Agriculture and Achieving The Millennium Development Goals Agriculture and Rural Development Department THE WORLD BANK Agriculture & Rural Development Department World Bank 1818 H Street, N.W. Washington, D.C. 20433 http://www.worldbank.org/rural REPORT NO. 32729-GLB Agriculture and Achieving The Millennium Development Goals Agriculture and Rural Development Department THE WORLD BANK This volume is a product of the staff of the International Bank for Reconstruction and Development/The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accu- racy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www. copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Contents A C R O N Y M S A N D A B B R E V I AT I O N S v E X E C U T I V E S U M M A RY vii 1 Introduction 1 2 Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals 3 MDG 1--Eradicate Extreme Poverty and Hunger 5 Poverty Reduction Through Agriculture-Led Economic Growth 7 Agriculture's Contribution to Food and Nutrition Security 8 MDG 2--Achieve Universal Primary Education 9 MDG 3--Promote Gender Equality and Empower Women 10 MDG 4--Reduce Child Mortality 10 MDG 5--Improve Maternal Health 12 MDG 6--Combat HIV/AIDS, Malaria, and Other Diseases 13 MDG 7--Ensure Environmental Sustainability 13 MDG 8--Develop a Global Partnership for Development 15 3 Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on Poverty 19 Country-Level Assessment of Selected MDGs: Ethiopia and Zambia 19 Ethiopia: Agriculture-Led Growth and 22 Zambia: Facing the Challenge of the Millennium Development Goals-- The Role of Agriculture 29 Conclusions 34 Global Quantified Assessment of Selected MDG Targets: IMPACT-WATER 36 Results 41 Implications for Investment 46 Conclusions 48 iii iv Contents 4 Global trends and Emerging Issues in Meeting the MDGs: a Trade, Macroeconomic, and Policy Perspective 49 Policy Actions and Their Impacts on the MDGs 49 Trade Policies 50 Policies of Developed Countries 50 Policies of Developing Countries 52 Policies in Development Assistance: The Case of Food Aid 53 Macroeconomic Reforms and National Government Infrastructure Investments 54 Importance of Rural Infrastructure 55 On Public Investment and Infrastructure 58 The Role of the Private Sector in Supporting Agriculture to Meet the MDGs 58 The Rural Nonfarm Private Sector 60 Effective Legal, Regulatory, and Institutional Environments 61 Private-Public Partnerships and Investment Synergies: The Case of Water and Agricultural Research 61 The Role of Foreign Direct Private Investment 65 The Role of Governance Structures in Agriculture for Achieving the MDGs 65 Conclusion 67 5 Conclusions 69 E N D N O T E S 7 5 R E F E R E N C E S 7 7 Appendix 1. The Millennium Development Goals, Targets, and Indicators 81 Appendix 2. IMPACT-WATER Developing Countries and Region 84 Contents v Figures 2.1 Proportion of Population Living on Less than $1 Per Day (PPP), Most Recent Estimates 4 2.2 Agriculture Value Added, as a Percentage of GNP, 2001 4 2.3 Progress in Attaining MDG 1 Targets for Poverty and Undernourishment, by Region (Trend Lines Based on Recent Performance and Targets for 2015) 5 2.4 Child Malnutrition by Urban or Rural Residence, Stunting (Low Height for Age) Prevalence Among Preschoolers, Surveys Since 1999 6 2.5 Prevalence of Undernourishment--Proportion of the Population Unable to Acquire Sufficient Calories to Meet their Daily Caloric Requirements, 2003 Estimate 8 2.6 Child Mortality by Urban or Rural Residence, Surveys Since 1999 11 2.7 Estimated Prevalence of Iron Deficiency Anemia in Women Aged 15 to 49, 2004 12 3.1 Poverty Rate in Ethiopia Under Different Growth Scenarios 29 3.2 Poverty Reduction Under the Current Growth Path (2001­2050) 32 3.3 Per-Capita Kilocalorie Availability in 1995 and Projections to 2015, Baseline and MDG Scenarios 43 3.4 Child Malnutrition, Developing Countries, 1995, 2015 Baseline, 2015 MDG Scenario, and 2015 Target 43 3.5 Child Malnutrition, Developing Country Regions, 2015 Baseline, 2015 MDG Scenario, and 2015 Target 44 3.6 Number of People Without Access to an Improved Water Source, 1995, and Projected 2015 Baseline and MDG Scenarios 46 3.7 Cost Estimates for Implementing the Baseline and MDG Scenario, 2015, Developing Countries 47 4.1 Changes in Paved Roads, Telephone Lines, and Electricity Production Over Time, Selected Regions 57 4.2 Sample of Public-Private Investment Arrangements 62 Tables 2.1 Summary of Links Between the Agricultural Sector and the Millennium Development Goals, Principally at Household Level 18 3.1 Comparison of Poverty Across Selected African Countries 20 3.2 Poverty Lines in the Country Case Studies 20 3.3 Economic and Demographic Indicators, 2001 21 3.4 Average Trends Across Countries, 1991­2001 Averages 21 3.5 Cereal Production in Selected African Countries 22 3.6 Growth and Poverty Reduction Under Agricultural Growth Options 24 3.7 Poverty in Zambia During the 1990s 31 vi 3.8 Growth and Poverty Reduction Under the Alternative Zambian Growth Scenarios (2001­2015) 35 3.9 Child Malnutrition Estimates (Underweight), 1990 and 1995, with 2015 MDG Indicator Target 38 3.10 Access to Safe Drinking Water, 1990 and 1995, with 2015 MDG Target Indicator and 2015 IMPACT-WATER Baseline Estimates 39 3.11 Parameters for Developing Countries, Baseline and MDG Scenarios 41 3.12 Income Growth Requirements for Baseline and MDG Scenarios, 1995­2015 42 3.13 Southeast Asia, Share of Malnourished Preschool Children, 1995, and 2015 Baseline, MDG Scenario, and Indicator Target 45 4.1 Latin America and the Caribbean: Main Features of Changes in Agricultural Policies 55 4.2 Possible Scenario Regarding Financing Sources for Agriculture under NEPAD 64 This report was prepared by Mark W. Rosegrant, Claudia Ringler, Todd Benson, Xinshen Diao, Danielle Resnick, James Thurlow, Maximo Torero, and David Orden International Food Policy Research Institute and coordinated by Nwanze Okidegbe, Derek Byerlee, and Sanjiva Cooke (World Bank). Editorial assistance was provided by Melissa Williams (World Bank) and Sarah A. Cline and Rowena Andrea Valmonte-Santos (consultants). This report was prepared under the general direction of Kevin Cleaver (sector director) and Sushma Ganguly (sector manager). Acronyms and Abbreviations ACC/SCN Administrative Committee on Coordination/ Subcommittee on Nutrition AfDB African Development Bank AIARD Association for International Agriculture and Rural Development BRI Bank Rakyat Indonesia CBOs Community-Based Organizations CPI Corruption Perceptions Index 2004 EC European Commission EPRDF Ethiopia Peoples' Revolutionary Democratic Front FAO Food and Agriculture Organization FAOSTAT Food and Agriculture Organization Statistical Database FDI Foreign Direct Investment GATT General Agreement on Trade and Tariffs GDP Gross Domestic Product GNP Gross National Product HIPC Heavily Indebted Poor Countries HIV/AIDS Human Immunodeficiency Virus/ Acquired Immune Deficiency Syndrome IFAD International Fund for Agricultural Development MAAIF Plan for Modernization of Agriculture MDGs Millennium Development Goals MFPED Poverty Eradication Action Plan MIGA Multilateral Investment Guarantee Agency MOFED Ministry of Finance and Economic Development MOPED Ministry of Planning and Economic Development ODA Official Development Assistance OECD Organization for Economic Co-operation and Development PPP Purchasing Power Parity PRSP Poverty Reduction Strategy Papers NEPAD New Partnership for Africa's Development UN United Nations vii viii Acronyms and Abbreviations UNDP United Nations Development Program UNESCO United Nations Educational, Scientific, and Cultural Organization UNICEF United Nations Children's Fund SAPs Structural Adjustment Programs WANA West Asia and North Africa WFP World Food Program WHO World Health Organization WTO World Trade Organization Vice President: Ian Johnson Sector Director: Kevin Cleaver Sector Manager: Sushma Ganguly Task Team Leader: Nwanze Okidegbe Task Team: Derek Byerlee, Sanjiva Cooke, Melissa Williams Executive Summary Relationship Between Agriculture and the Millennium Development Goals In 2000 the member states of the United Nations adopted the Millennium Declaration as a renewed commitment to human devel- opment. The Declaration includes eight Millennium Development Goals (MDGs), each with quantified targets, to motivate the interna- tional community and provide an accountability mechanism for actions taken to enable millions of poor people to improve their livelihoods. The MDGs are as follows: 1. Eradicate extreme poverty and hunger 2. Achieve universal primary education 3. Promote gender equality and empower women 4. Reduce child mortality 5. Improve maternal health 6. Combat human immunodeficiency virus/acquired immune deficiency syndrome (HIV/AIDS), malaria, and other diseases 7. Ensure environmental sustainability 8. Develop a global partnership for development. About 70 percent of the MDGs' target group live in rural areas, particularly in Asia and Africa, and for most of the rural poor agri- culture is a critical component in the successful attainment of the MDGs. Even though structural transformations are important in the longer term, more immediate gains in poor households' welfare can be achieved through agriculture, which can help the poor overcome some of the critical constraints they now face in meeting their basic needs. Thus, a necessary component in meeting the MDGs by 2015 in many parts of the world is a more productive and profitable agri- cultural sector. While the linkage with agriculture is particularly strong for the first MDG, or MDG 1--halving by 2015 the proportion of those suf- fering from extreme poverty and hunger--all MDGs have direct or indirect linkages with agriculture. Agriculture contributes to MDG 1 through agriculture-led economic growth and through improved nutrition. In low-income countries economic growth, which enables increased employment and rising wages, is the only means by which the poor will be able to satisfy their needs sustainably. MDG 2, on universal education, has the most indirect linkage to agriculture. A more dynamic agricultural sector will change the ix x Agriculture and Achieving the Millennium Development Goals assessment of economic returns to educating chil- based on alternative development scenarios from dren, compared to the returns from keeping chil- economy-wide models. Ethiopia is a predomi- dren out of school to work in household nantly subsistence economy with agriculture con- (agricultural) enterprises. Agriculture contributes tributing 52 percent to GDP and with almost to MDG 3 directly through the empowerment of 85 percent of the population living in rural areas. women farmers and indirectly through reduction Productivity in the sector is low, due to frequent of the time burden on women for domestic tasks. droughts, limited input use, and poor infrastruc- Agriculture contributes to reduced child mortality ture. In contrast, mining traditionally has domi- (MDG 4) indirectly by increasing diversity of food nated economic growth in Zambia, which in turn production and making more resources available has marginalized the country's agricultural sector. to manage childhood illnesses. Agriculture di- Zambia has both a higher proportion of its pop- rectly helps improve maternal health (MDG 5) ulation living below the poverty line and a higher through more diversified food production and concentration of individuals at the low end of the higher-quality diets, and indirectly through in- income distribution. creased incomes and, thus, reduced time burdens Model results for Ethiopia show that if the on women. Agriculture also directly helps to com- country stays on a business-as-usual growth path, bat HIV/AIDS, malaria, and other diseases (MDG poverty will increase by another 10 million people, 6) through higher-quality diets, and indirectly by and food security will be compromised even fur- providing additional income that can be devoted ther. The largest impact on poverty and food secu- to health services. Agriculture practices can be rity can be achieved through a focus on growth in both direct causes of and important solutions to staple crops, which today account for 65 percent of environmental degradation (MDG 7). More pro- agricultural value added as well as most small- ductive agricultural technologies allow the with- holder employment. Rapid growth in the livestock drawal of agriculture from marginal, sensitive sector has the most significant effect on overall environments. Developing a global partnership economic growth but has a relatively smaller for development (MDG 8) will help maintain the poverty alleviation effect. Very rapid growth in the steady increase in agricultural trade and signifi- nontraditional-export sector fuels total economic cant increases in development assistance offered growth as well, but has little impact on rural to the agricultural sector, increases that help sus- poverty levels. Accelerated growth in all three tain the benefits from agriculture in the longer sectors would help slash the poverty rate by term 16 percentage points to 27 percent by 2015. The simulated growth in staple food production could be achieved through a doubling of the irrigation Agriculture and MDG 1: Country-Level area by 2015, and by improving the efficiency of Analysis for Ethiopia and Zambia fertilizer use combined with enhanced seed use. Compared with other regions of the developing Moreover, as more than 50 percent of the poor live world, Sub-Saharan Africa faces the largest chal- in food-deficit areas where the availability of food lenge in terms of meeting MDG 1. In 2001, about staples per household is half the national average, 47 percent of the population was living below the market access and market development need to be international poverty line. Agriculture will be the integral parts of a national agricultural develop- primary means of addressing this challenge, as ment strategy. Enhanced market access, chiefly 65 percent of the people in the region derive their through large investments in improved and ex- livelihoods from the sector. However, the contri- tended road networks, would reduce the national bution that agriculture can make in achieving the poverty rate to 22.7 percent, and thus help MDGs in this region depends on the particular Ethiopia reach the MDG 1 target. constraints and opportunities prevailing within In Zambia, the poverty rate under business-as- each of the countries. usual growth would still be 68 percent by 2015, In this report, Ethiopia and Zambia, with rural only 7 percentage points lower than the current poverty rates of 45 percent and 86 percent, respec- poverty rate of 75 percent. Annual GDP growth of tively, are analyzed according to the level and type 8.8 percent would be required to halve poverty of agricultural growth required to meet the MDGs by 2015. Although agriculture accounts for only Executive Summary xi 25 percent of GDP, it is still the main source of was made, but from very high levels, and parts of livelihood for most of the country's population, Southeast Asia. Levels in West Asia and North including the majority of Zambia's poor who live Africa are comparatively low, but have remained in rural areas where the incidence and severity of virtually unchanged over the last 30 years. poverty is greatest. More rapid productivity The MDG scenario combines two broad courses growth under the Agriculture-Led Growth for improving food security and reducing poverty Scenario would lead to higher sectoral growth for in developing regions: the first way is through both staples and export crops. Under a focus on broad-based and rapid agricultural productivity nonagricultural growth, rural households would and economic growth to increase effective in- benefit from increased demand in urban areas, but comes, effective food demand, and food availabil- the overall effect on poverty would be relatively ity; and the second is through investments in small. Agricultural processing within the manu- education, social services, and health (proxied in facturing sector, however, does in fact represent a the model by female secondary enrollment rates, potential area for growth and poverty reduction. the female-male ratio of life expectancy at birth, Within the Agriculture-Led Growth Scenario, a and access to clean water). Staples-Led Growth Scenario combined with poor The changes in agricultural and complementary market access would again have little impact on social indicators result in a reduction in total child poverty among small-scale farm households. An malnutrition from 24 percent under business-as- Export-Crop-Led Growth Scenario would favor usual in 2015 to 17 percent under the MDG scenario, rural medium-scale households due to market and a reduction from 131 million children to credit constraints for the rural poor. 91 million children. Under the MDG scenario, to bring developing countries, particularly South Asia and Sub-Saharan Africa, within reach of the pre- Agriculture and MDG 1: Regional Analyses school malnutrition target indicator, total invest- with a Focus on Child Malnutrition ments in agricultural and supporting sectors during Malnutrition affects nearly one-third of all chil- 1995­2015 will have to increase by $161 billion* dren under five years of age in developing coun- based on IMPACT-WATER calculations. The three tries--174 million children in 1990. More than half main areas of investment for the MDG scenario in of childhood deaths are associated with under- percentage terms are rural roads, irrigation, and weight, and malnourished children who survive education. Together these three areas will require into adulthood are more likely to suffer from $403 billion between 1995 and 2015 to achieve the chronic illness and disability, and have a higher rapid reductions in childhood malnutrition sim- probability of reduced physical and intellectual ulated under the MDG scenario. Agricultural productivity. The IMPACT-WATER model is used research investments account for $109 billion, and to project the proportion of malnourished children $78 billion of investments are required toward under business-as-usual and an alternative MDG increasing access to safe water. Due to the long lags scenario that attempts to close the gap between in the generation of impacts from agricultural the MDG target rate of childhood malnutrition research, increases in research expenditures--even and business-as-usual outcomes. The gap is closed beginning now--will have relatively small impacts through growth in agriculture as well as comple- on crop yields by 2015. Increased investments in mentary investments in social sectors, and special agricultural research are likely to be essential to attention is given to those countries and regions meet crop and animal production needs beyond least likely to reduce malnutrition significantly by 2015, however. Other investments, such as roads other means. Under business-as-usual, the devel- and irrigation, have significant lags in impact as oping-country level of childhood malnutrition is well, so implementing the investment portfolio still 24 percent by 2015, down from 30 percent in required for the MDG scenario will require very 1990. Regions least likely to reach the MDG target rapid action. As relatively high levels of access to indicator on halving childhood malnutrition are clean drinking water are already achieved in the Sub-Saharan Africa, where the number of mal- baseline scenario, only $15 billion in investments nourished children has increased over the last are added for the MDG scenario; these investments 30 years, South Asia, where substantial progress have virtually no lag period in becoming effective. xii Agriculture and Achieving the Millennium Development Goals Role of Trade, Policies, tance in place of food if existing food-aid pro- and Governance Systems grams were terminated. Thus, attention needs to focus on how its effectiveness can be maximized In addition to the targeted investments required to and potential harms mitigated. help agriculture achieve its potential contribution Infrastructure is of particular concern as one of for the MDGs, policies and governance systems the key inputs entering into the "production func- need to be supportive for agriculture to achieve tion" of the MDGs and the achievement of many maximum impact. of the MDG targets, from poverty reduction to Supportive systems and policies include trade environmental sustainability targets. In Sub- and domestic support polices for agriculture in Saharan Africa in particular, a lack of adequate developed and developing countries, macro- infrastructure, typically attributed to geography economic reform and public-sector infrastructure and poor initial conditions, clearly impedes more and other investments, the role of the private sec- productive agriculture. Achieving the health and tor and public-private partnerships, and general education MDGs will require more than health good governance. and education interventions; in particular, infra- Price support policies and border protection of structure services have a crucial role to play. Piped wealthy Organization for Economic Co-operation water is crucial to reduce diarrhea in young chil- and Development (OECD) countries, valued at dren, while electricity allows for more hours of hundreds of billions of dollars each year, cause studying and road access promotes easier estab- harm to agriculture in developing countries. lishment of schools and higher attendance. Where Removal of OECD protection could boost rural the government believes that service should be value added in low- and middle-income countries provided beyond what a well-functioning market by $60 billion per year. The increase in world will offer, subsidies may be justified to promote prices from removal of OECD protection would additional investment to achieve these govern- lead to larger agricultural production in develop- ment goals. ing countries. An important area for developing To improve the effectiveness of public invest- countries to increase participation in international ment, increased coordination at the country, markets is through the buildup of capacity to pro- regional, and donor levels is necessary because the duce for exacting standards of importing markets. linkages and complementarities of infrastructure Developing countries themselves retain substan- investment have often not been realized. More- tial trade barriers and have varying requirements over, the traditional approach of top-to-bottom regarding agriculture liberalization, due to differ- infrastructure development has to be changed to a ences in trade specialization and needs of net food more demand-driven approach. Finally, the im- imports. When developing countries join in agri- pediments to efficient markets in rural areas need cultural trade liberalization, they can achieve to be addressed through regulatory reforms in overall welfare gains of $20 billion annually, twice order to increase the availability and effectiveness the gains in national welfare compared to reforms of resources to address the real access gap in these in the developed countries only. Overall, a suc- areas. cessful conclusion for agriculture in the World Public intervention alone is not sufficient to Trade Organization Doha Development Round deliver the services and investments required to trade negotiations can make an important contri- achieve the MDGs. To alleviate rural poverty in bution to achieving the MDGs, by establishing developing countries, the private sector can con- sustainable positive incentives for agricultural tribute to economic growth through job creation production among developing countries. both on and off farm. The private sector also has Food aid is another component of international an important role in supporting timely and effi- transactions that directly and indirectly affects cient credit availability to fuel agricultural devel- rural poverty in a globalized agricultural economy opment and growth. Finally, the private rural and therefore could have a significant impact in nonfarm sector is an important engine for rural achieving the MDG targets. While the provision of development. In Asia, for example, the rural non- aid in the form of food is not the optimal form for farm economy accounts for 20­50 percent of total development assistance, donors would probably rural employment and 30­60 percent of total rural not provide equivalent cash development assis- income. Executive Summary xiii Private financing of investment and MDGs may poor governance and poor institutions typically be explored as a general way to ensure availability have poor public services, including those for agri- of basic services, particularly as the official devel- cultural extension and research, and have particu- opment assistance or aid for the water and other larly poor social services like water provision and sectors has declined in recent years. Financing education. Good governance can be built through options should also include income redistribution the development of social capital, particularly in and tax efforts in China, India, and other poor rural areas, where participation by individuals in countries to co-finance the MDGs. social networks increases the availability of infor- Foreign direct investment (FDI) and other long- mation and lowers its cost, helps enforce property term, relatively stable investments have a signifi- rights, and reduces opportunistic behavior in nat- cant impact on agricultural and overall economic ural resource use--thus supporting all MDGs. growth. However, the impact of this investment Effective community-based organizations, such on hunger and poverty may be limited. Little of as farmer associations or cooperatives, water user this investment goes directly into agriculture or groups, and farm and other microcredit and lend- rural areas. Moreover, FDI is largely concentrated ing groups can improve governance, for example, in just a few countries. by educating and sensitizing the public about their Ideally, the public and private sectors comple- rights and entitlements under public programs; by ment each other, with the government providing acting as a conduit to the government for public an appropriate enabling environment for private opinion and local experience; by influencing local initiatives to develop. Public-private partnerships agricultural development policies; and by helping are an important way to increase financial, human, government and donors fashion a more effective and social capital in rural areas. Partnerships can development strategy through strengthening include publicly provided training for small- and institutions, staff training, and improving man- medium-scale enterprises, partnerships in educa- agement capacity. At a higher level, donor coun- tion, agricultural research, the provision of infor- tries and international aid agencies should focus mation and communication technologies, the their resources on those countries where good or expansion of rural infrastructure including roads, improving governance structures will ensure that and the development of rural industrial clusters. the rural poor are reached. Finally, at the interna- Moreover, through partnerships, public research tional level, enhanced governance and commit- institutions can gain access to advanced scientific ment could help bring about improvements in the knowledge and technologies held by the private global trade and environmental agendas. Progress sector, mechanisms for developing, processing, in the trade agenda will come from regarding the marketing, and distributing final products to needs of the poorest countries, including en- farmers and consumers, and financial resources hanced access to both agricultural and other mar- that are increasingly difficult to obtain. In the area kets; progress in the global environmental agenda of agricultural research, however, a sustained pub- will come from the gains of developing countries lic role in funding agricultural research will be from enhanced environmental standards, includ- essential, particularly for crops and regions that ing protection of remaining biodiversity in these are unlikely to be served by the private sector, countries, or from participating in and gaining such as those in less favorable environments. from climate mitigation policies. Good governance is typically defined under the terms of accountability, transparency, predict- Conclusions ability, and participation. These principles are only meaningful and supportive of agricultural devel- Given that the majority of poor people live in rural opment toward achieving the MDGs if adequate areas or rely on agriculture, and that agriculture institutional and social structures are available. paves the way for economic growth in the poorer Countries with a good governance structure and nations, agricultural and rural development will adequate institutions tend to ensure political and underlie progress on the broad array of economic economic stability, possess reasonable state capac- and social indicators emphasized by the MDGs. ity, enforce property rights and contracts, provide The most effective strategy for making steady, sus- sufficient public goods, and limit government cor- tainable progress toward the MDGs is to serve all ruption and predation. Conversely, countries with the goals in an integrated way. However, each goal xiv Agriculture and Achieving the Millennium Development Goals will need a well-defined package of technologies greater number of private entrepreneurs to and services for success at the field level. explore opportunities in agribusiness. A healthy Of the eight Millennium Development Goals, market and private sector would provide value- the first goal is the one whose attainment most added, skilled work to the landless poor and gen- clearly involves the agricultural sector: The poor erate multiple livelihood opportunities in both the around the globe are disproportionately farmers farm and nonfarm sectors. Other important and herders, and, perversely, the hungry also most reforms in the trade area include the elimination of commonly find their livelihoods through agricul- export subsidies; the move toward measures that ture. By increasing food availability and incomes support income instead of stimulating production and contributing to asset diversity and economic in developed countries, and the reform of the growth, higher agricultural productivity and sup- international and national governance of food aid portive pro-poor policies allow people to break out programs. of the poverty-hunger-malnutrition trap. As the As the financing requirements for realizing the country-level model simulations revealed, broad- MDGs are substantial, the private sector is increas- based agricultural growth is the key for decreasing ingly called upon to fill investment gaps. Its com- poverty and increasing growth in Sub-Saharan plementary and supporting role in the provision Africa. A global assessment of Target 2 of MDG 1 of basic services in water, land, health, and other (halving child malnutrition levels) shows that the infrastructure development that is lacking in combination of agricultural and economic growth most developing countries cannot be ignored. It together with larger investments in social sectors, will take a particular kind of private-sector including health and education, can substantially involvement to generate the necessary economic narrow the gap between the business-as-usual out- transformations. Private entrepreneurs are now comes for 2015--24 percent of developing-country increasingly held to environmental, social, and preschool children malnourished--and the target corporate governance principles that stress sus- indicator--15 percent children malnourished--to tainable business practices and adherence to labor reach 17 percent. However, the outcome varies sig- standards. Without these standards and practices, nificantly by country and region. Latin America, the private sector and disadvantaged groups West Asia and North Africa, and China will, on cannot mutually benefit from consumer, employ- average, likely get close to the target indicator by ment, and entrepreneurial activities. Government 2015, even under business-as-usual; however, the agencies in developing countries urgently need likelihood that Sub-Saharan Africa and South Asia to revisit the legal, regulatory, political, and in- will come close to their respective target rates is stitutional framework in agriculture, research, much smaller. The total increase in investments extension, and industrial sectors to facilitate pri- estimated is $161 billion in agricultural and sup- vate-sector involvement. Moreover, both the porting sectors during 1995­2015. In addition to private and public sectors must foster private- these investments, significant policy and gover- public­sector partnerships and cultivate this rela- nance reform is required. tionship with the end objective of addressing the To achieve faster agriculture-based growth MDGs. Moreover, to allow the agriculture sector rates, there must be in place favorable macro- to develop its full potential for achieving the economic and trade policies, good infrastructure, MDGs, the share of ODA spent on agriculture and access to credit, land, and markets. These con- needs to increase significantly. ditions create level playing fields and give farmers It is a promising development that the review of incentives to adopt new and sustainable technolo- progress--and lack thereof--in achieving the gies and diversify production into higher-value MDGs has reached global attention. Calls for crops, actions that raise incomes and lift house- accountability and action that have real impact on holds out of poverty. An improved domestic regu- people are growing because of that attention. latory framework would intensify competition Policy action and increased investment in the crit- among suppliers of essential inputs, such as seeds ical arenas of sustainable agriculture productivity and fertilizer. In addition, the elimination of trade and food and nutrition security will be essential barriers for agricultural products, especially the for responding effectively and responsibly to high-value-added products, would encourage a reach the Millennium Development Goals. 1 Introduction Today, 1.1 billion people Today, 1.1 billion people in the world are living on less than $1* per day--430 million in South Asia, 325 million in Sub-Saharan Africa, in the world are living on 260 million in East Asia and the Pacific, and 55 million in Latin America (World Bank 2004a). Although in aggregate, human material condi- less than one dollar per tions have improved greatly over the past century, the continued daily deprivation of the basic needs of millions of people testifies to day and most depend on the fact that the global challenge of human development is far from agriculture for their met. Too many children around the world today lead lives character- ized by hunger, illness, and, all too often, early death. livelihoods. In order to establish a renewed commitment to human develop- ment, in 2000 the member states of the United Nations adopted the Millennium Declaration. Emerging from the Millennium Summit, the declaration recommits the community of nations to a broad range of steps to lead to a "more peaceful, prosperous and just world" and "to free our fellow men, women and children from the abject and dehu- manizing conditions of extreme poverty . . . to making the right to de- velopment a reality for everyone and to freeing the entire human race from want." In contrast to previous global policy statements, the Millennium Declaration also established eight goals--the Millennium Development Goals (MDGs)--each with quantified targets, to moti- vate the international community and provide a mechanism for ac- countability in undertaking action to enable millions of poor people to live lives of dignity, free from extreme want. The goals are: 1. Eradicate extreme poverty and hunger 2. Achieve universal primary education 3. Promote gender equality and empower women 4. Reduce child mortality 5. Improve maternal health 6. Combat human immunodeficiency virus/acquired immune de- ficiency syndrome (HIV/AIDS), malaria, and other diseases 7. Ensure environmental sustainability 8. Develop a global partnership for development. For each goal, one or more targets have been set, most for 2015, using 1990 as a benchmark. The goals, targets, and specific indicators are described in Appendix 1 of this document. Since the Millennium Summit, the MDGs have come to serve as key objectives in guiding the planning and implementation of a broad *All dollar amounts are U.S. dollars. 1 2 Agriculture and Achieving the Millennium Development Goals range of global and national development efforts. human capital needed to reach those targets will also Progress toward the attainment of the MDG targets result in a considerably more productive and resilient is being monitored, and policies and programs are agricultural sector. There is, therefore, a need to being modified to bring about the changes needed undertake a critical review of the pathways by which to attain those targets. agriculture can contribute directly or indirectly to The vast majority of people whose lives need to attaining the MDGs. change the most, in order to attain the targets speci- Following an explanation of the MDGs and a dis- fied in the MDGs, depend on agriculture for their cussion of their progress to date, we assess goal-by- livelihoods. Coming up with strategic options for as- goal the potential contributions of agriculture to sisting these individuals and their households is a achieving the MDGs in the second chapter. The necessary component for improving global perfor- third chapter examines more closely how the targets mance in meeting the MDGs. However, there are of MDG 1 can be achieved in Sub-Saharan Africa little or no diagnostics of the direct and indirect links (Ethiopia and Zambia), and how agricultural and between the MDGs and agriculture. Moreover, some economic growth, together with larger investments of the indirect effects will come about through broad, in social sectors including health and education, can economy-wide processes, while others, particularly substantially narrow the gap between business-as- those operating at the household and community usual and MDG outcomes for Target 2 of MDG 1 level, will be felt more immediately. Additionally, (halving child malnutrition levels). This report then cause-and-effect relationships between the agricul- examines how trade and macroeconomic policies tural sector and the MDGs are not all one way. While for agriculture can improve the attainment of the the agricultural sector provides critical inputs for at- MDGs, particularly the goal of halving poverty. The taining the MDG targets, the broad improvements in final section sets out some conclusions. 2 Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals The poorest countries of About 70 percent of those living on less than $1 a day live in rural areas (World Bank 2002).1 For the majority of the poor in the world, the world are those in agriculture is a critical component in the successful attainment of the MDGs. Although the rural poor pursue a range of strategies to assure which agriculture is the their livelihoods, the dominant strategy is food production through cropping or raising livestock.2 The vast majority of people whose predominant source of lives need to change the most to attain the targets specified in the MDGs are farmers and herders. The material well-being of these employment. To a large individuals and households is dependent upon the productivity of degree, the poverty their cropping and livestock husbandry activities. As shown in figure 2.1 the poorest countries of the world are also those in which agricul- experienced in these ture is the predominant sector of employment. To a large degree, the poverty experienced in these countries is a product of unproductive countries is a result of agriculture. Moreover, the dominance of agriculture in the economies of the poorest countries (as shown in figure 2.2) often is more a reflec- unproductive agriculture. tion of a poorly performing economy in which subsistence agriculture is serving as a safety net of last resort for populations with limited economic options, rather than as an effective engine of economic growth. Although in the longer term a broad transformation and diversification of rural economies away from a strong dependence on agriculture is desirable, more immediate gains in the welfare of poor households are most likely to come through the poor overcoming some of the critical constraints they now face in meeting their basic needs through agriculture. Thus, a necessary component in meeting the MDGs by 2015 in many parts of the world is a more productive and profitable agricultural sector. In this chapter, we will make a broad assessment on an MDG-by- MDG basis to consider how more productive agricultural activities and a more vibrant agricultural sector in general might significantly advance a country's efforts to attain each MDG. How might improved agricultural performance at household, community, and economy- wide levels bring progress toward the MDGs? A summary of the discussion on how the agricultural sector can contribute to the attain- ment of the MDGs, as well as how, in turn, progress in attaining the MDGs can animate the agricultural sector is provided in table 2.1 at the end of this chapter. Direct and indirect effects need to be distinguished. We will pay more attention to those MDGs that are directly influenced by agricul- tural activities--most notably, MDG 1, halving by 2015 the proportion of those suffering from extreme poverty and from hunger. Moreover, some of the indirect effects will come about through broad economy- 3 4 Agriculture and Achieving the Millennium Development Goals Figure 2.1 Proportion of Population Living on Less than $1 Per Day (PPP), Most Recent Estimates less than 20 percent 20 ­ 30 30 ­ 40 40 ­ 50 50 ­ 60 more than 60 percent no data Source: World Bank (2003). Note: PPP = purchasing power parity. Figure 2.2 Agriculture Value Added, as a Percentage of GNP, 2001 less than 20 percent 20 ­ 30 30 ­ 40 40 ­ 50 50 ­ 60 more than 60 percent no data Source: World Bank (2003). Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals 5 wide processes across sectors, while others, particu- MDG 1--ERADICATE EXTREME larly those operating at the household and commu- POVERTY AND HUNGER nity levels, will be more immediately felt. Of course, for the more localized indirect effects of agriculture Progress in meeting the targets of the first MDG is to contribute effectively to achieving the MDGs, encouraging in East Asia, generally adequate in such effects must be spread widely throughout the South Asia and Latin America, but disturbingly rural population. Additionally, cause-and-effect poor in Sub-Saharan Africa. Based on the simple relationships between the agricultural sector and trend line shown in figure 2.3, the prevalence of the MDGs are not all one way. While the agricul- dollar-a-day poverty in East Asia in 2015 should tural sector provides critical inputs to attaining the be considerably less than the 50 percent reduction MDG targets, the broad improvements in human from 1990 levels. In Latin America and South Asia, capital needed to reach those targets will also pro- poverty rates should be close to the targets, if vide an important foundation from which a consid- slightly above. erably more productive and resilient agricultural However, for Sub-Saharan Africa, the propor- sector can be developed. Finally, while most MDG tion of the population living on less than $1 per targets are complementary, some might actually day in 2015 is quite likely to have increased from involve tradeoffs. For example, enhanced access to that of the reference year 1990. The regional trends improved drinking water sources might collide, in in reduction of undernutrition, while all moving some regions, with the goal of reduced hunger downward, are not as clear-cut as the trends in through increased irrigated agriculture. Similarly, poverty. The undernutrition target is quite likely several indicators of MDG 7, ensuring environmen- to be met in East Asia and in Latin America (see tal sustainability, might well be adversely affected also the global assessment of childhood malnutri- by efforts aimed at increasing agricultural and eco- tion in chapter 4). However, in both South Asia and nomic development that are important for the Sub-Saharan Africa, undernutrition will likely achievement of MDG 1. remain at levels considerably above the targets for Figure 2.3 Progress in Attaining MDG 1 Targets for Poverty and Undernourishment, by Region (Trend Lines Based on Recent Performance and Targets for 2015) 50 40 (%) Sub-Saharan Africa Sub-Saharan Africa day a 40 (%) 30 US$1 than 30 South Asia MDG less South Asia Targets 20 on East Asia & Pacific undernourishment 20 of East Asia & Pacific living people 10 of 10 Prevalence Latin America & Caribbean Latin America & Caribbean Share 0 0 1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015 Source: World Bank (2004b) (left), and FAO (2003) (right). 