S U C C E S S E S IN AFRICAN AGRICULTURE LESSONS FOB THE F i n RE Sub-Saharan Africa is one of the poorest regions of the world. Because most Africans work in agricul- ture, increased agricultural productivity offers an unusually powerful lever for raising the incomes of the majority of poor households, lowering food prices, and stimulating growth in other eco- nomic sectors. Per capita agricultural production in Sub-Saharan Africa has fallen, however, for much of the past half-century. Successes in African Agriculture in- vestigates how to reverse this decline. Instead of cataloging failures, as many past studies have done, this book identifies episodes of successful agri- cultural growth in Africa and identifies processes, practices, and policies for accelerated growth in the future. The individual studies follow developments in, among other areas, the farming of maize in East and Southern Africa, cas- sava across the middle belt of Africa, and cotton in West Africa; horticulture in Kenya; and dairying in East Africa. Drawing on these case studies and on consultations with agricultural specialists and politicians from across Sub-Saharan Africa—undertaken in collaboration with the African Union's New Partnership for Africa's Develop- ment—the contributors identify two key determinants of positive agricultural performance: agricultural research to provide more productive and sustain- able technologies to farmers and a policy framework that fosters market incentives for increasing production. Successes in African Agriculture Other Books Published in Cooperation with the International Food Policy Research Institute IFPRI Conditional Cash Transfers in Latin America Edited by Michelle Adato and John Hoddinott From Parastatals to Private Trade: Lessons from Asian Agriculture Edited by Shahidur Rashid, Ashok Gulati, and Ralph Cummings Jr. Public Expenditures, Growth, and Poverty: Lessons from Developing Countries Edited by Shenggen Fan The Dragon and the Elephant: Agricultural and Rural Reforms in China and India Edited by Ashok Gulati and Shenggen Fan WTO Negotiations on Agriculture and Developing Countries Anwarul Hoda and Ashok Gulati Transforming the Rural Nonfarm Economy: Opportunities and Threats in the Developing World Edited by Steven Haggblade, Peter B. R. Hazell, and Thomas Reardon Agricultural Research, Livelihoods, and Poverty: Studies of Economic and Social Impacts in Six Countries Edited by Michelle Adato and Ruth Meinzen-Dick Information and Communication Technologies for Development and Poverty Reduction: The Potential of Telecommunications Edited by Maximo Torero and Joachim von Braun What's Economics Worth? Valuing Policy Research Edited by Philip Pardey and Vincent H. Smith Land and Schooling: Transferring Wealth across Generations Agnes R. Quisumbing, Jonna P. Estudillo, and Keijiro Otsuka Household Decisions, Gender, and Development: A Synthesis of Recent Research Edited by Agnes R. Quisumbing Ending Hunger in Our Lifetime: Food Security and Globalization C. Ford Runge, Benjamin Senauer, Philip G. Pardey, and Mark W. Rosegrant The Triangle of Microfinance: Financial Sustainability, Outreach, and Impact Edited by Manfred Zeller and Richard L. Meyer Successes in African Agriculture Lessons for the Future EDITED BY STEVEN HAGGBLADE AND PETER B. R. HAZELL Published for the International Food Policy Research Institute The Johns Hopkins University Press Baltimore © 2010 International Food Policy Research Institute All rights reserved. Published 2010 Printed in the United States of America on acid-free paper 9 8 7 6 5 4 3 2 1 The Johns Hopkins University Press 2715 North Charles Street Baltimore, Maryland 21218-4363 www.press.jhu.edu International Food Policy Research Institute 2033 K Street, NW Washington, DC 20006 (202) 862-5600 www.irpri.org IFPRI's research, capacity strengthening, and communications work is made possible by its financial contributors and partners. IFPRI gratefully acknowledges the generous unrestricted funding from Australia, Canada, China, Finland, France, Germany, India, Ireland, Italy, Japan, the Netherlands, Norway, the Philippines, Sweden, Switzerland, the United Kingdom, the United States, and the World Bank. L I B R A R Y O F C O N G R E S S C A T A L O G I N G - I N - P U B L I C A T I O N D A T A Successes in African agriculture : lessons for the future / edited by Steven Haggblade and Peter B. R. Hazell. p. cm. Includes bibliographical references and index. ISBN-13: 978-0-8018-9502-9 (hardcover : alk. paper) ISBN-10: 0-8018-9502-2 (hardcover : alk. paper) ISBN-13: 978-0-8018-9503-6 (pbk.: alk. paper) ISBN-10: 0-8018-9503-0 (pbk. : alk. paper) 1. Agricultural development projects—Africa, Sub-Saharan—Case studies. 2. Agricultural productivity—Africa, Sub-Saharan—Case studies. 3. Agricultural assistance—Africa, Sub-Saharan—Case studies. I . Haggblade, Steven. I I . Hazell, P. B. R. HD2117.S84 2010 338.1'867—dc22 2009035572 A catalog record for this book is available from the British Library. Special discounts are available for bulk purchases of this book. For more information, please contact Special Sales at 410-516-6936 or specialsales@press.jhu.edu. The Johns Hopkins University Press uses environmentally friendly book materials, including recycled text paper that is composed of at least 30 percent post-consumer waste, whenever possible. All of our book papers are acid-free, and our jackets and covers are printed on paper with recycled content. Contents List of Figures ix List of Tables xi List of Boxes xv Foreword xvii Acknowledgments xix Acronyms and Abbreviations xxi PART I Overview 1 1 Challenges for African Agriculture 3 S T E V E N H A G G B L A D E , P E T E R B . R . H A Z E L L , A N D E L E N I G A B R E - M A D H I N PART II Success Stories 27 2 The Cassava Transformation in West and Southern Africa 29 F E L I X N W E K E A N D S T E V E N H A G G B L A D E 3 "Seeds of Success" in Retrospect: Hybrid Maize in Eastern and Southern Africa 71 M E L I N D A S M A L E A N D T H O M A S S . J A Y N E 4 Mali's White Revolution: Smallholder Cotton, 1960-2006 113 J A M E S T E F F T 5 Are Horticultural Exports a Replicable Success Story? Evidence from Kenya and Cote dTvoire 163 N I C H O L A S M I N O T A N D M A R G A R E T N G I G I 6 Smallholder Dairying in Eastern Africa 209 M A R G A R E T N G I G I , M O H A M E D A B D E L W A H A B A H M E D , S I M E O N E H U I , A N D Y E M E S R A C H A S S E F A vi i i Contents 7 Sustainable Soil Fertility Management Systems 262 S T E V E N H A G G B L A D E , G E L S O N T E M B O , D A N I E L K A B O R E , C H R I S R E D , O L U Y E D E C . A J A Y I , S T E V E N F R A N Z E L , P A R A M U M A F O N G O Y A , A N D F R A N K P L A C E PART II I Lessons for the Future 321 8 Lessons from Past Successes 323 S T E V E N H A G G B L A D E 9 Implications for the Future 349 S T E V E N H A G G B L A D E , P E T E R B . R . H A Z E L L , A N D W I L B E R F O R C E K I S A M B A - M U G E R W A References 373 Contributors 421 Index 423 Figures 1.1 Trends in the value of agricultural production per capita, 1961-2006 5 1.2 Trends in agricultural factor productivity, 1961-2003 10 1.3 Case study countries 18 1.4 The DE-A-R framework 20 2.1 Africa's cassava belt, 1980 30 2.2 Cassava production trends in Nigeria and Ghana, 1961-2006 32 2.3 Technology adoption in Nigeria's 65 COSCA villages 38 2.4 Inflation-adjusted price of gari in Edo State, Nigeria, 1971-98 44 2.5 Trends in cassava production in Malawi and Zambia, 1961-2006 51 3.1 Per capita grain production in Zimbabwean communal lands, 1914-94 80 3.2 Public expenditures supporting smallholder maize yields in Zimbabwe during episodes of success and decline, 1970-2000 93 3.3 Maize area and maize marketed by the Zambian small-scale and medium-scale farming sector, 1984-2005 95 3.4 Changes in area distribution among maize and competing major crops in Malawi, 1982-2005 96 3.5 Maize production, hybrid maize diffusion, exogenous shocks, and policy changes in Malawi, 1981-2001 100 4.1 Cotton zones of Mali, 2003 116 4.2 World market price of cotton fiber, 1960-2006 118 4.3 Effect of world market price on farmers' price share, 1960-2005 119 4.4 Cotton area and production in Mali, 1961-2006 120 4.5 Cotton yield by variety and year, 1960-2002 124 4.6 Trends in seed cotton yields in Mali, 1961-2006 134 4.7 Input-output price ratios among Malian cotton farmers, 1980-2008 134 ix x Figures 4.8 Estimated total unit cost of cotton fiber production in Mali, 1986-2002 146 5.1 Value of Kenyan fruit and vegetable exports and share of agricultural export revenue, 1961-2006 170 5.2 Trends in the composition of Kenyan fruit and vegetable exports, 1961-2006 173 5.3 Export marketing channels for Kenya's green beans, 2001 180 5.4 Value of Ivorian fruit and vegetable exports and share of agricultural export revenue, 1961-2006 196 5.5 Trends in the composition of Ivorian fruit and vegetable exports, 1961-2007 198 6.1 Population density in the three study countries, 2005 210 6.2 Growth in Kenya's annual smallholder milk production, 1957-64 220 6.3 Kenya's milk supply pattern, 1986-91 227 6.4 Kenya's milk marketing channels, 2002 230 6.5 Trends in Kenya's annual milk production, 1961-2006 235 6.6 Trends in Kenya Cooperative Creamery milk intake, 1971-2000 235 6.7 Change in sources of milk supply for greater Kampala, 1961-68 243 6.8 Uganda's milk marketing channels, 2002 246 6.9 Ethiopian dairy marketing channels, 2002 248 6.10 Total and per capita annual milk production in Ethiopia, 1961-2006 253 7.1 Timing of land preparation, by tillage method, Zambia, 2001 275 7.2 Declining labor requirements over time for digging conservation farming basins, 2003 279 7.3 Distribution of zai in Burkina Faso, 2002 288 7.4 Number of farmers planting improved fallows in Zambia, 1995-2006 304 7.5 Adoption of improved fallows in Zambia, by wealth category, 2001 309 9.1 Domestic and border wholesale prices for white maize grain, Randfontein, South Africa, 1996-2006 358 9.2 Trends in donor aid for African agriculture, 1975-2007 364 9.3 Mean farm size by continent, 1930-90 366 Tables 1.1 Indicators of agricultural performance and welfare 4 1.2 The successes in African agriculture research process 15 1.3 Expert survey success nominations 16 1.4 Case study selections 17 1.5 Measuring results 21 1.6 Case study protocol 22 2.1 Dynamics and drivers of change in Nigeria's cassava transformation 34 2.2 Income elasticity of demand for cassava and other food staples in Nigeria and Ghana 36 2.3 Returns to cassava production and processing in Nigeria, 1991 43 2.4 Retail price of 1,000 calories from fresh cassava roots and maize in rural market centers of Nigeria and Ghana, 1992 44 2.5 Percentage of cassava fields in which women and men provide the bulk of each kind of labor 47 2.6 Dynamics and drivers of change in Zambian cassava 52 2.7 Release of improved cassava varieties in Malawi and Zambia 57 2.8 Changing incentives for cassava producers in Zambia 62 2.9 Impact of household labor availability and HIV/AIDS on cassava and maize production 64 2.10 Size distribution of cassava area in Malawi and Zambia (percent) 66 2.11 Gender distribution of cassava production in Malawi, 1998 66 2A. 1 Trends in cassava production (5-year trailing averages) 70 3.1 Selected production and consumption statistics for the maize case study countries, 2000-2005 73 3.2 Dynamics and drivers of change in maize production in Eastern and Southern Africa 74 3.3 Percentage of national maize area in improved maize for the case study countries, 1990-2007 102 xi xi i Tables 3.4 National maize