AGRICULTURE GLOBAL PRACTICE DISCUSSION PAPER 01 AGRIBUSINESS INDICATORS: SYNTHESIS REPORT WORLD BANK GROUP REPORT NUMBER 91133-AFR DECEMBER 2014 AGRICULTURE GLOBAL PRACTICE DISCUSSION PAPER 01 AGRIBUSINESS INDICATORS: SYNTHESIS REPORT © 2014 World Bank Group 1818 H Street NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Email: feedback@worldbank.org All rights reserved This volume is a product of the staff of the World Bank Group. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of World Bank Group or the governments they represent. The World Bank Group does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of World Bank Group concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. World Bank Group encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone: 978-750-8400, fax: 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank Group, 1818 H Street NW, Washington, DC 20433, USA, fax: 202-522-2422, e-mail: pubrights@worldbank.org. Cover Photo: Courtesy of Jonathan Ernst/World Bank, overlooking the central Kumasi market, Ghana. CONTENTS Foreword v Acknowledgments vii List of Abbreviations ix Executive Summary xi Chapter One: Introduction 1 Chapter Two: Methodology 3 Chapter Three: Policy Environment 9 Chapter Four: Seed 13 Chapter Five: Fertilizer 19 Chapter Six: Farm Mechanization 25 Chapter Seven: Agricultural Finance 31 Chapter Eight: Rural Transport 39 Chapter Nine: Summary and Conclusion Country Results 45 Chapter Ten: Conclusions and Discussion 59 Appendix A: Selected Comparative Country Statistics Related to the Agriculture Sector 61 Appendix B: ABI Indicators Used in Initial African Country Studies, 2011–12 63 Bibliography 77 BOXES Box 2.1. A Note on Triangulation 6 Box 4.1. Example of Successful Reforms: Turkey 17 Box 6.1. Ghana’s Mechanization Scheme: An Example of Public Mechanization Program 30 Box 7.1. Establishment of a Credit Reference Bureau in Ghana 37 Box 8.1. Registration as a Trucking Association in Ethiopia 40 FIGURES Figure S.1. Agricultural Transformation and Nine ABI Pilot Countries xiii Figure 4.1. Seed Imports in Thousands of Metric Tons, 2011–12 15 Figure 4.2. Seed-to-Grain Price Ratio in the Pilot Countries, 2011–12 Cropping Seasons 15 Figure 4.3. Number of Seed Companies and Percent of Certified Seeds Supplied by the Private Sector 16 Figure 5.1. Nutrient/Output Price Ratio 21 Figure 5.2. Average Retail Price of Urea in Rural Areas (US$ per Metric Ton), 2011–12 22 Figure 5.3. Agro-Input Dealer Density: Agro-Input Dealers per 10,000 Farmers 23 Agribusiness Indicators: Synthesis Report iii Figure 6.1. Farm Power Sources (percentages) in Sub-Saharan Africa, Asia, and Latin America, 2006 26 Figure 6.2. Tractors per 100 km2 in Selected Countries 26 Figure 6.3. Average Horsepower per Hectare of Arable Land 28 Figure 6.4. Percentage (%) of Tractors Imported by the Private Sector, 2011–12 29 Figure 7.1. Proportion of Commercial Bank Lending to Agriculture 34 Figure 7.2. Bank Branches per 100,000 of the Rural Adult Population 35 Figure 7.3. Average Interest Rates on Lending for Agriculture through Commercial Banks in 2010 36 Figure 7.4. Lending-Deposit Spread 36 Figure 8.1. Percentage of the Road Classified as Being in Poor Condition 42 Figure 8.2. Figure Comparisons across Countries: Quality of Infrastructure (Scale of 0 to 5) 43 TABLES Table S.1. Key Agribusiness Indicators Used: Farming Sector xiv Table S.2. Key Agribusiness Indicators Used: Private Sector xv Table S.3. Key Agribusiness Indicators Used: Policy Environment xvi Table S.4. Agribusiness Indicators: Farming Sector xvii Table S.5. Agribusiness Indicators: Private Sector Perception xx Table S.6. Agribusiness Indicators: Policy Environment xx Table 2.1. Sources for Indicators 5 Table 2.2. Type of Data 7 Table 3.1. Enabling and Policy Environment Indicators 11 Table 4.1. Summary of Seed Indicators 14 Table B4.1.1. Effects of Regulatory Reforms on the Introduction of New Varieties in Turkey 17 Table 5.1. Summary of Fertilizer Indicators 20 Table 6.1. Summary of Farm Mechanization Indicators 27 Table 6.2. Estimate of Total Tractor Horsepower (HP) per 100 Square Kilometer in Kenya, 2011 27 Table 6.3. Average Cost of Hiring Tractors for Farm Operations in Kenya, 2011–12 28 Table 6.4. National Strategies for Agricultural Mechanization 30 Table 7.1. Financial Landscape of the ABI Pilot Countries 33 Table 7.2. Rural Finance Indicators 33 Table 8.1. Rural Transport Indicators 41 iv Agriculture Global Practice Discussion Paper FOREWORD Ethel Sennhauser Director Agriculture Global Practice The World Bank African agriculture is undergoing an unprecedented period of change. Across Sub- Saharan Africa, economic growth rates are up, projected to strengthen to 5.2 percent during 2015–16, up from 4.6 percent in 2014 and rising to 5.3 percent in 2017. Rapid urbanization, and spiraling demand for food represent an economic opportunity—the value of Africa’s food and beverage markets are expected to top $1 trillion by 2030, up from only $313 billion today. Urban demand for food products is projected to increase at a compound annual growth rate of around 4 percent. In lock step, con- sumer demand is also changing, and new marketing channels are emerging. Strong growth opportunities exist for agribusiness provided the farming sector can deliver agricultural and food products at competitive prices. These projected changes bring with them risks and opportunities both for farmers and the larger agricultural sector. In particular these changes require policy makers and planners to think differently: to factor these changes into their planning for new policies, regulations, and public investments, while paying attention to the critical need for improving the enabling environment so that agriculture and agribusinesses can thrive. This report covers nine Sub-Saharan countries and aims to provide policy makers with the tools and data with which they can better understand how their own country is faring in the market orientated agricultural sector while also learning from those countries which are performing well. The report looks at the policy environment as well as key sub-sectors such as seed, fer- tilizer, finance, mechanization, and transport. It brings into the public domain, often for the first time, key comparisons of costs, regulations, perceptions by the private sec- tor of Government policies, and differences in farmers’ access to inputs at the country level. We believe such information can help policy makers and planners make informed decisions about their policies, introduce key data for an improved policy dialogue and contribute to the larger aim of encouraging evidence-based policy making in Africa. Readers and users of the information are invited to look at the findings and use their judgment and knowledge of local conditions as an input for sound policy-making. We expect that this analytical work will have a role in creating the next generation of poli- cies that help to unleash the potential of the agriculture and agribusiness sectors and contribute to the twin goals of ending poverty and boosting shared prosperity. This report benefited from the support of the Bill and Melinda Gates Foundation. Agribusiness Indicators: Synthesis Report v ACKNOWLEDGMENTS The report was funded by the Bill and Melinda Gates Foundation (BMGF). However, the views of the authors do not necessarily reflect the opinions or strategic program thrusts of the BMGF. The grant money was generously given to the World Bank (WB), Agriculture and Environmental Services Department (AES) to conduct this initial pilot study of agribusiness indicators in Sub-Saharan Africa. The BMGF Agriculture Department staff members, in particular, Prabhu Pingali, Chris Gingerich, Mumukshu Patel, Alan Rennison, and Greg Traxler have provided guidance throughout the grant implemen- tation, as well as incisive comments on the outputs of the AES/Agribusiness Indicators (ABI) team. At the World Bank, the fieldwork, country studies, and synthesis report have been a collaborative effort involving qualified agriculture specialists and editors. The team was composed of the following people: Grahame Dixie, Task Manager and Agribusi- ness advisor to AES; John Holtzman, program coordinator for ABI and principal investigator on the Mozambique, Burkina Faso, and Rwanda country studies; James Mbata, senior agricultural economist and principal investigator on the Ethiopia, Kenya, and Nigeria country studies; Samjhana Thapa, economist and principal inves- tigator on the Ghana, Tanzania, and Zambia country studies; Raissa Adomayakpor, research assistant and analyst; Damian Milverton, editor; and Alina Alvarezm, graph- ics designer. The document was edited by Gunnar Larson, Patricia Van de Velde, Chaitri N. Hapugalle, and Christopher Dielmann. Agribusiness Indicators: Synthesis Report vii LIST OF ABBREVIATIONS ABI Agribusiness Indicators ha hectare ABS African Seed and Biotechnology Program IFDC International Fertilizer Development Center ADC Agricultural Development Corporation IFPRI International Food Policy Research Institute ADP Agricultural Development Project ISTA International Seed Testing Association AES Agriculture and Environmental Services KNFAP Kenya National Federation of Agricultural Department, World Bank Producers AISE Agricultural Input Supply Enterprise, Ethiopia KSC Kenya Seed Company AMSEC Agriculture Mechanization Services KTZ Kenya, Tanzania, Zambia Enterprises Center LICs Low-Income Countries AU African Union LPI Logistic Performance Index BBA Benchmarking Business of Agriculture MAFAP Monitoring African Food and Agricultural BMGF Bill and Melinda Gates Foundation Policies CAADP Comprehensive Africa Agriculture MFI Microfinance Institution Development Programme MICs Middle-Income Countries CAGR Compound Annual Growth Rate MINAGRI Ministry of Agriculture, Rwanda CIF Cost Insurance and Freight mmt Million Metric Tons CILSS Permanent Interstate Committee for MoA Ministry of Agriculture Drought Control in the Sahel MoFA Ministry of Food and Agriculture CNFA Citizens Network for Foreign Affairs mt Metric Tons COCOBOD Ghana Cocoa Board NEPAD New Partnership for Africa’s Development COFOG Classification of Functions of Government NGO Nongovernmental Organization COMTRADE Commodity Trade Statistics Database (of NPK Nitrogen Phosphate Potassium Fertilizer the United Nations) NPLs Nonperforming Loans CSO Civil Society Organization OECD Organisation for Economic Co-operation CTA Confederação das Associações Económicas and Development de Moçambique PER Public Expenditure Review ECE Ethiopian Seed Enterprise RAI Rural Access Index ECOWAS Economic Community of the West African REC Regional Economic Community States RTA Ethiopian Road Transport Authority ECX Ethiopian Commodity Exchange SADC Southern African Development EE Enabling Environment Community EU European Union SSA Sub-Saharan Africa FAO Food and Agriculture Organization UEMOA Union Economique et Monétaire Ouest FAST Forum for African Seed Testing Africaine FO Farmer Organization UNPCB Union des Producteurs du Coton Burkinabè GDP Gross Domestic Product USAID U.S. Agency for International GGC Ghana Grains Council Development GIS Geographical Information System VAT Value Added Tax GLSS 5 Ghana Living Standard and Measurement WB World Bank Survey, Fifth Round, 2008 WBG World Bank Group GoE Government of Ethiopia WRS Warehouse Receipt System GoR Government of Rwanda ZANFU Zambian National Farmers’ Union All dollar amounts are U.S. dollars unless otherwise indicated. Agribusiness Indicators: Synthesis Report ix EXECUTIVE SUMMARY OBJECTIVE OF THE AGRIBUSINESS INDICATORS PROJECT The need for countries in Sub-Saharan Africa to build more productive, modern, and market-oriented farming sectors is one of our most pressing development challenges. In coming years, African agriculture will have to increase food production and expand and intensify value chains in order to meet changing demand on the part of a rap- idly expanding and urbanizing consumer base. The process of doing this will enable African countries to begin pushing back against their currently growing reliance on food imports. An essential precondition for bringing this transformation to pass is to increase and improve the information on which farmers and agribusinesses base their production and investment decisions, and on which public sector institutions base their policies. The purpose of the Agribusiness Indicators (ABIs) Project is to provide this kind of empirical information in the form of a series of metrics and indicators that can be used to measure change over time and to make direct comparisons between coun- tries, especially policy makers. These indicators will be used to inform policy dialogue, including dialogue between representatives of the private and public sectors. It will provide a common framework of reference with which to communicate their respec- tive concerns, priorities, and intentions. This will facilitate better communication that leads to constructive interaction between public officials, farmers’ producer organiza- tions, private investors, civil society organizations (CSOs), and others. Ultimately, it will be their decisions that determine the course of agricultural development and com- mercialization in their respective countries. The Agribusiness Indicators are intended to furnish them with information from sources within both the private and public sectors which can be cross-referenced and correlated. This type of information has generally not been available in the past. Some of the indicators are particularly useful in revealing the attributes of countries with policy portfolios that are supportive of agribusiness investment. Agribusiness Indicators: Synthesis Report xi AGRIBUSINESS INDICATORS AS A PILOT agriculture”: agriculture-based, transforming, and urban- OF A LARGER GLOBAL-LEVEL PROGRAM ized countries. This classification is based on two factors: This ABI report is the first pilot phase of a larger program the proportion of a country’s total gross domestic product called Benchmarking the Business of Agriculture (BBA). (GDP) that comes from the agriculture sector, and the pro- Building on the lessons of ABI and of the World Bank’s portion of a country’s workforce that is employed in the Doing Business report, some of the comparisons of the nine agriculture sector. As manufacturing and service sectors of original pilot countries are planned to be extended to 80 the economy grow, agriculture accounts for a diminishing countries in all regions of the world and at all stages of proportion of employment and GDP—a process known agricultural transition. This more extensive sample will to development economists as the agricultural transformation. provide an ongoing reporting process that will enable This depiction of where countries are situated along the countries to measure, compare, and contrast performance agriculture-based to urbanized continuum is illustrated over time, and to correlate this performance with different in the figure S.1, in which the nine ABI pilot countries policies and policy environments. are represented as triangles. The squares represent other countries in Sub-Saharan Africa, and the diamonds rep- resent countries in regions elsewhere in the world. For AREAS OF FOCUS the purposes of the ABI studies and the larger Bench- The ABI country studies synthesized in this report ana- marking the Business of Agriculture (BBA) program, lyze six areas that relate to the state of development of the three worlds of agriculture were further divided into agribusiness in a given country: policies, seed and fertilizer two successive phases of transition and into urbanizing inputs, mechanization, finance, infrastructure, and trans- and developed economies—a set of distinctions that will port services. The report devotes one chapter to each, assume greater significance as the BBA expands into focusing in particular on small and medium size farms additional developing regions. and the network of mostly small and medium enterprises that are necessary to support the emergence of a more modern farming sector. STUDY APPROACH The ABI country study teams visited each country two In the executive summary the issues have been clustered dif- or three times, spending between four and eight weeks in ferently, covering the perspectives of the farming sector and each. Local consultants were employed to help the teams the agribusiness community regarding the policy environ- identify important sources of information, including ment. The key indicators used in this report are set out in existing surveys, research papers, and reports. Data on 54 table S.1, along with their mean, median, and high and low indicators were collected, 44 of which are presented in scores. These are intended to reflect the findings from the this report. The local consultants also assisted the study field, the key issue areas that the study has identified, and teams in preparing meetings and interviews with respon- the quantified measures used. dents representing public officials, private sector investors, and farmers’ organizations. Their perspectives about the COUNTRY FOCUS factors that actively restrict agricultural modernization The countries selected for this study were Burkina Faso, and commercialization were useful in enabling the study Ethiopia, Ghana, Kenya, Mozambique Nigeria, Rwanda, teams to compose summaries of the different investment Tanzania, and Zambia. The individual country studies climates in the nine countries. of these nine countries were carried out and published in 2012 and 2013. The country studies sought to identify and Because the nine country study reports that served as measure the root causes to which the problems being expe- background material for this synthesis were peer reviewed rienced within the respective countries can be attributed. by individuals with extensive experience in the respective countries, the data collected by monitoring the indicators The 2008 World Development Report, Agriculture for will become more reliable and refined over time. Much of Development, classified countries into “three worlds of the information presented here has never been chronicled xii Agriculture Global Practice Discussion Paper FIGURE S.1. AGRICULTURAL TRANSFORMATION AND NINE ABI PILOT COUNTRIES Rest of world SSA ABI 70 Agricultural sector value-added as % of total GDP Agriculture 60 Transitioning 50 Urbanizing 40 30 Developed 20 10 Pretransition 0 0 10 20 30 40 50 60 70 80 90 100 Agricultural employment as % of total employment Source: ABI/BBA analysis based on Food and Agriculture Organization (FAO) and World Bank (WB) data. before and may therefore be time bound, yielding insights in a timely fashion. This will, in turn, help generate greater into the initial conditions covered by the studies but not marketable surpluses that lead to improved incomes, some necessarily representing trends and developments over of which will be reinvested in agriculture-related activities time. For instance, data on fertilizer prices were gathered both on and off the farm. during a two-year period during which international fer- tilizer prices greatly fluctuated. How representative this Access to affordable seed, fertilizer, mechanical inputs, flux is will only become clear over time, which is of course and inputs supplied by the private sector are key factors one of the purposes of having indicators to monitor. The affecting farm productivity and profitability. Significant limited size of the sample makes it impossible to deter- differences are revealed between the nine countries. Retail mine how representative the study’s findings are, and there prices of unsubsidized fertilizer in the ABI countries vary are important concerns about how comparable scored by some 200 percent. Particularly in the rain-fed (nonirri- perception indicators are across countries. That said, the gated) farming systems that make up most African agricul- responses of investors and prospective investors indicating ture, fertilizer use tends to be extremely sensitive to price. how they perceived the business environment demonstrated A number of countries are moving toward allowing private substantial alignment with other, more objective measures sector participation in the multiplication of foundation of whether or not the private sector is seen as partner in seed, and this direction may lead to major breakthroughs economic development. in the countries where the public sector monopoly over foundation seed is correlated with outstanding seed scarcity. The emergence of new seed laws and regulations is impor- RATIONALE OF THE INDICATORS tant in creating a clear path toward a thriving seed industry. SELECTED In this regard, the performance of ABI countries has been Farming Sector Indicators (Table S.1). Typical cereal patchy and generally slow. yields by a country are given to provide context to the indi- cators. The first grouping of indicators covers the economic Africa lags behind the rest of the world in farm mechani- land use of inputs. It is vital for farmers’ productivity to zation. Mechanization is typically most closely associated have access to high-quality inputs and services provided with rising labor costs and rural labor shortages that are Agribusiness Indicators: Synthesis Report xiii TABLE S.1. KEY AGRIBUSINESS INDICATORS USED: FARMING SECTOR Farming Sector Indicators Mean Median High Low Input use and economics Average cereal yield (kg/ha) 1,651 1,660 2,693 694 Fertilizers Fertilizer nutrients (kg/ha) (Abuja target 50+) 20 17 60 3 Nutrient/output ratio (good < 5) 6 5 2 13 Seeds % Use of certified maize seeds (%) 31 19 70 6 Maize seed to grain price ratio (good < 5) 7 7 2 13 Tractors Tractor density (SSA average 13 per 100 km2) 11 9 27 1 Input sourcing and costs Fertilizers Average fertilizer (unsubsidized) rural retail price 721 760 475 1,023 US$/ton Fertilizer subsidy—Y/N? 7 Y, 2 N Y N Fertilizer import taxes (%) 1.8 0.0 0.0 8.5 Seeds Private sector production of foundation seed (%) 14.8 2.5 100.0 0.0 Imported seed as % total certified seed 59 54 100 12 Private sector production of certified seed (%) 16 15 56 0 Time (years) for seed testing and registration 2.7 2.0 2.0 5.0 Existence/implementation of national seed 2.9 2.5 4.9 2.0 regulations/laws (0–5) Tractors Tractors imported by private sector (%) 65 60 10 100 Tractor import tariffs as % CIF prices 2.3 0.0 0.0 16.0 Parts import tariffs as % CIF prices 16.4 16.0 0.0 27.0 Cost of tractor rental (pilot average $82) 82 68 46 163 Distribution Agrodealer network (number per 10,000 farms) 3.0 2.1 8.4 0.3 Financing Access Number of bank branches per 100,000 adult 1.9 1.4 5.0 0.8 population Percentage of commercial banks lending 7 8 14 2 to Ag (3 years) (%) Cost Average lending rates for Ag loans (real rates) 10 12 5 22 Percentage of nonperforming Ag loans 20 13 4 59 Road access Rural Access Index (%) 23 24 36.1 10.3 Quality Road network in poor condition (%) 34 34 8.7 69.0 Logistic Perfomance Index (LPI) infrastructure 2.10 2.16 2.41 1.53 quality (0–5) Cost Cost of transportation ($/mt/km)—main routes 0.10 0.10 0.05 0.15 Cost of transportation ($/mt/km)—secondary 0.26 0.24 0.13 0.47 routes Note: High scores are indicated by green and emboldened figures (that is, 14.5%) and low scores by pink and a black surround (that is, 2.0%). Source: ABI Country Reports, 2012. driven by migration to urban areas. However, the possibil- include (i) who imports tractors; (ii) import taxes on trac- ities of a sustainable private sector providing machinery tors and spare parts, which are essential to maximizing contracting services can be choked off by the unintended the useful life of the national tractor fleet; and (iii) cost consequence of public policy. Conversely, supportive poli- of hiring plowing services, which is a useful method for cies can help mechanization take root as has been the enabling smaller-scale producers to access the benefits example in China and Pakistan. The indicators used to of machinery without having to make these investments measure policy environment as it affects mechanization themselves. xiv Agriculture Global Practice Discussion Paper TABLE S.2. KEY AGRIBUSINESS INDICATORS USED: PRIVATE SECTOR Private Sector Indicators Mean Median Low High Perception: policy environment Enabling environment (0–5) 3.1 3.0 2.5 4.3 Policy consistency (0–5) 2.5 2.5 1.5 3.5 Private sector advocacy (0–5) 2.4 2.0 1.0 3.7 Perception: business environment Fertilizer business environment (0–5) 2.6 3.0 0.0 4.2 Ease of entry/operations in seeds (0–5) 2.9 2.5 1.5 4.9 Mechanization business environment (0–5) 2.8 2.8 1.0 5.0 Note: High scores are indicated by green and emboldened figures (that is, 14.5%) and low scores by pink and a black surround (that is, 2.0%). Source: ABI Country Reports. The ABI study used another metric to measure the den- The presence of fora in which representatives of the pub- sity of input dealers. These businesses are important not lic and private sectors are able to interact and consult one only in delivering inputs, but increasingly in providing another regarding policies and commercial strategies, technical advice. The number of agricultural input sup- and to express their concerns and priorities, appears to plying businesses was found to vary considerably between be strongly correlated to countries in which agricultural countries, by as much as 20-fold. Rural finance indicators performance is better than otherwise might be expected.2 reflect the relative importance of agricultural credit in the The perception indicators are given on a scale of 0 to 5, overall loan portfolio of commercial banks, the interest with 0 indicating very negative impressions and 5 indicat- rates charged to agricultural borrowers, and the density ing very positive impressions. of commercial banks in rural areas. Rural transport indi- cators cover the issues of access, quality of roads, and unit AGRICULTURAL POLICY ENVIRONMENT costs expressed as U.S. dollars per kilometer per metric (TABLE S.3) ton on both primary and secondary rural roads. The policy indicators include measures of public invest- ment in agriculture as well as levels of support in fertil- Private Sector (Table S.2). Public expenditure can izer, banking, and transport. The indicators include (i) provide only a small proportion of the investment needed the CAADP target of 10 percent of government spend- by agriculture. Progressive thinking, for example in the ing on agriculture, (ii) measures of the proportionality of Comprehensive Africa Agriculture Development Pro- government agricultural spending in relation to the sec- gramme (CAADP), focuses on how to leverage greater tor’s overall importance in the economy, and (iii) whether private sector investment, both by producers and by the expenditure on research and development exceeds a tar- business networks supporting them. The report aims to pro- get of 1 percent of agricultural GDP. vide governments with insights into how the private sector perceives the business environment generally, as well as how Indicators are generated on the relative importance of that environment affects farm production’s most important fertilizer subsidies; country membership of international inputs: seed, fertilizer, mechanization, and finance.1 seed agreements; and the presence of instruments that facilitate agricultural lending, such as credit reference Inconsistent and unpredictable public policy is an anathema agencies and warehouse credits. Market access is closely to private investment at all levels. It introduces an element of related to the availability, cost, and travel time using the often arbitrary risk that discourages investments and causes rural road network. A perception indicator measures the delayed investment decisions on the part of producers, small ease of entry into the trucking business in terms of costs, and medium enterprises, and multinational corporations. waiting periods, corruption, and so on. 1 The perception/opinion indicators are correlated with each other. Perception 2 International examples include the Netherlands, Denmark, and Chile, while is based in part on reality that can influence the private sector behavior. These Wiggins and Keats come to similar conclusions on the basis of their analysis of findings should be treated as indicative and illustrative, as they are based on experiences in African countries, see “Leaping & Learning: Linking Smallhold- perceptions of a small sample of private firms. ers to Markets” (Wiggins and Keats 2013). Agribusiness Indicators: Synthesis Report xv TABLE S.3. KEY AGRIBUSINESS INDICATORS USED: POLICY ENVIRONMENT Policy Environment Indicators Mean Median Low High Investment in agriculture Ag spending as percentage of budget—current 7.9 6.5 2.0 14.5 Trend in Ag spending budget—2001–10 3≠, 3 Flat, 3 Ø N/A N/A Ag public expenditure ratio to Ag GDP 0.26 0.30 0.06 0.43 Ag R&D spending as percentage Ag GDP, 0.55 0.43 0.26 1.53 where when 1, Ag PE% = Ag GDP% Input support Fertilizer subsidy as percentage of 20.5 18.5 4.0 38.1 Ag Ministry budget ISTA membership or adherence to OECD 2 Yes, 7 No N/A N/A requirements Banking support Evidence of credit reference (0–5) 1.6 1.5 0.0 3.0 Presence of collateral registry for loans 2 Yes, 7 No N/A N/A Existence of warehouse receipts (0–5) 1.4 1.0 0.0 4.0 Transportation support Time (days) to get truck registered and licensed 13 7 5 29 Ease of entry into trucking (0–5) 3.7 4.0 4.0 2.5 Note: High scores are indicated by green and emboldened figures (that is, 14.5%) and low scores by pink and a black surround (that is, 2.0%). ISTA = International Seed Testing Association; OECD = Organisation for Economic Co-operation and Development. Source: ABI Country Reports. FARMER ACCESS TO INPUTS African agriculture is highly dependent on animal and AND SERVICES INDICATOR human power, which provides 25 and 65 percent of power RESULTS (TABLE S.4) in farming, respectively. Countries in the rest of the world The nine ABI pilot countries exhibit major differences in are much more reliant on mechanical power in farming. farmers’ access to and use of inputs, though in general Countries like Brazil and Tunisia have around 10 times their use of fertilizer and certified seed is low. The level the tractor density of the ABI average. The level of trac- of mechanization is extremely low (often about 1/10) torization is affected by regulations and policies, and it is compared to that of comparator countries. Input-output more directly influenced by the cost of labor. International ratios of seed and fertilizer are mixed and often depend experience has shown that mechanization does not replace on the presence of subsidies to be financially favorable labor but rather substitutes for rural labor shortages, and enough to encourage farmers to use these inputs. its success is largely a response to rural labor rates rising. Kenya and Zambia have the highest level of fertilizer use, The price of fertilizer varies widely between the ABI although only Kenya exceeds the Abuja Declaration tar- pilot countries. Mozambican growers have to pay over get of 50 kg of nutrient equivalents per hectare. Kenya $1,000 per metric ton for urea. In contrast, Ghanaian also outperforms other countries in terms of using certi- growers pay about half that price when subsidies are fied maize seeds and the density of tractors. Mozambique removed. Most countries have introduced fertilizer sub- and Nigeria are the weakest performers in both seed and sidies and have kept import tariffs to zero, although other fertilizer. Tanzania is the ABI country that is most reliant miscellaneous taxes are often charged. The most recent on imported seed. and comprehensive study of agricultural subsidies in Africa characterizes fertilizer subsidies as a mixed bless- Major differences appear between the nine countries in ing at best.3 Their introduction leads to quick and highly their use of on-farm mechanization. For instance, there is a 17-fold difference in tractor density between Rwanda 3 Jayne, T.S., Shahidur Rashid. 