94910 Innovative Agricultural SME Finance Models November 2012 © International Finance Corporation 2012. All rights reserved. 2121 Pennsylvania Avenue, N.W. Washington, D.C. 20433 Internet: www.ifc.org The material in this work is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. IFC encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly, and when the reproduction is for educational and non-commercial purposes, without a fee, subject to such attributions and notices as we may reasonably require. IFC does not guarantee the accuracy, reliability, or completeness of the content included in this work, or for the conclusions or judgments described herein, and accepts no responsibility or liability for any omissions or errors (including, without limitation, typographical errors, and technical errors) in the content whatsoever or for reliance thereon. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The contents of this work are intended for general informational purposes only and are not intended to constitute legal, securities, or investment advice, an opinion regarding the appropriateness of any investment, or a solicitation of any type. IFC or its affiliates may have an investment in, provide other advice or services to, or otherwise have a financial interest in, certain of the companies and parties (including those named herein). All other queries on rights and licenses, including subsidiary rights, should be addressed to IFC’s Corporate Relations Department, 2121 Pennsylvania Avenue, N.W., Washington, D.C. 20433. International Finance Corporation is an international organization established by Articles of Agreement among its member countries, and a member of the World Bank Group. All names, logos, and trademarks are the property of IFC and you may not use any of such materials for any purpose without the express written consent of IFC. Additionally, “International Finance Corporation” and “IFC” are registered trademarks of IFC and are protected under international law. Innovati v e Agric u ltu ral S ME Finance Models 1 Table of Contents Acknowledgements 5 Introduction 7 Executive Summary 9 CHAPTER 1 Major Challenges and Opportunities 14 1.1 The Challenge of Lending to Agriculture 14 1.2 The Opportunities in Lending to Agriculture 16 1.3 Target clients 18 CHAPTER 2 Innovative Farmer and Agricultural SME Financing Models 25 2.1 Financing Farmers 27 2.2 Financing movable assets 29 2.3 Financing Farmers in Value Chains 31 2.4 Risk Management Models 35 2.5 Distribution models 38 CHAPTER 3 Observations 41 3.1 Observations from the Case Studies 41 3.2 Enabling Country Environments 44 3.3 Patterns, Challenges, and Solutions 51 CHAPTER 4 Conclusion 57 CHAPTER 5 Case Studies Summaries 60 ANNEX A Case Overview 100 ANNEX B NPLs and GDP Dependency on Agriculture 103 ANNEX C Cases with Color Coded Ratings 106 Bibliography 109 2 Glob al Partn er ship for Financial I nc lu si on List of Figures Figure 1  Where the funding opportunities lie 18 Figure 2  Farmer segmentation and key characteristics 19 Figure 3  Farm Size and Productivity — Evidence from Uganda 23 Figure 4  Financing Models and Cases 26 Figure 5  Actors on the Supply Chain 27 Figure 6  Risk transfer models and cases 35 Figure 7  Distribution models and cases 39 Figure 8  Channels for Farmers 41 Figure 9  Farmers and Value Chains 42 Figure 10  Three Business Environments 46 Figure 11  Environment I Models 48 Figure 12  Environment II Models 49 Figure 13  Environment III Models 50 Figure 14  Framework to Implement Innovative Models. 55 Figure 15  Agricultural lending share and NPLs in 5 countries 103 Figure 16  NPLs of high, medium and low agricultural GDP share countries 104 List of Cases CASE 1  Equity Bank, Kenya — “Kilimo Biashara” 60 CASE 2  Opportunity International, Africa — Informed Lending 61 CASE 3  HDFC Bank, India — Correspondent Banking 62 CASE 4  Zanaco, Zambia — Munda Smallholder Scheme 63 CASE 5  Finterra, Mexico — Emerging Farm Business Financing 64 CASE 6  Zanaco, Zambia — Emergent Farmer Finance and Support Program “ZEFP” 65 CASE 7  NMB, Tanzania — Kilimo Account Product “KAP” 66 CASE 8  Banco De Lage Landen, Brazil — Equipment Finance 67 CASE 9  Mahindra & Mahindra Financial Services, India — Equipment Finance 68 CASE 10  Jain Irrigation, India — Equipment Finance 69 CASE 11  Development Finance Company Uganda Leasing, Uganda — Leasing 70 CASE 12  NMB, Tanzania — Warehouse Receipt Finance 71 CASE 13  HDFC Bank, India — Warehouse Receipt Loans Facility 72 CASE 14  Ghanaian Financial Services, Ghana — Collateral Management 73 CASE 15  Dunavant Zambia Ltd. and Cargill Zambia Ltd. — Farmer Input Credit 74 CASE 16  Palabana Dairy Cooperative Society & Parmalat, Zambia — Value Chain Finance 76 CASE 17  ECOM, Africa & Asia — Capital Improvement Loan Facility 77 CASE 18  Ghana Grains Partnership, Ghana — Value Chain Finance 78 Innovati v e Agric u ltu ral S ME Finance Models 3 CASE 19  CRDB, NMB, Kilombero Sugar & Mtibwa Sugar, Tanzania — Outgrower Finance 79 CASE 20  NMB, Tanzania — Agro-Dealer Scheme 80 CASE 21  Bayer, Raiffeisen Aval Bank, Ukraine — SME Farmer Input Credit 81 CASE 22  ITC & State Bank of India — Smallholder Input Finance 82 CASE 23  Centenary Bank & Technoserve, Uganda — Factoring 83 CASE 24  Kenya Gatsby Trust — Factoring 85 CASE 25  Root Capital, Latin America & Africa — Export Trade Finance 86 CASE 26  Microensure & Kilimanjaro Native Coffee Union, Tanzania — Health Insurance 88 CASE 27  Taytay Sa Kauswagan, Inc (TSKI) , Philippines — Index Insurance 89 CASE 28  Syngenta Foundation & UAP Insurance, Kenya — “Kilimo Salama” Index Insurance 90 CASE 29  PepsiCo, ICICI Lombard & WRL, India — Index Insurance 91 CASE 30  Bagsa Agricultural Commodity Exchange, Nicaragua — Commodity Price Risk Management 92 CASE 31  M-Pesa and M-Kesho, Kenya — Mobile Banking 93 CASE 32  Banque Populaire du Rwanda — Mobile Banking 94 CASE 33  Refresh Mobile WING, Cambodia — Mobile Banking 95 CASE 34  United Bank Ltd., Pakistan — “Omni” Branchless Banking 96 CASE 35  Opportunity International Bank, Malawi — Mobile Banking 97 CASE 36  Dunavant Zambia Ltd. — Mobile Payment Systems 98 CASE 37  Africa Agriculture and Trade Investment Fund (AATIF) 99 Innovati v e Agric u ltu ral S ME Finance Models 5 Acknowledgements International Finance Corporation (IFC) is the lead technical advisor (implementing partner) to the G-20 Global Partnership for Financial Inclusion’s (GPFI) SME Finance Sub-Group. This report was produced by IFC on behalf of the GPFI. The GPFI is the main platform for implementation of the G-20 Financial Inclusion Action Plan. The Group engages partners from G-20 and non-G-20 countries, the private sector, civil society, and others. It is chaired by the G-20 troika countries, currently France, Mexico, and Russia. The GPFI is supported by four Implementing Partners: the Alliance for Financial Inclusion (AFI), the Consultative Group to Assist the Poor (CGAP), IFC, and the World Bank. www.gpfi.org The report “Innovative Agricultural SME Finance Models” was developed under the overall guidance of Peer Stein (IFC) and Susanne Dorasil (BMZ). The IFC working group team for the report was led by Panos Varangis and comprised Ulrich Hess, Heather Miller, Ghada Teima, Anushe Khan, and Patricia van de Velde. The work was undertaken in close collaboration with GIZ contributors Roland Gross, Achim Deuchert, and Jan Meise (BMZ). The team would like to thank the World Bank-IFC internal peer reviewers Rolf Behrndt, Dave Chalila, Gary Reusche, Hans Dellien, Raiomand Billimoria, Ajai Nair, Marc Sadler, Don Larson and Maria Pagura. IFC would also like to thank external peer reviewers Claudia Schmerler (KfW), Birgit Holderied-Kress (KfW), Michael Hamp (IFAD), Eugenia Serova (FAO), Calvin Miller (FAO), Emilio Hernandez (FAO), and Anita Campion (AZMJ). IFC especially thanks all of the organizations whose case studies are included in this report for sharing their information. The case study research was undertaken by Rabo International Advisory Services (RIAS) on behalf of IFC in 2011. The report has also benefited from a broader collaborative effort with contributions from individuals at the following organizations: Agence Française de Développement (AFD), Alliance for a Green Revolution in Agriculture (AGRA), German Federal Ministry for Economic Cooperation and Development (BMZ), the Food and Agriculture Organization of the United Nations (FAO), Gesellschaft für Internationale Zusammenarbeit (GIZ), the International Fund for Agricultural Development (IFAD), Kreditanstalt für Wiederaufbau (KfW), Making Finance Work for Africa (MFW4A), Stanbic Bank, Technoserve, United Nations Capital Development Fund (UNCDF), and World Bank. Last, but not least, this work was completed under the leadership of the co-chairs of the G-20 SME Finance Sub-Group: Susanne Dorasil (Germany), Ayşen Kulakoglu (Turkey), and Jonathan Rose (United States). Innovati v e Agric u ltu ral S ME Finance Models 7 Introduction What innovations can help bankers in developing countries who wish to finance agricultural small and medium enterprises (SMEs)? This report tries to answer this question by isolating promising cases of emergent and innovative financing, risk mitigation, and distribution models. The paper identifies key elements observed across case studies. In the report, case studies are documented, models are observed, and patterns are determined. The report primarily addresses private sector financial institutions in devel- oping countries and therefore focuses on models from the private sector perspective. A previous G-20 report (“Scaling up Access to Finance for Agricultural SMEs — Policy Review and 4 Recommendations” ) addressed policy makers and discussed public sector banks and related policy issues at length. That paper was designed to contribute to the formulation of an agricultural SME finance policy framework. The current report elaborates further on promising and innovative approaches to agricultural SME finance in three types of agricultural SME finance country environments. It therefore endeavors to summarily assess both the salience of these approaches, called “models” in this text, given various country environments. A banker in Malawi faces fundamentally different challenges from a banker in Mexico, and therefore different models might be more useful for her. At the same time, a banker in Ghana has quite different challenges in financing cocoa rather than maize. All of these innovative models can help the banker finance agriculture by: (i) replacing traditional collateral with new types of security (“financing” models); (ii) mitigating risks more effectively (“risk mitigation” models); or (iii) lowering transaction costs (“distribution” models). Each of these models is illustrated by the cases included in the annex. Most of the case studies are based on a background stocktaking report compiled by Rabobank International Advisory Services for IFC, as well as a databank with information compiled by IFC from other sources, on more than 100 cases. Lending to agricultural enterprises can be an important opportunity for growth due to a variety of factors. There is a rapid sector expansion due to increasing demand for food and commodities in general. There is increasing emergence and development of profitable agricultural value chains. Small and medium- sized agricultural enterprises can be more productive and efficient if markets for their goods function properly. There appears to be a lack of correlation of the agricultural business with financial markets, and therefore an opportunity to diversify the bank’s portfolio. There continues to evolve and emerge innova- tive lending and risk mitigation models that help to better manage the agricultural portfolios. And finally, data from 93 countries reveals comparatively promising lending spreads and similar non-performing loans (NPLs) in those countries with significant agricultural sector shares. 4 IFC (2011) 8 Glob al Partn er ship for Financial I nc lu si on The need for agricultural commodities grows as populations grow and adopt dietary habits that demand higher protein content and higher quality foods for emerging middle classes in urbanizing populations. The total value of all agricultural activities in the world increased by more than a third from 2002 to 2010, to reach 7,043 billion dollars. With agricultural demand estimated to grow 50 percent by 2030, the world’s 450 million smallholder farms will play an increasing role in food provision. Highly organized value chains with strong buyers such as food processors, distributors, and commodity traders have emerged in many markets and can help to secure lending to those farmers supplying to these buyers. It must be noted, however, that the greatest benefits may be secured by those larger farm- ers capable of delivering high volumes at precise intervals while meeting stringent quality standards, 5 thus reducing transaction costs for high-volume buyers. However, the vast majority of farmers in emerging economies are outside these high-value supply chains. Providing financial services to the agriculture sector has many challenges that multiply significantly as financial institutions move from larger farmers and high value chains to smallholder farmers and lower value crops, particularly subsistence food crops. It is indeed this very high degree of heterogeneity of farmers that makes it difficult to think of a single model and approach that can make a difference. Understanding and better classifying farmers while dealing with their specific challenges is the first step in thinking about potential solutions and innovative approaches. Lending to farmers has a number of challenges. Recent innovative models show some promising signs in dealing with risk assessment and mitigation, lowering transaction costs, exploring delivery channels to farmers, and enabling better flow of information for banks to assess opportunities. The objective of this report is to collect information from case studies, draw observations, and derive lessons learned thus far. This report is organized into three main sections. Section 1 sets out the context by describing the par- ticular challenges and opportunities related to financing agriculture, followed by defining the target group of agricultural SMEs. Section 2 describes a set of innovative financing models. Section 3 then pro- vides a framework for an indicative assessment of these models by gathering and forming observations from the case studies. The model observations are then highlighted in three types of country contexts in which bankers in developing countries may find themselves. Preliminary assessment results are pre- sented. The report concludes with key lessons learned on innovative agricultural financing, relevant case studies, and an outlook on further work in this area. Annexes present methodologies and case studies representing the various models. 5 Wiggins, Kirsten, and Llambi (2010) Innovati v e Agric u ltu ral S ME Finance Models 9 Executive Summary Financial institutions face opportunities as well as chal- poverty over other sectors.2 Though widely recognized lenges in providing financial services to the agricultural for its social impact, agricultural investment — particu- sector. The sector clearly lacks financing, with the one larly to small and medium enterprises (SMEs) — is also percent commercial lending share to agriculture in recognized by the financial sector as a profitable growth Africa often cited as an example. Farmers are a very het- business. Agricultural enterprise finance offers banks and erogeneous group with varied plot sizes, production financial institutions a major growth opportunity for the capacity, mechanization, resources, and expertise. following four reasons. First, global food demand is However, they all share a limited ability to access appro- likely to grow 50 percent by 2030, led by emerging priate financial services for their farming activities and middle classes in urbanizing populations. Rapid sector overall household expenses. In this context, opportuni- expansion through strong buyers with profitable value ties to expand financial services to farmers are high- chains will drive product procurement from SMEs. lighted by innovative financing, risk mitigation, and Secondly, financing allows farmers to invest in new tech- distribution models, observed in 100 case studies exam- nologies and access better inputs, thus increasing yields ined for this purpose. Innovation is defined by: (i) new significantly and contributing to food security and better models that are not widely used yet; (ii) adaptation of incomes. Thus, access to finance will help farmers move existing models in a developing country context; and, from the subsistence/semi-commercial level to become (iii) downscaling models for smallholders. Overall, these commercial farmers. Third, agricultural lending pro- innovative models mobilize additional resources for agri- vides the opportunity to diversify larger portfolios; data culture through the private sector institutions that from the economic crisis of 2008 particularly supports finance agricultural SMEs and farmers. The model cases this point. Fourth, innovative financing, risk mitigation, also show the need to forge partnerships between vari- and distribution models hold some promise that the ous private sector actors along agricultural supply chains, risks and costs of agricultural SME lending can be man- as well as between private and public sector institutions. aged. Given these factors, lenders are beginning to rec- ognize the growing potential and profitability of lending The Opportunity to these “generally feared but little understood” agricul- tural enterprises.3 Three-quarters of the developing world lives in rural areas, and about nine out of every ten depend upon agri- The Challenge culture for their livelihoods.1 Agricultural investment is often regarded as one of the most efficient and effective The leading challenges facing lenders that want to engage ways to promote food security and reduce poverty, with in the agricultural sector may be broken down into some studies demonstrating a four-fold reduction in three main areas: the unique problems of agriculture, 1 World Bank (2007) 2 Oxfam (2009) 3 Ibid. 10 Glob al Partn er ship for Financial I nc lu si on high transaction costs, and sub-optimal policy and persist between farmers and buyers, and formal or regulatory environments. informal contracts provide security to lenders. Unlike with typical short-term loan schemes, agricultural Main Risk Management Models loan products must reflect the unique characteristics of agricultural production. Namely, products must cater Although the financing models detailed in this report are to seasonal production with long and diverse gestation designed to minimize the risk of default, various risk periods. Lenders face irregular payments and slow rota- mitigation models may be a useful complement to trans- tion of invested capital and are sometimes challenged to fer key risks to markets. Insurance products, such as design appropriate financial products due to lack of credit life products, have been mainstreamed in the sufficient knowledge of local agricultural and environ- market. There are, however, emerging health, produc- mental characteristics, as well as complex liquidity tion, and weather insurance products that can signifi- management. Agricultural lending also involves sys- cantly improve the security package and ultimately temic, covariate risks — all of which are intensified for reduce the default risk to lenders. Personal insurance term finance or for enterprise development in general. products may be formally tied to financing opportunities Farmer diversification does little to divert risk from those through health credit products, or informally tied as systemic risks that affect all of the creditor’s activities and micro-insurance coverage expands. Crop and weather potentially the entire agricultural finance portfolio. insurance products under certain preconditions and cir- cumstances could provide solutions to dealing with crop While agriculture in general poses many unique risks to losses. There are also risk management instruments that lenders, agricultural finance usually also involves high can deal with commodity price risks, but their use in transaction costs due to low population densities, low most low-income emerging markets is still very limited. infrastructure quality, and distant locations. Inefficient agricultural markets can limit the viability of rural finan- Main Distribution Models cial services. Distortion in the production and financial markets can also affect the profitability. Distribution models — including mobile banking, branchless banking, and mobile payment systems — Main Financing Models help support the financing models. As banks provide low-cost financial services to the rural agricultural The case studies used herein revealed a large number sector, they connect with their clientele through a of financing models that reflect the heterogeneity of transaction history, learn about their needs, and countries and commodities. To facilitate observations, develop relationships — all of which are essential to these financing models are divided according to their build and maintain a profitable loan portfolio. repayment source or collateral into three categories: Distribution models may also reduce banks’ transaction farmer, movable collateral, and buyer. In financing costs through efficient loan disbursement and repay- models targeting the farmer or groups of farmers, ment systems. Overall, distribution models provide collateral generally involves cash flow analysis by access to a clientele that previously was out of reach. banks in order to underwrite anticipated earnings, overall savings, and/or group guarantees. Financing Observations models using movable assets as collateral often include leased equipment or harvested commodities in ware- Close examination of key elements within the collected houses. Financing models that rely on buyers as the case studies reveal information under two main head- repayment source are based upon an overall value ings. First, channels of distribution to farmers (such as chain analysis in which strong business relationships banks, buyer, inputs, local cooperatives, or microfinance Innovati v e Agric u ltu ral S ME Finance Models 11 institutions) can vary and should be tailored to specific II. In Environment II (strong business environment, needs and capabilities, depending on the type of farmer low agricultural productivity), we find more diver- and their production (type of crops or livestock). In sified models, including significant movable asset order to access farmers in value chains effectively, it is models, some farmer risk models, and buyer risk necessary to look at farmer linkages in value chains and models, while warehouse receipt financing, indi- to involve farmer organizations. This is particularly true rect lending, mobile banking, and certain risk for smallholder farmers. Providing technical assistance insurance models are most relevant given the better and/or extension services to farmers also potentially business enabling environment. adds value to financing and may be associated with higher yields and incomes. The second observation III. In Environment III (high agricultural productivity), derived from the case studies relates to characteristics of there is a vast array of models, including a prepon- innovation and its role in dealing with credit risk. Some derance of farmer risk and buyer risk models, but notable examples found in the cases were an increase in most of the innovative models discussed can be rel- use of first loss guarantees, cases with credit risk assess- evant for this environment. ment combined with agronomic models, and use of spe- cific credit scoring systems for agriculture lending. Noted It is important to acknowledge that the inventory of more extensively was the use of movable collateral case studies used in this report did not include many and a more flexible approach to credit requirements. cases from countries in Environment III, as the focus of Alternative channels (e.g., mobile banking, payments, this report was on examining models in lower-income, etc.) play an important part in mobilizing different types emerging markets. Therefore, if the inventory were to of collateral, savings, and repayment options. There are include more cases from countries in Environment III, some cases where insurance, particularly weather insur- we would expect to see significant reach numbers in all ance, seems to have played an important role as part of a model types. broader package that included both access to improved inputs/technology and access to finance. Lessons Learned Country Environments Across cases, some patterns emerged in terms of what seem like good practices. For one, farmer segmenta- On the basis of three generalized types of country envi- tion is important to enable bankers to start differenti- ronments, this report provides an indication of where we ating classes of farmers. Segmentation allows banks likely encounter the various models described in Section and other financial institutions to locate specific 2. This country environment framework provides a strate- growth opportunities for distinct farmer groups, and gic view of the cases that have been compiled for this this is best accomplished by deepening knowledge of report. Case information has been gathered through a local conditions and understanding of the needs of combination of primary and secondary sources, and cer- agriculture. Looking at farmers within their value tain case examples have only limited information avail- chains or organizations and combining the resulting able. Different models are more or less relevant in the information into financial packages is most likely to three types of country environments. Key findings are: be effective in reducing risks and costs. Financing agriculture is more effective when it is part of a I. In Environment I (weak business environment, low broader package that combines both financial and agricultural productivity), buyer-driven financing non-financial services to the farmers with the objec- models tend to reach larger numbers of farmers, tive of improving yields and quality (through access and “tight” value chain financing seems to be the to better inputs and extension) and ensuring access to most relevant and sustainable model. markets for selling their produce. A third element 12 Glob al Partn er ship for Financial I nc lu si on visible across case studies is that risk management also promote further innovation. This paper is a first matters: insurance and risk-sharing can be important, attempt to collect existing innovative models in order although they need to be appropriate for the specific to draw some early observations. situation. Furthermore, it seems that financial insti- tutions would be better equipped to provide these It can also be concluded that no single innovation can with pre-existing risk management capabilities — be considered the miracle or “silver bullet” solution, meaning that banks need to have the capacity to assess and this is despite the various calls over time to come farmer credit risk and be able to identify bankable up with grand schemes and search for big solutions. opportunities. Insurance and risk-sharing arrange- Observations