area, yield, and production growth rates during and after periods of rapid production growth 103 3.5 Illustrative partial budget for smallholder maize producers in Malawi, during and after the periods of maize growth 105 3.6 Concentration of maize sales among smallholder farm households in Kenya (1999/2000), Malawi (2003/04), and Zambia (2000/2001) 107 4.1 Share of francophone Africa in world cotton production and trade (percent) 114 4.2 Annual growth rates in Malian cotton, by period (percent) 121 4.3 Dynamics and drivers of change in Malian cotton, by period 123 4.4 Cereal production, by zone, 1998 (kilograms per person per year) 130 4.5 Change in area planted by Malian cotton producers 13 8 4.6 Evolution of farm equipment levels and land holdings among Malian cotton farmers 138 4.7 Costs and returns to cotton production in Mali 141 5.1 Dynamics and drivers of change in Kenyan horticulture, 1895-present 166 5.2 Composition of Kenyan exports of fruits and vegetables, 2005 177 5.3 Summary of measures of importance of each crop, 2000 (percentage of production) 182 5.4 Fruit and vegetable production by income category, 2000 184 5.5 Fruit and vegetable production by farm size, 2000 185 5.6 Comparison of French bean growers and other farmers, 2000 191 5.7 Gross margins in Kenyan horticulture and staple food crop production, 2005 192 5.8 Percentage of farms growing selected fruits and vegetables by income category, 2000 194 5.9 Percentage of farms growing each crop by farm size, 2000 194 6.1 Selected indicators for the dairy sector in Kenya, Uganda, and Ethiopia, 2006 211 6.2 Dynamics and drivers of change in Kenya's dairy subsector, 1900-2006 214 6.3 Trends in artificial insemination charges, 1980-90 224 6.4 Marketing margins in Kenya's formal and raw milk supply channels, 2000 233 6.5 Long-term trends in Kenyan dairy production, 1980-2000 236 6.6 Value of dairy production in Kenya as a percentage of total farm household income, 2000 237 6.7 Distributional impact of dairy income in Kenya, 2000 238 6.8 Prevalence of the main dairy production systems in Kenya, 2000 239 Tables x i i i 6.9 Milk production costs in Kenya by feed management regime, 1999 240 6.10 Factors influencing the choice of dairy feed management system in Kenya, 2000 241 6.11 Structure of demand for milk products in Ethiopia, 2000 247 6.12 Trends in Ethiopian milk production, 1961-2006 251 6.13 Changing structure of Ethiopian milk production and distribution systems, 1985 versus 2000 255 6.14 Gross margin analysis for cross-bred and local cows in Ethiopia, 2003 257 6.15 Profile of adopters and nonadopters of improved, market-oriented dairy cattle in Holetta, Ethiopia, 2002 258 7.1 Dynamics and drivers of change in Zambian conservation farming, 1964-present 265 7.2 Adoption of conservation farming basins in Zambia, 2001/02 and 2002/03 271 7.3 Differences in output and inputs for cotton and maize under conservation farming and conventional tillage, central and southern Zambia, 200.1 /02 273 7.4 Determinants of yield for cotton and maize under conservation farming in central and southern Zambia, 2001/02 276 7.5 Sources of yield gains in Zambian conservation farming of cotton and maize, 2001/02 278 7.6 Returns to conservation farming and conventional farming 280 7.7 Returns to animal draft technologies in maize and cotton plots, eastern and southern Zambia, 2004/05 281 7.8 Dynamics and drivers of change in the zai planting pits in Burkina Faso, 1930-2005 283 7.9 The impact of planting pits (tassa) on cereal yields in Niger's Illela district, 1991-96 288 7.10 Crop production budgets with zai planting pits in Burkina Faso, 2003 290 7.11 Dynamics and drivers of improved fallows in Eastern Zambia and Western Kenya, early 1960s-2003 294 7.12 Trends in soil fertility restoration systems in non-pilot project areas of Western Kenya, 1997-2001 305 7.13 Returns to labor and land under alternative soil fertility restoration systems in Eastern Zambia, 1996-2002 (constant 2005 US$) 306 7.14 Factors affecting the use of soil fertility replenishment systems in 17 villages in Western Kenya, 1997-2001 310 7.15 Effects of land use system on soil physical properties after eight years of improved fallow-crop rotations at Msekera, Chipata-Zambia, 1998 313 xiv Tables 7.16 Nutrient balances of improved fallows at Msekera Research Station, following eight years of fallow-crop rotation, 1998-2002 (kg/ha) 314 7.17 Contrasts and similarities in the four soil fertility case studies 315 8.1 Case study summaries 326 8.2 Case study impact summary 338 9.1 Trends in agricultural research and development spending in Sub-Saharan Africa, 1960s-2000 351 9.2 African government budget allocations for agriculture, 2004 363 Boxes 2.1 A small farmer tests the new cassava varieties 58 2.2 Commercial cassava production in central Malawi 59 2.3 Hedging against cassava mosaic virus 63 3.1 Farmers'perceptions 86 5.1 An accounts clerk turns to passion fruit 181 5.2 A long-time horticultural grower adjusts to changing conditions 183 5.3 Diversification from staples to vegetables 188 6.1 The tree that bears fruit every day 222 6.2 A Kenyan retiree turns to commercial dairy production 225 7.1 A recent convert to conservation farming basins 269 7.2 A six-year conservation farming veteran 272 7.3 A pioneer experimenter with improved fallows 298 7.4 A farmer's comparison of improved fallows and mineral fertilizer 312 xv Foreword Africa stands at the dawn of a new era. In spite of the global financial meltdown and spiraling food prices, the continent has shown remarkable resilience. On the economic front, Africa has achieved remarkable progress in recent years. Reforms have opened up political and economic space for the private sector, including traders and farmers, and have allowed civil-society organizations to actively participate in national agriculture development. Eighteen countries in the region have maintained average economic growth of 5.5 percent per year during the past decade, while more than half of the continent's countries have recorded agricultural growth exceeding 4 percent during the past five years. Africa is moving toward social and political stability, albeit slowly. Africa's impressive gains, however, were not enough to compensate for more than two decades of economic stagnation and decline before the recent recovery. As a result, many African countries are still facing high levels of poverty and malnutrition. Increasingly volatile world markets, coupled with global environmental change and increased population pressure on arable land, wi l l increase the challenges for rural communities in Africa. To successfully combat hunger in the region, future efforts must focus more on agriculture. More than two-thirds of Africa's poor work primarily in agriculture, yet agricultural productivity in Africa remains the lowest in the world. Clearly, agricultural productivity must increase i f Africa is to improve the welfare of the majority of its citizens. Only increased agricultural produc- tivity can simultaneously reduce food prices—which govern real incomes and poverty in urban areas—and increase the incomes of the majority of Africa's poor. For this reason, agricultural growth provides a central thrust around which the battle against African poverty must be waged. In recognition of this fact, Africa's leaders adopted the Comprehensive African Agricultural Development Programme (CAADP) in 2003 and agreed to raise budget allocations for agri- culture to a minimum of 10 percent of public spending. To help translate this new financial commitment into more effective ac- tion on the ground, the International Food Policy Research Institute (IFPRI) and the New Partnership for Africa's Development (NEPAD) Secretariat have col- xvii xviii Foreword laborated with a host of other partners to systematically examine past episodes of successful agricultural development in Africa. Indeed, in the past, African agriculture has seen moments of great promise. In 1960, maize breeders in Africa released the first commercially grown single-cross hybrids in the world. In 1977, cassava breeders in Nigeria released the first of the Tropical Manioc Selection (TMS) varieties that have subsequently revolutionized breeding lines and cassava productivity across the tropical midsection of Africa. Additionally, West African farmers have increased cotton production and exports at a com- pound annual rate of 9 percent per year for more than four decades. This book explores these case studies, as well as others, in an effort to identify key ele- ments driving successful agricultural growth in Africa. By identifying the in- stitutions, investments, processes, and policies that have made past successes possible, the authors provide a basis for fostering partnerships and an enabling environment in which African farmers can raise their productivity and welfare more consistently in the future. Given newfound commitment to agriculture by African leaders and the community of development partners, we believe that the time is ripe to review, reflect, and build upon what has worked well in the past. We hope this book wi l l contribute to these ongoing efforts to invigorate and accelerate growth in African agriculture. Joachim von Braun Richard Mkandawire Director General, IFPRI Agriculture Advisor, NEPAD Secretariat Acknowledgments This volume covers a broad geographic and historical landscape. Given the complexity and diversity of African farming systems, a review of this sort would not have been possible without strong support from a variety of sources. In launching this work, we are grateful to Per Pinstrup-Andersen, who strongly supported this major review effort while director general of the Inter- national Food Policy Research Institute (IFPRI) and allocated core IFPRI fund- ing to finance it. His successor, Joachim von Braun, has likewise proven highly supportive, particularly in facilitating IFPRI support for the series of stake- holder consultations and coordination with the New Partnership for Africa's Development (NEPAD) Secretariat. Eleni Gabre-Madhin served as one of the two original research coordinators. In that capacity, she played a key role in helping to execute the expert survey, develop the analytical framework, launch the case study teams, and organize the case study workshop. Our distinguished external advisory group included Michel Benoit-Cattin, Josue Dione, Simeon Ehui, Carl Eicher, and Wilberforce Kisamba-Mugerwa. We are grateful to all of them for the thoughtful, constructive guidance they pro- vided throughout this effort. Their broad experience proved particularly help- ful in guiding the selection of a portfolio of case studies that would offer both contrast and illumination. The empirical foundations of this research have depended largely on ac- cessing and analyzing existing data sources, published material, and a large gray literature. In this effort, the case study authors proved highly entrepreneurial. Ghada Shields likewise provided exceptionally able backstopping and refer- ence support to the field teams. In assembling aggregate data of various sorts, the analytical team acknowledges critical assistance from Ming Chen, Michael Johnson, Stanley Wood, and Ulrike Wood-Sichra. To solicit feedback and practical insights from policymakers, farmers, agribusiness, and other stakeholder groups, the analytical team has worked closely with Richard Mkandawire, Agriculture Advisor at the NEPAD Secre- tariat, in conducting a series of three major stakeholder consultations. He and his colleagues at the Secretariat, Ingrid Kirsten and