2013. “Input Subsidy Programs in Sub-Saharan and Zambia. More generally, the contrast between Sub- Africa: A Synthesis of Recent Evidence.” Agricultural Economics, Michigan State Saharan Africa and the rest of the world is significant. University, August. xvi Agriculture Global Practice Discussion Paper TABLE S.4. AGRIBUSINESS INDICATORS: FARMING SECTOR Farming Burkina Sector Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Input use and Average cereal yield 1,651 1,230 2,169 694 1,970 1,314 1,363 1,660 2,693 1,768 economics (kg/ha) Fertilizers Fertilizer nutrients (kg/ha) 20 10.1 18.3 2.6 17.0 10.1 6.5 59.6 37.0 19.8 (Abuja target 50+) Nutrient/output ratio (good 6 5.1 3.4 11.0 2.4 6.7 3.5 8.0 12.7 2.6 < 5) Seeds Use of certified maize 31 50 12 6 18 27 7 70 65 19 Agribusiness Indicators: Synthesis Report seeds (%) Hybrid maize seed to grain 7 13.0 NA 10.0 7.0 11.0 4.0 5.0 10.0 2.0 price ratio (best between 2 and 5) Tractors Tractor density (SSA 11 8.9 1.3 12.7 4.0 7.4 5.7 27.0 23.0 11.0 average 13 per 100 km2) Input sourcing and costs Fertilizers Average fertilizer 721 808 800 1,023 614 682 809 514 760 475 (unsubsidized) rural retail price US$/ton Fertilizer subsidy—Y/N? 7 Y, Y Y N N Y Y Y Y Y 2N Fertilizer import taxes (%) 1.8 8.5 0 2.5 0 0 5 0 0 0 Seeds Private sector production of 14.8 0 0 13 5 NA 0 0 100 0 foundation seed (%) Private sector production of 66 100 NA 15 12 79 54 10 100 100 certified seed (%) Imported seed as % total 14 NA NA 15 1 56 15 15 1 7 certified seed Time (years) for seed testing 2.7 5.0 3.0 3.0 2.0 2.0 2.0 3.0 2.0 2.0 and registration Existence/implementation 2.9 2.5 2.0 2.0 2.0 4.0 2.0 3.5 4.5 3.0 of national seed regulations/laws (0–5) (Continued ) xvii TABLE S.4. AGRIBUSINESS INDICATORS: FARMING SECTOR (CONTINUED) xviii Farming Burkina Sector Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Tractors Tractors imported by 65 59 42 60 100 10 10 100 100 100 private sector (%) Tractor import tariffs as % 2.3 16 0 5 0 0 0 0 0 0 CIF prices Parts import tariffs as % 16.4 20 30 25 0 10 5 16 15 27 CIF price Cost of tractor rental (pilot 82 51 163 62 81 68 93 50 125 46 average $82) Distribution Agrodealer network 3.0 2.1 1.8 0.3 2.2 1.3 2.1 5.8 NA 8.4 (number per 10,000 farms) Financing Access Number of bank branches 1.9 1.2 NA 1.6 0.8 2.0 1.3 1.4 NA 5.0 per 100,000 adult population Percentage of commercial 7 9 3 6 11 14 2 6 9 5 banks lending to Ag (3 years) (%) Cost Average lending rates for Ag 11 10 12 19 9 5 15 8 17 22 loans (real rates) Percentage of 20 NA 16 NA NA 58 6 9 10 21 nonperforming Ag loans Road access Rural Access Index (%) 23 24 36 24 10 24 20 32 17 24 Quality Road network in poor 34 9 69 32 34 29 34 34 48 18 condition (%) Logistic Performance Index 2.14 2.40 1.88 2.04 2.22 2.41 2.27 2.16 1.83 2.05 (LPI) infrastructure quality (0–5) Cost Cost of transportation 0.10 0.05 0.09 0.10 0.11 0.10 0.06 0.15 0.11 0.10 ($/MT/KM)—main routes Cost of transportation 0.26 0.14 0.37 0.23 0.24 0.13 0.47 0.24 0.13 0.35 ($/MT/KM)— secondary routes Agriculture Global Practice Discussion Paper Note: High scores are indicated by green and emboldened figures (that is, 14.5%) and low scores by pink and a black surround (that is, 2.0%). Source: ABI Country Reports. visible results as farmers respond. However, their benefits have the highest density of input suppliers, have the high- in terms of yield increases wane over time, they are costly, est levels of fertilizer use, and are among those with the and once introduced, governments find it politically very highest rates of certified seed use. Both countries took a difficult to discontinue them. proactive approach to encouraging the emergence of a knowledgeable private import supplier sector. In the seed sector there are major differences between countries in terms of the level of involvement by the gov- In Zambia, Ghana, Rwanda, and Burkina Faso, the ernment or private sector. The public sector dominates percentage of commercial bank lending made to the agri- foundation seed multiplication in all of the countries except culture sector exceeds 9 percent. The lowest agricultural Zambia. This can lead to serious difficulties because the lending is reported in Nigeria and Kenya. The highest quantities, qualities, and varieties being propagated may real interest rates on agricultural loans are in Ghana and not be those that farmers prefer. Furthermore, there Mozambique. Interest rates on agricultural loans gen- is no market mechanism in place to align supply with erally exceed those to urban enterprises for a variety of demand—often changing demand. The private sector reasons that will become clear throughout this report. dominates the production of certified seed in Ghana, Zam- bia, Burkina Faso, and Tanzania. Tanzania is the only Burkina Faso appears to be particularly well served by its country in the sample where imported seed plays a signifi- rural transport infrastructure, including road quality and cant role in the national seed market. Like Zambia, which logistics, and in terms of the unit costs of moving farm is a major seed exporter, Tanzania has modern seed laws inputs and outputs along primary and secondary roads. and regulations. Nigeria, Ethiopia, Rwanda, and Mozam- bique are perceived as having seed laws that are either Private Sector and the Business Environment outdated or underdeveloped. (Table S.5). The agribusiness sector had the most positive perspective of the agribusiness enabling environment in In Kenya, Zambia, Ghana, and Ethiopia, tractor impor- Zambia, followed by Tanzania, Rwanda, Burkina Faso, and tation is exclusively a private sector activity, while in Ghana. The most negative views were expressed in Nigeria Tanzania and Nigeria it is dominated by the public sector. and Ethiopia. The consistency of policy in particular correlates With the exception of Burkina Faso, none of the ABI coun- to the willingness of both agribusiness and farmers to make tries charge significant import tariffs on tractors. However, long-term investments. Rwanda and Kenya both scored the nearly all countries, with the exception of Ethiopia, charge highest, followed closely by Zambia in policy consistency. tariffs on tractor parts, such as air filters and tires. These Burkina Faso was viewed as having an especially volatile are key requirements for maintaining and maximizing policy environment. Zambia, Kenya, and Tanzania were the useful life of the national tractor fleet. Tariffs on trac- rated highest on quality and effectiveness of the interaction tor parts exceed 20 percent in Ghana, Mozambique, and between the public and private agribusiness sector. Private Rwanda. Contract machinery services are a promising sector perception of the business environment in fertilizer, means by which to extend the benefits of mechanization seed, and mechanization was seen as especially positive in to smaller-scale producers. Burkina Faso, Ghana, and Zambia, followed by Kenya. The positive perception of the Kenya have plowing hiring rates of $50 per hectare or farm machinery sector in Ethiopia reflects the strong role less. Rwanda has the highest rates at $163 per hectare, the private sector plays in importing tractors as well as the as well as one of the lowest tractor density rates in the absence of tariffs on imported tractor parts. study. Government attempts to operate these services at below cost have not only largely failed but have generally Government Policies (Table S.6). Three ABI pilot choked off private sector participation in the delivery of countries have achieved the CAADP target of spend- these services. ing 10 percent or more of the government’s budget on agriculture: Burkina Faso, Rwanda, and Ethiopia. Ghana Significant differences exist in the number of input suppli- is moving toward achieving this target, as well. Public ers in relation to the number of farms. Ghana and Kenya spending on agriculture in Burkina Faso, Rwanda, Ghana, Agribusiness Indicators: Synthesis Report xix xx TABLE S.5. AGRIBUSINESS INDICATORS: PRIVATE SECTOR PERCEPTION Private Burkina Sector Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Perception: Enabling environment (0–5) 3.1 3.0 3.0 2.8 2.5 3.6 2.5 3.0 4.3 2.9 policy environment Policy consistency (0–5) 2.5 1.5 3.5 2.0 2.0 2.5 2.0 3.5 3.0 2.6 Private sector advocacy (0–5) 2.4 2.0 3.0 1.5 1.3 3.3 2.0 3.5 3.7 1.0 Perception: Fertilizer business environment (0–5) 2.6 2.5 2.0 3.0 0.0 4.0 1.0 4.0 4.2 3.0 business Ease of entry/operations in seeds (0–5) 2.7 3.5 2.0 2.0 1.5 3.7 2.0 2.5 4.9 2.5 environment Mechanization business environment (0–5) 2.8 1.0 2.0 3.0 5.0 3.6 1.0 5.0 2.8 1.9 Note: High scores are indicated by green and emboldened figures (that is, 14.5%) and low scores by pink and a black surround (that is, 2.0%). Source: ABI Country Reports. TABLE S.6. AGRIBUSINESS INDICATORS: POLICY ENVIRONMENT Policy Burkina Environment Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Investment in Ag spending as percentage of 7.9 14.5 10.2 5.1 13.8 6.5 2.0 4.3 6.0 9.0 agriculture budget—current Trend in Ag spending 3≠, 3Æ, 3Ø down same same up same down down up up budget—2001–10 Ag public expenditure ratio to 0.26 0.43 0.32 0.16 0.30 0.24 0.06 0.15 0.30 0.36 Ag GDP Ag R&D spending as percentage 0.55 0.36 0.57 0.37 0.26 0.53 0.43 1.53 0.29 0.60 of Ag GDP Input support Fertilizer subsidy as percentage of 20.5 4.0 30.0 NA NA 18.5 NA NA 38.1 12.0 Ag Ministry budget ISTA membership or adherence 2Y, 5N N N N N N N Y Y N to OECD requirements— Y/N? Banking support Evidence of credit reference (0–5) 1.6 0.0 1.5 1.5 0.0 1.5 2.0 3.0 3.0 2.0 Presence of collateral registry 2Y, 5N N N N N N N Y N Y agency for loans Existence of warehouse receipts 1.4 1.5 0.0 0.0 3.0 4.0 0.0 0.0 1.0 3.0 (0–5) Transportation Time (days) to get truck 13 6 7 5 7 6 28 29 21 5 support registered/operating license Ease of entry into trucking (0–5) 3.7 3.5 4.0 4.0 4.0 3.6 3.5 2.5 4.0 4.0 Agriculture Global Practice Discussion Paper Note: High scores are indicated by green and emboldened figures (that is, 14.5%) and low scores by pink and a black surround (that is, 2.0%). Source: ABI Country Reports. Zambia, and Ethiopia more closely parallels the sector’s Zambia are both members of the OECD and ISTA inter- relative importance in the overall economy. Kenya is the national seed schemes in order to ensure that their seed only ABI country achieving the target of investing over industry is well harmonized with international standards. 1 percent of agricultural GDP in agricultural research In terms of the emergence of a more diverse range of and development. Seven countries operate fertilizer financial instruments, significant differences are observed. subsidy schemes. Zambia has the highest percentage of Both Kenya and Zambia operate credit reference agen- public agricultural expenditure dedicated to this one item, cies. Only Kenya and Ghana currently have a collateral followed by Rwanda. Of the four countries in which data registry agency for movable assets in place. Tanzania, could be collected, fertilizer subsidies alone accounted Ethiopia, and Ghana also have relatively advanced ware- for about one fifth of the agricultural budget. Kenya and house receipt systems in place. Agribusiness Indicators: Synthesis Report xxi CHAPTER ONE INTRODUCTION THE SECTORAL CONTEXT: THE POVERTY, GROWTH, AND AGRICULTURE NEXUS Agriculture is extremely important in the nine pilot countries. On average, the agriculture sector accounts for 65 percent of employment and 75 percent of domes- tic trade in Sub-Saharan Africa. For decades, investment in the sector was considered a strategic focus for spurring economic growth, generating employment, and reducing poverty. Despite efforts by governments, donors, the private sector, nongovernmental organizations (NGOs), and civil society, the pilot countries continue to confront poverty, malnutrition, and food insecurity—all of which are exacerbated in times of external and internal shocks such as droughts as well as food and fuel price spikes. Meeting the demands of burgeoning urban markets will require vastly improved food production systems. Research by the FAO projects that the increase in population and change in diets will lead to a 70 percent increase in global food demand by 2050. Urban populations in all nine ABI pilot countries are growing rapidly, comprising sizable middle-income and low-income segments. Like elsewhere in the developing world, an increasing proportion of food demand will be manifest in cities, and much of it will be supplied by smallholder, and in some instances medium- size, farmers (see figure S.1). With increasing proportions of food production serving urban consumers, addressing vulnerability to food insecurity in rural areas will remain a persistent priority for policy makers. Increased use of modern seed, fertilizers, and farm machines is instru- mental in bringing about increased agricultural yields. Together with access to financial and transport services, access to these inputs is a vital element of agricultural modernization. No region in the world has been able to expand agricultural growth rates and tackle hunger without improved seed, use of chemical fertilizers, and other modern inputs. Mechanization enables timely field preparation, precision planting, and fertilizer application which often assume greater importance among modern varieties. Farmers’ ability to access seed, fertilizer, pesticides, herbicides, and other inputs relies in large measure on their proximity to agro-input dealers. If these dealers are located Agribusiness Indicators: Synthesis Report 1 within reasonable distances that do not entail prohibitive and Zambia also have per capita incomes averaging over travel times, access is, of course, more likely. And where $400 per person per year. these dealerships operate within dense networks of com- peting input suppliers, the prices of these inputs tend to be Two of the countries, Ghana and Kenya, appear to be more competitive. Distance, however, is not the only con- in the second stage of development in which the econ- straint facing more remotely located farmers. A variety of omy is still largely agricultural but transitioning toward other policy-related, financial, institutional, and regulatory urbanization. Six of the countries fall in stage one of the factors can also militate against the adoption of modern agricultural development continuum, characterized by a inputs as well as including limited commitment of public low value of output per agricultural worker (<$300 per and private investment. Substantial impetus for increased worker in six countries in constant 2,000 US$ prices), a public and private investment is likely to result from gains very high proportion of the labor force in agriculture, in productivity that generate marketable surpluses and lead and agriculture contributing at least 30 percent of GDP. to higher incomes—surpluses and incomes that can be Nigeria is an outlier in that it has a large extractive sector reinvested in additional value-adding activities both on and and more industry than the other eight countries. (See off the farm. These in turn can be instrumental in estab- table S.1.) lishing forward and backward linkages that “fill out” and greatly strengthen agricultural value chains, creating new Nigeria also differs markedly from the other pilot coun- investment opportunities for agribusinesses which cumula- tries in that it has a large land area (second only to Ethio- tively lead to commercialization throughout the sector and pia) and the largest total area of agricultural land, a large beyond it in the greater rural economy. population (158.4 million in 2010), and high population density (174 people per square kilometer), trailing only CHARACTERISTICS OF THE ABI Rwanda (at 431 people per square kilometer). Nigeria PILOT COUNTRIES is the only very large country (on all measures) in the Even though the nine pilot ABI countries differ in terms sample. Five other countries (Ethiopia, Kenya, Mozam- of geography, climate, and socioeconomic development, bique, Tanzania, and Zambia) have large land areas but improving the competitiveness of agribusiness is a common vary significantly in other characteristics, such as popula- goal. Six of the nine countries—Burkina Faso, Ethiopia, tion, population density, urbanization, GDP per capita, Mozambique, Rwanda, Tanzania, and Zambia—are low- and arable land (cultivated in annual crops). On aver- income countries with large agricultural sectors, a large age, the agricultural sector’s contribution to GDP is 25 to rural population (five countries have ≥74 percent of popu- 35 percent in Sub-Saharan Africa, while it varies in pilot lation in rural areas), and a high proportion (≥70 percent) countries at about 20 percent in Zambia, and as high as of the economically active labor force employed in agri- 45 percent in Ethiopia. While the sector contributes about culture. Ghana and Nigeria are emerging middle-income 26 percent of GDP, in countries like Kenya, it also indi- countries with about half of their populations in urban rectly contributes to about 25 percent of GDP through areas and 4.1 percent urban population growth rates over the creation of backward and forward linkages in the food the period from 2000 to 2010. Nigeria, Kenya, Tanzania, manufacturing, distribution, and service industries. 2 Agriculture Global Practice Discussion Paper CHAPTER TWO METHODOLOGY An important part of the value of the agribusiness indicators is in the weight ascribed to the indicator and how the reader may understand what the indicated response answers and what it does not. In this chapter, we describe the creation of the indica- tors, the methodology, and the particularities and integrity of the responses. The pur- pose of this chapter is to enable readers to make more informed decisions on the value and weight of each indicator by explaining its origins, interpretation, and limitations. AGRIBUSINESS INDICATORS IN AFRICA (ABI): APPROACH, METHODS, INDICATORS The grant agreement called for pilot testing the ABI approach intensively in one country. The team did this in Ghana in fall 2010, focusing on two staple grain value chains—maize and rice. The fieldwork was highly instructive and helped the team narrow the range of (future) inquiry to a core set of cross-cutting constraints and issues. The team felt that the value chain work was not fully satisfactory. More time and effort would have been required to do comprehensive value chain analyses. The fieldwork was useful in identifying the following cross-cutting areas that shed light on the ease or difficulty of agriculture becoming more market forward: » Access to and availability of certified seed » Availability of and access to fertilizer » Access to farm machinery, particularly tractor hire services » Access to agricultural production and agro-enterprise finance » Cost and efficiency of rural transport, particularly trucking » Various policy measures (public expenditure on agriculture, percentage of a key export crop free on board [FOB] price captured by producers, role of pri- vate agribusiness advocacy groups, tariffs on imports of capital goods and spare parts for tractors) » Measures of consistency of the enabling environment for agribusiness INDICATOR DEVELOPMENT The key areas of investigation were fleshed out by developing a set of indicators for each group, defining each clearly, specifying how each indicator would be calcu- lated, identifying likely sources of information and how the team would collect such Agribusiness Indicators: Synthesis Report 3 information, and signaling what potential data collection available in readily usable form. In some cases, agricultural problems or issues might be encountered. The team devel- production figures are not released for years or are not oped an indicator matrix in early 2011. This was revised released at all if actual figures are well below targets. Crop and refined several times over the course of the project. production forecasts and estimates may also be subject to Technical specialists in the World Bank Group were con- manipulation, resulting in upward “adjustments.” Linear sulted in this process. Most of the indicators are descriptive extrapolation from past trends is also not uncommon. or positive indicators that are designed to provide a base- line against which a country can benchmark its own per- ABI developed a hybrid approach with which information formance over time, and allow for a country to compare is collected from multiple sources so that findings can be its performance on specific indicators with other African triangulated: countries. A few of the indicators are normative, implying » International data sources, typically on accessible specific policy or regulatory prescriptions or actions. They websites are designed to suggest performance shortcomings after » African government secondary databases and doc- comparison with other countries, as well as stimulate dis- uments reporting data relevant to ABI indicators, cussions and debate among public, private, and donor/ typically not found on websites NGO stakeholders, founded on comparative data. » Special studies that have generated and analyzed primary data, usually done by universities, research STUDY APPROACH institutes, or consultants, and often funded by do- As described in the Executive Summary, the ABI coun- nor agencies try study teams visited each country two or three times, » Key interviews with private sector agribusiness spending between four and eight weeks in each. Coun- managers, leaders of trade or industry associations tries were selected to give breadth and cover as many and farmer organizations (FOs), selected represen- particularities as possible. Both lusophone and franco- tatives of academia and the donor community, and phone countries were added as well as countries with key technocrats, agency department heads, and pol- geographically and climatically different areas. Local con- icy makers in government ministries and agencies sultants were employed to help the teams identify impor- » Available records from trade, industry and com- tant sources of information, including existing surveys, modity associations, FOs, parastatal agencies, and research papers, and reports. Data on 54 indicators were selected government-run or donor-funded projects collected, 44 of which are presented in this report. Local These sources are listed by topic in table 2.1. consultants assisted the study teams in preparing meetings and interviews with respondents representing public offi- ABI’s approach did not include launching formal surveys, cials, private sector investors, and farmers’ organizations. though small, purposively chosen samples were selected Their opinions and perspectives about the factors that for opinion/perception questions in which informants are restricting agricultural modernization and commer- were asked to provide ordinal ratings on a 0 to 5 scale. cialization were useful in enabling the teams to compose A shortcoming of these types of questions is that the summaries of the different situations in the nine countries. sample tends to be small and biased toward larger agro- enterprise firms that are easily accessible, literate, and INFORMATION-GATHERING APPROACH understand the concepts behind ratings. The working hypothesis at the outset was that approxi- mately 80 percent of the information required to inform ABI analysts were able to cross-check and triangulate the indicators already exists in some form, albeit not information from multiple sources in each country and always readily available and often available from dispa- internationally. A select sample of key informants was rate sources. There may also be public sector gatekeepers interviewed, covering participants and service providers who are reluctant to release information because it is not along the main staple crop value chains of interest (maize final, because the underlying data are problematic, or the and rice, and sometimes wheat or a key cash crop such findings are politically sensitive. Information is often not as cotton or coffee), as well as knowledgeable observers. 4 Agriculture Global Practice Discussion Paper TABLE 2.1. SOURCES FOR INDICATORS Category Key Informants and Documents Access to certified Ministry of Agriculture (departments dealing with seeds, fertilizer, and agricultural machinery); or improved seed agricultural research institutes (and any umbrella scientific research-type institution); any agency production and use or parastatal responsible for foundation seed production (and any agencies doing certified seed multiplication); any cross-agency bodies, such as a Seed/Varietal Approval Committee; any Ministry of Agriculture agency responsible for inspection, quality control, or certification; private seed multipliers and any private seed association; selected FOs (using improved seed) Fertilizer importation, Ministry of Agriculture (departments dealing with agricultural inputs, typically seeds, fertilizers, and mixing, distribution, agrichemicals); agricultural research institutes (and perhaps an umbrella scientific research-type and use institution); any agency or parastatal responsible for fertilizer importation and distribution; private agrodealer association; selected FOs (using fertilizer with improved seed); any government agency responsible for inspection, quality control, or testing of fertilizer; any private fertilizer importers, producers, or mixers/blenders; selected private wholesale traders/distributors (probably members of agrodealer association) Agricultural Ministry of Agriculture (division that works on mechanization); any government agency that imports mechanization (with and distributes tractors (for example, parastatal); private sector importers of tractors, spare parts, special attention to and tires; Ministry of Commerce/Customs for detailed data on imports of tractors and for tractors) information on import duties, value-added tax (VAT), and other charges/fees applied to imports of tractors and spare parts; any private sector association for tractor importers, distributors, or service centers; selected owners of tractors—FOs or cooperatives, NGOs, large farmers (who probably rent out their equipment); selected users of tractor hire services (FOs, NGOs); any agricultural machinery centers that provide tractor hire services and maintain/repair agricultural equipment (may not exist) Access to agricultural Central Bank (CB), commercial banks, microfinance institutions (MFIs), and rural banks; a sample of credit and credit users—FOs that take loans and on-lend to smaller agro-enterprises or small farms, medium agribusiness finance sized commercial farms, agroprocessors, importers of agricultural inputs (especially fertilizer and equipment), traders of agricultural commodities (and exporters); investment funds (if exist), program or project credit lines, grant components (typically of projects) targeting agribusiness Secondary research: Central Bank website and statistical bulletins, International Monetary Fund (IMF)/bank financial sector assessment program document, government financial sector strategy/ plan, financial sector surveys (if any) Transport indicators Traders, (grain) trade associations, transporters, truckers’ union or shippers’ associations, freight (mainly trucking) forwarders/international shipping companies Secondary sources of information: Transport Ministry planning and monitoring and evaluation (M&E) documents, transport surveys (if any), Rural Access Index (RAI) and Logistic Performance Index (LPI) measures, World Bank Group (WBG) Africa infrastructure assessments (ABI staff can get the last two). In some countries, a transport regulatory authority, road fund, or a road/highway administration may produce annual reports with much useful information. International organizations such as the World Food Program have logistics units that can provide a list of transporters that they use and negotiated transport rates. Bilateral donors (such as the Department for International Development [DFID]) providing technical assistance to the Ministry of Transport may be good sources as well. Focus on collecting information from grain traders and input suppliers on transport prices paid to ship agricultural products and inputs; traders’ (shippers) opinions/perceptions of transport services (provided by truckers) Policy-related Seed multipliers/dealers; fertilizer/agro-input dealers; fertilizer importers; importers/sellers of indicators tractors; agricultural machinery service centers (or individuals/large farmers who provide custom- hire tractor services); traders of agricultural products (especially grain); agroprocessors (especially millers); agricultural exporters (of staple and/or cash crops) Agribusiness Indicators: Synthesis Report 5 In cases where quantitative data were generated, such as be used by countries as a point of departure for further in estimates of rural transport costs, ABI analysts inter- investigation, consultation, and verification rather than as viewed samples of transporters, traders, and agribusi- a reference for making immediate policy changes. ness firms, using triangulation techniques to successively approximate plausible estimates. Given disincentives to public service, government institu- tions that generate and report agricultural and economic In order to guide information gathering in the field, the ABI statistics struggle to fulfill their mandates despite high lev- team developed guidelines for questions and checklists for els of foreign assistance. Out of necessity, this study relied each indicator group. These guidelines, which have under- heavily on information of variable quality, reliability, and gone continual refinement, are quite detailed and may accuracy, including secondary data from government and appear to go well beyond the scope of what is required to international sources as well as on some primary data (for inform particular indicators. The guidelines were designed, example, from the national agricultural census and periodic however, to pull together relevant background information national sample surveys of rural households). In relation of a policy, regulatory and institutional nature, which would to our best efforts to interpret incomplete and sometimes aid in interpreting data gathered solely to inform or calcu- ambiguous data, the critical reader is asked to exercise his late indicators. In many cases, background information is or her own judgment. We have somewhat more faith in found in recent government planning and strategy docu- information we generated in interviews with key inform- ments, agriculture sector and agribusiness studies, donor ants or in their responses to pointed e-mailed questions. strategies, program design papers, or other documents. Perception indicators have been used to gauge the private Table 2.2 indicates the probable typology for each indica- sector’s view of government policies and their implementa- tor. The nature of this research and attempts to get the tion. These can be criticized because of limited sample size most correct information for each indicator may mean that and concerns about consistency across countries of scored those sources vary by country. For example, monthly rural perception indicators. Nevertheless, these are included in retail prices for urea were available through a pilot bulletin the report because they demonstrate substantial alignment that was published monthly for some countries and not for with other, more objective measures of whether or not the others. For certain indicators, detailed customs informa- private sector is seen as partner in economic development. tion was shared, and in other countries the information was received as redacted data from the government. So as to provide the reader with as many tools as possible to exercise judgement on the agribusiness indicators, the QUALIFICATIONS: BE AWARE following section with by-country results provides informa- OF FACTORS tion on the type of informants and the type, by indicator. This synthesis report focuses on possible causes that are restricting agricultural production and highlights a num- BOX 2.1. A NOTE ON TRIANGULATION ber of indicators and metrics that will help to create bet- For quite a few of the indicators, the paper mentions the ter understanding and manage change. Much of this use of triangulation in the results. The purpose of trian- information has not been collected before. Rarely have gulation in qualitative research is to increase the credibil- cross-country comparisons at this scale and with these ity and validity of the results. Several scholars have aimed types of data been attempted. In compiling this report, to define triangulation throughout the years, but the defi- the team has had to make a number of judgment calls, nition most closely used for the purposes of this study is especially when presented with conflicting or incomplete according to O’Donoghue and Punch (2003), a “method of cross-checking data from multiple sources to search for information. Every effort has been made to triangulate regularities in the research data.” data. The individual country reports were peer reviewed by individuals with experience in each of the countries. Where multiple different sources were used, such as pub- The quality and accuracy of a number of these indicators lished data from different sources, or several perceptions to aggregate into, triangulation meant taking various sources will improve over time with repeated collection through and creating an average of each response. programs like the BBA. We recommended that this report 6 Agriculture Global Practice Discussion Paper TABLE 2.2. TYPE OF DATA Farm Indicators Description Input use and economics Fertilizers Average cereal yield (kg/ha) Government published data/international organizational data Fertilizer nutrients (kg/ha) Government published data/international (Abuja target 50+) organizational data Unsubsidized nutrient/output ratio (about 5) Investigator calculations based on government published data/interview; expressed as “Calculate nitrogen price from urea price and use maize price during the postharvest period. Urea is 46% N.” Maize price is an average of weekly prices over several months postharvest. Our interpretation is that lower ratios are better. Seeds Use of certified maize seeds (%) Government published data Hybrid maize seed to grain price ratio (best Informed opinions/interviews between 2 and 5) Tractors Tractor density (SSA average 13 per 100 km2) Government published data/international organizational data Input sourcing and costs Fertilizers Average fertilizer (unsubsidized) rural retail price Market information/ International Fertilizer US$/ton Development Center (IFDC) Fertilizer subsidy—Y/N? Government published data Fertilizer import taxes (%) Triangulated data from sources (customs, FAO) Seeds Private sector production of foundation seed (%) Informed opinions/perception triangulated