seek to increase produc- impacts and strategies for each farm type are tivity broadly within this group and/or help summarized in Table 3. The focus of agricultural these farmers diversify into production systems research programs must depend on a diagnosis in which they have a competitive advantage. of the relative importance and dynamics of the Table 3: Strategies for Enhancing Poverty Reduction Impacts of Agricultural Research by Farm Type Major institutional Sector Direct impacts Indirect impacts actors Commercial Increase micronutrient content Promote labor-intensive, high- Private agribusiness farmers of food to provide nutritional value industries (for example, Public regulatory benefits. cut flowers and horticulture). framework Increase productivity to reduce Producer and trade food prices for nontradable organizations food staples. Small, market- Diversify production to increase Increase agricultural productiv- Public research oriented value-addition and high value ity to stimulate overall econ- Public-private farmers crop/livestock production. omic growth. partnerships Develop technologies to reduce Develop labor-intensive produc- Producer organiza- production risks. tion systems to generate tions Strengthen producer organiza- employment. NGOs tions to improve demand for Increase productivity to reduce research and build human food prices for nontradable and social capital. food staples. Subsistence- Increase productivity and reduce Build human and social capital Public research oriented production risks for food for necessary to address a range Producer and com- farmers home consumption. of livelihood opportunities. muniity organiza- Encourage market access for tions higher value crops/livestock. NGOs Improve natural resource man- agement and develop envi- ronmental services markets. Key Issuesfor Research and Poverty Reduction 9 effects on each of these farm groups. In situations needed research will vary in each case. The pub- where small-scale family farms in moderate to lic sector usually has the mandate to conduct re- high potential areas dominate the sector, these search relevant to the needs of the poor, and NGOs should be the focus of research programs. How- and community organizations can be important ever, there are situations where the poor are con- for the poorest groups (C). centrated in unfavorable production environments A further issue, especially important to poor and-to the extent that these areas have potential farm households, is that of risk and vulnerabil- to raise productivity-they must be the focus of ity. For these households, research efforts should research. In any event, the institutional actors with seek to minimize risk and develop more diver- comparative advantage for implementing the sified livelihood strategies. 3 Good Practicefor Enhancing Poverty Reduction Impacts of Investments in Research Systems Good practice in developing research programs surveys and poverty reduction strategy pro- for enhanced poverty reduction requires atten- grams (PRSPs) are good starting points for this tion to strategy development, priority setting, analysis. The analyst should look for patterns institutional development, uptake of findings, of geographical concentration of poverty, ar- and monitoring and evaluation. eas where agriculture is especially important for the poor, and major causes of risk and vul- nerability (for example, droughts, markets, ANALYZING THE STRUCTURE AND NATURE disease epidemics). More information on pov- OF POVERTY erty analyses and PRSPs is available on the World Bank PovertyNet website . enhancing the poverty reduction effects of agri- Within agriculture, an analysis of com- cultural research investment is the analysis of modities important to poor producers and the different types of poverty (rural and urban), consumers will guide priority setting. The their determinants, the contexts in which they analysis must focus on whether increased ag- occur, and the livelihood response strategies ricultural productivity is likely to contribute adopted by the poor. Poverty assessment to poverty reduction and what innovations are 10 Good Practicefor Enhancing Poverty Reduction Impacts 11 likely to be the most effective. Finally, a diag- * Where small-scale farming dominates and nosis of the institutional strengths and weak- rural poverty is pervasive, broad-based pro- nesses of research systems should assess how ductivity investment strategies may maxi- effectively different institutions respond to de- mize poverty reduction both directly mands of the poor (see Box 3 for a summary through increased incomes of farmers and of some of the key issues for an assessment). indirectly through food prices, employment generation, and wider growth linkages. SELECTING A BROAD STRATEGY . Where most of the poor are net consumers of agricultural products, investment in agricul- The wide variation in country situations re- tural research might