from the innovative models show that ments can then increase the bank’s level of comfort success takes patience, careful planning, understand- and enable them to increase their reach to more ing of the local context, and attention paid to details farmers that would otherwise be on the margin of the during implementation. There are many small ingre- decision to lend. dients that, when put together, make the innovative case work. As seasoned bankers often say, you need to The challenges of lending to small and medium use “shoe leather” to make things happen. Thus solu- agriculture are not insurmountable for institutions tions, rather than reliance on large schemes, need to that rely on innovative and targeted approaches. be based on a number of coordinated actions aligned Important among the lessons to apply in emerging with the overall policies to improve access to finance solutions is to make use of value chains, local knowl- in the agricultural sector. These, in turn, must have edge, and producer organizations to lower risk. The the objective of improving the livelihoods of farmers key issue is addressing the variety of risks in agricul- and promoting food security. ture lending while keeping transaction costs con- tained. Emerging success stories of innovative Potential Areas for Policy approaches depend on the farmers, types of crops, Interventions locations, and conditions. In-depth knowledge and analysis of these can lead to the most added value for The findings from the case studies support the policy the financing of farmers. recommendations made in IFC’s previous report (“Scaling Up Access to Finance for Agricultural SMEs: Policy Conclusions Review and Recommendations”, October 2011). They further highlight certain areas where policy interventions and We can conclude that many innovations in financing the G-20 convening power could indeed strengthen for the agricultural sector already exist, but they are not the effectiveness and scaling up of financing for agri- widely known nor have they been systematically moni- cultural SMEs. These areas include: tored and evaluated. Many of the innovative models are still relatively new, but through time and the use of 1. Support for first loss/guarantee funds for agricul- appropriate systems to monitor and evaluate their ture, particularly focusing on smallholder farmers achievements, we will be able to draw more complete and agricultural SMEs. This support should lever- lessons that can help in scaling up and replicating age the Global Agriculture and Food Security them. This will help us better understand what works Program (GAFSP) as well as the Global SME and what does not, and under what conditions. What Finance Initiative, both of which have been sup- seems to be missing at this point is some repository of ported by the G-20 already, rather than starting a innovative models, systems to monitor, and methodol- new initiative. However, the effort may require ogies to evaluate them. In addition, we need to think of additional resources if the scale of the activities is incentives to strengthen existing innovative models and to expand significantly. Innovati v e Agric u ltu ral S ME Finance Models 13 2. Provide support for catastrophic insurance 6. Promote Private Public Partnerships (PPPs) by approaches to protect farmers and financial insti- which governments could leverage private sector tutions from severe losses. Since this industry is funding and management to improve longer-term still evolving, donor and partner interventions can investments in agriculture infrastructure and provi- play a critical role in accelerating its development sion of technical services. Agriculture-related infra- and deployment in emerging markets. structure could include warehouse facilities for improved storage of commodities, cold storage, 3. Promote the creation a forum of large agribusi- irrigation infrastructure, basic processing of certain nesses that could be encouraged to leverage their food commodities for local consumption, etc. networks in emerging markets and create oppor- tunities for attracting financial institutions that 7. Support capacity building for financial institutions could fund parts of their value chain, like local in emerging markets and facilitate its further sup- small traders, processors, farmers, etc. Financing port by donors, development finance institutions/ could be linked and become the catalyst for tech- international finance institutions (DFIs/IFIs) and nology improvements and promotion of envi- foundations. Capacity building is critical to provide ronmental and social standards along specific necessary skill transfer to financial institutions in value chains. order to better understand the agriculture sector, analyze risks, develop appropriate lending and 4. Create mechanisms to promote the adoption of other financial products, find cost-effective distri- technologies for agriculture (“agriculture pull bution channels to reach smallholder farmers, and mechanisms”) that could increase yields and develop the skills to forge value chain partner- improve quality for crops, particularly food crops. ships. Experiences thus far have indicated that it is There is a huge capacity to increase yields and also important to help financial institutions iden- improve quality, particularly in the African con- tify bankable opportunities in the agriculture text. Mechanisms could be modeled after a a 2005 space to quickly develop a pipeline of projects to effort in the health sector to promote vaccinations provide financial services. in Africa. 5. Strengthen producer organizations as important aggregators for delivering financial and non-finan- cial services to smallholder farmers. This can involve capacity building for financial and mana- gerial skills as well as improved corporate gover- nance. There are already a number of NGOs and initiatives that work to strengthen producer orga- nizations, but a more conscientious effort and a bigger scale is perhaps needed. 14 Glob al Partn er ship for Financial I nc lu si on CHAPTER 1 Major Challenges and Opportunities 1.1 The Challenge of Lending to and price risks have a large impact on the profitability Agriculture and repayment capacity of the borrower.6 Moreover, risk mitigation mechanisms such as crop insurance or Seasonality with long gestation hedging are rarely available. periods Some of these risks, in particular weather and price Agriculture is very seasonal, from planting or live- risks are systemic, which means that ultimately stock birth to harvest or slaughter with long gesta- whole agricultural finance portfolios are affected in tion periods. The result is that cash flows are highly addition to individual farm-level income losses. seasonal and sometimes irregular, with earnings While the activities may be diversified (crops and concentrated in certain times of the year. As such, livestock combinations, for example), the risks are there is a slow rotation of the invested capital as still often concentrated (a drought would affect all investments are spread over longer time horizons activities and their market prices). Unless the banker than for non-seasonal businesses. For the banker, manages to protect the loan portfolio against the this means that short-term agricultural credit may most systemic risks, the lack of true risk diversifica- need to be repaid in “lumpy installments,” some- tion exposes the bank to the risk of default or at least times over multiple seasons. It also means that frequent rescheduling. farmers require flexible and targeted savings and term finance products to meet their specific needs. The origins of this lack of true risk diversification lie From the banker’s point of view, irregular repay- in the risk-return dilemma that farmers face. In order ment schedules make liquidity management more to maximize profits, producers need to take on challenging and require costly investments in higher concentration and price risks by specializing, developing customized loan products in an unfa- often by adopting high-yielding varieties, focusing miliar sector. on lucrative niche products, and generating econo- mies of scale. By doing so, the farmer may have to Exposure to systemic risks adopt alternative risk mitigation strategies such as contract farming instead of diversifying into a range Most agricultural SMEs, in particular producers, are of small-scale activities. By contrast, a well-diversified not truly risk diversified. Emerging farm businesses portfolio including off-farm activities might be and SMEs tend to be either very concentrated in one much less profitable but safer through less exposure activity or to have a portfolio of activities that are all to the risk of livelihood threatening losses and exposed to similar key risks like droughts. Production resulting loan defaults. 6 Supply disruptions mainly caused by weather vagaries along with high price elasticity for major agricultural commodities lead to significant price swings. For example, world sugar prices soared to a 29-year high of nearly 30 cents a pound in early 2010 before falling back to half that level by early summer, remaining at 50 percent higher than average over the past 20 years. McConnell, Dohlman, and Haley (2010) Innovati v e Agric u ltu ral S ME Finance Models 15 This risk diversification vs. specialization returns exceed the profits they can make with these relatively dilemma is magnified for term finance for investment small loans. If farmers evolve from smallholders to projects or for enterprise development in general. Here, more specialized farmers, the lender must analyze the the exposure is longer and the risks are much harder to SME in all its details (e.g., the ability and character of assess, as outcomes depend more on the character and the management, the prospects for the product, cash capacity of management, because the loan is not based flow forecasts, the position of this SME relative to on specific transactions. competitors, etc.) in order to understand the risks involved. To cover such costs, loans must be signifi- Limited collateral cantly larger, reaching a size that substantially exceeds the absorption capacity for capital of the SME — Agricultural financial service providers have few instru- hence the financing gap. ments at their disposal to manage these various risks; they therefore tend to protect themselves through exces- Farming is also very heterogeneous, and deep sector sive credit-rationing and by relying heavily on traditional information is often not readily available. Farming land collateral. However, agricultural borrowers’ assets households in particular often have a wide range of are less suitable as collateral than for example, urban real crops and activities that can make the assessment of estate. In fact, farmers and their producer associations creditworthiness more complex and costly. frequently lack the collateral traditionally required by banks for larger and longer-term loans. Due to legal and Banks’ competing priorities administrative impediments as well as cultural factors, rural assets are often not registered and consequently Many banks in emerging markets face a number of may be more difficult to foreclose and sell. Even where priorities such as expanding their product offering these constraints are less binding, collateral is a poor mostly to urban SMEs and consumers or leveraging protection against massive defaults due to covariant risks. their branch networks and presence in urban loca- The result is that required collateral ratios are much tions. They also need to improve their systems (e.g., higher than they would be otherwise. IT, MIS, investment in risk management, etc.). In this context, expanding to the agricultural sector with all Higher transaction costs its particularities, without presence in rural areas, and with a lack of technical expertise, seems a significant Agricultural financing involves higher transaction costs challenge and appears to be a lower priority. The key than in urban areas given the distances, lower popula- issue here is for banks to understand the sector but, tion densities, and lower quality infrastructure. Together, more importantly, to identify bankable opportunities these factors make it hard to aggregate agricultural loans in the agricultural space. into portfolios that make branches viable. In addition, it can be costly to have branches and staff in remote areas, Limited access to long-term funding handling small transactions. One of the most prominent gaps in developing financial services particularly for External investment and long-term loans, other than rural Africa is poor infrastructure — for example, bad informal loans from family and friends, are available roads, erratic electricity provision, and lack of communi- only for a tiny proportion of SMEs in all economies. cations systems — which impedes effective outreach to Formal external investment is appropriate only for the customers or drastically increases the costs. minority of SMEs that are both growth-oriented and have the business model and management to achieve Financial institutions also face high creditworthiness the necessary growth. Professional investors taking assessment costs with agricultural SMEs that might significant shares in enterprises have to recover the 16 Glob al Partn er ship for Financial I nc lu si on transaction costs of making and monitoring their understanding of banking requirements compound investments and of absorbing the losses from the these problems. enterprises in their portfolio that fail. These costs are largely independent of the size of the investment, In addition, as a previous G-20 report on agriculture which discourages small investments. Investors need finance policies8 has highlighted, there are gaps in reg- one or more reliable exit routes so that they can sell ulations that inhibit the development of private sector their stakes to realize profits and recycle their capital.7 instruments that could provide financing solutions in the agriculture sector. For example, the lack of a legal Long-term finance can be inaccessible, denominated and regulatory environment for inventory financing in international currencies, and therefore expensive. and warehouse receipt lending inhibits the use of these Financial institutions face a timing and currency mis- financing mechanisms. Additionally, addressing the match. Lending maturities are shorter than funding regulatory and taxation issues that discourage the maturities; SMEs usually generate local currency earn- development of leasing could improve mechanization ings and therefore require local currency loan and and upgrading of equipment in agriculture. saving products. 1.2 The Opportunities in Lending to Potentially sub-optimal policy and Agriculture regulatory environments Three-quarters of the developing world lives in rural Government activity in promoting food security indi- areas, and about nine out of every ten individuals rectly affects agricultural markets through input and depend upon agriculture for their livelihoods.9 output prices and the overall credit culture. Agriculture Agricultural investment is often regarded as one of the is politically sensitive, because it is at the heart of food most efficient and effective ways to promote food security, a primary concern of governments, and security and reduce poverty, with some studies dem- therefore prone to government interventions. These onstrating a four-fold reduction in poverty over other interventions can include mandatory lending quotas, sectors.10 Though widely recognized for its social preferential lending programs for specific target impact, agricultural investment — particularly to groups, interest rate subsidies, mandatory loan farmers and agricultural SMEs — is also recognized rescheduling, or even loan forgiveness in some cases. by the financial sector as a profitable growth business. Agricultural enterprise is identified as a major oppor- Inefficient agricultural markets can be a barrier to tunity for banks and financial institutions for the fol- developing rural financial services. Agricultural value lowing four reasons: chains can be poorly organized and lack transparent pricing. In some cases, the financial environment can First, global food demand is expected to grow 50 percent be distorted by the presence of state banks and subsi- by 2030, led by increasing global population (expected dized credit, casting agriculture more as a social issue to reach 7.5 million by 2020), particularly in emerg- rather than an economic activity, and thereby creating ing markets where the middle class is growing as barriers for the evolution of private sector solutions in well. According to FAO figures, by 2018, world food financing the agriculture sector. Low financial literacy consumption is expected to increase by approximately rates, especially among small farmers, and a limited 30 percent compared to the 2005 reported figures. 7 Oxfam (2009) 8 IFC (2011) 9 World Bank (2007) 10 Oxfam (2009) Innovati v e Agric u ltu ral S ME Finance Models 17 In addition to the population growth, the per capita very profitable opportunities. In fact, between 2004 caloric consumption is increasing and there is a major and 2008 commodity prices almost doubled and, shift in caloric sources with a projected doubling of despite a dip in 2009, prices had again exceeded again meat consumption in China, India, and Africa by their 2008 levels by 2011. Another possible indication is 2030.11 The consumption of non-staple crops such as that the NPLs in countries with a high agriculture coffee, cocoa, and tree nuts is also expected to increase sector share in GDP saw their NPLs decline during the in emerging markets. To meet the growing demand for recent economic crisis, compared to developed coun- staple food and non-staple crops, there needs to be sig- tries with low agriculture sector share in GDP, which nificant investments in the agriculture sector. Some saw their NPLs significantly increase (see Annex B). estimates indicate that additional investments of US$83 billion per annum may be needed, most of which Fourth, innovative financing, risk mitigation, and dis- would have to come from the private sector.12 tribution models hold some promise that the risks and costs of agricultural lending can be managed. Given Second, farmers in emerging markets can contribute these, lenders are beginning to recognize the growing to food security and improve their incomes by potential and profitability of lending to these “generally increasing their productivity and the quality of the feared but little understood” agricultural enterprises.14 crops they produce. For this they would need to invest in new technologies, access better inputs, improve The above factors indicate that there is an unmet farm and off-farm practices, and invest in sustainable demand for credit in the agriculture sector, while that production methods for their crops. In particular, the credit is needed to address the growing demand for world’s 450 million smallholder farmers, over 90 per- agricultural commodities and shifting preferences cent of whom are in Asia and Africa, could play an towards higher value food sources. At the same time, important role in food security and also improve their the supply of agricultural commodities is coming incomes. Access to credit can play a key role: without under pressure stemming from growing water scar- credit smallholder farmers use sub-optimal inputs city, climate change impacts that can affect produc- and farming practices that lead to low yields and often tion in some areas of the globe, and fears of further resort to unsustainable practices of production. Access deforestation. These pressures indicate that expansion to finance will contribute thereby to moving farmers in agriculture production needs to happen with the from subsistence/semi-commercial into commercial use of resources more efficiently. Thus, investments in farmers and improve the livelihoods of those farmers sustainable production systems and methods will be a that are already commercially oriented. According to a key driver in the agriculture sector. recent report, the demand for credit by smallholder farmers globally was very roughly estimated to be In addition to resource efficiency, another growing nearly as high as $450 billion.13 trend is the reliance on smallholders. Many buyers consider sourcing from small farmers critical for Third, for financial institutions, agricultural lending securing adequate supplies and diversifying their provides the opportunity to diversify into larger and sources. In addition, consumer preferences for sustain- broader portfolios. For example, during the economic ably produced agricultural commodities create incentives crisis of 2008, agricultural commodities were enjoying for buyers to shorten supply chains and source more high prices and commodity sectors were showing some directly from farmers to ensure that the goods are 11 FAO (2009) 12 According to IFC internal estimates. 13 Dalberg, Citi Foundation, and Skoll Foundation (2012) 14 Ibid. 18 Glob al Partn er ship for Financial I nc lu si on sustainably produced. Also, buyers may source specific ƒƒ Climate change impacts that are straining the qualities of commodities that rely on smallholder pro- supply of agriculture commodities. Financing is duction. The trend for greater traceability of production needed for investments in sustainable production for quality purposes and for verification of sustain- systems and climate adaptation technologies. ability creates stronger linkages along supply chains, ƒƒ Emergence of new markets for niche products, improves information flow, and enables buyers and higher value crops, and certain crops/food products financial institutions to get closer to smallholder farm- with characteristics valued by consumers, such as ers. Smallholder financing models can contribute to certified products. Responding to new markets and farmers adopting better on- and off-farm practices that meeting emerging consumer preferences and lead to sustainable production, improve the quality of demands requires investments that bring small- crops produced, and reduce post-harvest losses. holder farmers in particular closer to value chains and key growth markets. In summary, the opportunities to lend or, more gen- erally to provide financial services to agriculture, stem 1.3 Target clients from the following trends: ƒƒ Increased demand for agriculture commodities This report defines agricultural finance for SMEs as due to increase of the population and the change in financial services for small and medium enterprises dietary habits. Financing is needed, among other engaged in agriculture-related activities such as farm- solutions, to enable the use of improved inputs and ing/production, input supply, trade, and processing. better on-farm practices to increase supplies and Agribusiness not involved on the production side can improve yields, as well as to generate improve- be segmented similarly into non-agricultural SMEs in ments in quality and better post-harvest practices to terms of classification based on the number of lower post-harvest losses and add value to the crops employees or annual turnover, and thus differentiated produced (further processing). from microenterprises and large agri-businesses. Figure 1  Where the funding opportunities lie Lower Climate Explore New Higher Post-Harvest Change Increase Access Market Objectives Productivity Losses Adaptation to Markets Trends Farmers Input Suppliers Traders/ processors What to finance Technology, Post-harvest Sustainable Market Value chains inputs, systems production infrastructure mechanization systems and value chains = Very relevant  = Relevant Innovati v e Agric u ltu ral S ME Finance Models 19 Figure 2  Farmer segmentation and key characteristics Key characteristics Large Farmer Land Size of cultivated land is large (>500 ha) Labor Mainly depending on skilled labor Annual farm net income Technology Fully mechanized (as function of skilled laborer (SK) income) Resources Formal bank loans and or external capital, skilled management Production Fully commercial and often dollarized Capacity Good market access, own storage/logistics, and market information > 2 * SK Value Chain Well positioned within the value chain Large Farmer Medium-sized Farmer Land Cultivated land is medium-sized (20–500ha) < 0.8–2 * SK Medium-sided Labor Combination of family members and external labor Farmer (emerging) Technology Partly mechanized Resources Limited access to formal bank loans Production Largely commercial Capacity Reasonable market access but limited access to information Commercial < 0.8 * SK Value Chain Weaker position, stronger in cash crops Smallholder Commercial Smallholder Land Size of cultivated land is small (2–20ha) < 0.3 *SK Semi-commercial Smallholder Labor Primarily family labor Technology Minimal mechanization Resources Mainly informal finance Production Partly commercial (at least one cash crop) Capacity Marketing through group structures Subsistence Farmer Value Chain Position depending on group strength Semi-commercial Smallholder Land Size of cultivated land is relatively small (e.g., <2ha) Labor Primarily family labor Technology Low technology, little access to know-how Resources Limited resources (capital, skills, labor, risk mgt, etc.) Production May produce subsistence or commercial commodities, with on-farm and off-farm sources of income Capacity Limited capacity of marketing, storage, and processing Value Chain Often vulnerable in supply chains Subsistence Farmer Land Size of cultivated land is relatively small (e.g., <2ha) Labor Primarily family labor Technology Low technology, little access to know-how Resources Limited resources (capital, skills, labor, risk mgt, etc.) Production Subsistence commodities, with part of their income from off-farm activities Capacity Limited capacity of marketing, storage, and processing Value Chain Extremely limited, no linkages to supply chains 20 Glob al Partn er ship for Financial I nc lu si on Who are agricultural SMEs? and large), large enterprise farm operators, agriculture production cooperatives, and other forms of producer Agricultural sector SMEs not involved in the pri- organizations. For simplicity, we will call them farmers. mary production side (i.e., traders, processors, input When it comes to primary production (farming), the suppliers) have their own financing needs just like traditional SME segmentation is more challenging in any other SMEs. Working capital, funding for acquisi- these cases, as structural differences in farm size and tion of assets (movable and real estate), cash flow income result in varying earning potential for farmers. management services, and insurance are often needed For example, cash crops or generally high value crops by these agricultural SMEs as well. As with SMEs in (e.g., fresh fruits and vegetables, spices, etc.) generate other non-agriculture sectors, non-primary produc- much higher income on smallholdings compared to staple tion agricultural SMEs face similar obstacles in access- crops. Thus, it is useful to use a range of characteristics ing financial services. There are, however, some key to segment farmers into subsistence/semi-commercial differences. For one, non-primary production agricul- smallholders, commercial smallholders, medium- tural SMEs face some of the systemic risks that affect sized farmers, and