Augustin Wambo, demon- xix xx Acknowledgments strated extraordinary organizational skills, an exceptional capacity for hard work, and a high degree of diplomacy and congeniality in managing these work- shops. Jergen Richter and his colleagues at Internationale Weiterbildung und Entwicklung (InWEnt) provided outstanding professional facilitators for the three stakeholder workshops. They ensured that all participants' viewpoints were expressed in the professionally facilitated small-group break-out sessions that formed the backbone of these workshops. The team is grateful to InWEnt, the Technical Centre for Agricultural and Rural Cooperation (CTA), the International Fund for Agricultural Development, Centre de Cooperation Internationale en Recherche Agronomique pour le Developpement, and the host governments of Kenya and South Africa for material and financial support in running the three stakeholder consultations in Pretoria, Nairobi, and Somerset West. Finally, we thank the conference participants—the government policy- makers, farmers, private agribusiness groups, agricultural researchers, donors, and operational project personnel—who took the time to review our preliminary research findings and help us distill the key lessons emerging from this work. Acronyms and Abbreviations A A D I Addis Ababa Dairy Industry ACP African, Caribbean, and Pacific ACU Acceleration of Cassava Utilization Task Force ADMARC Agricultural Development and Marketing Corporation ADPs agricultural development programmes AER agroecological region A I artificial insemination AR African Reserve ASARECA Association for Strengthening Agricultural Research in Eastern and Central Africa A U African Union AVs associations villageoises Bt Bacillus thuringiensis CAADP Comprehensive African Agricultural Development Programme CAIS Central Artificial Insemination Service CF conservation farming CFA Communaute Financiere Africaine CFAF CFA franc CFDT Compagnie Fran9aise de Developpement des Textiles CFU Conservation Farming Unit CGM cassava green mite CIMMYT Centra Internacional de Mejoramiento de Maiz y Trigo CIRAD Centre de Cooperation Internationale en Recherche Agronomique pour le Developpement CLUSA Cooperative League of the USA CM cassava mealybug CMDT Compagnie Malienne pour le Developpement des Textiles CMV cassava mosaic virus xxi xxii Acronyms and Abbreviations COMESA Common Market for Eastern and Southern Africa COPACO Compagnie Cotonniere CORFRUITEL Cooperative de Producteurs pour la Commercialization des Fruits et Legumes de la Cote d'lvoire COSCA Collaborative Study of Cassava in Africa CTA Technical Centre for Agricultural and Rural Cooperation DDA Dairy Development Agency DDE Dairy Development Enterprise DE-A-R framework Decisionmaking Environment (DE)-Action (A)-Results (R) framework DFCS Dairy Farmer Cooperative Societies DRC Democratic Republic of Congo DRSK Dairy Record Service of Kenya EAC East African Community EPAs Economic Partnership Agreements FAO Food and Agriculture Organization of the United Nations FARA Forum for Agricultural Research in Africa FPEAK Fresh Produce Exporters Association GART Golden Valley Agricultural Research Trust GATT General Agreement on Trade and Tariffs GDP gross domestic product GMB Grain Marketing Board GMOs genetically modified organisms HCDA Horticultural Crops Development Authority HYV high-yielding variety ICAC International Cotton Advisory Council ICRAF World Agroforestry Centre ICRISAT International Crops Research Institute for the Semi-Arid Tropics IER Institut d'Economie Rurale IFPRI International Food Policy Research Institute IITA International Institute of Tropical Agriculture InWEnt Internationale Weiterbildung und Entwicklung GmbH, or Capacity Building International IOPVs improved open-pollinated varieties IP intellectual property IPM integrated pest management IRCT Institut de Recherche Cotonniere et des Fibres Textiles Exotiques ISI import-substituting industrialization KAIS Kenya Artificial Insemination Service Acronyms and Abbreviations xxii i KARI Kenya Agricultural Research Institute KCC Kenya Cooperative Creameries KDB Kenya Dairy Board KEFRI Kenya Forestry Research Institute KFP Kenya Fruit Processing KNAIS Kenya National Artificial Insemination KSB Kenya Stud Book LPA Lagos Plan of Action LRC Livestock Recording Centre MCB Maize Control Board MDG Millennium Development Goal M K Malawi kwacha NAMBOARD National Agricultural Marketing Board NEPAD New Partnership for Africa's Development NGOs nongovernmental organizations NRM natural resource management NSCM National Seed Company of Malawi OCAB Office Centrale des Producteurs-Exportateurs d'Ananas et de Bananes OECD Organisation for Economic Co-operation and Development PICPE Presidential Initiative on Cassava Production and Export R&D research and development RTIP Root and Tuber Improvement Programme SACCAR Southern African Centre for Cooperation in Agricultural Research and Training SAP structural adjustment program SCAER Societe de Credit Agricole et d'Equipement Rural SDFs small-scale dairy farmers SFC State Farms Corporation Shell-BP Shell BP Petroleum Company of Nigeria SIDA Swedish International Development Cooperation Agency SP Starter Pack SRDP Smallholder Rehabilitation and Development Programme SSA Sub-Saharan Africa swc soil and water conservation SYCOV Union of Cotton and Food Crop Producers TIP Targeted Input Program TMS Tropical Manioc Selection ULV ultra-low-volume xxiv Acronyms and Abbreviations UNICEF United Nations Children's Fund USAID United States Agency for International Development WECARD/CORAF West and Central African Council for Agricultural Research and Development WFP World Food Programme WTO World Trade Organization ZIAP Zambia Integrated Agroforestry Project ZNFU Zambia National Farmers Union P A R T I Overview 1 Challenges for African Agriculture STEVEN HAGGBLADE, PETER B. R. HAZELL, AND ELENIGABRE-MADHIN Motivation Needs Sub-Saharan Africa faces critical challenges.1 More than 40 percent of all Africans live on less than US$1 per day, and a staggering one out of three is undernourished (Table 1.1). Yet 70 percent of the continent's poor work in agri- culture.2 Clearly, Africa's agriculture has underperformed. Under "business- as-usual" projections for agriculture, Africa wi l l increasingly depend on food imports to meet its basic consumption requirements. Only in Africa, among all developing regions, are the numbers of malnourished children projected to in- crease over the next two decades (Rosegrant et al. 2001, 2005). Although many factors contribute to Africa's persistent hunger and poverty, poor agricultural performance lies at the heart of the problem. On av- erage, agriculture accounts for 65 percent of full-time employment in Africa, 25-30 percent of gross domestic product (GDP), and over half of total export earnings (IFPRI 2004; World Bank 2007b).3 It underpins the livelihoods of over two-thirds of Africa's poor and assumes even greater importance in the conti- nent's poorest countries, such as Burundi, Ethiopia, Malawi, and Tanzania. Yet Africa's agricultural performance over the past 45 years has ranked worst in the world according to most conventional measures. Given low levels of land and labor productivity, African farmers produce output per capita valued at only half 1. For brevity, the discussion hereafter will use Africa when referring to Sub-Saharan Africa. 2. Complete data on the sectoral composition of poor households' income do not exist for all of Sub-Saharan Africa. This estimate relies on work by Ravallion, Chen, and Sangraula (2007), who project the percentage of poor households residing in rural and urban parts of Sub-Saharan Africa; by Valdes et al. (2009), who provide breakdowns of the sectoral composition of income among poor rural households; and Garret (2004) and others, who provide evidence on the preva- lence of urban agriculture among poor African households. 3. Weighting by GDP for all countries outside of South Africa (which alone accounts for 40 percent of Sub-Saharan Africa's aggregate GDP) produces an agricultural share of 25 percent. Weighting by population produces an agricultural GDP share of 30 percent. 3 4 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin T A B L E 1.1 Indicators of agricultural performance and welfare Sub- Saharan Africa South Asia East Asia Latin America Agriculture Cereal yields, 2005 (tons/ha) 0.9 Value of agricultural production per farm population, 2005 (US$) $225 Food aid cereals, 2000-03 (kg/capita) 4.6 Welfare Poverty headcount, 2004 (% living on less than US$1 per day) 41 31 Undernourishment, 2004 (% of population) 32 21 Malnutrition, 2004 (% children under 5 underweight for age) 29 45 Per capita income, 2005 (US$) $746 $692 Aid per capita, 2005 (US$) $44 $6 Population Population growth rate, 2006 (% per year) • 2.3 1.5 Population, 2005 (millions) 743 1,470 2.8 $446 1.1 12 15 $1,630 $5 0.8 1,886 3.0 $2,105 1.1 9 10 7 $4,045 $11 1.3 551 S O U R C E S : World Bank (2007b), FAOSTAT (2008). N O T E : The agricultural data for South Asia and East Asia refer to developing Asia. the level achieved in developing Asia, while cereal yields in Africa attain only one-third of the level prevailing in developing Asia and Latin America (see Table 1.1). Africa likewise remains the only developing region where per capita agricultural production has fallen over the past four and a half decades (Figure 1.1). Today, in comparison with other developing regions of the world, Africa lays claim to the world's lowest agricultural productivity, its highest incidence of poverty, and per capita food aid quadruple that of other developing regions (see Table 1.1). Opportunities Increased agricultural productivity offers a potentially powerful tool for spear- heading broad-based income gains among Africa's poor (see World Bank 2007a; Diao et al. 2008). On a continent where 70 percent of the poor work in Challenges for African Agriculture 5 F I G U R E 1.1 Trends in the value of agricultural production per capita, 1961-2006 Index(1961 = 100) 250 I Sub-Saharan Africa Developing Asia 200 Latin America 150 - 100 - 50 - 0 I 1 1 ' 1 ' 1 ' ' >- A * A qS o!' # cf> # (fc -=f ^ ^ ^ % ^ ^ ^ S O U R C E : FAOSTAT (2008). agriculture, an upsurge in farm productivity contributes directly to raising rural living standards. In addition, a prosperous agriculture generates power- ful growth linkages to the rest of the economy, providing cheap food, raw ma- terials, and a growing demand for nascent processing and service industries (Mellor 1976; Haggblade, Hazell, and Brown 1989; Haggblade, Hazell, and Dorosh 2007). Even the urban poor, who spend the majority of their income on food, see their real incomes rise when growing agricultural productivity and output enable reductions in staple food prices. Likewise, the many rural households that remain net purchasers of food—a majority in many locations— benefit from rising farm productivity and falling food prices (Jayne, Mather, and Mghenyi 2006). Consequently, growing agricultural productivity attacks poverty from three different directions. It increases the productivity and in- comes of the majority of Africa's poor, who work primarily in agriculture. It re- duces food prices, which govern real incomes and poverty in urban areas. And it generates important spillovers to the rest of the economy. Despite Africa's weak past performance, many agricultural specialists see significant potential for agricultural growth in Africa. 