be justified irrespective quires adapting research strategies to the coun- of the parthcipation of the poor in production. try or local situation. Table 4 taken from Hazell * Where the poor are concentrated in certain and Haddad (2001) shows how quite different geographical regions or production systems, priorities and strategies emerge between and focused investment on particular commodi- within countries depending on: (i) the income ties, regions, and technology traits and on level of the country, (ii) market and trade poli- empowering rural communities might be cies, (iii) agro climatic potential, (iv) rural infra- appropriate (see next section). structure, service provision and market access, * Where the poor lack access to sufficient land and (v) the labor situation. Although it is com- and other assets, or the agricultural poten- mon wisdom that research for the poor must tial of those assets is very low, investments focus on marginal areas, this analysis shows that in agricultural research may have little at least for low-income countries, research in direct effect on poverty reduction. high potential areas will often provide higher payoffs (Box 4). To tailor national research strategies to meet Examples of common strategies include: needs of the various farm types, while respond- Box 3: Assessing Agricultural Conditions and Research Systems Key points to consider: * Location, type of poverty and of poor households- net food purchasers/ sellers, labor users/ sellers * Role of agriculture in the livelihoods of the poor, and their main crop and livestock activities * Farm types based on size and production systems, disaggregated by region, ethnicity, and gender * Income by farm type and livelihood strategy and their variability. * Status and trend in condition of natural resource base (water resources, pasture, land) * Market and infrastructure base-transport costs and degree of integration with markets * Research providers and types of producers served * Linkages (effectiveness and efficiency) to research delivery systems and input markets. Information sources: * Household surveys, especially poverty assessments * Rapid rural appraisals * Participatory assessments of production systems and service delivery * National statistics and research and extension reports 12 Designing Investnents in Agricultural Researchfor Enhanced Poverty Impacts Table 4: Strategies for Agricultural Research to Reduce National Poverty Good/infrastructure Poorinfrastructure Surplus labor Scarce labor Surplus labor Scarce labor Low High Low High Low High Low High potential potential potential potential potential potential potential potential Mid-income country Markets 1,2,3,5 2,3,5,8 1,4,6 4,6,8 1,3;5,7 3,5,8 1,4,6,7 4,6,8 liberalized Markets not 1,2,3,5 1,2,3,5,8 1,4,6 1,4,6,8 1,3,5,7,9 1,3,5,8 1,4,6,7,9 1,4,6,8 liberalized Low-income country Markets 3,5 1,2,3,5,8 2,4,5,8 1,2,4,5,6,8 1,3,5,7,9 1,3,5,7,9 1,4,5,7,9 1,4,5,7,9 liberalized Markets not 1,3,5,8,9 1,4,5,8,9 1,4,5,8,9 1,3,5,7,9 1,3,5,7,9 1,4,5,7,9 1,4,5,7,9 1,4,5,7,9 liberalized Priorities for agricultural research: 1 - Staple food production; 2 - High-value crops, trees and livestock; 3 - Employment intensive growth; 4- Increased labor productivity; 5 - Smallholder farms; 6- Medium and large farms; 7- Low external-input farming; 8 - High external-input farming; 9 - Nutritional content of food staples. Shaded areas are the high priority regions within the country group. Source: Hazell and Haddad (2001). ing to pressures of different political interest levels, since research typically generates many groups and attempting to meet the needs of poor spillovers and indirect effects (anticipated and producers and consumers is a major challenge. unanticipated). However, resources can be al- National strategies will have to be pragmatic, located to specific priorities-by commodity, recognizing key issues but planning for invest- crop trait, and ecological zone-that are impor- ments within the constraints of available re- tant to poor households (both producers and sources and information. consumers). When based on good diagnostic work, pro-poor priority setting may improve allocation of public funds in many situations, SETTING SPECIFIC PRIORmES FOR by targeting research investments by (Byerlee INVESTMENTS 2000): Pro-poor research investment faces difficulties * Geographic regions with high rates of pov- in that benefits of research activities cannot di- erty, especially less-favored production ar- rectly target individual households or income eas (for example hills, rainfed areas, arid Good Practicefor Enhancing Poverty Reduction Impacts 13 Box 4: Research Strategies: Marginal versus Favorable Areas Many policymakers and donors call for more re- in favored areas through the operation of labor search on marginal areas as a way of reducing pov- markets (that is, migration from marginal areas) and erty and arresting resource degradation. This