AGRICULTURAL PRODUCTIVITY IN AFRICA Trends, Patterns, and Determinants EDITED BY SAMUEL BENIN About IFPRI The International Food Policy Research Institute (IFPRI), established in 1975, provides research-based policy solutions to sustainably reduce poverty and end hunger and malnutrition. The Institute conducts research, communi- cates results, optimizes partnerships, and builds capacity to ensure sustainable food production, promote healthy food systems, improve markets and trade, transform agriculture, build resilience, and strengthen institutions and gover- nance. Gender is considered in all of the Institute’s work. IFPRI collaborates with partners around the world, including development implementers, public institutions, the private sector, and farmers’ organizations. About IFPRI’s Peer Review Process IFPRI books are policy-relevant publications based on original and innova- tive research conducted at IFPRI. All manuscripts submitted for publica- tion as IFPRI books undergo an extensive review procedure that is managed by IFPRI’s Publications Review Committee (PRC). Upon submission to the PRC, the manuscript is reviewed by a PRC member. Once the manuscript is considered ready for external review, the PRC submits it to at least two external reviewers who are chosen for their familiarity with the subject mat- ter and the country setting. Upon receipt of these blind external peer reviews, the PRC provides the author with an editorial decision and, when necessary, instructions for revision based on the external reviews. The PRC reassesses the revised manuscript and makes a recommendation regarding publication to the director general of IFPRI. With the director general’s approval, the manuscript enters the editorial and production phase to become an IFPRI book. Agricultural Productivity in Africa Trends, Patterns, and Determinants Edited by Samuel Benin A Peer-reviewed Publication International Food Policy Research Institute Washington, DC Copyright 2016 International Food Policy Research Institute. All rights reserved. Contact ifpri-copyright@cgiar.org for permission to reproduce. Any opinions stated herein are those of the author(s) and are not necessarily representative of or endorsed by the International Food Policy Research Institute. International Food Policy Research Institute 2033 K Street, NW Washington, DC 20006-1002, USA Telephone: 202-862-5600 DOI: http://dx.doi.org/10.2499/9780896298811 Library of Congress Cataloging in Publication Program 101 Independence Avenue, S.E. Washington, DC 20540-4283 Library of Congress Cataloging-in-Publication Data Names: Benin, S. (Samuel), editor. Title: Agricultural productivity in Africa : trends, patterns, and determinants / edited by Samuel Benin. Description: Washington, DC : International Food Policy Research Institute, [2016] | Includes bibliographical references and index. Identifiers: LCCN 2015041365 | ISBN 9780896298811 (pbk. : alk. paper) Subjects: LCSH: Agricultural productivity--Africa. Classification: LCC S472.A1 A346 2016 | DDC 338.1/6096--dc23 LC record available at http://lccn.loc.gov/2015041365 Cover Design: Anne C. Kerns, Anne Likes Red Project Manager: Patricia Fowlkes, IFPRI Book Layout: David Peattie, BookMatters Contents Tables, Figures, and Boxes vii Abbreviations and Acronyms xv Foreword xxi Acknowledgments xxiii Chapter 1 Introduction 1 Samuel Benin, Stanley Wood, and Alejandro Nin-Pratt Chapter 2 Intertemporal Trends in Agricultural Productivity 25 Samuel Benin and Alejandro Nin-Pratt Chapter 3 Spatial Patterns of Agricultural Productivity 105 Stanley Wood, Zhe Guo, and Ulrike Wood-Sichra Chapter 4 Typology of Agricultural Productivity Zones 133 Bingxin Yu and Zhe Guo Chapter 5 Agricultural Intensification and Fertilizer Use 199 Alejandro Nin-Pratt Chapter 6 Factors Influencing the Effectiveness of Productivity-Enhancing Interventions: An Assessment of Selected Programs 247 Joseph Karugia, Stella Massawe, Paul Guthiga, Maurice Ogada, Manson Nwafor, Pius Chilonda, and Emmanuel Musaba Chapter 7 Conclusions and Implications for Raising and Sustaining High Agricultural Productivity in Africa 335 Samuel Benin Authors 349 Index 355 Tables, Figures, and Boxes Tables 1.1 Annual average agricultural growth, productivity, and public spending in Africa and other selected developing regions of the world, 1970– 2010 5 1.2 Stated budget allocation to the top three programs in selected African countries (percentage of total NAIP budget) 14 2.1 Description of variables and data used in estimating partial and total factor productivity 29 2.2 Countries by geographic region and country’s share in region’s total agriculture value-added (%) 31 2.3a Land and labor productivity (annual average level, 1961– 2012) 34 2.3b Land and labor productivity (%, annual average growth rate, 1961– 2012) 38 2.4 Total factor productivity growth, efficiency change, and technical change (%, annual average, 1961– 2012) 50 2.5 Input and capital per worker and technical change, annual average (1995– 2012) 60 2.6 Correlation coefficients between land, labor, and total factor productivity (TFP) growth by technical change and input intensity (1961– 2012) 62 2B.1 Annual average TFP growth rates for Africa using different TFP index methods, 1971– 2012 80 2B.2 Annual average TFP growth rates for African countries using different TFP index methods, 1971– 2012 83 2C.1 Countries by economic development classification and country’s share in group’s total agriculture value-added 85 2.C2 Countries by Regional Economic Community (REC) and country’s share in REC’s total agriculture value-added 86 2.C3 Countries by size and growth of agriculture sector 88 3.1 Spatial data used in exploring the spatial patterns of partial productivity in crop production in Africa south of the Sahara 110 3.2 Land productivity: Average value of annual crop production ($) per hectare cropland by subregion and farming system 2005– 2007 118 3.3 Labor productivity: Average value of annual crop production ($) per agricultural worker, 2005– 2007 122 3A.1 Distribution of value of crop production by farming system ($ millions), 2005– 2007 127 3A.2 Distribution of cropland area by farming system (1,000 hectares), 2005 128 3A.3 Distribution of rural population headcount by farming system (number), 2005 129 4.1 Comparison of simplified and FAO-defined farming systems 139 4.2 Share in Africa south of the Sahara and average by farming systems 148 4.3 Summary statistics of the cluster analysis for the tree-root crop farming system 151 4.4 Number of APZs and typologies of APZs by farming system 152 4.5 Description of the typologies of APZs in the tree-root crop farming system 153 4.6 Description of the typologies of APZs in the forest-based farming system 155 4.7 Description of the typologies of APZs in the highlands farming system 157 4.8 Description of the typologies of APZs in the cereal-root crop farming system 158 viii Tables, Figures, and boxes 4.9 Description of the typologies of APZs in the maize mixed farming system 160 4.10 Description of the typologies of APZs in the pastoral- agropastoral farming system 161 4.11 Description of the typologies of APZs in the irrigated farming system 162 4.12 Description of the typologies of APZs in the coastal farming system 164 4.13 Description of the typologies of APZs in the large commercial and smallholder farming system 165 4.14 Typology of APZs in Ethiopia 167 4.15 Typology of APZs in Ghana 168 4A.1 Average annual NDVI by farming system 171 4A.2 Number and size of agricultural productivity zones (APZs) by country 172 4A.3 Cropland area by farming system, in 1,000 hectares in 2005 173 4A.4 Travel time by farming system, in hours to cities with population greater than 50,000 inhabitants in 2005 174 4A.5 Rural population density by farming system, in people per km2 in 2005 175 4A.6 Typology of major subsystems in Africa south of the Sahara 176 4A.7 Typology of major subsystems (within systems) by country in Africa south of the Sahara 180 4A.8 Typology of minor subsystems by country in Africa south of the Sahara 184 4A.9 Typology of marginal subsystems by country in Africa south of the Sahara 187 5.1 Population density and output per hectare of agricultural area (average values for 1995– 2000) 216 5.2 Population density and output per hectare of different measures of agricultural area by quantile of population density and correlation values, 1995–2000 217 5.3 Population density and inputs per hectare of different measures of agricultural area (average values for 1995– 2000) 218 Tables, Figures, and boxes ix 5.4 Population density and inputs per hectare of different measures of agricultural area by quantile of population density (average values for 1995– 2000) 220 5.5 Decomposition of total output per hectare of potential agricultural land (2008– 2011) and growth rates of its different components by per quantile of population density, 1995–2011 222 5.6 Decomposition of total output per hectare of potential agricultural land (2008– 2011), and contribution of its different components to growth during 1995– 2011 227 5.7 Correlation coefficients of different components of intensification and fertilizer use 231 5.8 Population densities, output per hectare of harvested land, and fertilizer per hectare of arable land, 1995– 2011 232 5.9 Variables expected to affect fertilizer use, showing countries with high population density and low fertilizer use (all countries in G4) 234 5.10 Variables expected to affect fertilizer use, showing countries with intermediate levels of population density and high fertilizer use (all countries in G3) 236 5.11 Variables expected to affect fertilizer use showing countries with low population density and high fertilizer use (all countries in G1 and G2) 237 5.12 Comparison of average values of fertilizer use per hectare between the maize mixed farming system and other farming systems (2005– 2011) 238 5.13 Total land suitable for crop production under maize mixed and highland temperate mixed systems, compared with other systems by country 239 6.1 Conceptual factors and empirical indicators used in performance assessment 261 6.2 Likert scales and associated scores 261 6.3 Interventions selected for assessment by countries and farming systems 262 6.4 Performance of the interventions in meeting criteria for effectiveness in implementation 264 6.5 Overall performance in implementing the interventions 265 x Tables, Figures, and boxes 6.6 Distribution of projects in meeting the overall productivity target 266 6.7 Performance in indicators of implementation by performance in overall productivity 267 6A.1 Productivity-enhancing interventions, their locations and objectives, and sources of information 273 6A.2 Instrument used to collect information from agricultural and rural development practitioners 277 6A.3 Agricultural productivity impact pathways: How the 13 factors identified in the conceptual framework affect productivity 278 6A.4 Description of methodology used in rating performance against the criteria of successful project implementation 280 6A.5 Summary of review of performance in implementation of selected agricultural productivity-enhancing interventions in Africa south of the Sahara 282 6A.6 Performance in meeting the overall productivity objective or target 316 Figures 1.1 Public expenditure on agricultural research and development in selected African countries, 1996– 2008 (annual average % of agricultural value-added) 12 1.2 Figure 1.2 Stated budget allocation to selected agricultural functions in selected African countries (percentage of total NAIP budget) 15 2.1 Line plots of land and labor productivity by geographic region (1961– 2012) 42 2.2 Land and labor productivity for selected countries (average 2000– 2012) 46 2.3 Growth rate in land and labor productivity for selected countries (annual average 2000– 2012) 47 2.4 Levels of total factor productivity, efficiency, and technology by geographic region (1961– 2012: indexed at 1961=1) 54 2.5a Total factor productivity growth decomposition by group (%, annual average 1961– 1985) 56 Tables, Figures, and boxes xi 2.5b Total factor productivity growth decomposition by group (%, annual average, 1985– 2012) 