4 The continent is blessed with abundant natural resources, 12 times the land area of India with only two-thirds as many people to feed. Growing urbanization, from currently low levels, coupled with high rates of overall population growth, portends rapidly growing domestic and regional food markets. With few exceptions, African 4. See, for example, InterAcademy Council (2004), FAO (2005), and World Bank (2009). 6 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin land distribution is still equitable by international standards (Valdes et al. 2009). Small farms, efficient but poor, dominate the continent. Agricultural proponents likewise take heart from recent improvements in Africa's policy environment. The structural adjustment programs of the 1980s and 1990s have removed the worst of the previous biases against agriculture and have improved incentives for agricultural investments (Binswanger-Mkhize and McCalla 2008; Ander- son and Masters 2009). Agricultural specialists similarly see opportunities for raising Africa's currently low yields through technological change. Some of the required technologies are already available, and modern science is opening up new opportunities to increase agricultural productivity, even in countries and regions that have not benefited significantly from new technologies in the past. The recent surge in world food prices, at least partly sustained by the biofuels policies of the United States and the European Union, seems likely to create new market opportunities for African farmers while at the same time adding ur- gency to the imperative for more rapid agricultural growth. Recognizing these potential gains, a growing contingent of African lead- ers and technical specialists has become convinced that enhanced agricultural performance wi l l constitute a necessary centerpiece for broad-based poverty re- duction efforts. African leaders, through the African Union's (AU's) New Part- nership for Africa's Development (NEPAD), have highlighted the critical im- portance of accelerating agricultural growth in Africa. In 2003 they launched a Comprehensive African Agricultural Development Programme (CAADP) to spearhead agricultural development efforts at a continental level (AU/NEPAD 2003a). Many donors and Africa specialists, likewise, consider agriculture fun- damental to broad-based economic growth and poverty reduction in Africa. 5 Impediments But past lackluster performance in African agriculture has discouraged some professionals, even those who remain convinced of agriculture's potential im- portance as a poverty fighter. They question whether African agriculture can grow as rapidly as required, given a variety of historical, geographic, epidemio- logical, political, and world market handicaps.6 Geography and natural resource endowments give rise to a series of con- cerns. Africa's old and weathered soils, primarily kaolinitic clay, provide poor nutrient and water retention capacity. Its tropical climate precludes freezing winter temperatures that in temperate latitudes help to control pests and frac- ture soil clods and plow pans to facilitate plant root development (Lowe 1986; 5. Borlaug (1996), Partnership to Cut Hunger (2002), AU/NEPAD (2003a), InterAcademy Council (2004), FAO (2005), Valdes and Foster (2005), World Bank (2007a, 2009), Alliance for a Green Revolution in Africa (AGRA 2009). 6. See, for example, Bloom and Sachs (1998), Diamond (1998), Maxwell, Urey, and Ash- ley (2001), and Ellis (2005). Challenges for African Agriculture 7 Masters and McMillan 2001). Endowed with a paucity of domesticable plant and animal species, African farmers operated for many millennia with a re- stricted agricultural genetic base (Diamond 1998). Given the continent's lim- ited irrigation potential, most farmers depend on rainfed cultivation under dif- ficult climatic conditions. Endemic diseases such as malaria, yellow fever, and HIV/AIDS have weakened Africa's labor force, while debilitating livestock dis- eases such as trypanosomiasis have severely limited livestock rearing, animal traction, and mixed cropping in the tropical zones (Bloom and Sachs 1998; Sachs 2001). Africa's land surplus and its consequently low population density have limited incentives for agricultural intensification, raised transport costs, and made it more difficult for viable input, output, and credit markets to emerge (Hyami and Platteau 1997). Small countries pose problems of scale and market access. With a welter of political boundaries, many delineated in Berlin in 1885, independent Africa inherited a continent of nearly 50 independent states yet a population two-thirds the size of India's. For agricultural research and technology development, small countries imply high fixed costs. Africa, with the same cultivable area as the United States, operates eight times as many public agricultural research insti- tutes. The resulting small size of Africa's nearly 400 public research agencies limits prospects for achieving economies of scale and spillovers in agricultural research (Pardey et al. 2007). A constellation of small countries, coupled with disease-ridden coasts and temperate inland plateaus, has led to an unusually high concentration of inland population and landlocked countries in Africa (Collier and Gunning 1999). This spatial configuration results in long distances to world markets and high transport costs. Yet coastal countries have limited in- centives to invest in the road and transit infrastructure needed to serve their in- land neighbors (Bloom and Sachs 1998; Chigunta, Herbert, and Mkandawire 2003; Collier 2007). Weak institutions, poor governance, and bad policies also give pause. Skep- tics highlight Africa's weak rural development institutions and poor rural infra- structure, with road densities today even lower than in India during the 1950s (Spencer 1994). Weak states, regional conflicts, and poor governance compro- mise the efficiency of public interventions in agriculture as well as in other sec- tors of the economy (Chigunta, Herbert, and Mkandawire 2003; Collier 2007). The skeptics likewise point to volatile and generally falling commodity prices for Africa's major agricultural exports. Until the world commodity boom of2007 and 2008, they note that slumping international prices posed long-term disincentives for African farmers. They further argue that Africa's low agricul- tural productivity, coupled with high transport costs and growing world market liberalization, make it increasingly difficult for African farmers to compete in global markets. Instead of focusing on what they see as slow-growth, low- return agriculture, some development specialists see urban-based manufactur- ing and services as more likely to stimulate broad-based economic growth in 8 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin many African settings (Maxwell 2003; Ellis 2005; Collier 2009). In short, the skeptics see attempts to develop African agriculture as too expensive and too late. They consider the prospects for success to be bleak. Objectives These concerns, coupled with the substantial benefits anticipated should agri- cultural growth accelerate, motivate interest in exploring the conditions under which African agricultural performance might improve. To do so, the following chapters examine a series of instances in which African policymakers and farm- ers have succeeded in sustaining agricultural growth over long periods of time. This book explores the conditions under which Africa can successfully ac- celerate agricultural growth and thereby contribute to broad-based economic expansion and poverty reduction. A welter of past reviews has focused on Africa's failures and asked why African agriculture has performed poorly.7 This book asks the opposite question. Instead of cataloging failures and constraints, it identifies episodes of successful agricultural growth, a series of region- and commodity-specific booms, many of which have lasted for decades. By exam- ining a series of instances in which important advances have occurred in the past in African agriculture, this book aims to identify promising avenues for achieving success more consistently in the future. Past Performance Early Developments Agricultural production across the continent has changed considerably since the beginning of domesticated agriculture in Africa 7,000 years ago. Today, im- ported plant species—such as maize, cassava, groundnuts, bananas, cocoa, po- tatoes, sweet potatoes, tea, and imported varieties of cotton and rice—account for over two-thirds of the value of Africa's gross agricultural output (Gabre- Madhin and Haggblade 2003). Even more striking, the continent's 600 million head of livestock and 700 million head of poultry descend almost exclusively from imported species, with the lone exception of the guinea fowl (Diamond 1998). Despite a virtual absence of indigenous domesticable livestock species, and with a limited range of indigenous plants, African farmers have built up di- verse agricultural systems based largely on imported plant and animal species. This transformation has taken place in spite of the formidable ecological con- straints imposed by Africa's old and weathered soils, limited irrigation poten- tial, and debilitating endemic diseases such as malaria, tapeworm, and yellow fever. Livestock diseases, such as trypanosomiasis, have likewise severely lim- ited livestock rearing, animal traction, and mixed cropping in the tropical zones (Bloom and Sachs 1998). 7. Berg (1981), World Bank (1989, 2000), Bloom and Sachs (1998). Challenges for African Agriculture 9 The first half of the 20th century brought with it profound changes in smallholder agriculture all across Africa. Migrant smallholder farmers spread cocoa across much of West Africa (Hill 1963). Others introduced cassava to re- place cocoyams, receiving important assistance from rural artisans who devel- oped the necessary processing equipment along the way (Nweke 2004). Maize, cassava, and sweet potatoes gradually replaced sorghum and millets, leading to productivity gains across much of Africa (Jones 1957; Miracle 1966; McCann 2005). Tree crops and growing population pressure led farmers to abandon shifting cultivation and reduce fallow periods. Outside the endemic trypanoso- miasis zones of Central Africa, ox plowing took root among many small farm- ers and commercial settler farmers. As Carr (2001,331) has observed, "A strik- ing feature of these developments was the speed at which many of these major innovations were adopted by large numbers of smallholders." Sluggish Recent Performance Since the middle of the 20th century, however, aggregate performance in Africa has lagged behind that of other developing regions. Over the past 45 years, the value of aggregate agricultural output has increased by 2.4 percent annually in Sub-Saharan Africa compared to 2.8 percent in Latin America and 3.6 percent in developing Asia (FAOSTAT 2008). Both labor and land productivity have stagnated, remaining far below levels achieved in other developing regions (Figure 1.2). African farmers apply an average of 8 kilograms of fertilizer per hectare compared to between 80 and 100 kilograms in developing Asia and Latin America (Morris et al. 2007). Given stagnant productivity, Africa's mea- ger output gains have come mainly from area expansion. This extensification, coupled with shortened fallow periods and minimal input use, has led to nutri- ent mining and declining soil fertility (Cleaver and Schreiber 1994; Smaling, Nandwa, and Janssen 1997). In international markets, Africa has lost market share in all of its traditional export crops (World Bank 2007a; Hazell and Wood 2008). Since 1960, Africa's