may food markets (that is, lowered cost of food) (Renkow be appropriate, but the situation is usually much 2000; Otsuka and David 1993). more complex. While the poorest of the poor are Nonetheless, there is some evidence from In- often located in marginal areas, the number of dia suggesting that3/4because of declining marginal people living in poverty is often higher in more fa- returns for investments in irrigated areas-invest- vored areas (Kelly and Parthasarathy Rao 1995). In ments in technology for medium potentia/rainfed some of the most difficult areas, the rate of gain areas might now generate greater impacts on from improved technology may be very slow. Pro- growth and poverty (Fan, Zhang and Zhang 2000). viding infrastructure, education and non-farm op- In fact, research resources in India have shifted portunities may bring earlier, greater, and more sharply toward rainfed and more fragile environ- sustainable benefit (to both poverty reduction and ments over the past decade (Pal and Byerlee 2002). conservation of natural resources) than is possible In other countries, there is a strong trend toward with agricultural intensification. This is especially true diversification of research demands due to increas- for areas with very high production risks and/or a ing exploitation and trade in niche markets for pro- fragile resource base. In addition, there are often duce from marginal areas, and emerging substantial spillover benefits from productivity gains environmental concerns. regions) provided there is demonstrated * Low external input production systems, potential to achieve productivity gains. A suited to farm households unable or unwill- high proportion of the poor in Vietnam, for ing to purchase production inputs. Poor example, live in upland and mountainous farmers in upland areas of Mexico and Cen- areas, and research investment focused on tral America for example, have extensively these areas offers opportunities for en- adopted Mucuna, a leguminous cover crop hanced poverty impacts (there are currently that reduces their need to purchase nitrog- almost no research institutes located in these enous fertilizer. areas or focused on their problems). * Diversification into high-value, labor- Commodities produced or consumed by intensive crops and livestock or upgrading the poor, especially staple foods and nutri- of traditional niche products. Developing ent-enhanced foods high in micronutrients. options for postharvest handling and pro- In Nigeria, for example, cassava with a cessing can reduce postharvest losses and higher yield and enhanced disease resis- quality loss and can offer producers oppor- tance was widely adopted, resulting in a tunities for value-added processing. Small substantial fall in real prices to consumers. farmers in Mexico, for example, successfully Since cassava is the staple food of the poor export organic coffee with the help of local in Nigeria, poor consumers captured a large and international NGOs. share of the benefits (Box 5). Similarly, en- * Environmental and natural resource con- hancing the nutritional value of products servation important to the poor (forests, consumed by the poor can produce major pastures, fisheries). Beekeeping, for ex- benefits (Box 6). ample, is often an important niche activity 14 Designing Investments in Agricultural Researchfor Enhanced Poverty Impacts Box 5: Distribution of Benefits of Cassava Research in Nigeria Cassava is the most important food staple in Ni- (1999) estimated that consumers captured 72% geria. The Nigerian Roots and Tubers Research in- of the benefits of this research through lower stitute, in collaboration with IITA introduced prices and that poorer consumers captured a dis- improved cassava varieties in the 1980s that re- proportion share of these benefits. And since poor sulted in yield increases of about 30 percent on farmers consume most of their produce, they 50 percent of the cassava area. Afolami and Falusi gained relative to larger farmers. Percentage increase in household income 10 - 8 6 4 2-- _ ,_,_,____,__ , < 2 2-4 4-6 6-8 8-10 >10 Income level ('000 Naira) of the very poor in Africa. Development of risks, substitutes for external inputs, and adds quality standards and market research can product traits (such as enhanced rnicronutri- allow beekeepers to expand into urban, re- ent content in foods important to the poor) (see gional and global markets (Christoplos, Table 5 for examples). Farrington, and Kidd, 2000). Investments in more basic sciences, such as Priority setting and the design of technolo- biotechnology, can also be pro-poor, especially gies relevant to poor farmers requires greatly if it reduces vulnerability to pests and clirnatic increased capacity in the social sciences in all Box 6: Eradicating Childhood Blindness in Africa Agricultural research is helping address the prob- white-fleshed varieties with little or no beta-caro- lem of vitamin