57 2.6 Total factor productivity growth decomposition at country level (%, annual average 1985– 2012) 58 2.7 Total factor productivity growth decomposition at country level (%, annual average 2000– 2012) 59 2.8 Land, labor, and total factor productivity growth in Africa (%, annual average 1961– 2012) 61 2A.1 Input possibility set, periods t and t+1 68 2B.1 Percentage of zero shadow prices for different inputs, annual average (1971– 2012) 79 2B.2 Average TFP indexes for Africa using different index methods, 1971– 2012 80 2B.3 Scatter plots of TFP growth rates from different index methods for Africa south of the Sahara (annual averages, 1995– 2012) 81 2B.4 Scatter plots of TFP growth rates from different DEA- Malmquist index methods (annual averages, 1995– 2012) 82 2.C1 Line plots of land and labor productivity by economic classification (1961– 2012) 89 2.C2 Line plots of land and labor productivity by Regional Economic Community (1961– 2012) 90 2C.3a Line plots of land and labor productivity by size or rate of growth of agriculture sector (1961– 2012) 91 2C.3b Line plots of land and labor productivity for selected countries by size or rate of growth of agriculture sector (1961– 2012) 92 2C.3c Line plots of land and labor productivity for selected countries by size or rate of growth of agriculture sector (1961– 2012) 93 2C.4 Levels of total factor productivity, efficiency, and technology by economic classification (1961– 2012: indexed at 1961=1) 94 2C.5 Levels of total factor productivity, efficiency, and technology by Regional Economic Community (1961– 2012: indexed at 1961=1) 96 2C.6 Levels of total factor productivity, efficiency, and technology for selected countries (1961– 2012: indexed at 1961=1) 99 xii Tables, Figures, and boxes 3.1 Major farming systems of Africa 113 3.2 Land and labor productivity of crop production in Africa south of the Sahara (circa 2006) 119 4.1 Distribution of agricultural productivity zones 141 4.2 Spatial patterns of key factors influencing agricultural production and productivity at the system level 144 4.3 Plots of the cluster analysis for the tree-root crop farming system 152 5.1 Contribution of new arable land, cropping intensity, and output per hectare of harvested area to growth of crop output per hectare of potential cropland by quantile of population density, 1995– 2011 223 5.2 Shadow price of labor relative to land at different levels of population density, 1995–2011 224 5.3 Patterns of the contribution of different components to growth of total crop output per hectare of potential crop land by quantile of population density. 1995– 2011 225 5.4 Contribution of new arable land, cropping intensity, and output per hectare of harvested area to growth of crop output per hectare of potential cropland by country and quantile of population density, 1995– 2011 229 5.5 Number of SSA countries with more than 50 percent of their national agriculture in a particular farming system, 2005– 2011 240 6.1 Factors influencing the success or failure of agricultural productivity-enhancing interventions 250 7.1 Land, labor, and total factor productivity (TFP) growth, and TFP growth decomposition in Africa (%, annual average 1961– 2012) 337 7.2 Distribution of agricultural productivity zones in Africa 340 Boxes 3.1 Land productivity: An appropriate denominator? 116 6.1 Operation Mwolyo Out intervention and selected performance indicators 269 6.2 Kenya Animal Health Services Rehabilitation Programme 270 Tables, Figures, and boxes xiii ABBREVIATIONS AND ACRONYMS AEZ agroecological zone AFDB African Development Bank AFSI L’Aquila Food Security Initiative AHSRP Animal Health Services Rehabilitation Programme AI artificial insemination APEP Agriculture Productivity Enhancement Programme APZs Agricultural productivity zones ASARECA Association for Strengthening Agricultural Research in Eastern and Central Africa ASDP Agricultural Sector Development Programme ASDS Agricultural Sector Development Strategy AU African Union AU-NEPAD African Union– New Partnership for Africa’s Development BMGF Bill & Melinda Gates Foundation BXW banana Xanthomonas wilt C3P Crop Crisis Control Project CA conservation agriculture CAADP Comprehensive Africa Agriculture Development Programme CAP1 Conservation Agriculture Project 1 CBOs community-based organizations CDC Commonwealth Development Corporation CEDP Cassava Enterprise Development Project CEN-SAD Community of Sahel– Saharan States CFU Conservation Farming Unit CMD cassava mosaic disease COMESA Common Market for Eastern and Southern Africa CPA stock of land suitable for crop production CRS constant returns to scale CSPR Civil Society for Poverty Reduction DANIDA Danish International Development Agency DEA data envelopment analysis DVS Department of Veterinary Services EAC East African Community EADD East Africa Dairy Development Project ECA eastern and central Africa ECCAS Economic Community of Central African States ECOWAS Economic Community of West African States EIA environmental impact assessment EPRC Economic Policy Research Centre ERPs economic recovery programs FADGIP FARM Africa Dairy Goat Improvement Project FAO Food and Agriculture Organization of the United Nations FFs farmer field schools FISBP Farm Input Subsidy Program FISPP Farmer Input Support Program FPIS Fuve Panganai Irrigation Scheme FTSP Fodder Trees and Shrubs Project G1– G4 Groups 1 through 4 G8 Group of Eight G20 Group of Twenty GAFSP Global Agriculture and Food Security Program GDP gross domestic product xvi abbreViaTions and aCronYMs GIS geographic information system GRUMP Global Rural Urban Mapping Project ha hectare I$ international dollars ICIPE International Centre of Insect Physiology and Ecology ICRAF World Agroforestry Centre IDA International Development Association IFAD International Fund for Agricultural Development IFPRI International Food Policy Research Institute IGAD Intergovernmental Authority on Development IITA International Institute of Tropical Agriculture ILO International Labor Organization ILRI International Livestock Research Institute IMF International Monetary Fund IPCC Intergovernmental Panel on Climate Change ISO International Organization for Standardization ISPs input subsidy programs IWMI International Water Management Institute JICA Japan International Cooperation Agency KARI Kenya Agricultural Research Institute KASCOL Kaleya Smallholders Company Limited KASFA Kaleya Smallholder Farmers’ Association KDDP Kenya Dairy Development Programme kg kilogram kg/ha kilogram per hectare KIP Kaleya Irrigation Project KIPPRA Kenya Institute for Public Policy Research and Analysis km kilometer km2 square kilometer km/hr kilometers per hour LP linear programming abbreViaTions and aCronYMs xvii m meter m2 square meters M&E monitoring and evaluation MAFC Ministry of Agriculture, Food Security and Cooperatives MAUP modifiable areal unit problem MCE multicriteria evaluation MDTF Multi-Donor Trust Fund Mha million hectares MI middle income MM metafrontier Malmquist index MSE mean square error NAADS National Agricultural Advisory Services NAEIP National Agricultural Extension Intervention Program NAFSIPs National Agricultural and Food Security Investment Plans NAIPs national agricultural investment plans NAIVS National Input Voucher System NARS National Agricultural Research System NDVI normalized difference vegetation index NEPAD New Partnership for Africa’s Development NERICA New Rice for Africa upland rice NFSD Novartis Foundation for Sustainable Development NGO nongovernment organization O&M operation and maintenance OECD Organisation for Economic Co-operation and Development OMO Operation Mwolyo Out OPEC Organization of Petroleum Exporting Countries PADETES Participatory Demonstration and Training System PAPSTA Support Project for the Strategic Plan for the Transformation of Agriculture PCU Projects Coordinating Unit PFP partial factor productivity xviii abbreViaTions and aCronYMs PIDP Participatory Irrigation Development Programme PMSU Project Management Support Unit PPS production possibility set PPT Push– Pull Technology PRSPs Poverty Reduction Strategy Papers R&D research and development RECs Regional Economic Communities RELMA Regional Land Management Unit RESAKSS Regional Strategic Analysis and Knowledge Support System SADC Southern African Development Community SAP structural adjustment program SAPRIN Structural Adjustment Participatory Review International Network SCP Specialty Coffee Program SG 2000-AP Sasakawa Global 2000 Agricultural Program SOAS School of Oriental and African Studies SOFA The State of Food and Agriculture SPAM Spatial Production Allocation Model SRI System of Rice Intensification SSA Africa south of the Sahara TFP total factor productivity TGR technology gap ratio TI tropicality index TPA total potential agricultural area UMA Union du Maghreb Arabe UNDP United Nations Development Programme UNFFE Uganda National Farmers Federation URT United Republic of Tanzania USAID United States Agency for International Development VRS variable returns to scale WARDA West Africa Rice Development Association abbreViaTions and aCronYMs xix WHO World Health Organization WSS within sum of squares WUA water use association WWIDP Wei Wei Integrated Development Project ZSC Zambia Sugar Company xx abbreViaTions and aCronYMs FOREWORD A gricultural Productivity in Africa: Trends, Patterns, and Determinants presents updated and new analyses of land, labor, and total productivity trends in African agriculture. It brings together analyses of a unique mix of data sources and evaluations of public policies and development projects to recommend ways to increase agricultural productivity in Africa. This book is timely in light of the recent and ongoing growth recovery across the continent. The good news is that agricultural productivity in Africa increased at a moderate rate between 1961 and 2012, although there are variations in the rate of growth in land, labor, and total factor productivities depending on country and region. Differences in input use and capital intensities in agricul- tural production in the various farming systems and agricultural productivity zones also affect advancements in technology. One conclusion based on the book’s research findings derives from the substantial spatial variation in agri- cultural productivity. For areas with similar agricultural productivity growth trends and factors, what works well in one area can be used as the basis for for- mulating best-fit, location-specific agricultural policies, investments, and inter- ventions in similar areas. This finding along with others will be of particular interest to policy- and decisionmakers. By asking and answering pointed questions, Agricultural Productivity in Africa offers succinct recommendations for specific situations as well as broad development objectives. How can Africa further raise labor productivity to reduce mass poverty? Can increasing land productivity (yields) make a dif- ference in averting future food crisis? How does Africa effectively take full advantage of regional and subregional alliances that promote and disseminate appropriate technologies capable of reversing the declining growth in land productivity, sustain or strengthen the recent rapid growth in labor productiv- ity, and expand the technological frontier on an ongoing basis? The authors have successfully framed their research questions, mapping out the body of evidence they present which has resulted in an informative book that will assist researchers in understanding the various ways agricul- tural productivity can increase and help policymakers and those in decision- making positions determine what options are best for their country, subregion, and region. Shenggen Fan Director General ACKNOWLEDGMENTS Funding was provided by the United States Agency for International Development and the Bill & Melinda Gates Foundation, through their support to the Regional Strategic Analysis and Knowledge Support System program in Africa. Several people have contributed to the production of this book, including Greg Traxler and Ousmane Badiane, whose assistance during conceptualization of the agricultural productivity study