6 Agriculture and Achieving the Millennium Development Goals 2015. Indeed, in Sub-Saharan Africa, using recent market, increased agricultural income will directly performance as a guide, little progress in reducing improve both household consumption levels today, undernutrition will have been made. and household asset levels to improve production We have already observed that poverty has a and better weather economic shocks in the future. strong rural and, hence, agricultural dimension: Increased food production will lead to real reduc- Approximately 70 percent of the world's poor live in tions in food prices, which will improve the pur- rural areas and primarily pursue agriculture-based chasing power of the poor throughout the economy, livelihood strategies. Hunger is endemic in most whether they are engaged in agriculture or some rural areas of the developing world. Consequently, other sector. More important, agriculture can serve as shown in figure 2.4, malnutrition levels are con- as the basis for broad pro-poor economic growth sistently higher in rural areas than in urban zones. to bring about permanent reductions in poverty. Improving the productivity of and the economic Second, with complementary nutritional factors in returns of agriculture for farming households will place, both subsistence farming households and have immediate effects in eradicating extreme those purchasing food in local markets will enjoy poverty and reducing hunger. First, through the immediate physiological benefits from increased Figure 2.4 Child Malnutrition by Urban or Rural Residence, Stunting (Low Height for Age) Prevalence Among Preschoolers, Surveys Since 1999 Nigeria Zambia Malawi Uganda Ethiopia Rwanda Burkina Faso Zimbabwe Benin Tanzania Mali Mauritania Guinea Gabon Egypt Rural Haiti Urban Peru Colombia Bangladesh Cambodia Nepal Turkmenistan Armenia Kazakhstan 0 10 20 30 40 50 Percent Source: ORC/Macro, 2004. MEASURE DHS+ STATcompiler. http://www.measuredhs.com. Most recent survey. Accessed May 2004. Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals 7 food production. As the first MDG is the one in reducing poverty. A recent cross-country analysis which the impact of a more dynamic agricultural by Thirtle and others (2002) found that, at the sector will be felt most directly, we will discuss national level, a 1 percent increase in agricultural in more detail these two issues--poverty reduc- yields decreases the percentage of population living tion through agriculture-led economic growth and on less the $1 per day by 0.64 to 0.91 percent, with improved nutrition through agriculture. a slightly higher reduction for the countries in Africa. Notably, in analyzing the effect of growth in Poverty Reduction Through Agriculture-Led the manufacturing and service sectors on poverty, Economic Growth no significant change in poverty was associated with growth in these sectors. Income redistribution and economic growth are the This description of the process is necessarily two economic mechanisms for reducing poverty. simplistic and ignores a broad range of potential While redistribution is a viable option to address impediments to agriculture-led economic growth. poverty in many parts of the developed world, in One of the most critical of these at present is low the developing countries of the world with large prices for staple food crops globally, due to the suc- segments of populations unable to meet their basic cess of the Green Revolution (as well as high levels needs, income redistribution policies are unlikely of subsidies in the developed world). As a con- to have much effect on general welfare levels. There sequence, it now is difficult to generate profits are simply insufficient resources in such economies through staple food crop production even at high to assure the basic needs of all. In these countries, levels of productivity, particularly for smallholders higher economic productivity--that is, economic who are unable to achieve economies of scale in growth enabling increased employment and rising their production. Increasingly, in order to obtain wages--is the only means by which the poor will an adequate return from their efforts in agriculture, be able to satisfy their needs sustainably. farmers need to diversify their production into For both long-developed countries and for the high-value, but knowledge-demanding, specialized handful of more recently developed countries, sig- crops. Similarly, agriculture cannot be expected to nificant increases in agricultural productivity were be an engine of economic growth for those coun- a critical early step in building sustained economic tries that have no comparative advantage for agri- growth. Initial growth in staple food production by the small-scale, labor-intensive agricultural sector cultural production or face significant barriers to through the use of improved technologies resulted producing for global markets. Many of the drier in reduced food prices, increased real wages, and, inland countries of Africa face important challenges consequently, lower poverty. Reduced food prices in this regard. enabled greater access to food, resulting in better The initial distribution of agricultural assets in nutrition for the general work force while also free- the economy is also a critical feature in whether or ing up additional household resources from food not agriculture-led growth will reduce poverty. to other expenditures, including productive invest- Where land ownership is concentrated, as most ments. In rural areas, investments initially went notably in parts of Latin America, such economic into cash crop production and agricultural pro- development is unlikely to reduce poverty greatly cessing activities, but, as the economy grew, rural (Timmer 2003). In contrast, in most of Africa and nonfarm and urban activities became increasingly many parts of Asia where the poor continue to have profitable. As this process of economic transfor- access to land, agriculture-led growth should lead mation advanced, the agricultural sector tended to to significant reductions in poverty. Finally, where play a decreasing role in sustaining economic significant movement of the population out of rural growth.3 Movement of labor out of the agricul- areas to the cities and out of agriculture into urban- tural sector occurs as employment opportunities in based economic sectors has already occurred, as in other economic sectors grow. Depending on rural many of the middle-income countries of the devel- population growth, larger-scale commercial enter- oping world, there is little potential for the agricul- prises may become more characteristic of the agri- ture sector to catalyze broad economic growth and cultural sector. reduce poverty. In these countries, poverty reduc- There is considerable empirical evidence that tion efforts should focus on the industrial and ser- demonstrates the significant contribution that vice sectors, while not ignoring the key role that growth in agricultural productivity can make to agriculture plays in their rural economies. 8 Agriculture and Achieving the Millennium Development Goals While these caveats are important to recognize out significantly increasing the productivity of the in tailoring economic development strategies, nev- agricultural sector. However, agricultural produc- ertheless, such a process of agricultural growth has tivity gains alone are not sufficient to bring about proven in the past to be a common means by which sustained economic growth. No country has been to spur broad-based pro-poor economic growth. able to sharply reduce poverty only through agri- Yet, for the poorest countries, the process of small- cultural strategies. Institution building in the agri- scale agriculture-led economic growth leading to cultural sector and parallel developments in other strong economies and minimal poverty will not sectors of the economy are needed to transform the occur by 2015, the target date for most of the MDG foundational contributions from the agricultural targets. For these countries, the relevant time frame sector into sustained broad economic growth in the is one of decades and generations. However, a economy. Agricultural strategies alone will not lead process of building sustained economic growth to success. However, the converse also applies: For requires that productivity increases in agriculture the poorest countries, economic growth and sus- be achieved in the early stages. Moreover, the ben- tained poverty reduction are unlikely to be achieved efits of the initial steps of agricultural productiv- without initially stimulating sustained agricultural ity increases accrue primarily to the farmers and production growth. herders, among whom the poor are concentrated. By pursuing such an economic growth strategy Agriculture's Contribution to Food within the context of the MDGs, we can achieve and Nutrition Security both the significant improvements in well-being that the MDGs seek to promote, and lay the founda- The indicators for the second target of the first MDG tion for the sustainable economic transformations include the reduction by half between 1990 and needed to attain, in the longer term, the broader aim 2015 of the prevalence of underweight (low-weight- of the Millennium Declaration of "freeing the entire for-age) children and the proportion of the popula- human race from want." tion whose food intake falls below the minimum Only a few countries have dramatically reduced level of dietary energy requirements (undernutri- poverty and achieved rapid economic growth with- tion). As shown in figure 2.5, the levels of under- Figure 2.5 Prevalence of Undernourishment--Proportion of the Population Unable to Acquire Sufficient Calories to Meet their Daily Caloric Requirements, 2003 Estimates less than 20 percent 20 ­ 30 30 ­ 40 40 ­ 50 50 ­ 60 more than 60 percent no data Source: FAO (2003). Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals 9 nutrition are high throughout the developing world, rity to attain nutrition security. To reduce malnu- and particularly in many of the countries in which trition in a comprehensive manner, agricultural agriculture is also the dominant livelihood. To attain strategies must be implemented as part of a broader the MDG 1 second-target goals, food and nutrition set of actions that involve the health, water and san- security needs to be enhanced for the poor. A house- itation, and education sectors (see also chapter 5). hold is food secure if it can reliably gain access Of the eight MDG goals, the first is the one to food in sufficient quantity and quality for all whose attainment most clearly involves the agri- household members to enjoy a healthy and active cultural sector: The poor around the globe are life. It is possible, however, for individuals in food- disproportionately farmers and herders, and, per- secure households to have deficient or unbalanced versely, the hungry and undernourished also most diets. Nutrition security is achieved when secure commonly find their livelihoods through agri- access to sufficient, safe, and nutritious food is culture. The impact that a dynamic agricultural coupled with a sanitary environment, adequate sector will have on the attainment of the other health services, and knowledgeable care to ensure a seven goals is less direct. Nonetheless, important healthy and active life for all household members. gains in achieving these goals can be made through Clearly, food and nutrition security is closely explicit attention to agriculture. We now consider tied to agricultural productivity. Increased food these other MDGs. production increases local food availability. Higher production from one's own farm or herds increases one's access to food and enhances household food MDG 2--ACHIEVE UNIVERSAL security. The nutritional quality of the food pro- PRIMARY EDUCATION duced is also an important consideration in reduc- Primarily in an indirect fashion, investments in ing malnutrition, particularly for households who agriculture will advance progress toward attaining acquire most of their food from their own fields by 2015 the goal of enabling children everywhere, and herds. For food purchasers, higher production boys and girls alike, to complete primary school. generally means lower food prices and access to a Higher productivity in agriculture leading to higher greater quantity of food in the market for a given incomes will enable either the use of hired labor for income level. Particularly in South Asia and Africa, agricultural operations or the use of labor-saving the most potent force for reducing malnutrition is raising food availability through increased agri- technologies in place of the labor of school-age chil- cultural productivity, as well as trade (Smith and dren in farming households. However, a critical Haddad 2000, p. 84). component of this equation is the value that farm- Many countries, particularly in Eastern and ers perceive that they or their children will obtain Central Africa, are characterized by a declining or by sending their children to school. These benefits slowly growing food crop sector and very low pur- are to a large degree determined by the vibrancy chasing power. While stable access to food through of the economy and the extent to which the higher the market requires that the food marketing system economic capacities of trained individuals are re- is effective in supplying food while also benefiting warded. In stagnant economies and particularly in those who have food to sell, the systems in these rural areas where the range of employment oppor- countries are unable to provide effective markets. tunities is narrow, perceived returns of education People living on less than $1 per day are unable to commonly are judged to be significantly less than pay the prices necessary to import all of the staple the opportunity costs associated with keeping a food they require. child in school, and so unable to work full time in Consequently, if hunger is to be addressed effec- the family's agricultural enterprise. The broad- tively, a range of complementary actions are needed based agriculture-led growth discussed earlier is in addition to those aimed at enhancing crop and necessary to alter the outcome obtained by farming livestock production. An important component households making this comparison. of these actions is the agriculture-led economic Such a pattern of growth, particularly as it growth described previously, whereby real incomes extends to the nonfarm and urban sectors, will and access to food are increased. However, a host of demand increasingly skilled labor, and will increase other institutional factors must be addressed, as the returns of investment in the schooling of one's well as several cross-sectoral challenges. The latter children. Moreover, the relationship between in- are particularly the case in going beyond food secu- creased educational attainment and a more active 10 Agriculture and Achieving the Millennium Development Goals agricultural sector runs both ways. The agricultural This being the case, if the benefits of a dynamic agri- sector in most poor countries is unlikely to continue culture sector are to result in sustained improve- to expand for very long on the basis of productiv- ments in the direct determinants of welfare-- ity increases of staple food crops alone. Increas- income, health, education, among others--it is nec- ingly, the sector will have to turn to the production essary that women have an important role in of high-value cash crops that usually have quite determining how the fruits of their agricultural specific production and marketing requirements. activities are used. Meeting the requirements to engage profitably There is considerable empirical evidence of the in their production requires a better-trained work importance of improving the status of women for force. As this sector of the local agricultural econ- improved general welfare. For example, in a broad omy develops, the returns of providing basic edu- cross-country analysis, Smith and others (2003) cation to one's children become considerably more found that women's decision-making power rel- compelling. ative to men's was significantly associated with improved nutritional status of their children. They MDG 3--PROMOTE conclude that sustainably improving nutritional GENDER EQUALITY AND status requires proactive efforts to improve the sta- EMPOWER WOMEN tus of women, particularly in South Asia, but also in Sub-Saharan Africa. The authors suggest programs Throughout the developing world, women are that will enable women to gain access to new re- farmers and find their principal productive activ- sources and promote girls' education and health ities in agriculture. Considerable research shows care, subsidize childcare for working parents, and that when men and women are able to use agricul- improve the nutritional status of adolescent girls tural inputs at equal levels of intensity, women are and young women. equally effective as men in profitably engaging in Moreover, regarding agriculture in particular, agriculture, being responsive to changing market an important dimension in the empowerment of conditions in the suite of crops they produce, and rural women is alleviating the labor burdens they effectively utilizing new technologies (Quisumbing experience so that they can adequately provide 2003). Agriculture provides key contributions to the for their children's needs. Domestic time demands economic empowerment of women. upon women are greater than for men. Agricultural Moreover, the relationship between agriculture technology developed with close attention to alle- and the empowerment of women works both ways. viating some of the labor constraints experienced A dynamic agricultural sector that offers broad wel- by rural women has the potential to improve not fare benefits can be expected to emerge only when only the well-being of the woman farmer, but also women are given the opportunity to participate of others in her household who are dependent profitably in the sector. Where they have security upon her care. in their access to productive resources and control of their agricultural production--that is, where MDG 4--REDUCE women farmers are empowered to achieve their CHILD MORTALITY full economic potential within agriculture--the wel- fare effects of a productive agricultural sector can The linkages between agriculture and child mortal- exceed the simple economic productivity measures ity are indirect but strong. Agriculture is a critical for the sector. The economic empowerment of component in assuring food and nutrition security. women both in agricultural production and in other As was described earlier, levels of child malnutri- economic spheres can be expected to advance sig- tion are significantly higher in rural areas than in nificantly efforts to attain several of the MDGs. urban areas. It is estimated that 45­55 percent of all Several of the MDGs are directly determined by child deaths are due to malnutrition exacerbating the extent to which sufficient resources are pro- the negative effects of disease on a child's health vided to children as they develop and grow. The (Pelletier and others 1994). As a consequence of immediate provision of these resources--health poor nutrition security in rural areas, mortality care, feeding, life skills training, and so on--is inher- rates for children under five years of age are signif- ently a gendered task. In most societies, women icantly higher among rural children than for their are the principal caregivers within the household. urban counterparts (figure 2.6). Poor nutrition secu- Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals 11 Figure 2.6 Child Mortality by Urban or Rural Residence, Surveys Since 1999 Mali Guinea Ethiopia Malawi Tanzania Rwanda Zambia Benin Burkina Faso Nigeria Uganda Gabon Zimbabwe Egypt Bangladesh Cambodia Rural Urban Turkmenistan Nepal Kazakhstan Armenia Haiti Dominican Republic Peru Colombia 0 50 100 150 200 250 Deaths of children under five years of age per 1,000 live births Source: ORC/Macro, 2004. MEASURE DHS+ STATcompiler. http: www.measuredhs.com. Most recent survey. Accessed May 2004. rity is poor food security coupled with poor access However, poverty does not fully account for to quality health services, lower general knowledge child mortality. Care is also a critical element in of proper feeding and management of childhood reducing child deaths. As women are the primary illnesses, poor sanitation, and unprotected water caregivers, MDG 3 is relevant here. Time and sources. knowledge are both critical constraints in this A productive rural economy offering sufficient regard. Women need to have the time and the employment and rising wages is a necessary com- knowledge to appropriately meet the survival ponent in any effort to reliably and durably reduce needs of their children. Education, both the formal the number of children dying in a rural community. education sought through MDG 2 and informal Consumption poverty is an important part of the education from peers and public health services, explanation for why child mortality rates are so high must be provided to a child's parents so parents in many parts of the developing world. If efforts to can take appropriate action to assure the child's attain MDG 1 by following a broad-based, agricul- potential to live a healthy and active life. ture-led strategy of economic growth bear fruit, we The economy-wide effects of a dynamic agricul- should see a parallel decline in child mortality rates. tural sector can help reduce child mortality, because 12 Agriculture and Achieving the Millennium Development Goals more funding will be available for the public provi- broad nutrition security. Yet, agriculture has consid- sion of medical care, health-care facilities, sanitation erable potential to directly improve maternal health and clean water, and public health interventions. by improving the diets of both rural and urban While one should not expect that agriculture-led women, as well as the other household members economic growth necessarily will quickly provide for whom they are responsible. Micronutrient defi- economic surpluses that can be invested in this ciencies are particularly severe among young chil- way, making the investments needed to reduce the dren and women. Among women, the health effects number of children who suffer and die must be a of such deficiencies, particularly of dietary iron, are priority as surpluses are generated in the economy. most pronounced during pregnancy, at birth, and in the months following birth. As shown in figure 2.7, MDG 5--IMPROVE the spatial pattern of the prevalence of iron defi- MATERNAL HEALTH ciency anemia at the national level parallels similar patterns of poverty and malnutrition. It is estimated Agriculture can contribute to the goal of reducing that more than 65,000 women die annually due to maternal mortality in a way similar to its contribu- severe anemia. tion to attaining the previous three goals. Insofar as By increasing the micronutrient content of food agriculture can contribute to the economic empow- crops, deficiencies among women and children erment of women and enable them to participate should decline, and maternal and child mortality better in decision making in their households and rates drop. With this goal in mind, a global, inter- communities, women will have greater ability to disciplinary research program within the Consult- pay attention to their own physical well-being and ative Group for International Agricultural Research have access to increased resources to assure their seeks to increase the nutrient density of many global own good health. staple food crops, particularly by increasing the Agriculture can contribute to improved maternal levels of bio-available iron, zinc, and vitamin A that health in another way. For the most part, the qual- the crops contain. Although this research effort is ity of the food produced by farmers is assumed to just beginning, if such traits can be bred into high- be irrelevant to how agricultural activities affect yielding varieties that enjoy wide consumer accept- Figure 2.7 Estimated Prevalence of Iron Deficiency Anemia in Women Aged 15 to 49, Most Recent Estimates less than 20 percent 20 ­ 30 30 ­ 40 40 ­ 50 50 ­ 60 more than 60 percent no data Source: Micronutrient Initiative & UNICEF (2004). Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals 13 ability, farmers would be able to offer a direct, appropriately to maintain sufficient levels of pro- low-cost, and sustainable way to improve the nutri- ductivity in their agricultural activities will be better tional status of millions around the world and, in able to cope with the burdens of caring for HIV- particular, improve maternal health. positive members. Agricultural support service staff, working in close coordination with staff from MDG 6--COMBAT HIV/AIDS, the health and education sectors, have a role to play in stopping the transmission of HIV in farming MALARIA, AND OTHER DISEASES communities. However, the challenges in accom- Efforts made to reduce child and maternal mortal- plishing this are immense. ity levels to which agriculture contributes will also Moreover, the agricultural economy as a whole contribute to effectively combating HIV/AIDS, suffers from a heightened incidence of HIV infec- malaria, and other diseases. Although the indica- tion. Declining aggregate agricultural productivity tors for this goal deal very little with agriculture, will result from insufficient labor as farmers and and the direct links between agriculture and these herders fall ill or must devote their time to house- health issues are not immediate, they are impor- hold members who are ill. Savings will be depleted tant nonetheless. For agricultural households, as to meet increased health care costs and, conse- for all households, the productivity of their eco- quently, needed inputs for profitable agricultural nomic activities is an important determinant of production will be inaccessible to many farming whether people live in an environment that allows households. Any economic gains that might have them to enjoy a healthy and active life and can been made by the sector will be negated. In a rural acquire the health care required to do so. population ravaged by HIV, there is little scope for However, a newly dynamic agricultural sector agriculture providing the lead in building strong also has the potential to radically alter the disease economic growth. Although agricultural practices environment in a region. For example, wage-labor can be modified and support services provided to migration associated with agriculture may expose increase the resilience of farming households suf- populations to new diseases, increasing the health fering from HIV/AIDS and strengthen households burdens they bear. Changes in local water manage- threatened by the disease, the scope of the epidemic ment for irrigation may alter local disease ecologies, is such that it poses a major threat to the already particularly for malaria and water-borne diseases. tenuous welfare of the poorest, most agriculturally New health challenges likely will emerge with an dependent populations. evolving agricultural sector. The resources of the agriculture sector, particularly in the public sphere through extension services, can be used in a coor- MDG 7--ENSURE dinated fashion with those of the health sector to ENVIRONMENTAL address such issues. SUSTAINABILITY The agricultural sector has an important role in addressing the important challenge to human dev- This goal covers a broad sweep--biodiversity, crit- elopment of HIV. Of the 25 countries in the world ical natural habitats, energy use and global climate with an adult HIV infection level above 5 percent change, unsafe water and poor sanitation, and in 2001, all except two have predominantly rural urban slums. Agriculture is implicated both as a populations. The long and fatal pathway of chronic means to effectively address many of these prob- illness with HIV infection, particularly in young lems, and as a source of and a contributory factor adults, severely compromises the welfare of farm- to the problems that MDG 7 was formulated to ing households. Loss of labor power and farming address. It is unlikely that one can develop the agri- knowledge, and increasing nutritional require- cultural sector in such a manner that only benefits ments as the disease progresses, are among the and no negative externalities accrue. In this light, a most salient effects within the household (AIARD judicious, comprehensive, and participatory assess- 2003). Professional agriculturalists--researchers ment of the environmental costs and benefits must and extension workers in particular--must tailor be undertaken in the planning process for any agri- their work to assist such households to better cultural development efforts. meet the particular farming challenges they face. A productive agricultural sector will reduce pres- Households that are able to modify their practices sure on and contribute to ensuring environmental 14 Agriculture and Achieving the Millennium Development Goals sustainability in most of the areas considered by urban agriculture can contribute to the in- MDG 7. In particular: come growth and nutritional status of slum dwellers. · Productive agriculture requires less land per unit yield, leaving marginal agricultural lands However, agriculture can also exacerbate envi- to other uses, including forests and other crit- ronmental degradation, fuel perverse rural-to-urban ical habitats. migration, and deepen poverty. For example: · Proper agricultural policies will allow the full · Agricultural expansion is a principal factor costs of agricultural technologies, including contributing to tropical deforestation and their costs for the environment, to be con- increasing global levels of atmospheric carbon sidered as they are being used. Policies that dioxide. The underlying causes for agricul- induce transparent assessment of these costs tural expansion are case dependent. Some- will, for example, reduce the scope for exces- times expansion arises from accelerated sive nutrient runoff from agriculture, provide growth in the agricultural sector (capital- incentives for efficient energy and water use in driven), and at other times from poverty as the sector, and enable the ecologically sustain- poor farmers seek land with which to meet able use of a range of technologies, including their basic needs (Geist and Lambin 2002). pesticides and genetically modified organisms. Agriculture-led economic growth in itself is · As agriculture is inherently an organic, carbon- not environmentally benign. based enterprise, the sector is a potentially · A stagnant agricultural sector with low pro- important component in any systems estab- ductivity and profitability will result in the lished to manage global carbon stocks. unsustainable use of natural resources for agri- · A dynamic agricultural sector fostering broad- culture. Most commonly this is seen in the based economic growth, as with any economic mining of soil fertility down to a base state in expansion, should provide additional public which cereal yields, even under suitable agro- revenue to enable greater levels of public pro- climatological conditions and adequate labor vision of safe drinking water and improved inputs, attain only a few hundred kilograms sanitation. Increasingly productive and prof- per hectare. Such environmental exploitation itable farmers in many developing countries in agriculture can take many other forms, will be able more and more to provide these affecting a broad range of the components of amenities privately for their own households. local ecosystems. Moreover, water infrastructure in agricul- · Poverty can actually increase due to the ture, whether for irrigation or flood control, expansion of the agricultural sector of the will also have important applications for economy if the distribution of agricultural the provision of safe water and adequate assets--land, in particular--is skewed. If agri- sanitation, particularly in small-scale agricul- cultural expansion is accomplished in a man- tural systems. ner in which capital substitutes for labor and · Population pressures in urban slums will be little growth in employment occurs, aggre- alleviated to a significant degree if profitable gate welfare will be little improved. More- agricultural systems are developed in the rural over, further consolidation of land may occur, hinterlands. Although broad agriculture-led resulting in greater pressures on the rural poor economic growth should lead in time to a sig- to migrate to urban slums. nificant movement of workers out of rural, agricultural occupations and into the manufac- With appropriate regulatory institutions in place turing and service sectors located predomi- to safeguard the benefits that society as a whole nantly in urban centers, such migration will be draws from the environment, an emergent agri- of a different quality than that most commonly cultural sector need not lead to environmental seen at present. Today what is frequently degradation. The underlying driving force for observed is an unproductive agricultural sec- environmental degradation through agricultural tor that forces many farmers and herders to expansion and the harmful use of farming tech- search for employment in urban centers that nologies is frequently poverty, rather than factors are only marginally more attractive in terms inherent to agriculture itself. If farmers realize per- of economic opportunity. Moreover, enhanced sonal economic benefits and also recognize the Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals 15 social benefits from environmental protection, they oped provide due and relatively detailed attention will respond to these incentives and employ envi- to the economic foundation for most of the poor ronmentally sustainable production techniques. people at whom they are targeted--agriculture. However, sustainable agricultural practices must Most current PRSPs highlight the need for broad- be profitable for this to happen. Whether or not this based economic growth, and such growth is typi- is the case depends on the vibrancy of agriculture cally one of the four or five "pillars" of most PRSPs. within the overall economy. Moreover, agriculture is frequently noted as being the most important livelihood for the poor and is prioritized as a key economic sector. MDG 8--DEVELOP A GLOBAL However, the means by which agriculture will PARTNERSHIP FOR DEVELOPMENT lead to broad-based economic growth is frequently left unspecified in the documents. Indeed, a key Although extremely broad in scope, the final MDG criticism of the implementation of the PRSPs has has important implications for agriculture. There been that social expenditures tend to be given are several areas where agriculture can contribute priority over the investments, particularly in the to efforts aimed at achieving the targets set. agricultural sector, that would accelerate economic First, it is under this goal that an open, rule- growth (Gautam 2003). Given the MDG's strong based, predictable, nondiscriminatory trading and social focus, they perversely provide additional jus- financial system is called for. Given the demonstra- tification for countries privileging social expendi- ble centrality of the agricultural sector to people tures to the neglect of investments that would lead whose conditions of life must change if the MDGs to sustainable pro-poor economic growth. A more are to be attained, agriculture should be among the balanced expenditure pattern is needed. In most focus sectors of initial efforts in this area. Global HIPC countries that have prepared PRSPs, con- agricultural trade must be harmonized and ratio- siderably more effort must go into framing the nalized in a manner that includes consideration of mechanisms by which agriculture will bring about the special needs of poor agricultural producers the desired improvements in welfare. For exam- and how they might derive maximum benefit from ple, Uganda has developed the relatively detailed such trade. Plan for Moderniation of Agriculture (MAAIF & However, even with attention to these important MFPED 2000) as the principal cross-sectoral eco- aspects, the poverty impact of globalized agricul- nomic development strategy emerging from the tural trade remains unclear. Such trade will often Poverty Eradication Action Plan (MFPED 2000), exclude the smallholder farmer and herder, as it Uganda's Poverty Reduction Strategy Paper. requires knowledge, capital, and quality assurance Many more HIPC countries need to do likewise. levels, as well as access to marketing networks that Measures are currently being undertaken to re- most smallholder producers cannot attain on their orient the mechanisms for the delivery of bilateral own. Smallholders may be uncompetitive and un- and multilateral official development assistance to able to participate in many of the most profita- be consistent with efforts to achieve the MDGs. ble subsectors under a wholly free-trade system. MDG 8 specifies several related to official develop- Establishing appropriate institutions is necessary to ment assistance. The sectoral allocation of such enable broad welfare gains to be achieved through assistance is not specified in these targets. How- trade. These issues will be addressed further in ever, clearly this is not a tangential issue. If agri- chapter 4's discussion of trade and macroeconomic culture is to be effective in broadly improving policies. the human condition, particularly that of rural res- Second, the Heavily Indebted Poor Countries idents, considerably greater levels of resources (HIPC) initiative contributes to the targets under need to be made available to agricultural develop- this goal. Poverty Reduction Strategy Papers (PRSP) ment. Currently, levels of assistance to agriculture, have been prepared by many of the HIPC countries as well as budgetary allocations by governments to demonstrate to their development partners how themselves, are inadequate. the funds made available through debt relief would For example, the Comprehensive Africa Agri- be used to reduce poverty. The PRSPs potentially culture Development Programme strategy of the are a very effective means by which progress can be New Partnership for Africa's Development made in achieving the MDGs. It is critical that the (NEPAD) proposes investments of $251 billion over framework within which these PRSPs are devel- the period 2002 to 2015, or just under $18 billion per 16 Agriculture and Achieving the Millennium Development Goals year, to reduce the incidence of hunger and raise earning their livelihoods--in agriculture. Donor farm output (NEPAD 2002). Such a budget faces priorities should reflect this basic element of the some stark constraints. Notably, estimated total global poverty profile. annual government expenditures on agriculture in Finally, the roots of deficient agricultural devel- Africa in the late 1990s were roughly $6.2 billion opment found in so many poor countries often lies (FAO 2001; World Bank 2003). Moreover, through in power relationships in which the welfare of the the 1990s the major bilateral and multilateral donors population is not served, resulting in poor gover- annually committed globally only about $8 bil- nance, political and social weakness, and adverse lion to agriculture, broadly defined (FAO/IFAD/ incentives. Although the agricultural sector is not WFP 2002). The agricultural development needs blameless, such problems do not lie fundamentally of Africa alone cannot be met under current levels within the agricultural sector, but are reflective of of official development assistance, and the human broader destructive processes within national polit- needs that could be met through agriculture extend ical economies. The incentives for bringing about much beyond Africa alone. If the poor of the world an active agricultural sector are not there because of are primarily farmers and herders and we want to these other problems. Little progress in attaining see sharp improvements in their well-being in the the Millennium Development Goals or in vitalizing near term by 2015, then important gains can be agriculture can be anticipated in countries that are made if we start with where they are currently unable to confront these issues. Table 2.1 Summary of Links Between the Agricultural Sector and the Millennium Development Goals, Principally at Household Level Complementary Goal Direct Indirect Nature of Relationship Requirements 1. Eradicate extreme · Increased food · For both farming · Two-way, quite · Suitable agricultural poverty and hunger. production and nonfarming strong, generally production tech- · increased food con- households, positive. nologies available. sumption for sub- increased income · Less hunger · Relatively equitable sistence farming · increased capital · more productive distribution of farm- households. investments in workers in agricul- land across the · More diverse food existing economic ture. population. production activities or diversi- · Less poverty · Efficient, widespread · higher-quality diets. fication into other · more investment in rural markets For farming sectors agriculture. that are linked to households: · enhanced welfare regional and · Increased and increasing international trade production household eco- circuits. · increased income nomic resiliency. · Knowledge on through markets proper diet and · increased consump- nutritional care. tion and household Sanitation and assets. health services For nonagricultural available. households: · Increased production · reduced prices for agricultural products · increased consump- tion or reduction in share of income spent on food. (continued ) Perspectives on the Role of Agriculture in Meeting the Millennium Development Goals 17 Table 2.1 Summary of Links Between the Agricultural Sector and the Millennium Development Goals, Principally at Household Level (Continued) Complementary Goal Direct Indirect Nature of Relationship Requirements 2. Achieve universal · Few. · More dynamic agri- · Two-way, princi- · Increased returns primary education. cultural sector will pally indirect. to skilled labor in change assessments · Possibly some nega- agriculture. of the economic tive ramifications if · Primary schools returns to educating increased returns with adequate one's children com- from agriculture can quality of instruc- pared to returns of be achieved using tion are accessible. keeping children child labor or out of school to higher skills are not work in house- required. hold agricultural enterprises. 