large farmers, as the financial services agriculture production and products. Traders and pro- they require and the security they can provide will vary cessors are exposed to price swings of the commodi- significantly. The chart below illustrates a segmentation ties that they buy before selling them as processed or of primary producers. unprocessed goods. Price declines, for example, affect margins and the ability of these entities to repay their The segmentation of agricultural producers is a chal- loans. Catastrophic crop losses could create financial lenging but very important task that those who wish problems for processors and traders, as they will not to offer financial services need to undertake. The seg- find produce to trade and process. Similarly, if there is mentation is challenging because farmers, or farming a catastrophic crop failure, traders and processors will enterprises, are very diverse, and trying to categorize not find enough produce to purchase, which means them is not an easy task. This may be partly due to that they will be operating well below capacity based country differences or differences among crops even on fixed assets, and therefore operations will be nega- within the same country. For example, a small maize tively affected. These systemic risks relating to price farmer in Ukraine may have similar characteristics as and crop production have a significant impact on the a large farmer in Malawi or Zambia. The environment cash flows of non-primary production agricultural and context that a small rice farmer in Sri Lanka is SMEs since commodity prices and crop production facing may be very different from a small cinnamon tend to be much more volatile compared to prices and or tea farmer in the same country. A survey of differ- volumes of industrial goods or services. ent crops in Tajikistan showed that net income per hectare fluctuated from around $400–600 per hectare Aside from price and crop production risks, the rest of for wheat, barley, and corn to well above $12,000 per the issues for accessing finance are quite similar hectare for vegetables (onions, tomatoes, garlic, and between non-agricultural and agricultural sector cucumber). Within the same survey, a farmer with 15 SMEs. In terms of innovations in financial services, hectares under cotton production has the same net agricultural SMEs not involved in primary production income as a farmer growing carrots in a 1.5 hectare can access similar instruments as farmers, such as farm. Thus even among smallholder farmers within supply chain finance, equipment leasing, and ware- the same country, which crops and crop combinations house receipt/inventory finance. they choose to grow makes a great deal of difference in terms of income and cash flow. The importance of Agricultural sector SMEs involved in primary farmer segmentation is not so much to determine production are basically farmers (smallholders, medium, which farmers should or should not be offered Innovati v e Agric u ltu ral S ME Finance Models 21 financial services, but to learn what financial services Commercial smallholders are farmers (or farming are needed in each market segment and how best to households) most commonly growing cash and serve the various segments. Some financial institu- higher value crops (e.g., coffee, cocoa, cotton, tea, tions may also use segmentation to pick the “low sugar, spices, fruits, and vegetables, etc.), often inter- hanging fruit” and begin offering services to the cropping with some food crops and livestock to sup- farmer segments that are easier to understand and plement the household consumption. Their income is work with before proceeding to others. more dependent on the surplus they produce from these cash crops, while some income may come from Large and medium-size farmers tend to be commer- off-farm activities (e.g., working in other farms or in cially oriented and often are customers of formal the village, etc.). financial institutions. These farmers often produce large volumes and have a significant income from In the case of commercial or cash crops, larger farm- farming activities. They own farming equipment, ers generally depend less on a single cash crop as part have land titles, often employ labor, and usually have of their farm income, while smaller farmers depend sufficient financial information and collateral that to a much larger extent on a single cash crop as part of would satisfy formal financial institutions and attract their farm income. For example, medium and larger their attention. Large and medium farmers usually cocoa farmers in Sulawesi draw around 56–62 percent demand specialized financial products, like crop of their farm income from cocoa (their primary cash loans, crop insurance, and loans or leases for farming crop), while small cocoa farmers have around 97 percent equipment that are tailored to their needs and reflect of their farm income coming from cocoa. On the other the cash flow seasonality in terms of repayment. hand, larger farmers depend much less on non-farm However, large and medium commercially oriented income compared to smaller farmers, particularly the farmers are by far the minority in most emerging less commercially oriented ones. For the same sample economies, particularly the poorer ones. of cocoa farmers in Sulawesi, larger and medium cocoa farmers have no income outside of farming, while Smallholders are by far the largest and most diverse smaller cocoa farmers have around 7 percent of their category of farmers in emerging economies. In terms income outside farming. of land area, estimates indicate that 85–90 percent of all smallholders have less than 2 hectares of land. For When it comes to small farmers that do not have cash this reason, smallholders need to be segmented fur- crops, their reliance on a single crop can be much ther in order to best meet their financial services smaller: in a survey in Andhra Pradesh, where there needs. At the very bottom, we have the subsistence was no cash crop dominating the area surveyed, only and semi-commercial smallholder farmers. These 11 percent of small farmers planted a single crop such typically grow staple crops that are used primarily for as rice. As with cash crop farmers, the smaller the size household consumption while small surpluses are of the farm holding planted with non-cash crops, the sold in local, informal markets for cash or exchanged higher the percent of total income derived from for other goods. These farmers may also own some income earned off their own farms (in the case of the livestock, both for cash flow and for asset accumula- above sample, 8.5 percent of the smallholder rice tion purposes, and they may draw income outside farmer income came from non-farm activities versus their own farm (e.g., working in larger farms) or even 4.1 percent for the larger rice farmers). outside agriculture (e.g., working in a village shop). The income of these farmers tends to be diversified Commercial smallholders can be further sub-divided and it is more of a household income rather than indi- into those that grow crops that form part of a tight vidual farmer income. value chain or a looser value chain. Tight value chains 22 Glob al Partn er ship for Financial I nc lu si on capture the flow of goods and funds into a single chan- seeds or fertilizers. These smallholder subsistence and nel, generally due to the characteristics of the particular semi-commercial farmers also need savings, money commodity. For example, sugar is delivered to a spe- transfers (e.g., higher dependence on remittances), cific neighboring sugar mill due to the low value per and insurance (life and health). Most often, basic volume, which makes transporting further uneconom- financial products with perhaps some adjustments to ical. There are other examples of tight value chain reflect cash flow seasonality (if it exists) is all that is goods, such as fruits, vegetables, and milk that require needed. Innovation in this case would come more cold or atmospheric storage in order not to spoil from reducing the costs to serve these clients, as they because of their high perishability. Tea and spices have have very small transactions. strong, large buyers that set up the infrastructure and buy from small farmers. There are certain cases where Moving up to the more commercially oriented small- coffee and cocoa have tight value chains when buyers holders, their financial needs start becoming more want to buy specific qualities (specialty coffees or linked to specific farming activities: the need to pur- cocoas) or want to buy coffee and cocoa of certain chase inputs, lease machinery, or perhaps even hire standards (e.g., certified, organic, Fair Trade). The key some seasonal help. Thus their financial needs reflect issue is that tight value chains generally involve greater the higher percentage of their income drawn from the control of the flow of goods and funds to ensure repay- farming activities. The degree of financial product ment (via delivery of the crop) and limit the opportu- customization increases to reflect the particularities of nities of side selling (when the farmer delivers the crops and cash flow seasonality. Of course, com- somewhere else to avoid repayment of loans extended mercial smallholders tend to also have broader finan- under value chain financing models). cial needs similar to the subsistence and semi commercial farmers, but overall their financial needs It is difficult a priori to determine the “tightness” or become more dependent on the investments needed “looseness” of value chains, and one needs to be able to grow, harvest, and manage their crops after the to analyze each situation very well to understand the harvest. Funding for commercial smallholders moves dynamics and the behavior of the actors along these away from funding the household needs, and the dis- value chains before determining how tight or loose tinction between household and commercial agricul- they are. Strong relations between buyers/traders and tural enterprises becomes more explicit. producers enable formal financial institutions to lever- age these relations and reduce the asymmetry of In terms of the supply of financial services, small- information and ensure loan repayment. holder farmers can get financing through informal channels (e.g., local money lenders), through supply Demand and Supply chains (e.g., major buyers, input suppliers), microfi- nance institutions and, in some cases, banks. Usually, In terms of demand, there are vast differences in the looser the value, the more the reliance is on infor- farmers’ financial needs across the spectrum, even mal channels and perhaps in-kind credit in the form among smallholders. The less commercially oriented of inputs from local providers. Tighter value chains and more subsistence-based smallholders tend to have can attract bank financing and financing from buyers broader and more basic financial needs. Loans often (e.g., sugar mills, cotton ginners, milk companies). fulfill a variety of uses, related to both farming and non-farming needs (e.g., school fees, weddings). In looser supply chains where crops can be sold on These farmers treat money as fungible, borrowing to the side and where repayment is difficult to capture meet their overall household needs rather than financ- through delivery, lenders who are near farmers have ing specific farming activities such as the purchase of an advantage, as proximity closes the asymmetric Innovati v e Agric u ltu ral S ME Finance Models 23 Figure 3  Farm Size and Productivity — Evidence from Uganda Can Small Still Be Beautiful in Farming? Smallholders can be quite productive, and some evidence indicates that they are proportionately more productive than larger farmers. With the exception of plantation farming, an inverse relationship between plot size and productivity per hectare has been observed, and has found support in some empirical evidence. Recent research from Uganda has demonstrated that smallholding farmers typically over-report land holdings — discrediting longstanding claims that this inverse relationship is merely a function of under-reported plot sizes and inflated yield data. This same study found that average acre yields of medium farmers were 270 percent more productive than those of large farmers, while aver- age acre yields of small farmers were over 300 percent more productive. These diseconomies of scale may possibly be explained through their labor structures. Small family farms can be maintained by a self-motivated workforce with expertise in local ecology and a labor supply that is easily adjusted for seasonal variability. However, this may not apply for all crops and environments. For example, evidence from Brazil and Argentina indicate that, at least for grains, larger farms tend to be highly productive and cost efficient. There are crops that seem better suited for smallholder farmers, such as coffee, cocoa, and vegetables, amongst others, where yields and productivity amongst these smallholders can be very high. These are more labor intensive crops, compared to grains that rely on scale and mechanization. UGANDA: INVERSE FARM SIZE PRODUCTIVITY RELATIONSHIP 300 250 200 150 100 50 1 2 3 4 5 6 7 8 9 10 DECILES OF LAND CULTIVATED Land Self-Reported Land GPS information gap, facilitates credit assessment, and the crop. Banks can either lend directly to farmers, or makes repayment enforcement easier. Thus local receive repayment through the buyer when farmers money lenders, input suppliers in the area, local credit deliver, or lend to the buyer, who can then lend to unions, credit cooperatives, and microfinance institu- farmers. There are ways that banks and buyers could tions (MFIs) in that location are more appropriate in share risks, share information in credit assessment, reaching out to these farmers. and administer disbursements and repayment. These arrangements all depend on the level of comfort Banks and buyers tend to be involved with smallhold- among the three parties: the farmers, the buyer, and the ers when there is a tight control in the value chain bank. The main collateral in these transactions is the and repayment can be assured through the delivery of future crop deliverable and the cash flow it generates. 24 Glob al Partn er ship for Financial I nc lu si on Banks may ask for other forms of security, such as land, Financial innovation models and approaches aim at equipment, or third party guarantees, but mainly the overcoming these challenges and enabling the provi- reliance is upon the delivery of the goods. sion of financial services to farmers. Financial innova- tion here is related to ways that would improve one Based on results from surveys, often smallholder farm- or more of the following perceived constraints in ers find the banks’ processing takes too much time, lending to the agricultural sector: involves burdensome requirements, and requires col- ƒƒ Enable better risk assessment (e.g., through information lateral levels that are quite high. Distance to the bank by the value chain); branches is also a negative factor. However, a survey for ƒƒ Reduce administrative costs (e.g., through mobile mostly smallholder farmers in India, in Tamil Nadu technology, agency model); and Andhra Pradesh, found that 91 percent of the ƒƒ Combine with other financial services (e.g., savings, respondents listed banks as the most preferred option insurance) and non-financial services (e.g., extension, in receiving financing. There are cases where local technical assistance, certification); and credit unions or financial cooperatives are more flexi- ƒƒ Improve security of the collateral and cash flows ble in their terms and processing and do not often (e.g., warehouse receipt financing, price hedging, require “hard” collateral, while offering loans as mul- insurance). tiples of saving balances after accounts are maintained for some minimum time. Given the plethora of models that exist, we chose to classify them as follows: Given the diverse characteristics of farmers and the ƒƒ Models or approaches for lending directly to farm- nature of agricultural economic activities, financial ser- ers, particularly focusing on smallholders (these are vices for agriculture must tackle specific challenges in mostly for working capital purposes); addition to those inherent in any financial service offer- ƒƒ Models or approaches that finance movable assets; and ings in emerging markets. These specific challenges are ƒƒ Models or approaches for lending through value primarily related to the characteristics of farmers, the chains, involving one or more of several parties nature of the agricultural sector and sub-sectors, and the such as buyers or input suppliers (again, most of the policy and regulatory environment within countries. financing is used for working capital purposes). Innovati v e Agric u ltu ral S ME Finance Models 25 CHAPTER 2 Innovative Farmer and Agricultural SME Financing Models This report examines innovative models to finance ƒƒ Downscaling to smallholders those models and farmers and agricultural SMEs with the goal of find- approaches that have worked in other sectors, com- ing ways to deal with credit risks, given the lack of modity sub-sectors, and/or for the larger and financial information, track record, or acceptable col- medium-sized commercial farmers segment (e.g., lateral by these entities. The vast majority of farmers value-chain financing). and agricultural SMEs, particularly the smaller ones, operate in the informal sector and in rural areas that Financial innovation has the overall objective of using are not usually covered by financial infrastructure models that would mobilize additional resources to (e.g., credit bureaus) and where banks or other finan- the agricultural sector and increase the participation cial institutions have scant local presence in terms of of private institutions in financing agricultural SME’s branches. These factors, as well as the conditions and and farmers. Innovation could also foster new part- risks outlined in Section I, make it very challenging to nerships between various stakeholders, both within assess credit risks. For purposes of this report, an the private sector (e.g., agribusinesses, input suppliers, important objective of seeking innovations is to iden- farmers, financial institutions) and between the tify models and approaches that would help financial public and private sectors (PPPs). institutions find ways to reduce impediments and risks in lending to farmers, particularly smallholder As noted earlier, the innovations outlined in this farmers. Additional objectives of this report are to report are divided into three main types: financing find models that reduce the transaction costs of pro- models, risk mitigation models, and distribution viding financial services to smallholders and provide models. Within the types of financing models, the risk management instruments to smallholders in approaches are divided according to their repay- order to protect them against risks, mostly systemic ment source or collateral into three categories: ones such as price and weather risks. farmer, movable collateral, and buyer. In financing models targeting the farmer or groups of farmers, In defining innovation, we considered the following collateral generally involves cash flow analysis by three criteria: banks in order to underwrite anticipated earnings, ƒƒ New models and approaches not yet widely used overall savings, and/or group guarantees. Financing (e.g., weather insurance, parametric credit scoring, models using movable assets as collateral often mobile banking); include leased equipment or harvested commodi- ƒƒ Adaptation of established models and approaches in ties in warehouses. Financing models that rely on use elsewhere but adapted to the context of emerg- buyers as the repayment source are based upon an ing markets, particularly those relevant to lower overall value chain analysis in which strong busi- income (IDA) countries and smallholder farmers ness relationships persist between farmers and (e.g., warehouse receipts, price hedging, agricul- buyers; formal or informal contracts provide secu- tural equipment leasing); rity to lenders. A discussion of each of these types 26 Glob al Partn er ship for Financial I nc lu si on Figure 4  Financing Models and Cases Financing Models by Main Secondary Source of Repayment Farmer Movable Collateral Buyer Direct Smallholder lending Equipment finance Tight market Value Chain Finance (VCF) with output buyers Kilimo Biashara, Equity Bank, Kenya; Banco de Lage Landen, Brazil; Palabana Dairy Cooperative Society, Opportunity International (OI) Mahindra & Mahindra Financial Parmalat, Zambia; Informed Lending Model, Ghana, Services, India Rwanda, Mozambique, Malawi, Dunavant Cotton Contract Farming, Uganda Leasing Zambia; IMON Agricultural Leasing pro- ECOM Trading Coffee farmer Indirect lending through gram, Tajikistan; financing, Africa-Asia Facility FBOs/coops Development Finance Uganda Zanaco’s Munda Credit Facility, Loose market VCF with Leasing Company Zambia output buyers IFC Coffee farmer bicycle Leasing, Emerging Farmers Finance Ghana Grains Partnership, Ghana Rwanda Finterra, Mexico; Nucleus Farm/Outgrowers Infrastructure Finance Zanaco’s “The Zambia Emergent Mtibwa and Kilombero Sugar Cane Jain Irrigation Systems Limited Farmer Finance and Support Outgrower Schemes, Tanzania Microirrigation system financing, Program” (ZEFP), Zambia India VCF with input suppliers Savings account linked Warehouse Receipt Financing NMB Agro-Dealer Financing input finance NIB, Ethiopian Commodity Bayer/Raiffeisen Aval Bank, NMB’s Kilimo Account Product Exchange, Warehouse Receipt input financing, Ukraine; (KAP), Tanzania Financing for Coffee Farmers ITC-SBI Input finance for NMB WHR fin, cashew, coffee, smallholders, India maize, sesame, sunflower Tanzania Factoring Centenary Bank/Technoserve Collateral Management Uganda Ghanaian Financial services, Kenya Gatsby Trust, Kenya Ghana Trade Finance Trade finance for exporters from Root Capital of financing models follows in the first three parts collateral, either in the form of equipment or com- of Section II 1–3, after which both risk mitigation modities. And the third column indicates where the models and distribution models are discussed in bank looks to the buyer in value chain or trade financ- Section II 4–5. ing, specifically looking at the relations the buyer has with other actors along the supply chain. The classification of financing models is based on the sources of repayment or collateral on which the Before delving into the various financing models, it financial institution can rely. The first column in is important to put financing in context within the Figure 4, above, shows how the bank can look to the overall agricultural value chain. The various models farmer by relying on his or her overall cash flow, on of financing for agriculture can exist at many dif- his or her savings, or a group guarantee. The second ferent points along a given agriculture supply column outlines where the bank has access to movable chain, as depicted by the following figure. Innovati v e Agric u ltu ral S ME Finance Models 27 Figure 5  Actors on the Supply Chain Systemic Agriculture Risks Transaction Costs Input Suppliers Farmers Traders & Processors Marketing Distribution Pre Harvest Finance Inventory Finance Trade Finance The figure above includes all the actors along a farmer’s ability to generate cash flow or liquidate given agricultural supply chain and indicates the various assets to repay the loan. Key success factors place of various financial instruments in relation to generally involve investment by the bank to under- their target users. The target users include input stand the needs of the farmers and the primary cash supply SMEs, farmers, and agriculture SMEs in pro- flow strengths and weaknesses in order to adequately cessing, trading, marketing and distribution activi- underwrite cash flow and rely less on collateral. ties. The financial instruments cover pre-harvest Thus the following models innovate through new loans, inventory financing, and trade financing, as types of finance arrangements, such as group lend- well as ways to deal with systemic risks and trans- ing, parametric lending methodologies, emerging action costs. These risks and transaction costs apply farm business finance, out-grower models, or sav- to all actors along the supply chain but are much ings linked approaches. more pronounced in the case of financing farmers. Direct smallholder lending 2.1 Financing Farmers Direct smallholder finance models seem to be more This section looks at models for financing farmers, effective than indirect or wholesale models in pro- either directly or indirectly, through farmer-based viding access to financial services for agricultural organizations (FBOs) or cooperatives. The primary SMEs. The main advantage of the direct model is source of repayment is usually the farm’s conversion that it enables distribution of a full range of finan- of working capital into cash flow through the pro- cial services, whereas the wholesale model mainly duction season. If, for some reason, this conversion focuses on credit. Retail models also allow for a fails to generate sufficient cash flow to service the segmented approach to agricultural SMEs. For loan requirements, the bank has to consider other example, smallholders may be served only with options, many of which are still dependent on the small credits, whereas growing farmers could 28 Glob al Partn er ship for Financial I nc lu si on eventually apply for investment financing as well. have proven entrepreneurial skills and track records, Key risk mitigants for this model are: (i) deep as well as minimum farming sizes and assets. All of knowledge of the farmer and his or her business; (ii) these factors combined give them the potential to a cap on the exposure to a single farmer; (iii) group transform their business into larger, independent, lending (collective responsibility); (iv) integration commercial farms in the grains, livestock, and horti- into a supply chain; and, (v) providing cash to the culture sectors. This class of farmers emerges from farmer during the lean season to lower the side sell- the segmentation of farmers. At the bottom of the tri- ing risk. The direct model allows the bank to attract angle depicted in Figure 2, there is a large group of deposits as well, which lowers funding costs and semi-commercial and commercial smallholder farm- facilitates more effective asset/liability management. ers. Emerging farm businesses in that category are The Kenyan Equity Bank (Case 1, page 60) and generally medium-sized, situated between the large Opportunity International (Case 2, page 61) cases farmers and the semi-commercial and commercial are examples of direct lending, while the HDFC case smallholders. Generally, emerging farm businesses (Case 3, page 62) is a good illustration of an agency have access to reasonable plots of land (often >100ha) based model. but only cultivate a small portion (perhaps only 15–20ha) due to lack of working capital and lack of Indirect lending through agronomic, technical, and financial skills to grow FBOs/cooperatives their businesses. They may also lack or have uncertain land titles, which then prohibit them from access to This model, also known as a wholesale model, is based commercial bank financing. on a