was invalu- able. Melanie Bacou, Angga Pradesha, Linden McBride, and Heather Wyllie provided data and analytical support. Additional assistance was provided by the International Food Policy and Research Institute’s (IFPRI’s) Publications Unit of the Communications and Knowledge Management Division, espe- cially Patricia Fowlkes and Andrea Pedolsky. We would also like to thank par- ticipants of IFPRI’s conference on “Increasing Agricultural Productivity & Enhancing Food Security in Africa: New Challenges and Opportunities,” held on November 1– 3, 2011, in Addis Ababa, Ethiopia, for their feedback on a presentation of this study. We are especially indebted to the anonymous reviewers for providing insightful comments and directions for addi- tional work. Fostering higher agricultural productivity and accelerating agricultural growth in Africa are commonly seen as core strategies for overall devel- opment in the continent (Lewis 1954; Fei and Ranis 1961; Hayami and Ruttan 1985; Hazell and Haggblade 1991; Binswanger and Townsend 2000; World Bank 2007).1 Because the majority of Africa’s poor and malnourished population depends largely on farming, these strategies can be particularly effective in reducing poverty and hunger. Yet, agricultural growth in Africa lags behind overall economic growth, and the continent’s agricultural perfor- mance has fallen further behind that of other developing regions of the world. The development literature offers many hypotheses to help explain the chronic underperformance of Africa’s agriculture sector. One particularly fundamental perception by those making critical policy and investment deci- sions is the ambiguity of agriculture’s role in development. Additionally, the quality and relevance of data and analysis provided to those individuals to allow them to measure potential costs and benefits, consider trade-offs, and make informed decisions are questioned. The recent high global food prices of 2007– 2008 and later periods, which gave rise to food crises in many African countries and drew varied and often productivity-reducing responses from several governments across the continent (Benson, Mugarura, and Wanda 2008; Wodon and Zaman 2010; Headey et al. 2012; Benson et al. 2013), have renewed concern about knowledge gaps surrounding appropriate strategies for raising and maintaining higher levels of agricultural productivity. 1 Diao et al. (2007) provides a solid review of the literature on the role of agriculture in devel- opment, spanning the classical thinking of a passive role where agriculture serves as a reserve of labor and capital, to one where agriculture plays an active role through production and con- sumption linkages, including its role in rural, as opposed to national, development because of spatially differentiated constraints in production and market linkages. They also review more recent discourse about the agriculture– nutrition nexus, agriculture’s role in stabilizing food prices and ensuring food security, and the unique decisionmaking processes associated with managing the sector. INTRODUCTION Samuel Benin, Stanley Wood, and Alejandro Nin-Pratt Chapter 1 1 This book raises explicit questions for policy analysts in African countries and development agencies who advise policymakers on strategies to acceler- ate productivity growth and presents new and updated analyses of agricultural productivity trends for African countries and subregions. These analyses offer greater economic and spatially disaggregated insights than is typical for stud- ies encompassing all of Africa, and suggest some critical conclusions for the viability of a rapid acceleration of agricultural productivity and value addition in Africa. To fully contextualize the book, the remainder of this introductory chapter examines the competing hypotheses for Africa’s poor agricultural performance from a historical perspective, beginning in the colonial era, through the structural adjustment periods, to the current crop of agricul- tural development strategies guiding the continent under the auspices of the Comprehensive Africa Agriculture Development Programme (CAADP). Following an assessment of the challenges faced in implementing these strate- gies, the chapter concludes with a summary of the organization of the remain- der of the book. History of African Agriculture and Hypotheses Regarding Its Poor Performance A starting point for the contextualization of this book begins with a histori- cal overview that, by necessity, offers stylized facts about Africa’s development. These facts also bring into sharp focus the sweeping generalizations made about Africa that effectively led to some simplistic approaches to agricultural development that lacked an understanding of the continent’s diversity and variation, presaging the critical necessity of higher-resolution data and analysis that are the later focus of this book. During the colonial period in Africa, agriculture was the most important economic activity. Farmers were required or incentivized by many colonial administrations to grow cash crops for export, primarily to provide raw mate- rials for industrial production in the metropolitan countries (Anthony et al. 1979). The dominant cash crops for export included cocoa, coffee, tea, palm oil, and rubber in the rainforest areas of central and West Africa; ground- nuts and cotton in the Sahel belt of West Africa; sisal, tea, and coffee in East Africa; and sisal, sugarcane, and tobacco in southern Africa. In general, food crops were not promoted, and farmers grew them for subsistence only. Colonial administrations invested heavily in transportation systems to facili- tate the movement of cash crops from the interior to the coastal ports, as well 2 Chapter 1 as the flow of manufactured goods imported from the metropolitan coun- tries into the interior. To bolster their aims, administrations also invested in farm support, research, extension, and marketing infrastructure directed to those commodities. Also during the colonial period, Africa was developed essentially as an agricultural-exporting economy. This goal was achieved with some success, as evidenced by the number of African countries being top global producers of tropical cash crops.2 This orientation of agricultural production toward exports of primary products persisted during the 1960s, the era of Africa’s independence from colonial rule, except now the export revenues and develop- ment assistance in many countries were concentrated on financing ambitious domestic manufacturing activities under import substitution industrialization strategies and on developing the urban sector (Lawrence 2005). This was con- sistent with the “dual-economy” models of development, which viewed agri- culture as a low-productivity supplier of food, raw materials, and surplus labor to a modern and more urbanized industrialization process (Adelman 2001). As such, there was underinvestment in agriculture and in the rural sector (Fan 2008). Investments in agriculture were concentrated on input subsidies; government-provided services (marketing, infrastructure, extension, research); and the establishment of input and commodity marketing parastatals to pro- mote the export crops of the colonial era, which now provided African gov- ernments with their major source of foreign exchange (along with minerals in some countries). To ensure low food prices in the urban areas, food price controls and government-run estate farms and food marketing and distribution coopera- tives (which consumed the bulk of subsidies on farm inputs and machinery) were established. However, the import substitution manufacturing strategy was unsustainable for a variety of reasons. Protectionist policies employed by countries within and outside the continent constrained demand for man- ufactured goods to the size of the domestic market, which is small for many African countries. Groups of countries tried to overcome this constraint through customs unions. The East African Community, for example, had agreements concerning the location of specific manufacturing plants, so that 2 In the 1960s, the earliest periods when data were available, the highest-ranked African coun- try and the total number of African countries in the top 20 agricultural producers in the world were listed as follows: cocoa beans (Ghana was ranked number 1 in the world, with a total of 10 African countries in the top 20); green coffee (Côte d’Ivoire 3, total 8); unshelled groundnuts (Nigeria 3, total 12); palm oil (Nigeria 1, total 12); rubber (Liberia 6, total 6); sisal (Tanzania 2, total 10); tea (Kenya 7, total 6); tobacco (Zimbabwe 20, total 1); and cassava (Democratic Republic of the Congo 3, total 12) (FAO 2014). IntroduCtIon 3 production was not duplicated across the community; however, these agree- ments were not always adhered to. Furthermore, the factories were highly dependent on expensive imported capital and expatriate labor for producing mostly basic consumption goods (such as food processing, textiles and cloth- ing, and shoes) and processing primary products for exports, except in a few cases where intermediate goods were produced, such as fertilizer in Tanzania (Lawrence 2005). Neglect of smallholder farmers who produced the bulk of the food crops resulted in diminishing food production and rising food prices. These developments— in addition to leadership problems, economic mis- management, and corruption on the one hand and political turmoil and internal conflicts on the other— which many African countries experienced in the 1970s and 1980s, characterized the complex development issues in the continent at the time. The oil and drought shocks of the 1970s compli- cated the issues further. In general, the 1970s and 1980s are often associated with the beginning of the chronically poor performance of African agricul- ture. Between 1971 and 1980, for example, agricultural output in Africa south of the Sahara grew by only 1 percent per year on average, compared with 3 percent in Asia and other developing regions of the world, and land produc- tivity (output per unit area) was about two to three times lower (Table 1.1; Fuglie and Nin-Pratt 2013). The 1980s and 1990s ushered in the structural adjustment programs (SAPs) and the economic recovery programs (ERPs) of the International Monetary Fund (IMF) and the World Bank. The programs constituted con- ditions for receiving new loans or international development assistance, and involved cutting government expenditures, dismantling the parastatals, end- ing commodity and input subsidies, removing price controls, devaluing curren- cies, and stimulating private-sector investments to occupy the spaces left by the government-run agencies. While the overall impacts of the SAPs and ERPs are still debated, the prevailing view is negative, especially with regard to their impact on poverty (for example, Killick 1995; SAPRIN 2004; Easterly 2005).3 Because SAPs promoted economic output based on direct export and resource extraction, they also exacerbated the lack of attention on the rural sector, smallholder farmers, and food crops. For example, because devaluation makes local goods cheaper for foreigners to buy and foreign goods more expensive to 3 There is vast scholarly literature on the SAPs and ERPs. Killick (1995) provides a good review of the literature and highlights the difficulties in making generalizations about the effects of SAPs and ERPs because of data and methodological problems in dealing with the complex and varied instruments employed in the SAPs and ERPs on the one hand, and the many different, but con- nected, outcomes on the other. 