A deficiency that threatens hundreds tene (a precursor of vitamin A). Participatory meth- of thousands of African children with childhood ods with largely female and poor farmers, who are blindness and contributes to infant mortality. Re- the main producers and consumers of sweet pota- searchers have identified orange-fleshed sweet po- toes, have shown that the orange-fleshed varieties tato varieties that provide a year-round source of are also acceptable to consumers. The new variet- vitamin A, produce excellent harvests with little in- ies have now been extensively adopted. Research puts, and are acceptable to consumers. This is par- has shown that even small amounts of the new ticularly important in southern and eastern Africa sweet potato varieties in the diet can eliminate vi- where vitamin A-related diseases are widespread, tamin A deficiencies. and the traditionally grown sweet potatoes are Source: CIP 2002. Good Practicefor Enhancing Poverty Reduction Impacts 15 Table 5: Crop Biotechnologies to Meet Needs of Poor Farmers and Consumers Technique Biotechnological solution Tissue culture-plant micro- This relatively low-cost technology has been adopted by poor farmers propagation to produce healthy for bananas in China, taro in Samoa, multipurpose trees in Kenya, po- plants tato in Vietnam, and cassava in Colombia. Anther culture (a subcategory of Crosses between Asian and African rice varieties resulted in new Afri- tissue culture) to reduce time can rice varieties with higher yields and pest, weed, disease and cli- needed for interspecific crosses matic tolerance relative to traditional varieties. Poor farmers in upland rice areas of West Africa are now rapidly adopting these varieties. Molecular markers to identify Molecular markers are being employed in many breeding programs, more precisely genes for specific including breeding for sweet potato viruses (important to poor produc- characteristics, and thus speed ers) and for quality protein maize (important to poor consumers). conventional breeding Genetic engineering to insert Some genetically modified products are being commercialized and genetic material into crop and widely adopted (Bt cotton for insect resistance in South Africa, Mexico, livestock species and China); others are under field testing (i.e., virus-resistant sweet potatoes and potatoes, Bt maize for insect resistance, and rice with late blight resistance). Many others under development offer traits of spe- cial significance to the poor (potatoes with higher protein, enhanced vitamin A rice, and rice with salinity tolerance). research systems. Investment programs need accountability. Universities, NGOs and pro- to provide support for building such capacity ducer organizations can often play a special both within research organizations and by role in meeting the demands of the poor for contracting-in needed skills from universities. research services. In Nepal, for example, a local NGO along with farmers has under- STRENGTHENING THE INSTITUTiONAL BASE taken an extensive participatory varietal testing program that resulted in a sharp in- Inrecognizingthat thepotentialfortar g is crease in adoption of new varieties that were In recogmzmg thtthe potential for targethng iS highly case specific and not easily generalized, developed but not yet released by the pub- investment strategies must promote the devel- lic research system. opment of research institutions that address the * Promoting private sector research and free- needs of the poor in a sustainable way. These ing up and refocusing public resources for often require participatory approaches that pro-poor programs, by reducing policy and empower the poor at the local level to demand regulatory barriers to private investment. services. Specific institutional reforms might For example. privatization of hybrid maize include (Table 6): research should free resources for the pub- lic sector to produce open-pollinated vari- * Developing an institutionally pluralistic eties for poorer farmers and regions. system that decentralizes or privatizes ser- * Emphasizing institutional reforms that vices, and strengthens client orientation and employ contractual and competitive 16 Designing Investments in Agricultural Researchfor Enhanced Poverty Impacts Table 6: Summary of Pro-Poor Institutional Approaches Investment principle Objective Pro-poor implementation Examples Focus public programs To ensure that government Give priority to poverty reduc- Privatization of research on on public goods programs benefit the gen- tion and food security. appropriable technologies eral public and do not crowd (for example, hybrid maize). out private funding of re- Phase out public sector invest- search. ment in private goods. Farmer controlled levies on commercial crops and re- duced public funding. Institutional pluralism To encourage development Involve NGOs and producer or- Competitivefundingthrough of a range of public