with available data Private sector production of certified seed (%) Private sector/public sector published data Imported seed as total certified seed (%) Private sector/public sector published data Time (years) for seed testing and registration Government published data/international data Existence/implementation of national seed Government published data/international regulations/laws (0–5) organizational data Tractors Tractors imported by private sector (%) Government published data/international organizational data/interviews with private sector Tractor import tariffs as % CIF prices Government published data/international organizational data Parts imports tariffs as % CIF price Government published data/international organizational data Cost of tractor rental (pilot average $82) Opinion interviews/averaged Distribution Agrodealer network—number per 10,000 farms International organizational data/interviews and averages Financing Access Number of bank branches per 100,000 adult Central Bank/international data population Commercial banks lending to Ag (3 years) (%) Central Bank Cost Average lending rates for Ag loans (real rates) Central Bank/banks Nonperforming Ag loans (%) Published data—Central Bank (Continued ) Agribusiness Indicators: Synthesis Report 7 TABLE 2.2. TYPE OF DATA Farm Indicators Description Road access Rural Access Index (%) International organizational data Quality Road network in poor condition (%) International organizational data LPI infrastructure quality (0–5) International organizational data, World Bank Logistics Performance Index Cost Cost of transportation ($/mt/km)— Informal interviews—multiple sources; average main routes Cost of transportation ($/mt/km)— Informal interviews—multiple sources; average secondary routes Private Sector Indicators Perception of policy Enabling environment (0–5) Perception based: interviewee (private sector environment participant); rate between 0 and 5 Policy consistency (0–5) Perception based: interviewee (private sector participant); rate between 0 and 5 Private sector advocacy (0–5) Perception based: interviewee (private sector participant); rate between 0 and 5 Perception of Fertilizer business environment (0–5) Perception based: interviewee (private sector business participant); rate between 0 and 5 environment Ease of entry/operations in seeds (0–5) Perception based: interviewee (private sector participant); rate between 0 and 5 Mechanization business environment (0–5) Perception based: interviewee (private sector participant); rate between 0 and 5 Policy Environment Indicators Investment in Ag spending as percentage of budget—current Government published data/international agriculture organizational data/IMF Ag spending as percentage of budget—2001–10 Government published data/international organizational data/IMF Ag spending as percentage of Ag GDP Government published data/international organizational data/IMF Ag R&D spending as percentage of Ag GDP Government published data/international organizational data Input support Fertilizer subsidy as percentage of Ag Ministry Government source budget ISTA membership or adherence to OECD International organizational data: OECD requirements—Y/N Effectiveness of mechanization strategies Opinion interview; expressed qualitatively Banking support Evidence of credit reference (0–5) Perception based: rate between 0 and 5 Presence of collateral registry for loans Government source/interviews Existence of warehouse receipts (0–5) Perception based: rate between 0 and 5 Transportation Time (days) to get truck registered/operators’ Opinion interview/average based on interviews and support license possible published data Ease of entry into trucking (0–5) Perception based: rate between 0 and 5 8 Agriculture Global Practice Discussion Paper CHAPTER THREE POLICY ENVIRONMENT Private sector perceptions of public policy and the government’s role in the local econ- omy are defining characteristics and determinants of whether or not the business envi- ronment is an enabling environment. This relies on the government’s own attitudes toward business, and whether its commitment to needed policy reforms is genuine or simply a matter of politically expedient public rhetoric. Investors have a strong preference for stable, rational, and predictable policies that allow for long-term plan- ning, particularly when the investment has a long gestation period. Increased public spending on agricultural research and extension and purposeful reforms introduced to improve the capacity of the institutions responsible for governing and regulating the agriculture sector can reassure prospective investors regarding the government’s commitment. Rational and constructive policies relating to agribusinesses’ vital role in promoting agricultural development can likewise be instrumental in convincing inves- tors that the national government is an honest and reliable broker. Cultivating these types of perceptions on the part of investors and prospective investors is particularly important in contexts in which public expenditure on agriculture is limited, as is the case in the nine pilot ABI countries. It creates a more positive environment for public- private partnerships in which private investment complements and supplements public spending. A purposefully limited government role in input and product marketing functions such as distribution, storage, transport, and processing benefits agribusiness effectively and reduces the chances that private sector participation in these activities will be crowded out. It also enables government institutions to focus their attention and their efforts where intervention is warranted, for instance in settings characterized by market fail- ure and missing and incomplete markets. The government’s ability to identify and purposefully respond to these conditions with the right mix of incentives, regulations, controls, and strategic public investments is a vital element of a constructive public sector engagement in the agriculture sector. Although it is likely to take time, this con- structive and strategically selective engagement in the agricultural economy can foster increased investment on the part of agribusinesses and ultimately lead to development of the kind of competitive market that is characteristic of commercialized agriculture. Agribusiness Indicators: Synthesis Report 9 low-income countries where public resources are often KEY FINDINGS (TABLE 3.1) acutely scarce, 1 percent of agricultural GDP is gener- Agricultural Spending as a Share of Total Budget. ally seen as an appropriate level of public investment. In This proportion is an indicator used to reflect the fiscal higher-income developing countries this proportion tends to costs of agricultural policy, and the priority that a gov- increase to 3 or 4 percent. Among the ABI pilot countries, ernment assigns to agriculture relative to other sectors. Kenya is the only one for which the proportion of agricul- The Comprehensive Africa Agriculture Development tural GDP invested in publicly funded agricultural research Programme (CAADP) prescribes that African countries exceeds 1 percent, at 1.35 percent. allocate 10 percent of their total budget to agriculture. According to the results of a 2012 study, only 19 percent Private Sector Perception of the Enabling Envi- of Sub-Saharan African countries achieved this target; ronment (0–5). The way investors and prospective 57 percent of the countries allocated less than 5 percent investors subjectively apprehend the local environment to agriculture (World Bank 2012). Among the nine ABI for agribusiness differed markedly between the ABI pilot pilot countries, three achieved this level of commitment countries, and strong correlations emerged between as of 2010–11: Burkina Faso at 14.5 percent, Ethiopia these perceptions and the level of the local government’s at 13.8 percent, and Rwanda at 10.2 percent. In Ethio- involvement. Assigning this indicator a number rating of pia, Ghana, and Zambia, budgetary allocations have between 0 and 5, in which 0 is very unfavorable and 5 increased recently. The lowest performers with regard to is highly favorable, Zambia received the highest rating this indicator have been Nigeria at 2 percent and Kenya at 4.1, followed by Tanzania at 3.6. Governments in both at 4.3 percent—both proportions that declined between Zambia and Tanzania were seen as imposing a limited set 2000 and 2010. of regulations that made for an operating environment in which businesses had the necessary latitude to make Agricultural Spending as a Proportion of Agri- important investment decisions. Ethiopia and Nigeria cultural GDP. Public expenditure on agriculture rela- received the lowest ratings, both at 2.5. Investors in both tive to agricultural GDP is important because it provides countries saw themselves as crowded out of input and scale to the agricultural expenditure discussion. In larger, output markets by high levels of government intervention. more diverse economies (of middle- and higher-income The private sector perceptions captured by the pilot ABI countries), public expenditure on agriculture is typically are noteworthy, but they are based on a limited sample lower than it is in low-income countries, where agriculture and are not necessarily representative of a larger cohort contributes proportionally more to GDP. In Burkina Faso, of investors or of the private sector in general. Ghana, Rwanda, Ethiopia, and Zambia, the ratio is 0.3 and above. Nigeria is the outlier with public expenditure Private Sector Perception of Policy Consistency in agriculture which is only 0.06 of the percentage value and Predictability (0–5). In three countries, private of agriculture in the country’s total GDP.4 sector perception was positive: Kenya and Rwanda both at 3.5 and Zambia at 3.0. The rating of consistency was Public Spending on Agricultural Research as a based in large part on one or a small number of trade Share of Agricultural GDP . Agricultural research is a policy issues, and particularly on the ability of traders particularly strategic area in which to invest public resources. to export maize or rice across borders to neighboring This research is essential for sustained productivity growth countries. In Burkina Faso (1.5), Ethiopia (2.0), Mozam- and for delivering technologies and methods that enable bique (2.0), and Nigeria (2.0), private sector perceptions producers to adjust to the effects of climate change. Among were generally less favorable due to export bans imposed by the government as measures to curb price volatility 4 To make sound judgments about whether public expenditure in agriculture is in the international commodity markets during 2007 too low or too high, in-depth analyses of the composition of public expenditure and 2008. Some of these barriers have gone against are required. Both the World Bank and FAO have received grants from the BMGF to do such in-depth expenditure pattern reviews in some 20 African the spirit of the regional trade agreements signed with countries. the Economic Community of the West African States 10 Agriculture Global Practice Discussion Paper TABLE 3.1. ENABLING AND POLICY ENVIRONMENT INDICATORS Policy Burkina Environment Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Investment in Ag spending as 7.9 14.5 10.2 5.1 13.8 6.5 2.0 4.3 6.0 9.0 agriculture percentage of budget—current Trend in Ag spending 3≠, 3Æ, 3Ø down same same up same down down up up budget—2001–10 Ag public expenditure 0.26 0.43 0.32 0.16 0.30 0.24 0.06 0.15 0.30 0.36 ratio to Ag GDP a Agribusiness Indicators: Synthesis Report Ag R&D spending as 0.55 0.36 0.57 0.37 0.26 0.53 0.43 1.53 0.29 0.60 percentage of Ag GDP Input support Fertilizer subsidy as 20.5 4.0 30.0 NA NA 18.5 NA NA 38.1 12.0 percentage of Ag Ministry budget ISTA membership 2Y, 5N N N N N N N Y Y N or adherence to OECD requirements— Y/N Banking support Evidence of credit 1.6 0.0 1.5 1.5 0.0 1.5 2.0 3.0 3.0 2.0 reference (0–5) Presence of collateral 2Y, 5N N N N N N N Y N Y registry agency for loans Existence of 1.4 1.5 0.0 0.0 3.0 4.0 0.0 0.0 1.0 3.0 warehouse receipts (0–5) Transportation Time (days) to get 13 6 7 5 7 6 28 29 21 5 support truck registered/ operating license Ease of entry into 3.7 3.5 4.0 4.0 4.0 3.6 3.5 2.5 4.0 4.0 trucking (0–5) a When 1, then percentage of public expenditure on agriculture is the same as the agricultural GDP. Source: ABI Country Reports. 11 (ECOWAS) and the Union Economique et Monétaire Ouest in Zambia ABI respondents rated their effectiveness at Africaine (UEMOA).5 In Nigeria and Ghana, rice import 3.7, Kenya at 3.5, and Tanzania at 3.3. In a few countries, tariffs and related fees have been subject to many umbrella organizations of these groups represent a wide changes, undercut operations of some private import- range of private agribusiness interests across commod- ers, and discouraged private investment in domestic rice ity groups and agro-industries. In other countries where production and processing. apex organizations do not exist, commodity associations have been effective, focusing more narrowly on repre- Private Sector Perception of Advocacy Groups senting the interests of the participants in specific value (0–5). Advocacy groups often play an important role chains. In Zambia, the national farmers union (ZANFU), in lobbying for policy reforms and in persuading legisla- and in Kenya, the National Federation of Agricultural tors and other public officials, making them an impor- Producers (KNFAP) are seen as having been influential tant source of leverage for agribusinesses in developing in this capacity. These farmers’ unions cut across value countries. Their role in representing business in policy chains and are not tied to commodity-specific associa- dialogue makes them a vital asset to investors who wish tions (often called interprofessional associations or trade to have a voice in that dialogue, and a force in crafting associations dominated typically by exporters). There an enabling environment for investors. Most typically, are strong producer groups tied to a particular key com- the benefit of these types of dialogues is that they better modity, for instance cotton in Burkina Faso. As a result, enable government policies to be aligned with commer- respondents in Zambia rated them quite favorably, at cial realities and market opportunities. In Africa, these 3.7 out of 5. Kenyan and Tanzanian respondents gave groups have generally been less influential, although with favorable ratings as well, at 3.5 and 3.3, respectively. In the help of some bilateral donors, this has recently begun Rwanda, on the other hand, where the chamber of com- to change. In three of the ABI pilot countries, agribusi- merce represents the interests of multiple sectors and nesses rate the capacity and effectiveness of these advo- industries, respondents found its influence on behalf of cacy groups in influencing public policy quite favorably: agribusiness to be less effective. 5 ECOWAS is regional economic community (REC) of 15 West African coun- tries, including all three West African ABI study countries—Burkina Faso, Ghana, and Nigeria. UEMOA covers a subset of the ECOWAS countries—six francophone countries plus Guinea-Bissau in West Africa (including Burkina Faso)—and Chad. 12 Agriculture Global Practice Discussion Paper CHAPTER FOUR SEED Improved, high-quality, often certified seed is a critical input for increasing crop yields and improving agricultural productivity. The low rates of improved seed use among smallholders in the nine pilot ABI countries are a major cause of their low yields, par- ticularly when considered in combination with their low rate of use of other inputs. Maize yields throughout the continent are a case in point. The average yield of 1.5 tons maize per hectare is roughly one fifth the yield typical of maize production in industrialized countries. The factors believed to prevent the majority of African farmers from using improved seeds include lack of awareness, cost, and often most critically, the shortage of improved seed at the right time and place. Awareness and cost can be closely related. The farmer who does not understand the potential effects of an improved seed variety on his or her productivity and income is likely to see the higher price of that input as being unwarranted and therefore may be much less willing to adopt the variety. Yet the cost of improved seed is usually less than 5 to 10 percent of his or her total production costs. Although the yield gaps that characterize so much of African agriculture may seem daunting, the enormous potential to increase productivity by effectively target- ing these constraints must inform policy formulation because the opportunity is great and the opportunity costs of business as usual are unsustainable. The selection of seed variety is a critical decision for farmers. Helping to inform those decisions with better information is one of the most important services a farmer can be provided. The seed industries in many African countries limit plant breeding and testing to a limited number of crops, principally maize and sorghum, with little if any attention to tubers, grains, legumes, and horticultural plants that are food staples within many countries. Public sector and parastatal agencies often play a dominant role in seed multiplication and marketing, raising concerns that private firms are being crowded out. Lengthy and complicated varietal testing requirements, certification procedures, regulations, and price-setting policies inhibit the emergence of private seed production and distribution. A number of these concerns appear to be corroborated by the find- ings of the ABI pilot study. Agribusiness Indicators: Synthesis Report 13 14 TABLE 4.1. SUMMARY OF SEED INDICATORS Burkina Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Seeds Use of certified maize seed (%) 31 50 12 6 18 27 7 70 65 19 Hybrid maize seed to grain price ratio 7 13.0 NA 10.0 7.0 11.0 4.0 5.0 10.0 2.0 Private sector production of foundation seed (%) 14.8 0 0 13 5 NA 0 0 100 0 Private sector production of certified seed (%) 66 100 NA 15 12 79 54 10 100 100 Imported seed as percentage of total certified 14 NA NA 15 1 56 15 15 1 7 seed Time (years) for seed testing and registration 2.7 5.0 3.0 3.0 2.0 2.0 2.0 3.0 2.0 2.0 Existence/implementation of national seed 2.9 2.5 2.0 2.0 2.0 4.0 2.0 3.5 4.5 3.0 regulations/laws (0–5) Ease of entry/operations in seeds (0–5) 2.7 3.5 2.0 2.0 1.5 3.7 2.0 2.5 4.9 2.5 ISTA membership OECD 2Y, 5N N N N N N N Y Y N Agrodealer network—number per 10,000 3.0 2.1 1.8 0.3 2.2 1.3 2.1 5.8 NA 8.4 farms Source: ABI Country Reports. Agriculture Global Practice Discussion Paper FIGURE 4.1. SEED IMPORTS IN THOUSANDS FIGURE 4.2. SEED-TO-GRAIN PRICE RATIO OF METRIC TONS, 2011–12 IN THE PILOT COUNTRIES, 56 2011–12 CROPPING SEASONS 13 10 10 10 7 15 15 15 5 4 7 1 1 Tanzania Kenya Mozambique Nigeria Ghana Ethiopia Zambia Source: ABI Country Reports. Burkina Faso Tanzania Zambia Mozambique Ethiopia Kenya Nigeria Source: ABI Country Reports. KEY FINDINGS (TABLE 4.1, trained and licensed agents who are usually certified by FIGURES 4.1, 4.2) the regulatory authorities. Data on this indicator were The use of certified maize seed is measured in terms of available from only three ABI countries: Ethiopia, Kenya, the percentage of a country’s total maize area planted and Nigeria. Kenya was the only country in the sample with certified seed. Of the nine ABI pilot countries, in which the majority of farmers (over 60 percent) used Kenya and Zambia have the largest proportion formal sources. In Nigeria, 90 percent of farmers used of maize-cultivated areas planted with certified informal sources. maize seed at 70 and 65 percent, respectively. In Burkina Faso the proportion is 50 percent. The seed industries in Imported Seed as a Percentage of Total Certified the three countries are characterized by large networks of Seed in Pilot Countries. With the exception of Tan- agro-input dealers, a role that is often dominated by gov- zania, where 56 percent of certified seed was imported ernment and parastatal agencies in other countries. in the 2011–12 growing season, imports were low at only 15 percent in Kenya, Mozambique, and Nigeria. In Ethi- Mozambique, Nigeria, and Rwanda have the opia and Zambia this proportion was extremely low, at smallest proportion of maize area planted with 1 percent. Ghana imported 7 percent of its certified seed. improved seed at 6, 7, and 12 percent, respectively. These low import levels appear to be the result of laws These three countries are also notably subject to certified that require imported seed to undergo lengthy and cum- maize seed shortages and have seed industries that are bersome processes of testing and certification. unable to adequately supply farmers at reasonable prices and in a timely manner. There is evidence of a disconnect Seed-to-Grain Price Ratio. The ratio of the price of a between demand for improved seed and agencies’ ability to unit of seed to the price of a unit of grain is an indicator supply farmers at those volumes and with the types of seeds used to gauge how efficient the seed industry is in a par- that farmers want, suggesting that these seed systems are ticular country. A useful benchmark is the seed-to-grain not demand driven. This may warrant a reexamination of price ratio in the region of 5:1 for maize cultivated in Afri- the role of the public sector in the seed industry. can countries. At low seed-to-grain price ratios, hybrid maize seed is attractive to smallholder farmers, even if Certified Seeds as a Percentage of All Seed Used in yields remain low (Lopez-Pereira and Morris 1994). At a Pilot Countries. Most of the seed used by African farm- high seed-to-grain price ratio of 20:1, the yield advantage ers is generated from within the informal sector. Informal must be very high for a hybrid seed business to be attrac- seed is most consistent with locally sourced, unimproved tive and profitable. In Kenya, the seed-to-grain price ratio varieties, and recycled seed that generally does not lead has improved progressively from about 10:1 in early 2000 to yield increases. Formal seed is typically purchased from to 5:1 in 2012 (Odame and Muage 2010). Agribusiness Indicators: Synthesis Report 15 FIGURE 4.3. NUMBER OF SEED COMPANIES AND PERCENT OF CERTIFIED SEEDS SUPPLIED BY THE PRIVATE SECTOR 100 100 100 90 79 52 54 35 34 32 16 15 13 12 10 10 Burkina Faso Zambia Ghana Tanzania Nigeria Mozambique Ethiopia Kenya Percent of certified seed multiplied by the private sector Number of seed firms Source: ABI Country Reports. Note: Although KSC is a government parastatal, it has been restructured to operate like a private seed firm. Proportion of Foundation Seed Produced by the 79 percent. In Ethiopia, Mozambique, and Kenya, this Private Sector. Foundation seed is produced by pub- proportion is much lower, at 12, 15, and 10 percent, lic sector institutions in most of the ABI pilot countries. respectively. In Kenya, the Kenya Seed Company (KSC), Zambia is the only pilot country in which foundation seed a government parastatal, operates more like a private firm is supplied entirely by the private sector. Private sources and is the dominant player in the production and market- in Ethiopia and Mozambique supply 5 and 13 percent of ing of certified seed. The Ethiopian seed enterprises, the foundation seed, respectively. The predominance of the development enterprises, and regional seed companies are public sector in this seed production makes the capacity the predominant sellers of certified seed. In combination, of the responsible public agencies a critical determinant these players have more than 80 percent of total market of supply. Do the agencies have the necessary person- share of the Ethiopian seed market. The Ethiopian gov- nel, laboratory infrastructure, and financial resources to ernment uses the cost-plus approach to set a ceiling price produce enough foundation seed to adequately multi- for certified seed which constricts the potential profits of ply and distribute certified seed to farmers? If they lack private seed companies. In Tanzania, taxes and lack of these resources, their predominance in this industry is access to finance discourages private sector participation. likely to be a major factor restraining the adoption of certified seeds. Existence and Implementation of Seed Law (0–5). A comprehensive and functioning set of seed laws and Proportion of Certified Maize Seed Supplied by regulations facilitate access to improved seed. The exis- the Private Sector, and the Private Sector’s Share tence of seed laws is seen as an opportunity to harmo- of the Certified Seed Market (figure 4.3). Private nize laws and facilitate levels of policy coordination that sector involvement in certified seed is far more pronounced lend themselves to the development of the seed industry. than it is in foundation seed. All of the certified maize seed A number of initiatives and processes are in place among in Burkina Faso, Ghana, and Zambia is produced and mar- regional groups, mostly regional economic communities, keted by the private sector. In Tanzania the proportion is to facilitate the production and distribution of seed within 16 Agriculture Global Practice Discussion Paper BOX 4.1. EXAMPLE OF SUCCESSFUL REFORMS: TURKEY Turkey is seen as an example of how changes in govern- lines available under the system of rigid price fixing and the ment regulations and control can enable the active partici- minimum margin structure of the industry. pation of private sector actors and help in the development of the seed sector. Before reforms in 1980, the government In 1982 Turkey introduced reforms allowing private com- dominated all stages of the seed industry in the country. panies to do their own variety performance tests, reporting The government variety testing program made it extremely results to the government while the testing requirements from difficult to introduce genetic advances in seed technology the previous minimum of three years were lowered to one that were available elsewhere in the world. The time span to three years, depending on the crop and type of seed (e.g., for new variety introduction, including hybrids, and their one year for hybrids, two years for vegetables). Since Decem- approval by the key Variety Registration Committee, was ber 1983 each seed firm has been allowed to set its own seed prices (Gisselquist and Pray 1999). Reforms aimed at relaxing a minimum of three years. Seed companies operating out- restrictions have had positive effects on the seed sector in Tur- side of Turkey had been reluctant to undergo the lengthy key. Table B4.1.1 illustrates the increased availability of new time period required to obtain commercial introduction of cultivars in Turkey. hybrids in Turkey or to make the local production of parent TABLE B4.1.1. EFFECTS OF REGULATORY REFORMS ON THE INTRODUCTION OF NEW VARIETIES IN TURKEY Total Number Number of New Number of New Approved Prior to Cultivars between Cultivars between Crops 1982 1982 and 1987 1987 and 1992 Wheat 21 22 38 Sunflower 3 29 45 Maize 44 95 99 Sugar beets 11 18 6 Potatoes 8 13 35 Soybeans 2 43 27 Tomatoes 43 32 203 Cucumbers 1 8 115 Tobacco 31 7 2 Cotton 9 8 11 Source: Gisselquist and Pray 1999. Technology Transfer and Private R&D to Support Africa’s Green Revolution. and between member countries. These processes mostly Seeds Act is in compliance with the East African Commu- involve harmonization of laws and regulations, particu- nity harmonization agreement, Kenya and Uganda have larly in terms of movements and trade policies that allow still not adopted their laws to allow full implementation of cross-border trade. This type of regional harmonization the agreements. of seed legislation, regulation, and trade policy remains incomplete in a number of pilot countries, despite years Since 2002, the 17 member states of the West African Eco- of effort. nomic and Monetary Union (UEMOA), the Economic Community of the West African States (ECOWAS), and Tanzania has been a party to several harmonization the Permanent Interstate Committee for Drought Con- agreements with other countries in the East African Com- trol in the Sahel (CILSS) have been working on a regional munity (EAC), the Common Market for Eastern and agreement aimed at facilitating cross-border trade in seed. Southern Africa (COMESA), and the Southern African In 2008, the ECOWAS Council of Ministers adopted the Development Community (SADC). While Tanzania’s seed agreement but implementation remains limited. Agribusiness Indicators: Synthesis Report 17 When agreements like these are more fully implemented, performance trials ($500), and distinctness, uniformity, they should allow seeds of any variety that are registered and stability tests ($600 for two seasons). in one country to be produced and commercialized in all the participating countries without further testing and cer- Membership in ISTA and Adherence to OECD tification. All the pilot countries studied were at different Guidelines. Participation in international seed stan- stages of initiating, reforming, and/or updating existing dard organizations such as the OECD and the ISTA by seed laws. In Rwanda, the national seed law was passed in Kenya and Zambia has facilitated trade in certified seed. 2003, and national seed policies were developed in 2007; However, most of these organizations’ trading partners both have been reinforced by the decrees of the Minis- in Africa, including the other ABI pilot countries, do not try of Agriculture and Animal Resources (MINAGRI) of belong to OECD or ISTA. Although Tanzania does not 2010. However, the seed law and policies are still to be belong to either ISTA or OECD, the country benefits implemented. Similar situations exist in Nigeria, Kenya, from membership in the regional Common Market for Ethiopia, and Ghana. Eastern and Southern Africa (COMESA) and has been able to engage in international trade with the other coun- Time for Seed Testing and Registration (2-Year tries in the region. Standard). With the exception of Kenya (which has enabled international seed trade with other countries Private Sector Perceptions of the Ease of Entry under the OECD seed testing scheme), most ABI coun- and of Ongoing Operation in the Seed Business tries have a two- to three-year period of testing prior to (0–5). Private sector respondents expressed positive the release of new seed and imported varieties. Restric- impressions on the ease of entry into the seed business tions on imports limit seed supply. In most countries, it in Tanzania and Zambia. This may be related to the takes two to three cropping seasons in order to complete relatively limited presence of the public sector in the a full testing and registration cycle. The number of years industry in these two countries. In Ethiopia and Nige- required for new seed variety release is two years in Ethio- ria, where government agencies and parastatals have a pia, Ghana, Nigeria, Tanzania, and Zambia. The pro- significant share of the seed market, private sector per- cess is longer in Burkina Faso (five years), Kenya (three ception is far less positive. Government interventions in years), and Mozambique (three years). In Kenya, a variety setting seed prices administratively and subsidizing dis- approval process fee is levied and attributable to national tribution distort the seed market. 