bank lending indirectly to smallholders through an aggregator organization, such as a farmer-based Although the number of so-called emerging farm busi- organization or cooperative. In the wholesale model, nesses is difficult to determine, estimated between 1,000 the entire group is the borrower, and therefore group and 10,000 farmers in Zambia for example, growth in members guarantee each other. In the agent model, the agricultural production in many developing countries group’s organization only administers the loans, and may very well come from this class of farmers as well as individual group members are the borrowers. The ben- from smallholders. Unlocking their potential requires: efits of this approach are savings on costs of creditwor- (i) working capital finance and investment finance (irri- thiness assessment and loan administration. The gation and mechanization); and (ii) farm management security of the model can be enhanced by cash collat- skills, technical skills, and financial skills (cash flow eral requirements at the organization level, instead of planning). In addition, these farmers often need traditional collateral or claims on harvest proceeds at improved land security, which is particularly important the individual farmer level, as well as direct integra- for making significant investments in long-term capital tion with input suppliers to reduce the amounts of for these types of emerging farm business. The Mexican cash disbursed directly to farmers. The Zanaco case Finterra case (Case 5, page 64) and the Zambian Zanaco (Case 4, page 63) illustrates how these two factors cases illustrate both the difficulties and rewards of serv- combine into a zero-default lending scheme. Other ing these emerging farm businesses. success factors include strength of management, length of history, and commercial orientation of the FBO or Savings-account linked input finance cooperative through which the bank will lend. Savings are a very important part of the financial services Emerging farm business finance package that banks want to offer farmers. Savings accounts are a stepping stone to turning a smallholder “Emerging farm businesses” are those farmers who farm into a more commercial business. In addition, Innovati v e Agric u ltu ral S ME Finance Models 29 deposits are usually the most economical way for bank- multilateral institutions that have started to provide ers to fund their business; they are de facto long-term long-term funding in local currencies.15 savings. Finally, savings can be an effective part of the loan security package, and they can become the principal Equipment finance collateral to secure a loan. The Tanzanian case of NMB’s Kilimo account (Case 7, page 66) illustrates the design of Equipment finance denotes financing of usually movable an effective lending product linked to a savings account assets acquired as additions or supplements to more per- linked. Success factors of this model are strong checks manent assets. An important factor in this type of asset and balances that prevent farmers from “gaming” the finance is close collaboration between the equipment system. “Checks” include “know your customer” (KYC) providers (vendors) and the bank. The Banco de Lage signals, such as requirement of references or member- Landen case in Brazil (Case 8, page 67) underscores the ship of farmer associations, and “balances” include need for a deep understanding of farming and farm strong savings incentives and bonuses for high savings equipment markets. For the Mahindra & Mahindra (Case balances over longer periods of time. 9, page 68) and De Lage Landen (DLL) (Case 8, page 67) cases, asset finance is based on a loan and pledge struc- 2.2 Financing movable assets ture rather than a lease structure, due to specific local circumstances (tax issues) as well as the farmers’ prefer- This sub-section reviews the experience and cases ence to own their equipment. The Mahindra & Mahindra with movable assets as secondary repayment source. case reveals the following key success factors for equip- Movable assets can be anything from equipment to ment finance, which are also supported by the other small infrastructure and commodities (post-harvest). cases: (i) understanding the farmers’ payment capacity; This sub-section discusses term equipment finance (ii) avoiding intermediates; (iii) local network and local separately from leasing cases. Although there is essen- decision processes with short response times; (iv) prod- tially no economic difference between the two struc- ucts that suit farmers and account for seasonal payment tures, often the choice between the two is driven by patterns; (v) a platform for effective repossession and tax and preferences with regard to ownership. remarketing of equipment for defaulting farms; and, (vi) efficient handling of cash payments in the absence of Challenges to long-term financing for investments bank relationships with its clients.16 The Jain irrigation such as irrigation, replanting of cocoa or cashew trees, (Case 10, page 69) case in India also reveals how or farm equipment are even greater than providing equipment financiers can leverage government subsi- seasonal working capital loans to agricultural SMEs in dies for equipment. developing countries. In addition to constraints around enforceability of collateral, banks do not wish Leasing to have a long term local currency exposure and a mismatch with their liabilities on the funding side. A lease is a contractual arrangement between two par- Funding of long-term deposits in local currency is ties whereby a party that owns an asset (the “lessor”) often problematic. Some multilateral institutions are lets another party (the “lessee”) use the asset for a pre- interested in providing long-term funding to local determined time in exchange for periodic payments. banks, but these are often denominated in USD and Leasing focuses on the lessee’s ability to generate cash thus create a risk of a mismatch with the local cur- flow from business operations to service the lease pay- rency long-term loans. This is being addressed by ment, rather than on the balance sheet or on past credit 15 IFC and other IFIs have started offering longer-term loans in local currencies in a number of emerging markets. 16 An additional case, the BrazAfric case in Kenya, illustrates that Public Private Partnerships (including a partial credit guarantee) can be pivotal to the scaling-up of these projects. 30 Glob al Partn er ship for Financial I nc lu si on history. This explains why leasing is particularly advan- security of the goods until they have been sold and the tageous for young companies, as well as for small and proceeds collected. Given the limited collateral available to medium businesses that do not have a lengthy credit support farmers’ financing needs, such post-harvest com- history or a significant asset base for collateral. modities and warehouse receipts represent a liquid form Furthermore, the absence of traditional collateral of collateral against which banks can lend. When a well- requirements (such as land) offers an important advan- functioning warehouse receipt system is in place, farmers tage in countries with weak business environments, have a choice in deciding whether to sell immediately particularly those with weak creditors’ rights and col- after harvest (when prices are often lowest) or to store in a lateral laws and registries. Because the lessor owns the licensed warehouse and to apply for a short-term credit equipment, it can be repossessed relatively easily if the (thus enabling farmers to sell at a later date, when prices lessee fails to meet lease rental obligations; this is par- may be higher). Warehouse financing also enables aggre- ticularly advantageous in countries where secured lend- gators and processors to secure their sourcing throughout ers do not have priority in the case of default. the year and to purchase their raw materials. The leasing entities that do have a focus on the agricul- There is significant upfront work required to create, tural sector are often linked to manufacturers or distrib- operate, and monitor a full warehouse receipt system. utors of agricultural equipment in one way or another. Necessary preconditions for a warehouse receipts system Lease financing only partially overcomes the typical con- in which smallholder farmers can participate are many: straints to credit financing. Leasing firms often take addi- (i) a legal environment that ensures easy enforceability tional collateral from rural clients in developing of the security, and makes warehouse receipts a title doc- countries;17 this practice is different from the typical ument; (ii) reliable and high-quality warehouses that are lease transaction in developed economies, in which the publicly available; (iii) a system of licensing, inspection, leased asset itself is considered adequate security. The and monitoring of warehouses; (iv) a performance bond security deposit or down payment required tends to be and/or indemnity fund; (v) banks that trust and use the higher than typically demanded in developed econo- system; (vi) agricultural market prices that reflect carry- mies. In addition, a World Bank study finds that non- ing costs; (vii) supportive public authorities; and, (viii) farm enterprises account for a significant proportion of well-trained market participants. rural leases; rural leasing can be profitable, but jump- starting rural leasing may require government and donor Even with the necessary preconditions in place, there support; and, rural leasing companies may not always be remain risks in warehouse receipt systems, including: (i) viable. Given that leasing is a very specialized financial fraud or collusion; (ii) credit and counterparty risk; (iii) activity, economies of scale, cost, and risk factors may storage risk and misappropriation by warehouse opera- require leasing companies have large urban operations.18 tors; (iv) price risks, given the volatility in agricultural The Ugandan case of DFCU (Case 11, page 70) illustrates commodity prices and government price intervention; all of these challenges and limitations. (v) marketing or buyer risks; and, (vi) legal risks con- cerning perfection of security, registration of prior Warehouse receipt financing claims, and enforceability. Nevertheless, both the Tanzanian NMB (Case 12, page 71) as well as the HDFC Warehouse receipt finance is a form of secured lending to (Case 13, page 72) cases illustrate how warehouse receipt owners of non-perishable commodities, which are stored schemes can thrive sustainably. in a warehouse and have been assigned to a bank through warehouse receipts. Warehouse receipts give the bank the 17 World Bank (2006) 18 Ibid. Innovati v e Agric u ltu ral S ME Finance Models 31 Collateral management agreement suppliers, as well as flows from financial institutions financing into the chain, or combinations of both. The buyer security models are structured so that the bank relies A collateral management agreement (CMA) is a tripartite upon the buyer contracts (verbal or written) to help agreement between a collateral manager/warehouse secure its loans. From the bank’s perspective, having a operator, a named depositor or owner of the commodi- strong buyer in the chain in itself provides comfort, ties, and a bank. The collateral manager acts as the custo- because it helps to reduce or manage the risks of lim- dian of the commodities held in storage at the warehouse ited market access and price volatility, especially if the on behalf of the bank. The collateral manager will not farmer has an off-take agreement19 with a trusted release the goods to the depositor or a buyer until the counterparty, and is therefore less likely to default. bank provides a written form of release to the collateral Bankers may be further secured when the buyer helps manager, usually only upon receipt of loan repayment or to minimize default risk with the pledge of buyer other payment assurance against its loan secured by the receivables to the lender or some other form of guar- goods in storage. CMAs are generally costly and thus are antee, and by sale proceeds flowing through the bank. often not accessible to smallholder farmers and agricul- Under these models, bankers base lending decisions tural SMEs. Nonetheless, agricultural SMEs might benefit on the strength of the value chain as much as on the from CMAs. Banks in developing countries often provide creditworthiness of individual farmers. financing to aggregators, processors, and exporters backed by agricultural commodities held in warehouses The downside of these arrangements is the depen- under collateral management agreements in the absence dence of farmers on a single buyer: when the buyer of a fully-developed warehouse receipt system (accord- disappears or defaults on his or her obligations, the ing to the defined pre-conditions above). The same risks whole supply chain collapses and takes farmers repay- as outlined above for warehouse receipts also apply to ments with it. An additional constraint of value chain CMA-backed financing, such as fraud, collusion, storage finance is that it does not address other financial ser- risks, credit risks, price volatility, and buyer risks. vices needs of the farmers, given its focus on credit However, as the bank maintains physical control over only. These models do not facilitate development of the commodity in storage via its custodian (the collateral the smallholder into an emerging farm business. At manager) until its loan repayment is secure, there is lim- least in traditional contract farming models, the farm- ited risk that the bank’s security interest will not be per- er’s role is limited to execution of the production plan fected. The Ghana case (Case 14 page 73) tries to of the off-taker/processor. The advantage for the illustrate some of the issues involved with CMAs. farmer is that he or she hardly needs any working capital and that the income becomes predictable. The 2.3 Financing Farmers in Value Chains major benefit for the farmer and the bank is that cash flows become more predictable compared to stand- Rather than relying on the creditworthiness of indi- alone farmers, and that there is a risk of side-selling in vidual farmers, value chain financing and other tight value chains. A risk for the bank is that the buyer approaches that rely upon buyers are based on busi- gets into financial/operational problems and is no ness relationships in the value chain. Broadly speak- longer capable of buying the produce under the con- ing, value chain finance includes financial flows tract. In addition, there is often a strong monitoring between value chain actors, such as buyers or input role for the buyer and there are often high set-up 19 In Africa, many rural smallholder farmers are illiterate and often off-take agreements in the traditional sense are non-existent. However, there are aggregators that are fully entrenched in the local community and have an unwritten understanding that the farmers will sell their produce to them, usually because the aggregator has provided credit, inputs, and advice to those farmer. Nevertheless, the risk of side selling remains high, especially when unscrupulous traders prey on vulnerable farmers. The strength of such a value chain is based not on written agreements but on relationships with a foundation of respect and trust in local communities. 32 Glob al Partn er ship for Financial I nc lu si on costs, given that the financing structure, related con- who access financing from banks for their own work- tractual arrangements, and procedures for monitoring ing capital to finance their farmer customers. and enforcement need to be tailored to each specific value chain situation. Buyers are interested in involv- Tight value chain financing (TVCF) ing banks in the farmer financing, because they do with output buyers not want to use significant capital for the non-core business lending to farmers. Bringing banks into tri- Tight value chains are characterized by multiple “con- partite arrangements allows buyers to leverage banks’ striction” points for farmers that ultimately prevent balance sheets. Banks benefit from the buyer’s knowl- side-selling. These constriction points can be incen- edge of the chain and some level of buyer guarantee tives (technical assistance for farmers, loans, club of farmer risk, given its higher risk tolerance. membership, prizes, cash advances during the lean season, sustainable price premiums, etc.) as well as The value chain finance (VCF) models are divided into penalties and constraints (such as perishable crop or four categories, according to the characteristics of dif- enforced legal sanctions). Integrating the financing of ferent value chains: (i) tight VCF with output buyers; inputs into supply chain activities is more common (ii) loose VCF with output buyers; (iii) nucleus out- for “tight” value chains for a variety of reasons. Often, grower models; and, (iv) VCF with input suppliers. the values at stake are higher, including higher input These distinctions are made according to the tightness loan sizes for specialized seeds, fertilizers, etc. These of the value chain, which affects the magnitude of side- models are predicated upon strong commercial inter- selling risk, and according to the actor in the chain mediaries with a focus on the physical trade and opti- with which the bank interacts to implement its financ- mization of production, quality, logistics, storage, ing model (output buyers vs. input suppliers). The risk processing, and risk management functions in of side-selling is the biggest challenge for any actor that between. Successful commercial intermediaries with provides inputs, input finance, or working capital to integrated supply chain management recognize that a farmers in a value chain with the expectation to gener- profit-making opportunity exists in continuously ate repayment via sale proceeds, whether it is the bank, working with smallholders to increase productivity the buyer, or an input supplier. Tight value chains, such as and secure stable supplies. Thus providing finance to sugar and cotton, have integrated value chains where supplying farmers plays an important role in increas- farmers face only one de facto buyer for certain types ing production, yields, and quality for the benefit of of crops: highly specialized export crops; highly per- the buyers and farmers. Finance mechanisms may be ishable crops; and crops with constriction points in the either through the buyer or from the bank to the chain, usually transport costs or specialized processing. farmer directly with the security of a tri-partite agree- In these tight VCs, side-selling is very costly or even ment between bank, buyer, and farmer. Input finance impossible. These characteristics are also applicable to is a crucial added service that the buyer facilitates for most nucleus outgrower financing models, in which nucleus the farmer, one that ultimately increases loyalty and farms typically give outgrower farmers access to pro- more stable supplies. cessing, transport, and markets for cash crops. Loose value chains are typical of crops that are more easily market- There are several benefits of TVCF models. Value chain able and therefore attract third-party buyers to pur- actors tend to have better knowledge of the key risk and chase crops directly from farmers in the value chain. profitability factors in a particular sub-sector, and banks While farmers may have contracts with value chain can benefit from this knowledge of the value chain. buyers, they can be tempted to side-sell to these third These models often bundle finance with other services, party buyers. VCF for input suppliers includes farmer financ- such as improved inputs, extension services, and train- ing by other value chain actors, such as agro-dealers, ing, which can lead to increased cash flow for farmers Innovati v e Agric u ltu ral S ME Finance Models 33 and better quality for buyers. Tying credit with existing Outgrower schemes touch points and commodity flows can reduce the transaction costs of lending. Since buyers and other agri- Outgrower models, often based on a central processing business companies have a core interest in obtaining the unit or estate, can allow farmers to access input finance crop, they have every incentive to monitor the farmers thanks to the additional security the buyer provides to closely and ensure delivery of the produce, which also the lender. Such schemes bring together four elements: a will ensure the repayment of the loan. This provides central farm and facilities surrounded by growers who value chain buyers with an incentive to control delivery produce on their own land under contract; the provision and thus defaults. Value chain financing can be provided of inputs and technical assistance to growers by the either through the key buyer or through a financial nucleus farmer; guarantees to purchase the growers’ institution in close collaboration with the buyer. Close crops subject to meeting predefined standards; and, collaboration can involve various arrangements from growers typically receiving an agreed-upon percentage introducing farmers to the financial institution, to distri- of the final sales price of their products. Although this bution and collection of funds, to risk sharing arrange- still leaves growers exposed to price and weather risk, it ments between the parties. The Dunavant cotton case allows them to allocate a portion of their farmland to (Case 15 page 74), the Parmalat case (Case 16 page 76), growing a cash or export crop they otherwise would not and the Ecom case (Case 17 page 77) illustrate the grow due to limited market access. The nucleus farm is widely varying types of arrangements under this model. generally engaged in primary production on a large farm plot, but also has other operations such as storage, pro- Loose value chain financing (LVCF) cessing, transportation, and market distribution for its with output buyers own produce. However, engaging nearby farmers allows the nucleus farm to increase volume and achieve higher As described in the introduction to the buyer-based economies of scale than would otherwise be possible models, VCF for tight value chains is generally easier and through their own production. more prevalent than VCF for loose value chains, which typically feature easily marketable, staple crops. There are There are several key success factors for effective few success stories of value chain finance in staple crops nucleus farm models according to a Technoserve such as maize, cassava, wheat, and ground nuts. For review: (i) direct access to a viable market (local, these crops, the side selling risk is naturally higher, regional, global) for the end product; (ii) a clear, because there are many buyers and crops can be sold in transparent pricing mechanism, a price that is attrac- local markets. Additionally, government interventions tive to farmers, or both; (iii) avoiding mono-cropping are more frequent and sometimes unpredictable, causing systems, especially low-value, high-volume annuals; market distortions and price volatility. Even in the case of (iv) avoiding overreliance on credit to purchase rice, a value chain structure would only work if there inputs; (v) leveraging a competitive advantage in pro- were a strong relationship between the farmers and the duction, product attributes (e.g., brand, certifications), mill. However, in many countries there are multiple and/or proximity to the end market; and, (vi) credi- smaller mills and middlemen absorbing paddy produc- bility of the buyer and trust among farmers via regu- tion and undermining any potential value chain finance lar direct interaction between the buyer and the structure. Thus VCF for these loose value chains has been farmers. This review also notes evidence suggesting notoriously difficult, if non-existent. The Ghana Grains that ad hoc, opportunistic investments that do not partnership case study (Case 18 page 78) documents a pursue and sustain an integrated and comprehensive notable exception to this pattern. farm-to-market approach are likely to fail.20 20 Technoserve-IFAD (2011) 34 Glob al Partn er ship for Financial I nc lu si on Though similar to VCF for output buyers, outgrower via input suppliers, it is important to note that these schemes are distinguished by the centralized estate that agro-dealer arrangements do not inherently involve both sources from local farmers and acts as a primary buyer agreements and thus do not address a banker’s producer. Estates may have processing capabilities, but concern with strong, stable procurement arrangements. often sell aggregate production to end line processors. Tanzania’s NMB agro-dealer scheme (Case 20 page 80) Strong, local linkages offer additional security to lenders. and the Ukrainian Bayer guaranteed input credit scheme Proximity with outgrowers promotes supervision, limit- (Case 21 page 81) as well as the Indian e-choupal sup- ing the side selling that is often a function of distance. ported input finance scheme (Case 22 page 82) illustrate Local sourcing also simplifies the provision of extension this model. services and other supportive functions, providing addi- tional opportunities to build trust and establish working Factoring relationships.21 The Tanzanian sugar case of Kilombero (Case 19 page 79) seems to fit this pattern. Tanzanian Factoring can be a powerful tool in providing financing banks lend to sugar outgrowers based on the additional to high-risk, opaque agricultural SMEs. Factoring is security the sugar estates and mills provide. based on a company selling its accounts receivable (A/R) to a bank or factoring company at a discount. Value chain finance with input Factoring differs from the VCF models described previ- suppliers ously, because the A/R are only generated once goods have been delivered but cash payment is still forthcom- Most commercial banks have limited branch networks ing. The company selling its A/R benefits by receiving outside of major urban centers, and no branches in rural cash earlier than it would under the terms of the areas. Banks interested in financing smallholders may receivable and is thus able to immediately utilize the choose to pursue lending directly to local agricultural cash received to invest in working capital needs. From input dealers, but leave the provision of credit to indi- the bank’s perspective, the key virtue of factoring is vidual farmers completely in the hands of the agro-deal- that underwriting is based on the risk of the receivables ers themselves. Lending through the agro-dealer (i.e., the buyer) rather than the risk of the seller of the leverages the benefits of farmer facing trusted parties. A/R; there is no delivery risk as in VCF models. Lending decisions are made through local knowledge of Therefore, factoring may be particularly well suited for farmer capacity and commitment as overall transaction financing receivables from large or foreign firms when costs are reduced. Value chain finance with input dealers those receivables are obligations of buyers who are is a special type of model, because the lender generally more creditworthy than the sellers themselves. assumes the agro-dealer risk, which requires a very dif- ferent type of creditworthiness assessment and security Factoring can provide important export services to SMEs package, often