and pri- ganizations that tend to have a producer associations (Co- vate institutions with unique pro-poor orientation. lombia, Guinea). comparative advantages in provision of research. Use of participatory To empower local communi- Mobilize the poor and develop Local Agricultural Research approaches ties, tap local knowledge and their initiative and problem- Committees (CIALs) (Latin resources, and provide ser- solving ability. America). vices that suit local condi- tions. Work through, and strengthen, producer organizations and NGOs. Build culture, incen- To enhance responsiveness Improve institutional arrange- National and regional inno- tives and methodolo- and effectiveness in dealing ments for collaboration among vation councils and regional gies for pro-poor with pro-poor programs un- research and extension actors technology fora. responsiveness and der conditions of increasing and their clients within pro- collaboration among livelihood vulnerability. poor programs. agencies Conduct socioeconomic studies of poor client needs and re- sources. Decentralization of op- To increase user influence Encourage regional governance Regional fora with strong erational authority and over programs by devolving mechanisms with stakeholder participation of minority responsibilities responsibilities to the re- voice representing the poor. groups and women (Peru gional level. and Colombia). Building human and To promote emergence of a Finance capacity development Investment in capacity build- social capital new generation of institu- of producer organizations. ing of producer organiza- tions representing the rural tions and private service poor. Strengthen agricultural educa- providers. tion and training programs. Cost sharing by major To enhance financial sus- Recognize trade-offs between Decentralized research funds stakeholders tainability of research pro- financial sustainability and par- (Peru and Mexico). grams by sharing the burden ticipation of the poor. of funding among central and local governments, the Use graduated payment private commercial sector schemes to require larger farm- and farmers. ers to pay a greater percentage of program costs than poor farmers. Good Practice for Enhancing Poverty Reduction Impacts 17 mechanisms to improve efficiency, and pro- countries where domestic "market size" is mote pro-poor effectiveness. Many of these insufficient to justify comprehensive re- mechanisms can be directed toward poverty search programs. enhancement. For example, a competitive * Building capacity of producer organiza- grants program in Colombia has awarded tions to integrate user groups into research research and extension grants giving high and extension systems, to strengthen tech- weight to their likely benefits to poor farm- nical and managerial capacities, and to en- ers and landless laborers. courage inclusion of women and the poor. * Promoting financial sustainability through Farmers in Burkina Faso and Guinea, for reforms that improve confidence in public example, have received support to organize institutions, and building political support and contract adaptive research according to for research especially from poorer groups. their needs. Cost recovery and cost sharing are impor- tant to financial sustainability, and even While all of these approaches may be rel- quite poor farmers have shown willingness evant, depending on the specific situation, the to cofinance services that meet their de- general consensus from experiences in World mands (for example, cofinancing of research Bank lending is that institutional reforms require services by small-scale farmers in Peru). The long-term commitments over many years. They tradeoff, however, between sustainable must also give greater attention to strengthen- cofinancing and participation of the poor ing the demand for research through decentrali- must be recognized, and the level of zation and building capacity of producer cofinancing needs to be adjusted to ability organizations and user groups, complemented to pay. by opportunities to express demands through * Decentralizing research programs to put competitive approaches to funding. Use of ad- iustable program lending, combined with decen- scientists closer to users and enhance user )ust r l participation in setting the research agenda. tralized capacity building, is showing promising This is especially relevant for adaptive and results in terms of reducing poverty in a num- location-specific research activities, as dem- ber of countries. onstrated by programs in Peru and Colom- bia that have established regional fora with PROMOTING INCLUSIVE PARTICIPATION strong representation of the poor. This brings all technology users and suppliers P r Participatory research approaches are impor- within a region together to agree on priori- tant tools to assess the relevance of technolo- ties for small-scale farmers. Decentralization gies, to identify client needs, and