3. Promote gender · Increasingly prof- · Broader economic · Two-way. · Security of female equality and itable agriculture improvements · Increased willing- access to agricul- empower women. · potential to eco- through dynamic ness of women to tural resources. nomically empower agriculture invest in agriculture · Secure female women farmers. · increased public · more dynamic agri- control over own expenditures on cultural sector. agricultural output. water and sanita- · Possibly negative tion, health, ramifications if energy sectors more dynamic agri- · reduced time bur- cultural sector den on women for · increased male domestic tasks. domination of agri- cultural activities. 4. Reduce child · Few. · More diverse food · Principally one-way. · Knowledge of mortality. production proper diet and · better nutrition nutritional care. · increased child · Accessible and survival. effective health · More dynamic agri- services. cultural sector · increased income · more resources available to manage childhood illnesses. 5. Improve maternal · More diverse food · Primarily through · Two-way, but not · Degree of control health. production same mechanisms strong. women have over · higher-quality diets as MDG 3 on · Improved maternal resources to assure · improved health. empowerment of health will result in their own health. women. more productive · Availability of agricultural labor, nutrient-dense both from women food crops. and from their children. (continued ) 18 Agriculture and Achieving the Millennium Development Goals Table 2.1 Summary of Links Between the Agricultural Sector and the Millennium Development Goals, Principally at Household Level (Continued) Complementary Goal Direct Indirect Nature of Relationship Requirements 6. Combat HIV/AIDS, · Greater and more · More dynamic agri- · Two-way, princi- · Effective health sys- malaria, and other diverse food cultural sector pally indirect. tem, both curative diseases. production · increased income · Reduced health and public health · higher-quality diets · more resources to burden enables services. · improved health. devote to health more productive · Effective interven- services. agriculture. tions to limit HIV · Possible negative infection. ramifications if agri- · Particularly for HIV cultural investments infection in sub- or labor migration sistence farming patterns exacerbate households, avail- or extend diseases. ability of nutritious food crops that are not labor-intensive. 7. Ensure environmental · Agriculture prac- · More productive · Two-way. Both · To minimize nega- sustainability. tices can be both agricultural direct and indirect. tive environmental direct causes of and technologies · Agricultural sector externalities of important immedi- · withdrawal of is as likely to have agricultural invest- ate solutions to agriculture from negative ramifica- ments, participatory environmental marginal, sensitive tions on the envi- planning processes degradation. environments. ronment as positive. required. · More profitable Unprofitable agri- · Relatively equitable agricultural sector cultural systems distribution of agri- · reduced migration tend to unsustain- cultural assets across to urban slums. ably mine environ- the population. mental resources. · Environmental · Declining environ- costs of agricultural mental resource production incorpo- base is an erosion rated into economic of the foundation assessments of for the agricultural production systems. economy. 8. Develop a global · Expanding global · More profitable · Two-way, but prin- · Sufficient knowl- partnership for agriculture trade agricultural sector cipally toward agri- edge, capital, and development. increases need for · expectation of culture. Primarily access to markets to formal trading part- better governance direct. enable agricultural nerships and rules. and provision of · Globalization is as producers to engage · Capital require- public goods by likely to have nega- in regional and ments for compre- governments to tive as positive ram- global trade. hensive agricultural sustain the benefits ifications on development from agriculture in agricultural produc- · Significant increases the long term. ers, particularly in development small-scale subsis- assistance offered tence farmers, in to the agriculture the short term. sector. 3 Alternative Scenarios to 2015 Prospect and Policies for Meeting the MDGs on Poverty If Sub-Saharan Africa COUNTRY-LEVEL ASSESSMENT OF SELECTED MDGS: ETHIOPIA AND ZAMBIA follows current trends, Compared with other regions of the developing world, Sub-Saharan it will be the only Africa faces a huge challenge in terms of meeting the two targets constituting the first MDG--halve, between 1990 and 2015, the pro- developing region portion of people whose income is less than one dollar a day and halve, between 1990 and 2015, the proportion of people who suffer where the number of from hunger. poor people actually Indeed, as of 2001, approximately 46.5 percent of the inhabitants of Sub-Saharan Africa, or 314 million people, were living below the increases from the international poverty line (World Bank 2004b). If the region follows current trends, 39.3 percent of the population will remain below this 1990 level. line by 2015, and Sub-Saharan Africa will be the only developing region where the number of poor people actually increases from the 1990 level (AfDB/OECD 2002; UNDP/UNICEF 2002). Poverty both contributes to and is reinforced by hunger and malnutrition. During the last decade, the percentage of under- nourished people in Sub-Saharan Africa decreased only margin- ally, from 35 to 33 percent, while the number of undernourished people increased from 166 million to 198 million. At the same time, the average per-capita energy intake only increased by 100 calo- ries and actually fell in some countries. Children are particularly vulnerable to the effects of hunger. During the 1990s, the propor- tion of children under five years of age who were underweight remained relatively stagnant, averaging about one-third, while the number of underweight children actually increased by 8 mil- lion (UNDP/UNICEF 2002). The extent of child stunting (propor- tion of children with height for age under 2 standard deviations from the reference population median) is even higher, affecting approximately 41 percent of preschool children in Sub-Saharan Africa (UNICEF 2004). Agricultural growth is often identified as a means for ameliorat- ing Sub-Saharan Africa's hunger and poverty. Indeed, agriculture is the primary livelihood of approximately 65 percent of people in the region, represents between 30 and 40 percent of the region's GDP, and accounts for almost 60 percent of its income from exports (IFAD 2003). Increased growth of the agricultural sector offers direct bene- fits to poor farmers, such as income and food, contributes to broader food security objectives, and helps establish forward linkages with higher value-added industries. 19 20 Agriculture and Achieving the Millennium Development Goals Nevertheless, the ability of the agricultural of the decline occurred in the early period and the sector to contribute significantly to the MDGs in poverty rate only changed 1 percent between 1996 Sub-Saharan Africa will vary according to the con- and 2000. Table 3.2 shows the U.S. dollar conversion straints and opportunities prevailing within each of the national poverty lines on which these poverty country. This chapter focuses on the level and type rates are based and highlights that Ethiopia's poor of agricultural growth required in two countries live on less than $85 a year. of the region that are currently far from achieving Economic and demographic indicators, along the MDGs: Ethiopia and Zambia. As shown in with broad growth trends averaged over the period table 3.1, the percentage of the population living from 1991 to 2001, further elucidate the similarities below the international poverty line of $1 a day is and differences between the two countries. In U.S. higher in both countries than for Sub-Saharan dollar terms, Zambia's per-capita income is almost Africa as a whole. triple that of Ethiopia and lies between the average Compared to Ghana and Uganda, progress at for Ghana and Uganda. This is predominantly due reducing poverty in Ethiopia and Zambia has been to the higher urbanization rates in Zambia com- relatively slow over the last decade. In Zambia, pared with Ethiopia and the higher incomes in these the most recent household survey (1998) shows a urban areas. However, once per-capita incomes are 6 percent increase in the national poverty rate since adjusted to purchasing power parity (PPP), both 1991. In Ethiopia, the poverty rate declined from Zambia and Ethiopia are far below the average 51 percent in 1993 to 44 percent in 2000, but most in the region (table 3.3). In Ethiopia, agricultural Table 3.1 Comparison of Poverty Across Selected African Countries Trends in national poverty rates Rural poverty Urban poverty Population under (survey year in parentheses) rate rate $1 US/day Ethiopia 51.1a 45.5 44.2b 45.4b 36.9b 81.9c (1992/93) (1995/96) (1999/00) (1999/00) (1999/00) (1999/00) Zambia 68.9d 79.4d 75.4d 85.6d 58.3d 63.7c (1991) (1996) (1998) (1998) (1998) (1998) Ghana 51.7e 39.5e 49.9e 18.6e 44.8e (1991/92) (1998/99) (1998/99) (1998/99) (1999) Uganda 56.0f 44.0f 35.0f 39.0f 10.0f n.a. (1992) (1997) (1999/00) (1999/00) (1999/00) Africa 49.0g (2000) Sources: aMOPED (1994); MOFED (2002); World Bank, World Development Indicators (2003); Thurlow and Wobst (2004); PRSP for b c d e Ghana (2003); PRSP for Uganda (2003); and UN Millennium Development Goals Database. f g n.a. = Not applicable. Table 3.2 Poverty Lines in the Country Case Studies Annual poverty line Country Local currency U.S. dollarsa Survey year Type of poverty line Ethiopiab Br 696 84.65 1999­00 absolute Zambiac K 555,432 298.28 1998 basic needs Sources: aConversions calculated from exchange rates given in the IMF's International Financial Statistics Database; MOFED (2002); Thurlow b c and Wobst (2004). Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 21 Table 3.3 Economic and Demographic Indicators, 2001 GDP per GDP PPP per Agricultural GDP Total population Rural population Agricultural share capita (US$) capita (US$) per capita (US$) (millions) (millions) of GDP (%) Ethiopia 121 810 63.21 65.8 55.4 52 Zambia 405 780 94.71 10.3 6.2 22 Ghana 421 2250 253.41 19.7 12.5 35 Uganda 355 1490 167.79 22.8 19.5 37 Africa 567 1826 166.21 673.9 403.7 33a Source: World Bank (2003). Notes: aExcludes South Africa. growth is lower than the country's population has been hindered by a number of factors, including growth, which is worrisome considering that agri- protracted conflict with neighboring Eritrea, declin- culture constitutes 50 percent of Ethiopia's GDP ing world prices for its coffee exports, and the coun- and provides the livelihood for approximately try's vulnerability to natural disasters, particularly 85 percent of its population. By contrast, agricul- droughts (AfDB/OECD 2002). Due to the preva- ture provides only around 22 percent of GDP in lence of droughts, Ethiopia remains Africa's largest Zambia, much lower than the African average. recipient of cereal food aid, which equaled approxi- This is because the now faltering copper industry mately 1.2 million tons in 2002. The effects of food has traditionally dominated the economy. Although insecurity are particularly acute for children, as agricultural growth has, on average, outstripped Ethiopia has Africa's highest rate of children who Zambia's population growth, total economic are underweight and the second highest rate of growth has not (table 3.4). child stunting after Burundi. The unique challenges facing Ethiopia and In Zambia, a series of structural adjustment pro- Zambia partially explain these trends. Since the grams have been adopted that emphasize, among early 1990s, both countries have been engaged in a other things, privatization of the copper mines, process of economic liberalization. In Ethiopia, the trade liberalization, and agricultural reforms. How- government has been trying to transform the coun- ever, because of the inadequate design and piece- try from a centrally planned to a market economy meal implementation of many of these reforms, and remains committed to maintaining macroeco- coupled with a sizable external debt burden, nomic stability. Nonetheless, the country's growth growth has not improved. In fact, the bias against Table 3.4 Average Trends Across Countries, 1991­2001 Averages Average annual growth rate (%) GDP Agricultural GDP Population Ethiopia 5.28 2.31 2.47 Zambia 1.18 5.27 2.45 Ghana 4.27 3.36 2.18 Uganda 6.55 3.82 3.00 Africa 3.00 3.00 2.00 Source: World Bank (2003). 22 Agriculture and Achieving the Millennium Development Goals agriculture created by the mining sector is still the construction of more roads, will be essential for prevalent, and insufficient input and output growth in these areas to materialize. markets have circumscribed farmers' ability to improve production. Like Ethiopia, agricultural pro- Ethiopia: Agriculture-Led Growth duction in Zambia is also hampered by insufficient and Poverty Reduction and unevenly distributed rainfall. Similar to many of its Southern African neighbors, the HIV/AIDS With a per-capita income that is only one-fifth of pandemic, which affects approximately 23 percent the African average, Ethiopia classifies as one of of the population, has severely diminished labor the world's poorest countries. In addition to daunt- productivity and placed even greater pressure on ing poverty, persistent food crises have left a large food security. A combination of all of these fac- portion of the population suffering from food in- tors has contributed to a deterioration in per-capita security. Despite significant amounts of food-aid cereal production (table 3.5), culminating in 2002­03 assistance over recent years, there has been little in a major food crisis that caused an estimated progress at reducing this food insecurity. 2.3 million people to require humanitarian assis- Ethiopia's current circumstances reflect the tance (Samatebele 2003). cumulative challenges faced by the country over Clearly, addressing the problems of hunger and the past decades. In particular, the country is ex- poverty deserves priority in both countries. Results tremely vulnerable to drought and since the early from the scenarios of economy-wide models, which 1980s, has experienced seven major droughts, five are discussed in detail reveal that neither country of which resulted in famines. The most recent will be able to meet the MDGs if they continue drought, which occurred in 2002­03, affected ap- along their current growth trajectories. Improved proximately 30 million people (EM-DAT 2004). performance in the agricultural sector will be cru- In addition to climatic factors, the country has cial for substantial hunger and poverty reduction suffered under the misguided economic policies of to occur, and fortunately each country's govern- the socialist Dergue regime, which ruled from 1974 ment is promoting development strategies targeted to 1991. When the Ethiopia Peoples' Revolutionary at enhancing the sector's growth. As the scenar- Democratic Front (EPRDF) replaced the Dergue in ios highlight, Ethiopia has the potential to meet 1991, a number of market-oriented reforms were the MDGs through combined growth in the sta- implemented, including those aimed at stimulating ples, livestock, and nontraditional export sectors agricultural and rural growth (World Bank 2004d). if there are concurrent improvements in the trans- For example, the country liberalized its foreign port sector and further market development. exchange markets and dramatically decentralized Halving poverty in Zambia by 2015 will be more the public administration to the woreda (district) challenging to achieve. Nonetheless, agricultural level. In rural areas, grain markets were liberalized growth is a necessary condition for poverty reduc- and fertilizer markets were opened up to participa- tion in rural areas. As in Ethiopia, attention to mar- tion from the private sector. In 1992, the EPRDF ket access issues in Zambia, particularly through also established the Agricultural Development Table 3.5 Cereal Production in Selected African Countries Production (kg per capita per year) 1991 1994 1997 2000 2001 Ethiopia 99.55 86.69 143.83 112.94 132.02 Zambia 138.48 123.30 109.24 94.38 65.72 Ghana 83.29 85.33 83.90 80.38 74.88 Uganda 84.61 94.26 72.19 86.27 91.92 Africa 130.65 135.17 127.11 120.94 118.50 Source: Calculated from FAOSTAT (FAO 1998) and World Development Indicators data (World Bank 2003). Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 23 Led Industrialization Strategy, which emphasizes superior role in reducing poverty. The study shows the role of the agricultural sector as a catalyst for that increasing national staple food availability by immediate improvements in food security and for 50 percent by 2015 will significantly help poverty long-term growth in the broader economy. reduction. The feasibility of achieving this goal However, the outbreak of conflict with Eritrea relies on reducing the productivity gap between between 1998 and 2000 created a humanitarian traditional and modern technologies that have emergency in the northern part of the country and been adopted in the country. Achieving sustain- reduced the availability of resources to finance able agricultural growth in Ethiopia also requires many of these reforms. Not only did donors and supporting investments in roads and other market investors reduce their support to the country, conditions. but also increases in official defense expenditures In addition, the study emphasizes the need for decreased the availability of funds for other sec- regionally differentiated strategies, given the coun- tors as well as for antipoverty programs (World try's size as well as heterogeneous conditions in Bank 2004d). both natural resource and economic environments. With the return to peace, the government has Indeed, more than 50 percent of the poor live in the reaffirmed its commitment to generating growth food deficit area, where the availability of food sta- and reducing poverty, especially through a strong ples per household is half the national average. The focus on the agricultural and rural sector. Since poverty and food security challenge is huge in more than 85 percent of the country's population such areas. On the other hand, more than 50 per- lives in rural areas where agriculture is the main cent of food staples are currently provided from economic activity and where the poverty ratio is sig- food surplus areas where food staples availability nificantly high, and since the nonagricultural sec- per household is already 70 percent higher than the tor is extremely small in Ethiopia, any strategy for national average. In these food surplus areas, there slashing poverty and hunger has to focus on gener- is a need for greater diversification in agricultural ating rapid growth in the agricultural sector. To this production. Consequently, market access and mar- end, the Ethiopian government has not only contin- ket development should be an integral part of a ued to support the Agricultural Development Led national agricultural development strategy. Industrialization Strategy but has also launched a series of development and poverty reduction pro- The current growth path results in more poverty grams, such as the Sustainable Development and Poverty Reduction Program in 2001 and the Food In order to first demonstrate the necessity for Security Program in 2004. Agricultural growth, food increased agricultural growth, the International security, and accelerated rural development are the Food Policy Research Institute's model simulates cornerstones for all these programs. the impact on poverty if Ethiopia continues along To identify which investments can have the its current growth trajectory. Between 1995 and largest impact on agricultural growth and in turn 2002, about 90 percent of increases in total crop pro- drive broader growth and poverty reduction, a duction and 70 percent of increases in cereal pro- greater understanding is needed of the linkages duction were due to area expansion. The annual between agriculture, growth, and poverty reduc- growth rate of cereal production was about 2 per- tion. To this end, researchers at the International cent, lower than the 2.5 percent population growth Food Policy Research Institute have developed a rate. The productivity growth rate (yield) for total spatially disaggregated, economy-wide model for crops and cereals was also low, about 0.2 and Ethiopia based on recent national household sur- 0.6 percent per year, respectively. If the crop area veys, agricultural sample surveys, global informa- expansion and growth in yields continue according tion system data, and other national and regional to their current trends for the next 12 years, together data. The model analyzes the growth and poverty with the growth trends in livestock production reduction linkages at both national and regional (4.2 percent per year) and non-agriculture (4.6 per- levels. The study shows that broad-based growth cent), a business-as-usual scenario shows that in the in agriculture is the key for Ethiopia's success in next 12 years the national economy (GDP) and the meeting the objective of halving poverty. Within agricultural sector grow at 3.2 and 2.5 percent the agricultural sector, growth in cereals and other annually, respectively. Since agriculture, especially staple crops should receive priority, given their cereal production, will grow either close to or more 24 Agriculture and Achieving the Millennium Development Goals slowly than the population growth, the national respectively, compared with 3.1 and 2.5 percent, poverty rate will keep at its high current level of respectively, in the baseline scenario. 45.7 percent by 2015. With a 2.5 percent population More importantly, this staple-crops scenario con- growth rate, the number of the poor living under tributes more toward poverty reduction than the the current poverty line will increase to 40 million other agricultural subsectors or the nonagricultural by 2015, equivalent to 10 million more than today's sector, even though these other sectors may have a number. Increases among the poor will mainly similar impact on the growth of the overall econ- come from the food deficit area, where the current omy (table 3.6). Growth in staple crops also bene- poverty ratio is already very high. The majority fits consumers, as staples are the most important of Ethiopians will still be struggling to meet their sources of food energy for poor rural and urban basic food needs, especially since average daily consumers alike. The national household survey per-capita caloric intake will be even lower than data indicates that the rural poor, whose income is its current level. already below the poverty line, spend about 70 per- cent of their total income on staple-crop food, which is 30 percent higher than the rural average. Growth in staples is the most important In the urban areas, households with incomes below for poverty reduction the poverty line spend almost 50 percent of their Cereals and other staple crops account for 65 per- incomes on staple-crop foods, which is 65 percent cent of agricultural value-added, and the majority higher than the urban average. Thus, raising pro- of small farmers are producers of staple crops. ductivity in staple crops will increase the food sup- Thus, this sector should have the potential to sub- ply, lower food prices, and help reduce the poverty stantially alleviate poverty. The model simula- rate in the urban and rural areas. tions show that if the average yield of staple crops However, it needs to be highlighted that in- grows by an additional 1.5 percent from its current creased staple production often exceeds farmers' rate of 0.6 percent to 2.1 percent annually, and is own consumption. Hence, expanding markets for combined with the 1.3 percent expansion in crop these commodities is a necessary condition for farm- area assumed in the business-as-usual scenario, ers to benefit from growth. Moreover, to improve cereal production can grow at 3.4 percent per year. crop yield, farmers need to increase the use of mod- By taking into account the demand-supply, agri- ern inputs, which are purchased from markets. cultural-nonagricultural, and cross-sectoral link- Development of input and output markets can ages in agriculture, such a growth rate in staple strongly support the growth of agricultural produc- crops (combined with the base-run growth trends tion. If marketed staple crops increase too rapidly in the other sectors) results in an annual GDP and in the absence of market development, prices for agricultural GDP growth rate of 3.9 and 3.5 percent, these crops can be significantly depressed, nega- Table 3.6 Growth and Poverty Reduction Under Agricultural Growth Options Annual growth rate Poverty rate (2015) Add. growth Agric. share Scenarios ratea (2002) GDP Agric. Nonagric. All Rural Urban Initial value 44.4 45.7 36.9 1. Current growth 3.1 2.5 3.7 45.7 48.0 33.0 Additional growth 2. Staple crops 1.5 65.0 3.9 3.5 4.3 36.7 37.7 31.0 3. Livestock 3.6 26.0 3.9 3.5 4.2 40.0 42.0 26.8 4. Nontraditional 8.7 4.8 3.6 3.4 3.8 40.2 41.4 33.7 5. Three sectors (2 to 4) 5.1 5.3 4.9 27.5 27.8 25.6 6. Market investment (2 to 4) 5.8 5.4 6.1 24.4 25.0 21.2 Source: Results from Ethiopian multimarket model. aAdditional annual percentage yield or productivity growth rate in the relevant agricultural sectors. Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 25 tively affecting farmers if declines in prices are nontraditional exportable products has been very larger than increases in marketed products. It is rapid in recent years, indicating that it is a booming important to realize such risks, and hence to grad- sector in the economy. Therefore, a much more ually rather than abruptly increase production in rapid growth rate of 13.3 percent is simulated for order to avoid overwhelming the market. the sector in the "nontraditional growth" scenario. After staple crops, the livestock sector is the To support such production growth, nontraditional country's second largest agricultural sector. The exports have to grow at 29 percent annually. This historical growth rate of 4.8 percent (1995­2002) in expansion of nontraditional exports can result in an the livestock sector is higher than the growth rate in overall economic growth rate of 3.6 percent and staple crops or in agriculture in total, implying that an agricultural growth rate of 3.4 percent, which is the sector has a strong growth potential. The "live- comparable to the projected annual growth rate stock growth" scenario simulates an additional under the "staple crops" scenario. 3.5 percent annual growth rate in the livestock sec- However, the nontraditional export sector's con- tor, which raises the annual growth rate in the live- tribution to poverty reduction is relatively small. In stock sector to a total of 8.3 percent. Holding the fact, the rural poverty rate falls to 41.1 percent, only growth rate in the other agricultural and nonagri- 4.6 percentage points below the current level. This cultural sectors the same as in the baseline scenario, is because nontraditional export growth is often this simulation shows that growth in the livestock concentrated around cities where there is greater sector has the most significant effect on overall eco- access to transportation and other market facilities. nomic growth, causing GDP and agricultural GDP Given the technical and financial constraints that to grow at 3.9 and 3.5 percent per year, respectively. they face, the majority of poor, rural farmers are However, an 8.3 percent annual growth rate in unable to adopt the technology required for produc- the livestock sector has a relatively smaller poverty ing nontraditional crops. alleviation effect than a 3.4 percent annual growth On the demand side, increased production of rate in the staple crops. Although the urban poor nontraditional goods provides few benefits to poor benefit more, the rural poverty rate falls to 42 per- consumers in both rural and urban areas, especially cent by 2015 in this scenario as opposed to the since such products are often intended for export 36.7 percent achieved in the "staples growth" markets. The greatest constraint to growth in such a scenario. sector is inadequate market access. Rapid growth While livestock is a relatively large agricultural in the nontraditional-export sector requires a timely sector in the economy, it accounts for a relatively expansion of markets for such commodities. Indeed, small share of income for poor farmers. On the achieving the projected annual growth rate of consumption side, poor consumers in both the 13.3 percent in the production of nontraditional rural and urban areas consume fewer livestock exports is unlikely without huge investments in products. Calculated from the household survey infrastructure and other market conditions. For this data, the rural households living under the poverty reason, this discussion regarding the sector's con- line spend less than 4 percent of their incomes on tribution to the overall economy and to poverty livestock and dairy products, which is 40 percent reduction may actually be too optimistic. lower than that for an average rural household. By focusing on individual agricultural sub- Thus, while increased livestock production de- sectors, the preceding analysis emphasizes that presses prices in domestic markets, farmers do not growth options among different agricultural sub- experience the same level of benefits that they do sectors have different effects on poverty reduc- in the staples scenario. tion. Obviously, growth in any single subsector The nontraditional-export sector, which includes alone cannot help the country meet the MDGs. exportable vegetables, fruits, other horticultural Admittedly, growth in staple crops is critical for products, chat, cotton, sugar, and sesame seed, cur- poverty reduction, but it needs to be supported by rently accounts for 5 percent of agricultural GDP. growth in other subsectors. Growth in staple crops An additional 8.7 percent annual growth rate in the and nontraditional exports can increase domestic sector's productivity is equivalent to the 1.5 per- demand for livestock products, which helps stabi- cent growth rate in staple crops or the 3.5 percent lize the livestock prices and raise livestock farmers' growth rate in livestock described in the two pre- incomes. Similarly, growth in the livestock sector ceding scenarios. Nonetheless, growth in some generates feed demand. Increased income from 26 Agriculture and Achieving the Millennium Development Goals growth in livestock and nontraditional exports can we analyzed above. Moreover, many researchers help stabilize the prices for food crops. Combining have shown strong diminishing returns in the large- the growth rate in the three major agricultural sub- scale irrigation investment, which implies that sectors results in a 5.3 percent annual growth caution is needed in promoting large irrigation rate for agriculture. Such a diversified agricultural investment projects (Fan and Hazell 2001). growth strategy will significantly reduce poverty, We simulate an increase in irrigation area accord- causing the poverty rate to fall by 17 percentage ing to the country's irrigation development pro- points to 27.5 percent by 2015. gram drawn from the country's Water Sector Development Plan (Ministry of Agriculture and Rural Development 2005). The model simulates an Assessing investment options increase in irrigated land according to the plans and Achieving the requisite growth rates examined in shows that even by doubling irrigated cereal areas these scenarios requires increased public invest- by 2015, irrigation will only account for 3 percent ments. Various public investments, such as agri- of total cereal production. As a result, the average cultural research and development, extension, irri- growth rate of cereal output will increase modestly, gation, and other infrastructure, can directly from 2.2 percent in the baseline scenario to 2.4 per- improve agricultural productivity, raise farmer cent in the irrigation scenario, which is equivalent incomes, and reduce poverty. Although all of these to about 0.2 percent of additional annual growth. Irrigated cash crop areas will triple by 2015, and investments are needed, each investment can have will account for 5 percent of total cash crop areas, a differential impact on agricultural growth and compared with the current 2 percent. This will poverty reduction. Given limited government bud- benefit exports. For example, horticultural exports gets and international donor funds, as well as will increase by four-fold by 2015. A similar effect the extremely broad areas in which public invest- is also observed for coffee exports. Because the ments are needed, priority matters in planning medium- and long-term projects are completed investment strategies. Accordingly, a broad analy- only toward the end of the period analyzed, the sis of investment strategies will be helpful in iden- potential returns from the newly irrigated areas tifying which kinds of investments can bring are not fully captured in the simulation. pro-poor agricultural growth to the country. Although irrigation has the potential to raise crop productivity, the ability of increased output, Irrigation. We examine irrigation investments first especially in the export sector, to reach international because increased irrigation is important for re- markets depends on simultaneously improving the ducing climate risk, which is one of the greatest underlying market conditions. The gains we ana- constraints to agricultural growth in Ethiopia. More- lyzed here should not be understood as solely from over, reducing climate risk helps induce the use investments in irrigation. Without investments in of modern inputs, such as fertilizers and improved markets, national prices for increased output might seeds, which can further increase agricultural pro- decline, which may reduce the realized gains from ductivity. Currently, there are about 200,000 hectares the investment in irrigation. of irrigated area in the country, accounting for With growth in both staple and cash crops due to slightly more than 2 percent of total crop areas. doubled irrigation area, together with cross-sectoral Among the 200,000 hectares of irrigated land, linkage effects, the annual growth rate for GDP and 60 percent is dedicated to cereal production. agricultural GDP rises to 3.6 and 3.0 percent, respec- According to the Agricultural Sample Survey data tively, from 3.1 and 2.5 percent in the baseline. The of 1997 and 2000, the yield gap between irrigation national poverty rate falls to 39.4 percent from its and rainfed crop production is 40 percent; that is, current 44.4 percent. While irrigation has a modest on average, irrigation can increase cereal yield up effect on national poverty reduction, its effect on the to 40 percent. Obviously, significantly increasing food staples deficit area is significant. Since pro- irrigation area can stimulate growth in cereal pro- jected increases in irrigated land are mainly located duction. However, as irrigation currently accounts in the food staples deficit area, the rural poverty for less than 2 percent of total cereal production rate in the area falls to 51.5 percent by 2015 from its and slightly more than 2 percent of other crop pro- current 58.4 percent. There is a much smaller effect duction, it is unrealistic to expect that investment on poverty reduction in the food surplus areas due in irrigation alone can generate the growth rate to a smaller increase in irrigated land in the area. Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 27 Adoption of improved seed and greater efficiency in crops. Although meeting the objective of halving fertilizer use. Compared with the international stan- poverty requires more than improving staple crops' dard, the current yield level of crop production, productivity, growth in the staple food sectors is especially for grains, is quite low in Ethiopia. Both obviously a necessary condition for any significant the low utilization and low efficiency of modern reduction in poverty. Exploiting the growth poten- inputs can partially explain such low yield levels. tial of staple crops from dissemination of modern However, there are many factors restraining the technology requires not only investment but also dissemination of modern inputs. Survey data changes in farm management and a transition from show that while fertilizer has been used on about current farming traditions to more modern farm- 40 percent of the grain area nationwide, only a ing systems. small portion of it is combined with the use of other inputs, especially with improved seed. As a result, Combined with infrastructure investments that reduce the average yield gap in grain production due to marketing costs, the agriculture-led growth strategy fertilizer use is quite small. Yet, there are significant has the potential to meet the MDG of halving poverty by gains from combining fertilizer use with improved 2015. An agriculture-led growth strategy does not seeds in grain production, especially in maize imply that investments should only be in agricul- production. ture. Many studies have shown that poor infrastruc- Thus, the model simulates a situation in which ture and dysfunctional markets prevent farmers the technology combining improved seeds with from accessing markets and hence diminish agricul- fertilizer is expanded to all of the cereal area that ture's profitability. It is important to remember that is currently fertilized. At the same time, we increase institutional barriers also constrain farmers from the efficiency of fertilizer use by 50 percent over becoming actively involved in market activities and 12 years. This technology results in an additional that market development does not solely imply annual growth rate of 0.9 percent for cereal pro- infrastructure investment. Nonetheless, in this sec- duction. As a result, the rural poverty rate falls to tion we focus specifically on investments in roads 40.3 percent by 2015, which is 4 percentage points and other infrastructure that could reduce the trans- lower than the current rate. portation costs of agricultural trade and improve market access for farmers. Increased irrigation combined with the adoption of The Ethiopian road density is 27 kilometers per modern seed and improved efficiency in fertilizer use. 1,000 square kilometers, which is only half of the Since the returns to technology adoption are low if average for Africa. Seventy percent of farmers are modern inputs are used in isolation and not sup- reported to be more than half a day's walk from an plemented by other technologies, modern technol- all-weather road (MOFED 2002). Such poor mar- ogy needs to be disseminated in a package. Thus, ket access conditions and high transportation costs we simulate a situation in which the adoption of significantly increase the price gap received by modern seed varieties is combined with improve- farmers and paid by consumers. The average grain ments in the efficiency of fertilizer use and com- price gap is estimated to be about 30 to 70 percent bined with the expansion of irrigated area. By across regions, and domestic marketing costs often simultaneously investing in the three areas, the account for more than 50 percent of fertilizer prices annual growth rate of cereal production rises to paid by farmers. These costs significantly reduce 3 percent, resulting in an annual growth rate of farmers' profitability from increased production. 3.8 and 3.4 percent for GDP and agricultural GDP, To address the constraints in the road sector, the respectively. Growth in the cereal sector, combined Government of Ethiopia formulated a two-phase with increased cash crop production from irriga- road sector development program (Ministry of tion projects, helps the poverty rate fall to 37 per- Agriculture and Rural Development 2005). Under cent, which is 7.4 percent lower than its current the first phase, which was completed in 2002, level and in line with the simulation result that the focus was on the rehabilitation of the core analyzed the "staple crop growth" scenario. networks. Substantial progress was achieved in The above analysis shows that through technol- reopening nearly all of the classified roads, accessi- ogy adoption and dissemination, combined with bility was improved, and the percentage of roads increases in irrigation area, it is possible to have a in good condition also increased. The second more than 3 percent annual growth rate in staple phase of the program involves the rehabilitation of 28 Agriculture and Achieving the Millennium Development Goals 1,168 kilometers, the upgrading of 2,045 kilome- poverty rate is significantly lowered, and the coun- ters, and the construction of 8,383 kilometers of try will be quite close to meeting the objective of roads. halving the poverty rate by 2015. In fact, the national Improving the road network also includes road poverty rate falls to 24.4 percent by 2015. maintenance. For example, the removal of rainwa- ter from the surface of the road as well as from the adjacent ground is crucial in Ethiopia because Summary for Ethiopia most of the roads (about 90 percent) are gravel and Ethiopia faces serious challenges in attempting to earth. Rainwater can wash away the road surfac- meet the MDGs. Along a business-as-usual growth ing construction, creating a significant inconve- path, the country's food security will further deteri- nience for vehicles and travelers. Because of poor orate. Without additional growth in agriculture, the maintenance, many existing roads have become poverty rate will actually rise, resulting in 12 million impassable during the rainy season and even in more people living in poverty by 2015 (figure 3.1). the dry season. Growth in staple crops contributes the most Lack of market infrastructure also constrains to poverty reduction. With a 3.4 percent annual the market accessibility of small farmers. For growth rate, (1.5 percent of additional growth instance, lack of storage and marketing facilities rate in productivity), growth in staple food helps and less developed formal trading systems signif- the economy and the agricultural sector grow at icantly increase farmers' transaction costs and force 3.9 and 3.5 percent, respectively. The country's many smallholders back to the subsistence mode poverty rate will reduce to 36.6 percent by 2015 of farming. from its current 44.4 percent. Combined growth While it is known that the cost of building and in staple crops with growth in livestock and non- maintaining roads is high in Ethiopia because of the traditional exports leads to much more rapid rugged topography and torrential tropical rains, growth in agriculture and poverty reduction. Such there is no approximate cost information for such a growth strategy results in an annual growth rate investments. Consequently, we have to make two of 5.3 percent for the agricultural sector, which main assumptions before conducting the simula- will help the country reduce its poverty rate to tion. First, investment is modeled as a lowering of 26.6 percent by 2015. the marketing margins between food staples sur- Increasing national food staple availability by plus and deficit areas. We assume that the market 50 percent by 2015 will significantly advance pov- prices across zones will converge due to improved erty reduction. This goal is feasible by reducing transportation and market conditions, and the the productivity gap between traditional and mod- price gap between surplus and deficit areas will ern technologies that have been adopted in the be 70 percent lower by 2015 than its current level. country. By doubling the irrigation area by 2015, We further assume that lowered marketing costs improving the efficiency of fertilizer use, and dis- are due to the improvement in the service sector's seminating the technology to combine improved productivity, and by 2015 the productivity in the seed with fertilizer use, the model results show service sector will be 20 percent higher than the that growth in staple foods is feasible with the level in the baseline's 2015, which is equivalent to right investment strategies. However, as more than an annual growth rate of 1.5 percent. 