involving cash collateral. in both developed and developing countries. Like tradi- tional forms of commercial lending, factoring provides Over time, the bank may be able to begin to lend to SMEs with working capital financing. Factoring only individual farmers, while still using the agro-dealer to requires the legal environment to sell, or assign, receiv- support borrower screening to address “Know Your ables and depends relatively less on the business environ- Customer” concerns and handle administration of loans ment than do traditional lending products, because to reduce distribution costs. This may also enable the factored receivables are removed from the bankruptcy bank to begin to provide non-credit services to farmers estate of the seller and become the property of the factor. by using agro-dealers as agents in the village. Once a In this case, the quality and efficacy of bankruptcy laws bank advances to this type of direct lending to farmers are less important. However, factoring may still be 21 Ibid. Innovati v e Agric u ltu ral S ME Finance Models 35 hampered by weak contract enforcement institutions method to win sales against competitors or import and other tax, legal, and regulatory impediments. For goods. The primary goal for each export sale is getting example, factoring generally requires good historical paid; therefore, an appropriate payment method must be credit information on all buyers; if unavailable; the chosen carefully to minimize the payment risk while factor takes on a larger credit risk and will apply a larger also accommodating the needs of the buyer. There are discount against the A/R portfolio. In general, a small four primary methods of payment for international firm sells its complete portfolios of receivables in order transactions: cash in advance, letters of credit, documen- to diversify its risk to any one seller. In fact, many factors tary collections, and open account. Cash in advance is require sellers to have a minimum number of customers the most secure for the seller, the open account the least in order to reduce the exposure of the factor to any one secure. Banks may assist by providing various forms of buyer and to the seller’s ability to repay from receipts support. For example, the importer’s bank may provide a from other buyers, in the case that a buyer defaults. letter of credit to the exporter or the exporter’s bank, However, this diversified portfolio approach requires providing for payment upon presentation of certain doc- factors to collect credit information and calculate the uments, such as a bill of lading. Alternatively the export- credit risk for many buyers. In many emerging markets, er’s bank may make a loan by advancing funds to the the credit information bureau is incomplete (i.e., may exporter on the basis of the export contract. The Root not include small firms), or non-bank lenders, such as Capital case (Case 25 page 86) shows how powerful factoring companies, are prohibited from joining. In the trade finance can be to ultimately finance farmers. case of exporters, it might be prohibitively expensive for the factor to collect credit information on firms 2.4 Risk Management Models around the world. The Ugandan Centenary Bank case (Case 23 page 83) illustrates the factoring model for The various risks for the agricultural borrower could matoke farmers, the Kenyan Gatsby Trust case (Case lead to default risk. Some of these risks, in particular sys- 24 page 85) for SMEs more generally. temic risks such as weather or price risks, can be man- aged through insurance (e.g., yield risks); forward Trade Finance contracts, futures, and options (e.g., price risks); or other similar contracts. This section outlines innovative ways Trade finance in this report refers to financing interna- to manage key risks encountered in the agricultural tional trading transactions of agricultural SMEs. In such a space, in particular related to health, weather, yields, and financing arrangement, the bank or other institution of price. There are other more traditional risk management the importing SME provides for payment for goods mechanisms, in particular credit-life products, but given traded on behalf of the importer. Similarly, exporting that they are “mainstreamed,” they are not discussed agricultural SMEs must offer their customers attractive here. The following box provides an overview of the risk sales terms supported by the appropriate payment management models and cases. Figure 6  Risk transfer models and cases Risk Transfer Models Credit-Health Insurance Credit-Weather Insurance Commodity Price Risk Management Kilimanjaro Native Cooperative TSKI MFI, Philippines; Bagsa Agricultural Commodity Union, Tanzania Exchange, Nicaragua UAP Insurance, Syngenta OI and APED, Ghana Foundation for Sustainable Agriculture, Kenya; PepsiCo, India 36 Glob al Partn er ship for Financial I nc lu si on Personal insurance assets and lives are lost to hurricanes and floods. These risks are particularly burdensome to agricultural SMEs. The personal risks of the borrower and the borrower Systemic risks, especially those that involve catastrophic household affect both household income and the bor- losses, pose special difficulties and costs. Production rower’s repayment capacity. Sometimes, severe health insurance, such as crop insurance, can be a solution. shocks or deaths in the family that lead to large health- Recently, the innovative index-based insurance products care bills or funeral expenses can jeopardize the viability that target agricultural SMEs have reached market matu- of the farming business for years to come, as households rity and significant scale in several countries, for exam- often have to sell essential livelihood assets such as live- ple India and Kenya. Such an index-based insurance stock or resort to borrowing from money lenders to product involves writing contracts against specific perils obtain the necessary liquidity quickly and simply. Thus, or events (such as drought, hurricane, or flood) or an insurance to cover these main personal risks can signifi- area-yield index defined and recorded at local area levels cantly enhance the security package and ultimately lower (usually at a local weather station or by government). default rates for the banker. The principal three products Thus the index insurance payouts depend not on the that help to secure farmer credit are: (i) credit life insur- individual losses of each policyholder, but rather on the ance; (ii) credit health insurance, and; (iii) health insur- locally recorded weather event or index of loss, which ance. Life insurance tends to become a mandatory part serves as a proxy for the losses in a region. of the loan package. For example, NMB bank Tanzania and Basix in India connect life insurance cover to their Because all buyers in the same region pay the same pre- loan; NMB has incorporated the premium for the life mium rate per dollar of coverage and receive the same insurance into the interest payment so that borrowers rate of payment, index insurance avoids adverse selection are automatically insured. Some banks created “credit and moral hazard problems. Also, since there are no on- life Plus” insurance products that also cover funeral costs, site inspections or individual loss assessments to per- disability, acute illness, and even property. form, it can be relatively cheap to administer. It relies only on local area index data, which are generally reli- Health insurance is less common given the moral hazard able and available either through ground level data mea- issues that make health insurance a difficult insurance surements or remote sensing.22 A review of 37 pilot and product to design and administer. The two innovative full market cases of weather index insurance revealed cases — OI in Ghana (Case 2 page 61), and KNCU, a that the potential is large, although scale-ups have Tanzanian Coffee Cooperative (Case 26 page 88) — occurred only in India (3.5 million farmers insured by therefore focus on credit health and health insurance. private and public insurers), Mexico (government pro- The KNCU health insurance plan links cooperative gram provides a weather risk safety net to 3.2 million member beneficiaries to a Dutch non-profit health orga- farmers), and Kenya (around 50,000). Certain key suc- nization and to a micro-insurance broker. cess factors determine scale-up in other countries: (i) focus on real value proposition of the insured; (ii) a Production risk insurance competent local champion who effectively overcomes for farmers set-up issues and barriers; (iii) efficient and trusted deliv- ery channels that handle cash; (iv) weather data infra- Farmers face a variety of production risks that make their structure (primarily weather stations); (v) risk transfer incomes volatile from year to year. In many cases, farmers into international markets (reinsurers mostly); (vi) train- also confront the risk of catastrophe, for example, when ing of all implementation partners; and, (vii) insurance crops are destroyed by drought or pest outbreaks, or when premium for farmer priced at pure risk rate thanks to 22 Hess and Hazell (2011) Innovati v e Agric u ltu ral S ME Finance Models 37 commercial sponsorships (input suppliers, mobile phone actor to simply absorb, producers or commercial actors operators, banks) or premium finance.23 Weather-based who are negatively affected by price volatility must agricultural insurance can both expand lending opportu- turn to the market and find mechanisms to transfer the nities for banks and agro-dealers as default risks are risk to market actors who are better equipped or more reduced and increase sales for agro-dealers and mobile willing to manage it. Theoretically, commodity market phone operators. As shown in the cases of TSKI in the instruments exist so that market actors unwilling to Philippines (Case 27 page 89) and Kenya’s Syngenta input carry price risk can transfer it to actors who are willing supplier-based weather insurance (Case 28 page 90), to carry or manage the risk based on expectations of loan packages are often effectively bundled with the opportunity to make a profit by doing so. Such weather insurance products. activity takes place either on a physical basis, through commercial trade of the actual commodity itself (e.g., Weather insurance for physical delivery forward contracts), or on a financial contract farming basis through instruments specifically developed for the purpose of risk transfer. Financial instruments are Weather insurance tends to support a value proposi- exchange-traded futures and options, over-the-counter tion for the farmers, such as a loan package or even a (OTC) options and swaps, commodity-linked bonds, contract farming operation, by serving to mitigate and other commodity derivatives. Generally, the finan- production quantity and quality risks stemming from cial instruments are only developed in commodity adverse weather. Thus, when risk insurance is bun- markets with established exchanges. The primary func- dled with these other products, it can effectively pro- tions of commodity exchanges are to serve as clearing- tect the farmer, the bank, and the off-taker behind the houses for the transfer of risk from one commercial loan. Basis risk represents a challenge for index-based participant to the other and to provide a transparent insurance, given the possibility of mismatches price discovery mechanism. Forward contracts, futures, between payouts and actual losses if the correlation and options allow sales prices to be locked in prior to between the index and actual farm yields is not suffi- the actual delivery of the product. ciently high. Another challenge is the lack of solid data in many emerging markets for sound actuarial This transfer of risk can be done through futures con- modeling and a limited physical infrastructure of tracts, which are similar to forward contracts in that they weather stations. Additionally, small producers often are agreements to buy or sell a specific quantity of a do not understand the concept of insurance or do not commodity, at a specific price, on a specific date in the trust insurance due to prior negative experiences. The future. Unlike forward contracts, however, futures con- PepsiCo case in India (Case 29 page 91) illustrates this tracts do not necessarily imply physical delivery to fulfill type of weather insurance application. the contract. For commercial intermediaries in develop- ing countries, futures contracts have an advantage in that Commodity price risk management they can lock in a sales price in advance of the actual delivery of the product. In essence, a commercial Market-based price risk management has the potential intermediary losing on the physical sale should be to help farmers or commercial intermediaries manage gaining on the financial, while a commercial inter- the risk of adverse price movements on commodities mediary losing on the financial side should be gain- markets through the use of physical or financial instru- ing on the physical. The major disadvantage for use of ments. Because the financial impact of price volatility this system in developing countries, however, is the has proven to be too large for government or any other credit risk inherent in trade of these contracts and 23 Adapted from key success drivers in Hazell and Hess (2010). 38 Glob al Partn er ship for Financial I nc lu si on associated margin requirements whereby the party at This creates buyer reluctance to offer fixed price for- risk has to deposit funds as de facto collateral. ward contracts because they may not get the goods if prices rise. In cases of very tight value chains where the A second type of contract traded on international buyer has a very strong control on the physical delivery exchanges, an options contract, can also be used to of goods, such as in many cases of contract farming, manage risk. Option contracts are similar to physical then a fixed price forward contract may be feasible. minimum price forward contracts in that they are Instead of fixed price forward contracts, some buyers agreements to buy or sell a specific quantity of a com- may offer a minimum price contract, meaning that modity, at a specific price, on a specific date in the they commit to purchase at a minimum price, but if future, but they also provide an opportunity to take prices are higher they will be paying the higher price. advantage of favorable price movements in the future. In effect, they are offering a put option to the farmers. Unlike minimum price forward contracts, however, options contracts do not necessarily imply physical 2.5 Distribution models delivery to fulfill the contract. The instrument is valu- able because it avoids absolutely locking in a price The primary purpose of these models is to offer mobile level as happens with a futures contract, and it pro- payment and other mobile banking services to reach vides the user with an opportunity to take advantage customers, in particular rural and more remote cus- of favorable price movements that may occur between tomers and farmers, and thereby build new relation- the time of purchasing the instrument and the time of ships. The mobile banking relationships also help the its expiration. Because premiums are paid up front, banker to understand these new clients with adapted there is no credit risk.24 However, futures contracts “know your customer” (KYC) approaches and to learn and options are often beyond the reach of agricultural their business patterns through payment transaction SMEs and commercial smallholders, due to the size of histories. Finally these services can encourage savings these contracts, procedures to access these instru- and deposits and thereby lower the bank’s funding ments, need for margins, and the overall high level of costs. Because they must distribute some financial ser- knowledge needed to operate these instruments. vices, these distribution models may not be considered as standalone. These models can therefore support the Another issue is basis risk, meaning the possibility of a financing models, in particular by helping KYC, but weak correlation between the price of the commodity they do not alter the lending models per se. The inno- in the domestic market and the price of the commodity vation here lies in the use of a new channel and the fact in the international exchange where usually futures that the models discussed in this report distribute a and options are traded. There are domestic commodity more complete range of products, such as payments, futures and options exchanges that reduce or eliminate deposits, and credit. The following table provides an this basis risk in some large emerging markets such as overview of all distribution models and cases. in Argentina, Brazil, China, India, and South Africa, but this is not so in smaller markets. Mobile banking Forward contracts with physical delivery can be writ- Currently there are over 5.6 billion mobile phone cus- ten for any amount and offer more flexibility to small- tomers worldwide, with the vast majority of the scale operators. However, a major risk is that farmers growth since 2005 taking place in the developing may deliver elsewhere if prices are higher at the time of world.25 The rural poor in Africa, South America, and delivery compared to the pre-agreed forward price. South and Southeast Asia have greatly increased their 24 World Bank (2005) 25 World Factbook (2011) Innovati v e Agric u ltu ral S ME Finance Models 39 Figure 7  Distribution models and cases Distribution Models Mobile banking Branchless banking Mobile Payments M-Pesa, SafariCom, Kenya United Bank Ltd. “Omni”, Pakistan Dunavant, Zambia BPR Mobile, Rwanda ANZ’s WING, Cambodia Opportunity International Bank, Ghana use of mobile phones as coverage has improved and mobile network operators, leveraging their national costs have come down. Mobile phone service growth is agent networks for prepaid airtime sales. There are ben- substantial and continuing in many developing mar- efits and drawbacks for each of the deployable technol- kets, with countries like India reaching 1.2 billion ogies, mainly with regard to security and control on mobile phone subscribers in December 2011, with 307 the part of the banks. Key hurdle for banks and mobile million in rural areas.26 In the Philippines, a country network operators are the lack of mobile phones in- with more than 7,000 islands, over 95 percent of the country that can support mobile Internet or “apps,” the land area is covered, and over 98 percent of families, cost of mobile data, the regulatory environment accounting for around 98 million family members, enabling the actors in these deployments to take con- have at least one mobile phone.27 The potential for huge trol, and cultural issues around trust, usage patterns, cost savings through full-service mobile phone banking and convenience.28 For small and mid-size farmers, was proven by successful examples in Africa as well as mobile technology has been a way to ensure access to a South and Southeast Asia. Mobile phone companies bank account 24/7, and receive payments and alerts. have realized that these services could potentially be This usage has potential to form the platform from very profitable and banks have realized that these tech- which banks, MFIs, and mobile network operators nology deployments represent an opportunity for rural could add value-added services to a farmer’s bank rela- bankers to reach more customers. Banks in developing tionship, including co-payment for agricultural input markets throughout the world have been launching vouchers, access to instant micro-credit capabilities, mobile phone banking platforms in earnest since 2008 and more advanced banking products such as insur- across a wide range of countries, including Zambia, ance.29 Mobile banking services might also encourage Tanzania, Rwanda, South Africa, Kenya, Mozambique, savings and could help lenders to use cash flow and the Brazil, Paraguay, India, Pakistan, the Philippines, new savings as collateral. Bangladesh, Uganda, Ghana, Zimbabwe, and Nigeria. The spectrum of possibilities for banks, mobile opera- The easing of traditional regulatory rules has facili- tors, MFIs, and other financial service providers spans a tated mobile service developments in developing number of possible services, from traditional mobile countries. The upside has been seen in increased banking to remote payments and “Mobile Money” — liquidity, increased competition, electronic money normally remittance and payment services offered by usage, and progressive leap-frogging of traditional 26 Telecom Regulatory Authority of India (2011) 27 Philippines Telecomresearch (2012) 28 Armstrong (2011) 29 For example, the mobile phone operator Tigo provides loyal customers in Ghana and Tanzania with free life insurance and offers customers the opportunity to sign up for life insurance through their post-paid bill, according to TigoGhana (2010). 40 Glob al Partn er ship for Financial I nc lu si on technologies, making it easier to enforce know-your- between rural clients and banks by providing cheap customer and anti-money laundering, anti-terrorism, transaction services, electronic savings accounts and, in and access-to-information policies. limited cases, even credit functions. The most promi- nent example is the service M-Pesa, provided by the Branchless banking mobile network operator Safaricom in Kenya, which has developed into one of the largest banks in eastern The term “Branchless Banking” usually refers to the Africa. In countries lacking the technical and commer- capability to offer a full array of banking services or a cial infrastructure for ATMs and point-of-service limited set of services either at bank-owned locations devices, mobile phone banking in particular can be a aside from branches, with bank-owned equipment low-cost way to expand access to financial services in (such as trucks or ATMs), or in partner-owned loca- rural areas. tions (by the bank employees or by partner employ- ees or agents). While branchless banking may include Mobile payment systems can benefit farmers by mobile banking services, as the following examples allowing them to receive payments as electronic credit demonstrate, this is not always the case. Branchless into their mobile phone-based account (or “m-wal- banking has great potential for increasing access to lets”) instead of waiting or having to travel to obtain financial services in the agricultural sector, given cash payment. Farmers then have more flexibility and lower costs, and for reaching customers in many loca- choice of when and how they use their credit. From tions previously unable to justify the expense of a the bank perspective, an additional benefit of provid- full-scale branch. To date, there have been a number ing such low-cost financial services is that smallholder of branchless banking successes throughout the devel- farmers can gain a transaction history with a bank oping world, although the sustainability of these busi- that could enable them to access loans, insurance, and ness models is still under discussion, given the savings products. As in the case below, where mobile start-up subsidies these programs have received, ques- payments are provided by a value chain actor, contract tions regarding the profitability of the customers they farming operators can improve their service offering serve, and the high logistics and management capabil- and build stronger relationships with the farmers and ities required for branchless banking programs to run generate greater loyalty. with both personal contact and trust and security levels similar to traditional branches. A challenge with these innovations is that mobile banking is relatively new within the financial infra- Mobile payment systems structure system, and there is no existing legislation for mobile phone banking in many countries. As suc- Information and communications technology innova- cessful and proportionate regulation in Kenya has tions, including mobile payment services, have strong demonstrated, it is possible to strike the right balance potential to enhance rural outreach by reducing trans- between supervisory requirements and the develop- action costs. Mobile phones operate at the intersection ment of financial access. Innovati v e Agric u ltu ral S ME Finance Models 41 CHAPTER 3 Observations 3.1 Observations from the Case Studies Channels to reach farmers The case studies examined, while rich in information, In all cases, the channels that financial institutions use did not have consistent data across them that would to reach farmers are critical to any financing scheme. have enabled a more quantitative type of analysis. As For both commercial and semi-commercial farmers, such, we rely on close examination of the informa- these channels tend to depend on the structure of tion collected and try to observe certain patterns. production and trade of the specific commodities that These observations were classified under two main the farmers produce, on the existence of strong and categories. First are observations about the channels reliable producer organizations, or on any type of that were used to deliver financial services to farmers, aggregator that can group together a number of small- primarily smallholders, in a number of emerging holders for both credit and other services. economies across the globe, and for a variety of crops, both cash and staple crops. Second are observations Farmer linkages in value chains are very important. about the innovative ways in which the services were The vast majority of cases relied on relations between applied, ranging from new ideas to systems in devel- farmers and other stakeholders along certain supply oped countries adapted to the local context. chains. Mostly, the reliance was on buyers, but input Figure 8  Channels for Farmers Channels Banks Buyer Inputs Local Coops/MFIs Larger to Medium Farmers Commerical smallholders Tighter value chains Commerical smallholders Looser Value Chains Semi-commerical smallholders = Large occurrence  = Medium occurrence  = Low occurrence  (Blank) = Very Rare, the exceptio n 42 Glob al Partn er ship for Financial I nc lu si on suppliers and providers of technical assistance were schemes was relatively low, well under 10,000 farm- also involved, as value chains are important for more ers. There seems to be some capacity constraint that than just commercial smallholders producing cash or these cases hit, given the needed administrative higher value crops. The case studies found a number arrangements (e.g., close monitoring, proximity) of small farmers producing basic grains, like maize, and implied costs in putting these schemes together. that became part of value chains. In the majority of these cases, the linkage to value chains went through Overall, involving producer organizations as aggre- an aggregator, such as a producer organization, a gators for smallholder farmers to channel and collect cooperative, or some other form of grouping farmers credit, gather produce, distribute inputs, provide (e.g., resembling group lending in microfinance). extension, and so on, was quite important in about 20 percent of the cases and particularly in Africa. To reach smallholders in less organized value Organizing farmers for more than just credit can chains, farmer organizations or a system that have big advantages, as it can lower