to develop may alsobuild political supportfor research, coordinated responses to those needs. They as shown by the rallying of small farmers in also help to build capacity of individuals and Colombia to protest threatened closure of a local institutions, and to encourage demand research and extension project. for services where there has been little in the * Promoting intemational strategic alliances past (Box 7). and regional research collaboration, be- Typically women plav a major role in agri- cause links to the outside world are impor- cultural production, but have much less influ- tant for all systems to access new ence on technological innovation than do men. knowledge, and enhance system efficiency Poverty rates are significantly higher for through collaboration. Such linkages are women-headed households. Because of past especially important for the many small neglect of women's needs, gender analysis is an 18 Designing Investments in Agricultural Researchfor Enhanced Poverty Impacts Box 7: Farmer-Run Research: Experience with the CIALs Turning responsibility for on-farm testing over to from the organizing institution assist in planning and farmers is an attractive alternative that has been ex- analysis of research trials and a paraprofessional (a tensively tested by Centro Internacional de Agricul- CIAL-experienced farmer) monitors and advises on tural Tropical (CIAT) in several NARSs of Latin the research. Technical staff visit 2-3 times per sea- America. Under the CIAL program (CIAL-Spanish son after the first 2-3 seasons. acronym for "Local Committee for Adaptive Re- CIALs operate with a small CIAL Fund (US$500 search), begun in 1990, an institution with interest per community) to cover the risks of crop failure or in technology dissemination (usually a state agency, to subsidize the costs of trials. These CIAL funds have NGO, or cooperative) facilitates a meeting in which been consolidated into a corporation at the national a community analyzes potential needs for local tech- level, but each CIAL manages its own fund. The nology testing. If the community is interested in un- funds, like the whole program, are "owned" by the dertaking local research, it selects a four-member community, and managed by the Committee. Committee (the CIAL) from the community to co- ordinate the research work. Outside technical staff Source: Ashby and others, 2000. essential base for planning pro-poor research * Allocating resources for work on crops, investments (Fong and Bhushan 1996). products, and production systems most rel- Minorities such as tribal groups, religious evant to these groups, and minorities, caste groups and other disadvan- * Promoting and working through traditional taged groups, are also disproportionately rep- and other organizations of these groups. resented among the poor and extreme poor. Frequently, they reside in the least favorable production environments in remote mountain- LINKING TO INNOVATION UPIAKE PATHWAYS ous or arid regions. They often have limited re- sources and restricted social and economic Even where technologies are appropriate, poor contact with the wider society. For these rea- farmers are often disadvantaged because they sons and because of language barriers, inten- lack access to the findings from research pro- tional neglect or discrimination, and distrust or grams. Larger, wealthier farmers have more outsiders, they are often difficult to reach with channels for accessing information and inputs agricultural services. to increase productivity. Even the public exten- Agricultural research programs can im- sion services charged with serving small farm- prove their relevance to these groups through ers are more likely to serve larger farmers as the following types of initiatives: active clients. Public research programs must therefore develop pro-active strategies to make * Increasing the number of women and mi- the results of research available to the poor by norities (with appropriate language skills) developing partnerships and alliances with a in staffing research institutions: variety of intermediaries that can work with * Ensuring representation of these groups in small farmers. Public extension services remain program planning, priority setting, and gov- important, but private companies, NGOs, ernance farmer organizations, and the media are equally Good Practicefor Enhancing Poverty Reduction Impacts 19 important in facilitating uptake of new technolo- PUTIING IN PLACE AN M&E SYSTEM gies. The International Rice Research Institute (IRRI), for example, and the Bangladesh Rice Good monitoring and evaluation (M&E) sys- Research Institute are collaborating with NGOs, tems for research programs are needed to track extension agencies, seed companies, and farmer implementation in terms of pro-poor orienta- groups to test mechanisms for speeding the tion. Potential indicators for impact, outcome, uptake of rice technologies. Experiences