50 percent of food staples are currently provided Once growth in the agricultural sectors is com- from the food surplus area, where per-household bined with improved marketing margins, the cross- food availability is already 70 percent higher than sectoral linkage effects cause the growth rates of the national average, market access and market GDP and agricultural GDP to increase to 5.8 and development will be especially important and 5.4 percent, respectively. Improving marketing should be integrated into Ethiopia's agricultural conditions makes the terms of trade favorable to development strategy. agriculture. Specifically, reducing marketing costs As the model simulations reveal, broad-based mainly benefits smallholders through increased agricultural growth is the key for decreasing pov- prices they receive for their goods, which enables erty and increasing growth in Ethiopia. Within the them to increase their income from producing the agricultural sector, growth in cereals and other same amount of output. Due to such strong, cross- staple crops should receive priority. Due to strong, sectoral linkages and positive price effects, the cross-sectoral linkage effects, an agriculture-led Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 29 Figure 3.1 Poverty Rate in Ethiopia Under Different Growth Scenarios National poverty ratio (%) under different growth scenarios 45 Current growth path 43 Additional growth in nontraditional exports 41 39 Additional growth in livestock 37 Additional growth in staples crops 35 33 Combined growth in 31 all three subsectors 29 27 Growth in all three subsectors combined with market investments 25 23 2003 2005 2007 2009 2011 2013 2015 Source: Results from Ethiopian multimarket model. growth strategy combined with investments in decades ago. The marginalization of agriculture lies roads and other market conditions can significantly at the center of this development failure. increase rural incomes. Only under such a scenario can Ethiopia potentially halve poverty by 2015. The marginalization of agriculture Two policy-induced biases determined the condi- Zambia: Facing the Challenge of the tion of the Zambian economy at the start of the Millennium Development Goals-- 1990s. The first of these was a general bias toward The Role of Agriculture urban areas. During the 1970s the government While Zambia shares with Ethiopia a high poverty adopted an inward-oriented development strategy rate, specifically in the rural sector, the circum- based on nationalized and protected state enter- stances contributing to Zambian poverty are quite prises. So comprehensive was this strategy that by different.4 In particular, Zambia's economic history 1991 over three-quarters of GDP was being gener- has been shaped by the misuse of its abundant nat- ated by the public sector (Chanthunya and Murinde ural resources. It has failed to translate its consider- 1998). The subsequent dependence on copper earn- able mineral wealth and agricultural potential into ings, as a source of both foreign exchange and sustained growth and the improved well-being of public revenues, created an economy that was vul- its population. Even during periods of growth the nerable to crisis. The first of these crises took place poor population has had limited opportunities to in the mid-1970s when world copper prices fell participate in the growth process. As such, not only dramatically. Rather than undergo structural re- has poverty in Zambia remained high, but also form, the government chose instead to borrow trends in social indicators suggest that the economy from abroad to maintain current consumption. has continued the declining trend that began three This marked the beginning of escalating foreign 30 Agriculture and Achieving the Millennium Development Goals debt, which would by the early 1990s make Zambia exported few agricultural commodities and was a one of the most indebted countries in the world.5 net importer of food. This is a significant indicator Despite foreign borrowing, the continued dete- of the failure of agricultural policies and of the rioration in the terms of trade and falling revenues country's severe food insecurity. Burgeoning pub- led to a reduction in social spending during the lic debt and the resulting fall in rural infrastruc- 1980s. The substantial gains in social outcomes that ture investment exacerbated the situation. Many were achieved during the first decade after inde- more remote areas of the country became isolated pendence slowly began to deteriorate. During the from input and output markets. Together the bias global recession of the 1980s, the government again toward urban areas and maize production created refused to embrace public-sector reform, this time an untenable situation, which forced the govern- opting to reduce public investment, first in rural ment to implement a series of far-reaching struc- infrastructure and later in its own industrial enter- tural reforms. prises. This contraction of public investment and the small size of the private sector explain the country's poor growth performance into the 1990s. Growth and poverty under structural adjustment Beyond supporting urban-based industrial and Due to the poor performance of the Zambian econ- public-sector employment, the government's re- omy at the end of 1980s, the newly elected govern- liance on the mining sector directed social spend- ment in 1991 chose a political platform based on ing and political favor toward the urbanized the implementation of a comprehensive structural Copperbelt and Lusaka provinces (Bigsten and adjustment program. This program, which was Kayizzi-Mugerwa 2000). Perhaps the most impor- implemented during the 1990s, included macro- tant display of this urban bias is the food subsidies economic stabilization, trade liberalization, priva- for urban areas. Ostensibly to ensure food security, tization, and agricultural reforms. Each of these these subsides became entrenched and were even- policies was to play an important role in determin- tually perceived as a right by the urban population. ing the growth and poverty outcomes of the 1990s. Attempts to reduce the scale of subsidies was met The government implemented a stabilization with considerable opposition and led to riots in program aimed at curbing inflation and creating the Copperbelt province. an environment conducive to private enterprise. More broadly speaking, the government's agri- Despite eventual success, the impact of deregu- cultural policies, of which food subsidies formed lated financial markets, and the removal of food part, had a profound effect on poverty and vulner- subsidies under agricultural reform, led to rapid ability in rural areas. Largely driven by its desire to increases in consumer prices. Inflation during the protect urban food prices, the government chose to early 1990s undermined real incomes and raised support maize production throughout the country. the cost of living, especially in urban areas. Trade This was done through publicly provided input liberalization and privatization also led to wide- subsidies and marketing support, and through pan- spread job losses and contributed to rising urban territorial price controls. The effect was to distort unemployment. Many semiskilled workers who the pattern of agricultural production, such that had previously been employed in state enterprises over 80 percent of the land planted was devoted to were forced into the informal sector, where job maize (Saasa 2003). Pan-territorial pricing prompted security and wages are substantially lower. Many many farmers to grow maize in areas that were not unskilled urban workers and their households ideally suited to this crop. This was particularly moved to rural areas, thus reversing the long- true for the more drought-prone southern prov- standing migrant labor system that had urbanized inces, which are better suited to drought-resistant 40 percent of the total population by 1991. Almost sorghum and millet, and whose inhabitants there- 10 percent of the urban population moved to rural fore became highly vulnerable to climatic changes areas, mostly into small-scale farming. This col- (World Bank 2004e). lapse of the formal economy explains much of the Apart from concentrating staple production in a substantial increase in the incidence and depth of single crop, the maize bias, together with the over- poverty in urban areas during the 1990s (table 3.7). valued exchange rate caused by copper, effec- Agricultural reforms were also pronounced. tively undermined incentives to produce exportable The government abandoned its support of maize cash crops. At the beginning of the 1990s, Zambia by removing subsidies and decontrolling prices. Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 31 Table 3.7 Poverty in Zambia During the 1990s Incidence of Poverty (P0) Depth of Poverty (P1) 1991 1996 1998 1991 1996 1998 National 68.9 79.4 75.4 41.7 45.4 40.0 Rural 88.0 90.1 85.6 61.3 56.7 49.7 Urban 46.0 61.2 58.3 18.1 25.9 23.7 Central 69.8 84.1 78.9 38.2 47.8 43.9 Copperbelt 55.5 70.8 67.0 22.3 31.7 31.3 Eastern 84.3 89.0 82.7 58.8 58.2 45.3 Luapula 83.9 88.4 85.4 53.3 51.2 47.8 Lusaka 31.0 51.3 54.4 12.1 21.2 22.7 Northern 83.6 90.8 85.0 55.5 58.3 47.7 Northwestern 77.8 89.0 76.0 48.7 52.8 40.0 Southern 78.4 86.1 78.4 52.5 51.1 44.2 Western 84.5 89.2 90.3 59.1 59.7 54.3 Source: Thurlow and Wobst (2004). The loss of protection revealed the artificial prof- spatial dynamic Computable General Equilibrium itability of maize and led to its rapid decline in and microsimulation model for Zambia, by the importance. Production halved during the 1990s, International Food Policy Research Institute analy- leading to rising poverty within rural areas. How- sis shows that under its current average of 4 percent ever, farmers shifted production toward more GDP growth, the country's poverty rate will be appropriate crops (Haggblade and Zulu 2003). 68 percent by 2015, only 7 percentage points lower While the rain-fed Northern and Luapula provinces than the current poverty rate of 75 percent. Looking reverted to cassava, the drier southern provinces further into the future suggests that Zambia will planted millet. Production of these two crops dou- not be able to halve poverty until after 2040 unless bled in response to agricultural reforms. Further- pro-poor growth is accelerated (figure 3.2). The more, the correction of the overvalued exchange model estimates that annual GDP growth of 8.8 per- rate--through privatization and falling copper cent is needed to achieve the target of halving prices--made export agriculture more internation- poverty by 2015. These findings are consistent ally competitive. Cash crop production rose accord- with estimates based on static growth-poverty ingly, with cotton, sugar, and horticulture showing elasticities, which suggest that the necessary GDP rapid growth. Although initially hindered by adjust- growth rate lies between 7 and 9 percent (Thurlow ment costs and droughts, the incidence of poverty and Wobst 2004). declined in rural areas in the late 1990s, driven Rural areas are expected to perform better than mainly by improvements in cash crop production. urban areas, given the rising importance and export Perhaps more importantly, the depth of poverty potential of the agricultural sector, and the gradual also declined, reflecting the lower vulnerability decline of copper production and earnings.6 The and improved livelihoods of the poorest among new diversification, achieved at the cost of struc- the rural population. tural adjustment, appears to have corrected some of the long-standing bias against agriculture and rural development. Small- and medium-scale households Prospects for halving poverty by 2015 within rural areas benefit the most from diversifica- Although poverty reduction was evident in the late tion into agricultural export production. At the sub- 1990s, especially in rural areas, the observed over- sectoral level, it is small-scale-intensive cotton and all changes in poverty during the 1990s suggest medium-scale-intensive horticulture that grow that Zambia is facing serious challenges to meet fastest. Both sectors are expected to grow at around the MDG of halving poverty by 2015. Using a 10 percent per year, a rate that is consistent with 32 Agriculture and Achieving the Millennium Development Goals Figure 3.2 Poverty Reduction Under the Current Growth Path (2001­2050) 100 90 85.6 Rural 80 78.5 75.4 (%) 70 68.3 (P0) National 60 58.3 51.4 50 Urban 48.9 headcount 40 38.2 30 20.4 Poverty 20 10 0 2000 2010 2020 2030 2040 2050 Source: Thurlow and Wobst (2004). recent trends in these crops. However, export crop that the overall GDP growth increases from its cur- growth is largely limited to the Eastern, Central, rent 4 percent to 5 percent per year.7 and Lusaka regions where access to input and out- Increased productivity stimulates economic put markets is better. Accordingly, while the current growth and lowers poverty. However, there are growth path suggests that agriculture-led devel- considerable differences in the impact of targeting opment is desirable, additional efforts targeting different sectors. The first two scenarios contrast the a majority of smallholders and broad agricultural poverty effects of accelerating sector-wide growth sectors are necessary for achieving significant either in all agricultural sectors or all nonagricul- poverty reduction. tural sectors. More rapid productivity growth under the Agriculture-Led Growth Scenario leads to higher sectoral growth for both staples and export Agricultural growth is pro-poor growth crops. Resulting declines in nonagricultural growth Although agriculture accounts for only 25 percent are partly offset by cheaper agricultural inputs into of GDP, it is still the main source of livelihood for downstream textiles and food processing. Export most of the country's population, including the agricultural growth stimulates urban investment majority of Zambia's poor who live in rural areas by relaxing the foreign exchange constraint. Rural where the incidence and severity of poverty is great- households benefit directly from higher incomes est. Apart from shifts toward more diverse staple and falling poverty, although smaller-scale staple production, the performance of cash crops has producers benefit less than medium-scale exporters improved greatly. However, with the exception of due to the domestic market constraints they face cotton, which is grown in the Eastern province, for their food crops. Urban households also bene- most of the farmers engaged in cash crop produc- fit from lower food prices, leading to rising real tion are located close to Lusaka or the country's incomes and falling poverty. main transport routes. Thus, the recent gains from Accelerating productivity growth under the agricultural growth have tended not to reach the Nonagriculture-Led Growth Scenario has very dif- more remote areas of the country. Given the need ferent implications for poverty reduction.8 Non- for faster economic growth, and in order to assess agricultural production is more dependent on the constraints and opportunities for significant imported intermediates, thus exacerbating the poverty reduction, we consider the impact of accel- foreign exchange constraint and undermining erating productivity growth in selected sectors such import-intensive investment. High levels of exter- Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 33 nal debt and limited opportunities for further bor- and poverty reduction through the expansion of rowing mean that the large increases in imports staple production. must be matched by more rapid export growth. Exportable commodities allow farmers to get Although urban poverty declines, the shift toward access to foreign markets, and, hence, increased greater trade in capital goods offsets private con- supply results in increased exports without depress- sumption growth. Furthermore, nonagricultural ing prices. This can be seen in the Export-Crop-Led growth reestablishes the bias against agricultural Growth Scenario. The changes in poverty rate fol- exports and generates few backward linkages lowing export crop expansion indicate that rural into agriculture. Therefore, while rural households medium-scale households would be better off if benefit from increased demand in urban areas, the productivity of traded commodities could be raised. overall effect of nonagricultural growth remains Again it should be stressed that only less-remote relatively small. households that already have adequate market Despite the less optimistic results for broad access are likely to benefit. This is evident in the nonagriculture-led growth, agricultural process- strong declines in poverty within the Eastern and ing within the manufacturing sector does in fact Central provinces. represent a potential area for growth and poverty The scenarios presented so far suggest that reduction. Agribusiness benefits smallholders enhancing productivity within agriculture gener- through greater demand for their crops, and pro- ates pro-poor outcomes. Given that a majority of vides a source of income for the growing nonfarm the poor live in rural areas, it appears that agricul- population. For example, hand-milling in smaller ture has an important role to play in helping the and more remote urban areas has shown consider- country achieve faster growth and poverty reduc- able growth in recent years. Furthermore, on-farm tion. However, the benefits from agricultural processing allows smallholders to raise the value growth are unlikely to reach all rural households. added of their produce. Scenarios that focus on the Poor access to markets and credit has been a per- agroprocessing sectors take into account these sec- sistent problem in rural areas and continues to tors' stronger backward linkages into agriculture. undermine rural development. Recently achieved Although not shown here, model simulations sug- macroeconomic stability and the increasing par- gest that increased productivity in the agroprocess- ticipation of the private sector in the rural econ- ing sectors is likely to benefit urban and nonfarm omy suggest that market access might already be households more than other households, largely improving, at least in the less remote provinces. because they have better access to the required cap- Private-sector to initiated outgrower schemes have ital (Thurlow 2004). Relieving the credit constraint proven highly successful in the cotton sector and, facing many farmers should extend the potential through credit provision, have allowed smallhold- benefits of this sector. ers to become the largest suppliers of raw cotton in Since agriculture-led growth appears to offer Zambia. However, the high cost of capital needed greater opportunities for broad-based poverty for horticultural exports and agribusiness produc- reduction, two additional scenarios contrast the tion has effectively limited the participation of effects of accelerating economic growth through smallholders in these sectors. Furthermore, low either the staples or export crop sectors. The Staples- productivity and labor shortages at harvesting Led Growth Scenario concentrates agricultural limit production, indicating the need for new tech- growth within the staples-producing sectors at the nologies and capital investment (Deininger and expense of export agriculture. However, increased Olinto 1999). Inadequate financing sources, poor productivity of staple crops has little effect on market access, and low levels of investment there- poverty among small-scale farm households. This fore represent the major constraints facing agricul- is due to current market constraints, which cause tural growth in Zambia. higher production to translate into falling prices. Lower prices in staple crops do not suggest that Improving market access to encourage domestic demand for staples is insufficient, but pro-poor growth rather that the current structure of the domestic market limits its ability to absorb substantially The agriculture-led growth scenarios have empha- higher levels of supply. Poor market access there- sized the importance of market access and its role fore represents the dominant constraint to growth in determining whether rural farmers can benefit 34 Agriculture and Achieving the Millennium Development Goals from economic growth. Transportation costs cur- households. Therefore, while infrastructure and rently account for between 60 and 70 percent of market access consistently strengthen poverty the cost of production (Lofgren, Thurlow, and reduction, especially in rural areas, there appears Robinson 2004). These costs, which are high even to be a tradeoff between poverty reduction and by regional standards, limit farmers' ability to growth. market their produce and contribute to high prices and, hence, the prevailing poverty in the country. Summary for Zambia Although Zambia has an extensive road network, which supports the dominant modes of trans- Zambia has undergone substantial reforms, which portation, the system has deteriorated over the the country's recent positive performance suggests last three decades. The government has identified might have been a prerequisite for renewed growth. increased investment in transportation infrastruc- Furthermore, the rising poverty and falling social ture as critical for growth and poverty reduction. outcomes of the pre-reform period suggest that Market access is especially limited in many rural structural adjustment might also have been a pre- areas. Only 18 percent of rural households are requisite for poverty reduction. However, short-run within 5 kilometers of input markets, and few of adjustment costs raised poverty during the 1990s, the more remote households have access to health moving Zambia further away from achieving the and education facilities. These households' access goal of halving poverty by 2015 (table 3.8). A very is limited due to either a lack of roads or the poor high and unlikely growth rate would have to be quality of the existing network. achieved if the MDG poverty target is to be met. This section contrasts the impact of building new Although meeting target 1 of MDG 1 appears to feeder roads in rural areas with new paved and be beyond Zambia's grasp, the country's success at gravel roads in less remote rural and urban areas. encouraging diversification is likely to increase its The scenarios consider a 10 percent increase in rate of pro-poor growth. However, several con- the provision of either feeder or paved roads. straints remain. The most important of these con- Improved infrastructure provision reduces transac- straints is slow agricultural growth, which has been tion costs in specific sectors in the economy. In largely undermined by low productivity and in- the case of feeder roads, lower transaction costs adequate rural infrastructure. are likely to benefit only rural agricultural pro- duction. Paved roads, on the other hand, should Conclusions benefit nonagricultural production and export agriculture. The latter is due to the existing con- As the two case studies illustrate, reaching the centration of export agriculture along the coun- MDGs, and particularly MDG 1, will require much try's main road networks. Furthermore, based on greater growth than both Ethiopia and Zambia have the current geographic distribution of produc- currently achieved. In Zambia, the GDP growth rate tion, paved roads will reduce the transaction costs required to halve poverty is around 8.8 percent, in both the domestic and export markets, whereas much higher than the rate of 6 percent needed in feeder roads will reduce the transaction costs in Ethiopia. The unique characteristics of each coun- domestic markets only. try's economy and poverty profile account for this The construction of both feeder and paved roads variation in growth rates. Specifically, Ethiopia is a increases growth and reduces poverty. However, predominantly subsistence economy with agricul- improving market access for export crops has a ture contributing 52 percent to GDP and with much stronger growth effect due to the strong pos- almost 85 percent of the population living in rural itive externalities that less-remote paved roads pro- areas. By contrast, mining traditionally has dom- vide to other nonagricultural sectors. Despite slower inated economic growth in Zambia and, in turn, growth and less impressive export performance, has marginalized the country's agricultural sec- building feeder roads and improving market access tor. Indeed, 60 percent of Zambia's population for staple crops is better at reducing poverty since live in rural areas while agriculture contributes the benefits accrue to the larger and poorer small- only 22 percent to the country's GDP. Moreover, scale farm population. Export crops, by contrast, Ethiopia's national poverty rate is 44 percent with benefit the smaller population of medium-scale a concentration of poverty in the rural areas. In farm households, and to a lesser extent, urban Zambia, however, the collapse of mining exacer- Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 35 Table 3.8 Growth and Poverty Reduction Under the Alternative Zambian Growth Scenarios (2001­2015) Current Agriculture- Nonagriculture- Staples- Staples Export-Crop- Export-Crops growth path led growth led growth led growth market access led growth market access Average annual growth rate, 2001­2015 (%) Gross domestic product 4.0 5.0 5.0 5.0 5.1 5.0 5.6 Agriculture 4.6 7.7 4.2 7.8 8.1 7.1 8.6 Staple crops 4.1 7.3 4.0 8.0 8.4 4.0 3.4 Export crops 10.2 13.4 7.1 7.0 3.5 22.8 27.2 Mining 1.9 1.9 1.9 1.9 1.9 1.9 1.9 Manufacturing 4.5 4.3 5.8 4.3 4.7 4.2 4.6 Services 3.9 4.1 5.6 4.0 4.0 4.7 4.7 Final year poverty headcount, 2015 (%) National poverty 68.3 59.4 63.9 59.5 54.5 62.0 55.8 Rural 78.4 68.1 76.4 68.1 61.2 72.3 64.2 Small-scale 79.0 68.1 77.2 68.0 60.4 73.0 64.4 Medium-scale 69.5 56.3 65.2 59.0 54.5 55.9 45.1 Urban 51.4 45.0 42.9 45.2 43.3 44.8 41.8 Central 73.8 66.4 70.1 66.5 64.2 67.7 61.9 Copperbelt 61.6 55.9 54.5 56.1 54.1 56.2 52.1 Eastern 67.1 56.0 68.3 62.5 58.1 51.0 36.7 Luapula 79.2 68.4 76.1 65.8 61.8 74.1 68.4 Lusaka 45.5 40.3 36.3 40.2 39.1 40.7 39.1 Northern 79.7 65.9 76.5 62.5 55.1 75.8 70.9 Northwestern 71.1 54.2 67.5 52.6 47.9 66.5 61.3 Southern 72.7 65.6 68.4 65.6 58.9 68.0 59.2 Western 87.3 77.9 83.9 76.6 56.7 83.3 78.0 Source: Zambia CGE-micro model results from Thurlow and Wobst (2004). bated urban poverty and contributed to a national rity in both rural and urban areas. Yet, in order for poverty rate of 75 percent. Not only is the incidence growth in staples to have its intended impact, sig- of poverty higher in Zambia than in Ethiopia, but nificant market access constraints need to be over- the depth of poverty, which measures the distrib- come. Otherwise, increased staple production will ution of the poor, is higher as well. In fact, the cause a decline in local food prices and diminish poverty gap in Ethiopia is 12 percent compared farmers' incomes. with 26 percent in Zambia. In other words, Zambia In Zambia, the expansion of export crops offers has both a higher proportion of its population living opportunities for increased growth because farmers below the poverty line and a higher concentration of do not encounter the same constraints in access- individuals at the low end of the income distrib- ing foreign markets as they do domestic markets. ution. Consequently, Zambia requires faster GDP Therefore, increased supply of export crops, even in growth than Ethiopia to halve poverty by 2015. the face of domestic market constraints, would typ- Achieving these rates of growth specifically ically not result in a decline in farmers' incomes, requires greater productivity in the agricultural although the poverty impact is lower than for sta- sector. The unique structure of each country's econ- ples crops. Agro-processing also has the potential to omy determines the impact of this agricultural complement growth in other agricultural sectors. growth on poverty reduction as well as which agri- By contrast, in Ethiopia, growth in nontradi- cultural subsectors are the most promising. In both tional exports alone will not have a large impact countries, domestic demand for staples is high, on poverty reduction. However, a combination of and growth in the staples sector could increase the nontraditional export growth with livestock and incomes of the rural poor and improve food secu- staple production growth can significantly reduce 36 Agriculture and Achieving the Millennium Development Goals poverty. For both countries, the level of technol- dren who die from causes related to malnutrition ogy and capital investment required to engage in were only mildly or moderately undernourished, nontraditional export production, as well as agro- showing no outward sign of their vulnerability. processing, circumscribes the benefits to smallhold- More than half of childhood deaths are associated ers. Moreover, growth in nontraditional exports with being underweight, and malnourished chil- particularly benefits those farmers concentrated dren who survive into adulthood are more likely around cities and where market access and infra- to suffer from chronic illness and disability, and structure are well developed. have a higher probability of reduced physical and Indeed, both case studies suggest that the lack of intellectual productivity (de Onis and others 2004; physical infrastructure represents a serious prob- Pelletier and others 1994; Smith and Haddad 2000; lem that needs to be addressed, not only for agri- UNICEF 2004). Poverty, low levels of education, cultural growth to occur but also for agricultural and poor access to health services are major con- growth to reduce poverty, particularly in the most tributors to childhood malnutrition, a complex issue remote rural areas. By concurrently improving that requires tackling on a wide number of fronts, rural infrastructure and market development, agri- including (UNICEF 2004): cultural growth might help Ethiopia to achieve · "Ensuring food security for poor households, MDG 1. In Zambia, on the other hand, reaching both enough food and the right kinds of food; MDG 1 will be more difficult, particularly since the · Educating families to understand the special country is starting with a higher national poverty nutritional needs of young children, notably rate. Nonetheless, increased construction of roads, the value of breastfeeding and the impor- especially feeder roads, is no less crucial. tance of introducing suitable complementary Notwithstanding Sub-Saharan Africa's vast foods at the right age; diversity, the similarities shared by Zambia and · Protecting children from infections, by such Ethiopia have important implications for Sub- measures as immunization against common Saharan Africa as a whole. In particular, agricul- childhood diseases and provision of safe water tural growth is absolutely essential for alleviating and sanitation; the most severe poverty, especially in rural areas. · Ensuring that children receive quality care However, the nonagricultural sector has an impor- when they fall ill; tant role in assuring that this growth reaches the · Shielding them from the micronutrient defi- intended beneficiaries. Consequently, efforts to ciencies that can bring death and disabil- increase yields and expand production areas must ity, especially iodine, iron and vitamin A be appropriately sequenced with investments in deficiencies; roads, irrigation, and storage facilities as well as greater credit and fertilizer provision for small- Paying special attention to the nutritional needs holders and an enabling macroeconomic environ- of girls and women, since chronically undernour- ment that encourages private sector involvement. ished women tend to bear low-birthweight babies and so perpetuate the vicious cycle of malnutrition into the next generation." These devastating effects GLOBAL QUANTIFIED of malnutrition led world leaders to choose "to ASSESSMENT OF SELECTED halve, between 1990 and 2015, the proportion of MDG TARGETS: IMPACT-WATER people who suffer from hunger" as one of the tar- gets of MDG 1. One of the specific indicators chosen Malnutrition affects nearly one-third of all children for this target and goal is the prevalence of under- under five years of age in developing countries, weight children under five years of age. 174 million children in 1990. Malnourished children We use the IMPACT-WATER model to project have lowered resistance to infection; they are more the proportion of malnourished children under the likely to die from common childhood ailments like baseline and one alternative scenario, titled the diarrheal diseases and respiratory infections, and MDG scenario.9 Malnutrition is defined here as for those who survive, frequent illness affects their underweight (proportion of under-fives falling be- nutritional status, locking them into a vicious cycle low minus 2 standard deviations from the median of recurring sickness and faltering growth. Their weight-for-age standard set by the U.S. National plight is largely invisible: three-quarters of the chil- Center for Health Statistics and the World Health Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 37 Organization). This standard is adopted by many trition came from the World Health Organization's United Nations agencies in assessing the nutri- Global Database on Child Growth and Malnutri- tional status of preschool children in developing tion (WHO 2004), with other sources including countries. Other, less commonly used indicators the United Nations Administrative Committee on for child malnutrition include stunting (propor- Coordination­Subcommittee on Nutrition (ACC/ tion of children under five falling below minus 2 SCN 1996) and World Development Indicators (World standard deviations from the median height-for-age Bank 1997). Sources for explanatory factor data of the reference population), and wasting (pro- include calorie availability from the Food and portion of children under five with weight-for- Agriculture Organization FAOSTAT database (FAO height falling below minus 2 standard deviations 1998); female secondary enrollment data from the from the median weight-for-height of the reference United Nations Educational, Scientific and Cultural population). Organization UNESCOSTAT database (UNESCO IMPACT-WATER generates projections of the 1998) and World Development Indicators (World Bank percentage and number of malnourished preschool 2004c); and female-to-male life expectancy ratios children (zero to five years old) in developing coun- from World Development Indicators (World Bank 1997, tries. Projections for the proportion and number of 2004c). For greater detail on sources, data coverage, malnourished children are derived from an esti- specific observations used, and model estimation mate of the functional relationship between the per- procedures and tests, see Smith and Haddad (2000). centage of malnourished children, the projected Table 3.9 presents 1990 and 1995 estimates for average per-capita kilocalorie availability of food, child malnutrition (low weight-for-age) for the and nonfood determinants of child malnutrition, reporting regions for this item in IMPACT-WATER, including the quality of maternal and child care as well as the 2015 MDG target estimates (halving, (proxied by females' status relative to men as cap- between 1990 and 2015, the proportion of people tured by the ratio of female-to-male life expectancy [that is, under fives] who suffer from hunger) cal- at birth), education (proxied by the share of females culated based on 1990 values. Table 3.10 presents undertaking secondary schooling), and health and the same values for access to safe drinking water, sanitation (proxied by the percentage of the popu- related to Target 10 of MDG 7, to "halve, by 2015, lation with access to safe drinking water). The equa- the proportion of people without sustainable access tions used to project the percentage and numbers of to safe drinking water and basic sanitation." As can malnourished children are as follows: be seen in Table 3.9, child malnutrition estimates between 1990 and 1995 declined, increased, or stag- %MALt = 25.24 ln(KCALt ) - 71.76 LFEXPRATt nated. This is one of several caveats that need to be - 0.22 SCHt - 0.08 WATERt - taken into account when assessing the possibility of achieving the MDGs based on the indicators. and For example, while the WHO global database on NMALt = %MALt × POP5t child growth and malnutrition for Myanmar reports a proportion of 32.4 percent preschool children where %MAL is the percentage of malnourished malnourished in 1990, the value for 1995 was children, KCAL is per-capita kilocalorie availability 42.9 percent, and the latter value is used for the estimated in IMPACT-WATER, LFEXPRAT is the IMPACT-WATER base year calculations. Further- ratio of female to male life expectancy at birth, SCH more, several countries do not have estimates for is the percentage of females with secondary educa- child malnutrition going back to 1990. In these tion, WATER is the percentage of the population cases, country assessment reports often suggest a with access to safe water, NMAL is the number of halving of child malnutrition levels by 2015 with malnourished children, and POP5 is number of chil- reference to the earliest year with reliable data. For dren zero to five years old. example, in the case of Afghanistan, the country The regression equation was derived based on report suggests a reduction in child malnutrition a fixed-effects model of pooled, cross-section time- (underweight) from 48 percent in 2000 to 24 percent series data from 63 developing countries covering by 2015 (UNDP 2004a) as an indicator for MDG 1. the 1970s, 1980s, and 1990s from a variety of sources Moreover, the data presented in table 3.9 do for both dependent and independent variables. The not capture rural-urban disparities, but as figure 2.4 majority of the data on prevalence of child malnu- shows, child malnutrition rates are typically higher 38 Agriculture and Achieving the Millennium Development Goals Table 3.9 Child Malnutrition Estimates (Underweight), 1990 and 1995, with 2015 MDG Indicator Target 1990 estimates 1995 estimates 2015 target Percent Latin America 8.7 9.1 4.4 Sub-Saharan Africa 26.8 13.4 Nigeria 35.3 39.1 17.7 Northern Sub-Saharan Africa 40.3 40.0 20.2 Central and Western Sub-Saharan Africa 27.9 27.8 13.9 Southern Sub-Saharan Africa 28.5 26.9 14.3 Eastern Sub-Saharan Africa 26.7 27.6 13.4 West Asia and North Africa* 11.0 13.2 5.5 South Asia 54.2 27.1 India 54.7 53.4 27.3 Pakistan 40.2 38.2 20.1 Bangladesh 67.1 56.3 33.5 Other S Asia 49.0 41.2 24.5 South East Asia 35.2 17.6 Indonesia 36.8 34.0 18.4 Thailand 22.0 18.9 11.0 Malaysia 25.0 20.1 12.5 Philippines 33.5 29.6 16.8 Vietnam 43.6 44.9 21.8 Myanmar 32.4 42.9 16.2 Other SE Asia 47.0 40.0 23.5 China 19.0 17.4 9.5 Developing 30.2 15.1 Notes: *The value for West Asia and North Africa was estimated, as details for several countries in the region are not available. Latin America includes the Caribbean; N SSA stands for Northern Sub-Saharan Africa; C and W SSA stands for Central and Western Sub-Saharan Africa; S SSA stands for Southern Sub-Saharan Africa; E SSA stands for Eastern Sub-Saharan Africa; WANA stands for West Asia and North Africa; Other S Asia stands for Other South Asia; Other SE Asia stands for Other Southeast Asia. The regional disaggregation is presented in Annex 2. No estimates for child malnutrition are made for parts of East Asia (Democratic People's Republic of Korea, Macao, and Mongolia) and for the rest of the world, including Cape Verde, Fiji, French Polynesia, Kiribati, New Caledonia, Papua New Guinea, Seychelles, and Vanuatu. 1995 values are incorporated in IMPACT-WATER. The 2015 target is based on 1990 estimates. Sources: 1990 estimates based on the WHO global database on child growth and malnutrition (WHO 2004); 1995 estimates also based on the WHO global database on child growth and malnutrition, but accessed in 1997, as well as additional sources (World Bank 1997 and ACC/SCN 1996). for rural areas. As table 3.9 shows, on a regional same year (WHO 2004). Even more important than basis, the incidence of low weight-for-age in 1990 point estimates are trends in malnutrition over time. was lowest in Latin America and the Caribbean Over the last 30 years, South Asia, where the among all developing regions, at just under 9 per- largest number of malnourished children reside, cent, followed by the West Asia and North Africa has made substantial progress, with the share of (WANA) region, at around 11 percent. On the other malnourished children declining from well over hand, the level was highest for South Asia, at 70 percent in 1970 to under 50 percent by 1996­97. 54.2 percent in 1990, followed by Southeast Asia. However, the absolute number of malnourished The incidence of low weight-for-age was some- children declined by only 7 million children over what lower in Sub-Saharan Africa, at 26.8 percent. this period due to continued rapid population Regional averages mask large differences within growth. The most impressive developments took regions; for example, the average for West Asia place in East Asia, where the number of malnour- and North Africa includes Lebanon, with a share of ished children decreased by 50 percent, or 40 mil- 3 percent malnourished children in 1996, but also lion children, between 1970 and 1997. Latin America Yemen, where the share was 38.1 percent in the also reduced the number of malnourished children Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 39 Table 3.10 Access to Safe Drinking Water, 1990 and 1995, with 2015 MDG Target Indicator and 2015 IMPACT-WATER Baseline Estimates 2015 IMPACT-WATER 1990 estimates 1995 estimates 2015 target baseline estimates Percent Latin America 82.0 77.5 91.0 80.2 Sub-Saharan Africa 53.0 76.5 Nigeria 53.0 50.0 76.5 64.9 Northern Sub Saharan Africa 41.6 37.9 70.8 51.3 Central and Western Sub-Saharan Africa 48.8 54.5 74.4 66.6 Southern Sub-Saharan Africa 47.8 51.2 73.9 66.7 Eastern Sub-Saharan Africa 48.3 47.2 74.2 61.9 West Asia and North Africa 84.0 81.9 92.0 88.2 South Asia 72.0 86.0 India 68.0 81.0 84.0 89.6 Pakistan 54.2 60.0 77.1 72.5 Bangladesh 83.0 79.0 91.5 82.9 Other South Asia 54.0 60.3 77.0 71.0 South East Asia 70.1 85.1 Indonesia 71.0 60.0 85.5 73.3 Thailand 80.0 89.0 90.0 94.3 Malaysia 80.9 77.0 90.5 87.2 Philippines 87.0 84.0 93.5 90.3 Vietnam 55.0 43.0 77.5 62.4 Myanmar 64.0 60.0 82.0 69.4 Other SE Asia 29.1 59.6 64.5 67.7 China 71.0 67.0 85.5 75.6 Developing 74.0 87.0 Notes: Latin America includes the Caribbean; N SSA stands for Northern Sub-Saharan Africa; C and W SSA stands for Central and Western Sub-Saharan Africa; S SSA stands for Southern Sub-Saharan Africa; E SSA stands for Eastern Sub-Saharan Africa; WANA stands for West Asia and North Africa; Other S Asia stands for Other South Asia; Other SE Asia stands for Other Southeast Asia. The regional disaggregation is presented in Annex 2. No estimates are made for child malnutrition for parts of East Asia (Democratic People's Republic of Korea, Macao, and Mongolia), and for the rest of the world, including Cape Verde, Fiji, French Polynesia, Kiribati, New Caledonia, Papua New Guinea, Seychelles, and Vanuatu. 1995 values are incorporated in IMPACT-WATER. The 2015 target is based on 1990 estimates. 1995 and 2015 IMPACT baseline estimates were used for the baseline run. Sources: 1990 estimates based on the World Development Indicators (World Bank 2002) and WHO/UNICEF (2001); 1995 estimates also based on World Development Indicators, but accessed in 1997. by half, but the number of malnourished children dren has fallen much less sharply. In addition, in WANA today is virtually the same as it was in despite the long-term improvement in most regions, the 1970s. The most troubling region, however, progress has slowed in recent years, causing addi- is Sub-Saharan Africa, where the number of mal- tional concern over future prospects for reducing nourished children increased by over 75 percent or child malnutrition (Rosegrant and Meijer 2002). 