18 Agriculture Global Practice Discussion Paper CHAPTER FIVE FERTILIZER Fertilizer use is estimated to account for yield increases of between 40 and 60 percent of global food production. Chemical or “inorganic” fertilizers are widely used in many of today’s farming systems, and their use supports scalable agricultural production. Their use has made it possible for the global population to expand from 1.6 billion in 1900 to more than 7 billion today (Smil 2001). Inorganic fertilizer is tailored to provide crop plants with specific nutrients and is a critical complementary input to improved seed. Maize is an important case in point—obtaining high yields for hybrid maize requires the right type of fertilizer to be applied at both the right times and in the right quantities. Increasing chemical fertilizer use in African countries has been a priority for decades, and the food and fuel crisis imbued the objective with new urgency. In the longer term, replenishing soil fertility using various forms of fertilizer is a solid option for agricultural sustainability in a region where land-use practices have often mined the soil of nutrients, leading to widespread land degradation. KEY FINDINGS (TABLE 5.1) Fertilizer Nutrient Use per Hectare (kg/ha). At 59.6 kg/ha, Kenya is the only ABI pilot country that has met the Abuja Declaration target of 50 kg/ha as of 2011–12. Kenya’s high average fertilizer application rate is influenced by policies that have facili- tated a competitive private fertilizer trade and was further helped by donor investments that helped to expand networks of fertilizer distributors and other agrodealers. Coun- tries with more agrodealers selling fertilizer and a denser network of distributors, such as Kenya and Ghana, have higher fertilizer use per hectare, as well as lower rural retail prices. As of 2010, Zambia was within striking distance of the Abuja Declaration target at 37.0 kg/ha, having expanded use of fertilizer with the help of high fertilizer subsidies and attractive maize support prices. Three other countries—Ethiopia, Rwanda, and Ghana—fall in the 17 to 20 kg/ha range, only 35 to 40 percent of the Abuja target. Fertilizer use in many African countries is concentrated on the most profitable and widely grown cash crops, such as cotton, tobacco, and cocoa, including in Burkina Faso, Ghana, and Mozambique. In some of the ABI countries, input credit is provided and recaptured by parastatal marketing agencies or large private firms when they later purchase the farm output. Fertilizer use on staple food crops is limited in many African Agribusiness Indicators: Synthesis Report 19 20 TABLE 5.1. SUMMARY OF FERTILIZER INDICATORS Burkina Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Fertilizer Fertilizer nutrients (kg/ha) (Abuja 20 10.1 18.3 2.6 17.0 10.1 6.5 59.6 37.0 19.8 target 50+) Nutrient/output ratio (good < 5) 6 5.1 3.4 11.0 2.4 6.7 3.5 8.0 12.7 2.6 Average fertilizer (unsubsidized) retail 721 808 800 1,023 614 682 809 514 760 475 price US$/ton Fertilizer subsidy—Y/N? 7 Y, 2 N Y Y N N Y Y Y Y Y Fertilizer import taxes (%) 1.8 8.5 0 2.5 0 0 5 0 0 0 Fertilizer business environment (0–5) 2.6 2.5 2.0 3.0 0.0 4.0 1.0 4.0 4.2 3.0 Fertilizer subsidy as percentage 20.5 4.0 30.0 NA NA 18.5 NA NA 38.1 12.0 of Ag Ministry budget Agrodealer network—number 3.0 2.1 1.8 0.3 2.2 1.3 2.1 5.8 NA 8.4 per 10,000 farms Source: ABI Country Reports. Agriculture Global Practice Discussion Paper FIGURE 5.1. NUTRIENT/OUTPUT PRICE RATIO 12.7 Unsubsidized 11 Subsidized 8 6.7 5.1 4 3.4 3.5 3.6 3.4 3.5 2.4 2.6 1.7 1.75 1.5 Ethiopia Ghana Rwanda Nigeria Burkina Faso Tanzania Kenya Mozambique Zambia Source: ABI Country Reports. countries, unless there is a well-funded subsidy program as Kenya (8:0), Mozambique (8:14), and Zambia (12:7). The there is in Rwanda and Zambia. A major factor limiting ratio ranged from 1:5 (Ghana) to 4:0 (Kenya) using sub- use of fertilizer on food crops in Africa is the high cost of sidized urea prices, which means that subsidized fertilizer fertilizer in rural areas. African farmers generally pay far use on maize is very attractive financially for producers. In more for their fertilizer than do Asian farmers. Zambia, the combination of subsidized input prices and price supports for outputs improved the nutrient/output The Nutrient/Output Ratio (figure 5.1). ABI’s ratio to 1:8, compared to the ratio of 12:7 without sub- main proxy for determining whether fertilizer use is prof- sidies or price supports. The ratios shown in figure 5.1 itable is the nutrient/output cost ratio. This ratio com- use both nonsubsidized (blue bar) and subsidized (red bar) pares the price per kilogram of nitrogen fertilizer to the input prices and market output prices.6 cost of maize—the primary grain on which it is applied. The study designates, as a general rule, a ratio of 5:1 or The Value/Cost Ratio. This ratio measures the techni- less to encourage fertilizer use, while acknowledging that cal response to fertilizer use in terms of the O/N ratio, or other determinants such as physical response of crop to units of crop output from one unit of nutrient to the nutri- fertilizer make this ratio complex. Ratios in the 6:1 to 9:1 ent/output price ratio [Pn/Po], noted above. The value/ range are less likely to encourage fertilizer use, although cost ratio (VCR) uses average rather than marginal pro- it may still be profitable. Ratios of 10:1 or over actively ductivity data and is therefore an approximate measure. discourage fertilizer use (Morris et al. 2007), although Although the VCR is a better indicator of profitability relative prices are only one determinant; the other deter- than the nutrient-output ratio, the necessary data are not minant is the physical response of the crop to fertilizer available in most African countries, which suggests that application. Our interpretation is that lower ratios are most African ministries of agriculture and agricultural better. Fertilizer subsidies and price supports reduce research institutes have no empirical basis with which to (improve) this ratio, making fertilizer use proportionately more attractive to farmers. 6 In order to calculate the nutrient/output ratio in economic as opposed to financial terms, real, undistorted prices are required. Hence, ABI uses unsub- The ratio of unsubsidized prices varies widely across the sidized fertilizer prices and spot market prices for cereals as opposed to high ABI pilot countries. ABI findings show that the ratio was support prices set for grain by grain boards or food security agencies. The latter low in Ethiopia (2:4) and Ghana (2:6) but very high in prices are typically not available to most sellers of grain. Agribusiness Indicators: Synthesis Report 21 FIGURE 5.2. AVERAGE RETAIL PRICE OF UREA IN RURAL AREAS (US$ PER METRIC TON), 2011–12 Black Sea 424 Middle East 472 Ghana 475 Kenya 514 Ethiopia 614 South Africa 637 Tanzania 682 Zambia 760 Rwanda 800 Burkina Faso 808 Nigeria 809 Mozambique 1,023 0 200 400 600 800 1,000 1,200 Source: ABI Country Reports. Note: The Black Sea and Middle East regions are cost insurance and freight (CIF) variant. assess the potential economic benefits of using fertilizer. per hectare. African farmers generally pay much more for In the ABI countries where data were available, Kenya fertilizer than farmers in most other parts of the world. and Ethiopia, the VCR was rather low, at 1:3 to 2:4 and 2:3, respectively. Typically, a ratio greater than or equal to Urea Prices in All ABI Countries Vary Widely two suggests that fertilizer use should be profitable. and Are Generally Well Above FOB. Prices in the Middle East ($472) and in the Black Sea region ($424) It is important to note that the calculated values of both were used as comparisons.7 Among the nine ABI pilot ratios are extremely sensitive to the agricultural product countries, Ghana ($475) had the lowest-priced fertilizer in prices chosen for the calculation and can fluctuate dramat- part due to competition among at least five well-financed ically according to when the ratios are calculated. Grain importers, as well as a large number of rural agrodeal- prices vary across months during the marketing season, ers. Kenya has nine importers and a dense input supplier and maize sales prices vary seasonally. An unweighted network. High fertilizer prices at retail outlets in Mozam- annual average price across months is not ideal. Prices bique ($1,023), Nigeria, ($809), and Burkina Faso ($808) should reflect levels during the period of most frequent discourage farmers from purchasing urea and contribute and plentiful sales, which in Africa is typically during the to low average application rates. Burkina Faso’s retail urea three to four months after the main harvest. price in market towns was the third highest among study countries at $808 per metric ton in 2011–12, in part due Average Retail Price of Urea in Rural Areas (US$/ to the country being landlocked, with high transport costs ton, figure 5.2). There is significant variability in fertil- from coastal ports, as well as due to limited commercial izer prices for the ABI countries. In Ghana and Kenya, import volume. competition among private sector actors is believed to have helped to drive down prices in rural areas where the input distribution network is denser. There is an inverse 7 Ghana’s fertilizer prices were collected in 2012, significantly before other relationship between fertilizer prices and fertilizer use data, which may account for the low cost. 22 Agriculture Global Practice Discussion Paper FIGURE 5.3. AGRO-INPUT DEALER DENSITY: AGRO-INPUT DEALERS PER 10,000 FARMERS 8.4 5.8 2.2 2.1 2.1 1.8 1.3 0.3 Ghana Kenya Ethiopia* Nigeria Burkina Faso Rwanda Tanzania Mozambique Source: ABI Country Reports. *In Ethiopia, agrodealers are cooperatives. Agrodealer Density (Agrodealers per 10,000 vary, some countries have imposed miscellaneous Farmers, figure 5.3). The density of input suppliers taxes on the declared value of fertilizer, such as import varied between the best and worst countries by a factor declaration fees, high duties on micronutrients used in of over 20. Ghana (8.4) and Kenya (5.8) have the best- fertilizer blending, and VAT on imported raw materials. developed agrodealer networks. Agrodealer density is Mozambique has a 2.5 percent duty for imports originat- inversely related to the distance and time a farmer must ing outside of the Southern African Development Com- travel to an input market. In both Kenya and Ghana, munity (SADC) zone. Burkina Faso has a 5 percent duty the policy environment has been more conducive to the on fertilizer imports from outside ECOWAS, as well as emergence of private sector dealers, input dealer associa- 3.5 percent in additional fees. Ghana adds miscellaneous tions, and dealers capable of providing input buyers with taxes totaling 4.5 percent of the declared fertilizer value, technical advice. In Kenya the high number of agrode- while Kenya requires payment of a 2.25 percent import alers per 10,000 farmers is in good part because of the declaration fee. AgMark project financed by the U.S. Agency for Inter- national Development (USAID), reducing the average Some countries charge higher duties on imports of micro- distance farmers traveled to fertilizer sellers from 8.1 km nutrients, thus penalizing domestic fertilizer blenders. As (1997) to only 4.0 km (2010). Over the same period, the a result, domestic blenders have trouble competing with use of fertilizer applied per hectare in Kenya rose from international suppliers. For example, Zambia applies an average of 58 kg to over 100 kg. Moreover, In Burkina 15 percent duty plus 16 percent VAT; Nigeria also applies Faso, only 12 percent of farmers traveled from less than a 5 percent VAT for fertilizers that are used in blending 10 km to buy inputs. within the country from imported raw materials. These practices have undercut the emergence of a domestic Fertilizer Import Taxes (%). In accordance with the fertilizer industry in those two countries and restrict the Abuja Declaration, fertilizer imports are tax exempt in capacity of the local fertilizer blenders to create fertilizers the ABI countries with two exceptions: Mozambique designed to match the nutrient requirements of specific and Burkina Faso. While the types of taxes and amounts crops in particular areas. Agribusiness Indicators: Synthesis Report 23 Fertilizer Subsidy Programs. Most of the ABI pilot (4 percent). The most recent and comprehensive study countries introduced fertilizer subsidy programs to accel- identifies their benefits as providing a quick and highly erate adoption rates. The two exceptions were Ethiopia visible response and a “power political tool.”8 However, and Mozambique. Seed and fertilizer subsidy programs in the benefits tend to wear off over time, and the costs are Africa focus mainly on fertilizer-responsive cereal crops, high and can divert public funds from investments that such as (hybrid) maize, rice, and to a lesser extent, wheat. provide better returns over a longer period of time. Once The proportion of fertilizer that is subsidized and used by started it is very challenging for governments to discon- farmers varies from low levels, less than 5 percent of total tinue subsidies. fertilizer use in Mozambique in 2009–10, to 15 percent in Burkina Faso, and almost 90 percent of the fertilizer used Private Sector Perception of the Fertilizer Pol- by smallholders in Zambia. Correspondingly, the mag- icy and Business Environment. In Zambia (4.2), nitude of the subsidies received by farmers varies from Kenya (4), and Tanzania (4), private sector importers and nearly one-quarter of the fertilizer cost in Burkina Faso agrodealers perceive their countries’ fertilizer policy and to 50 percent in Rwanda to 75 percent in Zambia. The business environment positively. These impressions appear timeliness of fertilizer distribution varies across countries, to relate mainly to the ease of participation, in spite of and subsidized fertilizer has often been distributed late to active participation by their governments. In Rwanda (2), farmers. The subsidies typically target small and medium- Burkina Faso (1.5), Nigeria (1), and Ethiopia (0), the pri- sized farms. vate sector was far less optimistic. Each of these countries is characterized by heavy public sector involvement in the Fertilizer (and Seed) Subsidy Program as Percent importation and distribution of fertilizer, as well as subsi- of Ministry of Agriculture Budget. In keeping with dies that discourage private sector participation. In Ethio- the Abuja Declaration, public expenditure on fertilizer pia, where the private sector once participated actively (and seed) subsidies as a proportion of total public expen- in fertilizer distribution, the government’s reentry into diture on agriculture is high in Rwanda (30 percent) importation and distribution during the later 1990s led to and Zambia (38 percent) compared to Burkina Faso a general private sector exit from the activity. 8 Jayne, T.S., and Shahidur Rashid. 2013. “Input Subsidy Programs in Sub- Saharan Africa: A Synthesis of Recent Evidence.” Agricultural Economics, Inter- national Association of Agricultural Economists, vol. 44(6), November, pp. 547–562. 24 Agriculture Global Practice Discussion Paper CHAPTER SIX FARM MECHANIZATION The use of farm machinery is associated with agricultural modernization, produc- tivity growth, and the need to compensate for less farm labor throughout much of the world. In general, countries that perform best in terms of reducing hunger are also countries that manifest higher net investment rates per agricultural worker (FAO and UNIDO 2008). Throughout the 1990s, the value added per worker in the group of countries with less than 2.5 percent of the population being undernourished was about 20 times higher than in the group with more than 35 percent being undernour- ished. This disparity has been attributed to the differences in agricultural investment in the two groups of countries. Mechanization plays a critical role in realizing the full benefits of improved seed, fertilizers, and pesticides, as well as irrigation water. Studies by the United Nations Industrial Development Organization (UNIDO) suggest that African agriculture’s heavy reliance on human power (65 percent), compared to that of Asian agriculture (30 percent), Latin American agriculture (25 percent), and North American agriculture (20 percent) is an important factor contributing to low produc- tivity and low rates of commercialization (figure 6.1). Sub-Saharan farmers use nearly three times as much human power as farmers in North Africa, and tractor power in North America is six times higher. In Sub-Saharan Africa, the rate of capital deprecia- tion exceeds the rate at which capital is invested in agriculture, and this applies directly to farm machinery. As a result, annual capital stock per worker has been declining over the last 20 years (FAO 2012). This is emblematic of the very low rates of agricultural investment that are typical throughout much of the region. Unless this investment can be significantly increased, smallholder farmers in Africa will continue to rely over- whelmingly on their own power and that of their draft animals. KEY FINDINGS (TABLE 6.1) Tractor Density (figure 6.2). The number of tractors per 100 square kilometers of arable land increases as more farms become larger and more commercialized. Kenya and Zambia have the highest number of tractors, at 27 and 21 per 100 square kilome- ters, respectively. This appears to be a reflection of these two countries moving toward larger-scale, commercialized farming. The lowest tractor densities among the ABI pilot Agribusiness Indicators: Synthesis Report 25 FIGURE 6.1. FARM POWER SOURCES (PERCENTAGES) IN SUB-SAHARAN AFRICA, ASIA, AND LATIN AMERICA, 2006 78 Human power Animal power Tractor power 65 60 55 55 50 40 35 30 30 30 25 25 25 20 20 15 15 10 10 7 East Africa North Africa Asia Latin America Sub-Saharan Africa Nigeria Tanzania Source: FAO 2008; FMA&RD 2011. FIGURE 6.2. TRACTORS PER 100 KM2 IN SELECTED COUNTRIES 200 27 21 8.9 11 12.7 13 4 5.7 7.4 1.3 Rwanda Ethiopia Nigeria Tanzania Burkina Faso Ghana Mozambique Zambia Kenya Sub-Saharan World Africa Source: FAOStat 2012. countries are Rwanda at 1.3, Ethiopia at 4, and Nigeria at Zambia (0.21) have some of the highest tractor densities 5.7. Comparisons with countries in other regions reveal among the pilot countries, they lag behind international just how far behind African countries are in the area of comparators such as Tunisia (1.43) and Pakistan (0.9). mechanization. The numbers of tractors per 100 square Both countries would need to increase the fleet of tractors kilometers of arable land in Tunisia and Brazil are 143 18 times in order to reach the average level of horsepower and 129, respectively; the global average is about 200. of Tunisia. Horsepower per hectare is generally very low in Africa (0.13). In most of the pilot countries, tractors Average Tractor Horsepower per Hectare are predominantly small- and medium-range horsepower (table 6.2, figure 6.3). Although Kenya (0.27) and (typically 30 to 70 horsepower), which make up more than 26 Agriculture Global Practice Discussion Paper TABLE 6.1. SUMMARY OF FARM MECHANIZATION INDICATORS Burkina Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Mechanization Tractor density (SSA average 13 per 100 km2) 11 8.9 1.3 12.7 4.0 7.4 5.7 27.0 23.0 11.0 Tractors imported by private sector (%) 65 59 42 60 100 10 10 100 100 100 Tractor import tariffs as percentage 2.3 16 0 5 0 0 0 0 0 0 Agribusiness Indicators: Synthesis Report of CIF prices Parts import tariffs as percentage 16.4 20 30 25 0 10 5 16 15 27 of CIF prices Cost of tractor rental (pilot average $82) 82 51 163 62 81 68 93 50 125 46 Mechanization business environment (0–5) 2.8 1.0 2.0 3.0 5.0 3.6 1.0 5.0 2.8 1.9 Source: ABI Country Reports. TABLE 6.2. ESTIMATE OF TOTAL TRACTOR HORSEPOWER (HP) PER 100 SQUARE KILOMETER IN KENYA, 2011 Probability of Estimated Numbers Midpoint of Horsepower Total Horsepower in Tractor Capacity (HP) Distribution (Pi) for Each Group for Each Class of Tractors Each Group 70–80 0.20 2,880 75.0 216,000 81–120 0.70 10,080 100.5 1,013,040 121–200 0.10 1,440 160.5 231,120 Total 14,400 1,460,160 Average horsepower/ha 0.28 HP/ha arable landb Average horsepower/ 2,755 HP/100 km2 100 km2 of arable land Source: Kenya Ministry of Agriculture 2011; Field Survey. a About 14,400 tractors were estimated to be in Kenya in 2011. b Total arable land under cultivation in Kenya in 2011 was about 53,000 km2. 27 TABLE 6.3. AVERAGE COST OF HIRING Tanzania, most private sector tractor hire services are not TRACTORS FOR FARM economically viable. Countries with lower tractor density include Nigeria and Ethiopia that have the highest trac- OPERATIONS IN KENYA, 2011–12 tor hiring costs, $93/ha and $81/ha, compared to coun- Amount Charged by tries with higher tractor densities such as Kenya ($50/ha), ADC and Other Private Sector Parastatals Actors Burkina Faso ($51/ha), Mozambique ($62/ha), and Tan- zania ($68/ha). Nigeria, Rwanda, and Mozambique have K Sh US$ K Sh US$ high tractor hire rates but also have the lowest tractor den- First plowing 2,500 31.25 4000 50 sities among the pilot countries. Zambia is the exception Second plowing 800 10 2000 25 to this rule, which may be the result of high tariffs rates Harrowing 1200 15 2500 31.25 and fuel costs. Spraying and 800 10 1200 15 herbicides Tractor Imports by Private Sector as a Percentage Source: Kenya Ministry of Agriculture 2011; Field Survey and ABI Kenya of Total Imports (figure 6.4). In Ethiopia, Zambia, Country Report. and Kenya, all tractors were imported by the private sector. In Tanzania, Ghana, and Mozambique, there are spikes in imports of tractors that often reflect the availability of sub- 70 percent of the total number of tractors. Horsepower sidized credit from foreign suppliers and/or cheap tractors. per hectare in Rwanda is 0.01, in Ethiopia 0.04, and in Time series data for tractor imports disaggregated by pri- Tanzania 0.08. vate and public sectors were largely unavailable, however, the overall trend appears to show that tractor imports have Average Cost of Plowing 1 Hectare of Land (US$ started gathering momentum and that the private sector is per Hectare, table 6.3). The cost of hiring a tractor in beginning to show more interest due to the deregulation of Rwanda is about four times higher than in Kenya—$163 this sector by most governments in ABI countries. per hectare compared to $50 per hectare. This disparity in cost can be attributed mainly to the tractor density, and Duties and Tariffs on Tractor Imports as a Per- the availability of the tractor hire services to nonown- centage of CIF Prices. While Tanzania and Zambia ers. Even in countries with strong incentives for farmers are VAT exempt, all countries except for Burkina Faso and entrepreneurs to import tractors, such as Ghana and and Mozambique are zero rated (item is taxable but input FIGURE 6.3. AVERAGE HORSEPOWER PER HECTARE OF ARABLE LAND 200 27 21 8.9 11 12.7 13 4 5.7 7.4 1.3 Rwanda Ethiopia Nigeria Tanzania Burkina Faso Ghana Mozambique Zambia Kenya Sub-Saharan World Africa Source: Field surveys, FAO 2008, FAO and UNIDO 2011. 28 Agriculture Global Practice Discussion Paper FIGURE 6.4. PERCENTAGE (%) OF TRACTORS IMPORTED BY THE PRIVATE SECTOR, 2011–12 100% 100% 100% 100% 60% 59% 42% 10% 10% Kenya Zambia Ghana Ethiopia Mozambique Burkina Faso Rwanda Nigeria Tanzania Source: ABI Country Reports. supply tax rate is nil). The ways the tax exemptions are Small farms typically do not have access to mecha- administered vary, and they are not always administered nization services, although the provision of contract efficiently. Tractor imports and other agricultural invest- machinery services is seen as a way that the benefits of ment goods are exempt from import duties in all countries, mechanization can be delivered to smallholder farmers. with the exceptions of Burkina Faso and Mozambique. Mechanization is increasingly becoming an important Actual import duties plus additional tax payments are means of expanding cultivated areas and is improving the higher than what is on the books. (This implies that cus- timeliness of farm operations. Use of machinery also off- toms agents use their own discretion in applying these sets labor shortages resulting from youth migration. There import charges.) In Ethiopia, tractors are exempt from are varying degrees of government participation in the taxes and other excise duties, on the condition they are tractor markets in the pilot countries. In some countries bought and cleared by customers within six months. the government is designing public-private partnerships that could devolve tractor acquisition and distributions Duties and Tariffs on Spare Parts as a Percent- to the private sector. In some countries, however, the age of CIF Prices. High duties are charged on the government is “innovating” other methods of distributing imports of tractor spare parts and completely knock- and renting tractors to certain segments of the farming down (CKD) parts in all of the ABI pilot countries. population, particularly smallholders. Tractor parts are necessary to maximize the national tractor fleet’s operational life. This policy conflict In Burkina Faso, a government agency has imported between tractor imports being zero rated and tractor tractors from India (in a government-to-government spare parts attracting (sometimes quite high) duties will export-import [EXIM] deal) and provided them to contribute to the relatively short life span compared to a variety of users with a subsidy. In Mozambique, the other tractors in Africa. government has imported tractors from China on easy terms, and although these appear to compete with private Effectiveness of Mechanization Strategies. The sector suppliers, most, if not all, of the government trac- ABI pilot countries are at various stages in drafting tors appear to be destined for use on parastatal operated mechanization strategies. In the past, their strategies of land. The government of Rwanda (GoR) was negotiat- subsidizing tractors and their services have not achieved ing a joint venture with a Korean supplier of power till- the desired results despite heavy public investment. ers, though the deal was not completed as of mid-2013. Agribusiness Indicators: Synthesis Report 29 BOX 6.1. GHANA’S MECHANIZATION SCHEME: AN EXAMPLE OF PUBLIC MECHANIZATION PROGRAM In Ghana, the government has embarked on a program of (AMSECs) to offer tractor-hire services to small-scale farmers mechanization to relieve labor shortages. In the five years across the country. Current demand in the country is primar- between 2004 and 2008, the government imported about 3,000 ily focused on land preparation services, especially plowing. tractors, mostly 30- to 50-kW tractors from India, China, Japan, Experience elsewhere shows that soil cultivation alone, being and the Czech Republic. They are being sold to farmers on a seasonal, is insufficient to provide enough hired tractor hours full-cost recovery basis with a 50 percent down payment and to cover costs. term payments, without interest, over three years (although evi- dence suggests that the credit recovery rate is very low and par- This AMSEC program evaluation was undertaken by the ticularly worrisome for the government). International Food Policy Research Institute (IFPRI) to ascer- tain whether AMSEC enterprises are viable business mod- To qualify for the scheme, farmers have to convince the district els that are attractive to private investors. It has shown the agricultural authorities that they have the necessary land and program to be ineffective, even with the current level of sub- business skills to justify the addition of a tractor. In addition sidy. One of the drawbacks to the program is the low capac- to providing subsidized agricultural machines, the govern- ity utilization, which is the most important constraint to the ment has assisted private enterprises to establish specialized profitability of investment in specialized agricultural mecha- Agricultural Mechanization Services Enterprise Centers nization service provision (IFPRI 2013). TABLE 6.4. NATIONAL STRATEGIES FOR AGRICULTURAL MECHANIZATION Country Year Launched Effective Issues/Status Kenya 1994 No No political will despite privatization agenda Ghana 2004; revised 2008 No Inefficient and unprofitable despite heavy subsidies Tanzania 2006 In progress but no visible impact Rwanda To be launched Draft To be ratified Nigeria 2011 Draft To be ratified and adopted Ethiopia None No strategy as of 2011 Burkina Faso N/A N/A N/A Zambia 2004–15, under the National No Agriculture Policy Côte d’Ivoire 2006 Ongoing Sierra Leone 2006 No To be implemented Uganda 2008 No Draft only Source: Tokida 2011. (The government was asking the Korean investor to put there was no visible government presence. It was lowest up more money, but with less than 50 percent managerial in Nigeria (1.0) and Burkina Faso (1.0), where there is a control.) significant government crowding out of the private sector service providers. This is consistent with earlier findings Private Sector Perception. The score on private sec- that direct government interventions may be crowding tor perception of the business environment in the tractor out the private sector and preventing the emergence of industry (including procurement and service provision) an economically sustainable system for delivering the ben- was highest in Kenya (5.0) and Ethiopia (5.0), where efits of farm mechanization. 30 Agriculture Global Practice Discussion Paper CHAPTER SEVEN AGRICULTURAL FINANCE Most of the ABI countries have been experiencing high rates of growth in finan- cial assets, as well as in the numbers of loans, depositors, and lending institutions. Although there does not appear to be any particular policy that prevented the growth of financial sector services to agriculture, agricultural lending, nevertheless, is found to be lagging when compared to nonagriculture lending. For the agriculture sector to be competitive, capital investments are needed to promote the uptake of improved technologies, such as hybrid seed, fertilizer, and machines to raise production and feed an expanding population. Investments are also required to finance storage, transport, and other postharvest activities that are essential for transforming subsistence produc- tion into the commercial production of agricultural commodities. The limited access that many farmers and agribusinesses have to these financial resources is a serious constraint to agricultural development. The key findings of the ABI studies of finance can be summarized as follows (see also table 7.1): » Access to and availability of finance for agriculture was poor. Only 2.3 percent of rural households in Mozambique had access to formal finance. Similarly, in Ethiopia only 1 percent of rural households had access to formal credit. » Commercial banks’ coverage is low as there was only one bank branch per 100,000 of the rural adult population in countries like Ethiopia. Except for Ghana, all countries had less than two bank branches providing services per 100,000 adults in rural areas. » Agricultural credit is costly. Banks charge nominal interest rates of more than 20 percent, with real rates well over 10 percent to agricultural borrowers. This discourages borrowing and does not lead to investments in the sector. In addi- tion, interest rate spreads in commercial bank lending were found to be high, reflecting inefficiency in the banking sector. » There is growth in the supply of financial services (that is, commercial banks) in the countries studied, but their lending to agriculture is quite low. In six out of nine countries, commercial bank lending to agriculture is less than 10 percent. Even within the small proportion of commercial lending that goes to agriculture, primary farm activities in production and marketing receive very little attention. Agribusiness Indicators: Synthesis Report 31 » Financial service providers state that agricultural sector consists of banks and nonbank financial institutions loans are considered riskier than urban-based, non- and has also seen steady growth in recent years. agricultural lending. Other factors contributing to limited lending are a general absence of registered, Ethiopia still has one of the strictest regulatory frame- titled land with secure tenure, an absence of valu- works, which prohibits the entry of foreign banks into the able fixed assets on the land (such as buildings or financial market. In addition, it has strict requirements storage structures), the limited financial assets of for the import of foreign currency. Foreign exchange borrowers, their inability to put together business accounts, payments, and currency transfers are still sub- plans with bankable loan applications (with cash ject to significant controls and restrictions. These meas- flow projections), and the hesitancy of banks in ures are believed to have seriously affected inflows of some countries to accept movable collateral as a foreign direct investment (FDI) and have potentially neg- guarantee. ative implications for credit availability and the cost of » Nonbank financial institutions like microfinance credit in the long run. institutions (MFIs) and savings and credit coopera- tives exist and are an important source of credit in Given the importance of the rural economy and the agri- some countries, but due to the lack of data, their culture sector for economic growth and poverty reduction, contribution to the agriculture sector is difficult to rural and agricultural finance is now widely acknowl- assess. edged as one of the main frontiers for financial systems » There have been some positive initiatives in re- development (table 7.2). Going forward, improved access cent years in some of the study countries that have to finance in both rural and urban areas will be critical in introduced institutional mechanisms to address enabling the commercialization of agriculture in response credit risks—establishment of credit reference bu- to (i) increased demand for agriculture commodities due reaus, credit guarantee schemes, warehouse receipt to population increases and changes in dietary habits, par- systems, and collateral registries for movable assets. ticularly in urban areas; (ii) climate change impacts that These are examples of instruments and innovative are affecting the supply of agricultural commodities; and practices that facilitate agriculture financing, and (iii) the emergence of new markets for higher value, often it would be worthwhile to monitor their impact on certified and niche commodities that are increasingly agriculture lending over time. demanded by urban consumers (International Finance Corporation 2012, p. 18). Since the early 1990s, several of the ABI pilot coun- tries, including Ghana, Tanzania, Zambia, Rwanda, and Mozambique, have liberalized their financial sec- KEY FINDINGS tors and have lifted barriers to entry for private banks, Commercial Bank Lending to the Agriculture both domestic and foreign. Removal of minimum bal- Sector (figure 7.1). A variety of banks, MFIs, sav- ance requirements, licensing for deposit-taking MFIs, and ings and credit associations, and other nonbank financial liberalization of exchange rates are a few examples of institutions finance agriculture in addition to commercial measures that have facilitated the entry of financial insti- banks. Many of these are unregulated, and it is difficult tutions into the banking sector. As a result, in 2011 Ghana to collect useful data from them. Although they are not had 26 commercial banks, 135 rural development banks, the only source of agricultural loans, formal sector com- followed by many nonbank financial institutions such as mercial banks play an important role in the agricultural credit unions, finance companies, and MFIs operating in finance market, and the proportion of their overall lend- the market. Similarly, since the 1990s, the banking net- ing portfolios that goes to farmers and agribusinesses is work in Mozambique grew rapidly from two state banks an important indicator of credit availability in the sector. and one private bank to the current 18 private commercial In Ghana and Kenya, banks channeled about 6 percent and investment banks, with the majority having an initial of their total lending in 2010–11 to the agriculture sector. infusion of significant foreign capital. Zambia’s financial (Although commercial banks in Kenya are legally required 32 Agriculture Global Practice Discussion Paper TABLE 7.1. FINANCIAL LANDSCAPE OF THE ABI PILOT COUNTRIES Burkina Ghana Mozambique Ethiopia Tanzania Faso Kenya Zambia Nigeria Rwanda Population (millions) 24.9 23.9 84.7 46.2 16.9 41.6 13.5 162.5 11.2 Types of Financial Institutions Commercial banks 26 13 14 47 12 44 19 24 9 Rural and community banks 135 821a MFIs NA 30 30 NA 263 NA 32 NA Savings and credit unions/cooperatives/banks 500 7 5,900 5,000 NA NA 1 416 Source: World Development Indicators (WDI), ABI Country Reports. Note: NA = not available. Agribusiness Indicators: Synthesis Report a Includes microfinance bank. TABLE 7.2. RURAL FINANCE INDICATORS Burkina Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Finance Number of bank branches per 100,000 1.9 1.2 NA 1.6 0.8 2.0 1.3 1.4 NA 5.0 adult population Percentage of commercial banks lending 7 9 3 6 11 14 2 6 9 5 to Ag (3 years) Average lending rates for Ag loans (real rates) 11 10 12 19 −9 5 15 8 17 22 Percentage of nonperforming Ag loans 20 NA 16 NA NA 58c 6 9 10 21 Evidence of credit reference (0–5)a 1.6 0.0 1.5 1.5 0.0 1.5 2.0 3.0 3.0 2.0 Presence of collateral registry agency 2Y, 5N N N N N N N Y N Y for loans Existence of warehouse receipts (0–5)b 1.4 1.5 0.0 0.0 3.0 4.0 0.0 0.0 1.0 3.0 Source: ABI Country Reports. a Evidence of credit reference: 0 = Credit reference bureau (CRB) does not exist; 1 = CRB planned, under design; 2 = CRB underway, but used by small number of financial institutions with limited number of farms/firms covered; 3 = most commercial banks participate; 4 = widespread use with point-of-sale additions (stores/suppliers that sell goods on credit); 5 = most commercial farms and firms covered in reporting system on bank credit histories and point-of-sale on credit. b Existence of warehouse receipts: 0 = no warehouse receipt system (WRS) in place; 1 = WRS under development; 2 = warehouse receipts laws/regulations developed and passed/approved; 3 = warehouse receipts laws/regulations implemented by commercial banks; 4 = warehouse receipts accepted by commercial banks (farmers/traders able to use as collateral); 5 = WRS expands (increased number of banks and certified warehouses, increased grain stored in certified warehouse against receipts issued and used as collateral). 