of the and output relevant to research investments are various partners are shared in an "Uptake Fo- given in Table 7 (Alex and Byerlee 2000), but rum." Private companies (some with large re- more work is needed to develop quantitative search budgets), the agro-food industry, and and qualitative indicators to improve M&E of environmental groups are other potential part- research impacts on poverty reduction. Evalu- ners for promoting pro-poor development and ating investments solely on the basis of eco- for financing research efforts. nomic returns can bias investment decisions Table 7: Examples of Indicators for Poverty Reduction in Agricultural Research Projects Measurement Project Impact-Level Indicator: * Distribution of economic benefits of research by income level of producers Share of increased benefits and consumers (using economic surplus methods) to the poor * Sustainable and active community groups / civil society or farmer organiza- Number tions Outcome Indicators: * Adoption rates for new practices or technologies, by sex and farm size Percent of all farmers * Farmer attitudes toward research Percent positive responses * Change in productivity (cost/unit) as a result of new technology or practices, Percent change from base by sex and farm size * Change in yield variability due to new technologies Percent change from base * Share of funds managed and co-financed by producer organizations/com- Percent change from base munity groups * Extent of on-farm participatory research Percent of scientist time and research budget Output Indicators: * Change in research resource allocation to poor regions and commodities of Percent change from base importance to the poor * Changes in orientation of research (i.e., to multidisciplinary, multi-institu- Percent of total cost tional) * New technologies released with specific applicability to poor regions or farmers Number * Number of projects executed by producer organizations or NGOs with a Number pro-poor orientation * Number of projects executed that have an orientation toward women and Number minority groups 20 Designing Investments in Agricultural Researchfor Enhanced Poverty Impacts toward commercial crops and high potential ability, risk reduction, greater social capital). areas. Evaluations on the basis of cost per ben- Even economic impacts are difficult to mea- eficiary or cost per household removed from sure because of the variability between small poverty are more appropriate to assessing pov- farms and the range of livelihood strategies erty reduction investments, but are more diffi- small farm households pursue. For these rea- cult to estimate. sons, beneficiary assessments and participa- Many research benefits to the poor are dif- tory evaluations by the poor are valuable tools ficult to quantify in economic terms (greater for assessing research impacts on poverty (See food self-sufficiency, environmental sustain- Box 8). Box 8: Beneficiary Assessment Beneficiary assessment is used to gather infor- been used for extension programs, where they mation on users' views of programs and services. have been useful in understanding weaknesses Beneficiary assessment is primarily a management and strengths of the programs and have led to tool to assist managers in improving quality and important changes in policy for service delivery. relevance of services. Conversational interviews EMBRAPA, the national research organization in and group discussions elicit information from ser- Brazil, conducts an annual survey to provide feed- vices, users, providers, and stakeholders to iden- back from users. tify program strengths and weaknesses. Beneficiary assessments have most commonly Source: Salmen (2000) References Adato, M., and R. Meinzen-Dick. 2002. Assessing the Byerlee, D. 2000. "Targeting Poverty Alleviation in Impact of Agricultural Research on Poverty Using Priority Setting for Agricultural Research." Food the Sustainable Livelihoods Framework. EPTD Dis- Policy, 25: 429-45. cussion Paper 89, FCND Discussion Paper 128, CEP. 2002. Eradicating childhood blindness in Africa: The IFPRI Washington D.C. March. promise of sweet potato. http://www.cipotato.org/ Alex, G.,and D. Byerlee. 2000. Monitoring and Evalu- News/pressreleases/english/VlTA.htm. Intema- ation for AKIS Projects: Framework and Options: tional Potato Center,Lima, Peru. AKIS Good Practice Note. Agricultural Knowl- Christoplos, I., J. Farrington, and A. Kidd. 2000. Ex- edge and Information Systems Thematic Team tension, Poverty and Vulnerability: A Contribution. (research), World Bank, Washington, D.C. Working Paper 144, ODI., London. Alston, J.M., Chan-Kang, C., Marra, M.C., Pardey, Dixon, J., A. Gulliver, and D. Gibbon. 