transaction costs aggregates small farmers are very beneficial. A and increase the efficiency of reaching many small number of cases involved smallholders in food farmers for a variety of services such as savings, pro- crops, like maize in Ghana or Zambia, for example. vision of technical assistance/extension, insurance, In these cases, financing to smallholders was made input, marketing, etc. possible via reliance on joint liability groups of farmers, the use of producer organizations, various Although not examined among the case studies, partnerships along the value chain, and close moni- another model of aggregating smallholder farmers is toring of farmer activities. In these cases, local the nucleus farmer model. In this model, a large knowledge, proximity to farmers, involvement of farmer can provide inputs, credit, and marketing of various stakeholders, and close monitoring were the final product to nearby smallholder farmers. Some very important in ensuring the client assessment, examples include rubber in Indonesia and coffee in loan use, and repayment. As such, the reach of such Vietnam, among others. Figure 9  Farmers and Value Chains Subsistence Farmers Risks Operational Costs Global number of farmers Lower value, looser value chains Information Asymmetry Tight value chains High value crops Innovati v e Agric u ltu ral S ME Finance Models 43 There are still challenges in scaling up signifi- guarantees, enhanced credit risk assessment systems cantly and reach large number of farmers. The tailored for farmers, or use of movable collateral. reach in terms of number of farmers affected in Additionally, innovation focused on alternative chan- about half of the cases reviewed was low (under nels, such as mobile banking, and on ways to insure 10,000 reached), which to some extent indicates that crop losses due to adverse weather events. reaching farmers through value chains or groups has its limitations. In general, value chains have certain Use of first loss guarantees is increasing. The guar- physical limitations linked to infrastructure, the antees were present in about 20 percent of the cases market size, and the physical capacity of the value but proved quite important in launching the financ- chain (how much it can handle, etc.) ing schemes to farmers. There were five cases in which credit guarantees or first loss guarantees were There were six cases involving either large global agri- used. One case was in Ukraine (Bayer), with private businesses (e.g., global commodity trading compa- sector and IFC participation. The other four cases nies) or large banks and leasing companies in large were in East Africa (Equity Bank, Centenary, and markets (e.g., India, Brazil) that managed to have NMB). Three cases involved donors covering the reached a high number of farmers (defined in excess credit guarantee/first loss, while one involved the of 100,000 per case). Within a given country, the government providing such guarantees. In all cases, reach of a financial institution seems to be larger guarantees seemed to have played a role in getting compared to a local agribusiness, as financial institu- credit flowing to farmers. tions can deal with several value chains, producer organizations, and individual farmers. However, Credit risk assessment. There was not much found financial institutions can achieve greater reach when in terms of sophistication in regard to credit risk they leverage the knowledge, information, and exper- assessment techniques. There were several cases in tise of local agribusinesses and try to join forces in which financial institutions came up with a combi- reaching to smallholders. nation of agronomic models and credit scorecards. However, in all cases, the financial institutions Providing technical assistance/extension to farmers involved had invested in learning about the agricul- along with credit is found to improve yields and ture sector to which they were supplying credit. incomes. In most cases, extension services and other They also forged strong linkages with buyers, pro- technical assistance (e.g., financial literacy training) cessors, and traders in value chains in order to gain offered to farmers were found to be very valuable, from their knowledge and expertise. This helped in forging stronger relationships in the value chains and assessing the credit risk of farmers. linkages with farmers. In the majority of cases, the provision of finance to farmers through these schemes In only a few cases was there explicit mention that was associated with higher yields and incomes. farmers received credit in kind (rather than cash) for inputs in order to control how credit would be used. Aspects of Innovation Credit in kind would be used for productive purposes to create the needed cash flow from the sale of the As discussed earlier in this report, innovation can be crop to repay the credit. the adaptation of existing and established models to the local context of lower-income emerging markets. Thus, Use of movable collateral. A key complaint of farm- innovation is not only about new models or those not ers dealing with formal financial institutions has been yet used globally. The important areas of innovation that the institutions ask for hard collateral, usually focused on dealing with credit risk through first loss real estate (farms) and with good titles. The cases 44 Glob al Partn er ship for Financial I nc lu si on examined indicated that financial institutions willing and correspondence banking, play an important role to lend to the agriculture sector are increasingly for mobilizing savings, processing, distributing and taking a more flexible approach to the collateral collecting loans, and providing insurance to farmers. In requirements. For example, in cases of value chain about 30 percent of the cases reviewed, some form of financing, particularly with tight value chains or con- alternative system discussed above is used. tract farming, the collateral is in the “soft” form of the promise for the farmer to deliver the crop and Dealing with crop losses due to adverse weather. repay the loan. There are cases that have used invento- Traditional crop insurance for small farmers has high ries as collateral, either through formal warehouse administrative costs and suffers from adverse selection receipts or through collateral management agree- and moral hazard problems. Assessing individual crop ments. Typically, the experiences from using such col- losses in small plots can be very costly to administer. lateral have been quite good, resulting in very low Alternative methods based on weather index insur- NPLs (less than 1 percent) and with few notable ance can provide an alternative under certain circum- exceptions of large losses due to fraud, though not in stances. Increasingly, there has been a proliferation of the cases examined for this paper. pilot projects, mostly in Africa and Asia and supported by donors and IFIs. Some of these have reached com- Experiences in leasing agriculture equipment exam- mercial scale, such as in the case of India. There were ined in Brazil and India were found to be quite posi- three such cases examined, one in Kenya, another in tive, with very large reach in terms of number of the Philippines, and the third in India. In Kenya, farmers using these financial services. Another case of insurance was part of the input supply package, while agriculture leasing in Uganda is relatively recent and in India it was part of the contract farming arrange- of course has not yet achieved significant reach. ment. In the Philippines, it is linked to loans to agri- However, like with the provision of financial services culture producers by a local microfinance institution. to agriculture, the leasing companies in our sample cases that are successful in the agriculture sector In all of the cases examined, insurance to protect report that the key to their success is good knowledge against unexpected crop losses due to weather was used of the agricultural sector, dedicated resources to serve in relatively few instances. In all of the cases that used agricultural producers in proximity to them, flexibil- insurance, it was an important ingredient in the whole ity in leasing payments to match farmers’ cash flow, package of financing farmers. However, the majority of and good credit risk assessment systems adapted to financing for farmers did not have such insurance. the agriculture sector. 3.2 Enabling Country Environments Cash collateral in terms of savings was used in two cases in Africa. In the case of Zambia, farmers’ savings Innovative models need to be relevant to the type of through District Farmer Associations are used as key country environment in which a given lender works. collateral (50 percent of the value of loan), and in the A warehouse receipt system, for example, requires a case of Tanzania, the targeted clients were farmers’ legal and regulatory environment (licensing system, primary cooperative societies. In both cases, it should security enforceability) that will not be present in all be noted that the targeted smallholder farmers are developing, rural economies. Similarly, contract farm- members of a producer organization: an association ing requires an appropriate environment in order to or cooperative society. enforce contracts. Rather than make broad generaliza- tions about the effectiveness of models across disparate Alternative channels. Increasingly, alternative chan- business environments, this report will establish corre- nels such as mobile banking, payments, and branchless lations between the specific country environment in Innovati v e Agric u ltu ral S ME Finance Models 45 which a model is active and the type of the model. In crops. This report finds that the nature of a business- other words, the report examines the type of specific enabling environment can be hard to capture or mea- country environment in which we tend to encounter sure. However, for an agricultural lender, there are two each of these models. The results from this exercise are drivers that seem to be important: the overall level of by no means a proxy for a feasibility assessment for value-add in agriculture and the business environment these models. Therefore, based on the number of case as measured by, for example, the World Bank’s “ease of studies collected, we assess the likelihood of encoun- doing business” ranking.30 The rationale for choosing tering specific models within three types of country these two indicators is based on the risk and return, environments. The models encountered within each namely that banks need to make loans in a cost-effec- country environment are discussed below, based on a tive way and get repaid. It is easier to achieve both in a review of all the case materials included in this report. context with high agricultural productivity and a good business environment. Thus, countries fall at various As explained in Section I, providing profitable finan- levels of development in terms of either or both of cial services, and in particular lending to farmers and these important determinants. agricultural SMEs, is very demanding in any context. The combination of systemic risks (weather shocks, We have mapped countries according to these two key fluctuating input and output prices), legal environ- determinants using measures of agricultural produc- ment risks (collateral rights that are uncertain and tivity expressed in agriculture value added per worker hard to enforce), and potential government interven- in USD31 and the quality of the business environment tions are challenging for any bank. through the “ease of doing business” rank of the country. Mapping all countries with a significant agri- Given these challenges, it is imperative that financial cultural sector that have a value added of at least 1 bil- service providers recognize not only what models may lion USD in agricultural productivity into the space be appropriate to emphasize in general, but also specif- opened by these two indicators generates a graphic ically where they may be more appropriate based on where countries tend to cluster in three areas, as illus- the incidence of encountering the various models in trated on the following figure. the various country environments examined. This report recognizes that there are significant differences On the basis of the analysis of the typology and the between types of enabling environments for the busi- three types of country environments, this report pro- ness of agricultural lending, and therefore the applica- vides an indication of where one is likely to encounter bility of different models for each environment will the various models described in Section 2. This also differ. It is important to note here that the criteria framework is designed to provide an overview of are generally applied on a country level of analysis where we see certain models, in which cases a given (country environments), though within countries there approach has worked as a type of financing, and the can be variations between states or provinces, espe- reasons behind the various outcomes. cially in large countries. In addition, even within a given country there could be significant variations Case information has been gathered through a combina- amongst crops, more specifically between export-ori- tion of primary and secondary sources, and certain case ented cash crops and domestically consumed food examples have only limited information available. It is 30 “Ease of doing business” ranks economies from 1 to 183, with first place being the best. A high ranking (a low numerical rank) means that the regulatory environment is conducive to business operation. The index averages the country’s percentile rankings on 10 topics covered in the World Bank’s Doing Business report. The ranking on each topic is the simple average of the percentile rankings on its component indicators as taken from the Doing Business Indicators. World Bank (2010a). 31 Agricultural value added figures are based on 2009 numbers in constant 2000 USD, derived from national accounts files and FAO in the World Development Indicators. World Bank (2010b). 46 Glob al Partn er ship for Financial I nc lu si on Figure 10  Three Business Environments Environment I — low productivity in agriculture, weak business environment This area illustrates an environment characterized by relatively low productivity in agriculture as well as a relatively weak business environment. Countries (or states in large countries such as India or China) in this type of environment have an ease of doing business rank of 100 or more and agricultural productivity per worker of less than US $1,600 (in constant 2000 USD). By way of illustration, we note that Ethiopia, Mozambique, Tanzania, along with the average for India, fall into this category. Environment II — low productivity in agriculture, strong legal rights This environment is characterized by low productivity values in agriculture as well as relatively strong legal rights. Countries (or possibly states in large countries) in this type of environment have an ease of doing business rank of 100 or less and agricultural productivity per worker of less than US $1,600 (in constant 2000 USD). This category has the largest number of countries; examples include Thailand, Rwanda, Ghana, and Zambia. Environment III — high productivity in agriculture This type of environment is characterized by high value-add in agriculture (at least US $1,600 per agricultural worker and year) and a wide range of business environments. Typical examples are Mexico, Ukraine and the Republic of South Africa. EASE OF DOING BUSINESS AND AGRICULTURAL PRODUCTIVITY IN CASE COUNTRIES 6,000 5,000 (VALUE ADDED IN USD/WORKER) Environment III AGRICULTURAL PRODUCTIVITY 4,000 RSA 3,000 2,000 UKR PER AZE 1,000 Environment I Environment II THL IND CHN GHA MOZ TZA ZAM RWA 0 180 160 140 120 100 80 60 40 20 0 EASE OF DOING BUSINESS (120 WEAK TO 1 STRONGEST) Innovati v e Agric u ltu ral S ME Finance Models 47 critical to acknowledge that this set of cases is not neces- the environments. It is important to note that the fol- sarily a representative sample. Thus, this assessment is lowing assessment is not indicative of whether the not meant to be construed as statistically valid or scien- different models are feasible or existent in the partic- tifically based. This assessment framework does not set ular environments, but rather show the probability out to strongly judge models, but instead seeks to pro- of encountering this type of financial scheme based vide some initial conclusions that can be further studied on the environment in which it is identified. The and refined through additional research and experience. color rating system illustrates the qualitative nature of the assessment. However, the reader should consider The framework examines each model within the vari- that there is more of a spectrum of shades than there ous country environments in order to provide some are true changes between colors. As noted above, guidance to financial institutions wishing to under- rather than providing strict parameters for each stand the relative value and applicability of innovative model, the colors indicate that it may take more effort models to the market conditions in which they operate. and time to apply a different model. The following This does not mean that a financial institution operat- table offers a guide to the color coding in the Figures ing in, say, Environment I (low productivity and weak 11 through 13. business environment) will not be able to apply an innovation that is more likely to be encountered in Less likely found in this environment Environment III (high productivity in agriculture). What it means is that the specific model will require More likely encountered more effort to make it work and/or will depend on the Situation dependent situation (e.g., finding a specific strong organization of smallholders or a good collateral manager, etc.). Although certain models are harder to match in cer- Before beginning this analysis, the total reach of the tain environments, coordination is not impossible, 100 cases by country environment was divided into given very specific circumstances. The assessment is the three categories of source of repayment or collat- therefore: 1) based on observations of occurrence, eral (farmer, movable asset, or buyer security). and 2) more indicative of the ease of implementation Descriptive statistics of the cases are in the annex. for a given model. The observations are by no means conclusive, nor do they constitute a rigid guide to It is important to acknowledge that the case inventory applying certain models in certain environments. The used in this report did not include many cases from classification should be understood as a continuum as countries in Environment III, as the target audience opposed to a strictly binding protocol. for the report is bankers in countries from the other two environments. Therefore, if the inventory were Models within Environment I sufficiently robust to include more cases from these (weak business environment, countries, we would expect to see significant reach low agricultural productivity) numbers in all model types. This type of environment presents the biggest chal- MODELS WITHIN COUNTRY ENVIRONMENTS lenges for farmers and agricultural SME finance, because in addition to the agriculture sector challenges The following sections discuss in more detail the for lenders, there are relatively low returns to agricul- overall salience of single models within each of the ture as well as limited legal protections. In this type of three country environments. The discussion looks environment, the cases show that most of the agricul- more closely at the correlation between the cases and tural finance activities are likely to be donor driven. 48 Glob al Partn er ship for Financial I nc lu si on Figure 11  Environment I Models Farmer Movable Assets Buyer Financing Direct Smallholder Lending Equipment Finance Tight Market VCF with output buyers Indirect Lending through Leasing Loose Market VCF with FBO's Output Buyers Emerging Farmers Finance Warehouse Receipt Financing Outgrower Schemes Savings Account linked Collateral Management VCF with Input Suppliers input finance Factoring Trade Finance Risk Mitigation Personal Insurance Commodity Price Risk Management Weather Insurance for Weather Insurance for Farmers Contract Farming Distribution Mobile Banking Mobile Payments Branchless Banking This is one of the reasons why most models in this the real business driver for the buyer in the value chain country context are rated as either yellow or orange. is obtaining access to the cotton; therefore, financing the farmer is a good investment in future business and The only green model is the tight value chain. This farmer loyalty. There are a few promising yellow models seems to be because, as a relatively self-contained model, that are not green due to one criterion only. A case in it depends less on a strong legal and market environ- point is satellite-index-based and cell-phone-delivered ment; strong buyers with strong chains depend on farm- weather risk insurance to secure input loans in the ers who generally honor their contracts and limit Philippines. Such low-cost and effective delivery and risk side-selling. The models categorized as a “tight VCF with mitigation appear to be sustainable. The model, how- output buyers” appear to be most relevant for this envi- ever, still requires index customization and has not been ronment. This type of case is exemplified in cotton tested thoroughly in this environment. Additionally, col- financing as part of a high value and tight value chain by lateral management appears very replicable, as it can be the Gulu Agricultural Development Company (GADC) in applied in almost all country environments as long as Northern Uganda, which, in spite of Uganda’s general there is a good and reputable collateral manager, yet it is country ranking in Environment II, is a difficult place in mostly used for higher value exports in countries with- which to do business. GADC covers 20,000 farmers in a out appropriate enabling environments. Therefore, it challenging, conflict-affected part of the country. This receives a yellow ranking due to relatively limited appli- case highlights a general pattern: financing inputs works cability. Warehouse receipt financing is orange, because where the core business is a high value proposition that this model relies heavily on a specific enabling legal and includes input finance as a cost of doing business. Here regulatory system (warehouse receipt as title). Innovati v e Agric u ltu ral S ME Finance Models 49 Figure 12  Environment II Models Farmer Movable Assets Buyer Financing Direct Smallholder Lending Equipment Finance Tight Market VCF with Output Buyers Indirect Lending through Leasing Loose Market VCF with FBO's Output Buyers Emerging Farmers Finance Warehouse Receipt Financing Outgrower Schemes Savings Account-Linked Collateral Management VCF with Input Suppliers Input Finance Factoring Trade Finance Risk Mitigation Personal Insurance Commodity Price Risk Management Weather Insurance for Weather Insurance for Farmers Contract Farming Distribution Mobile Banking Mobile Payments Branchless Banking Models within environment II (good financial or other services to farmers. Working capital business environment, low for agro-dealers is another model that could be viable, agricultural productivity) but there is not yet sufficient experience and donor support is too important to fully judge the sustainabil- Given the stronger legal rights in this environment, it ity of this model. Warehouse receipt financing is easier to build a viable alternative to traditional (WHR) can be a viable model, as highlighted by the farmer collateral, which is necessary in almost any NMB case in Tanzania that supports a series of crops environment and to scale up coverage and finance for around 110,000 farmers. A 50 percent government farmers profitably. Savings-linked loan accounts and guarantee made this scheme possible, but it has never cooperative-based input finance are emerging as alter- been called; thus banks may continue to offer WHR native opportunities for individual farmers to access financing even if the guarantee was rescinded. credit (NMB, Zanaco). Value chain financing — However, it is important to recognize that the inclu- mostly tight VCF and outgrower schemes for export- sion component of warehouse receipt finance depends able cash crops — again show that solid, on a legal and regulatory system that provides for business-driven value chains can catalyze input more than simply creditor rights. In addition, risk financing. Generally speaking, agribusinesses are mitigation models — in particular personal insurance more likely to set up and prosper in countries with models like credit health and credit weather insurance better enabling environments, while VCF and other that secure borrower and bank alike — appear to be buyer-linked models do not depend as much on the evolving into sustainable and scalable products with legal and general business environment to facilitate mature reinsurance markets behind them. Loose value 50 Glob al Partn er ship for Financial I nc lu si on Figure 13  Environment III Models Farmer Movable Assets Buyer Financing Direct Smallholder Lending Equipment Finance Tight Market VCF with Output Buyers Indirect Lending through Leasing Loose Market VCF with FBO's Output Buyers Emerging Farmers Finance Warehouse Receipt Financing Outgrower Schemes Savings Account Linked Collateral Management VCF with Input Suppliers Input Finance Factoring Trade Finance Risk Mitigation Personal Insurance Commodity Price Risk Management Weather Insurance for Weather Insurance for Farmers Contract Farming Distribution Mobile Banking Mobile Payments Branchless Banking chain models are rated at an improved yellow rank- opportunities. Farm sizes remain small; there is not ing, because the challenges with side-selling persist often consolidation of farmland into larger units to even as contracts and agreements provide additional increase economies of scale. As such, tight value chain security. One loose VCF case with output buyers for finance and lending models through FBOs/coopera- maize growers (Ghana Grains Partnership, led by the tives work best and thus are rated green in this fertilizer company Yara and distributor Wienco) com- environment. bines a series of very innovative features: ownership by farmers in profits generated up the chain, block Models within environment III (high farming through joint liability groups with no-tillage agricultural productivity) techniques, and the involvement of input suppliers and off-takers throughout the chain to establish trust. As expected with this type of high agricultural value- Other yellow models include emerging farm busi- add and highly conducive legal environment, we tend ness, leasing, and equipment finance. While the legal to encounter most models. However, even within this system protects assets and property rights, there may environment, direct lending to individual smallhold- not be the high values and types of value added activi- ers, particularly in lower value crops, may still be ties in agriculture that generate sufficient economies challenging and thus keep the color yellow. Certain of scale for asset financing