12.9 million over the last 30 years. In 1970, roughly Regarding the proportion of people with access 1 out of 10 malnourished children resided in Sub- to safe drinking water, rural-urban disparities Saharan Africa, today 1 in 5 do. are important and are typically reported for Despite Sub-Saharan Africa's discouraging dev- developing-country regions (WHO/UNICEF 2001). elopments, globally significant improvements in An important issue regarding the access to safe child malnourishment have been made, as shown drinking water data is that different surveys under- by a sharp reduction in the proportion of malnour- taken result in very different estimates even for ished children. However, because of population the same year. For example, in Ghana, the Ghana growth, the absolute number of malnourished chil- Living Standards Survey of 1988 estimated that 40 Agriculture and Achieving the Millennium Development Goals 36 percent of both the urban and rural population Saharan Africa at 11 and 12 percent, respectively had access to an improved drinking water source (Appendix Table 1). (with different water sources for rural and urban For the analysis here, two scenarios were used. areas), while the Ghana Demographic and Health The first is a baseline scenario that reflects our best Survey of 1988 reported that 84 percent of the urban estimates of future trends in food supply, demand, population and 28 percent of the rural population and trade, as well as changes in the noncaloric had safe access to drinking water. Finally, the parameters determining child malnutrition. The International Drinking Water Supply and Sanita- second scenario, titled MDG Scenario, attempts to tion Decade Review of National Progress (as of reach the child malnutrition indicator of MDG 1 December 1988) estimated that in 1988, 93 percent by 2015 through increases in economic and agri- of the urban population and 39 percent of the rural cultural growth and complementary improvements population had access to safe drinking water in in the service sectors, particularly for those coun- Ghana (WHO/UNICEF 2001 for Ghana). tries and regions least likely to make sufficient On a regional basis, in 1990 access to an im- improvements for this target if current trends pre- proved source of drinking water was estimated vail. Basic parameters for both scenarios are listed to be highest in West Asia and North Africa at in tables 3.11 and 3.12. 84 percent, followed by Latin America with 82 per- Table 3.12 presents the increases in income cent and Asia with around 70 percent. Coverage growth required between 1995 and 2015 under the was lowest in Sub-Saharan Africa with 53 percent. MDG scenario compared to the baseline scenario. The data for the other two noncaloric parame- Nonagricultural income growth rates for Latin ters incorporated in the child malnutrition equation America and Asia, with the exception of South Asia, described above--the ratio of female-to-male life are increased to 25 percent. Rates for West Asia and expectancy at birth, and the proportion of females North Africa are increased to 6 percent per year, and attending secondary schooling--are presented in rates for the two regions considered least likely to Appendix Table 1 for 1995 and 2015 (estimates for meet the MDG targets, South Asia and Sub-Saharan baseline and MDG scenarios). The indicator for the Africa, are increased to 8 percent per year. share of female secondary schooling is a reflec- Thus, the following items have been included tion of Target 3--to "ensure that, by 2015, children in the MDG scenario: everywhere, boys and girls alike, will be able to complete a full course of primary schooling"--of · Nonagricultural income growth increases the second MDG, "achieving universal primary in developing regions by 25­150 percent, education," and of Target 4--"eliminate gender dis- with particularly rapid increases projected parity in primary and secondary education prefer- for those regions least likely to reach the ably by 2005 and to all levels of education no later malnutrition indicator, Sub-Saharan Africa than 2015"--of MDG 3, "promote gender equality and South Asia. and empower women." · Annual irrigated yield growth per hectare In 1995, the ratio of female-to-male life expect- increases by 50 percent and rainfed yield ancy at birth for 195 countries with available data growth increases by 25 percent, for all devel- ranged from less than 1 (0.9876 for Nepal and oping countries and regions (cereals, roots and 0.9982 for Uganda) to a high above 1.2 (Latvia, tubers, and soybeans). The increase of crop 1.2023, Estonia, 1.2042, and the Russian Federation, productivity is due to a large expansion of 1.2305). Among the developing-country reporting investments in agricultural research and irriga- regions, Latin America and the Caribbean had the tion infrastructure, enhanced property rights highest ratio of female-to-male life expectancy at to land and water, improved coordination 1.0989 in 1995, whereas Bangladesh had the low- among agencies, and more transparent and est value (1.0072), followed by India with 1.0088 accountable use of funds. Broad-based growth (Appendix Table 1). Regarding the proportion of in production of meat and livestock products female secondary schooling, in 1995, among the is assumed to be generated by more rapid IMPACT-WATER reporting regions, the share was expansion in animal numbers, with a 50 per- highest for the Philippines at 78 percent, followed cent increase in annual numbers growth for by Malaysia at 63 percent, and China at 62 percent. all developing countries and regions. Both The share was lowest in Northern and Eastern Sub- items, crop and livestock production growth, Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 41 Table 3.11 Parameters for Developing Countries, Baseline and MDG Scenarios Parameter Baseline scenario MDG scenario Nonagricultural income growth See table 3.12 Population growth UN medium variant UN medium variant Livestock numbers growth Output numbers growth: Increase in numbers growth of 1.43%/yr, 2005­2015 animals slaughtered in developing countries by 50% beginning in 2005 Food crop yield growth Output cereal yield growth: Increase irrigated yield growth by 1.39%/yr, 2005­2015 50% for cereals, roots, and tubers and soybean in developing regions; increase rainfed yield growth by 25% for same crops and countries, all beginning in 2005 Irrigated area growth Output irrigated harvested area Increase irrigated area growth by a growth: 0.16%/yr, 2005­2015 factor of 0.1 Access to water Interpolated from IFPRI estimates Following MDG Target 10 for 2020 (see table 3.10) (see table 3.10) Female secondary education Interpolated from IFPRI estimates Increased from 1995 by between for 2020 (see Appendix Table 1) 20­150% to reach malnourishment indicator, but not higher than 95% (see Appendix Table 1) Female-to-male life expectancy ratio at birth Interpolated from IFPRI estimates Adjusted to reach malnourishment for 2020 (see Appendix Table 1) indicator (see Appendix Table 1) Source: IFPRI-IMPACT parameters. will increase food availability in developing Results regions and therefore reduce the share of mal- The results of the 2015 baseline and MDG scenario nourished children. simulations based on IMPACT-WATER show that · Increased investment levels in health and the indicator for halving the prevalence of child access to improved drinking water sources malnutrition can be reached (or even overshot) in will help achieve MDG 7 Target 10. some countries and regions, even in the baseline · Increased investment levels in education will scenario, while it is unlikely to be met, even under help reach higher levels of secondary enroll- the MDG scenario, for others. ment of girls. The combined effect of the parameter changes · Increased investment in female and maternal is first, an increase in projected constant-price well-being will help to reach improvements (rainfed and irrigated) cereal yield growth for all in the ratio of female to male life expectancy developing regions, 2005­2015, from 1.39 percent at birth. annually under the baseline to 1.88 percent per year Thus, the MDG scenario combines two broad under the MDG scenario; second, an increase of courses for improving food security and reducing crop-irrigated harvested-area growth from 0.16 per- poverty in developing regions: the first way is cent per year under the baseline to 0.21 percent through broad-based and rapid agricultural pro- annually under the MDG scenario; and third, an ductivity and economic growth to increase effec- increase from baseline livestock numbers growth of tive incomes, effective food demand, and food 1.43 percent per year during 2005­2015 to 2.17 per- availability; and the second is through investments cent annually under the MDG scenario. in education, social services, and health (proxied in Under the MDG scenario, increased per-capita the model by female secondary enrollment rates, income and lower food prices resulting from food ratio of female-to-male life expectancy at birth, and production increases lead to higher levels of per- access to clean water). capita food consumption. Increased per-capita food 42 Agriculture and Achieving the Millennium Development Goals Table 3.12 Income Growth Requirements for Baseline and MDG Scenarios, 1995­2015 Country/Region Baseline scenario MDG scenario Percent per year Latin America (25% increase) Mexico/Brazil/Other Latin America and Caribbean 3.6 4.5 Colombia 3.8 4.8 Argentina 4.5 5.6 Sub-Saharan Africa Nigeria 3.8 Northern Sub-Saharan Africa 3.3 Central and Western Sub-Saharan Africa 3.8 8.0 Southern Sub-Saharan Africa 3.2 Eastern Sub-Saharan Africa 3.5 West Asia and North Africa Egypt 4.0 Turkey 4.5 6.0 Other West Asia and North Africa 3.5 East Asia (25% increase) China 6.0 7.5 South Korea 5.0 6.3 Other East Asia 3.5 4.4 South Asia India 5.8 Bangladesh 4.8 8.0 Pakistan 4.5 Other South Asia 4.5 South East Asia (25% increase) Indonesia 4.5 5.6 Thailand/Malaysia 5.2 6.5 Philippines 5.0 6.3 Vietnam 5.8 7.3 Myanmar/Other South East Asia 4.0 5.0 Notes: Latin America includes the Caribbean; N SSA stands for Northern Sub-Saharan Africa; CW SSA stands for Central and Western Sub-Saharan Africa; S SSA stands for Southern Sub-Saharan Africa; E SSA stands for Eastern Sub-Saharan Africa; WANA stands for West Asia and North Africa; O S Asia stands for Other South Asia; O SE Asia stands for Other Southeast Asia; O E Asia stands for Other East Asia. The regional disaggregation is presented in Annex 2. Source: IFPRI-IMPACT parameters. availability translates into higher per-capita caloric in Latin America. The daily kilocalorie availability intake, which in turn reduces the share of mal- increase is largest by far for the Sub-Saharan Africa nourished preschool children. Strong growth in region at 515 kilocalories, driven by very rapid public investment in the social sectors, including income growth (8 percent per year) and rapid agri- total social expenditures, and particularly expen- cultural growth. In South Asia, caloric availability ditures for education and water and sanitation, increases by 128 kilocalories, driven by the same result in further reductions in the share of mal- factors as in Sub-Saharan Africa. WANA, which nourished children. already has high levels of kilocalorie availability in The projected levels of calorie availability for the 1995 and the 2015 baseline scenario, is expected to major IMPACT-WATER regions for the two scenar- further increase availability by a smaller amount, ios are shown in figure 3.3. Compared to the base- 65 kilocalories. line scenario, per-capita kilocalorie availability Figure 3.4 presents the results for child malnutri- improves for all developing regions in 2015 under tion for the group of developing countries for 1995 the MDG scenario, apart from a very small decline and 2015 (baseline, MDG scenario, and MDG indi- Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 43 Figure 3.3 Per-Capita Kilocalorie Availability in 1995 and Projections to 2015, Baseline and MDG Scenarios 3,500 1995 2015 baseline 3,000 2015 MDG 2,500 2,000 1,500 Kilocalories/cap/day 1,000 500 0 SSA S Asia SE Asia E Asia LAC WANA Source: IMPACT-WATER Simulations, IFPRI 2004. cator target). On average, the MDG scenario misses the absolute number of malnourished children: the MDG 2015 indicator by only 1.9 percentage from 161 million children in 1995 to 131 million points. The changes in agricultural and comple- children under the baseline, and 91 million children mentary social indicators result in a reduction in under the MDG scenario. total child malnutrition from 24.4 percent under the As can be seen in figure 3.5, the Latin America baseline to 17.0 percent under the MDG scenario. and Caribbean region basically achieves the target These declining shares reflect large reductions in rate even under the baseline scenario (4.6 percent Figure 3.4 Child Malnutrition, Developing Countries, 1995, 2015 Baseline, 2015 MDG Scenario, and 2015 Target 35 31 30 24 25 20 17 15 Percent15 10 5 0 1995 2015 baseline 2015 MDG 2015 target Source: IMPACT-WATER Simulations, IFPRI 2004. 44 Agriculture and Achieving the Millennium Development Goals Figure 3.5 Child Malnutrition, Developing Country Regions, 2015 Baseline, 2015 MDG Scenario, and 2015 Target 45 2015 baseline 41 40 2015 MDG 35 2015 target 30 30 28 27 27 24 25 21 Percent20 17 18 17 15 15 13 11 10 10 8 7 7 5 4 6 5 2 0 LAC SSA WANA S Asia SE Asia China Developing Source: IMPACT-WATER Simulations, IFPRI 2004. proportion of malnourished children compared to and a high target rate for safe drinking water a target rate of 4.4 percent). Noncaloric parameters result in a rate of 6.8 percent preschool children of the child malnutrition equation were changed malnourished. very little for the MDG scenario for this region. However, other developing regions, includ- Similarly, China, where declines in child malnu- ing Sub-Saharan Africa, South Asia, and parts of trition have been very rapid over the last 30 years, Southeast Asia, are less likely to reduce the inci- is projected to come close to the 2015 target indi- dence of malnutrition in under fives by half from cator for child malnutrition even in the baseline sce- 1990 to 2015, although very rapid growth in both nario. Under the baseline, the share of malnourished income and agriculture can help move these coun- preschool children is expected to reach 10.9 percent tries and regions almost within reach of the tar- compared to a target rate of 9.5 percent. Under the get indicator. MDG scenario projecting rapid increases in agricul- Under business-as-usual conditions, Sub- tural and economic growth, and moderate increases Saharan Africa is unlikely to meet the MDG target in income growth, the target is reached easily, indicator of halving the prevalence of malnour- even under very slow improvements in noncaloric ished children by 2015 to 13.4 percent. In the base- parameters, an increase in the ratio of female- line scenario, a reduction from 33 percent in 1995 to-male life expectancy from 1.0467 to 1.0650, and to 28 percent in 2015 leaves the region still with an increase in female secondary school participa- more than double the target rate for child malnutri- tion from 62 percent in 1995 to 74 percent by 2015. tion levels. In absolute terms, this translates into an These increases result in an under-five malnutrition increase in the number of malnourished children rate of 7.4 percent under the MDG scenario, or from 32 million to 37 million children. Under the 7.0 million children malnourished by 2015. MDG scenario, in addition to very optimistic non- WANA is the third region where reaching the agricultural income growth rates and very rapid target indicator for malnourished children appears crop and livestock production increases, highly feasible, on average. Under the baseline scenario, optimistic developments for the noncaloric param- the incidence of child malnourishment declines to eters are simulated, including very rapid changes 8.4 percent compared to the target rate of 5.5 per- in the ratio of female-to-male life expectancy at cent. Under the MDG scenario, income growth of birth and a doubling or more of the share of female 6 percent annually for all countries in the region, secondary school enrollment. Even under these combined with rapid agricultural growth, limited highly optimistic assumptions, the 2015 MDG sce- improvements in female secondary schooling and nario results in 17 percent of under fives malnour- the ratio of female-to-male life expectancy at birth, ished, or 23 million children. Countries and regions Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 45 expected to struggle most with the MDG target (Thailand) and 1.4 percentage points (Malaysia), indicator are Nigeria and Northern Sub-Saharan despite low improvements in noncaloric malnu- Africa, where child malnutrition rates in 1995 had trition parameters. Other countries, including been 39 and 40 percent, respectively. Indonesia, Myanmar, the Philippines, Vietnam, In South Asia, the share of malnourished chil- Cambodia, and Lao People's Democratic Republic dren is projected to decline from 51 percent in 1995 are less likely to halve the proportion of malnour- to 41 percent in 2015 under the baseline, a very sub- ished children in the baseline scenario; Myanmar stantial decline, translating into a reduction in the and Vietnam miss the target by 21 and 15 percent- number of malnourished children by 20 million. age points, respectively. Compared to the baseline, This process is mostly driven by India, which is under the MDG scenario, rapid increases in agri- projected to reduce child malnutrition from 53 per- cultural and moderate additional economic growth, cent in 1995 to 43 percent in 2015, or an absolute together with additional investments in social reduction of 19 million children. The MDG scenario sectors--as proxied by improvements in the ratio incorporates much faster economic and rapid agri- of female-to-male life expectancy at birth, higher cultural growth, and very optimistic assumptions proportions of female secondary schooling, and for the noncaloric parameters, such as a jump in the enhanced access to water--lead to declines in the ratio of female-to-male life expectancy at birth from number of malnourished children by 2015. These 1.0088 for India (the fifth lowest rank among 195 declines are 6 percent in Indonesia, 3 percent in reported values in 1995) to 1.0850, and an increase the Philippines (where many social indicators are in secondary female schooling from 38 percent in already at high levels today), 11 percent in Vietnam, 1995 to 95 percent by 2015. These optimistic non- 13 percent in Myanmar, and 3 percent in other caloric parameters necessitate very large invest- Southeast Asia, narrowing the Southeast Asia ments in the social sector. However, the outcome of gap to just 3.3 percentage points from the target the MDG scenario is still 30 percent preschool chil- (table 3.13). dren malnourished in 2015, compared to the target Figure 3.6 presents the number of people with- rate of 27 percent. out access to safe drinking water in 1995, for the In Southeast Asia, finally, some countries like baseline and MDG scenarios, along the lines of Malaysia and Thailand almost reach their respec- MDG 7, Target 10, to "halve, by 2015, the propor- tive baseline target reductions of 12.5 percent and tion of people without sustainable access to safe 11.0 percent, and therefore surpass the indicator drinking water and basic sanitation." Rates of target rates under the alternative scenario by 2.5 access to clean water underlying this figure are Table 3.13 Southeast Asia, Share of Malnourished Preschool Children, 1995, and 2015 Baseline, MDG Scenario, and Indicator Target Countries/ 2015 2015 regions 1995 Baseline MDG Target Percent Indonesia 34.0 27.8 21.4 18.4 Thailand 18.9 12.1 8.5 11.0 Malaysia 20.1 13.8 11.1 12.5 Philippines 29.6 23.1 20.0 16.8 Vietnam 44.9 37.2 26.1 21.8 Myanmar 42.9 37.4 24.5 16.2 Other SE Asia 40.0 29.9 27.1 23.5 South East Asia 34.1 27.5 20.9 17.6 Note: Other SE Asia includes Brunei, Cambodia, and Lao PDR. Myanmar's child malnutrition indicators worsened considerably between 1990 and 1995, but have since improved again. 46 Agriculture and Achieving the Millennium Development Goals Figure 3.6 Number of People Without Access to an Improved Water Source, 1995, and Projected 2015 Baseline and MDG Scenarios 900 854 1995 800 2015 baseline 2015 MDG 700 600 567 (million) 500 485 people of 400 300 276 Number 198 200 165 108100 100 52 61 51 36 0 Latin America WANA SSA Asia Source: IFPRI calculations. based on estimates of likely improvements in access vision, and agricultural research. Total irriga- for the baseline and required improvements to tion investments are calculated by multiplying the reach Target 2 for preschool children for the MDG estimated increases in irrigated area, adjusted for scenario. In 1995, an estimated 1.3 billion people in cropping intensity, during 1995­2015 by the aver- developing countries lacked access to safe drink- age cost of irrigation per hectare, expressed in 1995 ing water according to the IMPACT-WATER rates real U.S. dollars. Rural road investments are calcu- and population numbers. Under the baseline sce- lated by multiplying the incremental road length nario, by 2015, 0.9 billion people still are estimated required in 1995­2015 by road investment costs to be without access to safe drinking water. The per kilometer. The incremental rural road length is decline in the number of people without access to calculated assuming first, that the density of roads safe water during 1995­2015 would be largest for is proportional to cropland area, and second, that Asia at 34 percent, followed by Sub-Saharan Africa crop yield growth also contributes to road expan- at 28 percent, and slowest for Latin America at sion. Expenditures for public agricultural research 7 percent. Under the much higher rates projected are based on expenditure trends and projections for the MDG scenario, the number of people with- and their relative contribution to crop yields. The out safe access would drop to 0.7 billion people, estimated incremental investment in education which also still falls short of the target. Under this is based on the cumulated annual costs of the addi- scenario, only Latin America and parts of Asia tional number of female students required to in- would reach the target. crease the percentage of females with access to secondary education to the levels projected in the scenarios (Appendix Table 1). It is calculated by Implications for Investment the cumulative annual required increase in enroll- What are the implications for public investment for ment multiplied by the annual cost of secondary the 2015 baseline and MDG scenarios, based on school education per student. Incremental invest- IMPACT-WATER calculations? The most important ment costs for clean water, finally, are based on the public investment drivers in IMPACT-WATER are investment required to increase the share of people irrigation, rural roads, education, clean water pro- with access to clean water to the levels projected Alternative Scenarios to 2015: Prospect and Policies for Meeting the MDGs on POVERTY 47 under the alternative scenarios. Per-capita costs of sponding with the higher yield increases achieved providing clean water and estimates of the num- under this scenario. The average irrigation cost ber of people gaining access to clean water are dis- per hectare increases from $4,850 in the baseline to aggregated by urban and rural areas. Additional $6,204 in the MDG scenario as more expansion details on calculation procedures are presented in occurs in areas where projects are more costly, such Rosegrant and others (2001). The cost of improve- as Sub-Saharan Africa. The increase in irrigation ments in the five sectors during 1995­2015 under the investments is thus not only due to the expansion of baseline and MDG scenarios are shown in figure 3.7. irrigated area, but also to the increase in the cost of As expected, the MDG scenario envisions large irrigation under the MDG scenario. Compared to increases in investment in the key drivers. Total the baseline 2015 net area, area under the MDG sce- estimated expenditures during 1995­2015 for the nario is only 4 million hectares larger. As relatively group of developing countries are $430 billion high levels of access to clean drinking water are under the baseline scenario and $591 billion under already achieved in the baseline scenario, only the MDG scenario. Investments in rural roads $15 billion in investments are added for the MDG account for 28 percent of total investment under the scenario. The increase in investment in agricul- baseline, followed by agricultural research and irri- tural research under the MDG scenario is relatively gation with 24 percent and 21 percent, respectively, small, $5 billion. Due to the long lags in generation and clean water and education with 15 percent and of impact from agricultural research, increases 12 percent, respectively. In the MDG scenario, the in research expenditures, even beginning now, share of investments in education increases to will have relatively small impacts on crop yields 20 percent, with levels more than doubling from by 2015. Increased investments in agricultural $51 billion to $118 billion due to the assumed rapid research are likely to be essential to meet crop and expansion in female secondary schooling, particu- animal production needs beyond 2015, but do larly in parts of Asia and Sub-Saharan Africa. There not have large direct impacts on achieving the are increases in investments for rural roads, irriga- MDG scenario. Other investments, such as roads tion infrastructure, and agricultural research, corre- and irrigation, have significant lags in impact as Figure 3.7 Cost Estimates for Implementing the Baseline and MDG Scenario, 2015, Developing Countries 700 Education Clean water 591 600 Irrigation Ag research Rural roads 500 430 400 billion US$300 200 100 0 Baseline MDG Source: IFPRI calculations. 48 Agriculture and Achieving the Millennium Development Goals well, so implementing the investment portfolio ished) and the target indicator (15 percent children required for the MDG scenario will require very malnourished) to reach 17 percent. rapid action. However, the outcome varies significantly by The total increase in investments estimated country and region. While Latin America, the West based on IMPACT-WATER calculations is $161 bil- Asia and North Africa region, and China will likely lion in agricultural and supporting sectors dur- come close to or possibly reach the target indicator ing 1995­2015 under the MDG scenario to bring by 2015, even under business-as-usual, chances developing countries and, in particular, two major that Sub-Saharan Africa and South Asia will reach regions unlikely to achieve the MDG target indica- their respective target rates are much smaller. In tor for malnourished children--South Asia and fact, on average, Sub-Saharan Africa and South Sub-Saharan Africa--within reach of the preschool Asia will not be able to reduce their shares of mal- malnutrition target indicator. nourished children by 2015 even under very favor- able agricultural, economic, and social conditions, but improved conditions in these sectors can bring Conclusions the countries in these regions much closer to the 2015 target and can facilitate further reductions in The alternative scenarios have shown that the malnutrition later on. In Southeast Asia, finally, combination of agricultural and economic growth, the picture is mixed, with some rapidly growing together with larger investments in social sectors economies likely to reach the Millennium target including health and education, can almost com- indicator on child malnutrition, while others, in- pletely eliminate the gap between the business- cluding Myanmar and Vietnam, will need sub- as-usual outcomes for 2015 (24 percent of stantial additional investments in agriculture and developing-country preschool children malnour- social sectors to get closer to the target. 4 Global trends and Emerging Issues in Meeting the MDGs A Trade, Macroeconomic, and Policy Perspective Current trade barriers POLICY ACTIONS AND THEIR IMPACTS ON THE MDGS and subsidies are high The prospects for meeting the MDGs of eradicating extreme hunger for agriculture. and poverty and achieving the health and education goals are essen- tial to and directly affected by the agriculture sector. Farm-sector Successful WTO support policies and border protection worldwide are some of the policies at the sectoral level that influence both agricultural and MDG agricultural trade outcomes. Since most of the world's poor are rural and depend on negotiations that lower agriculture for part or all of their livelihoods, agriculture is a critical sector in which the global trade system must work to their benefit. protection and subsidies Policies directed at achieving the targets specified under the MDGs should be particularly directed at smallholder agriculture. could boost progress While small individually, in many cases smallholders account for a large share of agricultural production. In Sub-Saharan Africa, over toward the MDGs. 90 percent of agricultural output comes from smallholders, who account for nearly three-quarters of the poor. In India, farmers with less than 2 hectares account for 40 percent of total food grain produc- tion (Narayanan and Gulati 2002). Macroeconomic policies of the developing countries also affect agriculture and rural poverty. A number of developing countries have implemented structural adjustment programs aimed at correct- ing fiscal imbalances, largely through reductions in public expendi- tures; redefining the role of the state in economic affairs; privatizing major sectors of the national economies; accelerating growth; and through trade liberalization (Goldin and Winters 1992). These struc- tural adjustment programs affected agriculture and especially small- holders in a number of ways. Beneficial effects often arise from lower inflation, reduced government debt accumulation, and falling inter- est rates, as well as from depreciation of overvalued exchange rates. Yet, redressing fiscal imbalances also typically involves reductions in both agricultural and nonagricultural subsidies (Schiff and Valdez 1998) and significant reductions in public spending, with potentially adverse impacts on agriculture. Relative impacts on agriculture depend in part on the targeting of the remaining allocation of public spending; if public investments are targeted to public goods and specifically to services that foster eco- nomic growth, then impact on agricultural output may well be posi- tive. For example, Van Blarcom, Knudsen, and Nash (1993) analyzed a sample of 32 countries in which public spending on agriculture had been cut some time after 1970. As mentioned by Schiff and Valdez 49 50 Agriculture and Achieving the Millennium Development Goals (1998), the 1993 analysis concluded that much policy issues, as mentioned in Orden, Torero, and of the public spending on agriculture in these coun- Gulati (2004), are whether agriculture will be tries had been directed toward relatively unpro- brought more fully under liberalized trade rules ductive purposes and, therefore, reductions were and how the outcomes will affect the rural poor. appropriate in those cases. Additional dimensions of the possible disci- A review of a range of national poverty reduc- plines on agricultural support and protection poli- tion strategy papers (PRSPs) showed evidence of a cies are their effects on food aid and the impact of link between poverty and lack of infrastructure ser- technical regulations and standards on agricul- vices. In particular, agricultural research, education, tural trade opportunities. The latter are posing and rural infrastructure are the three most effective challenges to market participation by smallholders public spending items in promoting agriculture just as high-value demands are creating potential growth as shown for India and China (Fan, Hazell, new income streams. and Hague 2000; Fan, Hazell, and Thorat 2000; Fan, In this chapter we examine four aspects of the L. Zhang, and X. Zhang 2002; Fan and Hazell 1999; policies affecting achievement of the MDGs. These Zhang and Fan 2000). include trade and domestic support polices for Fan and Rao (2004) compiled government expen- agriculture in developed and developing countries, ditures by type across 43 developing countries from macroeconomic reform and public sector infrastruc- 1980 to 1998 and found that structural adjustment ture and other investments, the role of the private programs had different consequences for different sector and public-private partnerships, and the sectors. In Africa, governments reduced expendi- importance of good governance. We argue that sup- ture shares for agriculture, education, and infra- portive government investments and well function- structure, while Asian governments reduced shares ing private and public market institutions, together for agriculture and health. In Latin America, educa- with foresight in the design of agricultural policies, tion and infrastructure suffered from reduction in are required to take advantage of market opportu- government expenditures. nities to sustain increased agricultural output and Trade liberalization has been another major com- raise rural incomes that will help achieve the tar- ponent of structural adjustment programs, and lib- gets formulated for the MDGs. eralization of nonagricultural trade has to some extent improved agricultural incentives through lower industrial prices and through the deprecia- TRADE POLICIES tion of the real exchange rate. However, the poten- Policies of Developed Countries tial benefits of agricultural trade liberalization have not been generally realized because agriculture Support policies and border protection of wealthy has long been treated as a special case left out- Organization for Economic Co-operation and Dev- side the multilateral trade-liberalization process elopment (OECD) countries, valued at hundreds in the General Agreement on Trade and Tariffs of billions of dollars each year, cause harm to agri- (GATT). As a result, extensive Organisation for culture in developing countries. Evaluating the Economic Co-operation and Development (OECD) overall effects of the subsidies and protection subsidies and border protection continue to block among developed countries, assessing the effects of opportunities for those poor people who depend these policies specifically on developing countries, on farming for their livelihoods. or even more specifically assessing their effects on The Uruguay Round of GATT negotiations poverty in developing countries, are complex chal- (1986­94) created the World Trade Organization lenges. The evaluation must rest on counterfactual (WTO) and also produced the first comprehensive simulation of alternative policy scenarios. A diverse framework of multilateral disciplines on agricul- set of policies has to be represented, and models to tural subsidies and trade policies. Yet, the Uruguay accomplish these tasks differ in assumptions about Round Agreement on Agriculture achieved only crucial parameters, levels of aggregation, scope of modest agricultural trade liberalization. A critical commodity and country coverage, and many other question is whether the current Doha Development dimensions. Round of WTO trade negotiations (2001) can build A number of model results were reviewed on this framework to deliver further market open- recently by Beierle and Diaz-Bonilla (2003) with ing and opportunities for trade. The overarching the objective of describing what is known and the Global trends and Emerging Issues in Meeting the MDGs 51 remaining knowledge gaps on whether trade lib- welfare. However, there are some studies that eralization (in the form of reduced protection differentiate impacts by types of households (for and export subsidies and lowered import restric- example, agriculture, self-employed, nonagricul- tions) would benefit smallholder farmers and others ture; male or female head), focus on food security in poverty in developing countries. Several key or poverty reduction, or, most recently, incorpo- findings from their review and other assessments rate detailed household survey data to evaluate follow: net and distributional effects of reforming trade policies. · Most models demonstrate negative impacts A few representative results illustrate the points of current developed country (OECD) trade above. The primary mechanism by which OECD protection policies and positive impacts from transmits agricultural protection and subsidies developed country liberalization on develop- around the world is commodity prices in world ing country welfare, agricultural production markets. The analytic studies suggest that OECD and incomes, and food security. agricultural subsidies and protection depress world · Impacts vary by country, commodity, and sec- prices of basic agricultural food crops in the rough tor, and for regions within countries. range of 1­5 percent. OECD protection (mainly · OECD market access restrictions harm dev- tariffs and tariff-rate quotas [TRQs]) depresses the eloping countries, but effects of production prices of other nonstaple commodities by a larger and income-support subsidies are more amount, such as 8 percent for sugar, 22 percent for ambiguous. sheep meat, and 24 percent for milk. Averaged · Developing countries tend to gain more from across all commodities, a common estimate of the liberalization of their own policies than from extent to which OECD policies depress prices is reforms by the OECD. Consumers in develop- 10 percent, with various studies estimating average ing countries benefit widely from developing price changes in the range of 5­20 percent.10 country liberalization reforms. The models show that the increase in world · Model results differ on the basis of assump- prices from removal of OECD protection will lead tions such as the scope of commodity cover- to larger agricultural production in develop- age, mobility of resources among alternative ing countries. Beghin, Roland-Holst, and van der crops and between farm and nonfarm employ- Mensbrugghe (2002) estimate that removal of ment, availability of underutilized labor, and OECD protection could boost rural value added in static versus dynamic analysis. low- and middle-income countries by $60 billion per · Multilateral liberalization reduces the benefits year. Tokarick (2002, 2003) arrives at lower num- derived from preferential trade agreements, bers, estimating that OECD market access barriers but these losses are relatively small compared and subsidy policies cost developing countries to the overall gains from the broader reforms. as a whole $8 billion in (overall) welfare annually · Most models have not had sufficient resolution (0.13 percent of developing country GDP). to analyze the impacts of reforms on small- In a model that allows for unemployment in holders, subsistence farmers, and other poor rural and urban sectors in developing countries and households, but there is an emerging literature for slight positive effects of technical change corre- attempting to do so (Beierle and Diaz-Bonilla lated with higher levels of trade openness, Diao, 2003; Hertel and Winters 2004; Narayanan and Diaz-Bonilla, and Robinson (2003) estimate that Gulati 2002; Tokarick 2002). OECD subsidies and border protection reduce agri- With the diversity in modeling approaches, no cultural exports from the developing world by single model has all the desired features that would $37.2 billion (25.3 percent) annually. Agricultural allow an examination of the impacts of trade protec- value added among developing countries is re- tion and liberalization on smallholder and sub- duced by $23.0 billion annually, while national sistence farmers and food security. Most of the welfare of developing countries is repressed by studies disaggregate only to the regional level-- $9.4 billion. For specific countries and specific com- Sub-Saharan Africa, for example. Many analyses modities, the effects can be critical, as in the case of with world general equilibrium models consider cotton for the rural poor in a number of African only an aggregate household per country and pre- countries. Minot and Daniels (2003), using house- sent the results mainly in terms of overall national hold survey data, find that a drop of world cotton 52 Agriculture and Achieving the Millennium Development Goals prices by 20 percent, as might be due to developed- what to do about the agricultural trade barriers in country subsidies, raises poverty in cotton export- developing countries. Developing countries are dependent Benin by 4 percentage points (an not homogenous with regards to agriculture liber- increase of 10 percent of the population under the alization. They have different trade specializations poverty line) through direct and indirect effects on (different export products facing varied degrees rural incomes. of protection or support in developed countries); Beyond the effects of tariff and farm-support some are net food importers, and others are faced policies captured in models as described, the fastest with different trade barriers. Those countries with growing world agricultural markets are for fruits strong agricultural export potential have called for and vegetables, livestock products, and other high- more open markets, but those fearful of negative value commodities. For these products, regula- effects on their poor farmers have been reluctant to tions related to safety and quality play a large role in endorse such moves. Many development advocates determining trade opportunities. The WTO embod- are adamant that developing countries be granted ies agreements to discipline agricultural and food room to retain agricultural trade barriers. safety and quality regulatory decisions that are What is at stake in the reduction of agricultural primarily sovereign prerogatives. These WTO disci- protection among developing countries is some- plines call broadly for countries to achieve legiti- what different than the stakes from developed- mate regulatory goals in the least trade-distorting country reforms. When developing countries join manner. Effectiveness of these disciplines is an in agricultural trade liberalization, reducing high important aspect of a rules-based agricultural trade levels of trade restrictions, relaxing quantitative system (Josling, Roberts, and Orden 2004). restrictions on imports, and lowering import tariffs, Developing countries have a lot at stake in the they can achieve overall welfare gains of $19.9 bil- area of food regulation. Production of high-value lion annually, according to Diao, Diaz-Bonilla, and products is potentially a source of higher incomes Robinson (2003). This is twice the gains in national among the rural poor. But stringent developed- welfare compared to reforms in the developed country regulatory measures to address health, countries only. The additional overall gain is pri- safety, and quality goals can close off market oppor- marily due to consumers facing lower internal tunities. There is a trend in quality regulation prices as their countries' own trade barriers are toward the required use of certain production meth- reduced. Developing-country trade policy reforms ods or required labeling of production and process- add an additional $14.9 billion annually to aggre- ing attributes. These process-focused measures gated agricultural exports of developing countries, often demand complex conformity assessment with but do not increase their aggregated agricultural high