share of the value of world agricultural commod- ity exports has fallen from 5.0 percent to 1.3 percent (FAOSTAT 2008). Africa's aggregate performance has lagged at the same time it confronts the most daunting demographic challenge of any developing region. Since 1960, Africa has contended with population growth rates of 2.6 percent per year, 0.5-0.7 percent greater than in Latin America and developing Asia (FAOSTAT 2008). Consequently, comparisons of per capita production performance across continents over the past 45 years reveal deteriorating agricultural performance in Africa alone (see Figure 1.1). As per capita food production has fallen, Africa has turned from a food exporter to a net food importer. Equally worrisome are signs of decapitalization of Africa's key agricul- tural resources—its soils, human talent, and support institutions. Nearly half of Africa's farmland suffers from erosion and nutrient depletion (Cleaver and Schreiber 1994; AU/NEPAD 2003a). HIV/AIDS, with over 70 percent of known 10 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin F I G U R E 1.2 Trends in agricultural factor productivity, 1961-2003 Land productivity (value of agricultural output per hectare in constant 2005 US$) Sub-Saharan Africa (including South Africa) Sub-Saharan Africa (excluding South Africa) Latin America and Caribbean Developing Asia * # s? ^ ^ ^ ^ Labor productivity (value of output per agricultural worker, US$) S O U R C E : FAOSTAT (2008). N O T E : The arrows at the tip of each time series indicate the value in 2003. cases concentrated in Africa, has likewise extracted a heavy toll on Africa's hu- man capital. Eroding civil service salaries and anemic recurrent budgets have immobilized agricultural extension and research staff, diminished staff incen- tives, and fueled an exodus of senior scientists from public research institutions (Pardey, Roseboom, and Beintema 1997; InterAcademy Council 2004). These disconcerting trends place Africa's natural, human, and institutional capital un- der pressure and threaten Africa's capacity to sustain agricultural productivity growth in the future. Bright Spots This bleak aggregate picture contrasts with periodic bursts of more promising region- and commodity-specific performance. A review of 26 village studies in Africa documents a series of impressive achievements, a collection of "small and not so small booms in production of food crops for the national and sub- national markets" (Wiggins 2000, 635). Since 2000, half a dozen different groups have set out to identify and explore instances of superior performance Challenges for African Agriculture 11 in African agriculture.8 Several wonder aloud i f a green revolution is already taking place in various parts of Africa (Wiggins 2007; Zachary 2008). Indeed, African farmers and agricultural policymakers have achieved a se- ries of substantial successes in agricultural development, although in recent decades these have proven inadequate in number and scale to counter Sub- Saharan Africa's heavy demographic pressure. Despite their temporal and re- gional dispersion, many have endured for decades, as the following thumbnail sketches suggest. B A N A N A S I N A F R I C A ' S C E N T R A L H I G H L A N D S . For over 800 years, begin- ning about A.D. 500, farmers in the Great Lakes region experimented intensively with imported bananas. Through assiduous selection of cultivars, farmers bred a wide range of varieties suitable for human consumption, launching an extraordi- nary agricultural and demographic revolution in the Central African Highlands, a region that today supports one of the highest population densities in Africa (Schoenbrun 1993; de Langhe, Swennen and Vuysteke 1996; Reader 1997). M A I Z E I N E A S T A N D S O U T H E R N A F R I C A . The development and diffusion of modern, high-yielding varieties of maize have transformed this imported ce- real from a minor crop in the early 1900s into the continent's major source of calories today. Maize breeding in Zimbabwe and Kenya launched the first ma- jor breakthroughs during the 1960s, when Africa's breeders produced the first commercially grown single-cross hybrids in the world. These breeding break- throughs have improved the productivity of millions of small and large farms throughout Africa while moderating food prices for tens of millions of urban consumers (Smale and Jayne 2003). C A S S A V A . Cassava breeding and pest control efforts over the past three decades have triggered broad productivity gains for producers of Africa's number-two staple food. A stream of improved cassava varieties—the Tropi- cal Manioc Selection series released beginning in 1977—has invigorated breeding programs across Africa, increasing on-farm yield gains by over 40 percent without purchased inputs, and permitting rapid responses to recurring viral attacks. Given the improved cassava's high productivity, low purchased input requirements, and well-recognized drought tolerance, Nweke, Spencer, and Lynam (2002) have dubbed this cassava transformation "Africa's best- kept secret." C O T T O N I N W E S T A F R I C A . Since independence in the 1960s, West African cotton production and exports have both grown rapidly, at a compound annual rate of over 8 percent per year over the ensuing four and a half decades. As a result, francophone Africa's share in world exports has grown from near zero 8. In addition to Wiggins (2000), see the recent reviews by Pretty and Hine (2001), Reij and Waters-Bayer (2001), Gabre-Madhin and Haggblade (2004), Noble et al. (2004), FAO (2005), and Reij and Smaling (2007), as well as earlier work by Leonard (1991). 12 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin to 14 percent, making it the world's third-largest cotton-exporting block after the United States and the former Soviet Union (Tefft 2003). H O R T I C U L T U R E E X P O R T S F R O M E A S T A F R I C A . From the early 1970s on- ward, Kenya's private traders have steadily expanded high-value exports of fruits and vegetables from Kenya. Over the past 30 years, horticultural exports have increased fivefold in real terms to become the country's third-largest source of foreign exchange after tourism and tea (Minot andNgigi 2003). D A I R Y P R O D U C T I O N I N K E N Y A . In recent decades, Kenya's dairy produc- tion has grown rapidly, resulting in per capita production double the levels found elsewhere on the continent. Today 600,000 small farmers operating with one to three dairy cows each produce 80 percent of Kenya's milk. By the year 2000, nearly 70 percent of Kenyan smallholders produced milk and it had be- come their fastest-growing income source, with net earnings from milk aver- aging US$370 per year (Ahmed, Ehuj, and Asseva 2003; Ngigi 2005). As a re- sult, dairy marketing offers an important pathway out of poverty for farms of all size (Burke et al. 2007). R I C E P R O D U C T I O N I N M A L I . Policy reform in rice milling and marketing has radically altered opportunities and incentives for Mali's rice producers over the past two decades. Price deregulation, together with the dismantling of public monopolies on paddy assembly, milling, and rice marketing, led to the emergence of small private dehuller mills operating at one-fourth the milling cost of the large state mills. The subsequent 50 percent devaluation of the Communaute Finan- ciere Africaine (CFA) franc (CFAF), in January 1994, further boosted producer incentives. Producers responded rapidly to these new options and incentives. As a result, Malian rice production more than tripled after 1985, growing by 9 per- cent annually over the ensuing 20 years (Diarra et al. 2000; FAOSTAT 2008). The Disconnect Why have these impressive individual episodes not added up to rapid aggregate growth? In part, sluggish aggregate statistics may understate Africa's agricul- tural accomplishments, particularly in cassava, bananas, and other difficult-to- measure perennial crops. Agricultural sample frames, often geared to measur- ing major grain staples, prove unreliable for tracking dairy, livestock, cassava, beans, and other minor food crops. Likewise, large countries such as Nigeria weigh heavily in continental aggregations, despite the questionable reliability of their statistical projections (Jaeger 1992; Wiggins 2000). More fundamentally, Africa's high population growth rates impose the largest demographic challenge of any developing region. Given its high fertil- ity rates, Africa wi l l have to run faster than the rest of the developing world just to keep pace with its growing population. Because of Africa's belated demo- graphic transition, dependency ratios remain higher and growth of the prime age workforce more limited than elsewhere. Past successes, impressive but Challenges for African Agriculture 13 episodic, have proven insufficient to keep pace with Africa's daunting demo- graphic challenge. Although episodes of robust agricultural performance have occurred too infrequently in the past to counter Africa's demographic pressures, their occur- rence suggests a possible diagnostic for improving future performance. By ex- amining a series of cases in which important advances have occurred in the past, it may be possible to identify promising avenues for achieving success more consistently in the future. Learning from Past Successes Goals Sometimes, things go right. And when they do, we may benefit from asking why. This book summarizes efforts by the International Food Policy Research Institute (IFPRI) to systematically identify instances of superior agricultural performance in Africa, to study them, and to learn from them. In doing so, the book aims to identify processes, practices, and policies that have successfully stimulated agricultural growth in Africa. Our goal is to learn from what has gone right in the past. We begin with the premise that performance in some periods and locations has proven supe- rior to that in others and that by comparing them it wi l l prove possible to iden- tify the causes of this variation. Using a comparative case study approach, the individual case studies that follow, in Part I I of this book, attempt to contrast instances of superior performance with similar cases in which results have proven lackluster. The resulting method resembles a positive deviance study, in which instances of superior agricultural performance become the role models to be understood and emulated. Historically, this work most closely resembles the detailed case studies of African agriculture assembled by de Wilde (1967) and colleagues over 40 years ago.9 Research Methods To identify common ingredients and processes that underlie agricultural "suc- cess," the research team first identified episodes of strong performance in African agriculture along with contrasting instances, over time or space, in which performance had lagged. They then recruited specialists to study these contrasting cases using a common research protocol. In a final stage, the ana- lytical team brought the individual case study results together before groups of experienced African agricultural specialists to enlist their expertise in helping 9. Methodologically, the focus on detailed case studies of positive deviance resembles the approach adopted by Leonard (1991) in his study of successful rural development managers in Kenya. 