33 c The 58 percent NPL in Tanzania warrants qualification as an outlier, neither representative nor typical of the indicator over time, but rather a reflection of a particularly bad year in terms of price volatility of key commodities in the wake of the financial crisis and severe drought throughout much of the country. FIGURE 7.1. PROPORTION OF COMMERCIAL areas, formal financial service providers are either not BANK LENDING TO AGRICULTURE available in rural areas or offer only limited services. 14% Much of the credit that is available for agriculture-related investment goes to input supply, trading, and agropro- 11% cessing—activities that are largely urban in location and 9% 9% typically not classified as “agricultural loans.” In Mozam- 8% bique, only 2.3 percent of rural households had access to 6% finance for agriculture according to an agriculture census 5% conducted in 2010. Further disaggregation showed that 3% proportionally larger farms had better access to finance 2% than smaller farms. In Ethiopia and Nigeria, only 1 and 2 percent of rural households, respectively, had access to Tanzania Ethiopia Zambia Burkina Mozambique Kenya Faso Ghana Rwanda Nigeria credit. In Ghana, 8 percent of rural households had access Source: ABI Country Reports. to agricultural finance (Ghana Statistical Service 2008). to invest between 17 and 20 percent of their loan portfolio In terms of the sources of formal credit, out of the house- in the agriculture sector, this law is not enforced.) In Nige- holds that received credit in rural areas of Mozambique, ria, where agriculture accounts for 40 percent of GDP, just 14 percent came from financial institutions like commer- 2 percent of commercial bank lending went to agriculture cial banks, agriculture development banks, and coopera- in 2011. In Ethiopia, where agriculture accounts for 45 tives, while input suppliers provided 42 percent (with a percent of GDP, commercial banks’ agricultural lend- number of individual crops, agribusiness firms such as ing represented just 11 percent of their overall lending the British-American Tobacco Company provide input portfolios between 2009 and 2011—about a quarter of credit, raising the proportion of producers who receive what it would be if it were proportionate to agriculture’s credit from their suppliers). Informal sources (self-help role in the economy. Even within the small proportion groups, relatives and friends, and others) were important of the commercial lending that goes to agriculture, pri- and accounted for 29 percent of credit. In Ghana, infor- mary farm activities in production and marketing receive mal sources provided loans to 50 percent of rural house- very little attention. Much of this is directed to off-farm holds, while state banks and private commercial banks agribusiness functions (such as processing, trading, retail advanced loans to 16 percent of households. establishments, and warehousing) and upstream agribusi- ness lending (input supply, importation, and distribution). Agriculture’s inability to attract formal finance generally, and finance from commercial banks in particular, relates Access to Financial Services (figure 7.2). Despite the in considerable measure to how prospective lenders per- growth of commercial banks and other financial institu- ceive risks that are seen as endemic to the sector. Among tions in Africa, access to and availability of finance remain these are risks relating to land tenure, weather variability, notable constraints for businesses, particularly for farmers and nonrepayment of loans, to name just a few.9 In some and small businesses in the agriculture sector. Countries of the ABI countries, the agriculture sector has often been like Ghana are doing better than others in addressing the affected by government intervention in agricultural input challenges by increasing the number of commercial bank and output markets. Unpredictable export and import branches and offering credit products suitable for the sec- bans on certain commodities have led to uncertain- tor. There are five branches of commercial banks per ties in the supply of, and demand for, agricultural prod- 100,000 of the rural adult population in Ghana. Kenya ucts. There are also inconsistencies and unpredictability has also been able to expand financial services with the use of mobile phones that have been shown to increase the number of rural people with access to at least some 9 The various mechanisms described in this section are further elaborated under formal financial and banking services. However, in many the Benchmarking the Business of Agriculture (BBA) project. 34 Agriculture Global Practice Discussion Paper FIGURE 7.2. BANK BRANCHES PER 100,000 OF THE RURAL ADULT POPULATION 5.0 2.0 1.6 1.4 1.3 1.2 0.8 Ghana Tanzania Mozambique Kenya Nigeria Burkina Faso Ethiopia Source: ABI Country Reports 2012; data for Zambia and Rwanda were not available. regarding tariffs, with unpredictable changes having a with an inflation rate of 33 percent12 that shows that their negative impact on the credit market. commercial lending rates must be subsidized. Cost of Agricultural Credit (figure 7.3). Commercial In addition to levying high interest rates, banks in most of credit to the agriculture sector in the countries studied was the countries require borrowers to be able to provide 100 found to have high interest rates, especially in Ghana and percent collateral (usually in the form of “urban” land or Mozambique. In 2010, the average nominal interest rate “titled” land), several years of audited financial records, for agriculture loans in Ghana was found to be 33 percent, and cash flow statements—all backed up by a host of doc- followed by Mozambique at 29 percent.10 In a country like uments whose preparation may require paid service pro- Kenya, which has a relatively advanced agricultural sec- viders. Smallholders rarely have formal land tenure and tor, loans to the agricultural sector attracted higher interest usually are unable to provide the required background rates of 20 to 25 percent. Four out of the seven countries documents, which makes it very difficult to access finance. had nominal interest rates that were higher than 20 percent, There are also additional fees that banks can charge for rendering agriculture loans unaffordable for most borrow- loan applications as well as for granting a loan, which ers. Even when adjusted for inflation (using the consumer adds costs to already expensive credit for most farmers price index),11 real interest rates for most of the countries and agro-enterprises. are high. Ethiopia has a negative real interest rate for 2011, The basis for high interest rates charged by commercial banks is the result of many factors, one of which is high 10 The rates referred to here are average rates. In most of the countries, farmer organizations and small-scale traders tend to be charged the highest rates, while rates are lower for bigger firms. Rates are also lower when there is some kind of guarantee, and they are influenced by the loan amount. 12 This was a one-time high, however, with lower inflation rates of previous 11 The inflation rate for all countries was calculated for 2011, except for Ghana, years at 8.5 percent (2009) and 8.1 percent (2010). The devaluation of the Ethi- where data were from 2010. opian currency in late 2010 contributed to the high inflation rate. Agribusiness Indicators: Synthesis Report 35 FIGURE 7.3. AVERAGE INTEREST RATES ON LENDING FOR AGRICULTURE THROUGH COMMERCIAL BANKS IN 2010 Average nominal interest rates Average real interest rates 33.0% 29.0% 26.0% 23.2% 22.5% 22.3% 18.7% 18.0% 16.8% 15.2% 13.0% 11.0% 10.2% 8.5% 5.3% Ghana Mozambique Nigeria Zambia Kenya Tanzania Burkina Faso Ethiopia –9.0% Source: ABI Country Reports; WDI. FIGURE 7.4. LENDING-DEPOSIT SPREAD South Africa 3% Thailand 5% Ethiopia 6% Mozambique 6% Tanzania 8% Rwanda 9% Burkina Faso 9% Kenya 11% Zambia 12% Nigeria 21% Ghana 22% Source: FinStats 2012 (except for Ghana and Nigeria, for which the source was the Bank of Ghana and Central Bank of Nigeria, respectively). interest rate spreads. Lending-deposit spread (also called was 21 percent. Other ABI countries have lower spreads interest rate spread) measures the gap between the average of 8.2 percent (Tanzania), 9.4 percent (Kenya), 11.5 per- deposit and average lending rates (one of the ABI indi- cent (Zambia), and 6 percent (Ethiopia). In comparison cators). In Ghana, the interest rate spread in 2010 was with South Africa (3.4 percent) and Thailand (4.9 percent), exceptionally high at 22 percent. Similarly, in Nigeria, it these rates are still high (figure 7.4). Such high interest rate 36 Agriculture Global Practice Discussion Paper BOX 7.1. ESTABLISHMENT OF A CREDIT the agriculture sector, are able to design products that are REFERENCE BUREAU IN GHANA well suited to the realities of the sector, and are capable of effectively appraising and evaluating loan applications. The government of Ghana enacted the Credit Report- ing Law (Act 726) in 2007. As a result, the first license was Loan Duration. Even where they are available, agricul- awarded to a private company named XDS Data Ghana tural loans were found to have short repayment periods to operate as a credit reference bureau. The bureau started operation in April 2010 and collects credit information that usually do not exceed one year. Such terms do not on borrowers, making it available for banks and nonbank make capital expenditure attractive to agro-enterprises financial institutions. Two new companies are currently in and commercial farms, where investments have to be the process of getting licenses. All commercial banks and made for a longer time with payback over an extended about 25 nonbank financial institutions in Ghana have period. Those that cultivate tree crops, for instance, have signed up for the services of XDS Data. The company little interest in a loan that must be repaid in less than a currently is discussing service fees and payment modalities year, given that their investment will take several years to with commercial banks. In June 2011, about 8,000 enqui- ries were recorded. So far, banks are only sharing negative yield returns. Access to loans is dependent upon the size information. It is hoped that in the future, the amount of of the firms. International agribusiness firms are able to information shared about clients with good credit histories raise capital offshore. Many domestic small and medium- will increase. In a market where the banks are hesitant to sized enterprises raise capital internally or from family, lend to new borrowers, this mechanism will help reduce friends, or others. This slows the rate of capital formation information asymmetries by providing the lending institu- and constricts the growth of agro-enterprises. tions with credit history information about borrowers. Source: Ghana ABI Study 2011. POLICIES AND INSTRUMENTS TO ENCOURAGE AGRICULTURAL FINANCE The ABI pilot countries have introduced a variety of mea- spreads reflect inefficiency in the banking sector, as well as sures intended to redress the lack of finance that is avail- a smaller depositor base. They also discourage borrowing able to farmers and agribusinesses, which are currently at from the banks. various phases of development. A number of indicators were conceived to capture these reforms, and to demon- Nonperforming Loans. The proportion of nonper- strate how effective these policies and instruments are in forming loans (NPLs) for agriculture-related financing in encouraging agricultural lending and borrowing. These the ABI countries varied from as low as 0.2 percent for indicators will be highly useful to reformers who advo- commercial banks in Ethiopia to as high as 58 percent cate for these measures in their dialogues with high-level in Tanzania in 2011. This was an exceptionally bad year policy makers, making their arguments more persuasive for agriculture with an unusually high level of NPLs that with firm empirical bases. (These indicators are being fur- apparently was not representative of the years before (but ther elaborated in the larger Benchmarking the Business for which data could not be obtained). During this period, of Agriculture project.) (i) companies defaulted on their loans when agriculture exports suffered due to the global financial crisis, (ii) there Credit Reference Bureaus. Financial institutions are was significant price volatility for key commodities such concerned about a general lack of information about as coffee, and (ii) drought was a problem in some areas the credit history of the prospective borrowers. With the in Tanzania. Ghana’s agriculture portfolio was also faced exception of Burkina Faso and Ethiopia, all the ABI coun- with high NPLs at 21 percent (2010), followed by 10.4 per- tries have established credit reference bureaus to address cent in Zambia (2011), and 8.5 percent (2011) in Kenya. the issue of lenders’ confidence. The existence of these The remaining countries did not have NPL data for agri- bureaus is therefore one of the agribusiness indicators. culture loans. The lower proportion of NPLs among Kenyan and Nigerian banks results from the benefits of Warehouse Receipt Systems. Another mechanism having well-trained staff who are knowledgeable about that has received attention in many African countries is Agribusiness Indicators: Synthesis Report 37 the establishment of a warehouse receipt system (WRS), linked to the central depository of the Ethiopian Com- which some countries are promoting as a means for modity Exchange (ECX). In Ghana, the Ghana Grains banks to shift their risk from the borrowers to the assets Council warehouse receipt system allows members of the (mainly grains) that are stored in the warehouses. Such council, including farmers and traders, to deposit grains an approach was recently introduced in Ethiopia, where in a GGC-certified warehouse and be issued a receipt that 16 warehouses in the country are electronically linked can be transferred to other members of the council, or to the central depository of the Ethiopian Commodity used as collateral against loans from GGC partner finan- Exchange (ECX). Ghana has also established a ware- cial institutions.13 Unlike other countries, Tanzania has house receipt system recently that has been promoted by 30 warehouses. the Ghana Grains Council (GGC), and the first grains receipt was issued in January 2013. The GGC Warehouse Leasing. Separate leasing laws are in effect in a number Receipts System allows members of the council, including of ABI pilot countries, including Ghana, Mozambique, farmers and traders, to deposit grains in a GGC-certified and Tanzania. In other countries, such as Zambia, bank- warehouse and be issued a receipt that can be transferred ing laws consider leasing as one of the financial services. to other members of the council or used as collateral Overall, the leasing of agricultural equipment was not against loans from GGC partner financial institutions. In found to be a common practice. Bigger agroprocessing Tanzania, a warehouse receipt system has been in place companies have leased a few machines, but this practice since 2007, and as many as 30 warehouses have been is not yet extended fully to the financing of agricultural given licenses so far. However, experience in Tanzania has machines like tractors, which would benefit the agricul- been mixed. For some cash crops, the system is working ture sector. relatively well. For staple crops like maize, producers are benefiting less due to government restrictions on exports Subsidies and Guarantees. Subsidized loans are and interventions in domestic trade. These actions can available in some countries where agriculture credit is undercut incentives and normal seasonal price patterns provided by government-owned banks. In Mozambique, in staple food crop prices. Management and operation lending rates for subsidized credit for agriculture are 10 of warehouses have also been problematic due to the low percent, while in Ghana, the Agriculture Development management capacity of warehouse operators. Bank offered loans to maize farmers in 2010 at 19 percent, and 22 percent to the rest of the agriculture sector. Usu- Findings show that Ethiopia, Ghana, and Tanzania have ally such government-operated banks are not efficiently WRS, but their experiences are mixed. Ethiopia is con- run, and loans are not provided based on clear and trans- sidered a best practice for the region, in Tanzania WRS parent rules. As a result, such banks do not perform opti- does not work well for staple crops like maize due to gov- mally, and the government is required to intervene from ernment restrictions and interventions. Management time to time to cover the cost of bad loans. Guarantee and operation of warehouses are also problematic due schemes funded by the government or donors have been to the low management capacity of warehouse opera- used, but these instruments, which are in essence indirect tors. Ethiopia has 16 warehouses that are electronically subsidies, are not sustainable over the long run. 13 “Bold Step to Help Grain Farmers.” 2013, January 21. GhanaWeb. http://www .ghanaweb.com/GhanaHomePage/NewsArchive/artikel.php?ID=262689. 38 Agriculture Global Practice Discussion Paper CHAPTER EIGHT RURAL TRANSPORT Transport accounts for a major proportion of overall agricultural costs in Sub-Saha- ran Africa, both for the shipment of farm inputs and the marketing of agricultural produce. The high cost of transport adds to the costs of doing business and is a source of competitive disadvantage for African farm exports. It is attributable to a number of factors including low population density throughout much of rural Africa, limited access to all-weather roads, unmaintained and poor-quality roads, and aging trucking fleets: » Government investment in transport in the ABI pilot countries has increased significantly in recent years, and much of this investment has been augmented by international donors and development agencies. Much of this investment has gone into expanding road networks and building bridges, of which the majority has focused on primary “trunk” roads, to the neglect of construction and maintenance of secondary rural roads that tend to be especially important for the transport of agricultural inputs and outputs. The ABI country studies found the price of shipping 1 ton, 1 km on secondary rural roads to be double or more the cost of shipping on trunk roads. » The transport costs paid by businesses in Africa are among the highest in the world. Roads are the most widely used form of transport in ABI countries. In Kenya, 96 percent of the passenger and goods traffic is by road. In Ghana, roads account for 94 percent of freight, and 97 percent of all traffic movement in the country. In Tanzania, roads account for 80 percent of passenger traf- fic, and over 70 percent of freight traffic. Agricultural commodities are mainly transported from rural areas to secondary towns by smaller 3- to 5-ton trucks, pickup trucks, or vans that operate informally. The poor condition of many of these roads increases shipping costs dramatically and can substantially offset the apparent advantages of countries like Rwanda, where a high proportion of rural people have access to roads, but where 69 percent of the road network is classified as being in poor condition. » Trucking market structures and regulations differ widely among subregions in Sub-Saharan Africa. Many of the transport prices along African corridors seem to depend on the political economy of freight logistics. Transport prices (but not necessarily transport costs) differ widely across subregions and corridors. Most Agribusiness Indicators: Synthesis Report 39 of the ABI countries in East and Southern Africa BOX 8.1. REGISTRATION AS A TRUCKING have competitive and only lightly regulated nation- ASSOCIATION IN ETHIOPIA al trucking industries. Market entry is easy in most of the countries. Getting registered and obtain- The Ethiopian Road Transport Authority issues guidelines for the registration of the trucking associations and for the ing an operating license for a truck takes less than regulation of the trucking industry. To be registered as a seven days in six of the nine countries. However, trucking association, the association must have a minimum limited access to finance among smaller trucking of 20 trucks of 40 tons each and must establish its legal- countries imposes a de facto barrier to entry, and ity as a purely business entity. On average, it takes about the existence of trucking cartels in some countries seven days to register a truck and about 21 days to register a restricts access to some popular routes. trucking association from the time an application is submit- ted for registration. Once operational, the association must In more sparsely populated countries like Mozambique submit monthly reports of its activities to the government or the license to operate will be revoked. Notwithstanding and Zambia, the per capita costs of investing in roads can these regulations and requirements, interviewees believe be very high. This can impose a major financial burden that there is freedom of entry and exit into the trucking on the country. This is characteristic of Sub-Saharan market, although one major external barrier is the prohibi- Africa as a whole, where population density is 34 peo- tive custom tariffs and taxes on imported trucks and spare ple per square kilometer, compared to East Asia and parts. the Pacific where it is 131 people. Africa also has a large On imported trucks, these impositions consist of custom number of landlocked countries that are distant from the duties, excise tax, transaction taxes, surtaxes, value-added nearest seaports which is a major structural impediment tax (VAT), and in some cases, withholding taxes. It is esti- to transport throughout much of the region. Inadequate mated that these taxes can add up to more than 70 percent seaport capacity relative to demand for shipping services of the CIF price of the imported trucks and parts depend- ing on the model. As a result, a good percentage of the and nonimplementation of regional trade and transport freight-carrying vehicles in Ethiopia are not in very good agreements likewise increase the cost of providing road condition, a situation that in part accounts for the high access. transportation cost in the country. Source: Ethiopia ABI study 2012. KEY FINDINGS (TABLE 8.1) Number of Days to Register a Truck and Obtain an Operating License. It takes seven days or less to was 5¢ and 6¢, respectively—the lowest price among the register and obtain an operating license for a truck in ABI countries. In Kenya the price was 15¢. Nigeria has Ghana (5), Mozambique (5), Tanzania (6), Ethiopia (7), the highest freight rates from primary to secondary mar- and Rwanda (7). In Kenya it takes 29 days, the longest kets and secondary to terminal markets. Global compari- wait period among the ABI countries. Cartels along the sons show that on average, transport prices along major most profitable major routes require prospective trans- transport corridors are more than two thirds less in Paki- porters to make informal payments in order to operate.14 stan than in major corridors in Africa. The cost of trans- porting goods along the Tema-Ouagadougou corridor COST OF TRANSPORTATION. is 35 percent higher than the average for other African Transport prices along the main trunk roads were more and Asian corridors, such as Dacca-Chittagong, Laem similar between the ABI countries than transport on sec- Chabang-Vientiane, Durban-Nelspruit, and Maputo- ondary roads, which varied more widely. The cost of trans- Nelspruit.15 portation is measured in terms of the U.S. dollar price of moving 1 ton, 1 km. In Burkina Faso and Nigeria this cost Transport price differences were more pronounced among rates for secondary roads. In Nigeria and Ghana, The ABI study did not focus on barriers to entry in intraregional trucking, 14 where borders are crossed. 15 Analyzed by Nathan Associates and the USAID 2011. 40 Agriculture Global Practice Discussion Paper Agribusiness Indicators: Synthesis Report TABLE 8.1. RURAL TRANSPORT INDICATORS Burkina Indicators Mean Faso Rwanda Mozambique Ethiopia Tanzania Nigeria Kenya Zambia Ghana Rural transport Rural Access Index (%) 23 24 36 24 10 24 20 32 17 24 Percentage of road network in poor 34 9 69 32 34 29 34 34 48 18 condition Logistic Performance Index infrastructure 2.14 2.40 1.88 2.04 2.22 2.41 2.27 2.16 1.83 2.05 quality (0–5) Cost of transportation ($/MT/KM)— 0.10 0.05 0.09 0.10 0.11 0.10 0.06 0.15 0.11 0.10 main routes Cost of transportation ($/MT/KM)— 0.26 0.14 0.37 0.23 0.24 0.13 0.47 0.24 0.13 0.35 secondary routes Time (days) to get truck 13 6 7 5 7 6 28 29 21 5 registered/operating license Ease of entry into trucking (0–5) 3.7 3.5 4.0 4.0 4.0 3.6 3.5 2.5 4.0 4.0 Source: ABI Country Reports. 41 FIGURE 8.1. PERCENTAGE OF THE ROAD CLASSIFIED AS BEING IN POOR CONDITION 69.0% 48.4% 34.0% 34.0% 33.8% 32.2% 28.6% 18.1% 8.7% Rwanda Zambia Ethiopia Nigeria Kenya Mozambique Tanzania Ghana Burkina Faso Source: World Bank 2011. the rates were as high as 47¢ and 35¢ per ton per kilom- out much of Africa. Poor-quality roads discourage trans- eter, respectively, while in Zambia, rates were much lower port and transportation services from expanding into rural at 13¢. In the case of Ghana, the transport along second- areas and make delivering commodities from the farm gate ary roads is dominated by old 5- to 10-metric-ton capac- to the market slow and costly. This is often the result of a ity trucks that each carry between 50- and 125-kg bags, lack of resources invested in periodic maintenance. The which raises per unit shipping costs. Poor-quality roads ABI pilot countries varied widely in terms of this indica- lead to high operating and maintenance costs which then tor. Burkina Faso has the lowest proportion of its road net- makes transportation very costly for traders and for agri- work classified as poor at 8.7 percent, followed by Ghana at businesses. 