compliance costs. Innovations are needed to production value added in the Diao, Diaz-Bonilla, ensure efficient implementation of such measures. and Robinson model. There is a great need to build up the capacity of These results suggest that trade policy reforms developing countries to produce up to the exact- among developing countries boosts their over- ing standards of importing markets. And there is a all welfare but also creates distributional impacts challenge to the MDG: agricultural and food regu- among those developing countries that are best able lations that are well-intentioned in some dimen- to gain from trade openness versus those less able sion can have the undesirable effect of reducing to do so. These reforms also create distributional income-earning opportunities or blocking technol- effects between food consumers and producers ogy adoption that would benefit the poor. within (poor) countries. There is a need to under- stand these distributional effects better. However, the overall welfare benefits are important as well; Policies of Developing Countries they extend to poor food consumers, including There is growing agricultural trade among devel- poor farmers and landless rural workers who use oping countries, but these countries also retain nonfarm income to buy food. substantial trade barriers on agricultural products. Three leading empirical studies that focus on Developing-country governments (and civil society Africa's growth performance--Easterly and Levine organizations) that are largely united in seeking (1997), Rodrik (1998), and Sachs and Warner lower agricultural subsidies and protection in the (1997)--conclude that the region performed poorly developed countries have been divided concerning after the reforms of the 1980s, even though the Global trends and Emerging Issues in Meeting the MDGs 53 reasons provided for the poor performance vary. Modern food aid emerged after World War II, Easterly and Levine emphasize the role of ethnic particularly with the United States P.L. 480, the fragmentation and poor quality of institutions, Agricultural Trade Development and Assistance while Sachs and Warner stress poor trade policies Act, of 1954. This law asserted multiple goals for and geography. Rodrik reports that, contrary to food aid. Food aid would combat world hunger Sachs and Warner, the fundamentals for long-term and malnutrition and promote agricultural develop- growth are human resources, physical infrastruc- ment, but it would also expand trade and develop ture, macroeconomic stability, and the rule of law. export markets for U.S. agricultural commodities. In an overall context, a successful conclusion for Food aid from the United States peaked in the 1960s, agriculture in the WTO Doha Development Round but the country has remained the largest single trade negotiations can be understood to make an donor of food aid, accounting for about 55 percent important contribution to achieving the MDGs by of the total during the 1990s. Food aid now accounts establishing sustainable positive incentives for agri- for less than 5 percent of global trade in agricul- cultural production among developing countries. A tural and food products. Yet many controversies July 30, 2004 WTO framework agreement to guide surround the use of food aid, either for emergency further negotiations came at the last possible hour crisis relief or as an instrument of humanitarian to avoid a collapse of the Doha round. For agri- and development policy. Given relatively fixed or culture, it narrowed the negotiating field some- slowly changing budgets, there is a built-in cycle of what, but still leaves wide latitude about how food aid: the volume of food aid available will be much trade will be opened under an agreement's lowest when commodity prices are highest (and final terms. Because of the way WTO disciplines need is, in that sense, greatest). Moreover, there is on tariffs and subsidies are defined and adminis- much evidence that the provision of food aid is tered, agreed reductions in protection and domes- subject to political pressures related to supporting tic farm support would have to be quite big to world commodity prices and other objectives. have much effect on trade (Anderson, Martin, and Food aid falls under emergency aid, project aid, van der Mensbrugghe 2004). or program aid. Emergency aid, which has been In summary, an increase in the share of national increasing its share in total food aid recently, is a income that is exported, or lower importing prices, response to natural disasters or conflicts that leave does not in itself generate growth in per capita vulnerable populations at risk of starvation or income and will not on its own set an economy on severe malnutrition. Project aid is associated with a sustained growth path unless it is accompanied the development of specific food security or devel- by other structural reforms.11 However, excessive opment projects, such as school feeding programs. levels of export taxation and import restrictions Program aid is the most general use of food as a can contribute to the relative decline in growth of form of foreign assistance, essentially providing the some countries. In addition, to achieve some of the monetized value of the food as a resource for use by positive gains, developing countries that will ben- a developing-country government, although some- efit from more open markets abroad need to par- times with conditionality requirements about how ticipate in multilateral agricultural trade policy this aid is utilized. reforms. Benefits for poor farmers in countries less Food aid can be procured in the donor country, able to compete as trade barriers come down will local markets of the recipient country, or from third- come not from multilateral trade policy reform country sources. It can be provided through bilat- itself, but from complementary investments and eral or multilateral channels, and these channels can policy improvements. encompass governments, multilateral agencies such as the United Nations's World Food Programme (WFP), and nongovernmental organizations. The Policies in Development Assistance: U.S. P.L. 480 requires the use of commodities pro- The Case of Food Aid duced in the United States; in 2002, for example, Food aid is another component of international nearly 90 percent of the total 9.6 million tons of transactions that directly and indirectly affects rural food aid delivered worldwide was procured in the poverty in a globalized agricultural economy and donor countries. Contributions in financial terms by therefore could have a significant impact in achiev- donors, rather than in the form of food commodi- ing the MDG targets. ties, allows greater flexibility in providing food 54 Agriculture and Achieving the Millennium Development Goals aid, which can be sourced in the donor country or ment assistance, but likewise that donors would elsewhere using the aid financing. Local or third- probably not provide equivalent cash development country (triangularization) purchases accounted for assistance in place of food if existing food-aid pro- a higher percentage of the food aid delivered by grams were terminated (Hoddinott, Cohen, and the European Commission's multilateral food aid Bos 2003). Thus, attention has focused on how its program (70 percent), when food aid is provided for effectiveness can be maximized and its potential emergency purchases (67 percent), and of food harms mitigated (Barrett and Maxwell 2004; von deliveries through the WFP (which accounted for Braun 2003). nearly 40 percent of total food aid in 2002, nearly 60 percent of which was procured in recipient or third countries) (Hoddinott, Cohen, and Bos 2003). Macroeconomic Reforms and National Once food aid is provided to recipient agencies, Government Infrastructure Investments it can either be delivered directly as food to targeted For the last two decades developing countries populations or "monetized" through sale in recipi- have experienced several changes in their macro- ent country markets. When monetized, the cash economic policies. Macroeconomic policies changed value of the food aid becomes a resource that the from development strategies aimed at the domes- agency can use to support various activities. Even tic market, with strong public-sector intervention, in the case of food aid for emergency purposes, to more outward-looking strategies with open- some of the food can be sold to cover nonfood costs ness to trade, deregulation of markets, more fis- of aid delivery. In the cases of project and program cal control, and the private sector as the main agent aid, a great deal of monetization occurs. There is for development. therefore concern that food aid can depress incen- Although it was expected that these new policies tives for local food production. But well directed would lead to an increase in growth, the results are food aid, or aid programs financed by monetization not clear and vary among regions. Results derived of food aid, can also provide essential emergency from cross-country comparisons may be indica- relief and reduce vulnerability of the poor to short- tive but are not very conclusive. Such comparisons term shocks that undermine their long-term human involve differences in the experience of each of the and physical assets (Barrett and Maxwell 2004; individual countries, not only in terms of the effects Hoddinott, Cohen, and Bos 2003). of the policy reform but also of the initial conditions Several international institutions provide guide- at the initiation of reforms and the timing and con- lines for food aid.12 The current rules exempt "bona sistency of them. However, many experts agree that fide food aid" from restrictions on export sub- there is a need to maintain macroeconomic bal- sidies, and the nonbinding 1994 WTO Marrakesh ances, to further trade and financial openness, to Ministerial Decision recommends increased food restrict activities of the state, and to deregulate aid as a means to help developing countries. But markets (Devarajan and others 2000; Edwards as direct export subsidies allowed under the Agree- 1995; Stiglitz 2000). ment on Agriculture are subject to increasingly dis- Similarly, agricultural policies changed signif- ciplined negotiation (and are even being phased icantly in the 1990s. An example is shown in table out), indirect forms of subsidization, including some 4.1 for Latin America and the Caribbean. Serious uses of food aid, are also under scrutiny. The July concerns have been raised regarding the effects of 2004 framework agreement calls for elimination "by policies on the agriculture sector and on the more a credible end date" of export subsidies and "paral- vulnerable parts of the population. Demand con- lel" elimination of "all export measures with equiv- traction, the abolition or reduced role of parastatals, alent effect." For food aid, the agreement calls for significant reductions in public expenditures, and elimination of "provision of food aid that is not in the withdrawal of state interventions--combined conformity with operationally effective disciplines with rigidities in resource mobility, slow response to be agreed. The objective of such disciplines will of the private sector, and lack of appropriate infra- be to prevent commercial displacement." structure and institutions--have resulted in an Food aid remains the subject of ongoing contro- increase of unemployment and therefore an increase versies in the context of the issues described above. in inequality between and within rural and urban It is widely recognized that provision of aid in the areas. Reductions in public investment outlays were form of food is not the optimal form for develop- not (fully) compensated by private investments, Global trends and Emerging Issues in Meeting the MDGs 55 Table 4.1 Latin America and the Caribbean: Main Features of Changes in Agricultural Policies Policies in the 1950s to 1980s Present Policies Strong state centralism Political, institutional, and financial decentralization aimed at serving local projects State planning and financing of "public" goods and services Demand-oriented approach, with cofinancing of investments Concern for the food security of the country, with research in infrastructure and services therefore focused on increasing production and yields Demand-oriented approach, with cofinancing of research of staple foods (rice, wheat, maize, potatoes, beans, both in public research institutes and universities lentils, etc.). Free technical assistance, with programs prepared by state Paid private technical assistance in response to demand bodies with little or no user participation from producers, subject to presentation of a project for competition when subsidies are involved Subsidized credit through special lines of credit for the sector Elimination of specific lines of credit for the agricultural sector, of development banks, and of the obligation of private banks to loan a certain proportions of their portfolio to the sector Labor reforms leading to the monetization of the rural econ- Deregulation of the labor market, although in the case of omy and the gradual disappearance of sharecropping- agriculture the rurales had always been less strict and type systems, with their labor force being replaced by more difficult to enforce permanent or, most often, temporary wage laborers Price controls for staple foodstuffs in order to check infla- Freedom of prices tion and keep minimum wages low so as to encourage industrialization Exchange rates favorable to imports In principle, equilibrium in exchange rates Tariff rates that were generally high but covered a wide Reduction of tariff rates and application of a uniform rate, range of rates albeit with some exceptions, especially for agricultural products Taxes on agricultural exports Elimination of taxes and other hindrances to exports Agrarian reform programs aimed at reducing the inefficiency Replacement of agrarian programs with purchase through of highly heterogeneous production systems (with abundant the land markets, aided by loans or subsidies for small land and capital for a few agents and abundant labor producers; promotion of formal proof of ownership in the case of the rest) and above all at reducing the through the issue of official property titles possibility of social disorder in rural areas Source: Dirven (1999). and this of course affected expenditures in infra- Importance of Rural Infrastructure structure, health and education services, and social programs. There is an increasing consensus that providing ade- Infrastructure is of particular concern as one of quate infrastructure is an important step in the the key inputs entering into the "production func- process of poverty alleviation and in providing a tion" of the MDGs and the achievements of many more equitable set of opportunities for all citizens. of the MDG targets, from poverty reduction to envi- Several authors have studied the aggregate-level ronmental sustainability targets.13 For instance, in links between poverty and rural capital-intensive Sub-Saharan Africa, where less than half of the pop- infrastructure, including Jimenez (1995), Lipton and ulation has access to safe drinking water (Fishbein Ravallion (1995), and Van de Walle (1996). In partic- 2001) (see also table 3.10), child mortality may ular, the literature on specific infrastructure compo- depend on the availability of clean water (Galiani, nents, such as the role of rural roads, telephones, Gertler, and Schargrodsky 2005), and attainment of or access to electricity on poverty alleviation is universal primary education for girls may also cru- very broad, including Binswanger, Khandker, and cially depend on access to piped water (Leipziger Rosenzweig (1993), Howe and Richards (1984), and others 2003). Jacoby (1998), and Lebo and Schelling (2001). 56 Agriculture and Achieving the Millennium Development Goals Recently Renkow, Hallstrom, and Karanja (2002) and electricity production. Increases in population estimated the fixed transaction costs (those not moved each group to the left by reducing its ratio dependent on commercialized volume) that impede of land to labor, and each group of countries also access to product markets for subsistence farmers moved upwards because of the expansion to access in Kenya. The authors estimate that high transaction in each of the infrastructures. It is obvious that the costs are equivalent to a value-added tax of approx- size of movement for Sub-Saharan Africa was sig- imately 15 percent, illustrating the opportunities to nificantly smaller than those for other country raise producer welfare with effective infrastructure groups. The major reason for this significant lag has investments. Smith and others (2001) show that, normally been attributed to geography (Diseases, for the case of Uganda, the rehabilitation of roads internal distance, and low populated areas were a increases labor opportunities in the service sector. much bigger obstacle.) and to the poor initial con- Moreover, based on an infrastructure index that dition of infrastructure in Africa. includes road, rail, and telecommunications density, Unlike Asia and Latin America, Sub-Saharan Limão and Venables (1999) found that infrastructure Africa inherited a highly dispersed and unevenly is a significant and quantitatively important deter- distributed infrastructure from its colonial past. minant of bilateral trade flows. Improving destina- There was little improvement of infrastructure, if tion infrastructure by 1 standard deviation reduces any, during the colonial era, and "in some impor- transport costs by an amount equivalent to a reduc- tant respects, it can even be said that colonial policy tion of 6,500 kilometers at sea or 1,000 kilometers of reinforced the handicaps of SSA [Sub-Saharan overland travel. According to their findings, most of Africa]" (Platteau 1996, p. 200). The limited infra- Africa's poor trade performance can be accounted structure that was built during that era was driven for by poor infrastructure. Moreover, lack of ade- by the objective of connecting natural resources to quate infrastructure in much of Sub-Saharan Africa export markets. For instance, "two-thirds of the impedes more productive agriculture. In Norman African railways built in the colonial period con- Borlaug's words, "an effective system to deliver nected mines to a coastal harbor" (Platteau 1996, modern inputs--seeds, fertilizers, crop-protection p. 200). The rest of the continent was virtually chemicals--and market output must be established. ignored and "only the Union of South Africa with If this is done, subsistence farmers, who constitute mass immigration of Europeans had more than six more than 70 percent of the population in most meters of railways per square kilometer in 1970, countries there [Sub-Saharan Africa], can have a and six countries had no railways at all" (Boserup chance to feed their people." (Borlaug 1999) 1981, p. 148). The skewed distribution of infrastruc- In addition, and as argued by Leipziger and ture was perpetuated even after independence. others (2003), achieving the health and education If structural impediments predominate in agri- MDGs will require more than health and educa- culture as in the case of Africa, it will be difficult to tion interventions; in particular, infrastructure ser- achieve sustained growth in production by price vices have a crucial role to play. In health, there incentives alone unless countries develop serious is clear evidence that better access to basic hard- strategies to reduce these impediments. In a recent infrastructure services has an import role to play in study, Fay and Yepes (2003) predicted the demand improving child health outcomes and therefore in for roads, railroads, telecommunications, electricity, complying with the three health-related MDGs. water, and sanitation. According to their estimates, For example, Jalan and Ravallion (2001) showed in order to meet the predicted demand, the coun- the importance of piped water to reduce diarrhea in tries of Sub-Saharan Africa will need to invest young Indian children; and Galiani, Gertler, and around $25.9 billion annually between 2005 and Schargrodsky (2005) found that child mortality fell 2010. Of this sum, $12.6 billion will be needed for 5­9 percent in areas that privatized their water ser- maintenance of the existing infrastructure and the vices. Brenneman and Kerf (2002) showed that elec- rest to build new infrastructure. This would require tricity allows for more hours of studying and road an annual investment of more than 5.5 percent of access promotes easier establishment of schools GDP. Wood (2002), taking into account the low pop- and higher attendance. ulation density of Africa, predicted that Africa will Sub-Saharan Africa seems to particularly lag need to invest at least twice as much of its GDP in behind in infrastructure investments, as is shown infrastructure as does low-income Asia, as well in figure 4.1 for paved roads, telephone lines, as meet higher recurrent charges for operation and Global trends and Emerging Issues in Meeting the MDGs 57 Figure 4.1 Changes in Paved Roads, Telephone Lines, and Electricity Production Over Time, Selected Regions Paved roads 1990­2000 4.5 4.0 OEC area) 3.5 SAS land 3.0 km 2.5 sq 2.0 1.5 (km/10 1.0 Road LAC 0.5 SSA 0 0 2 4 6 8 10 12 14 Land area (per 100 labor force) Telephone lines 1980­2000 700 OEC 600 population) 500 1,000 400 (per 300 lines 200 LAC telephone 100 SAS SSA Main 0 0 2 4 6 8 10 12 14 16 18 Land area (sq km/100 labor force) Electricity production 1980­2000 12,000 OEC 10,000 inhabitant) per 8,000 (kWh 6,000 4,000 production LAC 2,000 SAS SSA Electricity 0 0 2 4 6 8 10 12 14 16 18 Land area (sq km/100 labor force) Note: LAC = Latin America & Caribbean; OECD; SA = South Asia; SSA = Sub-Saharan Africa. Source: Torero and Chowdhury (2004). 58 Agriculture and Achieving the Millennium Development Goals maintenance. Therefore, strategies for infrastruc- demand driven. At present, the estimation of rural ture development are needed to be able to move infrastructure investment is generally based on the toward the required trends to reach the MDGs. needs assessed for each sector at the national level, with little or no assessment of demand and coordi- nation at the local level where the service ultimately On Public Investment and Infrastructure will be provided. More often than not, such invest- Even among countries that have advanced the most ment assessments do not reflect the preferences of in market-oriented reforms, as mentioned earlier in users of services and the contingencies of ser- this chapter, there is still a significant infrastructure vices. For instance, demand for secondary school- access gap, and reforms alone are not enough to ing may be contingent on access roads, and failing provide complete access in remote poor rural areas. to coordinate these two may result in a mismatch Some sort of public intervention is needed to close between availability of a service and its actual use. this gap and therefore to achieve the MDGs. Where Furthermore, it is important for communities to be the government believes that service should be pro- informed about the technology they want to use, vided beyond what a well-functioning market the service level they require, and especially to have will offer, subsidies may be justified to promote a clear understanding of long-term costs and main- additional investment to achieve these govern- tenance implications, so that communities can ment goals. But the government should also seek choose what is most appropriate for them under to improve the functioning of the market so that their budget constraints. There is evidence that if subsidies can provide a maximum benefit when provided with appropriate information and tech- and where they are needed. In particular, a failure nical support, communities can make informed to address the impediments to efficient working of choices about service options as well as clearly the market in rural areas through regulatory identify their willingness to pay, thereby assuming reforms will reduce the availability and effective- ownership and responsibility for the infrastruc- ness of resources to address the real access gap in ture (Brenneman and Kerf 2002; Estache, Foster, these areas. and Wodon 2001, 2002; Torero, Chowdhury, and Moreover, the potential best practice to reduce Galdo 2003). the access gap largely depends on the institutional Finally, public sector intervention alone is not framework existing in each country. Countries with enough; forging private-public partnerships seems sound regulatory institutions and legal frameworks to be the most efficient way of closing the access can adopt some solutions that will be out of reach gap in all services covered as we discuss later in for countries with weak institutions. However, this chapter. Public intervention alone is usually there could be institutional solutions adequate to not cost effective, and isolated private initiatives reduce the access gap while simultaneously devel- also fail to deliver all services. Despite the rise in oping the legal, institutional, and regulatory frame- private-sector involvement in infrastructure pro- work needed to advance different strategies. vision, the overall investment levels, particularly There are two other issues that concern public concerning small subsistence farmers, are far from investment on infrastructure. The first issue is the adequate. The need for the public sector to play a lack of coordination in public investments at the facilitating role has not been addressed. Since there country, regional, and donor levels. Where the link- is little evidence that rural infrastructure is com- ages and complementarities of infrastructure invest- mercially viable on a stand-alone basis, the role of ment have not been realized, it is common to find the public sector needs to be reinvented. fragmented approaches, lacking sufficient attention for substantive policies and development issues. In fact, in many cases, access to infrastructure has THE ROLE OF THE PRIVATE not been linked to poverty alleviation strategies SECTOR IN SUPPORTING or to the general development goals of countries. AGRICULTURE TO MEET Therefore, it is necessary to take an integrated THE MDGS approach, even if the actual design may vary from country to country. Public intervention alone is not sufficient to de- Second, investment in infrastructure in these liver the services and investments required to areas is done from top to bottom rather than being achieve the MDGs as outlined in the MDG scenario Global trends and Emerging Issues in Meeting the MDGs 59 described in chapter 3. Thus, the role of the private particularly important role to play in the provision sector for providing investments and delivering of rural financial services, while the public sector services is receiving increased attention in the agri- should provide an appropriate enabling environ- culture as well as the rural nonfarm sector. Ideally, ment for the private sector to operate efficiently. In the public and private sectors complement each particular, the public sector should facilitate access other, with the government providing an appropri- to broader financing options by continuing devel- ate enabling environment for private initiatives opment of domestic financial markets coupled with to develop. Good governance principles for both skill building for regulators and private financial public- and private-service provision are crucial to institutions. achieve both effectiveness and efficiency in resource The Bank Rakyat Indonesia (BRI) system has allocation. The following section describes how the shown that microfinance services can be provided private sector can support agriculture's role in to low-income and poor households at fully com- achieving the MDGs. mercial terms by a large national commercial bank, To alleviate rural poverty in developing coun- thus combining maximum outreach and sustain- tries, the private sector can contribute to economic ability. The BRI Unit Desa (Village Bank) system, growth through creating jobs both on and off farm. which has developed into the largest internation- The private sector can also empower the poor by ally acknowledged successful microfinance oper- providing a broad range of products and services ation in the developing world, reaches almost at lower prices (UNDP 2004b). Large local compa- 3 million borrowers and nearly 28 million small nies and multinational corporations can channel savers. As BRI Unit Desa practice shows, savings private initiatives into development efforts. This and credit interest rates are not among the most could be done, for example, through the expansion important considerations for micro-clients. The of public-private partnerships in the fields of edu- massive amount of rural savings mobilized by the cation, such as with vocational schools; in the retail unit desas demonstrates the demand for attractive sector, through contractual arrangements drawing savings products and the capacity of rural people to in small-scale farmers; and, generally, through the save. The BRI experience further demonstrates that diffusion of knowledge and information to small- policies and institutions can be designed to achieve and medium-scale enterprises working on forward high levels of outreach, serve the very poor, and and backward linkages of agriculture. The private attain financial and institutional sustainability using sector is also important in the finance sector, where an individual lending technology. The unit desas it can provide resources and relationships that can reach poor clients by using an individual loan tech- help establish these enterprises. nology requiring collateral, but group loans may Rural development that helps achieve the MDGs be useful for reaching even poorer clients. Among creates the demand for timely and efficient credit the key factors for BRI Unit Desa success are availability to support private investments in user-friendly products and services priced for insti- agriculture and related services. As development tutional viability; close and regular relationships proceeds, rural areas will generate large capital with micro-customers; convenient bank locations, surpluses that need to be captured and managed simple loan procedures, quick processes, flexible efficiently by the financial sector. Appropriate terms; offering savers a combination of security, development of the rural financial sector can also convenience, liquidity, confidentiality, service, and reduce household vulnerability through promotion return; simple management information systems of savings mobilization. Efficient rural financial and transparent accounting and reporting systems; markets that serve the full range of financial needs and close loan monitoring and incentives for repay- of farmers and nonfarm enterprises are usually ments (SEARCA/IFPRI/CRESCENT 2004; Meyer more important than targeted or subsidized credit and Nagarajan 1999). programs. Such programs should be demand In Africa the private sector, together with public- driven, not donor or supply-side oriented. These sector support, is crucial for expediting the rate of financial programs should provide a broad range market development; removing or reducing barriers of financial services, not just savings mobilization; to market access, both by providing special sup- emphasize viable and sustainable financial inter- port for markets that are slow to develop and easing mediation based on market interest rates; and pro- market participation by poorer producers; and vide credit for all purposes. The private sector has a establishing a more equitable set of market relations 60 Agriculture and Achieving the Millennium Development Goals between (small-scale) producers and market inter- nonfarm sources (Rosegrant and Hazell 2000). mediaries (NEPAD 2002). The following activities Nonfarm shares of income are strongly and neg- are suggested to further market development. atively related to farm size. Low-investment manu- facturing and services--including weaving, pottery, · The government can develop the skills and gathering, food preparation and processing, domes- organization for smallholders, for example, tic and personal services, and unskilled nonfarm through the promotion of producer groups wage labor--typically account for a greater share of or associations. income for the rural poor than for wealthier rural · The government can help the private sector to residents (Hazell and Haggblade 1993). The reverse develop and broaden its outreach. A greater is true of transport, commerce, and such manufac- private-sector presence will then provide more turing activities as milling and metal fabrication, competitive and efficient services to small- which require sizable investments. holders, particularly for input supplies, pro- As noted, effective financial institutions are duce, marketing, and agroprocessing. This required in order to promote trade, commerce, approach is also vital to support the develop- and manufacturing in the rural nonfarm economy. ment of micro-enterprises. Microfinance programs to help women and poor · Both the government and the private sector people develop nonfarm enterprises may contribute can finance the provision of essential con- to poverty alleviation. Microfinance also needs to necting infrastructure, such as market access be accompanied by appropriate training programs roads, communications, and price and mar- to give women and poor people the skills they need ket information. to compete in the market. · Most important, the private sector can assist Rural people need adequate training if they are smallholders in promoting a dialogue between to obtain relevant technical, entrepreneurial, and the main stakeholders to generate the policy, management skills. The rural nonfarm economy institutional, and legal context required for provides much of its own training through appren- enhanced market linkages. ticeship schemes and on-the-job learning, but in an increasingly technical and communications- oriented world, specialized training schemes (in The Rural Nonfarm Private Sector computing and accounting, for example) are The rural nonfarm sector is an important engine needed, including programs for women, who in economic development. During successful rural dominate many service and trading activities. economic growth, the emergence and rapid expan- In much of the developing world, apart from sion of the (mainly private-sector) nonfarm econ- some manufacturing activities, the rural nonfarm omy in rural areas and the towns they serve is a sector was largely ignored by policymakers until major source of growth in incomes and employ- recently. Furthermore, because it depended heavily ment. From a relatively minor sector, often largely on agriculture either directly or indirectly for much part-time and subsistence-oriented in the early of its demand, the rural nonfarm sector also suf- stages of development, the rural nonfarm economy fered as a result of macroeconomic policies that dis- can develop to become a major motor of economic criminated against the agricultural sector. Recent growth in its own right, not only for the country- macroeconomic policy reforms that have benefited side but for the economy as a whole. Its growth the agricultural sector should, therefore, have led to also has important implications for the welfare of positive growth-multiplier benefits for the rural women and poor households, sometimes helping nonfarm economy. The policy reforms have also to offset inequities that can arise within the agri- favored tradable-goods production in general, and cultural sector. In Asia, for example, the rural non- this should have been directly beneficial to much farm economy accounts for 20­50 percent of total rural industry. However, these benefits for the rural rural employment and 30­60 percent of total rural nonfarm sector are often limited by a continuing income. The rural nonfarm economy is especially bias toward capital-intensive industry at the ex- important to the rural poor. Landless and near- pense of trade and services. In many cases, small landless households everywhere depend on non- firms have effectively been placed at a competitive farm earnings; those with less than 0.5 hectare disadvantage to their larger rivals (for example, they earn as much as 90 percent of their income from do not receive the same subsidies and tax benefits), Global trends and Emerging Issues in Meeting the MDGs 61 and this has encouraged a more capital-intensive Private-Public Partnerships and Investment pattern of development than is optimal. Synergies: The Case of Water In spite of the constraints on the role of the pri- and Agricultural Research vate sector, development of this sector is essential to How should investments in agriculture and related progress in meeting the MDGs. We now examine the role of the public sector in enhancing the per- sectors be supported to achieve the MDGs? Public- formance of the private sector through provision of private partnerships are an important way to appropriate regulation and oversight, setting fair increase financial, human, and social capital in ground rules for the private sector, and through the rural areas. Partnerships include publicly provided provision of public goods. training for small- and medium-scale enterprises, partnerships in education, agricultural research, the provision of information and communication Effective Legal, Regulatory, technologies, the expansion of rural infrastruc- and Institutional Environments ture including roads, and the development of rural industrial clusters. Because public funds are Effective legal, regulatory, (for example, to secure property rights and enforce contracts) and institu- increasingly scarce, private-public partnerships are tional (for example, financial) environments are one important means to enhance investments in required to promote agricultural trade, commerce, rural areas. and manufacturing. At the farm level, effective reg- Private-public partnerships are considered par- ulatory systems need to be provided, particularly ticularly important to close the access gap in infra- in the areas of standards and quality control, to structure and social services. Public intervention ensure future competitiveness of farmers both in alone is usually not cost-effective, and isolated pri- the emerging supermarkets nationally and inter- vate initiatives also tend to fail to deliver all ser- nationally in the global markets, which demand vices. Despite the rise in private-sector involvement increasing labeling, standards, and quality control. in infrastructure provision, overall investment lev- At the firm level, in order to ensure the long- els are particularly far from adequate. According term viability of the private sector in development, to the Commission on Private Sector and Dev- the creation of an enabling environment is required elopment (UNDP 2004b), private-sector manage- that includes strong accounting and disclosure ment of small-scale energy production and water standards, bankruptcy and foreclosure processes, supply projects can be particularly effective. These and improved taxation and accounting laws. projects allow management by smaller domestic Garnaut (1998) summarizes the objectives of pru- companies whose final product is delivered to the dential regulation: accuracy, honesty, and trans- rural customer or informal sector. Decentralized parency in financial reporting; the avoidance of power production through distributed energy of related-party and other noncommercial transac- various kinds can also be contracted to the private tions; and the maintenance of relatively high capital- sector through agreements with the public-sector to-asset ratios--higher than the norm for developed grid. Examples are solar power and small hydro- countries. Prudent regulation should focus on sim- power plants. ple, enforceable targets, rather than attempting In particular, water resources have been recog- to regulate a wide range of indicators (Fane 1998). nized as under serious threat with growing national, Corporate governance also needs long-term reform, regional, and seasonal scarcities (Rosegrant, Cai, including, for example, monitoring of enterprises and Cline 2002). Over 1 billion people lack access by commercial banks. Enhanced disclosure and to improved water services. Worse, another 2.4 bil- accounting practices and strengthened enforcement lion people still live without improved water sani- of corporate governance regulations, especially tation. The need for investment, particularly from as they relate to capital markets, are also essen- the private sector, is obvious. To facilitate invest- tial components of corporate governance reform. ments, the World Bank, through the Multilateral Institutions need to be strengthened so that analysis Investment Guarantee Agency (MIGA), is engaged of corporate financing and monitoring of firm per- in public-private arrangements for the provision of formance and behavior are comprehensively cov- water and wastewater. However, the lack of suffi- ered by a combination of private, semipublic, and cient public funds and difficulties in attracting public organizations (World Bank 1998). private money in the sector makes achieving the 62 Agriculture and Achieving the Millennium Development Goals MDGs more difficult. There are a number of models hands of the public to ensure that the rights of the for public-private investment partnerships, rang- public are protected. ing from management contract to build-operate- The private tubewell revolution in South Asia transfer contracts (figure 4.2); their applicability since the 1980s is an excellent example of effective and success depend on the local conditions and private investment and private-public investment development status. synergies. Public investments have been an impor- Both public and private companies have been tant facilitator of private irrigation investment. able to effectively provide water service when they Private tubewells in South Asia have grown most are operated properly and certain guidelines are rapidly in areas with reasonably good roads, re- taken into consideration. Gleick and others (2002) search and extension systems, access to electric or spell out several principles that need to be consid- diesel energy, and access to credit; therefore, the ered for private water provision. Many of these are wells have been concentrated in and around the key elements of effective public water systems as command areas of large, publicly developed sur- well. These principles include ensuring that all face irrigation systems. Seckler (1990) notes three basic human and ecosystem needs for water are reasons for the complementarity between public met, with a basic amount of water provided at a and private irrigation investment in South Asia: subsidized rate for those too poor to afford to deep percolation losses from the surface systems pay; provision of water at reasonable rates, with recharge the aquifers for tubewells; tubewells are increases in price linked to improved service when often used together with surface irrigation water, possible; subsidies that are economically and so- which lowers pumping costs and concentrates cially sound; and new water projects that are these costs in periods of the highest marginal permitted only when shown to be less costly than returns; and the tubewells ride piggyback on the efficiency improvements. It is also crucial that infrastructure created for the surface systems. there be significant government oversight, with The role and structure of agricultural research government regulation of water quality, clear con- are changing over time. The relative importance of tracts that spell out the role of all parties, clear rules productivity-driven growth will increase, because for dispute resolution, and clarity and stakeholder growth in input use is declining as many regions in involvement during the negotiation