companies. Likewise, innovations, such as electronic warehouse receipts in emerging farm business finance requires a minimum South Africa, can be scalable in the originating coun- of market infrastructure and high value business try, but may prove more challenging to replicate in Innovati v e Agric u ltu ral S ME Finance Models 51 other countries. The highly effective Brazilian “cedula and other financial institutions to locate specific de produto rural” or CPR is especially unique to growth opportunities for those distinct farmer groups Brazil’s country conditions, and thus it can be quite often viewed as a single block with similar characteris- difficult to replicate elsewhere. tics and limitations. Important elements for targeting credit to farmers could be presented as follows: The high value added of these agricultural sectors in this ƒƒ Financial institutions need to organize appropriately environment is associated with higher levels of mecha- in order to provide financial services to agriculture. nization and specialization, more processing performed ƒƒ Financial institutions should segment and target in the country, and deeper and better functioning mar- appropriately selected types of farmers. Farmer seg- kets. These factors generate high value businesses that are mentation should go deeper than simply small, very profitable to bank with and are associated with medium, and large, also focusing on the type of better access to finance in a positive cycle. Thus, those crop, the environment, assessment of the commer- models requiring high value agriculture with corpora- cial orientation and prospects of these farmers, and tions, such as factoring (from agribusinesses to smaller the markets for which they produce and sell, among farmers and SMEs), trade finance, and leasing, are green other relevant factors. in the chart for this environment. ƒƒ Financial institutions need to invest in understanding agriculture, as well as specific agriculture sub-sec- 3.3 Patterns, Challenges, and Solutions tors, and adapt their products to meet farmer needs. Key patterns across cases Farmer organization and aggregation models can be critical to reducing risks and costs- to-serve. Under this category of observations, we tried to include ƒƒ Value chain financing, particularly for higher value the cross-cutting elements that repeat themselves crops and tighter supply chains, and specifically almost consistently from one case to another. These when it is part of a broader package of financial and elements tend to appear in cases that showcase some- non-financial services, can be a key factor in select- thing else; for example, the case of savings as collateral ing farmers and where to lend. involves the use of producer associations. Another ƒƒ For looser value chains, particularly in staple/food example is the case of leasing to finance agriculture crops or in lower value crops and for semi-commer- equipment, which relied on credit assessment systems cial farmers, producer organizations can play an adapted for agriculture. A third example can be the case important role in effectively reaching smallholders. of providing weather insurance to farmers with train- ing and improved inputs. From the cases examined, producer organizations play a more significant role in linking farmers to mar- Looking at these cases and picking up on these common kets (input and output) and credit in looser value themes or patterns is where we find that most financial chains compared to tighter ones. In looser value institutions need to be flexible in identifying distribu- chains, particularly for lower value staple crops, it tions channels to reach farmers, assessing risks, and bun- would have been considerably harder, if not impossi- dling products and services to add value to finance. ble, for individual farmers to link to value chains and buyers had it not been for the option of going through Farmer segmentation is important to enable bankers their producer organizations. For financial institutions to start differentiating various classes of farmers. reaching these farmers, producer organizations pres- Different financing, risk mitigation, and distribution ent, perhaps, the most important and viable means of products will facilitate financially sustainable growth linkage. The challenge here is to identify well-run for various farmer segments. Segmentation helps banks producer organizations that can provide effective 52 Glob al Partn er ship for Financial I nc lu si on intermediation of credit and other services, both ƒƒ Financial institutions need to adjust and adopt appro- financial and non-financial, to farmers. priate systems to assess farmer credit risks and understand the overall risks of the specific category A core value proposition matters. The next observa- of farmers they are targeting. Overall risks beyond tion that emerges from the assessment of cases is an old credit should include price, weather/yield, and one: farmers have a variety of needs, both financial and health, among the most important ones. non-financial. The provision of financial service should ƒƒ Approaches to credit risk assessment can include support the core business of farmers in terms of help- information about farmers from value chains or pro- ing them move up the ladder and improve their ducer organizations, building specific score cards for incomes and general welfare. Financing should be part farmers (in some cases for specific crops and types of of a larger package of services to farmers for the follow- farmers), and incorporating agronomic information ing reasons: and derived cash flows. ƒƒ Provision of extension and technical assistance to ƒƒ Financial institutions need to be flexible in terms of farmers in addition to credit is found to be very collateral requirements, relying less on real estate and important. In addition to extension services, farmers more on movable forms of collateral or, in certain also benefit from basic financial literacy training. cases, on future cash flows (value chain financing). A ƒƒ In most cases, financing targeting the purchase of more flexible approach to collateral can increase better inputs led to improved yields and farmer reach so that financial institutions can lend to more incomes. farmers. However, this should be done prudently, ƒƒ Access to non-financial services, in addition to exten- with the financial institution relying on robust and sion and financial literacy, can include farm certifica- appropriate credit assessment systems to ensure that tion for sustainability, market/price information, etc. farmers have the ability and willingness to repay. ƒƒ The provision of such non-financial services need not ƒƒ Although risk sharing arrangements (guarantees or be done by the financial institution that provides the first loss sharing) and insurance did not cut across credit, but can be implemented by other participants the case studies, they nonetheless seem to have and stakeholders along a specific value chain. played important roles when used in mobilizing credit to smallholders in the specific cases. Risk management matters, both at the portfolio ƒƒ Based on the cases examined, first loss guarantees and individual borrower levels. Experience with appear to have had some initial success in getting some of the innovative risk mitigation models shows credit schemes started, as they helped to establish a that risk mitigation and overall risk management can level of trust in a new system. However, it may still be the most important interventions for a profitable be too early to judge success and financial sustain- and sustainable agricultural lending portfolio. Risk ability over time. sharing can make a difference at the portfolio level, as ƒƒ Agriculture insurance, particularly that which is based it effectively moves the critical threshold downward, on weather events, seemed to play an important role which makes some models viable. For example, indi- in several cases in dealing with severe weather occur- rect lending through farmer-based associations and rences that can cause crop losses and hamper the abil- savings-account-linked input loans become viable for ity of the farmers to repay their loans. the financial institution. Risk mitigation through insur- ance can contribute to making a lending proposition It can be concluded that financial institutions would sustainable over time — new types of personal and need to have existing risk management capabilities production insurance are the important innovations in or skills pertaining to the agriculture sector and have this regard. Risk sharing and insurance schemes can identified opportunities in the first place, before play important roles, but not in every case: insurance and risk sharing facilities can play their Innovati v e Agric u ltu ral S ME Finance Models 53 role in moving credit to smallholder farmers. needs, rather than just farming needs, can be very Insurance and risk sharing arrangements can be very important for smallholder farmers, particularly the useful in increasing the reach of financial institu- less commercially oriented ones. tions to lend to more farmers that otherwise would ƒƒ Financial institutions need to explore appropriate be on the very margin of a decision to lend. Many channels for reaching farmers, including value such farmers, particularly smallholders, can fold on chains, producer organizations, mobile banking, this margin of a decision to lend because they lack correspondence banking, etc. appropriate collateral, do not have proven cash flows ƒƒ Receiving frequent information and being able to mon- and track records of financial information, and are itor farmers through proximity (e.g., rural branches or exposed to systemic risks. correspondence banking) or through other participants along a given value chain with aligned interest with the Good product mix, including savings and sequencing financial institution is important. to build a high-quality portfolio, are critical. Based on robust target group segmentation, the right mix of Roles for donors and DFIs within the products (lending and savings) and sequencing are case studies important in order to develop a good understanding of the value chains into which farmers are linked. On the In examining the 37 case studies listed at the end of lending product side, short-term finance usually comes this report, we identified the roles played by donors first. Overall, the main objective is to build a high- and DFIs within them. This is not a comprehensive quality portfolio (low NPLs, new deposits from farm- list, nor is it a representative sample of donor and ers). Volume and reach can follow on the heels of lead DFI involvement in agricultural finance, which is farmers promoting the bank that serves them well, quite extensive. Nevertheless, a brief survey of these combined with an increasing base of knowledge of the 37 cases sheds some light on the roles and inputs agriculture sector within the bank. Building a high- that have been present. quality portfolio entails intense monitoring of borrow- ers throughout the production cycle, especially at There was active donor and DFI involvement in harvest season. It also involves managing concentration about one third of the 37 case studies. These activi- risk in terms of crops and space. ties were varied, but a few key areas in which donor and DFI involvement was evident include: A good analysis of the key drivers of profitability for ƒƒ First loss guarantees supported by donor funding both the banker and the value chain in which farmers (eg, AGRA and IFAD). In some cases, providing act is critical. Distribution matters for costs and reach, funding for first loss guarantees allowed donors to but does not alter the nature of the lending business. participate without distorting the market. The mobile banking and branchless banking models ƒƒ Technology support, either by sharing, enhanc- can leapfrog traditional development patterns because ing, or providing scale-up and access to existing information and financial services travel fast and technology. This route is fairly straightforward, directly to farmers. In fact, they might allow banks to and constitutes a basic yet effective way to sup- generate more economies of scale as well as a better port financing institutions. An example is a case deposit base. in which one financial institution providing ƒƒ Savings, payment systems, insurance (personal and loans could capitalize on its larger subsidiary and crop), and loans for children’s education are all very the government facilitated all documentation and important in improving farmer livelihoods in addi- verification procedures, thereby reducing associ- tion to crop-specific credit. Taking a more holistic ated costs to the bank and allowing for more approach to serving the overall household financial favorable loans. 54 Glob al Partn er ship for Financial I nc lu si on ƒƒ Capital for risk-sharing schemes and insurance. Some there are emerging finance models that offer secu- NGOs and governments provided initial funds to rity through other sources. While pledging land as support insurance schemes, as well as knowledge collateral has its challenge, lenders are increasingly sharing and the provision of information. able to underwrite anticipated cash flows from the ƒƒ Seed capital, which enables innovative financing farmer and the buyer, and use other forms of collat- business models to begin. Contributions from eral, such as movable collateral. Inventory financing donors and DFIs pave the way for start-ups. using warehouse receipts and collateral management agreements provides a secure form of lending for Particularly in Africa, there is much room for innova- financial institutions. However, inventory financing tive donor financing to enable the private sector. is mostly useful for the post-harvest period, and Innovative models of donor support can be through more suitable for the needs of traders, processors, PPPs, such as the case of the Africa Agriculture and and producer organizations. Trade Investment Fund (AATIF)32, with contributions to first loss layers, as well as investments in agricul- Value chain finance models hold much promise, tural inputs, and value chains. It also provides a but require a great deal of engagement and cannot wholesale refinancing line to regional development apply everywhere. Robust and competitive value banks (Case 37 Page 99). chains represent good financing opportunities for financial institutions. The weakest link of all value Challenges and Emerging Solutions chains is side-selling by farmers — a risk that can be mitigated in some ways. Hard mechanisms of buffer- Systemic risks abound in rain-fed agriculture, but ing these weak parts are those “constriction points” in there are innovative insurance models that can the chain that simply drive up the real cost of side- effectively mitigate those risks. Reducing exposure selling for the farmer: sustainable price premiums, to or mitigating the systemic production risk in the perishable crops, enforceable legal sanctions, and joint creditor — borrower relationship can make that rela- liability groups. Softer measures to deal with this tionship a viable one. Weather-based insurance solu- aspect are those incentives that show the farmer that tions could become a viable mitigation technique that there are strong associated benefits together with can increase access to financing — but under certain finance: technical assistance, advance cash payments preconditions and if the necessarily public goods during the offseason, women’s clubs, personal insur- (data, weather stations) are provided. Dealing with ance packages, mobile payment systems. price risks for smallholder farmers is still a challeng- ing area. The more appropriate solution seems to be Serving farmers with small transactions can be costly. through contract farming or forward contracts (fixed However, producer organizations, value chains, and prices), but this requires an applicable legal and reg- mobile banking channels can provide lower cost distri- ulatory environment or strict control in delivering bution channels for financial products. the goods to enforce contracts. Futures and options contracts are viable for internationally traded com- Agriculture is quite heterogeneous and agricultural modities in commodity exchanges and are more expertise is needed by financial institutions to assess suitable for large farmers and more sophisticated opportunities and develop relevant financial prod- agribusinesses. ucts. Financial institutions should develop agricultural lending expertise over time. Expertise cannot be Smallholder farmers often lack the traditional acquired instantly, especially for multiple crops and/or collateral required by lending institutions, but value chains. It is important to start with the easier cases, 32 This donor supported investment fund is but one example of specifically donor-funded investments to the agricultural private sector. Innovati v e Agric u ltu ral S ME Finance Models 55 so that the learning phase does not jeopardize the devel- lending expertise, it is also important to help financial opment of a profitable portfolio. Character-based lending institutions identify bankable opportunities. techniques are combined with technical criteria in selecting borrowers, setting loan terms, and enforcing For innovative models, the key issue is implementation, repayment.33 In agricultural extension work, the strat- which depends on local conditions. A generic frame- egy often is to work with the best farmers, then obtain work for implementing innovative models is reflected a multiplier effect. In order to develop agricultural the following figure: Figure 14  Framework to Implement Innovative Models. The Who, the How, and the What Understand and analyze commodity subs-sectors ƒƒ Who are the key players? ƒƒ How are they connected to farmers? ƒƒ What are the existing financial arrangements? ƒƒ What are the main risks? Segment farmers ƒƒ Who are these farmers and key characteristics? ƒƒ How are they organized? ƒƒ What is the credit gap? ƒƒ What are their financial and non-financial needs? Determine distribution channels ƒƒ Who can provide the financial and non-financial products? ƒƒ How the delivery mechanism could work to reach farmers? ƒƒ What are the roles of the various parties involved? Pilot and scale up ƒƒ Who can be the first participants? ƒƒ How to access success of the pilot? ƒƒ What would it take to scale up and by how much? 33 Christen and Pearce (2005) 56 Glob al Partn er ship for Financial I nc lu si on Potential Areas for Policy particularly in Africa. Mechanisms could be mod- Interventions eled after that in the health sector in 2005 to pro- mote vaccinations in Africa. The findings from the case studies support the policy recommendations made in IFC’s previous report 5. Strengthen producer organizations as important (“Scaling Up Access to Finance for Agricultural SMEs: Policy aggregators for delivering financial and non-finan- Review and Recommendations,” October 2011). They high- cial services to smallholder farmers. This can light further certain areas where policy interventions involve capacity building for financial and mana- and the G-20 convening power could indeed further gerial skills as well as improved corporate gover- strengthen the effectiveness and scaling up of financ- nance. A number of NGOs and initiatives already ing for agricultural SMEs. These include: seek to strengthen producer organizations, but a more conscientious effort and a bigger scale may 1. Support for first loss/guarantee funds for agricul- be needed. ture, particularly focusing on smallholder farmers and agricultural SMEs. This should leverage GAFSP 6. Promote PPPs by which governments could lever- as well as the Global SME Finance Initiative, both of age private sector funding and management to which have been supported by the G-20 already, improve longer term investments in agriculture rather than a new initiative. However, this may infrastructure and provision of technical services. require some additional resources if the scale of the Agriculture-related infrastructure could include activities is to expand significantly. warehouse facilities for improved storage of com- modities, cold storage, irrigation infrastructure, 2. Provide support for catastrophic insurance basic processing of certain food commodities for approaches to protect farmers and financial insti- local consumption, etc. A recent example is the tutions from severe losses. Since this industry is Africa Agriculture and Trade Investment Fund still evolving, donor and partner interventions can (AATIF), which provides funding to the private play a critical role in accelerating its development sector as well as wholesale financing lines to and deployment in emerging markets. regional development banks. 3. Promote the creation of a forum of large agribusi- 7. Support capacity building for financial institutions nesses that could be encouraged to leverage their in emerging markets, and facilitate its further sup- networks in emerging markets and develop opportu- port by donors, DFIs/IFIs, and foundations. nities for attracting financial institutions that could Capacity building is essential to provide the neces- fund parts of their value chain, like local small trad- sary skill transfer to financial institutions in order ers, processors, farmers, etc. Financing could be to better understand the agriculture sector, ana- linked and become the catalyst for technology lyze risks, develop appropriate lending and other improvements and promotion of environmental and financial products, and find cost-effective distri- social standards along specific value chains. bution channels to reach smallholder farmers, including the skills to forge value chain partner- 4. Create mechanisms to promote the adoption of ships. Experiences thus far have shown that it is technologies for agriculture (agriculture pull mech- also important to help financial institutions iden- anisms) that could increase yields and improve the tify bankable opportunities in the agriculture quality of crops, particularly food crops. There is a space to quickly develop a pipeline of projects to huge scope to increase yields and improve quality, provide financial services. Innovati v e Agric u ltu ral S ME Finance Models 57 CHAPTER 4 Conclusion Lending to farmers, particularly those with poten- report demonstrate, banks and other financial insti- tial to become more productive, can contribute to tutions are beginning to recognize the growing higher incomes and push farmers up the pyramid potential and profitability of these enterprises that from subsistence or semi-commercial farmers to are often generally feared but little understood. commercial farmers. It can drive rural economic They are seeing that a unique challenge is some- and social growth. Although lending to farmers has thing to be understood, not feared — that once its unique challenges, it can be profitable, help to appropriate financing, risk mitigation, and distribu- diversify the risk profile of a bank’s overall lending tion products are offered to these highly productive portfolio, and lower funding costs through new and increasingly important producers and proces- deposits from farmers. There are some indications sors, lending portfolios may become more resilient that NPLs are largely independent of the size of the and profitable. agricultural portfolio. Countries with a high share of agriculture in their GDP show NPL ratios in their The case studies also indicate that there is no such a banking system that are no higher than those of thing as a magic bullet in financing the agriculture countries with much lower shares of agriculture in sector. There are no single key innovations that their GDP (see Annex B). Moreover, the global eco- alone can unlock the great potential for lending to nomic crisis of 2008 showed the benefits of bank farmers. Rather, banks and financial institutions diversification across various sectors, including should better invest in understanding agricultural agriculture. NPLs of countries with higher shares of activities and should segment farmers in order to agriculture in their GDP appear to have been less understand what they need. Financial institutions affected by the crisis, partly because of the growth should explore linkages that farmers may have with in agricultural prices and the fact that overall profit- supply chains, evaluate farmer-based organizations ability of agriculture during that same period was and cooperatives to act as aggregators for financial not linked to the financial crisis. Agricultural lend- and other services to farmers, and look into distri- ing can be a desirable business for banks that have bution channels that can lower the cost to serve the systems and experience to understand farmers farmers. Better understanding of the local context and supply chain risks. Rabobank, as one of the and the environment in which farmers operate can contributors to this study, has noted that agricul- provide unique solutions on how to provide finan- tural SME NPLs generally outperform those in other cial services with lower risks and lower administra- sectors. In particular, the more commercially ori- tion costs. Thus, solutions tend to be more localized ented farmers in the middle of the pyramid have based on crops, types of farmers, types of value the potential to become the “Mittelstand” that chains, etc. Insurance also plays an important role, played such a pivotal role in growing the German particularly in dealing with severe weather events economy in rural areas. As the case studies of this that cause crop losses. In most cases examined, 58 Glob al Partn er ship for Financial I nc lu si on however, insurance was not a pre-condition for needs of farmers. Savings, insurance, payment sys- credit: that is, the absence of insurance did not pre- tems, and loans for non-farming activities can be vent the flow of credit. very important in serving the farmers’ overall household financial needs beyond credit for the On the risk mitigation side, there has been some farming activities. Compared to larger commercial innovation in terms of solutions to deal with yield farmers, smallholder farmers in particular value risks, mostly focusing on weather or area-based approaches that address their overall household index insurance. Price risk management has proven financial needs. more challenging, mostly focusing on widely traded commodity prices in liquid exchanges and Because of the heterogeneity of farmers across in large markets (e.g., North America, Europe, countries as well as across crops within the same South Africa, Brazil, Argentina, China, and India). country, we have encountered many models and On the credit risk side, there have been some inno- approaches. Although we have tried to draw lessons vations on credit risk assessment using credit scor- from our observations of innovative models and ing systems based on statistical and behavioral approaches, the local context plays a key role, and observations in agriculture portfolios. These credit in implementing these models, a certain degree of scoring systems usually supplement other sources customization can be crucial. At the same time, we of information that financial institutions have about observe an evolution of approaches and models farmers, either from supply chains or producer within countries. A number of