2001. Farming P.G., and Wyatt, T.J. 2000. A Meta-Analysis of Systems and Poverty: Improving Farmers' Liveli- Rates of Return to Agricultural R&D: Ex Pede hoods in a Changing World. FAO, Rome, and the Herculem? IFPRI Research Report 113. Intema- World Bank, Washington, D.C. tional Food Policy Research Institute, Washing- Evenson, R.E., and D. Gollen (eds.) 2002. Crop Vari- ton, DC. ety Improvement and its Effect on Productivity-The Ashby, J. A., A. R. Braun, T. Gracia, M. Guerrero, Impact of International Agricultural Research. L.A. Hernandez, C. A. Quir6s, and J. I. Roa. 2000. Wallingford, U.K.: CABI. Investing in Farmers as Researchers: Experience with Falusi, A. O., and C. A. Afolami, 1999. Effect of Tech- Local Agricultural Research Committees in Latin nology Change and Commercialization on Income America. CIAT, Cali. Equity in Nigeria: The Case of Improved Cassava. Berdegue, J., and G. Escobar. 2001. Agricultural Knowl- http://www.ciat.cgiar.org/povertyworkshop/ edge and Information Systems and Poverty Reduc- papers.htm, International Centre for Tropical Ag- tion. AKIS/ART Discussion paper. Rural riculture,Cali,Colombia. Development. Dept, World Bank, Washington, Fan, S., L. Zhang, and X. Zhang. 2000. Growth and D.C. Poverty in Rural China: The Role of Public Invest- 21 22 Designing Investments in Agricultural Researchfor Enhanced Poverty Impacts ment: EPTD Discussion Paper No. 66. IFPRI, Otsuka K., and C. C. David (eds.) 1993. Modern Rice Washington, D.C. Technology and Income Distribution in Asia. Boul- Fan, S., P. Hazell, and S. Thorat. 1999. Linkages be- der, Col: L. Rienner Publishers. tween Government Spending, Growth, and Poverty Pal, S., and D. Byerlee. 2002. (Forthcomning) The Fund- in Rural India. Research Report No. 110. IFPRI, ing and Organization of Agricultural Research in Washington, D.C. India: Evolution and Emerging Policy Issues. Inter- FAO/AKIS. 2000. Agricultural Knowledge and Infor- national Food Policy Research Institute, Wash- mation Systemsfor Rural development (AKISIRD): ington, D.C. Strategic Vision and Guiding Principles. FAO, Pardey, P.G., and N.M. Beintema. 2001.Slow Magic: Rome, and World Bank, Washington, D.C. Agricultural research a Century after Mendel. Food Farrington, J., I. Christoplos, and A. Kidd with M. Policy Report, International Food Policy Re- Beckman. 2002. Extension, Poverty and Vulnerabil- search Institute, Washington, D.C. ity: The Scopefor Policy Reform3AlFinal Report of a Renkow, M. 2000. Poverty, productivity and produc- Studyfor the Neuchatel Initiative. Working Paper tion environment A review of the evidence. Food 155. Overseas Development Institute (ODI), Lon- Policy 25:463-78 don, U.K. Salmen, L. 2000. Beneficiary Assessments of Agricultural Fong, M., and A. Bhushan. 1996. Toolkit on Gender in Extension: A Manual of Good Practice. AKIS Good Agriculture: Gender Toolkit Series No. 1. World Practice Note. AKIS Thematic Group, World Bank, Washington, D.C. Bank, Washington, D.C. Hazell, P., and L. Haddad. 2001. Agricultural Research Sarris, A. 2001. The Role of Agriculture in Economic and Poverty Reduction. Food, Agriculture and the Development and Poverty Reduction: An Empirical Environment, Discussion Paper 34, IFPRI, Wash- and Conceptual Foundation. Rural Strategy Work- ington, D.C. ing Paper, Rural Development Department, IFAD. 2001. Rural Poverty Report 2001: The Challenge World Bank, Washington, D.C. of Ending Rural Poverty. Oxford University Press, Thirtle, C., L. Lin, and J. Piesse. 2002. The Impact of London, U.K. Research-Led Agricultural Productivity Growth on Irz, X. L. Lin, C. Thirtle, and S. Wiggins. 2001 Agri- Poverty Reduction in Africa, Asia, and Latin cultural productivity growth and poverty alle- America. Unpublished paper, Imperial College of viation, Development Policy Review, 19: 449-66. Science, Technology, and Medicine, London, Kelley, T.G.,andP. PartiasarathyRao. 1995. "Marginal U.K.. environments and the poor: Evidence from India." Timmer, C.P.1997. How well do the poor connect to the Economic and Political Weekly, 30(40): 2494-5. growth process?. Consulting Assistance on Eco- Kerr, J., and S. Kolavalli. 1999. Impact of Agricultural nomic Reform Discussion Paper No. 178, Research on Poverty Alleviation: Conceptual Frame- Harvard Institute for International Development, work with Illustrations from the Literature. EPTD Cambridge, MA. Discussion Paper No. 56, IFPRI, Washington, World Bank. 2001. World Development Report 2000/ D.C. 2001: Attacking Poverty. Washington, D.C. Titles in the Agriculture & Rural Development Working Paper Series[ Number 1-Investing in Science for Agriculture & Rural Development in the 21' Century Number 2-No-till Farming for Sustainable Rural Development Number 3-A Road Map from Conventional to No-till Farming Number 4-Europe & Central Asia-Food and Agriculture in the Slovak Republic: The Challenges of EU Accession Number S-Integrated Pest Management in Development: Review of Trends and Implementation Strategies aThe first three titles were not numbered at the time of printing, but they do represent the first products in this series. I i. 6 DEVELOPMENT Ag - Vm 19e. I siii1tll - so iZ. 20433 a CoS. L ,