4 Chapter 1 import, it provides incentives for SAP-implementing countries to export more and import less in the long run. However, by simultaneously devaluing the currency and removing subsidies, the immediate effect of structural adjust- ment was to raise the prices of agricultural inputs, especially those of yield- enhancing technologies, such as fertilizers, pesticides, and machinery, which are typically imported. The consequences were higher farm production costs, low adoption of high-yielding technologies, low agricultural productivity, and low incomes to smallholder farmers. Furthermore, private- sector investments did not materialize as expected, and new problems related to market failures surfaced (Dorward, Kidd, and Poulton 1998; Kherallah et al. 2002). Table 1.1 Annual average agricultural growth, productivity, and public spending in Africa and other selected developing regions of the world, 1970–2010 Indicator and region Years and values Agricultural output growth rate (%) 1 1971–1980 1981–1990 1991–2000 2001–2010 africa south of the Sahara 1.0 2.7 3.1 2.6 asiaa 3.0 4.1 4.0 3.5 Latin america and the Caribbean 2.9 2.4 3.1 3.2 Agricultural output per hectare of land (constant 2004–2006 US$) 1 1980 1990 2000 2009 africa south of the Sahara 163 182 192 219 asiaa 494 607 704 773 Latin america and the Caribbean 326 368 394 424 Government agriculture expenditure (% of total expenditure) 2 1981–1990 1991–2000 2001–2010 africa south of the Sahara 7.1 3.3 3.1 asiab 7.2 5.0 5.5 Latin america and the Caribbean 3.6 3.2 2.0 Government agriculture expenditure (% of agriculture value-added) 2 1981–1990 1991–2000 2001–2010 africa south of the Sahara 4.9 3.0 3.9 asiab 3.7 3.0 4.6 Latin america and the Caribbean 7.2 7.5 7.4 Agriculture R&D in Africa south of the Sahara 3 1971–1980 1981–1990 1991–2000 2001–2008 Growth rate in expenditure (%) 1.7 0.6 1.0 2.4 Growth rate in full-time-equivalent staff (%) 5.4 3.8 1.3 2.8 Source: authors’ calculations based on 1 Fuglie and nin-pratt (2013), 2 IFprI (2014a), and 3 Beintema and Stads (2011). Notes: a Made up of northeast, South, and Southeast asia. b South asia. r&d = research and development. IntroduCtIon 5 The austerity measures imposed by the SAPs led to a drastic reduction in government spending on agriculture in general (IFPRI 2014a), and an ero- sion of critical agricultural investments in national research and extension sys- tems in particular (Beintema and Stads 2011). For example, in Africa south of the Sahara, the share of government agriculture expenditure declined from an average of 7.4 percent per year of the total budget in the 1980s to 3.3 percent in the 1990s, whereas the growth rate in the amount spent on agriculture research and development (R&D) declined from an annual aver- age of 1.7 percent in the 1970s to 0.6 percent in the 1980s and 1.0 percent in the 1990s (Table 1.1). Therefore, although growth in African agriculture was higher in the 1980s and 1990s than in the 1970s— thanks largely to area expansion, rather than to the adoption of yield-enhancing technologies— agricultural productivity remained very low compared with levels achieved in other developing regions of the world, especially in Asia, where the Green Revolution was taking root. In 1980 and 1990, for example, agricultural output per hectare of land in Africa south of the Sahara was $163 and $180, respectively— about one-third of the values achieved in Asia (Fuglie and Nin- Pratt 2013; Table 1.1).4 It is important to remember that the Green Revolution in Asia occurred before the SAPs were established in Africa. After starting in Mexico, the Green Revolution quickly spread to Asia, where it is widely acknowledged to have doubled both output and yields of key food staples— rice and wheat— in just 20 years. These successes helped promote a broader reassessment of agri- culture’s role in Africa’s development, which we will return to shortly. The start of the new millennium introduced a greater emphasis on a more comprehensive approach to poverty reduction, in which agriculture was called on to play a more significant role. National strategies were formalized into Poverty Reduction Strategy Papers (PRSPs), required by the IMF and the World Bank for countries requiring debt relief and seeking new development assistance. While PRSPs have been described by some as simply an extension of SAPs (e.g., SAPRIN 2004), they are based— in theory if not always in prac- tice— on a more broadly based articulation of development, including the need for poverty-focused growth, participatory processes in strategic planning, public– private partnerships, and other principles that are expected to ensure that the benefits of growth are distributed to all members of society. Although the impact of the PRSPs in Africa is still being debated, their primary focus on poverty— a particularly prevalent phenomenon in rural 4 All currency is in US dollars, unless specifically noted as “international dollars.” 6 Chapter 1 areas— suggests that proper implementation of such plans should increasingly favor agricultural and rural development. For example, the use of agricultural input and farm support subsidies, which was discouraged under the SAPs, has returned strongly, particularly following the recent high food and input prices crisis. This is consistent with several studies prior to the start of the PRSPs, which recommended that the World Bank and IMF revisit their posi- tion on input subsidies by considering their merits in the broader context of agricultural intensification, in addition to their macroeconomic feasibility (for example, Lele, Christiansen, and Kadiresan 1989; Reardon et al. 1999; World Bank 1994). The successes of the Green Revolution in Asia also helped promote this movement, although this is not apparent in the PRSPs. For example, whereas most of the PRSPs state in various ways that raising agricultural output and productivity will be accomplished by promoting and supporting the use of yield-enhancing technologies and modern management practices, as done during the Green Revolution in Asia, only a few country PRSPs made direct reference to employing lessons or technologies from India (for exam- ple, Ghana 2003; IMF 2006), while Madagascar’s PRSP made explicit refer- ence to creating a Green Revolution there (Madagascar 2007). These PRSPs have promoted greater adoption of yield-enhancing technologies and modern management practices and helped return agricultural productivity to levels achieved prior to the decline in the 1970s— although still much lower than levels achieved in other developing regions of the world (Table 1.1). Against this historical narrative, the literature examining the poor perfor- mance of African agriculture has largely formulated hypotheses based on par- tial analyses, local contexts, and particular points in time. Associated findings and recommendations, therefore, often fall short of addressing the fundamental issues in their entirety. For example, it is reasonable to assume that Africa needs a movement similar to Asia’s Green Revolution. What we now know about that brief period in history is that it involved more than just high- yielding, semi- dwarf rice and wheat varieties. It also included investments in irrigation infra- structure, modernization of farm management techniques, supportive public policies, a strong geopolitical undercurrent, and a clear smallholder focus tied to its geopolitical motivation (Djurdfeldt et al. 2005), and had significant envi- ronmental consequences (for example, Shiva 1991). Nevertheless, the Green Revolution offers one— and only one, possibly irreplicable— model for the intensification and modernization of agriculture in Africa. Why may Asia’s Green Revolution not be replicable in Africa? This ques- tion derives from some of the arguments that have been advanced for the poor IntroduCtIon 7 performance of African agriculture, which also are consistent with differ- ent parts of the historical narrative or with different geographical contexts of the continent, including agroecological complexities and heterogeneity that make it difficult to exploit intercontinental technology spillovers (for exam- ple, Pardey et al. 2007); poor economic policies and, in particular, lack of openness to international markets or access to ports (for example, Sachs and Warner 1997); and the low productivity and high cost of labor (Karshenas 2001; Collier and Dercon 2009; Woodhouse 2009). Regarding the agroecological complexities and technology spillover con- straints, for example, many countries in Africa have small economies and lim- ited capacities and resources for adopting or adapting technologies that fit their own national interests and needs. Thus, although regional agricultural R&D systems can help fill these gaps and facilitate economies of scale,5 high transaction costs associated with political, institutional, and administrative barriers can rapidly erode the potential gains— gains that can differ substan- tially by commodity and by the degree of agroecological similarity between technology source and the target areas for technology application. The agro- ecological complexities mitigating the replicability of Asia’s Green Revolution in Africa are further complicated by climate change and global warming, a topic that, because of its recent emergence and its potential substantial effects on potential growth and development pathways, is logically absent in the his- torical narrative of the performance of African agriculture. We will address the implications of this topic for future development strategies after we have examined the current African agricultural development strategy. The labor constraint arguments for the poor performance of African agri- culture are representative of the issues related to the supply of key factors of production, including capital. All of these factors require an economywide, rather than a sectoral, approach to development because of the strong forward and backward linkages between the agriculture and nonagriculture sectors (Diao et al. 2012). For labor specifically, the possibility of increasing labor pro- ductivity depends on the availability of appropriate labor-saving technologies in agriculture, having profitable exit options out of agriculture into other sec- tors of the economy, and having high-quality labor to be able to earn a higher wage or return to labor in the other sectors. 5 See Omamo et al. (2006), Nin-Pratt et al. (2011), and Johnson et al. (2014) on the potential gains from implementing such regional agricultural R&D strategies in different subregions of the continent. 8 Chapter 1 Current African Agricultural Development Strategy In July 2003, African heads of state at the Second Ordinary Session of the Assembly of the African Union launched CAADP in Maputo, Mozambique. This agriculture-led integrated framework of development priorities in Africa is aimed at reducing poverty and increasing food security in the continent (AU-NEPAD 2003). The program shares many of the principles articulated in the PRSPs, including poverty-focused growth, participatory processes in strategic planning and implementation, country ownership, public– private partnership, and mutual accountability, to ensure that the benefits of growth are equitably distributed to all members of society. The main difference between CAADP and preceding development strat- egies in Africa is that it emphasizes the role of agriculture as the engine of economic growth and development in its compact-signing countries. Furthermore, CAADP deliberately categorizes investment into four mutually reinforcing pillars (land and water management, market access, food security, and agricultural R&D) and cross-cutting enabling factors, including institu- tional capacity strengthening. It also prescribes specific policies and programs to be implemented, in addition to specific targets to be achieved. CAADP has two overarching targets: (1) achieving an annual average agricultural growth rate of 6 percent, and (2) spending 10 percent of the national budget on agri- culture— popularly known as the Maputo Declaration (AU 2003). Various processes at national, regional, and continental levels have been put in place to ensure evidence-based planning, to facilitate implementation of CAADP according to the declared principles, to monitor and evaluate progress, and to promote mutual learning (AU-NEPAD 2014c). The impact of CAADP on agricultural and economic growth, poverty, and food and nutrition security is yet to be assessed. However, a little more than a decade since its launch in 2003, CAADP can point to several achieve- ments. For example, CAADP has significantly raised the political profile of