14 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin to extract cross-cutting general lessons emerging from this body of collective case study experience (Table 1.2). To identify a broad range of successful episodes in African agriculture, the team first launched an expert survey targeting IFPRI's extensive contact list of African and Africanist agricultural specialists. A one-page questionnaire posed the following single question: "What do you consider the most successful in- stances of improved agricultural performance in Sub-Saharan Africa?" To en- courage the respondents to think broadly, the questionnaire deliberately left the criteria for success, as well as the time and geographic scope, unconfined. For each success story nominated, the survey form asked respondents to provide both their selection criteria and the factors they considered crucial in deter- mining the favorable outcome. In response, the team received over 250 nominations ranging broadly in geographic and historical scope. Some highlighted recent cutting-edge techni- cal breakthroughs, while others went back thousands of years. Table 1.3 pro- vides an overview of the nominations received, while Gabre-Madhin and Hagg- blade (2003, 2004) provide greater detail on the methods and findings. In order to select instructive case studies and to help focus the analytical work, the team required a clear definition of "success." For purposes of this work, the Successes in African Agriculture team defined "success" as "a significant, durable change in agriculture resulting in an increase in agriculturally derived ag- gregate income, together with reduced poverty and/or improved environmental quality" (Haggblade 2004, 4). This definition retained the three vertices of the "critical triangle"—income, equity, and sustainability. Given the importance as- cribed to aggregate income growth, the definition calls for a significant increase in income coupled with improvement in either sustainability or equity, or both. With this definition in hand, the research coordinators and an external advi- sory group reviewed the expert survey success nominations in order to select a dozen cases for in-depth review (Table 1.4). The resulting selection provides a series of important contrasts among private and public instigators of change, a variety of intervention points, differing levels of subsidy, a mix of food and ex- port crops, regional diversity, and impacts of variable duration and scale. Geo- graphically, the case studies covered 11 countries in West, Eastern, and Southern Africa (Figure 1.3). In total, the analytical team produced 11 case study reports as well as an additional half dozen background and synthesis papers. These provide the substance on which the remaining chapters of this book are based. To provide an appropriate counterfactual comparison, the team attempted to identify paired comparisons of successful and unsuccessful efforts for each of the case study selections. This proved relatively straightforward with the dairy, horticulture, cassava, and cotton case studies.10 The other case studies— 10. Due to a debilitating illness suffered by one of the co-authors of the cotton sector study, only the "successful" Malian case study was completed. 3 o §• PH o § 2 ° « cd O S3 Pi < O l l 2 w < -S o CA j3 s .a a cd a? O ' a B 0 0 O 03 S3 " O o c« c/: cd cd O O GJ O - O 43 '> 3 o ^ 3 o o I u o o Q o o 7 4 3 o o o 5 1 >• 3 "S u H s s - O Q —> 3 - J. J - m 0 , < ca ca 42 Ih B ca « •o X a 43 ^ 1 s o cd O TO X 2 "° I •= 2 •a » u § •§ -a S 1^ I a 4> W . g J W « C J > c < H < g % cd a Z £ «> ^ S § .S ca £ "2 42 3 P * feb § .a J= tfi M < ca a K o o 2 > Z S 45 HH cy5 . 2 " 'I S .O 43 3 " C M C " r- _r £ o '5b I I I ^ m ts 1 0 ,£3 s QJ ^ ca 3 j j M 4̂ _3 43 ca ^3 •a u 0 I -S 2 -S, Z B =J 1 1 1 3 'So 3 ^ ^ 2 PH CQ ca - 3 6 0 § W "o .8 2 •! » «J !o o a-" o « M H i H c o T A B L E 1.3 Expert survey success nominations African agricultural successes identified (%) Region- Category Total Africa-wide specific21 Commodity-specific Maize 10.3 11.1 10.0 Cassava 6.7 15.3 3.3 Horticulture 6.6 1.4 8.6 Livestock 6.2 9.7 4.8 Cotton 4.5 1.4 5.7 Coffee 4.3 5.6 3.8 Dairy 3.4 0.0 4.8 Rice 3.3 5.6 2.4 Cocoa 2.5 2.8 2.4 Bananas 2.5 1.4 2.9 Beans 1.8 1.4 1.9 Other 9.5 6.9 10.5 Subtotal 61.6 62.5 61.2 Activity-specific Soil fertility enhancementb 7.1 5.6 7.7 Policy reform Agricultural markets 2.0 0.0 2.7 Macro policy 1.6 0.0 2.2 Irrigation development 2.4 1.4 2.7 Specific technology development0 1.6 1.4 1.6 Other 6.7 6.9 6.6 Subtotal 21.2 15.3 23.5 Institution building Agricultural research 5.5 12.5 2.7 Farmer organizations 3.1 1.4 3.8 Market institutions 2.4 1.4 2.7 Human capacity building11 1.6 5.6 0.0 Other institutions 3.5 1.4 4.4 Subtotal 16.1 22.2 13.7 Countries Ethiopia, 1990s 0.4 0.5 Ghana, 1990s 0.4 0.5 Ivory Coast, 1960s and 1970s 0.4 0.5 Subtotal 1.2 1.6 Total Share 100.0 100.0 100.0 Number of nominations 253 71 182 S O U R C E : Gabre-Madhin and Haggblade (2004). aSpecific to East, Southern, Central, or West Africa or to a specific country, includes improved fallows, crop rotations, conservation farming. "Biotechnology applications, vaccines. dFinance, management, business. Challenges for African Agriculture 17 T A B L E 1.4 Case study selections Paper Countries Regional no. Topic covered applicability Authors 1 Cassava Nigeria, Ghana West Africa Nweke 2 Cassava Zambia, Malawi Southern Africa Haggblade and Zulu 3 Cotton Mali Francophone Tefft West Africa 4 Dairy Kenya, Uganda East Africa Ngigi 5 Dairy Ethiopia Horn of Africa Ahmed, Ehui, and Assefa 6 Horticulture Kenya, Cote Africa Minot and d'lvoire Ngigi 7 Maize Kenya, Malawi, East and Smale and Zambia, Southern Jayne Zimbabwe Africa 8 SSFM: planting Burkina Faso Semiarid Kaboure and basins Africa Reij 9 SSFM: planting Zambia Semiarid Haggblade and basins Africa Tembo 10 SSFM: improved Kenya Africa Place, Franzel, fallows Noordin, and Jama 11 SSFM: improved Zambia Africa Kwesiga, fallows Franzel, Ajayi, and Mafangoya N O T E : SSFM, sustainable soil fertility management. including those for maize, cotton, and sustainable soil fertility management technologies—have instead relied on temporal comparisons within the selected case study countries. In both sets of case studies, periods of sluggish perfor- mance provide a contrasting scenario against which to compare periods of rapid growth. For this reason, the turning points and the forces driving change at those junctures become central to helping us understand what instruments have proven effective in turning around agricultural performance in a variety of African settings. Analytical Framework Eleven case studies provide the empirical foundation for this analytical work. In order to produce comparable narratives, the analytical team developed a common framework for guiding each of the case studies. As a first step, the an- 18 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin F I G U R E 1.3 Case study countries alytical team—a group of subject-matter specialists recruited to conduct each of the case studies—convened in Washington, D.C., for one week to develop a common analytical framework and to plan the case study work. They sought an analytical framework that would help them track sources of change in agricultural trajectories. Given the long lead times required in tech- nology development and the ongoing continuous evolution in biological sys- tems, the framework needed to track interactions between human and biologi- cal systems over time. In constructing such a framework, the analytical team has drawn key features from a number of different conceptual paradigms rang- ing across a wide spectrum of disciplines. The umbrella framework and core dynamic processes draw heavily on the co-evolution paradigm developed in the anthropological and archaeological literature to study the emergence and long- Challenges for African Agriculture 19 term evolution of agriculture.11 This literature focuses broadly on the inter- dependence between biological systems and the human institutions governing agriculture. To help refine and operationalize that core dynamic framework, a related body of work in the biological and social sciences has likewise proven useful. 1 2 Figure 1.4 summarizes the resulting analytical framework, which the analytical team refers to as the Decisionmaking Environment (DE)-Action (A)-Results (R) framework (the DE-A-R framework). D E C I S I O N M A K I N G E N V I R O N M E N T . Because future growth in African agri- culture wi l l depend on improved performance by millions of individual farm- ers, the analytical framework places farmer decisionmaking at its core (see Fig- ure 1.4). It focuses on the decisionmaking environment that shapes incentives and opportunities available to farm households. Farmers' productive capacity, or opportunity sets, are determined at any given time by household assets as well as a set of public and collective goods, including available technology, roads, and community land management systems. Biophysical features of the agroecological zone—such as rainfall, soil fertility, and the incidence of pests and disease—also clearly affect productive capacity. The second major com- ponent of the decisionmaking environment, the existing incentive structure, de- pends on prevailing prices, local values and culture, and institutions governing the marketing and processing of farm products. Public actors such as govern- ments, donors, and nongovernmental organizations (NGOs) influence the farm- ers' decisionmaking environment through investments in public goods, such as infrastructure and research, through support institutions and through policies that shape the incentives to which farmers respond. A C T I O N S . Farmers act, motivated by the prevailing incentive structure and constrained by their productive capacity. They allocate land, labor, draft power, and financial capital across a portfolio of farm and nonfarm activities. Within each, they select a given technology and level of input use. They ex- periment, monitor, and compare outcomes. At the end of each production cy- cle, they decide to market some of their farm output, process a portion, and con- sume or store the remainder. The natural biological environment likewise responds to changes in agri- cultural systems. Farm technologies and input use influence soil fertility, soil 11. See Rindos (1980, 1984) and Price and Gebauer (1995). 