process. Finally, Asia are reaching high levels of input use. Private the ownership of the water should remain in the investment in agricultural research--which gener- Figure 4.2 Sample of Public-Private Investment Arrangements PUBLIC-PRIVATE ARRANGEMENTS FOR THE PROVISION OF WATER AND WASTEWATER SERVICES PUBLIC PRIVATE Management Lease Build-operate-transfer Privatization/concession contracts contracts arrangements arrangements Public sector owns the system Public sector owns the system Private sector finances and Private sector owns and and makes final investment and is responsible for develops the project and operates the system. and funding decisions. Private investments and tariff setting. manages the utility until sector manages the utility and Private sector manages utility, transfer of ownership to the is paid on a profit-sharing collects bills, and pays an public sector. basis. operator's fee. CAPITAL COMMITMENT Source: MIGA (2004). Global trends and Emerging Issues in Meeting the MDGs 63 ates significant public benefits--will increase in New agricultural technologies in Asia--such as importance, if policy reforms continue to create technologies to improve pest management and the and improve the incentives for private investments nutrient balance and the timing and placement of by eliminating price distortions and strengthening fertilizer applications--are increasingly complex, property rights. Market failures and social objec- knowledge intensive, and location specific; they tives will continue to call for an important role for demand continued investment to create a better public investment in agricultural research, however. and more decentralized research and extension sys- Through partnerships, public research institu- tem. Because new technologies are more demand- tions can gain access to advanced scientific knowl- ing for both the farmer and the extension agent, edge and technologies held by the private sector, they require more information and skills for suc- mechanisms for developing, processing, marketing, cessful adoption compared to the initial adoption of and distributing final products to farmers and con- modern varieties and fertilizers. Decentralization of sumers, and financial resources that are increas- existing extension services structures that encourage ingly difficult to obtain. There are few examples, a bottom-up flow from farmers to extension and however, of successful public-private partnerships research could also help farmers cope with the addi- in agricultural research. Part of the problem stems tional complexity of efficiency-enhancing tech- from the sectors' inherently different research objec- nology. Bottom-up information flows, combined tives: while public institutions conduct research with adaptive, location-specific research, are according to their broad social mandates, firms particularly important in the transfer of complex pursue more narrow, profit-maximization goals. crop-management technologies. Other modern Furthermore, public institutions and private firms technologies, such as commercial poultry technol- compete over the ownership and use of proprietary ogy, will be transferred essentially intact from scientific knowledge and technologies, over scarce developed countries, without local adaptation, but financial resources for research, and over markets, will similarly require higher levels of education clients, and beneficiaries for their outputs. This and management skills than traditional livestock competition incurs real and hidden costs and operations. Finally, the increasing importance of risks that make partnerships difficult to create new, knowledge-intensive technology requires a or sustain--costs and risks that are exacerbated by market-friendly environment for the adoption and deep-rooted misperceptions and information gaps adaptation of new technologies and the removal between the sectors. To succeed, public-private of restrictions on technology imports, which must partnerships require that both parties identify be encouraged through continued progress in eco- research areas where their objectives are compati- nomic liberalization. Privatization of extension ble, and where both parties are willing to engage in through contracting to private companies can intro- detailed and often difficult negotiations for project duce incentives for higher efficiency, and has been planning and implementation (Spielman and von applied by seed companies, including for hybrid Grebmer 2004). maize. Privatization of extension is likely to be suc- Agricultural research is often long term, large cessful when extension is linked to the delivery of a scale, and risky, and while the returns to new specific technology (such as hybrid maize or poul- technologies are often high, the firm responsible try) and to larger, more homogeneous groups of for developing the technology may not be able to farmers. For commodities where private extension appropriate the benefits accruing to the innova- services cannot be self-supported, the govern- tion--as in the case of improved open-pollinated ment needs to continue providing assistance and rice and wheat varieties. The benefits of agricul- training. tural research often accrue to consumers (through In contrast to other developing regions of the reduction in commodity prices due to increased world, the private sector in Africa is not increas- supply), rather than to the adopters of the new ing its research efforts as government spending technology, so social returns may be greater than declines. Given a share of 2 percent in total spend- private returns to research. Therefore, a sustained ing, the private sector plays an exceptionally small public role in funding agricultural research will be role in funding agricultural research in the region. essential, particularly for crops and regions in less Increasing their contribution is highly unlikely favorable environments, which are unlikely to be because the potential profits from conducting re- served by the private sector. search on important crops are not sufficiently 64 Agriculture and Achieving the Millennium Development Goals high to attract the interest of either domestic or Nonetheless, the U.N. Conference on Financing for international private firms in Africa (NEPAD 2002). Development in Monterrey in 2002 declared that This is a marked contrast to industrial countries, the decline in ODA should be reversed and assis- where private enterprises fund over 50 percent tance should increase by 25 percent or $12.5 bil- of agricultural research. African governments have lion by 2006. To allow the agriculture sector to reduced their support to agricultural research and develop its full potential for achieving the MDGs, extension because of the pressures to reduce spend- the share of ODA spent on agriculture needs to ing in general. This may also be due to the shift in increase significantly. priorities when governments question the value of For the case of Africa, NEPAD (2002) reported research and extension, given the lack of improve- that estimates on the likely distribution of financ- ment in agricultural productivity in Africa. Donor ing between the public and private sectors remains assistance to agricultural research likewise declined highly hypothetical and requires specific country as a result of priorities that shifted from agricultural conditions to be taken into account. Table 4.2 pre- production to environmental protection, health, sents one set of assumptions on financing. The education, water and sanitation, and so forth. Many public sector is expected to take the lead in water question the need of continued public funding on and land development and rural infrastructure; it agricultural research and extension, thinking that is also expected to increase food supplies at an the world's food problems are solved, or con- approximated $7.5 billion, albeit with considerable strained by research systems or extension services, matching contributions from the farm sector. The or the private sector will do the job. In order to cor- total incremental investment requirement, includ- rect these perceptions, it is necessary to maintain ing operations and maintenance, is about $15.7 bil- and increase support to these services, which are lion per year from 2002 to 2015. This figure was fundamental to maintain the competitiveness of derived from both the national public and private agricultural economies in Africa (NEPAD 2002). resources in addition to international cooperation, Private financing of investment may be explored in line with the Monterrey 2002 commitments on as a general way to ensure availability of basic ser- financing for development. vices, particularly as the official development assis- A report by Atkinson (2003, cited in Reisen tance (ODA) or aid for the water and other sectors 2003) focused on the role of rich countries and on has been declining in recent years (Winpenny 2003). the sphere of public finances, but urged not to limit Table 4.2 Possible Scenario Regarding Financing Sources for Agriculture Under NEPAD Share of total investment (US$ billion) Immediate future Short term Medium term Total Source of investment (2002­2005) (2006­2010) (2011­2015) (2002­2015) Africa Public domestic sources 19.6 40.0 37.8 97.4 Private domestic 2.8 10.0 14.2 27.0 Subtotal 22.4 50.0 52.0 104.4 External Concessional assistance (such as ODA) 25.2 35.0 28.3 88.5 Nonconcessional loans 5.6 10.0 4.7 20.3 FDI (private) 2.8 5.0 9.5 17.3 Subtotal 33.6 50.0 42.5 126.1 Rounding adjustment -- -- -- 0.8 Total 56.0 100.0 94.5 251.3 Annual 14.0 20.1 18.9 17.9 --Not applicable. Source: NEPAD (2002). Global trends and Emerging Issues in Meeting the MDGs 65 perspectives to a resource transfer from rich to over long-term investments should be eliminated. poor countries. Rather, options should also con- Moreover, both prudential regulations on currency sider income redistribution and tax efforts in China, positions of banks and strengthened supervision India, and other poor countries to co-finance the of these regulations and other risk-management MDGs. The additional $50 billion considered nec- procedures are required. Finally, short-term and essary for achieving the MDGs could be financed unhedged borrowing by corporations should be through both traditional and innovative means, disclosed to reduce the credit risk. such as through the enactment by all major econo- mies of a global tax on carbon use; an agreement by the U.S. and other major economies to create addi- The Role of Governance Structures in tional SDRs for development purposes; the enact- Agriculture for Achieving the MDGs ment of a currency transaction tax (Tobin tax); the According to the Transparency International establishment of a Global Lottery in agreement Corruption Perceptions Index 2004 (CPI), cor- with national lotteries; measures to increase flows ruption is considered rampant in many develop- of remittances by immigrant workers for develop- ing countries. Among the 14 worst offenders-- ment purposes; other global taxes, such as a brain expressed as a level of 2 or less with 10 indicating drain tax, an international airport tax, taxation of a clean score--are five countries in Sub-Saharan ocean fishing, taxation of arms exports, a "bit" tax Africa, three in Asia, two in Latin America, and on computer use, or a luxury goods tax. four in the former Soviet Union (Transparency International 2004). Perceived levels of corruption are but one indi- The Role of Foreign Direct cator of lack of adequate governance structures in Private Investment these countries, and lack of governance is consid- International capital flows. Foreign direct investment ered one of the major stumbling blocks to achiev- (FDI) and other long-term, relatively stable invest- ing the MDGs in general, and to realizing the ment have a significant impact on agricultural and MDGs through improvements in the agricultural overall economic growth. FDI in developing coun- sector in particular. tries by companies based in industrial countries Indications abound that poor governance is has increased substantially with globalization and related with bad outcomes for the population, and now dwarfs foreign aid. Private investment was particularly the (rural) poor. Conflicts that are often five times greater in 1998 than in 1968, in constant associated with poor governance are linked with 1993 U.S. dollars. However, the impact of this hunger, both as a cause and an effect of food inse- investment on hunger and poverty may be limited. curity (Messer and Cohen 2004). Good governance Little of this investment goes directly into agricul- conveys stability and security, which again have ture or rural areas. Moreover, FDI is largely con- important linkages to agricultural and economic centrated in just a few countries. Only five countries development (Zhang 2004). Estache and Kouassi accounted for over half of all FDI in 1998 and just (2002) showed that high levels of inefficiency in 12 for all but a small amount of FDI (Runge and the African water sector are linked with weak gov- others 2003). ernance and institutions. In contrast to FDI, the benefits of short-term Good state governance is typically defined under international capital are relatively small and uncer- the terms of accountability, transparency, pre- tain because, unlike FDI, short-term capital does dictability, and participation. Accountability trans- not bring along technology and management lates into government institutions following clear innovations (World Bank 1998). Moreover, when lines of responsibility that ultimately end up with savings rates are already high and marginal invest- the electorate, from which authority flows. Trans- ment is misallocated, short-term capital greatly parency is a concept that makes it possible for gov- increases the vulnerability of the economy. Manage- ernments (and other institutions) to be accountable. ment of international capital flows should there- Predictability relates to the knowledge that the rules fore focus on the creation of an environment of the games are certain and predictable. Partici- conducive to long-term investments and discour- pation is both a principle and a means for achieving aging to short-term capital inflows. Tax incentives good governance. For example, enhanced partic- and other distortions that favor short-term inflows ipation of communities in local government can 66 Agriculture and Achieving the Millennium Development Goals increase accountability, transparency, and pre- capital is the only asset to which the rural poor dictability. These principles are only meaningful have easy access. given adequate institutional and social structures, · Participation in local networks and attitudes where they can be implemented. of mutual trust make it easier for any group to Countries with a good governance structure and reach collective decisions and implement col- adequate institutions tend to ensure political and lective action. Since property rights are often economic stability, possess reasonable state capac- imperfectly enforced in developing coun- ity, enforce property rights and contracts, provide tries, collective decisions on how to manage sufficient public goods, and limit government cor- common or communal resources (such as ruption and predation. On the other hand, coun- watershed land, irrigation, drinking water, tries with poor governance and poor institutions and urban waste disposal) can help maximize typically have poor public services, including for their access, use, and benefits. agricultural extension and research, and particu- · Networks and attitudes reduce opportunis- larly for social services such as water provision tic behavior by community members. Social and education (Wolfensohn and Bourguignon pressures and fear of exclusion can make 2004). In such countries, the neediest--typically individuals behave in certain community- poor farmers and other rural dwellers--are left beneficial ways. For example, farmers have out of the supply chain, as resources are deviated been known to use such networks and atti- to private pockets and to those who can bribe or tudes to exert mutual pressures to prevent otherwise coerce the government to supply their individual diversion of irrigation water, or to needs. Moreover, the voiceless rural people in prevent loan defaults in group borrowing sit- these cases tend to have no access to recourse or uations with joint liabilities (SEARCA/IFPRI/ complaints. At a higher level, the investment cli- CRESCENT 2004). mate in countries lacking good governance and In addition to the explicit implementation of quality institutions is adversely impacted, result- good governance principles, the development and ing in fewer and more expensive private-sector support of community-based organizations (CBOs) investments in both rural and urban areas. Good and general civic education can play an important public governance should include the attainment of development targets and delivery of crucial role in good-governance reforms in agriculture. public services for the rural poor. On the other Effective CBOs, such as farmer associations or coop- hand, good corporate governance essentially means eratives, water user groups, farm and other micro- responsible corporate citizenship, that is, respon- credit and lending groups, or other existing rural sible stewardship of the assets of the company by CBOs, can improve governance in several ways: by increasing the economic value added without encouraging government ministries to adopt suc- damaging the environment or abusing consumers cessful approaches developed within these groups; in the process. by educating and sensitizing the public about their How can governance structures in agriculture rights and entitlements under public programs; by be amended for agriculture to play its due role in acting as a conduit to the government for public achieving the MDGs? Good governance can be opinion and local experience; by collaborating with built through the development of social capital. official bodies; by influencing local agricultural Social capital affects economic development mainly development policies of national and international institutions; and by helping government and by facilitating transactions among individuals, donors fashion a more effective development strat- households, and groups in society. This facilitating egy through strengthening institutions, staff train- function can take the following forms: ing, and improving management capacity. Pitfalls · Participation by individuals in social net- for CBOs can be their own internal governance works increases the availability of information structures and processes, particularly their trans- and lowers its cost. This is true in the rural vil- parency and accountability. These organizations lages, especially when the information can should strive for financial sustainability, for exam- increase the returns from agriculture and trad- ple, through developing stronger links with the ing. Examples include crop prices, location of private corporate sector. This is especially workable new markets, sources of credit, or treatments for nonadvocacy CBOs, such as those that deliver for plant or livestock diseases. Often, social educational, healthcare, and humanitarian services. Global trends and Emerging Issues in Meeting the MDGs 67 As has been shown in Indonesia, decentraliza- Finally, at the international level, enhanced gov- tion can improve governance in rural areas, as ernance and commitment could help improve the local governments are usually more adept with global trade agenda by making progress regard- and knowledgeable about rural agricultural needs ing the needs of the poorest countries, including in particular. However, administrative decentraliza- enhanced access to agricultural and other markets. tion without preparatory capacity building and ade- Enhanced international governance and commit- quate fiscal support could force regional and district ment can also improve the global environmental governments to enact new tax laws and other agenda, as developing countries would gain from regulations that add to the costs of doing business enhanced environmental standards, including pro- without corresponding increase in public services, tection of remaining biodiversity in these countries, or improvement in the quality of governance or from participating in and gaining from climate (SEARCA/IFPRI/CRESCENT 2004). mitigation policies. Improved governance will help lessen con- flict and encourage greater levels of government CONCLUSION funding to be allocated to the agriculture sector. Although governance is important to sustainable To meet the MDGs, there are several concerns that agricultural development in all countries, the spe- all organizations, from the government to the pri- cific political situation in each country will deter- vate sector, in the developed and developing world mine the appropriate measures needed to ensure need to face. To begin with, the complementary good governance. roles of government agencies, the private sector, With rapid development of agricultural exports and public-private partnerships cannot be ignored and trade, as well as with the very rapid growth of when addressing the lack of basic services in water, the retail sector in many developing countries, it land, health, and other infrastructure development has become more important for the agriculture in most developing countries. Likewise, the assis- sector, including smallholders, to be in tune with tance of international and regional organizations corporate and retail sector development, increas- for financing investments, extending technologi- ingly the buyers of agricultural products. Because cal and intellectual support, and strengthening of these developments, improvements in corpo- skills and capacity will move forward economic rate governance should be promoted to support growth and progress in the developing world. agriculture as well, such as by consistently enforc- Government agencies in developing countries ing laws and providing some incentives for good need urgently to revisit the legal, regulatory, polit- governance practices. Good governance in the rural ical, and institutional framework in the agriculture, nonfarm sector is also essential, so that the rural pri- research, extension, and industrial sectors. A refo- vate sector, particularly small- and medium-size cusing of priorities and policies to meet the spe- enterprises, can exploit the emerging investment cific needs of poor people, especially on property and business opportunities in the rural sector rights in land and water to augment their require- offered by agriculture and related activities. ments of food and nutrition security, must be At a higher level, donor countries and interna- achieved to adequately address MDG 1. In addi- tional aid agencies should focus their resources on tion, government agencies, through the assistance these countries where good or improving gover- of international and regional organizations, must nance structures will ensure that the rural poor facilitate the development of alternative rural are reached. The poverty reduction strategy papers nonfarm livelihoods. (PRSPs) that the World Bank and others rely on Both the private and public sectors must foster increasingly include sections on addressing gover- private-public sector partnerships with the objective nance objectives. Whereas the MDGs provide a of addressing the MDGs, and specifically MDG 8-- long-term perspective, PRSPs can provide medium- to develop a global partnership for development. term targets to address these long-term goals. Both Several methods have been discussed here on how PRSPs and reaching the MDGs requires strong the private sector and government institutions can national planning and implementation capacity, a work together to create mutual opportunities and feature of good governance. Increased levels and benefits. It is also anticipated that the private sector direction of ODA toward agriculture are also impor- will increase its investment in agricultural research tant (Wolfensohn and Bourguignon 2004). and extension in developing countries. Use of food 68 Agriculture and Achieving the Millennium Development Goals aid has to be considered judiciously to avoid nega- ducive and sustainable international price incen- tive impacts. tives must be brought about through multilateral Good governance is also a cross-cutting issue. reform of agricultural trade and domestic sub- Governance plays a role in the decision making sidy policies. Current trade barriers and subsi- and implementation of policies both nationally dies are high for agriculture. Benefits will accrue and internationally, in public administration, and to poor farmers from reforms both in developed in service delivery, and therefore plays a role in and developing countries. Successful WTO agri- achievement of the MDGs. cultural trade negotiations that lower protection Finally, if the appropriate investments and and subsidies could thus boost progress toward governance are to help reduce global poverty, con- the MDGs. 5 Conclusions Agriculture paves the The most effective strategy for making steady, sustainable progress on the Millennium Development Goals is to serve all the goals in an inte- way for economic growth grated way. However, each goal will need a well-defined package of technologies and services for success at the field level. Pursuing in poorer nations and the goals separately without acknowledging their interlinkages will reduce the complex process of human and economic development to underlies progress on the a series of fragmented, conflicting, and unsustainable interventions. broad array of economic A comprehensive and harmonious development approach is in order. Given that the majority of poor people live in villages or rely on and social indicators that agriculture, and that agriculture paves the way for economic growth in the poorer nations, agricultural and rural development will under- the MDGs emphasize. lie progress on the broad array of economic and social indicators that the MDGs emphasize. In pursuing the MDGs, we should seek ultimately the elimination of hunger, poverty, and maternal and child malnutrition, with each MDG being an important step along the way to this ultimate target. An emphasis on healthy, productive individuals means that we must attend not simply to food security at the aggregate level, but to nutri- tion security (economic, physical, social, and environmental access to a balanced diet and clean drinking water) at the individual level of child, woman, and man. In the longer term, the goals should be mod- ified to promote a reduction in the absolute number of people living in unsuitable conditions across all countries, rather than a reduction in global proportions, and ultimately the elimination of hunger and malnutrition. Despite these limitations in framing the task at hand, the MDGs can be used to set a powerful agenda for developing countries and the international community, because they offer a guide for plan- ning and implementing a broad range of development efforts. Of the eight Millennium Development Goals, the first goal is the one whose attainment most clearly involves the agricultural sector: The poor around the globe are disproportionately farmers and herders, and, perversely, the hungry also most commonly find their livelihoods through agriculture. The impact that a dynamic agricultural sector will have on the attainment of the other seven goals is less direct. Nonetheless, important gains in these can be made through explicit attention to agriculture. 1. Eradicate extreme hunger and poverty. Improving the productivity of and the economic returns of agriculture will have immediate effects in eradicating extreme poverty and reducing hunger. In 69 70 Agriculture and Achieving the Millennium Development Goals addition, subsistence farming households will saving innovations are required to eliminate this enjoy immediate benefits from increased food potentially zero-sum tradeoff. production. But more importantly, increased 5. Improve maternal health. Agriculture can con- food production will lead to real reductions in tribute to this goal in a similar manner to how it food prices, improving the purchasing power of contributes to attaining all of the previous three the poor throughout the economy. Furthermore, goals. In addition, agriculture can contribute economy-wide effects will occur when the focus to improved maternal health in other ways. on agriculture increases. Agricultural growth Agriculture has considerable potential to directly will, therefore, have ripple effects into other improve maternal health by improving the qual- sectors--enhancing the productivity, increasing ity of the diets of women. Micronutrient defi- the returns, and improving the incomes of those ciencies are particularly severe among young working in sectors quite far removed from farm- children and women. Among women, the health ers, herders, and fishermen and women. effects of such deficiencies are most pronounced 2. Achieve universal primary education. Investments during pregnancy, at birth, and in the months in agriculture will advance progress toward following birth. By increasing the micronutrient this goal primarily in an indirect fashion. At the content of food crops the prevalence of micro- household level, the profitable use of labor- nutrient deficiencies among women should saving agricultural technologies will reduce the decline and maternal health would improve. opportunity costs to farmers who allow their 6. Combat HIV/AIDS, malaria, and other diseases. children to go to school. Broad economic growth The analysis for several of the other goals also likely will demand increasingly skilled labor, applies here. Although the indicators for this and will increase the returns of investing in goal have no or very little agricultural content schooling one's children, both in agriculture and and the direct links between agriculture and in other sectors. these health issues are not immediate, they are 3. Promote gender equality and empower women. important nonetheless. For agricultural house- Throughout the developing world, but particu- holds, as for all households, the productivity of larly in Sub-Saharan Africa, women are farm- their economic activities is an important deter- ers. To a greater degree than for men, women's minant of whether or not such households live principal productive activities are in agricul- in an environment that allows them to enjoy a ture. Agriculture also provides key contribu- healthy and active life. Besides, a dynamic agri- tions to the economic empowerment of women. cultural sector also has the potential to radically A dynamic agricultural sector that offers broad alter the disease environment in a region. New welfare benefits can only be expected to emerge health challenges likely will emerge with an when women are given the opportunity to par- evolving agricultural sector. For example, wage- ticipate profitably in the sector. However, if the labor migration associated with agriculture may benefits of a dynamic agriculture are to result in expose populations to new diseases, increasing sustained improvements in the direct determi- the health burden they bear. The resources of nants of welfare--income, health, and educa- the agriculture sector, particularly in the public tion, among others--it is necessary that women sphere through the extension services, can be have an important role in determining how the used in a coordinated fashion with those of the fruits of their agricultural activities are used. health sector to address such challenges. 4. Reduce child mortality. The linkages between 7. Ensure environmental sustainability. This MDG agriculture and child mortality are indirect but covers a broad sweep, including biodiversity, important. Agriculture is a critical component critical natural habitats, energy use and global in assuring food and nutrition security. It is esti- climate change, unsafe water and poor sanita- mated that 45 to 55 percent of all child deaths tion, and urban slums. A productive agricultural are due to malnutrition exacerbating the nega- sector will reduce pressure in all of these areas, tive effects of disease (Pelletier and others 1994). including the following: In addition, increased agricultural productivity · A productive agriculture requires less land, may come at the expense of women's ability leaving marginal agricultural lands to other to offer dedicated care to children in their care. uses, including forests and other critical Consequently, agricultural labor and time- habitats. Conclusions 71 · Proper agricultural policies will allow the full especially to achieve an open, rule-based, pre- costs of agricultural technologies to be consid- dictable, nondiscriminatory trading and financial ered as they are being used, reducing the scope system. A renewed focus on agriculture could for excessive nutrient runoff from agriculture, support improvements toward achieving this providing incentives for efficient energy use in goal. Global agricultural trade must be harmo- the sector, and promoting the ecologically sus- nized and rationalized. Moreover, the rationality tainable use of a range of technologies, includ- used must include a judicious consideration of ing pesticides and GMO materials. the special needs of poor agricultural producers · Agriculture has the potential to be an impor- and how they might derive benefits from global tant component in any systems established to agricultural trade. The sectoral (re)allocation manage global carbon stocks. of reoriented ODA should be specified and, with · A dynamic agricultural sector and the pub- the preponderance of poor people in rural lic revenues that accrue from the sector will areas, agriculture and rural development should enable greater levels of public provision of receive significant priority in allocations. safe drinking water and improved sanita- tion. Increasingly productive and profitable In the remainder of this conclusion, the focus is farmers in many developing countries will be on policies to meet MDG 1. Eradicating hunger and more able to privately provide these ameni- poverty requires an understanding of the ways in ties for their own households. For agricultural which these two injustices interconnect. Hunger, development programs to be environmen- and the malnourishment that accompanies it, pre- tally sustainable, their long-term environmen- vents poor people from escaping poverty because it tal costs and benefits have to be taken into diminishes their ability to learn, work, and care for account. Plus, policies and regulations need to themselves and for their family members. If left be in place to encourage efficient--instead of unaddressed, hunger sets in motion an array of out- excessive--energy, water, fertilizer, and pesti- comes that perpetuates malnutrition, reduces the cide use. Sound water management in agri- ability of adults to work and to give birth to healthy culture is critical for safe drinking water, as children, and erodes children's ability to learn and well as for prevention of water-borne dis- lead productive, healthy, and happy lives. This eases and wasting of water. At the same truncation of human development undermines a time, agriculture-led economic growth will country's potential for economic development--for provide public revenues that governments generations to come. can use to provide safe drinking water and There are strong, direct relationships between better sanitation, as well as higher incomes agricultural productivity, hunger, and poverty. that will allow individual farming households Three-quarters of the world's poor live in rural areas to invest in these basic needs. The issue of and make their living from agriculture. Hunger and better resource management arises in urban child malnutrition are greater in these areas than in areas as well. Urban water subsidies, for urban areas. Moreover, the higher the proportion of example, go disproportionately to the better the rural population that obtains its income solely off in most developing countries because they from subsistence farming (without the benefit are connected to the public system. The urban of pro-poor technologies and access to markets), poor, who must rely on water vendors, pay the higher the incidence of malnutrition. Therefore, many times more for water than better-off res- improvements in agricultural productivity aimed at idents. Removing such subsidies and using small-scale farmers will benefit the rural poor first. the available money to finance wider distrib- Increased agricultural productivity enables farm- ution of piped water would benefit the poor. ers to grow more food, which translates into better · Population pressures in urban slums will diets and, under market conditions that offer a level be alleviated to a significant degree if prof- playing field, higher farm incomes. With more itable agricultural systems are developed in money, farmers are more likely to diversify produc- the rural areas. tion and grow higher-value crops, benefiting not 8. Develop a global partnership for development. There only themselves but the economy as a whole. A are several areas where agriculture can con- larger supply of agricultural products also brings tribute to achieving the targets under this goal, prices down, allowing both the rural and urban 72 Agriculture and Achieving the Millennium Development Goals poor to purchase essential foods for less money. preschool children malnourished--and the target Smaller food bills mean that landless poor people indicator--15 percent children malnourished--to will have more money to invest in assets, which will reach 17 percent. help them increase income and survive future eco- However, the outcome varies significantly by nomic shocks. This income and asset security helps country and region. While Latin America, West build a solid foundation for economic growth, by Asia and North Africa, and China will, on average, enabling people to work free from the debilitating likely get close to the target indicator by 2015, even effects of hunger and undernutrition. A flourishing under business-as-usual, the likelihood that Sub- agriculture sector also facilitates job creation in other Saharan Africa and South Asia will come close to areas, such as the food processing and marketing their respective target rates is much smaller. In fact, sectors, and creates secondary economic effects in on average, Sub-Saharan Africa and South Asia will the nonfarm economy. not be able to halve their shares of malnourished By increasing food availability and incomes children by 2015 even under very favorable agri- and contributing to asset diversity and economic cultural, economic, and social conditions; but growth, higher agricultural productivity and sup- improved conditions in these sectors can bring the portive pro-poor policies allow people to break out countries in these regions very close to the 2015 of the poverty-hunger-malnutrition trap. A closer target and can facilitate further reductions in mal- examination of how the targets of MDG 1 can nutrition levels later on. be achieved in Sub-Saharan Africa (Ethiopia and Under the MDG scenario, to bring developing Zambia) illustrates that countries in the region will countries, particularly South Asia and Sub-Saharan need to achieve significantly larger economic Africa, within reach of the preschool malnutrition growth, and very soon. In Zambia, the GDP growth target indicator, total investments in agricultural rate required to halve poverty is around 10.5 per- and supporting sectors during 1995­2015 will have cent, and in Ethiopia, 5.8 percent. Achieving these to increase by $161 billion based on IMPACT- rates of growth requires greater productivity in the WATER calculations. The three main areas of invest- agricultural sector. The unique structure of each ment for the MDG scenario in percentage terms country's economy determines the impact of this are rural roads, irrigation, and education. Together agricultural growth on poverty reduction as well as these three areas will require $403 billion between which agricultural subsectors are the most promis- 1995 and 2015 to achieve the rapid levels in child- ing. In Ethiopia, investments in the staple food sec- hood malnutrition simulated under the MDG sce- tor have the largest potential impact on both growth nario. The increase in irrigation investments is not and poverty alleviation, while in Zambia, the expan- only due to the expansion of irrigated area, but also sion of traditional and nontraditional goods is more to the increased irrigation costs under the MDG important. Both countries have in common sig- scenario as a result of expansion in more costly nificant constraints to market access that need to areas, such as Sub-Saharan Africa. be overcome to attain projected rates of agricul- Agricultural research investments account for tural and economic growth and decreased poverty. $109 billion, and $78 billion of investments toward Multisector strategies, such as a combination of increasing access to safe water is also required. Due nontraditional export growth with livestock and to the long lags in the generation of impact from staple production growth, can significantly reduce agricultural research, increases in research expen- poverty in Ethiopia. By concurrently improving ditures, even beginning now, will have relatively rural infrastructure and market development, agri- small impacts on crop yields by 2015. Increased cultural growth could help Ethiopia to achieve investments in agricultural research are likely to MDG 1. In Zambia, on the other hand, reaching the be essential to meet crop and animal production poverty target under MDG 1 will be more difficult. needs beyond 2015, however. Other investments, A global assessment of Target 2 of MDG 1 (halv- such as roads and irrigation, have significant lags ing child malnutrition levels) shows that the com- in impact as well, so implementing the invest- bination of agricultural and economic growth ment portfolio required for the MDG scenario will together with larger investments in social sectors, require very rapid action. As relatively high levels including health and education, can substantially of access to clean drinking water are already narrow the gap between the business-as-usual out- achieved in the baseline scenario, only $15 billion comes for 2015--24 percent of developing-country in investments are added for the MDG scenario, Conclusions 73 but these investments have basically no lag period and generate multiple livelihood opportunities in in becoming effective. both the farm and nonfarm sectors. In addition to these investments, significant pol- In addition to pro-poor economic and agricul- icy and governance reform is required. How can tural policies, agriculture, like other sectors, needs the poor benefit most from greater investments good governance, absence of conflict, and well- and higher agricultural productivity? Experience functioning markets and private enterprise to flour- has shown that a number of key conditions help ish. As the financing requirements for realizing the maximize the benefits of a growing agriculture sec- MDGs are substantial, the private sector is increas- tor for poor people. To achieve faster agriculture- ingly called upon to fill investment gaps. Its com- based growth rates, favorable macroeconomic and plementary and supporting role in the provision trade policies, good infrastructure, and access to of basic services in water, land, health, and other credit, land, and markets must be in place. These infrastructure development that are lacking in most conditions create level playing fields and give farm- developing countries cannot be ignored. The devel- ers incentives to adopt new and sustainable tech- opment and business communities must increas- nologies and diversify production into higher-value ingly recognize that the MDGs cannot be achieved crops, actions that raise incomes and lift house- and private enterprise cannot flourish without holds out of poverty. greater and more equitable involvement of poor At the macroeconomic level, trade liberalization, people in markets. The idea of enticing global pri- particularly in developing countries, will enhance vate enterprise into developing-country markets is the targets under MDG 1. To achieve these overall not new but the expectations are different this gains, developing countries that will benefit from time around. In many respects they are driven by a more open markets abroad need to participate in greater understanding that not just any kind of eco- multilateral agricultural trade policy reforms. Im- nomic growth will improve the lives of the poor. It proved investments in infrastructure are another will take a particular kind of private-sector involve- important avenue to enhance MDG 1. However, ment to generate the necessary economic transfor- if there is a lack of coordination at the country, mations. Private entrepreneurs are now increasingly regional, and donor level, the linkages and com- held to environmental, social, and corporate gover- plementarities of infrastructure investment will nance principles that stress sustainable business not be realized. As a result, fragmented approaches practices and adherence to labor standards. Without that lack sufficient attention for substantive poli- these standards and practices, the private sector and cies and development issues do not help coun- disadvantaged groups cannot mutually benefit tries achieve their MDG targets. Moreover, local from consumer, employment, and entrepreneurial recipients need to be integrated fully in any invest- activities. ment plans. Likewise, international and regional organiza- Other important areas of reform in the trad- tions' assistance in terms of financing investments, ing area include the elimination of export subsi- extending technological and intellectual support, dies; the move toward income-support instead promoting skills, and strengthening capacity will of production-stimulating support measures in move forward economic growth and progress in the developed countries; the expedition of importing- developing world. Government agencies in devel- country risk assessments and regulatory changes oping countries urgently need to revisit the legal, that can help open market access for poor countries; regulatory, political, and institutional framework and the reform of the international and national in agriculture, research, extension, and industrial governance of food aid programs. An improved sectors to facilitate private-sector involvement. domestic regulatory framework would intensify Moreover, both the private and public sectors must competition among suppliers of essential inputs, foster private-public­sector partnerships and culti- such as seeds and fertilizer. In addition, the elimi- vate this relationship with the end objective of nation of trade barriers for agricultural products, addressing the MDGs. To allow the agriculture sec- especially the high-value-added products, would tor to develop its full potential for achieving the encourage a greater number of private entrepre- MDGs, the share of ODA spent on agriculture needs neurs to explore opportunities in agribusiness. A to increase significantly. healthy market and private sector would provide The role of good governance systems in achiev- value-added, skilled work to the landless poor ing the MDGs and agricultural and economic devel- 74 Agriculture and Achieving the Millennium Development Goals opment in general has long been recognized. Good ernance and commitment could help bring about governance can be built and strengthened through improvements in the global trade and environmen- the development of social capital. In addition to the tal agenda. explicit implementation of good governance princi- When good governance, equitable markets, and ples, the development and support of CBOs and the other key conditions noted above are absent, civic education, in general, can play an important poor farmers are unlikely to earn decent incomes role in good-governance reforms in agriculture. As and secure adequate diets for themselves and has been shown in Indonesia, decentralization can their families. If agriculture underperforms or improve governance in rural areas, as local govern- fails, nonfarmers will also feel the negative effects. ments are usually more adept with and knowledge- We need to keep uppermost in our minds that able about rural agricultural needs in particular. significant gains in agricultural productivity have Improved governance will help lessen conflict and provided the critical first steps in economic dev- encourage greater levels of government funding to elopment in many countries. It is a promising be allocated to the agriculture sector. Although gov- development that the review of progress--and ernance is important to sustainable agricultural lack thereof--in achieving the MDGs has reached development in all countries, the specific political global attention. Calls for accountability and situation in each country will determine the appro- action that have real impact on people are grow- priate measures to ensure good governance. At a ing because of that attention. Policy action and higher level, donor countries and international aid increased investment in the critical arenas of sus- agencies should focus their resources on those coun- tainable agriculture productivity and food and tries where good or improving governance struc- nutrition security will be essential for responding tures will ensure that the rural poor are reached. effectively and responsibly to reach the Millennium Finally, at the international level, enhanced gov- Development Goals. Notes CHAPTER 2 9. IMPACT-WATER is described in detail in Rosegrant, Cai, and Cline (2002). 