agriculture; has contributed to more specific, purposeful, and incentive-ori- entated agricultural policies; and has promoted greater participation of multiple state and nonstate actors in agricultural policy dialogue and strat- egy development (AU-NEPAD 2010). Some of the specific tools, mecha- nisms, and processes that have contributed to these achievements include the annual CAADP Partnership Platform and Business meetings since 2006 that bring the different stakeholders at different levels together to review progress and make plans for the future (AU-NEPAD 2014a); preparation of the four pillar framework documents to guide adaptation of the CAADP IntroduCtIon 9 principles and targets into national and regional policymaking (AU-NEPAD 2010); establishment of the knowledge systems to provide analyses that track progress, document success, and derive lessons for the implementation of the CAADP agenda (IFPRI 2014b); development of a monitoring and evalua- tion (M&E) framework (Benin, Johnson, and Omilola 2010) and a mutual accountability framework (Oruko et al. 2011); and establishment of the CAADP Multi-Donor Trust Fund (MDTF) to finance the CAADP pro- cesses at all levels (AU-NEPAD 2010). By the end of 2014, 40 African coun- tries had signed their CAADP compacts with their main stakeholder groups, and many of them had developed detailed country investment plans (or National Agricultural Investment Plans [NAIPs] or National Agricultural and Food Security Investment Plans [NAFSIPs]). Furthermore, a majority of the strategies and plans are based on economywide analysis in order to iden- tify coherent growth options and quantify the aggregate public agricultural resources required to support different growth paths (for example, Diao et al. 2012). Despite these and other achievements that can be attributed to CAADP, several challenges have arisen. First is assessing the impact of CAADP, where the major issue involves attributing change in the outcome indicators to CAADP. This assessment is difficult because many governments and coun- tries were already engaged in policy reforms in harmony with the CAADP principles, and much of the CAADP framework was derived from earlier strategies and successful agricultural reforms in those African countries. The issue, therefore, will be how to isolate CAADP’s specific contributions. A second challenge is the delayed response in adapting a continental-level agenda and commitments to fit regional- and national-level priorities or vice versa. For example, when CAADP was launched in 2003, the heads of state set a five-year timeline for implementation (see AU 2003, Declaration 7(II).2). By 2008, however, only Rwanda had a signed CAADP Compact to demon- strate a concrete implementation progress (AU-NEPAD 2014b). These delays reflected inherent political, institutional, and administrative barriers across national boundaries. Therefore, the heads of state renewed their commitment through a resolution at the 13th Ordinary Session of the Assembly of the African Union (AU) in Sirte, Libya, in July 2009 by requesting the AU Commission, the NEPAD [New Partnership for Africa’s Development] Secretariat and the RECs [Regional Economic Communities] to continue to mobilize the necessary technical exper- tise and financial resources to support capacity development and 10 Chapter 1 related policy reforms to accelerate CAADP implementation in all Member States, including the signing of country CAADP Compacts indicating the policy measures, investment programs, and required funding to achieve the six percent (6 percent) growth and ten percent (10 percent) budget share targets for the agricultural sector by 2015. (AU 2009, Declaration 2(XIII).5) Consequently, 12 more countries and the Economic Community of West African States signed their compacts in 2009, an acceleration spurred by the establishment of the MDTF to finance CAADP processes in 2008 and the establishment of the Global Agriculture and Food Security Program (GAFSP) in 2009 to assist in securing Group of Twenty (G20) pledges that would support the financing of NAIPs. Other challenges faced by CAADP related to achieving the 10 percent budget allocation and the 6 percent growth rate targets and to complet- ing development of the NAIPs. With regard to progress made toward the 10 percent budget allocation target for agriculture, Table 1.1 shows that Africa south of the Sahara managed to reach only 3.1 percent on average between 2001 and 2010 (IFPRI 2014a). Since 2003, only 13 countries in all of Africa have managed to surpass the target in any year (Benin and Yu 2013). NEPAD also has set a national agricultural R&D investment target of at least 1 percent of agricultural value-added, which only a few countries have been able to achieve so far (Figure 1.1)— especially Botswana, Mauritius, Namibia, and South Africa, all of which have relatively well-established and well-funded agricultural research systems and relatively small contributions of agricul- ture to gross domestic product (Beintema and Stads 2011). Regarding prog- ress toward achieving the 6 percent agricultural growth target, Table 1.1 shows that Africa south of the Sahara managed to reach only 2.6 percent on average between 2001 and 2010 (Fuglie and Nin-Pratt 2013). Between 2003 and 2009, for example, only six countries— Angola, Ethiopia, Guinea, Mozambique, Nigeria, and Rwanda— met or surpassed the target (Benin et al. 2011). Looking now at the plans for the future, results of the economic model- ing used in CAADP planning indicate that although it is possible for many African countries to reach the 6 percent annual average agricultural growth rate target, it will require substantial additional growth across different key subsectors and commodities. This in turn will require substantial addi- tional investments to stimulate the necessary acceleration in growth in the key subsectors (for example, Diao et al. 2012). In many cases, the additional IntroduCtIon 11 investments required are in excess of the 10 percent of total expenditures com- mitment agreed upon under the Maputo Declaration. Such large demands on fiscal resources are necessary because of inadequate or no technical change in the sector (for example, Irz and Thirtle 2004; Nin-Pratt and Yu 2008). As countries enter the operational stage of CAADP investment program design and execution, a fundamental question and technical and institutional chal- lenge is how to achieve and sustain significantly higher levels of agricultural productivity across different parts of Africa. Given the limits to boosting pro- ductivity that is achievable through area expansion, as has been experienced in many parts of Africa for long periods of time, agricultural productivity gains in the future must rely heavily on technological change. A review of the NAIPs shows that individual countries have formulated different strategic responses to these common policy, technology, and institu- tional challenges. Such variation is expected, for example, since climate and natural resource endowments that condition strategic agricultural develop- ment options differ considerably among countries. Table 1.2 and Figure 1.2 show some of the clear differences in investment and development approaches among NAIPs in terms of the proportion of the total agriculture budget that is allocated to different priority areas and investments. With regard to the general approach to agricultural development, for example, Table 1.2 records budget allocations according to the overall agriculture sector goal, the four 0 1 2 3 4 5 Central Region Eastern Region Northern Region Southern Region Western Region All Public agricultural R&D spending (annual average % of total agGDP) 1996–2003 2003–2008 NEPAD 1% target Ga bo n Co ng o, R ep . Bu ru nd i Su da n M ad ag as ca r Ta nz an ia Et hi op ia Rw an da Er itr ea Ug an da Ke ny a M au rit iu s Tu ni si a M or oc co M au rit an ia Za m bi a M oz am bi qu e M al aw i N am ib ia So ut h Af ric a Bo ts w an a Ni ge r Gu in ea Si er ra L eo ne Ni ge ria To go Bu rk in a Fa so Be ni n Ga m bi a, T he Cô te d 'Iv oi re Gh an a M al i Se ne ga l Al l FIgURe 1.1 Public expenditure on agricultural research and development in selected African countries, 1996– 2008 (annual average % of agricultural value-added) Source: author’s calculations based on IFprI (2013). 12 Chapter 1 CAADP pillars, the CAADP cross-cutting theme, and other areas. The table illustrates that achieving the agriculture sector goal of increasing agricultural productivity, growth, or income represents the dominant strategy in many of the African countries reported. However, in several of the other coun- tries— for example, Ethiopia, The Gambia, Liberia, Malawi, Niger, and Sierra Leone— food and nutrition security (pillar 3) and natural resource manage- ment (pillar 1) are given higher priority. Pillars 2 and 4 and the cross-cutting theme were accorded lower priority in terms of the stated budget allocations. Whereas drawing conclusions from these budget shares is difficult, because different countries may invest differently to achieve different goals and objec- tives, the disparate results are consistent with a fundamental knowledge gap about the drivers of high levels of agricultural productivity growth across the continent. A similar implication derives from Figure 1.2, which reports allocations to specific subprograms or functions that are known to be critical for over- all agricultural productivity growth, including research, extension, irrigation, natural resource management, and farm support subsidies. Although these represent the major functions that were articulated in the NAIPs, the results in Figure 1.2 show that budgets were not necessarily allocated accordingly. The figure also highlights differences across governments and stakeholders in individual countries in terms of making explicit resource allocation com- mitments to such specific agricultural functions. Clearly, commitments to invest in natural resource management and providing farm support subsi- dies were favored or seemed easier to make in many countries in terms of attracting large shares of the agriculture budgets. These commitments were followed by investment in irrigation. Although investing in research and extension has been found to have large and long-lasting impacts on agricul- tural growth and other development outcomes (for example, Fan, Hazell, and Thorat 2000; Fan 2008; Mogues et al. 2012), they were stated priori- ties in only a handful of countries, including Benin, Burundi, Côte d’Ivoire, and Uganda. Therefore, although we expect countries to have different strategic responses to achieving the CAADP targets and their own national objectives in ways that reflect their own national contexts that are also shaped by such noneconomic factors as political, cultural, social, historical, and linguistic fac- tors, the contrasting results shown in Figure 1.2 would suggest that there is a knowledge gap about the drivers of high levels of agricultural productivity growth across Africa. IntroduCtIon 13 Table 1.2 Stated budget allocation to the top three programs in selected African countries (percentage of total NAIP budget) CAADP pillar/theme African countries Sector goal Pillar 1 Pillar 2 Pillar 3 Pillar 4 Cross- cutting Other Benin, 2010–2015 51.9 2.7 — 44.7 — — 0.7 Burkina Faso, 2011–2015 67.9 — 17.7 — — 11.9 2.5 Burundi, 2012–2017 55.9 — 19.0 — — 20.1 4.9 Côte d'Ivoire, 2010–2015 41.8 — 14.9 — — 24.3 19.0 ethiopia, 2010–2020 3.4 57.4 — 17.1 — — 22.1 the Gambia, 2011–2015 — 27.9 30.3 15.2 — — 26.6 Ghana, 2011–2015 55.7 — — 36.9 3.4 — 4.0 Kenya, 2010–2015 36.0 42.0 13.1 — — — 8.9 Liberia, 2011–2015 — — 32.6 39.9 — 14.4 13.0 Malawi, 2011–2014 — 36.6 — 46.9 6.2 — 10.4 niger, 2010–2012 — 34.4 — — — 12.6 53.0 nigeria, 2011–2014 35.5 40.9 12.7 — — — 10.8 rwanda, 2009–2012 77.7 — 15.1 — — 4.9 2.3 Senegal, 2011–2015 59.4 31.0 — — — — 9.6 Sierra Leone, 2010–2014 17.3 — 23.6 33.7 — — 25.4 tanzania, 2012–2016 71.1 13.7 — — — 7.8 7.4 togo, 2010–2015 66.1 — — — 9.0 15.3 9.6 uganda, 2011–2015 68.6 — 25.0 — — 4.2 2.2 Key: Pillar 1: natural resource management (land, water, climate, etc.); Pillar 2: Competitiveness, market trade, and private-sector development; Pillar 3: Food and nutrition security and emergency preparedness; Pillar 4: Science and technology; Cross Cutting: enabling environment (policies, institutions, good governance) Source: authors’ calculations based on national agricultural investment plans (naIps). the plans can be viewed and down- loaded at www.resakss.org and http://www.caadp.net/library-country-status-updates.php. Notes: this table has been prepared to show allocations to the top three programs only. Because the budgets in the different naIps were presented in different formats, the six programs identified here try to capture allocations to the overall agricul- ture sector goal (productivity, growth, income); the four Caadp pillars; and the cross-cutting theme. Furthermore, because not all the naIps had budget allocations for these programs, the blank spaces are intentional, so as to not crowd the table. the calculations are based on the stated amounts allocated to different programs, in terms of share of total budget. a blank space means that the calculated share allocated to the related program is not in the top three programs when compared with the shares allocated to different programs. Because different naIps have different programs, a blank space may also indicate that the related program is not stated in that country’s naIp. therefore, the last column, labeled “other,” collects the remaining shares outside of the top three programs, so that the total for the row or country adds up to 100 percent. — = not applicable. 