12. On crop evolution, see, for example, Harlan (1992, 1995), Smith (1995), Evans (1996), and Diamond (1998). On nutrient monitoring, see Stoorvogel and Smaling (1990), Smaling, Nandwa, and Janssen (1997), de Jager, Nandwa, and Okoth (1998), and van den Bosch, de Jager, and Vlaming (1998). On induced innovations, see, for example, Boserup (1965), Hyami and Rut- tan (1971), and Binswanger and Ruttan (1978). On institutions and institutional innovation, see Binswanger and Mclntyre (1987), Nabli and Nugent (1989), North (1990), and Hayami and Ot- suka (1993). On development pathways, see Pender, Scherr, and Duron (2001) and Pender et al. (2004). For the related literature on economic livelihoods, see, for example, Ellis (2000), Pretty (2000, 2005), and Pretty and Hine (2001). OS 2 & < w Q © J3 H o Q 0> s £ 1 60 S 3 S s © u Challenges for African Agriculture 21 organic matter, and biotic activity. Plant and animal viruses evolve continu- ously as they adapt to changing agronomic practices, plant populations, input applications, and weather. Pests, diseases, and weed species change over time in response to human agricultural practices. R E S U L T S . These actions, together with natural variations in weather, gen- erate a set of results for individual households and, in aggregate, for the rural community. The results indicators adopted by the analytical team include em- pirical measures for each component of agricultural "success" (Table 1.5). The case study teams agreed that they would attempt to measure indicators along each dimension of success. While recognizing that this would not always be possible with available primary and secondary data, team members helped each other to identify possible sources of microdata necessary for making these calculations. F E E D B A C K A N D I N T E R A C T I O N . In a giant dynamic feedback loop, outcomes in each period influence both dimensions of the decisionmaking environment in the next season. Agricultural output and income influence farmers' capacity and inclination to finance productive investments. Asset accumulation in any one period increases farm households' productive capacity in the next. Marketed vol- umes likewise influence market price incentives as well as consumption and fu- ture production. Outcomes from one period affect soil fertility and other house- hold assets as well as the behavior of external actors such as agribusiness firms, farm organizations, government, donors, NGOs, and nonfarm consumers. The re- sulting responses by these actors, external to the farm household, alter the deci- sionmaking environment facing farm households in the following period. Pests, weed populations, and diseases likewise respond and adapt to changing circumstances. Over time, these adaptations lead to periodic violent T A B L E 1.5 Measuring results Success indicator Measure 1. Improved welfare a. Income Number of households adopting change (small, medium, large) Increased average income per household b. Assets c. Consumption d. Nutrition 2. Equity a. Smallholder participation in production growth b. Income change for smallholders c. Poverty reduction 3. Sustainability a. Ecological a. Net nutrient flows b. Financial b. Financial viability, with and without subsidies 22 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin upsurges in pest and disease infestation. The case studies in Part I I of this book routinely demonstrate the recurring difficulties posed by mutating diseases, viruses, pests, and weed populations. The case studies explore how these adap- tations in the biological system trigger corresponding responses on the part of farm managers and researchers on whom farmers rely for a steady infusion of new genetic and phytosanitary material. C A S E S T U D Y P R O T O C O L . The standard case study protocol emerging from the DE-A-R framework is described in Table 1.6. This standard protocol begins by defining the scope of the case study and explaining why agricultural spe- cialists generally considered it a success. The team then summarizes historical dynamics in order to define key turning points and phases in each given com- modity case study. Within each period, discussion and analyses focus on key shocks to the decisionmaking environment, the ensuing changes in farm house- hold actions, and key results in terms of income, equity, and sustainability. Each case study concludes with a discussion of policy implications. To provide some specificity and microeconomic grounding to these broad case study overviews, the analytical team agreed to prepare a series of short farmer case histories exploring how changing opportunities and incentives af- fected individual farm families in practice. Summarized in short boxes in the case study chapters, these profiles aim to help illustrate key dimensions of the changes under way. They likewise serve to focus attention on individual farm families, whose improved performance and welfare lie at the heart of agricul- tural development efforts. Key Policy Issues W H A T I N T E R V E N T I O N S ? Two key structural features of the agricultural decisionmaking environment govern human responses at any given point in time (see Figure 1.4). First, productive capacity places bounds on the scope of T A B L E 1.6 Case study protocol 1. Scope of the case study Commodity, time period, and location of the study Scale and scope of the "success" Why is this considered a success? 2. Historical turning points Change in aggregate production trends (output, yield, area) Identification of key turning points Definition of key phases 3. Dynamics and drivers of change during each phase Key shocks to the farmer decisionmaking environment Actions and responses by farm households and others Results: income, equity, sustainability 4. Policy implications Challenges for African Agriculture 23 action available to farmers. These opportunity sets depend on the available quantity, productivity, and distribution of key productive assets such as land, labor, capital, and water; on the stock of available biological and agronomic technology; on the state of physical infrastructure; and on supporting institu- tions for resource management, input supply, and marketing. Second, prevail- ing incentive structures govern which of the available options farmers wi l l se- lect from within their available opportunity sets. Incentives such as enhanced food security, social solidarity, or risk reduction influence individual and house- hold decisionmaking, while market prices affect input supply as well as the pro- duction, storage, processing, and marketing of outputs. Levers available for initiating change thus fall into two broad categories. Interveners can elect to expand productive capacity through development of new agricultural technology, provision of public and collective goods (roads, institu- tions governing production and input supply), or investment in private assets such as soil fertility, human capital, and physical and financial capital. Alterna- tively, they can focus on improving farmer incentives. Common tools include exchange rates, tariffs, subsidies, and taxes that influence market prices as well as investment in institutions governing output marketing and processing. W H O I N T E R V E N E S ? The case studies that follow identify key actors and agents of change. Who has taken the key initiatives? Given rapidly changing agricultural technology and the privatization of gains from biotechnology and genetic engineering, private and public roles may differ in the future. Trends toward agricultural market liberalization have opened up large agricultural mar- kets to private investment and trade. From a policy perspective, it becomes crit- ical to determine when and where public intervention is required and, con- versely, where the private sector can most usefully take the lead. Chapters 8 and 9 return to these key questions. Generalizing across the Case Studies To generalize from a dozen individual, historical case studies, the analytical team first compared the findings from all the case studies in order to identify genera] determinants of success. What, they asked, has driven past successes in African agriculture? What investments, technologies, policies, institutions, or- ganizational structures, and processes have proven key to enabling success in each of the case studies under review? How have public and private actions shaped these opportunities? Second, they explicitly asked how the future envi- ronment facing African agriculture might differ from that of the past. What principal changes are under way—in technology, communications, world mar- kets, and local institutions—that might influence future prospects for African farmers? Finally they asked, given these changes, how can farmers and policy- makers best apply the lessons of the past going forward? To answer each of these questions requires considerable judgment and col- lateral knowledge. Therefore the analytical team, in collaboration with the NEPAD Secretariat, convened a series of three groups of experienced agricul- 24 Steven Haggblade, Peter B. R. Hazell, and Eleni Gabre-Madhin rural specialists and political leaders from across Africa to provide input into this synthesis effort. Participants included a broad spectrum of farmers, gov- ernment officials, and private sector agribusiness operators. The first workshop, held in Pretoria in December 2003, involved roughly 70 senior African agri- cultural policy and technical specialists.13 The second, focusing exclusively on East and Southern Africa, took place in Nairobi in November 2004. 1 4 The third, drawing African parliamentarians from all across Africa, took place in Somer- set West, South Africa, in May 2006. 1 5 In all three settings, following brief plenary presentations by the case study teams, the participants spent the bulk of their time interacting in a series of professionally facilitated small-group working sessions. Through the facilitated small group sessions, participants worked together to complete a series of specific tasks: (1) summarize key lessons learned from past successes in African agriculture, (2) realistically assess the domestic and international policy environment within which African decisionmakers cur- rently operate, and (3) identify priorities for future policy action necessary to trigger sustained agricultural growth in Africa. Part I I I of this book, which sum- marizes key lessons for the future, has drawn heavily on these successive dia- logues with African agricultural stakeholders and policymakers. Organization of the Book Part I of this book has described the objectives, methods, and analytical frame- work used in reviewing agricultural successes in Africa. Part I I , the core of the book, reports in detail on the individual case studies selected for review. Chap- ter 2 begins by examining the cassava production surge that has occurred across much of Africa since the 1980s using case study material from Nigeria, Ghana, Zambia, and Malawi. Chapter 3 then reviews the causes and consequences of hybrid maize development and diffusion in eastern and southern Africa by ex- amining four contrasting country case studies from Zimbabwe, Kenya, Malawi, and Zambia. Given the long-term investments that underpin these productivity gains, this discussion covers roughly a 100-year period, from the early 1900s to 2005. Chapter 4 details the origins and impact of francophone West Africa's long-term cotton expansion over the four and a half decades following in- dependence, focusing on evidence and experience from Mali. Chapter 5, in turn, 13. The full conference program, proceedings, participant list, and all background papers are available at . 14. Titled "Agricultural Successes for the Greater Horn of Africa," these results are sum- marized by Haggblade (2005). A full set of conference materials is available at . 