18.1 percent. Rwanda has the highest at 69 percent, despite having good road access as a result of its relatively Rural Access Index. This indicator measures the per- high population density. Rwanda is followed by Zambia at centage of rural people living within 2 km—typically a 48.4 percent, and Tanzania with 28.6 percent of its roads walk of between 20 and 25 minutes—of an all-season rated as poor. road as a proportion of total rural population. It does not account for population density. Countries in which this Logistics Performance Index. The Logistics Per- proportion is higher typically have transport networks that formance Index (LPI) is an interactive benchmarking effectively enable farmers to sell the commodities they tool created to help countries identify the challenges and produce in local markets, which is an essential element opportunities they face in their performance on trade logis- of agricultural commercialization. In ABI countries the tics and what they can do to improve their performance access to rural roads in Rwanda (36 percent) and Kenya (World Bank 2012b). The LPI 2012 allows for compari- (32 percent) is better than in Zambia (17 percent) and sons across 155 countries. The LPI is the weighted average Ethiopia (10 percent). of the country scores on six key dimensions: (i) efficiency of the clearance process (that is, speed, simplicity, and Percent of Road Network Classified as Being in predictability of formalities) by border control agencies, Poor Condition (figure 8.1). The quality of existing including customs; (ii) quality of trade- and transport- roads is a major concern for the agriculture sector through- related infrastructure (for example, ports, railroads, roads, 42 Agriculture Global Practice Discussion Paper FIGURE 8.2. FIGURE COMPARISONS ACROSS COUNTRIES: QUALITY OF INFRASTRUCTURE (SCALE OF 0 TO 5) South Africa 3.79 Turkey 3.62 China 3.61 Tanzania 2.41 Burkina Faso 2.40 Nigeria 2.27 Ethiopia 2.22 Kenya 2.16 Ghana 2.05 Mozambique 2.04 Rwanda 1.88 Zambia 1.83 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 Source: World Bank 2012b. information technology); (iii) ease of arranging competi- within the scheduled or expected delivery time. In com- tively priced shipments; (iv) competence and quality of paring quality of transport infrastructure, most of the ABI logistics services (for example, transport operators, cus- countries perform poorly (figure 8.2). Ratings for all coun- toms brokers); (v) ability to track and trace consignments; tries are below 2.5 on a five-point scale. In comparison, and (vi) timeliness of shipments in reaching destination South Africa rates 3.79, higher than Turkey or China. Agribusiness Indicators: Synthesis Report 43 CHAPTER NINE SUMMARY AND CONCLUSION COUNTRY RESULTS GHANA Access to and Availability of Certified Seed. Relatively few farmers who culti- vate field crops have access to improved seeds, and seeds of high-yielding hybrid crops in particular. Seed supply is constrained by inadequate production of both breeder and foundation seed. A number of recent government initiatives are promising. Seed imports are now open to private companies, although only a few of them have begun to import hybrid maize seed. The number of certified seed growers is increasing, and private sector firms are exploring the possibility of producing hybrids and open- pollinated varieties for domestic use and for export to regional markets as well. Availability of and Access to Fertilizer. In the five-year period from 2006 to 2010, Ghana’s fertilizer imports increased from 189,878 to 308,786 metric tons, an increase of more than 60 percent. Fertilizer consumption has also increased to 40 kg/ha, which is just 10 kg/ha short of meeting the Abuja Declaration on fertilizer. The nutri- ent output ratio of 2.6 indicates that fertilizer use is profitable for Ghanaian farmers, particularly as the grains they produce command higher prices in local markets. In the 1990s, the government monopoly over fertilizer imports and distribution was abolished, and the resulting liberalized environment made market entry relatively easy for import- ers, distributors, and retailers, who are able to obtain a license in a short period of time. With eight major importers, between 35 and 50 distributors, and as many as 4,000 retailers, the market is already quite robust, and the density of agrodealers in some regions is high. Access to Farm Machinery. Agricultural production in Ghana is labor intensive, with little use of machinery. In recent years, however, the demand for tractors has been on the rise, owing to an expansion in the amount of land that is cultivated by large commercial farms. As a result, a number of importers and distributors of well-known tractor brands have come into operation. Tractor services are utilized for land prepa- ration and for other agricultural activities such as planting, postharvest processing, Agribusiness Indicators: Synthesis Report 45 and hauling. This trend toward mechanization has gener- sector exist. Budget allocations fall short of the CAADP ally not applied to smallholders, who are unable to afford compact that set targets for allocating at least 10 percent the purchase of tractors that have high financing require- of government expenditure on agriculture. ments. The timely availability of spare parts has also been problematic. Some are concerned that the government’s involvement in the agriculture machinery market could KENYA crowd out the private sector. Access to and Availability of Certified Seed. Kenya’s farmers obtain seed of most commodities (except Access to Agricultural and Agro-enterprise maize and rice) from the informal sector. About 40 per- Finance. Agricultural finance in Ghana is difficult to cent of farmers’ maize seed comes from the formal public obtain, and where it is available, it is expensive. Agricul- seed companies, and another 20 percent is supplied by pri- ture receives substantially less commercial bank lend- vate seed companies. The vast majority of hybrid maize ing than other sectors—barely 6 percent. Interest rates seed used in the country—90 percent—comes from the of commercial bank loans are typically in the range of Kenya Seed Company. Although Kenya’s seed laws are 25 to 40 percent. Many providers of financial services are under review, among African countries Kenya has one of hesitant to provide loans for agricultural purposes owing the most advanced seed laws in place to guide and regu- to land tenure issues, a history of nonrepayment of subsi- late the use of certified seed. Because the Seeds and Plant dized loans, and the overall riskiness of investing in rain- Varieties Act incorporates the International Seed Testing fed agriculture. Loan guarantee funds are being designed Association’s rules of seed testing and the OECD’s seed and implemented, and insurance products are being certification schemes, Kenya can trade seed with most tested. A warehouse receipt system is being developed, other countries in the world that belong to these interna- and more financial institutions are becoming members of tional seed certification organizations. This privilege has the credit registry bureau for increased transparency and enabled Kenya to import seed in areas where domestic information sharing. supplies are insufficient. For maize, although adoption estimates vary by year and ecological zone, the use of Cost and Efficiency of Transport. Though Ghana’s hybrid and improved seed has increased over the years. transport sector is relatively well developed according to While most farmers in Kenya used hybrid seed, farm the country’s Rural Access Index, rural and feeder roads survey results showed that only about 25 percent of the that are important for agriculture are not always in good land area was planted to retained hybrid seed. Based on a operating condition. Despite increases in funding for road weighted average of all adopters by ecological zone, it is maintenance, a number of serious challenges need to be estimated that the adoption of hybrid maize could be as overcome. One of them is overloading among commer- high as about 82 percent, one of the highest in Sub-Saha- cial vehicles, a practice that has caused road quality to ran Africa. It is further estimated that on average certified deteriorate, adding to the cost of transporting agricultural and improved seed is used on about 70 percent of the land goods. under maize and 54 percent of the land under wheat. This was made possible (particularly hybrid maize adoption) by Agribusiness Policy Environment. Changes in agri- an extensive network of agrodealers, developed through cultural and private sector development policy in Ghana the Citizens Network for Foreign Affairs (CNFA) and the have made for a more enabling environment for the pri- Alliance for a Green Revolution in Africa, which reduced vate sector and market development, although some in the distances that farmers had to travel to obtain seed. the private sector express concerns over specific policies such as the subsidy on fertilizer and mechanization. The Fertilizer. In the past decade, fertilizer application rates new Private Sector Development Strategy (PSDS II) in Kenya have gone up significantly. This increase has emphasizes the need to foster public-private dialogues. been attributed to the liberalization of the fertilizer mar- Many associations focusing either on specific commodi- ket, which essentially eliminated price controls and import ties or subsectors advocating the interests of the private licensing quotas, removed foreign exchange controls, 46 Agriculture Global Practice Discussion Paper and phased out external fertilizer donations, which had the Agricultural Finance Cooperation (AFC). The AFC disrupted commercial operations. Another significant fac- is now the leading government institution mandated to tor in Kenya’s greatly enhanced fertilizer consumption provide credit for the sole purpose of developing agricul- was the development of a private agrodealer network ture. Only a small proportion of loans from commercial through a program by CNFA and Agricultural Market banks are allocated to agriculture (about 5.3 percent in Development (AGMARK). The program provided train- 2010 and 5.6 percent in 2011), far below what the gov- ing and helped dealers obtain credit for the private com- ernment has mandated, indicating serious underfinancing mercial provision of agro-inputs to farmers. The greater of the agricultural sector. However, government efforts to accessibility of agrodealers made it far easier for farmers increase the accessibility of formal financial services have to obtain a bulky input such as fertilizer. Of particular started to yield dividends. The expansion of commercial interest is the fact that fertilizer consumption has grown banking networks since 2005, following enactment of a among Kenya’s smallholders, who account for more than government policy mandating banks to improve their 70 percent of the agricultural output in the country. Gov- services in rural areas, has been highly significant. Lack ernment importation and distribution of fertilizer invari- of collateral remains an issue in agricultural lending, par- ably undercuts the agrodealers and could once again ticularly for smallholders. Despite having the necessary result in market distortions and failures. infrastructure—a large network of warehouses—Kenya has yet to establish warehouse receipt financing, so this Mechanization. Small-scale farmers in Kenya, unlike potential mechanism for farmers to use stored commodi- their counterparts in most other African countries, have ties as collateral for loans is yet to be harnessed. as a matter of necessity embraced the use of tractors for land preparation. Use of tractor-drawn implements is par- Transport. Approximately 96 percent of the passen- ticularly high in the high-potential agricultural areas of ger and goods traffic in Kenya is through the road net- the Rift Valley and Western Lowlands, where heavy rains work, and only about 4 percent is by rail, air, sea, or lake. sometimes result in waterlogged and caked soils, which About 70 percent of Kenya’s classified road network is are difficult and labor intensive to prepare for planting estimated to be in good or fair condition and maintain- using simple hand implements. The present level of agri- able; the remaining 30 percent requires rehabilitation or cultural mechanization in Kenya ranges from 95 percent reconstruction. Yet very few rural routes (about 5 percent) on large farms to as little as about 4 percent on some small are judged to be in good condition. High taxes and tariffs holdings. On the whole, it is estimated that only about 30 continue to deter transporters from importing new trucks. percent of the operations on small farms are done using During the study period, it was estimated that the numer- tractors and motorized equipment. ous taxes on a newly imported truck could add up to more than 50 percent of the CIF price. Although the govern- Tractor importation is duty free in Kenya, but imports of ment has built more roads and improved the Rural Access spare parts are subject to duties and a value-added tax of Index, Kenya’s high transport costs account for about 35 about 16 percent, which has serious implications for the percent of the total logistics cost along the Northern Cor- life span of tractors and accounts in part for the fact that ridor in Kenya. at any time about half of the tractors in Kenya are out of commission. The market for tractors is largely in the hands of private tractor dealers and is very competitive. TANZANIA Stakeholders confirmed the absence of the public sector Seed. Since the liberalization of the seed sector in Tanza- in buying and selling tractors in the country. nia, there has been an increase in availability of improved seed for farmers. Tanzania enacted a new Seeds Act in Finance. The financial sector in Kenya consists of a large 2003, repealing the Seeds Regulation of Standards Act of number of formal, semiformal, and informal financial ser- 1973. The 2003 Act encourages private sector seed pro- vice providers. Formal provision of agricultural credit to duction and distribution in the country and has introduced farmers is dominated by a specialized agricultural bank— measures to ensure that the seed produced and imported Agribusiness Indicators: Synthesis Report 47 meet a set of required standards. As a result, there has Finance. In 2011, 15.4 percent of the commercial bank been growth in the number of private companies operat- lending portfolio consisted of agriculture-related loans. ing in the market, and during the 2010–11 season, nearly Although competition has driven some financial provid- 80 percent of the total commercial seed was supplied from ers to be more efficient, the supply of credit remains lim- the private sector. On the demand side, however, farmers ited, especially in rural areas. The National Panel Survey lack awareness about how the use of improved seed leads (2008) found that only 6.5 percent of rural households to higher yields. The seed-to-grain price ratio for maize have access to credit. Bank interest rates on loans to agri- crop using hybrid seed is 10:1, which is considered high. culture are high, and the commonly offered short-term As a result, many farmers cannot afford to buy seeds. loans are not attractive for farmers or agribusinesses. Fur- ther, in the absence of titles to land, smallholders have Fertilizer. Similar to the seeds sector, the supply and little if any collateral to offer. The warehouse receipt distribution of fertilizer in Tanzania is primarily in the system and a variety of credit guarantee programs have hands of the private sector. The government’s National been introduced to address these constraints, though with Agriculture Input Voucher Program (NAIVS) has been limited reported success to date. Initiatives are also under- instrumental in building the capacity of the dealers and way to establish a credit reference bureau and collateral in facilitating their expansion into rural areas. The private registries for movable assets. sector finds the policy environment quite conducive to doing business, and over the years, fertilizer imports with Transport. Over the years, government investments zero-rated duties have increased. Yet very few farmers in in Tanzania’s road network have increased substantially. Tanzania have access to fertilizer. Even with the NAIVS Yet, in 2010, only 24 percent of rural people had access program, significant numbers of farmers have difficulty to an all-season road. Poor connectivity of rural roads paying for the cost of fertilizer that is subsidized. Another to regional and trunk roads and limited maintenance of factor contributing to low use is limited practical informa- rural road networks have been serious constraints for the tion among farmers about the proper agronomic uses of agriculture sector. Only about 43 percent of trunk roads fertilizer. This is also a result of poor extension. Despite are paved, while less than 2 percent of district and feeder the increase in the number of agro-input dealers, many roads are paved. This has resulted in higher marketing farmers still need to travel long distances to buy fertilizer costs for agricultural inputs and outputs. The policy envi- because dealers are mainly based in district headquarters. ronment for transport business is favorable with an open market and relatively easy entry into the sector. Trans- Mechanization. The Mechanization Department of porters find the cost of credit to be one of the major con- the Ministry of Agriculture, Food Security, and Coop- straints for their businesses. eratives (MAFC) estimated that in 2010, there were 8,466 tractors in use in Tanzania, in a country with 11.5 million Policy Environment for Agribusinesses. The pri- hectares of arable land. Based on this estimate, there are vate sector generally has a positive view of the policy only seven tractors per 100 km2 of arable land in Tan- environment for agribusinesses in Tanzania. Incentives zania. In Tanzania, 92 percent of farmers still use hand for agriculture investors that include zero-rated duty on hoes and farm a few acres of land, with just 5 percent farm inputs including fertilizer, seeds, and tractors, and of farm households using tractors. Starting in 2009, there zero-rated VAT on agricultural exports are encouraging. has been an upward trend in the number of tractors being Tanzania is a signatory of the CAADP compact that imported. Since the sector opened up, private companies calls for the government to allocate at least 10 percent have set up distributorships of various tractor brands, and of the total annual budget to agricultural develop- there are 10 or 12 major importers of tractors in Tan- ment. Tanzania has not met this target so far, but over zania. Large farms are their main clientele, in addition the years, the agriculture budget has been growing in to farmer groups or savings and credit cooperatives that both nominal and real terms. Nonetheless, policies for have access to subsidized financing from public banks or the sector are not always consistent. Periodic export donor-financed programs. bans on a number of crops, chiefly maize, have led to 48 Agriculture Global Practice Discussion Paper major disincentives for producers and businesses. Duties Mechanization. Tractor use remains low, despite Zam- on imports and exports that affect the agriculture sector bia’s relatively well-developed agribusiness industry, the also change from time to time, with inadequate informa- positive enabling environment for the private sector, and tion available to private firms on which to base planning some promising initiatives linking smallholders to agri- or investment decisions. business firms through vertically integrated outgrower programs. There is no reliable figure for the total number of working tractors in Zambia, but it is estimated to be ZAMBIA around 6,000, or about 21 tractors per 100 km2 of arable Seed. It is estimated that around two thirds of the maize land. In comparison to other pilot countries, this num- area is planted with certified seed (specifically, hybrid seed), ber seems high, but tractors are used mainly by large and whereas an estimated 20 to 30 percent of the wheat and corporate farms, while smallholders’ access to mechani- soybean area is planted to certified seed. Zambia’s seed zation services is still extremely limited in most parts of sector is fully liberalized. The private sector plays a major the country. Small- and medium-scale farmers mainly use role in seed production and exports. Zambia is one of the hand hoes and animal traction. All tractors in Zambia largest seed exporters in Africa; aside from the domestic are imported. Due to the country’s landlocked geography, market, it exported a recorded total of 17,891 tons of retail prices can be higher than in other countries. certified seed to other African countries in 2011. Despite the positive policy environment and growth for the sector, Finance. Zambia has a sizable number of private com- fewer than 40 percent of small- and medium-scale maize mercial banks and nonbank financial institutions operat- farmers used hybrid seed during the 2009–10 season. If ing in the market. Total agricultural lending as of March a company wants to introduce a new variety of seed in 2012 amounted to $415.8 million or 16.5 percent of all Zambia, it can still take up to two years to obtain official credit from commercial banks and microfinance institu- approval. Another issue of current concern to the private tions combined. Private firms in Zambia still do not bor- sector is the presence of counterfeit seed in the market. row from banks and instead rely on retained earnings or informal financing for long-term investments. Banks Fertilizer. In 2010–11, total fertilizer use in Zambia admitted they generally do not loan for greenfield invest- was 300,414 tons, which has increased steadily from ments in agriculture and usually deal only with trade- the preceding years. The government-financed Farmer related businesses. Banks also state that a culture of Input Support Programme (FISP) has contributed to nonpayment in Zambia, coupled with the risks associated this growth by financing 61 percent of the fertilizer used with financing agriculture, do not incentivize the banks to in the country. There is also steady growth in the com- offer services to the sector. As a result, agribusinesses have mercial farm sector, equal to about 15 percent per year difficulty accessing commercial credit. between 2007 and 2011. Fertilizer companies selling to commercial farmers point to the large increases in com- Transport. Because Zambia is landlocked, costs of haul- mercial wheat, soybean, sugar, barley, and maize produc- ing agricultural inputs and goods are high. These high tion as drivers of increased demand for their product. costs in turn raise production costs and reduce competi- On a national level, despite the expansion of FISP and tiveness in foreign markets. It has made major progress in increase in total fertilizer use, only 39 percent of small- constructing a network of trunk roads and in linking the holders use inorganic fertilizer. There are nine major provincial capitals to Lusaka and Lusaka to main inter- importers, but domestic fertilizer distribution is handled national border crossings. The country has also success- mainly by district governments and cooperatives. The fully operated a road fund that provides stable allocations limited competition has led to complaints from firms of resources to the sector. It is one of the few countries excluded from the program, but with the introduction of in the region with a road sector budget surpassing what the e-voucher program, they are hopeful that the role of is needed to maintain the main road network and ade- the private sector may increase, not only in importing but quate to address the rehabilitation backlog. Regardless in distributing fertilizer in rural areas. of increased investments in the transport sector, access Agribusiness Indicators: Synthesis Report 49 to roads is poor, with only 17 percent of people in rural Private sector seed multiplication is expanding, though areas living within 2 km of an all-season road. Trucking slowly. The Basic Seed Production Unit (USEBA), firms say that police interference is not a major problem which is a parastatal subsidiary of the national agricul- and point to the fact that many roadblocks have recently tural research institute (IIAM), produces most basic seed been removed, making it easy to move goods around the (except for rice seed), but volumes are typically too low country. There are few barriers to entry in the trucking for sufficient multiplication and wide-scale distribution to industry. Government registration and licensing are not farmers. Many assert that demand for improved seed of major obstacles, though the process can take time. maize and a wide range of other field crops is very lim- ited. Seed costs are reportedly high (five times or more Policy Environment for Agribusinesses. Economic than the cost of the grain produced), and improved seed reforms of the early 1990s have improved the business is often distributed through donor- or government-funded climate for the agricultural sector in Zambia. The priva- projects and programs. tization of several parastatals made agriculture relatively free of major policy distortions and increased private sec- Fertilizer. The vast majority of fertilizer was applied to tor participation in input supply, finance, and transport leaf tobacco (51 percent) and sugarcane (42 percent) in the services. Some private firms admit that the policy envi- 2010–11 cropping season in Mozambique, with vegetables ronment can be unpredictable when there is a change in grown in peri-urban areas perhaps receiving more fertil- government, but these companies still consider that gov- izer in the aggregate than other food crops. Knowledge- ernment has done a commendable job of consulting the able observers report that applying fertilizer (urea, NPK) private sector, even if it could sometimes do a better job of to maize is unprofitable for most smallholders in most rural listening and taking private sector concerns into account. areas of Mozambique. Nutrient output ratios are one mea- With respect to the consultative process between the pri- sure of the feasibility of using fertilizer on maize. They were vate sector and the government, no single apex body for 8 to 14 in the postharvest period of 2011, which means agribusiness exists in Zambia. As in other countries where that fertilizer is too expensive to buy and apply relative to apex organizations do not exist, commodity associations the low maize prices prevailing in maize surplus zones. have been effective, focusing more narrowly on represent- ing the interests of the participants in specific value chains. Agricultural Mechanization. Some animal traction is In Zambia, the national farmers union (ZANFU) is seen used in the south but is virtually nonexistent in the north- as having been influential in this capacity. With respect ern half of the country, partly because of trypanosomia- to the government budget, Zambia is a signatory to the sis and also because cattle were decimated during many Comprehensive Africa Agriculture Development Pro- years of civil war. As with the provision of other produc- gram Compact and has been spending just about 6 per- tivity-enhancing inputs in Mozambique, the provision of cent of its total budget on agriculture. This level is below farm equipment is often subsidized. Mozambique had the 10 percent target agreed upon under the 2003 Maputo an estimated 12.6 to 14.2 tractors per 100 km2 of arable Declaration, but it has been quite constant over the years. land from 2000 through 2010. The demand for tractors has increased on medium to large farms that are strongly commercially oriented. Expansion of agricultural produc- MOZAMBIQUE tion should drive a vibrant agricultural sector led by the Seed supply is constrained by inadequate pro- private sector, but government intervention in importing duction of breeder seed and foundation seed. and distributing tractors on subsidized terms with non- The low use of certified seed for basic grains, particularly transparent selection criteria bodes ill for the emergence maize and rice, causes yields of rain-fed crops such as of a private sector-led agricultural machinery servicing, maize to be lower than yields in most other countries in maintenance, and custom-hire capability. Sub-Saharan Africa. Improved seed was used only on an estimated 14 percent of the 2010–11 rice area, largely in Agricultural Finance. In Mozambique, access to agri- irrigated production zones. cultural finance is difficult. Even when loans are available, 50 Agriculture Global Practice Discussion Paper they are expensive. Agriculture receives much less atten- policy. Government regulation and taxes are considered tion than other economic sectors from commercial banks; excessive. The legal and regulatory framework affecting lending to agriculture was a mere 6.5 percent in 2010, agriculture is perceived as not fully transparent. Some down from 9.4 percent in 2008. Many financial service private firms fear that the government’s interventions in providers hesitate to lend to the agricultural sector due input markets and tractor distribution, along with signs to a long history of nonrepayment of subsidized loans, that it may reenter cereal markets, could undercut private thorny land tenure issues, and the risky nature of rain-fed sector competitiveness. Mozambique’s budget for agri- agriculture. Warehouse receipt systems essentially do not culture as a percentage of total budgetary expenditure exist. Several banks do allow agricultural machinery (that ranged from 5.4 to 5.7 percent from 2007 to 2009 and is, movable assets) to be used as collateral, but at deep dis- included half of district development funds. Producers’ counts to their estimated value. Many foreign-owned agri- modest share of the cashew export price (39 percent from businesses and Mozambican/foreign joint ventures access 2006 to 2009) is a disincentive to replant the aging stock finance offshore in South Africa, Europe, or Asia, so they of cashew trees. Mozambique currently exhibits a reason- are less constrained by the shallow Mozambican finan- ably strong investment climate for foreign investors but cial sector, which is slow to lend to agribusiness through limited support for the vast majority of domestic produc- domestic financial intermediaries. ers and rural agro-enterprises. Transport. Along trunk roads in the Beira Corridor, transport is efficient, competitive, and reasonably low BURKINA FASO cost, but transport beyond trunk roads is costly. The Rural Seed. INERA, the Agricultural Research Institute in Access Index for Mozambique is between 24 and 32 per- Burkina Faso, has developed improved varieties of cereal cent. The numerous rivers, tributaries, and streams cut- crops, pulses and oilseeds, and cotton. The multiplication ting east to west make rural transport costly and render of improved staple crops for seed has increased markedly some rural roads impassable (often flooded) during cer- over the past decade so that the use of improved seeds tain months of the rainy season. Field surveys suggest that by smallholder farmers who do not grow cotton remains transport costs are a major component of delivered input limited. Only half of the demand for improved maize costs in rural areas and in marketing of agricultural pro- seed is satisfied (through local multiplication), while only duce. Rural and feeder roads that are important for agri- a quarter of the demand of improved varieties is satis- culture are often not in good operating condition. Despite fied. It should be noted that most of the maize seed multi- increases in funding for road maintenance, several chal- plied and almost all seeds of improved rice are distributed lenges need to be overcome. The overloading of commer- with heavy government subsidies and donor programs. cial vehicles has caused road quality to deteriorate and The government controls the production of basic seed, raised the cost of transporting agricultural goods. Flood- the six private seed companies and many seed produc- ing in many low-lying areas near rivers leads to major ers (approximately 3,000, organized in seven coopera- damage to roads as well. Domestic transporter unions tives) who take charge of all the multiplication of certified complain of unfair competition from foreign trucking seed. The seed supply does not meet demand. Some seed fleets, which operate in Mozambique with few controls, imports are allowed, but most (except rice seed) seem of but knowledgeable observers argue that fostering regional an informal nature from neighboring countries and thus competition in transport is the best policy. are not controlled. Also, imported varieties are not subject to a screening or routine testing. The medium-term trends Agribusiness Policy Environment. This is consid- in national yields are stagnant, and expected increases in ered reasonably conducive to private sector investment, performance from the use of improved seeds and subsi- although much investment in commercial agriculture dized fertilizers imported have not materialized. remains foreign. The government’s announcement of foreign exchange controls in mid-2011 was unexpected as Fertilizer. During the 1990s and much of the 2000s, the private sector was not consulted about the change in fertilizers were distributed using subsidized credit to Agribusiness Indicators: Synthesis Report 51 cotton farmers. The use of fertilizers on cotton was Access to Financing for Agriculture. In Burkina Faso, clearly linked to a system of input supply and market- agricultural loans are discouraged by the fact that only a ing of a business culture in which the parastatal cotton small part of agricultural production is irrigated, and there (SOFITEX) and, thereafter, two other cotton companies is only one long growing season, characterized by variabil- can recover the credit when the seed cotton is purchased ity. Total loans to agriculture include 9.2 percent (2009), from producers. Fertilizer suppliers estimate that at least 10.2 percent (2010), and 7.7 percent (2011) of the total 85 percent of fertilizer in Burkina Faso was imported outstanding to agribusiness, including production activi- and distributed to cotton farmers before the introduction ties (agriculture, hunting, forestry and logging, fishing) and of the fertilizer subsidy program in 2008. In 2010–11, the manufacture of food products. The total agricultural almost a third of all fertilizer was applied on crops other production has received less than half—4, 3.1, and 3.7 per- than cotton. Taking into account the diversion of fertil- cent, respectively. The nominal interest rate for short-term izer cotton to corn (for which there are no reliable esti- credit is 12 to 14 percent in the stable monetary environ- mates), this proportion has probably climbed. The poor ment of Burkina Faso where inflation was 2.8 percent in quality of roads and strong illegal taxes imposed along 2011. Credit to the cotton sector has never exceeded 8 per- trade corridors also serve to increase the cost of fertilizer cent (nominal) per annum, and this probably influenced distribution. Fertilizers were subsidized up to 28 percent the perceptions of farmers and companies regarding rea- of the cost of urea and 23 percent of the cost of NPK in sonable interest rates. Rural areas of Burkina Faso are not 2011, although the grant is officially 50 percent. There well served by the banking system because there are only is an active and growing association of agro-input deal- 1.16 commercial banks to 100,000 rural residents. Most ers, many of them turning away fertilizer to further spe- people in rural areas have no access to financial services, cialize in the trade of agrochemicals and vegetable seed although there are several institutions or MFIs (COR, where prices are not subsidized or controlled. The sub- PAMF, URC-Nazinon, URCCOM) that target customers sidy program appears to have contributed to increased with more rural bank business. A warehouse receipt system use of fertilizers since low levels of use in 2008 and is increasingly used in Burkina Faso by producer organiza- 2009, but cereal yields have not increased between the tions that store dry grain for five to six months after har- pre-grant (2005–07) and the grant period (2008–11). vest to take advantage of seasonal price increases. But such Subsidized fertilizers are delivered late through regional programs are not a means of granting loans to producers. and provincial authorities in the Ministry of Agriculture, There is no registry of movable collateral although such Hydrology and Fisheries (MAHRH). These long delays a registry exists for nonmovable property. There are no do not appear to be a problem with regard to imported credit reference bureaus in Burkina Faso. fertilizers for cotton production where the supply chain is managed more effectively. Transport. Burkina Faso is separated by long, distant seaports. The roads leading to these ports are of uneven Mechanization. Promoting mechanization in an envi- quality and carriers face difficulties crossing borders ronment with limited resources, dominated by small- where they are exposed to official harassment as well as holders, is a challenge in most