14 Chapter 1 Another challenge faced by CAADP is following through with the vari- ous and increasing number of policies, initiatives, and principles, particularly with those emerging since its initial launch in 2003. For example, the Global Agriculture and Food Security Program (GAFSP 2014), the L’Aquila Food Security Initiative (AFSI, G8 2009), and the New Alliance for Food Security and Nutrition (G8 2012), which have emerged and are expected to comple- ment CAADP, require additional funding-eligibility processes. Therefore, although these new initiatives clearly state that CAADP compliance is an essential prerequisite for securing potential country support, the total amount of additional resources they provide relative to the status quo is often unclear (see, for example, Benin 2014 on the contribution of AFSI). Because the goal of CAADP here is to develop partnerships to meet the necessary policy, bud- getary, and development assistance needs of the NAIPs, the cost of involve- ment in and management of multiple partnerships may not be apparent. FIgURe 1.2 Stated budget allocation to selected agricultural functions in selected African countries (percentage of total NAIP budget) 0 10 20 30 40 50 60 70 80 90 Be ni n, 2 01 0– 20 15 Bu rk in a Fa so , 2 01 1– 20 15 Bu ru nd i, 20 12 –2 01 7 Cô te d 'Iv oi re , 2 01 0– 20 15 Et hi op ia , 2 01 0– 20 20 Th e Ga m bi a, 2 01 1– 20 15 Gh an a, 2 01 1– 20 15 Ke ny a, 2 01 0– 20 15 Li be ria , 2 01 1– 20 15 M al aw i, 20 11 –2 01 4 M al i, 20 11 –2 01 5 Ni ge r, 20 10 –2 01 2 Ni ge ria , 2 01 1– 20 14 Rw an da , 2 00 9– 20 12 Se ne ga l, 20 11 –2 01 5 Si er ra L eo ne , 2 01 0– 20 14 Ta nz an ia , 2 01 2– 20 16 To go , 2 01 0– 20 15 Ug an da , 2 01 1– 20 15 Farm Support and Subsidies Natural resource management Irrigation Extension Research Source: authors’ calculations based on national agricultural investment plans (naIps). the plans can be viewed and down- loaded at www.resakss.org and http://www.caadp.net/library-country-status-updates.php. Notes: this figure has been prepared to show the stated budget allocations in the naIps for the five agricultural functions only, which represent the major functions articulated in the naIps and are consistent with the major tenets of the Green revolution. Because the budgets in the different naIps were presented in different formats, not all the naIps had budget allocations for these agricultural functions. however, because all five functions were identified in all of the naIps as being important for achieving their respective development objectives, a zero share applied to any of the five functions indicates that there was no information to estimate the share of the budget for that function. as a result, the percentages do not add up to 100, because the total budget was not allocated exhaustively to the five functions. IntroduCtIon 15 A different kind of challenge that the NAIPs may only now be starting to internalize is global warming and climate change, which could affect agricul- ture in several ways, including productivity effects in terms of the quantity and quality of outputs; husbandry effects through changes in water availabil- ity and use of yield-enhancing technologies; environmental effects, such as soil erosion, water pollution, and reduction of diversity; land use, such as through land valuation and speculation; and adaptation in response to changes in the functional characteristics of organisms and ecological systems. Several studies (Kurukulasuriya et al. 2006; IPCC 2007; Seo et al. 2008; Nelson et al. 2010) provide strong evidence that climate change caused by accumulating green- house gases is likely to impose serious costs on agricultural growth. Nelson et al. (2010), for example, show that the negative effect of climate change on crop yields will increase over time, whereas Seo et al. (2008) show that the impacts of climate change will vary across different agroecological zones in Africa— farms in the savanna areas are expected to be the most vulnerable to higher temperature and reduced precipitation, while those in subhumid or humid forests could gain even from severe climate change. Because of the agroecolog- ical complexities in Africa, having information on specific local regions will be critical for identifying climate-smart agricultural interventions among the numerous possibilities to increase the resilience of livelihoods and production systems and to maximize the effects of technological changes on growth and development in a sustainable and equitable manner. Objectives and Organization of This Book By improving understanding of the spatial and temporal patterns of a range of productivity measures assessed consistently and comparably across Africa, this book is intended to contribute to the knowledge base of how best to achieve and sustain significantly higher levels of agricultural productivity. While indi- vidual countries have taken various investment and development approaches in preparing their NAIPs, a critical question remains: Which strategies work best in which contexts, and do so cost-effectively? In addressing this question, we base our analysis on the now rapidly expanding base of agricultural data in Africa, including geographically spe- cific information on production system heterogeneity, quality of natural resources, population density, infrastructure, and market access. We present analyses and findings aimed at improving our understanding of the status of and trends in African agricultural productivity and its determinants and, on that basis, identifying opportunities for agricultural productivity growth that 16 Chapter 1 lead to more effective design and implementation of agricultural policies and strategies in Africa. The book’s unique mix of data sources, detailed in the relevant chap- ters, includes time-series data on agricultural production from the Food and Agriculture Organization of the United Nations, national accounts from the World Bank, and public expenditure and project M&E data from gov- ernments and multilateral agencies. We acknowledge the legitimate concerns about data reliability, as highlighted in Jerven (2013). We address these con- cerns by triangulating among a range of independent sources and types of data (for example, static cross-sectional and time-series data, spatial and nonspatial data), which we believe has reduced some potential data pitfalls. For exam- ple, because the spatial data used are based on observed measures of outcomes, rather than self-reported data, the measurement errors associated with captur- ing only the formal sector are eliminated, although measurement errors asso- ciated with the methodology used to collect or compile the data remain. The specific approaches taken to combine different data components and some of the challenges involved are described in the individual chapters. This introductory chapter is followed by analysis of intertemporal trends (Chapter 2) and spatial analysis of different indicators and measures (Chapter 3) of agricultural productivity. Taken together, Chapters 2 and 3 provide a broad overview of the contemporary landscape of African agricul- tural productivity, and highlight the relevance and utility of different mea- sures of agricultural productivity in M&E. Chapter 2 involves defining, calculating, and interpreting trends in partial and total productivity measures of agricultural productivity using time-series data. In contrast, Chapter 3 brings a more spatially explicit perspective on productivity using a harmo- nized collection of Africa-wide geographic information system data (some 300,000 10 x 10–kilometer grid cells), in order to explore different production systems (including rainfed and irrigated cropping systems and livestock sys- tems) and partial productivity measures. Building on Chapter 3, Chapter 4 uses statistical and econometric meth- ods, particularly spatial and cluster techniques, to develop a typology of agri- cultural productivity zones (APZs) according to similarity in their likely pathways of technology adoption and agricultural productivity growth. Chapter 5 zooms in to examine some of the dominant APZs developed in the preceding chapter, and analyzes the status of and recent trends in patterns of intensification, as well as changes in output composition and input use asso- ciated with different intensification patterns. In particular, the chapter exam- ines the use of fertilizer and its role in the intensification pathways followed IntroduCtIon 17 by different subregions in Africa in recent years, and the implications of those patterns for agricultural growth and policymaking. Chapter 6 examines case studies of agricultural investment programs and value chains in different parts of Africa that were intended for enhanc- ing agricultural productivity. Using a qualitative and narrative approach, the chapter aims to identify what did or did not work well where and why, by dis- tilling lessons on key factors contributing to the effectiveness of the invest- ment programs. Chapter 7 summarizes and synthesizes the key insights and findings pro- vided in the preceding chapters, focusing on major challenges to and opportu- nities for raising African agricultural productivity, including investments in agricultural R&D, cross-border technology spillover, and institutional capac- ity. While the unique mix of methodologies and data used is a major strength in the book, particularly in terms of its policy relevance, it also reflects the dif- ficulty of compiling coherent and comprehensive sets of sufficiently reliable and interoperable data. We close with a call for governments and development agencies to invest more in strengthening national data and statistical systems (especially for compiling national production and public accounts data), as well as the human capacity to use more accessible, timely, and reliable data to support policy and investment decisionmaking at all levels. References Adelman, I. 2001. “Fallacies in Development Theory.” In Frontiers of Development Economics, edited by G. M. Meier, and J. E. Stigliz. Oxford, UK: Oxford University Press. Anthony, K. R. M., B. F. Johnston, W. O. Jones, and V. C. Uchendu. 1979. Agricultural Change in Tropical Africa. Ithaca, NY: Cornell University Press. AU (African Union). 2003. “Declarations.” Assembly of the African Union, Second Ordinary Session, July 10– 12, 2003, Maputo, Mozambique. Accessed April 15, 2014. http: / /www .nepad .org /system /files /Maputo%20Declaration .pdf . —. 