15. The outcomes of this conference, Championing Agricultural Successes for Africa's Future: A Parliamentarians' Dialogue on NEPAD, are summarized in NEPAD (2010) and at . Challenges for African Agriculture 25 explores two contrasting experiences with horticultural export development, one in Kenya and the other in Cote d'lvoire, focusing on the years since 1970. Chapter 6 reviews the East African dairy experience, contrasting the century- long expansion and modernization in Kenya with the less impressive perform- ance in neighboring Ethiopia and Uganda. To wind up the case study reviews, Chapter 7 describes the development and impact of two pairs of sustainable soil fertility management technologies. The first pair contrasts systems of minimum-tillage planting basins developed independently in Burkina Faso and in Zambia, while the second compares the development of improved fallow sys- tems in Western Kenya and in Eastern Zambia. Part I I I of the book generalizes and looks forward. Chapter 8 begins that effort by summarizing, comparing, and contrasting the six case study chapters. In so doing, it aims to distill the general policy lessons that emerge and to sep- arate out actionable items from idiosyncratic factors responsible for superior agricultural performance. Chapter 9 looks forward by examining several key questions. Why has Africa not experienced more frequent agricultural suc- cesses in the past? How wil l recent rapid changes in technology, global mar- kets, and agricultural institutions affect future strategies for stimulating agri- cultural growth? What needs to change to get African agriculture growing more rapidly? The book sums up by examining how lessons from the past can help governments, farm leaders, and agribusinesses shape future strategies for suc- cessful agricultural development in Africa. PART II Success Stories 2 The Cassava Transformation in West and Southern Africa F E L I X NWEKE AND STEVEN HAGGBLADE There was once a king of Bushongo who was called Samba Mikepe who was the wisest man who had ever lived. Before he mounted the throne, he made long voyages toward the west; no one knows to what distance to the other side of the Kasai River he went, and it was from this that he acquired his wisdom. . . . During one season of his reign the harvests of the Bushongo were completely destroyed by locusts and the people were in imminent danger of perishing from hunger. But they were saved by Samba Mikepe who showed them the use of manioc which could not be destroyed by any amount of locusts.1 Torday and Joyce (1911, 249) Scope of the Study Cassava production has grown rapidly in Africa in recent decades. In the early 1960s, Africa accounted for roughly 40 percent of world cassava production. Forty-five years later, by the mid-2000s, Africa's share had risen to 50 percent of world output. In the process, Nigeria has surpassed Brazil as the world's lead- ing cassava producer. As a result of these gains, cassava now provides Africa's second-most-important calorie source, after maize, and one-third of all Africans consume cassava as a food staple. Nigeria has led Africa's cassava surge. Production tripled there in less than a decade, beginning in the mid-1980s, as improved cassava varieties, success- ful pest control efforts, and strong producer incentives unleashed a pent-up surge in productive potential. Subsequently, advances in plant breeding, pest control, and cassava processing technology have spilled over from Nigeria to stimulate production surges across much of the rest of the African cassava belt (Figure 2.1). Cassava, a tropical root crop, possesses several key properties that make its recent expansion significant. Because of cassava's well-known drought tol- erance and because farmers can harvest cassava as they need it, any time over 1. Manioc is another name for cassava. Brazilians refer to the crop as mandioca, while Mex- icans and Central Americans know it as yuca. 29 30 Felix Nweke and Steven Haggblade FIGURE 2.1 Africa's cassava belt, 1980 S O U R C E : Okigbo (1980). a two to three year period, cassava provides an important food security buffer, allowing fanners to compensate for cereal production shortfalls in drought years. Similarly, because farmers can harvest cassava throughout the year, in many settings cassava remains the only food staple available for harvest dur- ing the lean season, when cereal prices and on-farm labor demands peak, caloric intake falls, and undernutrition among vulnerable households becomes most acute. Expansion of cassava production, therefore, can potentially improve food security significantly both during the lean season and during drought years. Moreover, because new varieties have proven so highly productive, cas- sava offers a low-cost source of carbohydrates not only for human consump- Cassava Transformation in West and Southern Africa 31 tion but also for use in animal feeds and industrial processing.2 So increasing productivity among cassava producers offers potential gains in farm income, food security, and industrial processing. This chapter examines the causes and consequences of Africa's unfolding cassava transformation. It begins by focusing on performance in Nigeria, the continental leader in these efforts and catalyst for many of the technological changes that have subsequently enabled productivity-led surges elsewhere in Africa. Ghana provides a contrasting example from West Africa, while Zambia and Malawi offer a Southern African perspective, highlighting potential differ- ences in the dynamics of cassava production and commercialization across re- gions and agroecological zones. Data for these reviews come from the Collab- orative Study of Cassava in Africa (COSCA), conducted between 1989 and 1997 in six different African countries, including Nigeria and Ghana, as well as from national farm-level surveys in Zambia and Malawi. 3 The Nigerian Cassava Transformation Over the past century, a series of changes in the farmer decisionmaking envi- ronment have propelled several waves of cassava-led growth in Nigeria. Inter- mittent drought, insect attacks, wars, and the sudden emergence of major plant and human diseases have triggered short-run surges in cassava production, both upward and downward (Figure 2.2). Over the long ran, new breeding and pro- cessing technologies, together with favorable government policy, have pro- pelled several long waves of productivity-led growth. Key actors have included Portuguese traders, several cohorts of emancipated Sierra Leonean slaves, lo- cal artisans who have developed low-cost processing technology, oil compa- nies that have supported cassava research and promotion, and researchers who have generated a stream of highly productive new cassava varieties, particu- larly over the past 40 years. In responding to evolving opportunities and threats, 2. As Jones (1959,255,281) puts it, "Manioc's outstanding characteristic, from an economic standpoint, is its capacity for producing large amounts of food calories per hectare.... To say that manioc is a low-cost calorie source is also to say that cultivation of manioc increases a commu- nity's productivity, either because more calories can be produced with the same land and labor, or because the same amount of calories can be produced by smaller inputs, thus freeing resources to be used in producing other things." 3. Data on cassava production are subject to a wide margin of error, for two principal rea- sons. First, because cassava farmers can harvest any time over a 2- to 3-year period and because farmers typically maintain a portfolio of cassava plots of differing maturity, total cropped area does not provide a reliable barometer of annual cassava production. Second, because farmers typically harvest their mature plots throughout the year, and often over multiple years, yield estimates based on recall data are notoriously suspect. Even where researchers conduct crops cuts, uncertainties about harvested area complicate output projections. In recognition of these difficulties, this chap- ter has preferred to use COSCA data where they are available, because these come from crop cuts and individual plot measurements. 32 Felix Nweke and Steven Haggblade F I G U R E 2.2 Cassava production trends in Nigeria and Ghana, 1961-2006 Nigeria (MMT) a. National cassava production Ghana (MMT) 12 10 8 6 4 2 0 Ghana / Nigeria / - / * / ̂ - / r_S— / / / / • 1 • / • I - I I I I . o£ «tf 45 40 35 30 25 20 15 10 5 0 b. Per capita cassava production kg per capita 600 S O U R C E : FAOSTAT (2008); for summary of raw data, see Appendix Table 2A.1. these actors have traced out five distinct phases in Nigeria's ongoing cassava transformation (Table 2.1). Phase 1: Cassava as a Food Staple, 1910—45 In the late 16th century, Portuguese traders introduced cassava to the west coast of Africa from South America. By 1700, cassava had become an important food Cassava Transformation in West and Southern Africa 33 crop in Sao Tome, in Fernando Po, and at Warri (Jones 1959). But cassava did not spread much farther until early in the 20th century, because the people of West Africa enjoyed a comfortable food security based on indigenous root crops, such as yams and cocoyams. Early in the 20th century, several factors spurred a rapid diffusion of cas- sava in different parts of Nigeria. A war of resistance against the British, from 1899 to 1914, followed by the First World War, from 1914 to 1918, and then the influenza epidemic of 1918 all contributed to sharply constricted labor availability in the lower Niger Delta. The resulting labor shortage precipitated a major shift among farm families from cultivation of labor-intensive yams to less labor-intensive cassava. During wartime, cassava's flexible harvesting calendar and long-term in-ground storage also offered considerable attraction for intermittently displaced villagers. By the late 1920s, cassava production had spread to most parts of the lower Niger (Ohadike 1981; Chiwona-Karltun 2001). The dissemination of cassava processing technology further accelerated the adoption and marketing of cassava-based products in the Niger Delta. In 1916, pastors of the Niger Delta Pastorate, mostly emancipated slaves from Sierra Leone and Western Nigeria, brought gari processing technology, along with Christianity, to Ngwa land in Eastern Nigeria.4 Transformation of cassava into gari, by pressing out the water after grating and fermenting the roots, re- duces transport and handling costs by over 60 percent. Further processing, by drying and toasting the gari, extends the shelf life of cassava from 2 days for fresh roots to several months for gari. By lowering transport costs and im- proving shelf life, this new processing technology proved vital to the develop- ment of cassava as a staple food as well as a cash crop. Between 1930 and 1939 in Oyo and Ondo provinces, north of Lagos, an invasion of locusts inflicted considerable damage on the yam crop. During this period, many farmers turned to more insect-tolerant cassava. Cassava's drought tolerance likewise contributed to its steady inroads in the yam belt. In 1945 and 1946 in Ondo province, farmers planted cassava in areas where yams had failed to sprout because of the long dry season (Agboola 1968). Phase 2: Cassava as a Cash Crop, 1946-77 The period around World War I I marked a turning point in Nigerian cassava development, when the confluence of several major forces combined to stimu- late the rapid expansion of cassava as a cash crop. Growing urban markets in- creased the demand for gari, while the introduction of new cassava varieties 4. Gari is a granulated, precooked cassava-based convenience food. It is especially popular in West Africa, where customers appreciate its convenience as a ready-to-eat food. Processing in- volves grating, fermenting, drying, and toasting. After purchasing, consumers merely need to add hot or cold water to the gari granules to produce a thick porridge. cd •c -a « H B % 1) CD ^ Is ^ s a o P H £ > cd cd co J 3 > cd P H cd o O ft g o o u oi a •= a O Ctn O 3 r l l a l l is 1 s 1 s I Q (A GO CO £ o g co cd 3 o "3 $ 22 "2 ° O ft o . s ft — oo w g j> oo o R -o ft § -̂3 - i -S J ^ ^ -s oo cd t l >. "2 i f 3 „ w > t3 cd a o T 3 « cd cd cd o 3 w 2 OP ta rt O co CD CO o o o rH 1) > l P H S a .S o " cd C- o .•5 3 oo 2 d - r 5 c d S S < D i > > c d -H O > CO CO u oo § J 3 2 B § 3 3 cd . 3 g N 4 0 CD co 2 3 a.,2 g o S3, O o .2 s -s c£3 £ B cS CN S 1 CD is . •§ S Q g ca 0 0 p 2 > o g B 3 ?H CO 5 p 0 0 5 2 n 43 CO t " > P ^ hi i CD .2 P >> £ 43 43 » 1 .S P 60 3 43 § 2 - P _CD 60 CD C M - S3 6 o a CO CD cS CD CO CD CO "E. CCS g C S "E. CCS 0 ~0 CO 0 CD CD 43 p av a b CO c2 av a b 1 ra l CO CO CO cS CD O 1 o ia