countries of sub-Saharan “rent-seeking” behavior by uniformed officers. In addi- Africa. Burkina Faso is no exception, and it is only recently tion, delays in ports, roadblocks, and border controls are that the government, through the FEER (Water Fund all factors explaining high transport costs. Truckers report and Rural Equipment), has promoted tractor use in a that they cannot make timely deliveries from Burkina project funded by the government of India called TEAM Faso to the seaports, where there is a long waiting time for 9. Imports by the government of a large volume of trac- vehicles and slow rotation of ships in ports. In addition, tors give little incentive to the private sector to import the rail system has only a limited capacity, and a large tractors for private distribution. In addition, there is a number of products must travel by road, so a large num- rental market of agricultural machinery in which the ber of exporters and traders prefer shipments through owners of tractors charge the user for field preparation Tema or Lome. Carriers indicate that vehicle registration and transportation. and licensing are not expensive procedures. They tend to 52 Agriculture Global Practice Discussion Paper complain more about informal roadblocks along major multiply. The government-dominated system does not highways and border crossings. Carriers have also indi- appear to be demand driven. As a result, many Rwandan cated that the state of roads should be improved while farmers do not have access to improved, certified seed of the main roads are generally in good condition and are high quality, available in the right quantities and at the regularly maintained. Like most countries of the ABI right time with complementary inputs. study, rural access to transport is mixed. The Rural Access Index of 24 to 25 percent is low. Road access is better in Fertilizer Use. Fertilizer use is rising but remains lower the more populated central region of Burkina Faso and is than the Abuja Declaration goal of applying 50 kg/ha of limited in the drier Sahelian regions that have low popu- fertilizer nutrients to arable land. Gross estimated fertilizer lation density. consumption increased from 14.3 kg/ha in 2008 to 34.0 kg/ha in 2011 (which in nutrient terms is about half those Policies and Public Spending on Agriculture. The levels). Fertilizer use on staple food crops, particularly dry- private sector rates the favorable context of agribusiness land cropped cereals, is limited in many African countries, at 3 on a scale of 5, recognizing the efforts made by the unless there is a well-funded subsidy program. Rwanda’s government of Burkina Faso to promote the agricultural fertilizer subsidy program, combined with hybrid maize sector but not entirely satisfied with the measures to stimu- seed imports, has led to large increases in fertilizer use late increased production of improved inputs use. Consis- on maize and significant average yield increases. Private tent and rigorous application of policies remains a thorny participants in input distribution systems rated the busi- issue, as evidenced by seasonal bans on the export of cere- ness environment for dealing in fertilizer as quite mixed in als, implemented in 2007–08 to 2011–12. The ability of Rwanda. The recently abandoned system of preselecting private sector organizations to effectively represent their 10 to 15 wholesale fertilizer dealers who bid for govern- members in discussions on these policies is not rated very ment-supplied fertilizer each year at auction was criti- favorably, and there is no umbrella organization in the cized by the private sector. In early 2013, the government field of agribusiness. Various interprofessional associa- authorized three fertilizer trading enterprises to import tions have not yet had a lot of influence but are seeking fertilizer for wholesale distribution. to strengthen their analytical and advocacy skills policies with the help of programs funded by donors. Public sec- Agricultural Mechanization. Rwanda’s density of 1.3 tor spending for agriculture, according to the government tractors per 100 km2 is by far the lowest of the ABI study of Burkina Faso, is higher than for most other countries countries. Although tractor imports are very low, they in sub-Saharan Africa, about 14.5 percent in 2008, above are duty free. From 2008 to 2012, only 156 tractors were the target New Partnership for Africa’s Development imported, with 58 percent imported by government agen- (NEPAD)/CAADP 10 percent. cies. In Rwanda, imports of spare parts carry a 15 percent duty and 15 percent VAT, which together make for higher import taxes on spares than in other ABI countries. Provi- RWANDA sion of tractors to Village Mechanization Service Centers Seeds. Use of improved seed is limited in Rwanda other (VMSC) and a few cooperatives has been subsidized at than access to improved maize and wheat seed in produc- 50 percent of the import cost. This undercuts the emer- tion zones where crop intensification program has been gence of any private sector capacity to sell tractors and most heavily implemented. Most farmers retain their own provide tractor hire services. Public sector initiatives that seed for other crops, using open-pollinated varieties (OPVs), subsidize acquisition of tractors and power tillers by often well beyond their three to four recommended years of VMSC and cooperatives tend to undercut incentives for efficacy. Two foreign companies are doing trials and hope to private sector importers, distributors, maintenance shops, gain approval to produce hybrid maize seed in Rwanda. The and private providers of tractor hire services. Hence, the Rwanda Agricultural Board (RAB) controls almost all foun- enabling environment in Rwanda does not yet appear con- dation seed production for OPVs, which leads to an inad- ducive for the emergence of a private sector-led network equate supply for private seed producers and companies to of tractor distribution, service, and custom hire providers. Agribusiness Indicators: Synthesis Report 53 Agricultural Finance. In Rwanda, several initiatives government. Although most African governments have in the financial sector are helping to expand access to not reached the NEPAD target requiring 10 percent of agribusiness finance, including the creation of a private total budget commitments to be allocated to agriculture, credit reference bureau, limiting use of movable assets as Rwanda achieved 10.2 percent in 2011. This uses the clas- collateral, some leasing of agricultural machinery, intro- sification of functions of government [COFOG] method, duction of agricultural loan portfolio guarantees, and a which accounts for expenditure on agriculture across the government-run rural finance facility. Even though com- government (beyond expenditures by MINAGRI alone). mercial banks remain the major source of funding for agribusiness in Rwanda outside of informal sources, sav- ings and credit cooperatives (SACCOs) and microfinance ETHIOPIA institutions (MFIs) are expanding their agricultural lend- Seed. In the seed subsector, perennial shortages of both ing, given greater reach than commercial banks in rural basic and certified seeds have limited agricultural produc- areas. Yet agricultural lending is a very small proportion tivity in Ethiopia. With less than 5 percent of the total of commercial bank operations (3.4 percent as of June land area planted with improved certified seed, Ethio- 2012 in Rwanda). pia is well behind its peer countries, such as Zimbabwe, Zambia, and Kenya, where commercial maize seed use Road Transport. Entry into the market is straightfor- exceeds 70 percent of total maize seeds. Ethiopia’s Min- ward. The trucking industry in Rwanda is composed of istry of Agriculture (MoA) continues to intervene in the many small players and is therefore fragmented. Trans- production, marketing, and distribution of certified seeds. port prices (per metric ton per km) on secondary roads are Although there are about 34 registered private seed com- usually double (or more) the rates on major trunk roads panies in the country, their total share of the certified seed in African countries. Rwanda is no exception, with rates market is only about 12 percent. over primarily major roads that average $0.19/mt/km and $0.37/mt/km over largely secondary roads. Trucking Fertilizer. The main actors in the fertilizer input mar- prices over tertiary rural roads are even higher. Agricul- ket are the Agricultural Input Supply Enterprise, which tural commodities are mainly transported by smaller 3- to has a monopoly to import fertilizers, and the coopera- 5-ton trucks or pickups and vans that operate informally tive unions, which are the primary wholesale and retail and haul goods from rural areas to secondary towns. outlets for farmers in the country. Private sector dealers Investing in improved roads and other infrastructure is a and retailers appear to be generally absent in the fertilizer priority for government. Despite government expenditure market. Although the intensity of fertilizer use is on the on infrastructure, only 38.1 percent of the entire national increase, rates still remain low at 17 kg of nutrients and road system was considered in good condition in 2011. 29 kg/ha compared to 200 kg/ha that is generally recom- mended as optimal for crop production in the country. Policy Environment. Private sector perceptions of Foreign exchange constraints continue to be a limiting the overall agribusiness policy and enabling environment factor to the liberalization in the market for this input. are improving but not unequivocally positive. The gov- ernment regulatory and policy environment is consis- Mechanization. The mechanized input market in Ethi- tent, even if private actors do not like all the government opia can be described as competitive with no government rules and interventions. Government policy making is direct involvement in the import and distribution of trac- quite transparent, although greater consultation prior to tors, and the free entry and exit of firms from the market. decrees and interventions would be welcome. Although There was no apparent collusion among the importers there is no explicit umbrella organization for the agribusi- and suppliers of tractors, who instead are in stiff competi- ness system in Rwanda, there is an agribusiness chamber tion with one another. Although tractors are exempt from within the Private Sector Federation with some influ- import duties, most prospective buyers are unable to capi- ence. Farmer organizations appear weaker than in most talize on this savings because of the six-month time limit other ABI countries, with cooperative creation driven by that tractors can be bought tax free once in the port of 54 Agriculture Global Practice Discussion Paper Djibouti. The result is that tractors remain expensive and Policy Environment. In the area of public policy, sig- scarce, with high rental costs and long queuing periods. nificant progress has been seen in government programs This also has the tendency of delaying land preparation and institutions that encourage private sector agribusiness and other farm activities. development. The government of Ethiopia (GoE) has sur- passed the 10 percent budget allocation to the agriculture Access to Finance from commercial banks and the sector as mandated by the African Head of State and gov- availability of good road transport were also investi- ernment within the CAADP framework. It has also estab- gated as enablers of private sector involvement in the lished the Agricultural Investment Support Directorate agribusiness industry. In Ethiopia, commercial banks (AISD) to facilitate the process of land acquisition and the account for more than 60 percent of the loans and provision of information for domestic and foreign investors credit to businesses, and therefore play a pivotal role in interested in doing business in agriculture in the country. leveraging agribusiness. The findings of this study, how- The ECX has accelerated the flow of timely market infor- ever, suggest that commercial banks in Ethiopia remain mation, and greatly reduced delays in payments to coffee reluctant to provide credit services to agribusinesses, farmers. ECX also has the warehouse receipt financing and smallholder agribusinesses in particular. The pro- program that links farmers to commercial banks. When portion of the total loan portfolio of commercial banks fully developed, the program should enable farmers who that is directed to agriculture increased only marginally, lack immovable collateral to access loans. Above all the from 8 percent a decade ago to 9.6 percent in 2010. Yet newly created Agricultural Transformation Agency (ATA) contrary to expectations, loans to the agriculture sector has been designed to address systemic bottlenecks in the performed as well as loans to the other core sectors of agriculture sector by supporting and leveraging the capa- the economy. Collection rates for outstanding loans from bility of the MoA and other stakeholders. the agriculture sector were no different from sectors like industry and international and domestic trade sectors. And despite recent progress with the establishment of the Other major players in agricultural finance in Ethio- ECX and ATA, direct government intervention persists pia are MFIs and rural saving and credit cooperatives with seed “crash” and ad hoc policies and price control (RUSACCOs). mechanisms that distort the seed subsector. Policy change and more effective institutions will be needed to establish Rural Transport. Access to good and reliable road an enabling environment in which private Ethiopian agri- transport continues to militate against agricultural trans- businesses can thrive. formation in Ethiopia. Most of the expansion in the country’s network of roads has focused on highways in urban areas. Roads in rural areas, where most agricul- NIGERIA tural production takes place, remain in very poor condi- Seed. Certified seed use in Nigeria is very low. Overall, tion. The cost of transportation is therefore much higher only about 5 to 10 percent of cultivated land in Nigeria on rural routes than on urban routes and international was planted with certified seeds, and about 10 percent corridors. Access to rural roads is measured by the Rural of farmers use certified seeds. Nigerian farmers would Access Index. While Ethiopia’s Index of 27 percent is require an estimated 1 million metric tons of improved higher than the 21.7 percent average for all low-income seeds to cover each of these cereals and pulses, whereas countries, only 53 percent of the country’s rural roads the formal commercial seed industry currently supplies are classified as being in good condition, compared to just 20,000 to 50,000 tons of seed annually covering all a 57.6 percent average for low-income countries gener- crops. This represents only 2 to 5 percent of farmers’ ally. (In Ghana for instance, 74 percent of rural roads actual seed needs and indicates a significant shortage in are in good condition.) Prohibitive taxes on new trucks the supply of certified seeds in the country. The govern- and spare parts continue to retard the replacement and ment dominates the production of foundation seeds, and maintenance of old fleets. As a result, most trucks in the its seed policy is currently tilted toward the government- country are not in good condition. owned ADPs in the production and marketing of seed. Agribusiness Indicators: Synthesis Report 55 This tends to crowd out private sector participation and Rural Finance. Like most African countries, agricul- is largely responsible for the shortages in certified seeds in tural credit in Nigeria comes from both formal and infor- the country. mal credit sources. Many Nigerian farmers are served by informal money lenders, who generally provide easy Fertilizer. Although Nigeria has great potential for fertil- access to credit but at higher cost, charging poor borrow- izer production given its abundant phosphate deposits and ers nominal monthly effective interest rates that typically natural gas reserves, almost all the fertilizer currently used range from about 10 percent to more than 100 percent. in the country is imported. The two fertilizer manufacturing Other forms of informal source include farmers’ associa- companies both went on to fail as the result of poor man- tions and cooperative societies. Formal sources include agement on the part of the public sector. As a result, fertil- credit from formal financial institutions such as com- izer consumption in the country is low, estimated at about mercial banks and microfinance enterprises and credit 600,000 to 700,000 tons annually compared to the poten- unions. Limited physical access to bank branches keeps tial market size of about 10 to 12 million tons. The fertil- investment in agriculture low, especially among small- izer industry has been characterized by heavy government holders despite the mandate by the federal government interventions in the form of subsidies. Available informa- stipulating that a certain percentage of commercial banks tion suggests that only 11 to 30 percent of subsidized fer- branches must be in the rural sector. Despite the fact that tilizer reaches smallholder farmers at the subsidized price. the agricultural sector accounts for about 40 percent of The parallel sales of “subsidized” and “market” fertilizer the total GDP of the country, its share of credit from tend to create an avenue for the lower-priced subsidized the commercial bank responsible is only 2 percent. High fertilizer to be diverted for sale at higher market prices. This interest rates charged by commercial banks on agricul- situation tended to crowd out the private sector and create tural loans were also identified as an important barrier opportunities for rent-seeking individuals. It is therefore to investment. not surprising that average fertilizer application rates are low and are estimated at 13 kg/ha, or just about 6 kg/ha in Rural Transport. In Nigeria, this consists mainly of terms of nutrient content—much lower than in most other road transport, and while the country has an extensive African countries. Overall, the value cost ratios (VCRs) for network of roads, most of its rural roads are in disrepair. the main staples were calculated (at unsubsidized prices) to Limited rail service, poor road conditions, frequent bottle- be 2.5 for maize, 2.5 for sorghum, 3.5 for cowpea/beans, necks, and informal checkpoints have all been identified and 3.4 for rice, respectively. These ratios indicate that the as causes of inefficiency and contributing factors to the use of fertilizer is profitable for these crops, provided they slow pace of agricultural commercialization in Nigeria. are delivered to farmers in a timely manner. The Port of Lagos experiences severe congestion that may cause ships to be docked for up to 20 days, and containers Farm Mechanization. Farm power among small- to be delayed for as long as 35 days. This has been identi- scale farmers in most parts of Africa, including Nigeria, fied as one source of delay in getting fertilizer to the farm- is largely human or animal driven, and relies largely on ers. The country’s unpaved roads are in particularly poor the use of the hoe and other hand tools. This low level condition. The local government road network, which is of mechanization greatly limits the amount of land that responsible for transporting farm produce from the farm can be cultivated, and limits the productivity of individ- to the first point of sale, is described as highly dilapidated ual farmers. More than 70 percent of the Nigerian labor with more than 70 percent impassable. This poor state of force is currently employed in agriculture. The estimated rural roads increases travel time, postharvest losses, and 45,000 tractors in the country translate to a density of 5.7 transport prices. tractors per 100 km2. And although tractor imports are themselves duty free, high tariffs are charged on imported Policy. Public sector investment in agriculture is low spare parts, leaving tractors with a short average life span and has been declining in recent years. It averaged just of about six years, compared to other countries in which 3 percent of public spending between 2006 and 2012, tractor life spans can be as long as 15 years. well below the 10 percent goal set by African leaders in 56 Agriculture Global Practice Discussion Paper the Maputo Agreement. Budgetary allocations by state zations in the country, is often perceived as a creation governments are lower still and can reasonably be con- of the government, lacking any effective voice in policy sidered indicative of the relatively low priority that pub- dialogue. A number of organizations receive financial lic officials assign to agriculture in Nigeria, in spite of support from the government and are therefore seen as the sector’s proportionately large contribution to overall lacking the independence necessary to represent their GDP. The ratio of public expenditure on agriculture to members or to be critical or to demand an audience the sector’s share of GDP between 2005 and 2012 was with public officials. Seed and fertilizer supply was sin- 0.04—a ratio of 1 would indicate that public spending gled out by stakeholders as a problem area, with fer- is commensurate with agriculture’s contribution to the tilizer suppliers in particular complaining of a lack of country’s economy. The Apex Farmers’ Association of transparency in how tenders and contracts are awarded Nigeria, which is the umbrella of all farmers’ organi- and payments made. Agribusiness Indicators: Synthesis Report 57 CHAPTER TEN CONCLUSIONS AND DISCUSSION Private sector perceptions of the environment for doing agribusiness in the nine ABI pilot countries show considerable consistency across the different inputs and services examined. The seed and fertilizer industries, the sales and service of agricultural machinery and tractor hire services, transport services, and other agribusinesses have all been subject to increased government intervention since the 2008 food price crisis, and this intervention has impinged heavily on investors’ perspectives of their prospects. Much of this intervention, particularly where it entails direct provision or subsidiza- tion of farm inputs or government participation in product marketing, clearly either crowds out private investment or otherwise provides strong disincentives to investors. The government’s role in different segments of the agricultural input system tends to be similar across input types. In some instances, this direct government engagement in the rural economy responds to very real market failures or to incomplete markets, but it more often appears to displace and substitute for investment on the part of pri- vate interests who are better placed than government to provide inputs and services. Yet awareness of these realities within the public sector appears to be increasing, and the need to provide more supportive policy environments is the matter of widening consensus. Where the public and private sectors enter into more constructive dialogue and partnership, the benefits are likely to become quickly apparent through the use of these agribusiness indicators, and in so doing provide advocates of similar change in countries that are slower to reform with considerable leverage in their policy dialogues. It will also afford the public sector greater selective discretion in determining which areas of the rural economy to engage in and how, saving governments the often exor- bitant costs of direct public spending. And it will build into these economies greater responsiveness to the demands of farmers in supplying the specific kinds of inputs they need, when they need them—a responsiveness that generally only results from prices and other signals that relatively free markets can provide. Much of this market respon- siveness is likely to consist of convenience and economy of access in matters such as the location of input dealerships. Farmers desperately need that as they attempt the transition from mainly subsistence to mainly commercial production, and are likely to respond positively to it themselves. Agribusiness Indicators: Synthesis Report 59 A number of such positive examples have already become to have a noteworthy effect on the amount of modern clear during the course of the ABI pilot studies. Kenya, seed and chemical fertilizer used. Tractor density and the Tanzania, and Zambia, for instance, have seen real pro- availability of tractor hire services are also major factors gress in productivity and commercialization as the result in determining levels of input use. Kenya and Zambia, of increased and more effectively targeted government which had the highest tractor densities at 27 and 21 per expenditure combined with more consistent and pre- 100 hectares of arable land, also saw the highest rates of dictable policies. While agricultural productivity gener- fertilizer application, at 100 and 70 kg/ha, respectively. In ally remains low and access to improved inputs remains these settings, farm labor is released and becomes avail- uneven in the nine countries, these positive examples send able for other operations. powerful signals to public officials and policy makers. Access to finance remains a key constraint facing fertilizer The agribusiness indicators also provide reliable points of wholesalers and agro-input dealers, leading to late deliv- reference for specifically identifying the persistent prob- ery of fertilizer and a lack of seasonal input credit. While lems that constrain productivity and the drive for com- fertilizer is a costly production input, farmers need short- mercialization. Some of the problems point to the need term production loans to purchase inputs. Fertilizer needs for direct engagement on the part of the public sector, to be provided on seasonal credit, with the input supplier for instance, the significant quantities of fraudulent inputs (which could be an aggregator, processor, or marketing/ available in markets, including mislabeled seed and ferti- trading company) recapturing the financial outlay at har- lizer that falsely claims to be certified, as well as expired or vest. A survey undertaken in Ghana by the International banned pesticides. Getting these products off the market Fertilizer Development Center and the International is an important role for regulators and law enforcement Food Policy Research Institute found that 79 percent of officials who are most likely to be employed in the public registered agrodealers lacked working capital to run their sector. Other problems, like the failure to deliver the right businesses (Fuentes et al. 2012). This constraint was often inputs to farmers at the right time and in the right mixes, noted in structured informal interviews by agro-input may be ones that public sector involvement appears to dealers in other study countries. actively aggravate. In some cases, there are compelling reasons for the public sector to disengage and “get out ABI country studies found that good farmer access to inputs of the way” in order for production to achieve the preci- and markets is correlated with lower transport costs (and sion necessary to modernize. Better organization, man- time/distance to market), better maintained rural roads, agement, and coordination of input supply and product and denser networks of agrodealers (retail input suppliers). marketing will generally improve an agricultural system. Road transport is very expensive in Africa because second- ABI studies found that in varying degrees, low yields in ary and tertiary roads tend to be poorly maintained. Signif- Kenya, Ethiopia, and Ghana are often the result of delays icant investments in building or upgrading of trunk roads that are directly attributable to government involvement have made long-distance trucking costs closer to interna- in fertilizer importation. When this entails tying fertilizer tional levels. But shipping agricultural inputs and products distribution to a subsidy program, the reimbursement over nontrunk roads is costly, often unreliable, and slow, promised to farmers, distributors, importers, and partici- with a limited supply of vehicles plying rural roads. Infor- pating banks is often delivered too late for them to use. mation and communication technology (ICT) can improve Stimulating competitive markets is implicated as a practi- timing and coordination in input supply and product cal way to resolve this issue. assembly, but the underlying physical infrastructure needs to be in place. Rural roads in most African countries are Agrodealer density emerged as an area of wide diver- poorly maintained, and access to them is often limited (as gence between ABI pilot countries, and this too tended reflected in low Rural Access Index percentages). 