1. This is particularly the case in Africa and Asia, but less so in Latin America where poverty in several countries is con- centrated in the urban centers (Khan 2000). CHAPTER 4 2. Although here rural statistics are used as a rough proxy for the population that is dependent on agricultural livelihoods, 10. References and further details are provided in Beierle and it is important to highlight that agriculture is also a key liveli- Diaz-Bonilla (2003). hood strategy for the urban poor, particularly in Africa. 11. As mentioned by Rodrik (1998), extensive trade liberaliza- 3. Johnston and Mellor (1961), who provided one of the first tion during the 1980s along with other reforms helped some comprehensive descriptions of the role of agriculture in eco- of the region's leading reformers, such as Uganda and nomic development, note that "insufficient movement out Ghana, recover from long periods of economic decline. But of agriculture will perpetuate, or lead to, excessively small neither Uganda nor Ghana has yet reached the level of in- farms and serious underemployment of labor as the proxi- come per capita it attained in 1970s. mate causes of sub-standard farm incomes (p. 590)." 12. The Food Aid Convention is a voluntary agreement among donors that has attempted to establish global food aid tar- gets, eligible commodities, and other guidance criteria, but CHAPTER 3 has no enforcement capacity. The FAO also has an advisory committee on food aid. In addition, food aid is subject to 4. This section on Zambia draws on Lofgren et al. (2004), Thurlow limited rules under the WTO, and those rules may be sub- (2004), and Thurlow and Wobst (2004). ject to additional clarification in the ongoing Doha Round 5. Based on per-capita debt to GDP. negotiations. 6. The simulations assume somewhat optimistic projections for 13. Wharton was one of the first to emphasize the importance the copper mining sector, including the stabilization of world of infrastructure in the generation of positive externalities copper prices and the availability of private investment (see at the microeconomic level. The author recognized that Lofgren, Robinson, and Thurlow 2002 and Lofgren, Thurlow, agricultural development is not exclusively determined by and Robinson 2004). the "economic behavior of the producers," but also de- 7. Given that the purpose of this section is to identify differ- pends on the environment, which according to Wharton in- ences in the poverty-reduction potential across agricultural cludes physical-climatic, socio-cultural, and institutional sectors, no attempt is made to account for the fiscal costs of components in what he calls "the agricultural infrastruc- raising productivity under these scenarios. ture." We follow Wharton's (1967) definition of infrastruc- 8. The Nonagriculture-Led Growth Scenario excludes produc- ture that identifies three categories: one that is capital tivity gains in mining. For a discussion of the effects of re- intensive (like roads or bridges); one that is capital exten- newed mining growth see Lofgren, Robinson, and Thurlow sive (principally extension services or animal sanitation (2002) and Lofgren, Thurlow, and Robinson (2004) and services); and institutional infrastructure (consisting of for- Thurlow and Wobst (2004). mal and informal institutions). 75 References ACC/SCN (United Nations Administrative Committee on de Onis, M., M. Blössner, E. Borghi, E. A. Frongillo, and Coordination­Subcommittee on Nutrition). 1996. Update on R. Morris. 2004. "Estimates of Global Prevalence of Childhood the Nutrition Situation. ACC/SCN, Geneva, Switzerland. Underweight in 1990 and 2015." Journal of the American Medical AfDB (African Development Bank) and OECD (Organization for Association 291(21): 2600­2606. Economic Cooperation and Development). 2002. Achieving the Devarajan, S., W. Easterly, H. Fofack, D. Go, A. Izquierdo, and Millennium Development Goals in Africa: Progress, Prospects, and L. Pizzati. 2000. A Macroeconomic Framework for Poverty Policy Implications. Paris, France. Reduction Strategies. Development Research Group and World AIARD (Association for International Agriculture and Rural Bank Institute. Washington, DC: The World Bank. Development). 2003. The Agriculture, Nutrition, and HIV/AIDS Diao, X., E. Diaz-Bonilla, and S. Robinson. 2003. "Tell Me Where Connections in Developing Countries. Resource paper prepared it Hurts and I'll Tell You Who to Call: Industrialized Countries' for the US Agency for International Development. Agricultural Policies and Developing Countries." IFPRI Brief. Anderson, Kym, Will Martin, and Dominique van der International Food Policy Research Institute, Washington, DC, Mensbrugghe. 2004. "Price, Trade and Welfare Implications September. of the Doha Agenda." Presented at the World Bank confer- Dirven, M. 1999. "The Role of Agents in Agricultural Policies: ence Agricultural Trade Reform and the Doha Development Intentions and Reality." Cepal Review 68. Chile. Agenda. The Hague, December 1­2 (proceedings forthcoming). Easterly, W., and R. Levine. 1997. "Africa's Growth Tragedy: Atkinson, A. 2003. New Sources for Development Finance. Helsinki: Policies and Ethnic Divisions." Quarterly Journal of Economics Wider. CXII (4): 1203­1250. Barrett, C., and D. Maxwell. 2004. Food Aid After Fifty Years: Edwards, S. 1995. Crisis and Reform in Latin America. Oxford and Recasting Its Role. Draft book manuscript. New York: Cornell Washington, DC: World Bank, Oxford University Press. University. EM-DAT. 2004. The OFDA/CRED International Disaster Data- Beghin, J. C., D. Roland-Holst, and D. van der Mensbrugghe. base. www.em-dat.net--Université Catholique de Louvain-- 2002. Global Agricultural Trade and the Doha Round: What are the Brussels--Belgium Estache, A., V. Foster, and Q. Wodon. 2001. Making Infrastructure Implications for the North and South? Working Paper 02-WP Reform Work for the Poor: Policy Options based on Latin American 308, Center for Agricultural and Rural Development, June. Experience. LAC Regional Studies Program, WBI Studies in Ames, Iowa: Iowa State University. Development, Finance, Private Sector and Infrastructure Beierle, T., and E. Diaz-Bonilla. 2003. "The Impact of Agricultural Department. Washington DC: The World Bank. Trade Liberalization on the Rural Poor: An Overview." Estache, A., V. Foster, and Q. Wodon. 2002. Accounting for Poverty Background paper for the RFF/IFPRI Workshop on Agri- in Infrastructure Reform: Learning from Latin America's Experience. cultural Trade Liberalization and the Poor, 3­4 November, Washington DC: World Bank Institute, The World Bank Group. Washington DC. Estache, A., and E. Kouassi. 2002. "Sector Organization, Bigsten, A., and S. Kayizzi-Mugerwa. 2000. "The Political Governance, and the Inefficiency of African Water Utilities." Economy of Policy Failure in Zambia." Working Papers in World Bank Policy Research Working Paper 2890. The World Economics, Department of Economics, Göteborg University. Bank, Washington, DC. Binswanger, H. P., S. R. Khandker, and M. P. Rosenzweig. Fan, S., and P. Hazell. 2001. "Returns to Public Investments in 1993. "How Infrastructure and Financial Institutions Affect the Less-Favored Areas of India and China." American Journal Agricultural Output and Investment in India." Journal of of Agricultural Economics 83(5): 1217­22. Development Economics 41(2): 337­366. ------. 1999. "Are Returns To Public Investment Lower In Less- Borlaug, N. 1999. "Biotech Can Feed 8 billion in the Next Favored Rural Areas? An Empirical Analysis of India." EPTD Century--If Environmental Extremists Permit It." New Discussion Paper 43, Environment and Production Technol- Perspectives Quarterly. Accessed at http://www.digitalnpq. ogy Division, International Food Policy Research Institute. org/global_services/nobel%20laureates/10-06-99.HTML. Washington, DC, May. Boserup, E. 1981. Population and Technology. Oxford, Basil Fan, S., P. Hazell, and T. Haque. 2000a. "Targeting Public Blackwell. Investments by Agro-Ecological Zone to Achieve Growth Brenneman, A., and M. Kerf. 2002. "Infrastructure & Poverty and Poverty Alleviation Goals in Rural India." Food Policy Linkages, A Literature Review." The World Bank, Wash- 25(4): 411­428. ington, DC. Fan, S., P. Hazell, and S. Thorat. 2000b. "Government Spending, Chanthunya, C. L., and V. Murinde. 1998. Trade Regime and Growth and Poverty in Rural India." American Journal of Economic Growth. Vermont: Aldershot and Brookfield. Agricultural Economics 82(4): 1038­51. Deininger, K., and P. Olinto. 1999. Why Liberalization Alone Has Fan, S., and N. Rao. 2004. "Public Spending in Developing Not Improved Agricultural Productivity in Zambia: The Role of Countries: Trends, Determination, and Impact." Preliminary Asset Ownership and Working Capital Constraints. Washington, Working Paper. International Food Policy Research Institute. DC: The World Bank. Washington, DC. 77 78 Agriculture and Achieving the Millennium Development Goals Fan, S., L. Zhang, and X. Zhang. 2002. "Growth, Inequality, and ------. 2003. "Achieving the Millennium Development Goals Poverty in Rural China The Role of Public Investments." by Enabling the Rural Poor to Overcome their Poverty." Research Report 125. International Food Policy Research Discussion Paper, IFAD, Rome, Italy. Institute. Washington, DC. Jacoby, H. 1998. Access to Markets and the Benefits of Rural Roads. Fane, G. 1998. "The Role of Prudential Regulation." In East Asia Development Research Group Rural Development. Washing- in Crisis: From Being a Miracle to Needing One? R. H. McLeod ton, DC: The World Bank. and R. Garnaut (eds). London and New York: Routledge. Jalan, J., and M. Ravallion. 2001. "Does Piped Water Reduce FAO (Food and Agriculture Organization). 1998. FAO FAOSTAT Diarrhea for Children in Rural India." Policy Research databases. Accessed at http://apps.fao.org/default.jsp. Working Paper No. 2664. The World Bank, Washington, DC. ------. 2001. Mobilizing the Political Will and Resources to Banish Jimenez, E. 1995. "Human and Physical Infrastructure: Public World Hunger. Rome: FAO. Investment and Pricing Policies in Developing Countries." ------. 2003. The State of Food Insecurity in the World--2003. In Jere Behrman and T. N. Srinivasan (eds.) Handbook of Rome: FAO. Development Economics, Volume 3.Amsterdam: North-Holland. FAO/IFAD/WFP (Food and Agriculture Organization, Inter- Johnston, B. F., and J. W. Mellor. 1961. "The Role of Agriculture national Fund for Agricultural Development, and World in Economic Development." The American Economic Review Food Programme). 2002. "Reducing Poverty and Hunger: 51(4): 566­593. The Critical Role of Financing for Food, Agriculture, and Josling, T., D. Roberts, and D. Orden. 2004. Food Regulation and Rural Development." Paper prepared for the International Trade: Toward a Safe and Open Global System. Washington DC: Conference on Financing for Development, 18­22 March, Institute for International Economics. Monterrey, Mexico. Khan, M. H. 2000. "Rural Poverty in Developing Countries: Issues Fay, Marianne and Tito Yepes. 2003. "Investing in Infrastructure: and Policies." IMF Working Paper WP/00/78. International What is Needed from 2000 to 2010?" World Bank Policy Monetary Fund, Washington, DC. Research Working Paper 3102, World Bank: Washington, DC. Lebo, J., and D. Schelling. 2001. "Design and Appraisal of Rural Fishbein, Robert. 2001. "Rural Infrastructure in Africa: Policy Transport Infrastructure: Ensuring Basic Access for Rural Directions." Africa Region Working Paper Series No.18. World Communities." Technical Paper No. 496. The World Bank, Bank. Washington, DC. Galiani, S., P. Gertler, and E. Schargrodsky. 2005. "Water for Life: Leipziger, D., M. Fay, Q. Wodon, and T. Yepes. 2003. "Achiev- The Impact of Privatization of Water Services on Child ing the Millennium Development Goals: The Role of Mortality." Journal of Political Economy Vol. 113 (1), pp. 83­120. Infrastructure." Policy Research Working Paper 3163. The Garnaut, R. 1998. "Economic Lessons." In East Asia in Crisis: From World Bank, Washington, DC. Limão, N., and A. J. Venables. 1999. "Infrastructure, Geographical Being a Miracle to Needing One? R. H. McLeod and R. Garnaut Disadvantage, and Transport Costs." Policy Research Working (eds). London and New York: Routledge. Paper 2257. The World Bank, Washington, DC. Gautam, M. 2003. Debt relief for the poorest: an OED review of Lipton, M., and M. Ravallion. 1995. "Poverty and Policy." In Jere the HIPC initiative. The Operations Evaluations Division. Behrman and T. N. Srinivasan, (eds). Handbook of Development Washington, DC: The World Bank. Economics, Volume 3. Amsterdam: North-Holland. Geist, H. J., and E. F. Lambin. 2002. "Proximate Causes and Lofgren, H., Thurlow, J., and Robinson, S. 2004. "Prospects for Underlying Driving Forces of Tropical Deforestation." Bio- Growth and Poverty Reduction in Zambia, 2001­2015," Dev- Science 52(2): 143­150. elopment Strategies and Governance Division Discussion Gleick, P. H., G. Wolff, E. L. Chalecki, and R. Reyes. 2002. The Paper No. 11, International Food Policy Research Institute, New Economy of Water: The Risks and Benefits of Globalization Washington, D.C. and Privatization of Fresh Water. Oakland, CA: Pacific Institute. Lofgren, H., S. Robinson, and J. Thurlow. 2002. "Macro and Goldin, Ian and L. Alan Winters (Eds). 1992. "Open Economies: Micro Effects of Recent and Potential Shocks to Copper Structural Adjustment and Agriculture." Cambridge Uni- Mining in Zambia." Trade and Macroeconomics Division versity Press, 342p. London. Discussion Paper No. 99, International Food Policy Research Haggblade, S., and B. Zulu. 2003. "The Cassava Surge in Zambia Institute, Washington, DC. and Malawi." Conference Background Paper No. 9 prepared MAAIF & MFPED (Ministry of Agriculture, Animal Industry, and for the InWEnt, IFPRI, NEPAD, CTA conference Successes in Fisheries and Ministry of Finance, Planning and Economic African Agriculture, Pretoria December 1­3, 2003. Development). 2000. Plan for the Modernisation of Agriculture: Hazell, P., and S. Haggblade. 1993. "Farm-Nonfarm Growth Eradicating poverty in Uganda (Government strategy and opera- Linkages and the Welfare of the Poor." In Including the Poor, tional framework). Kampala, Uganda: MAAIF & MFPED. M. Lipton and J. van der Gaag (eds). Washington, DC: The Messer, E. and M. J. Cohen. 2004. Breaking the links between World Bank. pp. 190­204. conflict and hunger in Africa. 2020 Africa Conference Brief 10. Hertel, Tom, and Alan Winters. 2004. "Putting Development Back Washington, DC: IFPRI. into the Doha Agenda: Poverty Impacts of a WTO Agreement." Meyer, R. L., and G. Nagarajan. 1999. Rural Financial Markets in World Bank conference held in The Hague, December 2­4 (pro- Asia: Policies, Paradigms, and Performance. Oxford University ceedings forthcoming). Press. Hoddinott, J., M. Cohen, and M. S. Bos. 2003. "Food Aid in the MFPED (Ministry of Finance, Planning and Economic Dev- 21st Century: Current Issues and Food Aid as Insurance." elopment). 2000. Uganda's Poverty Eradication Action Plan-- Paper presented at the international workshop on Defining Summary and Main Objectives. Kampala: MFPED. the Role of Food Aid in Contributing to Sustainable Food Micronutrient Initiative & United Nations Children's Fund Security, Berlin, Germany. September. (UNICEF). 2004. Vitamin & Mineral Deficiency: AProgress Report. Howe, J. D. G. F., and P. J. Richards. 1984. Rural Roads and Poverty Ottawa: The Micronutrient Initiative & New York: UNICEF. Alleviation: A Study Prepared for the International Labour Office MIGA (Multilateral Investment Guarantee Agency). 2004. MIGA within the Framework of the World Employment Programme. and Water. Washington DC: World Bank Group. London: Intermediate Technology Publications. Ministry of Agriculture and Rural Development. 2005. Millennium IFAD (International Fund for Agricultural Development). 2001. Development Goals Based Needs Assessment: The Rural Dev- Rural Poverty Report 2001: The Challenge of Ending Rural Poverty. elopment and Food Security Sector, draft, Government of Ethiopia: Oxford University Press. Addis Ababa. Introduction 79 Minot, N., and L. Daniels. 2003. "Impact of Global Cotton Saasa, O. 2003. "Agricultural Intensification in Zambia: The Role Markets on Rural Poverty in Benin." Discussion Paper No. 48. of Policies and Policy Processes." Institute of Economic and Markets, Trade, and Institutions Division, International Food Social Research, University of Zambia. Policy Research Institute, Washington, DC. Sachs, J., and A. Warner. 1997. Sources of Slow Growth in African MOFED (Ministry of Finance and Economic Development). Economies. Boston: Harvard Institute for International 2002. Ethiopia: Sustainable Development and Poverty Reduction Development. Program. Addis Ababa, Ethiopia. Samatebele, H. 2003. "Overview on the Current Food Security ------. 2003. Ethiopia: Sustainable Development and Poverty Crisis in Zambia." In S. Drimie and M. Lafon (eds). Food Security Reduction Program (SDPRP). Annual Progress Report. Addis in Southern Africa: Causes and Responses from the Region. Ababa, Ethiopia. Johannesburg: Institut Français d'Afrique du Sud. MOPED (Ministry of Planning and Economic Development), Schiff, M., and A. Valdez. 1998. "Agriculture and the Macro- Ethiopia. 1994. Poverty Monitoring Mechanism for Ethiopia. economy." In B. Gardnet, and G. Rausser (eds). Handbook of Addis Ababa, Ethiopia. Agricultural Economics. Elsevier Science. Narayanan, S., and A. Gulati. 2002. "Globalization and the SEARCA/IFPRI/CRESECENT. 2004. Agriculture and rural devel- Smallholders: A Review of Issues, Approaches, and Impli- opment strategy study. Final Report. Prepared for the Ministry cations." MSSD Discussion Paper 50, International Food of Agriculture, Indonesia. ADB TA 3843-INO. July. Bogor, Policy Research Institute, Washington DC. Indonesia. NEPAD (New Partnership for Africa's Development). 2002. Seckler, D. 1990. "Private Sector Irrigation in Africa." In Comprehensive Africa Agriculture Development Programme. Rome: Irrigation in Sub-Saharan Africa: The Development of Public and Food and Agriculture Organization of the United Nations Private Systems, S. M. Barghouti and G. Le Moigne (eds). (FAO). World Bank Technical Paper No. 123. Washington, DC: The ORC/Macro. 2004. Demographic and Health Surveys STAT- World Bank. compiler. Online DHS survey data extraction tool. Accessed Smith, D. R., A. Gordon, K. Meadows, and K. Zwick. 2001. at http://www.measuredhs.com/start.cfm. "Livelihood Diversification in Uganda: Patterns and Orden, D., M. Torero, and A. Gulati. 2004. "Agricultural Markets Determinants of Change Across Two Rural Districts." Food and the Rural Poor." DCD/DAC/POVNET/A(2004) 1/RD3. Policy 26: 421­435. DAC Network on Poverty Reduction OECD, Paris. [Numbers Smith, L. C., and L. Haddad. 2000. Explaining Child Malnutrition are OECD internal tracking system] in Developing Countries: A Cross-Country Analysis. Research Pelletier, D. L., E. A. Frongillo Jr., D. G. Schroeder, and J.-P. Report 111. Washington, DC: International Food Policy Habicht. 1994. "A Methodology for Estimating the Contri- Research Institute. bution of Malnutrition to Child Mortality in Developing Smith, L. C., U. Ramakrishnan, A. Ndiaye, L. Haddad, and Countries." Journal of Nutrition 124(10 suppl.): 2106S­2122S. R. Martorell. 2003. The Importance of Women's Status for Child Platteau, J.-P. 1996. "Physical Infrastructure as a Constraint on Nutrition in Developing Countries. Research Report 131. Agricultural Growth: The Case of Sub-Saharan Africa." Oxford Washington, DC: International Food Policy Research Institute. Development Studies 24(3): 189­219. Spielman, D., and K. von Grebmer. 2004. "Public-Private PRSP. 2003. "Ghana Poverty Reduction Strategy Paper." Inter- Partnerships in Agricultural Research: An Analysis of national Monetary Fund: Washington, DC. Challenges Facing Industry and the Consultative Group PRSP. 2003. "Uganda Poverty Reduction Strategy Paper on International Agricultural Research." EPTD Discussion Annual Progress Report." International Monetary Fund, Paper No. 113. International Food Policy Research Institute, Washington, DC. Washington, DC. Quisumbing, A. (ed.). 2003. Household Decisions, Gender, and Sitglitz, J. E. 2000. Economics of the Public Sector 3rd ed. New York: Development. Baltimore: Johns Hopkins University Press for W. W. Norton & Co. the International Food Policy Research Institute. Thirtle, C., L. Beyers, L. Lin, V. McKenzie-Hill, X. Irz, S. Wiggins, Reisen, H. 2003. New Sources of Development Finance. An annotated and J. Piesse. 2002. The Impact of Changes in Agricultural conference report. OECD Development Centre. Accessed at Productivity on the Incidence of Poverty in Developing Countries. http://www.oecd.org/dataoecd/56/27/28321228.pdf Report to Department for International Development (DfID) Renkow, M., D. G. Hallstrom, and D. D. Karanja. 2002. "Rural no. 7946. London & East Kilbride, UK. Infrastructure, Transactions Costs and Market Participation Thurlow, J. 2004. "Growth and Market Opportunities in Zambian in Kenya." Journal of Development Economics, 73(1): 349­367. Agriculture." International Food Policy Research Institute, Rodrik, D. 1998. "Trade Policy and Economic Performance in Washington, DC. Sub-Saharan Africa." NBER Working Paper Series, No. 6562. Thurlow, J., and P. Wobst. 2004. "The Road to Pro-Poor Growth in Cambridge, MA. Zambia: Past Lessons and Future Challenges." Forthcoming Rosegrant, M. W., and S. Meijer. 2002. "Appropriate Food Policies Development Strategies and Governance Division Discussion and Investments Could Reduce Child Malnutrition by 43% Paper, International Food Policy Research Institute, Washing- by 2020." Journal of Nutrition 132: 3437S­3440S. ton, D.C. Input paper into the World Bank project "Opera- Rosegrant, M. W., X. Cai, and S. A. Cline. 2002. World Water tionalizing Pro-Poor Growth," Washington, D.C. and Food to 2025: Dealing with Scarcity. Washington, DC: Timmer, C. P. 2003. Agriculture and Pro-poor Growth. Pro-poor International Food Policy Research Institute (IFPRI). Economic Growth Research Studies, no. 12. Bethesda, MD: Rosegrant, M. W., M. S. Paisner, S. Meijer, and J. Witcover. 2001. Development Alternatives, Inc. & Boston Institute for Global Food Projections to 2020. Emerging Trends and Alternative Developing Economies, Ltd. Futures. Washington, DC: International Food Policy Research Tokarick, S. 2002. "How Do Industrial Country Agricultural Institute (IFPRI). Policies Affect Developing Countries." In World Economic Rosegrant, M. W., and P. Hazell. 2000. Transforming the Rural Outlook: Trade and Finance. Washington, DC: International Asian Economy: the Unfinished Revolution. Hong Kong: Oxford Monetary Fund. University Press. ------. 2003. "Measuring the Impact of Distortions in Agri- Runge, C. F., B. Senauer, P. G. Pardey, and M. W. Rosegrant. 2003. cultural Trade in Partial and General Equilibrium." IMF Ending Hunger in Our Lifetime: Food Security and Globalization. Working Paper, WP/03/110. International Monetary Fund, Baltimore: John Hopkins University Press. Washington, DC. 80 Agriculture and Achieving the Millennium Development Goals Torero, M., and S. Chowdhury. 2004. "Infrastructure for AFRICA: WHO/UNICEF. 2001. WHO/UNICEF Joint Monitoring Prog- An Alternative and a Constraint for Development." Brief ramme for Water Supply and Sanitation Coverage Estimates prepared for the 2020 Africa Conference, Assuring Food and 1980­2000. Accessed at UNICEF Statistics on water: http:// Nutrition Security in Africa by 2020: Prioritizing Actions, www.childinfo.org/eddb/water/database.htm. Strengthening Actors, and Facilitating Partnerships. Inter- Winpenny, J. 2003. Financing Water for All. Report of the World Panel national Food Policy Research Institute, Washington, DC. on Financing Water Infrastructure. World Water Council, Torero, M, S. Chowdhury, and V. Galdo. 2003. "Willingness to 3rd World Water Form, Global Water Partnership. Accessed at Pay for the Rural Telephone Service in Bangladesh and Peru." http://www.worldwatercouncil.org/download/Camdessus Information Economics and Policy 15(3): 327­361. Summary.pdf Transparency International. 2004. Transparency International Wolfensohn, J. D. and F. Bourguignon. 2004. Development and Corruption Perceptions Index 2004. Transparency Inter- poverty reduction. Looking back, looking ahead. Prepared national: Berlin, Germany. for the 2004 Annual Meetings of the World Bank and IMF. UNDP (United Nations Development Programme). 2004a. October. Washington, D.C.: World Bank. Opening Doors to Opportunity. Afghanistan's Millennium Dev- Wood, Adrian. 2002. Could Africa Be Like America? Department elopment Goals. New York: UNDP. Accessed at http://www. for International Development. processed. undp.org/mdg/afghanistan/afghanistan_mdg.pdf. World Bank. 1997. World Development Indicators 1997 on CD-ROM. ------. 2004b. Unleashing Entrepreneurship: Making Business Washington, DC: The World Bank. Work for the Poor. New York: UNDP. ------. 1998. East Asia: The Road to Recovery. Washington, DC: UNDP (United Nations Development Program) and UNICEF The World Bank. (United Nations Children Fund). 2002. The Millennium Dev- ------. 2002. World Development Indicators 2002. Washington, DC: elopment Goals in Africa: Promises and Progress. New York: The World Bank. UNDP and UNICEF. ------. 2003. World Development Indicators, 2003. Washington, DC: UNESCO. 1998. UNESCO Institute for Statistics. Accessed at The World Bank. [http://www.uis.unesco.org/ev_en.php?ID=2867_201&ID2 ------. 2004a. Partnerships in Development: Progress in the Fight =DO_TOPIC]. Against Poverty. Washington, DC: The World Bank. UNICEF. 2004. Monitoring the Situation of Children and Women. ------. 2004b. "Global poverty down by half since 1981, but UNICEF global database on child malnutrition. Accessed at progress uneven as economic growth eludes many countries." http://www.childinfo.org/eddb/malnutrition. News Release no. 2004/309/S, April 23, 2004. http://www. Van Blarcom, B., O. Knudsen, and J. Nash. 1993. "The Role of worldbank.org/data/wdi2004/index.htm. Public Expenditures for Agriculture." World Bank Discussion ------. 2004c. World Development Indicators 2004. on CD-ROM. Paper # 216. The World Bank, Washington, DC. Washington, DC: The World Bank. Van De Walle, D. 1996. Infrastructure and Poverty in Viet Nam. ------. 2004d. "Ethiopia Country Brief." Updated June 2004. Washington, DC: The World Bank. Accessed at: www.worldbank.org/ethiopia. von Braun, J. 2003. Berlin Statement. Conference summary rec- ------. 2004e. Zambia Country Economic Memorandum: Policies for ommendations drawn from the international workshop on Growth, Diversification and Poverty Reduction. Washington, Defining the Role of Food Aid in Contributing to Sustainable DC: The World Bank. Food Security, Berlin, September. Zhang, X. 2004. Security is like Oxygen: Evidence from Uganda. Wharton, C. 1967. "The Infrastructure for Agricultural Growth." DSGD Discussion Paper No.6. Washington, DC: IFPRI. In Agricultural Development and Economic Growth. Southworth Zhang, X., and S. Fan. 2000. "Public Investment and Regional and Johnston (eds). Ithaca, NY: Cornell University Press. Inequality in Rural China." Discussion Paper No. 71, Envi- WHO (World Health Organization). 2004. WHO Global Database ronment and Production Technology Division, Inter- on Child Growth and Malnutrition. Accessed at: http://www. national Food Policy Research Institute, Washington, DC. who.int/nutgrowthdb. December. Appendix 1 The Millennium Development Goals, Targets, and Indicators Appendix 1 The Millennium Development Goals, Targets, and Indicators Goal Target Indicators 1. Eradicate extreme 1. Halve, between 1990 and 2015, the 1. Proportion of population below US$1 poverty and hunger. proportion of people whose income is (PPP) per day. less than one dollar a day. 2. Poverty gap ratio [incidence × depth of 2. Halve, between 1990 and 2015, the poverty]. proportion of people who suffer from 3. Share of poorest quintile in national hunger. consumption. 4. Prevalence of underweight children under five years of age. 5. Proportion of population below minimum level of dietary energy consumption. 2. Achieve universal 3. Ensure that, by 2015, children every- 6. Net enrolment ratio in primary education. primary education. where, boys and girls alike, will be able 7. Proportion of pupils starting grade 1 who to complete a full course of primary reach grade 5. schooling. 8. Literacy rate of 15­24 year-olds. 3. Promote gender equality 4. Eliminate gender disparity in primary and 9. Ratios of girls to boys in primary, and empower women. secondary education preferably by 2005 secondary, and tertiary education. and to all levels of education no later 10. Ratio of literate women to men than 2015. 15­24 years old. 11. Share of women in wage employment in the nonagricultural sector. 12. Proportion of seats held by women in national parliament. 4. Reduce child mortality. 5. Reduce by two-thirds, between 1990 and 13. Under-five mortality rate. 2015, the under-five mortality rate. 14. Infant mortality rate. 15. Proportion of 1-year-old children immunized against measles. 5. Improve maternal health. 6. Reduce by three-quarters, between 1990 16. Maternal mortality ratio. and 2015, the maternal mortality ratio. 17. Proportion of births attended by skilled health personnel. 6. Combat HIV/AIDS, 7. Have halted by 2015 and begun to 18. HIV prevalence among 15­24 year old malaria, and other reverse the spread of HIV/AIDS. pregnant women. diseases. 19. Condom use rate. 20. Ratio of school attendance of orphans to school attendance of non-orphans aged 10­14. (continued ) 81 82 Agriculture and Achieving the Millennium Development Goals Appendix 1 The Millennium Development Goals, Targets, and Indicators (Continued) Goal Target Indicators 8. Have halted by 2015 and begun to re- 21. Prevalence and death rates associated verse the incidence of malaria and other with malaria. major diseases. 22. Proportion of population in malaria risk areas using effective malaria prevention and treatment measures. 23. Prevalence and death rates associated with tuberculosis. 24. Proportion of tuberculosis cases detected and cured under directly observed treatment short course (DOTS). 7. Ensure environmental 9. Integrate the principles of sustainable 25. Proportion of land area covered by forest. sustainability. development into country policies and 26. Ratio of area protected to maintain programs and reverse the loss of biological diversity to surface area. environmental resources. 27. Energy use (kg oil equivalent) per US$1 GDP (PPP). 28. Carbon dioxide emissions (per capita) and consumption of ozone-depleting CFCs (ODP tons). 29. Proportion of population using solid fuels. 10. Halve, by 2015, the proportion of people 30. Proportion of population with sustainable without sustainable access to safe access to an improved water source, drinking water and basic sanitation. urban and rural. 11. By 2020, to have achieved a significant 31. Proportion of population with access to improvement in the lives of at least improved sanitation, urban and rural. 100 million slum dwellers. 32. Proportion of households with access to secure tenure (owned or rented). 8. Develop a global 12. Develop further an open, rule-based, Official development assistance (ODA) partnership for predictable, nondiscriminatory trading 33. Net ODA, total and to LDCs, as percentage development. and financial system. Includes a of OECD/Development Assistance commitment to good governance, Committee (DAC) donors' gross development, and poverty reduction-- national income (GNI). both nationally and internationally. 34. Proportion of total bilateral, sector- 13. Address the special needs of the least allocable ODA of OECD/DAC donors to developed countries (LDC). Includes: basic social services (basic education, tariff- and quota-free access for least primary health care, nutrition, safe water developed countries' exports; enhanced and sanitation). program of debt relief for Heavily 35. Proportion of bilateral ODA of Indebted Poor Countries (HIPC) Initiative OECD/DAC donors that is untied. and cancellation of official bilateral debt; 36. ODA received in landlocked developing and more generous ODA for countries countries as proportion of their GNIs. committed to poverty reduction. 37. ODA received in small island developing 14. Address the special needs of landlocked states as proportion of their GNIs. developing countries and small island developing states. (continued ) Appendix 1 The Millennium Development Goals, Targets, and Indicators 83 Appendix 1 The Millennium Development Goals, Targets, and Indicators (Continued) Goal Target Indicators 15. Deal comprehensively with the debt Market access problems of developing countries 38. Proportion of total developed country through national and international imports (by value and excluding arms) measures in order to make debt from developing countries and LDCs, sustainable in the long term. admitted free of duties. 16. In cooperation with developing countries, 39. Average tariffs imposed by developed develop and implement strategies for countries on agricultural products and decent and productive work for youth. textiles and clothing from developing 17. In cooperation with pharmaceutical countries. companies, provide access to affordable, 40. Agricultural support estimate for OECD essential drugs in developing countries. countries as percentage of their GDP. 18. In cooperation with the private sector, 41. Proportion of ODA provided to help make available the benefits of new build trade capacity. technologies, especially information Debt sustainability and communications. 42. Total number of countries that have reached their HIPC decision points and number that have reached their HIPC completion points (cumulative). 43. Debt relief committed under HIPC Initiative, US$. 44. Debt service as a percentage of exports of goods and services. 45. Unemployment rate of 15­24 year-olds, each sex and total. 46. Proportion of population with access to affordable essential drugs on a sustainable basis. 47. Telephone lines and cellular subscribers per 100 population. 48. Personal computers in use per 100 population and Internet users per 100 population. Appendix 2 Impact-Water Developing Countries and Regions LAC--Latin and Central America West Asia and North Africa (WANA) 1. Argentina 1. Arab Republic of Egypt 2. Brazil 2. Turkey 3. Colombia 3. Other West Asian and North African countries: 4. Mexico Algeria, Cyprus, Islamic Republic of Iran, Iraq, 5. Other Latin America: Antigua and Barbuda, Jordan, Kuwait, Lebanon, Libya, Morocco, The Bahamas, Barbados, Belize, Bolivia, Chile, Saudi Arabia, Syrian Arab Republic, Tunisia, Costa Rica, Cuba, Dominica, The Dominican United Arab Emirates, and Republic of Yemen Republic, Ecuador, El Salvador, French Guiana, Grenada, Guadeloupe, Guatemala, Guyana, South Asian Haiti, Honduras, Jamaica, Martinique, Nether- lands Antilles, Nicaragua, Panama, Paraguay, 1. Bangladesh Peru, Saint Kitts and Nevis, Saint Lucia, 2. India Saint Vincent, Suriname, Trinidad and Tobago, 3. Pakistan Uruguay, and Venezuela 4. Other South Asian countries: Afghanistan, Maldives, Nepal, and Sri Lanka Sub-Saharan African Southeast Asia 1. Central and Western Sub-Saharan Africa: Benin, Cameroon, Central African Republic, 1. Indonesia Comoros, Democratic Republic of Congo, 2. Malaysia Republic of Congo, Gabon, The Gambia, 3. Myanmar Ghana, Guinea, Guinea-Bissau, Côte d'Ivoire, 4. Philippines Liberia, São Tomé and Principe, Senegal, 5. Thailand Sierra Leone, and Togo 6. Vietnam 2. Eastern Sub-Saharan Africa: Burundi, Kenya, 7. Other Southeast Asian countries: Brunei, Rwanda, Tanzania, and Uganda Cambodia, and Lao People's Democratic 3. Nigeria Republic 4. Northern Sub-Saharan Africa: Burkina Faso, China: Includes Taiwan and Hong Kong Chad, Djibouti, Eritrea, Ethiopia, Mali, Mauri- tania, Niger, Somalia, and Sudan 5. Southern Sub-Saharan Africa: Angola, Botswana, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Reunion, Swaziland, Zambia, and Zimbabwe 84 Appendix 2 Impact-Water Developing Countries and Regions 85 Appendix Table 2.1 Noncaloric Parameters of the Child Malnutrition Estimation Used for the Baseline and MDG Scenarios Ratio of female-to-male life expectancy at birth Share female secondary schooling 1995 2015 base 2015 MDG 1995 2015 base 2015 MDG Latin America 1.0989 1.0989 1.0900 59.0 60.2 70.3 Sub-Saharan Africa Nigeria 1.0603 1.0690 1.0975 28.9 36.6 57.7 Northern Sub-Saharan Africa 1.0527 1.0670 1.0920 10.6 11.8 26.4 Central and Western Sub-Saharan Africa 1.0534 1.0600 1.0870 17.7 20.9 35.5 Southern Sub-Saharan African 1.0498 1.0450 1.0750 20.6 30.3 41.2 Eastern Sub-Saharan Africa 1.0335 1.0468 1.0650 11.6 15.6 28.9 West Africa North Africa 1.0489 1.0498 1.0510 54.7 66.9 68.3 South Asia India 1.0088 1.0580 1.0850 38.0 50.8 94.7 Pakistan 1.0327 1.0388 1.0770 18.5 28.2 46.3 Bangladesh 1.0072 1.0550 1.0870 13.4 21.1 33.5 Other South Asia 1.0241 1.0380 1.0570 38.7 57.6 46.4 South East Asia Indonesia 1.0580 1.0598 1.0770 47.6 56.9 71.4 Thailand 1.0659 1.0772 1.0800 53.6 63.3 75.1 Malaysia 1.0692 1.0699 1.0850 62.6 75.1 78.2 Philippines 1.0639 1.0690 1.0746 77.8 88.4 95.0 Vietnam 1.0687 1.0777 1.0793 46.0 52.8 92.0 Myanmar 1.0849 1.0734 1.0988 33.1 41.0 82.7 Other South East Asia 1.0536 1.0690 1.0650 21.4 43.1 53.6 China 1.0467 1.0498 1.0650 61.8 70.2 74.1 Notes: Latin America includes the Caribbean; N SSA stands for Northern Sub-Saharan Africa; C and W SSA stands for Central and Western Sub-Saharan Africa; S SSA stands for Southern Sub-Saharan Africa; and E SSA stands for Eastern Sub-Saharan Africa; WANA stands for West Asia and North Africa; Other S Asia stands for Other South Asia; Other SE Asia stands for Other Southeast Asia. The regional disaggregation is presented in Annex 2. 2015 baseline values are interpolated from IFPRI IMPACT 2020 estimates. 2015 MDG scenario values are changed to attempt to reach the child malnutrition indicator of MDG 1. Sources: 1995 values for the ratio of female to male life expectancy and for female secondary schooling based on World Development Indicators (World Bank 2004c).