2009. “Decisions and Declarations.” Assembly of the African Union, Thirteenth Ordinary Session, July 1– 3, 2009, Sirte, Libya. Accessed April 15, 2014. www .au .int /en /sites /default /files /ASSEMBLY _EN _1 _3 _ JULY _2009 _AUC _THIRTEENTH _ORDINARY _ SESSION _DECISIONS _DECLARATIONS _%20MESSAGE _CONGRATULATIONS _MOTION _0 .pdf . AU-NEPAD (African Union– New Partnership for Africa’s Development). 2003. Comprehensive Africa Agriculture Development Programme. Midrand, South Africa. 18 Chapter 1 —. 2010. The Comprehensive Africa Agriculture Development Programme (CAADP) in Practice: Highlighting the Success. Midrand, South Africa. —. 2014a. “CAADP Partnership Platform.” Accessed April 2014. http: / /www .nepad-caadp .net /library-meeting-documents .php . —. 2014b. “Countries with Compacts/Investment Plans.” http: / /www .nepad-caadp .net /pdf / Table%201%20Countries%20with%20Investment%20Plans%20ver19 .pdf . —. 2014c. “Implementing the CAADP Agenda.” Accessed April 2014. http: / /www .nepad -caadp .net /implementing-caadp-agenda .php . Beintema, N. M., and G. J. Stads. 2011. African Agricultural R&D in the New Millennium: Progress for Some, Challenges for Many. Washington, DC: International Food Policy Research Institute. Benin, S. 2014. Aid Effectiveness: How Is the L’Aquila Food Security Initiative (AFSI) Doing? IFPRI Discussion Paper 1329. Washington, DC: International Food Policy Research Institute. Benin, S., M. Johnson, and B. Omilola. 2010. Monitoring and Evaluation (M&E) System for the Comprehensive Africa Agriculture Development Programme (CAADP). ReSAKSS Working Paper 6. Washington, DC: International Food Policy Research Institute. Benin, S., A. Nin-Pratt, S. Wood, and G. Guo. 2011. Trends and Spatial Patterns in Agricultural Productivity in Africa, 1961– 2010. ReSAKSS Annual Trends and Outlook Report 2011. Washington, DC: International Food Policy Research Institute. Benin, S., and B. Yu. 2013. Complying with the Maputo Declaration Target: Trends in Public Agricultural Expenditures and Implications for Pursuit of Optimal Allocation of Public Agricultural Spending. ReSAKSS Annual Trends and Outlook Report 2012. Washington, DC: International Food Policy Research Institute. Benson, T., N. Minot, J. Pender, M. Robles, and J. von Braun. 2013. “Information to Guide Policy Responses to Higher Global Food Prices: The Data and Analyses Required.” Food Policy 38: 47– 58. Benson, T., S. Mugarura, and K. Wanda. 2008. “Impacts in Uganda of Rising Global Food Prices: The Role of Diversified Staples and Limited Price Transmission.” Agricultural Economics 39 (suppl. 1): 513– 524. Binswanger, H. P., and R. F. Townsend. 2000. “The Growth Performance of Agriculture in Sub-Saharan Africa.” American Journal of Agricultural Economics 82 (5): 1075– 1086. Collier, P., and S. Dercon. 2009. “African Agriculture in 50 Years: Smallholders in a Rapidly Changing World?” Paper presented at the Organization of the United Nations Economic and Social Development Department Expert Meeting “How to Feed the World in 2050,” Rome, June 24– 26. IntroduCtIon 19 Diao, X., P. Hazell, D. Resnick, and J. Thurlow. 2007. The Role of Agriculture in Development: Implications for Sub-Saharan Africa. IFPRI Research Report 153. Washington, DC: International Food Policy Research Institute. Diao, X., J. Thurlow, S. Benin, and S. Fan, eds. 2012. Strategies and Priorities for African Agriculture: Economywide Perspectives from Country Studies. Washington, DC: International Food Policy Research Institute. Djurfeldt, G., H. Holmen, M. Jirstrom, and R. Larsson. 2005. The African Food Crisis: Lessons from the Asian Green Revolution. Cambridge, MA: CABI Publishers. Dorward, A. R., J. Kidd, and C. Poulton, eds. 1998. Smallholder Cash Crop Production under Market Liberalisation: A New Institutional Economics Perspective. Wallingford, UK: CABI Publishers. Easterly, W. 2005. “What Did Structural Adjustment Adjust? The Association of Policies and Growth with Repeated IMF and World Bank Adjustment Loans.” Journal of Development Economics 76 (1): 1– 22. Fan, S., ed. 2008. Public Expenditures, Growth, and Poverty: Lessons from Developing Countries. Baltimore: Johns Hopkins University Press. Fan, S., P. Hazell, and S. Thorat. 2000. “Government Spending, Agricultural Growth and Poverty in Rural India.” American Journal of Agricultural Economics 82 (4): 1038– 1051. FAO (Food and Agriculture Organization of the United Nations). 2014. “FAOSTAT: Commodities by Country.” Accessed April 15, 2014. http: / /faostat .fao .org /site /339 /default .aspx . Fei, J. C., and G. Ranis. 1961. “A Theory of Economic Development.” American Economic Review 51 (4): 533– 565. Fuglie, K., and A. Nin-Pratt. 2013. “Agricultural Productivity: A Changing Global Harvest.” In 2012 Global Food Policy Report. Washington, DC: International Food Policy Research Institute. G8 (Group of Eight). 2009. L’Aquila Joint Statement on Global Food Security: L’Aquila Food Security Initiative (AFSI). L’Aquila, Italy. Accessed April 15, 2014. http: / /www .g8italia2009 .it /static / G8 _Allegato /LAquila _ Joint _Statement _on _Global _Food _Security%5B1%5D,0 .pdf . —. 2012. “New Alliance for Food Security & Nutrition.” Accessed April 15, 2014. http: / / www .usaid .gov /unga /new-alliance . GAFSP (Global Agriculture and Food Security Program). 2014. “Global Agriculture and Food Security Program.” Accessed April 15, 2014. http: / /www .gafspfund .org /content /about-gafsp . Ghana (Government of Ghana). 2003. “Analysis and Policy Statement.” In Ghana Poverty Reduction Strategy, 2003– 2005: An Agenda for Growth and Prosperity. Vol. I. Accra. IMF. https://www.imf.org/external/pubs/ft/scr/2003/cr0356.pdf. 20 Chapter 1 Hayami, Y., and V. Ruttan. 1985. Agricultural Development: An International Perspective. Baltimore: Johns Hopkins University Press. Hazell, P., and S. Haggblade. 1991. “Rural-Urban Growth Linkages in India.” Indian Journal of Agricultural Economics 46 (4): 512– 529. Headey, D., F. Bachewe Nisrane, I. Worku, M. Dereje, and A. Seyoum Taffesse. 2012. Urban Wage Behavior and Food Price Inflation: The Case of Ethiopia. Ethiopia Strategy Support Program II (ESSP II) Working Paper 41. Addis Ababa: International Food Policy Research Institute. IFPRI (International Food Policy Research Institute). 2013. “Agricultural Science and Technology Indicators (ASTI) Database.” Accessed January 2013. www.asti.cgiar.org/data/. —. 2014a. DATASET. “Public Expenditure Database: Statistics on Public Expenditure for Economic Development (SPEED).” Accessed on March 15, 2014. http: / /www .ifpri .org / book−39 /ourwork /programs /priorities-public-investment /speed-database . —. 2014b. “Regional Strategic Analysis and Knowledge Support System (ReSAKSS).” Accessed April 15, 2014. http: / /www .resakss .org /about . IMF (International Monetary Fund). 2006. 2nd Generation PRSP, 2007– 2011. Mali: Poverty Reduction Strategy Paper. Washington, DC. IPCC (Intergovernmental Panel on Climate Change). 2007. State of the Science. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press. Irz, X., and C. Thirtle. 2004. “Dual Technological Development in Botswana Agriculture: A Stochastic Input Distance Function Approach.” Journal of Agricultural Economics 55 (3): 455– 478. Jerven, M. 2013. Poor Numbers: How We Are Misled by African Development Statistics and What to Do about It. Ithaca, NY: Cornell University Press. Johnson, M., S. Benin, L. You, X. Diao, P. Chilonda, and A. Kennedy. 2014. Exploring Strategic Priorities for Regional Agricultural Research and Development Investments in Southern Africa. IFPRI Discussion Paper 1318. Washington, DC: International Food Policy Research Institute. Karshenas, M. 2001. “Agriculture and Economic Development in Sub-Saharan Africa and Asia.” Cambridge Journal of Economics 25 (3): 315– 342. Kherallah, M., C. Delgado, E. Gabre-Madhin, N. Minot, and M. Johnson. 2002. Reforming Agricultural Markets in Africa. Baltimore: Johns Hopkins University Press. Killick, T. 1995. “Structural Adjustment and Poverty Alleviation: An Interpretative Survey.” Development and Change 26 (2): 305– 330. IntroduCtIon 21 Kurukulasuriya, P., R. Mendelsohn, R. Hassan, J. Benhin, M. Diop, H. M. Eid, K. Y. Fosu, et al. 2006. “Will African Agriculture Survive Climate Change?” World Bank Economic Review 20 (3): 367– 388. Lawrence, P. 2005. “Explaining Sub-Saharan Africa’s Manufacturing Performance.” Development and Change 36 (6): 1121– 1141. Lele, U., R. Christiansen, and K. Kadiresan. 1989. Issues in Fertilizer Policy in Africa: Lessons from Development Programs and Adjustment Lending, 1970– 87. MADIA Report. Washington, DC: World Bank. Lewis, W. A. 1954. “Economic Development with Unlimited Supplies of Labor.” The Manchester School of Economics and Social Studies 22 (2): 139– 191. Madagascar (Republic of Madagascar). 2007. “Madagascar Action Plan 2007– 2012: A Bold and Exciting Plan for Rapid Development.” http: / /www .un .org /esa /coordination /Alliance / MADAGASCAR%20ACTION%20PLAN .htm . Mogues, T., B. Yu, S. Fan, and L. McBride. 2012. The Impacts of Public Investment in and for Agriculture. IFPRI Discussion Paper 01217. Washington, DC: International Food Policy Research Institute. Nelson, G. C., M. W. Rosegrant, A. Palazzo, I. Gray, C. Ingersoll, R. Robertson, S. Tokgoz, et al. 2010. Food Security, Farming, and Climate Change to 2050: Scenarios, Results, Policy Options. Washington, DC: International Food Policy Research Institute. Nin-Pratt, A., M. Johnson, E. Magalhaes, L. You, X. Diao, and J. Chamberlin. 2011. Yield Gaps and Potential Agricultural Growth in West and Central Africa. Washington, DC: International Food Policy Research Institute. Nin-Pratt, A., and B. Yu. 2008. An Updated Look at the Recovery of Agricultural Productivity in Sub-Saharan Africa. IFPRI Discussion Paper 787. Washington, DC: International Food Policy Research Institute. Omamo, S. W., X. Diao, S. Wood, J. Chamberlain, L. You, S. Benin, U. Wood-Sichra, and A. Tatwangire. 2006. Strategic Priorities for Agricultural Development in Eastern and Central Africa. IFPRI Report 150. Washington, DC: International Food Policy Research Institute. Oruko, L., I. Randall, M. Bwalya, S. Kisira, and M. Wanzala. 2011. “Mutual Accountability Framework.” Accessed April 15, 2014. http: / /www .nepad .org /system /files /CAADP%20 MAF%20Final%20Draft%20Report%20%20March%202011 .pdf . Pardey, P. G., J. James, J. Alston, S. Wood, B. Koo, E. Binenbaum, T. Hurley, and P. Glewwe. 2007. Science, Technology and Skills. Working Paper 41371. Washington, DC: World Bank. Accessed September 2011. http: / /siteresources .worldbank .org /INTWDR2008 /Resources /2795087−1191427986785 /PardeyPEtAl _SciTech&Skills--ALL1 .pdf . 22 Chapter 1 Reardon, T., C. Barrett, V. Kelly, and K. Savadogo. 1999. “Policy Reforms and Sustainable Agricultural Intensification in Africa.” Development Policy Review 17 (4): 375– 395. Sachs, J. D., and A. M. Warner. 1997. “Sources of Slow Growth in African Economies.” Journal of African Economies 6 (3): 335– 376. SAPRIN (Structural Adjustment Participatory Review International Network). 2004. The Policy Roots of Economic Crisis and Poverty: A Multi-Country Participatory Assessment of Structural Adjustment. London: Zed Books. Seo, S. N., R. Mendelsohn, A. Dinar, R. Hassan, and P. Kurukulasuriya. 2008. A Ricardian Analysis of the Distribution of Climate Change Impacts on Agriculture across Agroecological Zones in Africa. Policy Research Working Paper 4599. Washington, DC: World Bank. Shiva, V. 1991. “The Green Revolution in the Punjab.” Ecologist 21 (2): 57– 60. Wodon, Q., and H. Zaman. 2010. “Higher Food Prices in Sub-Saharan Africa: Poverty Impact and Policy Responses.” World Bank Research Observer 25 (1): 157– 176. Woodhouse, P. 2009. “Technology, Environment and the Productivity Problem in African Agriculture: Comment on the World Development Report 2008.” Journal of Agrarian Change 9 (2): 263– 276. World Bank. 1994. “Feasibility of Phosphate Rock Use as a Capital Investment in Sub-Saharan Africa: Issues and Opportunities.” Unpublished report. Washington, DC. —. 