60 Agriculture Global Practice Discussion Paper APPENDIX A SELECTED COMPARATIVE COUNTRY STATISTICS RELATED TO THE AGRICULTURE SECTOR SELECTED COMPARATIVE ECONOMIC AND SECTORAL INDICATORS FOR ABI STUDY COUNTRIES Burkina Faso Mozambique Year Used Tanzania Ethiopia Rwanda Zambia Nigeria Source Ghana Kenya Mean Indicators Ag contribution to WDI 2006 33.3 47.9 30.4 26.8 27.9 32.0 38.4 30.4 22.4 32.2 GDP (%) Agricultural raw WDI 2011 51.7 8.6 4.9 10.9 5.1 6.1 4.0 5.6 1.9 na materials exports as percentage of total merchandise exports, 2011 Ag exports (as ITC 2012 22.2 73.8 28.9 35.5 4.3 0.9 35.3 nd 6.6 na percentage of total export value) Total land area WDI 2010 273,600 1,000,000 227,540 569,140 786,380 910,770 24,670 885,800 743,390 602,366 Ag land (percentage WDI 2009 43.7 35.0 68.1 48.1 62.7 81.8 81.1 40.1 31.5 54.7 of land area) Ag land (square WDI 2009 119,650 349,850 155,000 273,500 493,000 745,000 20,000 355,000 233,850 304,983 kilometers) Arable land (square WDI 2009 59,000 139,480 44,000 54,000 50,500 340,000 13,000 100,000 33,500 92,609 kilometers) Arable land as calc 2009 49% 40% 28% 20% 10% 46% 65% 28% 14% 30% percentage of agricultural land (Continued ) Agribusiness Indicators: Synthesis Report 61 SELECTED COMPARATIVE ECONOMIC AND SECTORAL INDICATORS FOR ABI STUDY COUNTRIES Burkina Faso Mozambique Year Used Tanzania Ethiopia Rwanda Zambia Nigeria Source Ghana Kenya Mean Indicators Ag land/rural calc 2009 977.8 506.1 1,309.4 867.5 3,423.2 936.8 232.0 1,075.4 2,778.6 1,345.2 population (hectares per person) Arable land/rural calc 2009 482.2 201.8 371.7 171.3 350.6 427.5 150.8 302.9 398.0 408.5 population (hectares/person) Arable land WDI 2009 21.5 13.9 19.3 9.4 6.4 37.3 52.6 11.2 4.5 19.6 (percentage of land area) GDP/capita WDI 2010 $283 $221 $360 $469 $384 $540 $337 $459 $432 $387 (constant $2,000) GDP/capita growth calc 2000-10 2.9 6.3 3.4 1.7 4.9 4.1 4.8 4.2 3.1 3.9 rate Net ODA/capita WDI 2010 $64.51 $42.50 $69.39 $40.20 $83.43 $13.02 $97.16 $65.97 $70.74 $60.77 (current $) Net ODA/GDP calc 23% 19% 19% 9% 22% 2% 29% 14% 16% 16% (using per capita data) Population (in FAO 2010 16,469 82,950 24,392 40,513 23,391 158,423 10,624 44,841 13,089 30,398 thousands) Population density WDI 2010 60.2 82.9 107.2 71.2 29.7 173.9 430.6 50.6 17.4 113.8 (population per square kilometer) Percentage of rural calc 2010 74% 83% 49% 78% 62% 50% 81% 74% 64% 68% population Rural population calc 2000-10 1.9% 2.2% 1.0% 2.3% 1.3% 1.1% 2.1% 2.2% 2.4% 0.6% growth rate Percentage of urban calc 2010 25.7 16.7 51.5 22.2 38.4 49.8 18.9 26.4 35.7 31.7 population Urban Pop growth calc 2000-10 6.8 3.5 4.1 3.8 4.9 4.1 5.7 4.5 2.8 4.5 rate Percentage of active calc 2010 92.0 77.3 54.5 70.6 80.5 24.8 89.3 75.8 63.2 69.8 population in agriculture Source: Authors. calc = calculation of data; FAO = Food and Agriculture Organization; WDI = World Development Indicators 62 Agriculture Global Practice Discussion Paper APPENDIX B ABI INDICATORS USED IN INITIAL AFRICAN COUNTRY STUDIES, 2011–12 How Key How Specified Calculated Assumptions Data Sources Observations Agricultural Productivity Measures Certified Seed Access and Use Improved seed use Percentage of staple Percentage of Percentage of MoA, agricultural Improved seed crop area planted area planted to cropped area research use tends to to certified seed. certified seed by estimates assume institutes be higher for Percentage of key staple crop an optimal Seed association irrigated crops cereal seed that (maize, wheat, seeding rate per Seed import data (rice), peri-urban is improved/ rice) relative to hectare; seed crops (F&V) and certified total cropped production and traditional cash (estimated from area imports are all crops (cotton) seed production, sold/used in year than for cereals, import data) of production/ oilseeds, legumes. import Improved seed Percentage of Percentage of Imperfect measure HH surveys or HH surveys (and access farmers using farmers, by farm as seed use and agricultural agricultural improved seed size category seeding rate per census censuses) hectare may be Special studies collect data on suboptimal percentage of farmers using improved seed (but often not on the quantity of seed used per farm or per hectare) (Continued ) Agribusiness Indicators: Synthesis Report 63 How Key How Specified Calculated Assumptions Data Sources Observations Seed law Existence of seed 0–5 scale Law in place MoA, ag research Year and law and extent of does not institutes, seed applicability of implementation equal effective association law and bylaws implementation are important to gauge Imported seed Are imports of Y/N Focus on Focus on imported MoA, ag research Refinement could certified seed key staples— seed for direct institutes be length of time allowed? cereals, oilseeds, sale and use Seed importers to get approval legumes (not as improved to import seed F&V, invariably seed (not for (case by case allowed) multiplication) basis versus one- time approval; regional lists of approved seed in trade—across RECs) Seed testing and Steps, time, and Number of If rain-fed crops, MoA, ag research Some improved registration costs required to cropping seasons each cropping institutes, private seeds in a register, test, and that tests are season equals seed companies subregion (of obtain approval required one year countries) are for new domestic on an accepted and imported within-region seed varieties trade list Private sector Percentage of Identify private Government MoA, agency for Foundation seed participation in foundation sector control over foundation seed production is seed production seed provided production, foundation seed production, monopolized by and distribution by government if any, and production is researchers, government-run organizations calculate very high in seed producers’ research institutes percentage of lower-income association and parastatals in basic/foundation countries SSA, but allowed seed that is by the private public sector in middle- income countries Number of private Number of Informal, Seed association Should number seed companies registered firms unregistered seed of firms be (that are seed producers or scaled to size association dealers are not of seed sector, members) fully accounted agricultural for sector, and so on? Volume and Private share is a Public shares MoA, seed Government percentage of residual once are small and producers’ monopoly or certified seed public shares declining as association, control may multiplied by known; focus on GDP/capita government not be efficient private firms key crops (maize, increases agencies in seed or desirable; and farms versus rice, and oilseed) multiplication government government and regulation lacks resources to entities multiply enough seed to meet farmer demand 64 Agriculture Global Practice Discussion Paper How Key How Specified Calculated Assumptions Data Sources Observations ISTA (International Does the country Y/N (and note Having an ISTA website ISTA has 12 “scope Seed Testing have an ISTA- number of accredited lab of accreditation” Association) accredited accredited labs) assumes good criteria for labs; accreditation laboratory? in-country seed these could be testing capacity used to develop a 0–5 scale Yield gaps Typical on-farm Percentage gap Large yield gaps MoA, ag research As major food crops yields, potential (one-typical reflect major institutes in most countries, yields, and yield yield/potential unrealized Sample surveys of maize and rice gaps for one to yield) for maize potential to farms have substantial two major food and rice expand food crop unrealized crops output potential Seed-to-grain price Price of certified/ Price of certified If ratio is less than MoA, ag research This is a measure of ratio improved seed seed divided by 5, improved seed institutes the profitability relative to the the average price use is considered Seed association or and risk in price of grain for of grain high cost and seed dealers producing and maize and rice risky using certified seed Fertilizer Access and Use Total fertilizer Total MT imported MT of fertilizer Volumes imported FAO, AMITSA Fertilizer imports consumption (and produced imports and local and produced MoC, customs data are often locally) by production/ locally are Large importers misclassified/ fertilizer type mixing, all consumed Local producer or identified as to differentiated by within year (no blender type; importers fertilizer type carryover); net are typically out exports of reluctant to local production provide detailed import volume breakdowns Fertilizer use rates Fertilizer Nutrient content National fertilizer FAO (to 2008 or Importers reluctant application rates is calculated import data 2009); customs, to provide in kilogram per by applying are available MoA or MoC volume data and hectare standard for recent years trade data; disaggregations Application rates conversion (often a challenge COMTRADE by fertilizer type for nutrients factors to main to get) data and prices (N, P, K) fertilizer types Allocation of Percentage of Total fertilizer Majority of Importers, export Typically difficult fertilizer to breakdowns of applied per fertilizer is used crop parastatal to establish with different crop fertilizer use crop type as on cash crops in or private firms any precision types by crop type: percentage of poorer countries; with outgrowers; unless very on food crops total supply maize and rice sample surveys good fertilizer versus cash (which equals are the key food (LSMS-ISA) for distribution data, crops (traditional imports in most grains in most field verification coupled with exportables) IDA countries) countries farm-level survey data on fertilizer use (Continued ) Agribusiness Indicators: Synthesis Report 65 How Key How Specified Calculated Assumptions Data Sources Observations Fertilizer access Percentage of Ideally Will not find National Farmers asked a farmers using disaggregated by fertilizer use agricultural Y/N question chemical farm size per crop or per census or without fertilizer hectare in most periodic farm providing details rural HH surveys HH surveys on quantities LSMS-ISA surveys obtained, use by crop, or type of fertilizer applied Fertilizer cost in Cost of 50-kg bag Converted to U.S. Prices are retail IFDC surveys; Cost of moving rural areas of NPK, urea, dollars per mt prices charged some national fertilizer from a and a third and compared by agro-input MIS (market info market town to a major fertilizer to cost of urea at dealers in market systems) farm village may in two to three port of entry towns add significantly main agricultural to fertilizer cost production zones in rural areas Agro-input dealer Number of agro- Calculate per ten Farmers are IFDC, CNFA; Listings of dealers density input dealers thousand farmers defined as those AGRA/IFPRI cover mainly (denominator) (not the number economically fertilizer studies; formal firms and and economically of farms) active in Min commerce understate total active people in agriculture Rural/agricultural numbers agriculture (as population data: the numerator) WBG, FAO Proximity of agro- Average distance Kilometers, if Requires recent Averages mask input dealers in kilometers known, or survey data: significant of farmer to an walking time LSMS, LSMS- variability among agro-input dealer converted to ISA, or other farms; percentage approximate HH survey of farms within, kilometers say, 7–8 km might be a better measure Ease of private 0 to 5 ordinal scale Obtained from Subsamples of Hard to get sector purposively importers, a sufficient participation in chosen sample of wholesalers, and sample without fertilizer market fertilizer industry retailers large players participants (importers) dominating Nutrient/output Calculate nitrogen Urea is 46% N; Maize prices in Urea retail price in Calculating a range price ratio (Pn/ price from urea maize price postharvest market towns; of ratios may Po) price and use as an average period of most maize price at be necessary, as maize price of weekly sales volume rural market/first both input and during the prices over influence farmers handler level product (maize) postharvest several months increase in prices prices vary across period postharvest at planting time time and location 66 Agriculture Global Practice Discussion Paper How Key How Specified Calculated Assumptions Data Sources Observations Profitability of Value cost ratio Increase in maize Assume can Agricultural Rely on research using maize (marginal value output due to disaggregate research results; other product of maize incremental increase in yield institutes, factors— produced over fertilizer use due to fertilizer CIMMYT, seed, water, the marginal response universities management, cost of using timely fertilizer) operations— influence yields (and need to be controlled for) Fertilizer subsidy Percentage of Need actual Actual outlays MoA, This captures the annual federal expenditure data may not MAFAP PERs cost of a fertilizer government from MoA and reflect planned subsidy program agriculture MoF expenditures in relative terms budget that goes to fertilizer subsidies Tariffs and taxes on Percentage of Ad valorem tariffs Declared values Customs, MoC Tend to be low; fertilizer declared import as a percentage are based on real distinguish value declared import import values between imports value from region and international imports Import cost of CIF price for two $ cost/mt of CIF price is the Importers, customs CIF cost of fertilizer to three key urea, NPK, border price and MoC imported fertilizers and another (not arrival at fertilizers varies commonly port price for significantly imported landlocked across countries fertilizer countries) (which may reflect decrease in competitive procurement) Fertilizer Ratio of CIF price Arrival at border Rural market price Customs data; Using a border distribution of fertilizer to price as a is a reasonable international price rather than efficiency rural market percentage of the proxy for farmer fertilizer a classic CIF price rural retail price; cost databases; MoC price (arrival for landlocked at seaport) countries this is preferred, would be CIF though port price plus costs and transport to of port clearance border costs must and transport to be accounted for border Note: getting CIF prices from importers is a challenge (Continued ) Agribusiness Indicators: Synthesis Report 67 How Key How Specified Calculated Assumptions Data Sources Observations Extent and Use of Agricultural Mechanization Tractor use Total number of Simple ratio. FAO Need to assume an FAO to 2007, 2008 Take FAO numbers tractors per 100 definition of average useful Mechanization as base, add km of arable 2 arable land life of tractors Department of imports, and land MoA, customs net out assumed number of tractors falling out of service to arrive at end-of- year stock Tractor hire service Cost of plowing 1 Convert local Size of hectare for Large farmers, Cost may vary cost ha of land currency cost to plowing purposes service providers, across production $/ha is consistent FOs, tractor zones with across countries importers different soil types; differentiate plowing and disc harrowing Tractor imports Number of tractors Simple ratio Customs/trade Customs data, Public sector imported by data include MoC, importers, import programs private sector as public sector MOA, are subsidized; percentage of the imports parastatals need to make total number of sure public tractors imported imports are fully accounted for Ease of private Perception of Average of This is a complex Private importers, Subsidized participation in participants responses composite across providers of government the agricultural (ranked on a three different hire services, import and machinery scale of 0–5) groups of firms; machinery repair parastatal market last two (next and maintenance programs tend to cell) hard to providers undercut private identify sector Tractor import Tariffs on new Percentage specified Import tariffs are Customs, MoC, Spares are subject taxation tractors and in tariff codes actually applied importers to higher tariffs spare parts than new tractor imports; can extend to other agricultural equipment 68 Agriculture Global Practice Discussion Paper How Key How Specified Calculated Assumptions Data Sources Observations Measures of Access to and Availability of Key Agricultural Sector Services Access to and Availability of Agribusiness Finance Credit allocation to Percentage of Percentage of loan Covers commercial Central Bank Typically available agriculture commercial portfolios of banks and not Department of for production banks lending to all commercial MFIs or other Supervision of agriculture agriculture banks in country sources of Commercial and not agro- (not just those credit/finance Banks enterprise lending to (classified agriculture) under lending to industry or commerce); much harder to obtain loan data on agribusiness finance Cost of credit Nominal and real Nominal rates Best to present Commercial banks Real interest rates interest rates less inflation a range that Agro-enterprises provide a better on commercial rates equals real includes rates estimate of credit bank loans to interest rates paid by firms of costs than do agriculture and different types/ nominal rates agribusiness scales Density of financial Bank branches per Bank branches Adult population is Central Bank Does not capture services 100,000 adult divided by adult defined as 15–59 Commercial banks MFIs, NGOs, population population years and rural projects with finance components Rural access to Percentage of rural Ideally Sample surveys May not capture credit HHs receiving disaggregated by Agriculture census loans for RNFE credit for farm size (rural nonfarm agriculture employment) Agribusiness access Percentage of Ideally Access of formal Finscope or FINAP ABI estimates to credit agribusinesses disaggregated firms greater surveys/studies may not be with access to by firm size and than for Special surveys representative, as credit type nonformal farms ABI estimates ABI interviews larger, more accessible firms Agricultural finance Existence of Scale indicates Other innovations Donors, banks, Requires some innovation warehouse stages of WRS exist; this targets warehouse digging and receipt system development cereals and operators judgment to (WRS): 0–5 other storable ascertain scale/ordinal commodities index (Continued ) Agribusiness Indicators: Synthesis Report 69 How Key How Specified Calculated Assumptions Data Sources Observations Law on leasing Existence of a law Y/N Leasing expands Central Bank Existence of law on leasing (and access to costly Commercial banks does not reflect extent of use of equipment extent of use leasing) (value of leases of equipment used in agriculture and agribusiness) Movable collateral Evidence of use of Y/N—Can also As many farms Commercial banks Could be refined to movable assets as specify types lack land title, estimate use of collateral this provides an movable assets alternative source in lending (and of collateral percentage of valuations by type of movable asset used as collateral) Collateral registry Presence of a Y/N for immovable Tracks repayment Central Bank This is already collateral registry assets and Y/N performance of Commercial banks an IFC/DB for movable firms and (larger) indicator; new assets farms. dimension is to distinguish between immovable and movable assets Credit rating and Existence of a Y/N with Databases need Credit rating Could add a reporting privately run discussion to include bureau measure of the credit reporting of extent of agriculture sector Central Bank scale of any system databases and & agribusiness Commercial banks credit reporting their use firms databases and accessibility of this info to lenders and vendors Access to and Cost of Transport Trucking prices Prices paid in per Local currency Prices do not Firms/traders/FOs Hardest to get mt/km grain or prices converted necessarily reflect using trucking transport prices fertilizer along to $/mt, full transport services; trucking on secondary and well-defined road differentiated by costs firms; WFP, tertiary roads itineraries road type NGOs in rural zones; highly variable 70 Agriculture Global Practice Discussion Paper How Key How Specified Calculated Assumptions Data Sources Observations Sea freight costs Maritime shipping Shipping quotes Shipping costs can Shippers (fertilizer Quotes from costs for standard obtained be quite different importers, maritime containers of 20′ from shippers for exported agriculture shipping and 40′ (loaded or maritime and imported product companies may with inputs and shipping firms containers exporters) be prices prior outputs ($ per International sea to negotiation; ton) freight firms— database already MSA, Maersk, captures port Delmas handling and customs/ other agency processing costs Entry into trucking Ease of entry Opinions of Transporters, Ease of entry is not into trucking transport firms trucking necessarily ease (of inputs and associations of operations agricultural products) (Scale: 0–5) Access to trucking Opinions of users Opinions of Ease of access Users of trucking Need to ask services of trucking traders, to trucks and services informants not services importers, other alternative Trucking to focus solely on (Scale: 0–5) shippers price quotes ≈ association cost but rather competitiveness on ease of access of trucking and availability of trucks for hire Infrastructure and Quality of trade LPI (logistics Shipping LPI rankings of Based on a set of logistics services and transport performance infrastructure WBG Trade opinions and quality related index); last and logistics Department perception infrastructure rankings from greatly matter (will they be ratings of service 2010 for exports to be continued?) providers and competitive users Rural access to Percentage rural Rural Access Index Proximity to an all- RAI rankings of HH survey based roads HHs within 2 km year practicable WBG based estimates are of a road road equates to on HH surveys dated (2003 good access and GIS; to 2006) and few countries may not match may produce GIS-generated estimates estimates in some countries (for example, Ghana) (Continued ) Agribusiness Indicators: Synthesis Report 71 How Key How Specified Calculated Assumptions Data Sources Observations Enabling Environment Indicators Private sector Private sector Opinions of a Positive EE Purposive sample Expert interviewing perception perception of cross section ≈ positive of firms (ABI and judgments agribusiness of agribusiness investment developing key are required enabling firms climate ≈ informant lists in to tease out environment greater ease of pilot countries) rankings; need (0–5 scale) agribusiness to ask a series opportunities of questions to and higher arrive at a score probability of (beware large agribusiness firm bias) investment Policy consistency Private sector Opinions of a The higher the Purposive sample Composite measure perception of cross section perception of of firms and for several policy consistency of of agribusiness consistency, knowledgeable domains—input agribusiness firms (plus expert the greater the observers subsidies, import policy and judgment) probability of policies, price/ regulatory private sector movement environment investment controls; could (0–5 scale) focus on recent policy events and shifts for key staples and inputs Policy advocacy Private sector Expert judgment Strong private Advocacy groups How handle capacity of the advocacy group on a 0–5 scale sector policy Policymakers and countries with private sector for agribusiness: advocacy keeps technocrats in interprofessional existence and public sector government organizations effectiveness honest and Knowledgeable organized on of an apex improves EE observers value chains and agribusiness without an apex organization agribusiness organization? Public expenditure Federal government Percentage Planned budgets do MoA, Ministry of PERs typically do on agriculture budget outlays of federal not necessarily Finance, WBG not capture all (PER) on agriculture government equal actual (recent PER), donor-funded as percentage of budget; expenditure MAFAP program projects or NGO total budget disaggregations patterns, which activities that are along functional ABI must try to not accounted lines and for get for in the federal recurrent versus budgetary investment costs process; note are very useful what some countries consider to be agricultural expenditure 72 Agriculture Global Practice Discussion Paper How Key How Specified Calculated Assumptions Data Sources Observations Producer returns Proportion of a key Need producer MoA, marketing Not recommended on export crops cash crop’s FOB prices and FOB parastatal, value to compare (to be dropped) export price paid export values chain studies percentage to producers (prices/costs return to along VCs) producers across commodities; transport costs a bigger share of FOB price (at seaports) for landlocked countries Degree of value Percentage of key Percentage value Increasing degree Customs data, Customs data for added to exports export crop that of commodity of VA in exports MoC trade exports may not is processed exports by forms increases export statistics; be sufficiently in which they are revenues COMTRADE disaggregated or shipped for import properly coded partner data Export policy Export taxes or Y/N Customs tariff code; Could develop a disincentives quotas MoA and MoC scale of 0–5 for approvals, where higher fees numbers mean a greater number of disincentive polices or mechanisms Export approvals Specify for key Steps/costs in High transactions Exporters of grain, Cereals, as food and restrictions agricultural process and costs discourage F&V, a key security crops, exportables: time required to exports and traditional export are the most a staple obtain approvals increase export crop problematic; grain, F&V, a marketing interdictions/bans traditional export costs (hence on exports are cash crop decrease in common in SSA, competitiveness) despite regional agreements that do not permit them Note: AGRA/IFPRI = Alliance for Green Revolution in Africa/International Food Policy Research Institute; AMITSA = Regional Agricultural Input Market Information and Transparency System; CIMMYT = The International Maize and Wheat Improvement Center; COMTRADE = Commodity Trade Statistics Data- base (of the United Nations); F&V = fruit and vegetable; FOB = free on board; HH = household; IDA = International Development Association of the World Bank Group; LSMS-ISA = Living Standards Measurement Study-Integrated Surveys on Agriculture; MoC = Ministry of Commerce; VCs = Voluntary Carbon Standards; WFP = World Food Programme. Agribusiness Indicators: Synthesis Report 73 INCENTIVES/PROGRAMS FOR AGRICULTURAL MECHANIZATION IN PILOT COUNTRIES Country Incentives Kenya Import tariff exception on all agricultural tractors and equipment Ethiopia Import tariff exception within six months of import, otherwise full duties (more than 100 percent of FOB price must be paid thereafter) Nigeria Import tariff exception on all fully assembled tractors but excluding completely knocked down parts Burkina Faso India provided 700 tractors through a Burkina Faso government agency that subsidized distribution Tanzania 20–50 percent shared by farmers, depending on the level of scale of operation, with smallholders receiving about 50 percent subsidies on tractor imports Rwanda Import tariff exemption 50 percent purchase subsidy Mozambique Tractors provided by donors or government have generous financial terms: low initial payments, below-market interest rates, and in some cases, grace periods of several years before interest due Ghana Procured by the government and sold to farmers at 50 percent down payment and the rest to be recovered without interest over a three-year period Zambia N/A Sierra Leone 50 percent purchase subsidy Benin 50 percent purchase subsidy Côte d’Ivoire Five-year loan Source: Tokida 2011; Field Surveys, ABI 2011–12. IMPORT DUTIES AND TARIFFS ON TRACTOR AND SPARE PARTS IMPORTS IN PILOT COUNTRIES, 2011–12 Duties /Tariffs on Spare Country Duties/Tariffs on Tractors Parts Issues/Comments Tanzania Zero rated, also exempt from VAT Import duty of 10 percent; VAT exempt Kenya Zero rated Import duty plus VAT of 16 percent Burkina Faso 5 percent import duty and VAT 20 percent import duty; 18 percent There are variations between what exempt (official); in practice taxes VAT (official); actual could go as the custom books stipulated and plus VAT could add up to 16 high as 34–38 percent what is actually paid percent Zambia Zero rated and exempt from VAT 15 percent import duty and 16 percent VAT for effective tax of 31 percent Ghana Zero rated Ethiopia Zero rated if buyers have import Zero rated if imported at the same Farmers complained of the time exemption papers; otherwise time as tractors and if bought synchronization between import 10 percent import duty plus 15 within six months of import, and loan disbursement; loans percent VAT, if bought within six otherwise 10–15 percent duties disbursed after government months of import plus VAT budget, which seems to fall well after the planting season Mozambique 5 percent duty on imports of 7.5 percent duty plus 17 percent tractors for agriculture VAT 74 Agriculture Global Practice Discussion Paper IMPORT DUTIES AND TARIFFS ON TRACTOR AND SPARE PARTS IMPORTS IN PILOT COUNTRIES, 2011–12 Duties /Tariffs on Spare Country Duties/Tariffs on Tractors Parts Issues/Comments Rwanda Zero rated for agricultural tractors 15 percent import duty and 15 percent VAT for effective duty of 30 percent Nigeria Zero rated for fully assembled Federal government has been trying tractors but there is a VAT of 5 to promote tractor acquisition; percent on the importation of tariffs on CKD were reduced completely knocked down (CKD from 25 to 5 percent in 2011 parts Source: Field Surveys, ABI 2011–12. Agribusiness Indicators: Synthesis Report 75 BIBLIOGRAPHY Access Capital Research. 2012. Sector Review: Investing in Ethiopian Agriculture. Agribusiness Indicators: Burkina Faso. May 2013. Agribusiness Indicators: Ethiopia. Report Number 68237-ET. April 2012. Agribusiness Indicators: Ghana. Report Number 68163-GH. April 2012. Agribusiness Indicators: Kenya. January 2013. Agribusiness Indicators: Mozambique. Report Number 68241-MZ. April 2012. Agribusiness Indicators: Nigeria. July 2013. Agribusiness Indicators: Rwanda. July 2013. Agribusiness Indicators: Tanzania. November 2012. Agribusiness Indicators: Zambia. December 2012. Bank of Ghana. 2011. “Quarterly Bulletin January 2011.” Bumb, Balu, and Carlos A. Baanante. 1996. “World Trends in Fertilizer Use and Projections to 2020.” IFPRI, Washington, DC. Dorosh, Paul, Hyoung Gun Wang, You Liangzhi, and Emily Schmidt. 2012. “Road Connectivity, Population and Crop Production in Sub-Saharan Africa.” Agricultural Economics 43(1): 89–103. FAO and UNIDO. 2008. “Agricultural mechanization in Africa: Time for action: plan- ning investment for enhanced agricultural productivity.” Report of an expert group meeting jointly held by FAO and UNIDO in Vienna on 29–30 November 2007. FAO and UNIDO. 2011. Investment in Agricultural Mechanization in Africa. Workshop organized and hosted by FAO and UNIDO, Centre for Agricultural Mechanization and Rural Technologies (CAMARTEC), 3–5 June 2009. Arusha, Tanzania. FAO. 2008. “Agricultural Mechanization in Africa: Time for Action, Planning Invest- ment for Enhanced Agricultural Productivity.” Report on Expert Group Meeting, January 2008. Vienna, Austria. FAO. 2010. Agricultural Mechanization in Mali and Ghana: Strategies, Experiences and Lessons for Sustained Impacts. Agriculture and Food Engineering Working Document No. 8. FAO. 2012. Current World Fertilizer Trends and Outlook to 2016. Rome: FAO. FAO, MAFAP. Monitoring African Food and Agricultural Policies. http://www.fao .org/mafap/. Field Survey. 2011–12. Cropping Seasons. Washington, DC: World Bank. FinStats. 2012. Tool to Benchmark Financial Sectors Across Countries and Time. Washington, DC: World Bank. FMA & RD. 2011. National Policy on Agricultural Mechanization Development (draft). Abuja, Nigeria, May. Fuentes et al. 2012. IFDC and IFPRI, “Improving Fertilizer Markets in West Africa: The Fertilizer Supply Chain in Ghana,” Paper Series IFDC—P-40 December 2012. Agribusiness Indicators: Synthesis Report 77 Gisselquist, David, and Carl E. Pray. 1999. “Deregulating Technology Transfer in Agri- culture: Reform’s Impact on Turkey in the 1980s” (March 1999). World Bank Policy Research Working Paper No. 2086. Available at http://ssrn.com/abstract=614976 IFDC. 2011. PRODIB (Projet de professionnalisation des distributeurs d’intrants agri- coles du Burkina Faso), Etude de géo-localisation et caractérisation des distributeurs d’intrants agricoles d’AGRODIA. Ouagadougou, August 2011. IFPRI. 2013. 2012 Global Food Policy Report. Washington, DC: IFPRI. IITA. 2011. “Research to Nourish Africa.” IITA Bulletin Issue No. 2095, 14–18 November. Improved Rice, Cowpea, and Maize Seeds in Nigeria: Policy Implications and Knowledge Gaps, Nigeria Strategy Support Program (NSSP). NSSP Working Paper No. 0016. Washing- ton, DC: International Food Policy Research Institute (IFPRI). International Finance Corporation. 2012. “Innovative Agricultural SME Financing Models,” published in November 2012 by IFC as the lead technical advisor to the G-20 Global Partnership for Financial Inclusion. Kenya, MoA (Ministry of Agriculture). Various years. Economic Review of Agriculture. Central Planning and Project Monitoring. Nairobi: Kilimo House. Lopez-Pereira, M.A., and M.L. Morris. 1994. Impacts of International Maize Breeding Research in the Developing World, 1966–1990. CIMMYT Economics Program, Mex- ico. Updated in 2002 by Morris, Michael. Morris, M., V.A. Kelly, R.J. Kopicki, and D. Byerlee. 2007. Fertilizer Use in African Agri- culture: Lessons Learned and Good Practice Guidelines. Washington, DC: World Bank. Nathan Associates. 2010. West Africa Transport Logistics Analysis Using FastPath— Tema-Ouagadougou Corridor. Washington, DC: USAID. Odame, H., and E. Muange. 2010. “Can Agro Dealers Deliver the Green Revolution in Kenya?” Working Paper No. 014. Future Agricultures, Sheahan, M.B. 2011. Analysis of Fertilizer Profitability and Use in Kenya. M.Sc. thesis, Michigan State University, East Lansing. O’Donoghue, T., and K. Punch. 2003. Qualitative Educational Research in Action: Doing and Reflecting. New York: Routledge. p.78. Raballand, Gael, Patricia Macchi, and Carly Petracco. 2010. Rural Road Investment Efficiency. Washington, DC: World Bank. Roberts, Peter, K.C. Shyam, and Cordula Rastogi. 2006. Rural Access Index: A Key Development Indicator. Transport Papers (TP-10). Washington, DC: World Bank. Smil, V. 2001. Enriching the Earth: Fritz Haber, Carl Bosch and the Transformation of World Food Production. Cambridge, MA: MIT Press. Sowing the Seeds of Food Security. 2011. Paul Van Mele, Jeffery W. Bentley Bolivia, Rob- ert G. Guéi, editors. Published by the Food and Agriculture Organization of the United Nations and the Africa Rice Center (AfricaRice) by arrangement with CAB International. Spielman, David J., D. Kelemework, and D. Alemu. 2011. “Seed, Fertilizer and Agri- cultural Extension.” Development and Governance Division, IFPRI, Ethiopia Strategy Support Program II (ESSP II). Working Paper 020, Addis Ababa, Ethio- pia (IITA, 2011). 78 Agriculture Global Practice Discussion Paper Surface and Marine Transport Regulatory Authority (SUMATRA). 2011. Tanzania. “Report on Evaluation of Truck Freight Charges.” November. Takeshima, H., A. Oyekale, S. Olatokun, and Sheu Salau. 2010. Demand Character- istics for Improved Rice, Cowpea, and Maize seeds in Nigeria. International Food Policy Research Institute, Series 16. Teravaninthorn, S., and G. Raballand. 2009. Transport Prices and Costs in Africa: A Review of the Main Trade Corridors. Washington, DC: World Bank. Tokida, K. 2011. “Public-Private Sector Models for Mechanization in SSA.” Work- shop on Boosting Agricultural Mechanization of Rice Cropping System in Sub- Saharan Africa, St. Louis, MO, 6–8 June. Van Mele, P., J.W. Bentley, and R.G. Guéi (eds.). 2011. African Seed Enterprises Sowing the Seeds of Food Security. Edited by Paul Van Mele Africa Rice Center (AfricaRice), Jeffery W. Bentley Bolivia, Robert G. Guéi Food and Agriculture Organization of the United Nations Italy. Published by the Food and Agriculture Organization of the United Nations and the Africa Rice Center (AfricaRice) by arrangement with CAB International (2011). Wiggins, Steve, and Sharada Keats. 2013. Leaping and Learning: Linking smallholders to markets in Africa. London: Agriculture for Impact, Imperial College and Overseas Development Institute. World Bank. 2011. Africa Infrastructure Country Diagnostic. Washington DC: World Bank. World Bank. 2012a. Analytical Support for Agricultural Public Expenditure Scale-Up in Sub- Saharan Africa. Presentation by Stephen Mink, Project Director. Washington, DC: World Bank. World Bank. 2012b. Logistics Performance Index. Washington, DC: World Bank. World Bank, Agriculture and Environmental Services. Country Studies for Agribusiness Indicators. Washington, DC: World Bank. World Bank, Agriculture and Rural Development Unit (AFTAR). 2009. “Eastern Africa: A Study of the Regional Maize Market and Marketing Costs.” Report No. 49831 AFR, December. Washington, DC: World Bank. World Bank. 2013. Strengthening National Comprehensive Agricultural Public Expenditure in Sub-Saharan Africa. February. Washington, DC: World Bank. World Bank. 2009. Project Information on Strengthening National Comprehensive Agricultural Public Expenditure in Sub-Saharan Africa. Sustainable Development Department, Africa Region. Prepared by Stephen Mink, Project Director. November. Washing- ton, DC: World Bank. Agribusiness Indicators: Synthesis Report 79 A G R I C U LT U R E G L O B A L P R A C T I C E D I S C U S S I O N P A P E R 0 1 W O R L D B A N K G R O U P R E P O R T N U M B E R 91133-AFR 1818 H Street, NW Washington, D.C. 20433 USA Telephone: 202-473-1000 Internet: www.worldbank.org/agriculture