2007. World Development Report 2008: Agriculture for Development. Washington, DC. IntroduCtIon 23 Introduction Changes or trends in agricultural productivity over time can shed light on the relative sources of agricultural growth as well as on resource and factor constraints to increasing agricultural production sustainably. Because improvements in agricultural productivity are important for reducing poverty and achieving other development objectives, it is essential to use the appropriate indicator and measure of agricultural productivity— partial factor productivity (PFP) or total factor productivity (TFP). However, because pro- ductivity embodies many different components, changes in productivity can catalyze a wide range of direct and indirect effects on the pathways to achiev- ing different development objectives. For example, output per worker or labor productivity, as indicators of PFP, may be better measures of productivity to identify linkages to nonagricultural growth, because it encapsulates the additional ways farm households earn income (Mellor 1999). Byerlee, Diao, and Jackson (2009) show that coun- tries with the highest agricultural growth per worker experienced the greatest rate of rural poverty reduction. Other measures of PFP have been found to be significant determinants of poverty. Datt and Ravallion (1998), for example, find that higher land productivity (measured by agricultural output per unit area) had greater effect in reducing absolute poverty than in reducing the pov- erty gap or squared poverty gap, suggesting that the gains from higher land productivity were via rising average living standards, rather than improved distribution. Because changes in PFP could be caused by change from a vari- ety of reasons, including change in the use of other inputs or change in out- put mix, the policy implications of changes in PFP measures are often unclear. Furthermore, changes in output and productivity also do not necessarily have similar impacts, and sometimes move in different directions, with differen- tial consequences for poverty (Schneider and Gugerty 2011), and productivity INTERTEMPORAL TRENDS IN AGRICULTURAL PRODUCTIVITY Samuel Benin and Alejandro Nin-Pratt Chapter 2 25 gains, depending on the distribution of assets, may have limited impact on poverty reduction (Thirtle, Lin, and Piesse 2003). Unlike these PFP measures, TFP measures provide a better sense of the changes in agricultural productivity that are attributable to technologi- cal change, which, for many policymakers and the Comprehensive Africa Agriculture Development Programme (CAADP), is a critical means of improving African agriculture. Known as the Solow residual, TFP measures the part of growth that is not accounted for by changes in conventional fac- tors of production, such as land, labor, or capital. As a residual, however, the source of TFP growth is varied. Fan, Hazell, and Thorat (2000) found that investments in roads, agricultural research and development, and education were significant determinants of TFP, which in turn had a substantial effect on reducing poverty via reduced prices and increased wages, but at the cost of increased landlessness. Rahman and Salim (2013) show that the different sources can have different effects on the various components of TFP, which, similar to the case with the PFP measures, also suggests that the policy impli- cations of changes in TFP can be complex. Deciding what indicator and measure of agricultural productivity to use is complicated by knowledge gaps across several dimensions of the different components embodied in productivity, including • Composition of agriculture— sector (all agriculture), subsector (crops, live- stock, fisheries, forestry), commodity group (such as cereal, export crops, meat), and commodity (such as maize, rice, beef, tilapia); • Type of factor (land, labor, capital), input (seed, fertilizer, feed), or hus- bandry (plant spacing, weeding, intensive livestock management); • Measure of output and input— physical quantity or monetary value, which is important when aggregating across several subcomponents, because summing over weights or volumes may not be meaningful; • Time (annual, long-term average, most recent years) and space (countries, regions, agroecologies, stage of development, endowment, etc.); and • Level of aggregation (plot, farm, household, subnational, national, regional, continental). The objective of this chapter is to assess changes over time and across dif- ferent parts of Africa, in both partial and total measures of agricultural pro- ductivity, to understand the relative sources of productivity growth. We begin the next section with a presentation of the partial and total measures 26 Chapter 2 of productivity, in addition to the data used in estimating the indicators and in conducting the analysis. This is followed by trends analysis of the indica- tors and key drivers, and then conclusions and implications for using different measures and indicators. Productivity Measures and Methodology Partial factor productivity is a ratio of output to a subset of the inputs, usu- ally one input, described as single-factor productivity. Two commonly used measures of PFP are land productivity (defined as the ratio of output to total harvested area) and labor productivity (the ratio of output to total number of hours worked). Obviously, these two PFP measures differ from one another by the variables they measure, as well as by the variables they exclude. PFP measures make it possible to focus on a given variable (that is, land or labor in the two examples above), to assess how that variable is changing relative to the output. Total factor productivity, conceptually also a measure of output to inputs, is commonly measured as an index of the ratio of total agricultural outputs to total agricultural inputs. As such, TFP analysis can be seen as an exten- sion of PFP analysis, since the variables used in measuring PFP are included in the variables used in measuring TFP. Use of TFP is favored in the analysis of productivity, because long-run agricultural growth depends on TFP growth, which can be decomposed into finer measures, including the three that are commonly estimated or presented in the literature: technical change, arising from movement of the technological frontier; technical-efficiency change, aris- ing from movement of observations toward or away from the technological frontier; and scale-efficiency change, arising from movement of observations about the technological frontier to capture economies of scale.1 In principle, measuring PFP is straightforward, and the data requirements are not complicated. Measuring TFP, however, can be challenging, especially for developing countries that lack data on prices to use in aggregating outputs and inputs. Several methods for and approaches to measuring TFP are avail- able, differing mainly in how outputs and inputs are aggregated. The methods can be classified into two broad groups: (1) nonparametric methods, including 1 Some studies have tried to decompose finer measures of efficiency change, distinguishing, for example, allocative efficiency change for inputs, allocative efficiency change for outputs, residual scale-efficiency change, etc. (for example, Rahman and Salim 2013). Details of the TFP decomposition in general, as well as the changes analyzed in the study (that is, technical- efficiency change and technical change), are presented in the appendix to this chapter. Intertemporal trends In agrICultural produCtIvIty 27 index-based growth accounting (for example, the Törnqvist-Theil index) and data envelopment analysis (DEA); and (2) parametric methods, including econometric estimation of the technology, often by stochastic frontier analy- sis. (See, for example, Coelli, Prasada Rao, and Battese [1998] and Coelli and Prasada Rao [2001] for review of the different methods and measurement issues.) This study uses the Malmquist index approach, where the index is cal- culated by DEA, using one output and six inputs, and assuming sequential technology that rules out technological regression or negative growth rates in technical change. This method, referred to as the DEA-Malmquist index, is fully documented in Nin-Pratt and Yu (2010), so we present the main aspects of it in Appendix 2A of this chapter. The literature has identified some issues with use of DEA methods to cal- culate distance functions. The major drawbacks include nonstochastic func- tions— that is, lack of including a random-error term to account for statistical noise; determination of implicit or shadow prices used in aggregating inputs; and dimensionality, or the number of inputs and outputs used relative to the number of observations in the cross-section. These issues and how they are dealt with in this study are also discussed in Appendix 2A. To generate greater confidence in the findings associated with this method, however, we compare the results with those obtained using three other approaches that differently address the issues with DEA. These include two other versions of the DEA-Malmquist index: one is calculated by using two outputs to deal with the dimensionality issue, and the other is calculated by including lower and upper bounds on the shadow prices. The third method is the more conventional growth-accounting TFP index, where inputs are aggre- gated using fixed-input shares for all countries and periods. A brief compari- son of the results is presented in Appendix 2B. Overall, the different methods yield similar TFP growth patterns, with the DEA-Malmquist−2-output-index giving higher growth rates, followed by the DEA-Malmquist−1-output-index, the DEA-Malmquist-bounds-index, and the growth-accounting-TFP-index. Data and Sources of Data The data used in the measurements of the different PFP and TFP indica- tors are drawn mostly from the Food and Agriculture Organization of the United Nations FAOSTAT database on agricultural production (FAO 2014), which covers the period 1961– 2012. The data, which are detailed in Table 2.1, include one output (total agricultural production) and six inputs (land, labor, fertilizer, animal feed, crop capital, and livestock capital). The two PFP 28 Chapter 2 TAbLE 2.1 Description of variables and data used in estimating partial and total factor productivity Variable Description output value of gross crop and livestock production expressed in constant 2004–2006 interna- tional dollars. In the case of nigeria, output for 2001–2012 was adjusted using agricultural value-added data from the World development Indicators (World Bank 2014) to better reflect recent growth measured at the country level. land hectares of land, including land under temporary crops (doubled-cropped areas are counted only once); temporary meadows for mowing or pasture, such as land used permanently (five years or more) for herbaceous forage crops, either cultivated or growing wild (wild prairie or grazing land); land under market and kitchen gardens; land temporarily under fallow (less than five years); land cultivated with permanent crops, such as flowering shrubs (coffee), fruit trees, nut trees, and vines, but excluding land under trees grown for wood or timber. labor total economically active population engaged in or seeking work in agriculture, hunting, fishing, or forestry, whether as employers, their own account workers, salaried employees, or unpaid workers assisting in the operation of a family farm or business. this is an uncorrected measure of labor that does not account for actual hours worked or labor quality (education, age, experience, etc.). data for nigeria were adjusted following Fuglie (2011). Fertilizer metric tons of nitrogen, phosphorus, and potassium nutrients consumed. animal feed metric tons (maize equivalent) of edible commodities (cereals, bran, oilseeds, oilcakes, fruits, vegetables, roots and tubers, pulses, molasses, animal fat, fish, meat meal, whey, milk, and other animal products) fed to livestock. Crop capital sum of gross fixed capital stock in constant 2005 us$: • Land development: major improvements in the quantity, quality, or productivity of land to prevent its deterioration, including (1) on-field land improvement undertaken by farmers (includes work done on the field, such as making boundaries and irrigation channels); and (2) other activities, such as irrigation works, soil conse