the models and organizations. approaches examined are relatively new. This means that there is a need to continue monitoring Finally, agricultural finance should be part of the and evaluating these experiences and keep dissemi- bigger picture to provide farmers with solutions to nating information related to how the implementa- improve their incomes and livelihoods. As such, tion is proceeding. Lessons and experiences from access to financial services is often linked to access these innovative models and approaches are evolv- to improved inputs, extension services, financial lit- ing, and this paper is a first attempt to synthesize eracy training, certification for sustainable prac- the observations so far. tices, and market/price information, which are among the factors that can contribute to improved In examining the case studies, we observed certain conditions for these farmers and ultimately lead to patterns and distilled some key elements that seem higher incomes. In many cases, we observed that to cut across them. Nevertheless, a limitation of this provision of credit, combined with access to better paper is that it cannot present the quantitative evi- inputs and extension services, led to improved dence that might back stronger conclusions. Many yields and incomes. Thus, bundling credit into a of these cases are relatively new in terms of their broader array of financial and non-financial services implementation period, so that their impact, finan- for farmers has a greater impact. Access to finance cial sustainability, scalability, and replication are yet combined with other services can facilitate the to be proven. However, these innovative cases show growth of farmers from semi-commercial to com- some promising signs and early lessons, which is mercial. Financial innovation can therefore be a why they are included in this report. A follow-up critical catalyst for change. study would involve gathering quantitative data in order to formulate and test hypotheses, thus pro- In addition to lending, financial institutions should viding quantitative evidence to back the lessons take a more holistic approach in serving the various learned and conclusions drawn so far. Innovati v e Agric u ltu ral S ME Finance Models 59 In addition to the above, there are some additional become effective in providing or facilitating credit areas that might warrant further exploration: to their member farmers. A first key area would be the issue of formal versus A third aspect worth noting is the role of govern- informal credit. Some surveys indicate that farmers ments, donors, and IFIs. In a number of cases, donors do indeed prefer formal financial services to infor- or IFIs provided first loss risk-sharing arrangements mal, although they find that dealing with formal to encourage lending by financial institutions. It will financial institutions is too complex and requires be important to analyse these cases to determine the too much collateral, and the lack of proximity is key ingredients for the success of such schemes. At often an issue. Informal credit has the advantage of the same time, there are governments that direct simplicity, close proximity, and almost instant credit to the agriculture sector, either through state availability. However, the same surveys indicate that agricultural banks, or through mandatory lending those farmers who do prefer formality may not quotas, and it would be important to assess the effec- necessarily be borrowing as a substitute to informal tiveness and performance of such directed lending credit but to meet their additional financial needs, on smallholder farmers. as most farmers indicate that their needs are only partially fulfilled. Research is needed to explore the Finally, while innovation has been mostly focused roles of formal and informal financial services to on commercial farmers and, to a certain extent, farmers, as well as to quantify, if possible, the semi-commercial farmers, it is essential to under- potential benefits from increasing reliance on stand the needs of all farmers and how to improve formal channels. access to finance to the subsistence and semi-com- mercial farmers of staple crops. Recent work on A second area to be explored is related to the role, behalf of CGAP aims to address this issue. An early and appropriate structures and strengthening, of observation from this work is that these farmers, producer organizations. The cases examined so far and rural households at the very bottom of the indicate that the targeting of semi-commercial pyramid, would benefit from general financial smallholders in mainly staple/food lower value products, while there may be few cases in which crops benefits from producer organizations. specific products need to be developed to cover Research is needed to investigate what role the pro- particular risks and cash-flow needs from agricul- ducer organizations can play and how they can tural activities. 60 Glob al Partn er ship for Financial I nc lu si on CHAPTER 5 Case Studies Summaries 34 CASE 1  Equity Bank, Kenya — “Kilimo Biashara” Direct Smallholder Lending Equity Bank is the largest bank in the region, with 5.7 million accounts, over 57 percent of all bank accounts in Kenya, and operations in Uganda and Southern Sudan. Equity Bank commenced business in 1984 and evolved from a building society and microfinance institution to an all-inclusive commercial bank listed on the Nairobi Stock Exchange and Uganda Securities Exchange. Equity Bank’s approach to agricultural financing is based on direct smallholder lending integrated into a larger supply chain partnership and supported by a first loss guarantee provided by donors. Equity Bank signed a partnership with AGRA, IFAD, and the Government of Kenya in May 2008. The deal includes a loan project of USD 50 million in agricultural SME loans for farmers with little or no collateral. AGRA and IFAD provide a 10 percent first loss guarantee. Under this partnership, Equity Bank developed the small- holder financing product “Kilimo Biashara,” which is designed to make financing available for 2.5 million farmers and 15,000 agricultural input retail businesses in rural areas. Equity Bank enhances security by (i) capping loan exposure at USD 17,000 per farmer, (ii) applying group lending terms, whereby six farm- ers act as co-guarantors, and (iii) reducing the cash amounts in farmers hands (farmers can pay agro- 35 dealers out of their Kilimo Biashara credit). By June 2008, USD 18.75 million in loans has been disbursed, reaching 37,000 beneficiaries. 36 The loans carry a 12 percent interest rate applied when the loans fall due — a rate well below Equity Bank’s stan- 37 dard lending rate of 18 percent. According to Equity Bank, the project is a success because it has changed the position of smallholders from food insecure to semi-commercial producers. One of the suc- cess factors is the technical assistance on financial literacy and farm management provided by the gov- ernment extension service bureau to the farmers. The repayment risk of the individual farmers is mitigated by their integration into supply chains, including WFP’s P4P program. 34 Unless otherwise referenced, all data included in case study summaries were taken from personal correspondence or interviews with the subject institutions by the authors of this study, generally either RIAS consultants or IFC & World Bank staff members. 35 AGRA (2009) 36 IFAD (2010) 37 AGRA (2009) Innovati v e Agric u ltu ral S ME Finance Models 61 CASE 2  Opportunity International, Africa — Informed Lending Direct Smallholder Lending in Ghana, Rwanda, Mozambique, Malawi, Uganda Opportunity International is one of the largest microfinance institutions in the world. Opportunity has started agricultural lending under the “Informed Lending” Production Finance Model. “Informed lending” is a parametric lending model anchored on: (i) the exact mapping of the borrower farm’s plots (plot sizes, altitude, access to water); (ii) a diagnostic of the borrower’s household profile (demographics of the family, breakdown of all farm enterprises such as crops/land used, other sources of income/activity, access to water/roads/bank, mobile phone use); and, (iii) the crop profile, including costs of inputs and labor, and returns based on yield and price data. Combined with targeted extension services, the financed farmer often improves food crop yields, allowing the reallocation of a portion of the land to cash crops. In addition to these assessment techniques, where appropriate Opportunity seeks to secure loan recovery by concentrating on cash crops with regulated output buyers (such as cocoa in Ghana and tobacco or chilies in Malawi). For these contract farming financing operations, Opportunity helps to reduce the risk of side-selling through advance cash provision to farmers during the lean season when farmers are most likely to succumb to the temptation of side-selling. The total number of target agricultural clients in 2011 was 41,000 farmers in Ghana, Malawi, Mozambique, Rwanda, and Uganda. The best estimate of the number of disbursed loans by the end of 2011 was only USD 29,000, however, with most of the shortfall occurring in the Malawi loan program. 62 Glob al Partn er ship for Financial I nc lu si on CASE 3  HDFC Bank, India — Correspondent Banking Direct Smallholder Lending HDFC was established in 1994 with a mission to be a world-class Indian bank. It currently has 2,201 branches and more than 5,000 ATMs in 1,174 Indian towns and cities, with a balance sheet size of USD 55 billion in December of 2011, of which approximately 14 percent is financed in agriculture. HDFC’s farming lending model features small value loans of less than USD 10,000 against the mortgage of agricultural land (75 percent of the farmers are landowners) and based on a credit feasibility study. This study — a form of parametric financing — takes into account land holding, crop selection, input cost, and the strength of the underlying commodities to generate cash flow estimates. Previous client data and past experience, combined with extension service use, is also considered. Monitoring occurs after every harvest to ensure that accounts are serviced and kept regular. Most of HDFC’s lending is limited to staple crops like rice, wheat, cotton, and sugarcane. In order to reduce transaction costs, increase exposure, and provide a “one stop shop” for farmers, HDFC promotes correspondent banking in partnership with actors across the agricultural value chain. Through their Correspondent Operation Centers, which are attached to fertilizer dealers and other input distributers, dairy cooperatives, and sugar mills, HDFC offers an “end-to-end” payment system, reducing 38 the number of transactions and increasing revenue. Besides offering credit and collecting payment, business correspondents often deliver inputs and offer extension services. HDFC’s holistic approach to farmer financing considers farmers’ anticipated cash flows and their overall participation in the value chain. HDFC’s model therefore assumes that the sector’s hesitation in financing the “missing middle” can be traced to a dearth of cash flow documents and high transaction costs, contributing to low profit mar- 39 gins for banks. Direct loans handled at the HDFC headquarters account for 90 percent of present business. Gross prof- its currently range between 15–20 percent, though this is lower than other segments. Even though losses are low, the transaction costs of working in remote areas, competition from state banks with subsidized loans, and costly standard accounting practices persist as key challenges. 38 Andrade (2011) 39 Ibid. Innovati v e Agric u ltu ral S ME Finance Models 63 CASE 4  Zanaco, Zambia — Munda Smallholder Scheme Indirect Smallholder Lending Zanaco (Zambia National Commercial Bank Plc) is one of the leading banks in Zambia in terms of customer deposits, total assets, and points of representation. It was partially privatized in April 2007 when Rabobank acquired 49 percent of the Government’s interest in the Bank. Zanaco was listed on the Lusaka Stock Exchange in 2008 and is considered “Citizen Owned” with over 50 percent of the shares owned by Zambians and the Zambia National Farmers’ Union (ZNFU). The Munda credit facility provides smallholder farmers better access to finance in order to help them to grow their business and to offer a practical alternative to the discontinued national Input Support Program, in which the Government had provided inputs to farmers. Zanaco lends to District Farmer Associations (DFAs) that are affiliates of ZNFU. Before each growing season, DFAs assess the total needs for inputs of their pre- dominately maize-growing smallholder members, then submit their requirements to ZNFU to tender for the accumulated need for fertilizer and seeds. Zanaco finances these inputs backed by 50 percent cash collat- eral, deposited by the DFAs. Farmers organized in co-operatives or associations and members of ZNFU through the DFAs are able to purchase seeds and/or fertilizer from input suppliers such as Omnia Fertilizer Zambia Limited and Seed Company/Zamseed. At the end of the maize season, the harvest is sold to the pro- cessor, who channels proceeds to Zanaco, which then deducts the outstanding loan balance from that amount, along with accrued loan interest and other associated costs, such as crop insurance premiums. The remaining surplus then flows back to the individual smallholders through their DFA. Via a DFA, the collective smallholders are responsible for the loan repayment of each individual, according to the principle “all for one”. If repayment is not made on time, participation in Munda for the DFA in the next season is prohibited. In the 2011/12 farming season, the scheme disbursed a total loan amount of USD 4 million to 25 DFAs repre- senting 4,026 participant farmers working on 10,088 ha, up from 600 farmers and 600 ha in the 2008/09 farming season. These farmers’ yields have also increased from an original 1.5 metric tons per hectare (MT/ ha) in 2008/09 to 3 MT/ha during the 2010/11 season on account of improved use of hybrid seeds and fertil- izers, and adoption of conservation farming practices spearheaded by ZNFU. Zanaco forecasts the number of financed farmers to increase to 10,000 in 2012/13. The current interest rate is ZMK Base Rate (16 percent 40 per annum (p.a.)) minus 5 percent (11 percent p.a.), which is a competitive rate in Zambia. The arrangement fee is USD 100 per DFA, and the borrower does not have to provide additional collateral beyond the cash. From a bank point of view, the portfolio performs well, and no defaults have been registered thus far. The reason for such a performance is twofold: first, a cross default is at stake since Zanaco finances DFAs, who collect the 50 percent cash collateral and use the “all for one” principle. Second, the Food Reserve Agency of the government sets the price early in the season at a particularly high level to cover the amount they buy for food security purposes. This price will hold for a minor part of the crop bought by the Government, and drives the general market price for maize above competitive levels. Currently, maize is the predominantly financed crop. The model heavily relies on the Zambia National Farmers’ Union organization in its operational and risk management activities; therefore, it might not be replicable in contexts with weaker smallholder organizations. The model’s sustainability also relies on crop diversification at the farmer level and enhanced corporate governance skills at the DFA level. Lastly, the 50 percent cash collateral — instead of a legal claim on the harvest as collateral — reduces the leverage of the farmers involved. Nevertheless, there is ample demand among farmers for the scheme. 40 Zanaco (2012) 64 Glob al Partn er ship for Financial I nc lu si on CASE 5  Finterra, Mexico — Emerging Farm Business Financing FinTerra is a leading non-bank financial institution in Mexico focused on lending to SMEs in the agribusi- ness sector, primarily producers of fruits, vegetables, grain crops, dairy, livestock, and specialty crops, along with various agricultural-related businesses. FinTerra was founded in 2004 with an initial capital of MXP$ 48.0 million. As of August 30th, 2010, FinTerra´s equity was MXP$ 295 million. Since 2007, Standard and Poors (S&P) has rated FinTerra as BB+ / Stable outlook. FinTerra has outstanding loans of MXP $1,075 million (approved credit facilities > MXP $1,400 million), and currently has 11 branches located in 10 Mexican states. FinTerra’s main lending product is the Individual Loan (87 percent of the portfolio). As of 2010, FinTerra had around 405 clients under this category. Individual Loans are both short- and medium-term, averag- ing the equivalent of around USD 1.9 million equivalent on average; the loans finance working capital and/or fixed assets investments of the clients. The individual loan borrowers are mostly SMEs, enter- prises with annual sales in the MXN 1–50 million (USD 0.90–USD 4.5 million) range. Other loan products include Grower Financing Loan Programs (MXP$ 300 million outstanding), through which FinTerra finances the suppliers of large food and beverage companies. Each loan program has a limit of total exposure, ranging from USD 1.0 to US$ 4.0 million equivalent, and 10,761 borrowers with “sub-loans” in the program, ranging from USD 10,000 to USD 100,000 equivalent. Innovati v e Agric u ltu ral S ME Finance Models 65 CASE 6  Zanaco, Zambia — Emergent Farmer Finance and Support Program “ZEFP” Zanaco (Zambia National Commercial Bank Plc) is one of the leading banks in Zambia in terms of cus- tomer deposits, total assets, and points of representation. It was partially privatized in April 2007 when Rabobank acquired 49 percent of GRZ’s interest in the Bank. Zanaco was listed on the Lusaka Stock Exchange in 2008 and is considered “Citizen Owned” with over 50 percent of the shares owned by Zambians and the Zambia National Farmers’ Union (ZNFU). The ZEFP seeks to combine access to finance with support services for emerging farm business. A pilot project was completed in August 2009, in which Zanaco provided the financing (both working capital and investment finance), IFC and Rabo Foundation financed the technical assistance program (via a grant that was channeled through ZNFU), and Rabo Development provided bank capacity building to Zanaco. This included agricultural credit skills and farm performance monitoring training to Zanaco. The farmers were also trained in farm management and financial skills. In addition, external specialists pro- vided support for individual farmer loan applications and business plans. An important aspect of the program is the involvement of key agricultural input suppliers and off-takers: the South African fertilizer company Omnia has a crucial role in soil sampling and determining the fertilizer program together with the farmers. Cropserve does the same for the agri-chemicals. Other partners such as Parmalat, Afgri, and Zamace (the local Agri Commodity Exchange) are committed to covering the marketing link to the program. The involvement of the project partners is commercially driven: all parties acknowledge the immense growth potential of this group. As of December 31, 2011, the program had provided loans to 123 farmers, with a total loan portfolio of USD 4.5 million. Some delays in achieving results have been related to the limited number of standalone farmers under the emergent segment. Zanaco has diversified by incorporating value chain financing in sectors with strong market linkages to develop its own segment of emergent farmers. Although most commercial farmers and small-scale farmers are members of ZNFU, the majority of farmers under this segment are not members. Therefore, the bank has diversified its marketing strategy by running adver- tisements on the program as a way of reaching out to them. The technical assistance program imple- mented by ZNFU/Rabobank and IFC has led to enhanced practices by the Bank and improved the productivity of the participating farmers. Zanaco hired and trained a group of new agri-loan officers to strengthen the agri-finance capacity of its branches. Without agri-finance capacity building of its rural branches, the ambitious growth targets for emerging farmers would not be feasible. The program plans to expand into sugar, pork, rice, and dairy production, as these sectors have relatively strong market link- ages that mitigate the risk of cash diversion by farmers and reduce reliance upon land collateral. 66 Glob al Partn er ship for Financial I nc lu si on CASE 7  NMB, Tanzania — Kilimo Account Product “KAP” Savings Account Linked Input Finance NMB (National Microfinance Bank) has become Tanzania’s largest financial services provider, with a growing customer base of more than 1.4 million people. In the KAP, farmers open a personal account, a NMB Kilimo (saving account), and apply for a loan account. After harvest, the farmer deposits part of the harvest proceeds in the Kilimo Account, which is then used as cash collateral for input financing in the following season. The NMB Kilimo Account is designed to encourage farmers to save the earnings from their harvest sales. The incentive to save comes in two ways: i) a high interest rate (3 percent) plus bonus (3 percent) when less than two with- drawals occur; and, ii) the ability of the farmer to take out a loan equal to three times his or her savings balance for the purchase of agricultural inputs. NMB targets those farmers that participate in the ware- house receipt financing schemes through their Primary Cooperative Societies. Given that maize, sun- flower seed, cashew, and coffee are less integrated with buyers, these farmers are typically not eligible for outgrower or value chain financing and therefore find it particularly difficult to access input finance. In order to qualify for the KAP, the farmer must have his or her own farm-produced non-perishable crops, harvest records of at least 3 recent years, be a member of MCOS/primary Cooperative Societies banking, and operate solely with NMB. In addition, the farmer needs a guarantee and reference letter from the AMCOS/Primary Cooperative Society. The scheme was launched with 1,080 farmers (total loans of around USD 200,000). Innovati v e Agric u ltu ral S ME Finance Models 67 CASE 8  Banco De Lage Landen, Brazil — Equipment Finance De Lage Landen International (DLL) is an international provider of leasing and asset finance, a fully owned subsidiary of the Rabobank Group. The equipment financing portfolio has almost entirely been generated in partnership with agricultural equipment vendors. Through Brazil’s national development bank (BNDES), general banks and financial institutions can provide finance to the agricultural sector at subsidized rates. Nonetheless, DLL has dis- tributed more funds via equipment finance to the agricultural sector than these general banks have done. Key for De Lage Landen’s approach are (i) a deep understanding of farming and of the agricultural value chain in Brazil; (ii) a thorough knowledge of agricultural equipment, control over its distribution chain; and, (iii) knowledge of the collateral value of the equipment and how to remarket it if needed. Most leases have a down payment or another form of client equity in the transaction and a cash collat- eralized partial guarantee from the dealer. Virtually all of DLL’s portfolio consists of loans. Brazilian farm- ers want to own their equipment and prefer loans over leasing. The legal options to repossess an asset under a loan and pledge structure are not fundamentally different from those for a similar asset under a leasing structure. De Lage Landen has built up an agricultural finance portfolio of around USD 3 billion in Brazil over the past 13 years. 68 Glob al Partn er ship for Financial I nc lu si on CASE 9  Mahindra & Mahindra Financial Services, India — Equipment Finance Mahindra & Mahindra Financial Services Limited (MMFSL) is a subsidiary of Mahindra & Mahindra Ltd. and part of the diversified industrial conglomerate Mahindra & Mahindra Group. The M&M agricultural division is one of the leading tractor brands in the world by volume. MMFSL has specialized in rural and agricultural finance with virtually no urban presence. MMFSL provides asset finance, because leasing has tax disadvantages in India. The financing is based on loans, with a pledge over the equipment. The standard down payment is 25 percent with no requirement for further collateral or credit or buy-back guarantees from dealers. The dealers refer their clients to MMFSL to finance used equipment as well, regardless of the brand. Most MMFSL clients have just a few acres of land. They make the business case for their tractor and other equipment by working the land of other farmers and hiring out equipment as well as via non-agricultural uses of it. Eighty percent of MMFSL’s clients are first-time borrowers with no relation with any other financial institution and with no financial statements that can be used for a credit analysis. Understanding farmer’s earnings capacity had to become a core capability of MMFSL. Starting in 1993, MMFSL has built up a USD 3 billion portfolio, with about 50 percent in equipment for farm- ers since 1996. MMFSL is very successful in terms of reputation, financial strength, and outreach. This suc- cess seems to be rooted in a distinctive rural strategy that extends beyond equipment finance. MMFSL is a non-bank financial institution, offering a broad range of financial services: Asset finance, including finance for used assets; and personal loans, housing finance, insurance (brokerage), fixed deposits, and mutual fund schemes (distribution). MMFSL operates via its own branch network of 556 branches with 10,300 employees throughout the country and has a presence in 80 percent of all the districts in India. Innovati v e Agric u ltu ral S ME Finance Models 69 CASE 10  Jain Irrigation, India — Equipment Finance Jain Irrigation Systems LTD (JISL) is the largest manufacturer of micro-irrigation systems (MIS) in India, and the second largest globally. In 120 districts with a total workforce of 7,000, JISL focuses on small farmers with holdings of less than 10 acres. JISL is listed on the Bombay stock exchange with a market capitalization of USD 2 billion. JISL pre-finances farmer micro-irrigation system purchases against the “collateral” of a government sub- sidy payment. Farmers who buy irrigation equipment from Indian MIS producers such as JISL obtain a subsidy of 50–70 percent of the equipment purchase price, with the national Government supplying 40 percent and the state Government providing the remaining 10–30 percent. The JISL dealer prepares the cost estimate, and the applicant farmer submits a subsidy application to the local subsidy Implementing Agency (IA). After approval, subsequent MIS installation, and IA inspection, the subsidy release requires a 180–360 day processing period. In 4 years, JISL expects to cover an additional 1 million hectares with drip irrigation systems. The delay in the Government processing of the subsidy limits working capital and creates uncertainty for JISL, therefore IFC invested in JISL to improve cash flows and ensure the working capital necessary to expand operations and grow sales. 70 Glob al Partn er ship for Financial I nc lu si on CASE 11  Development Finance Company Uganda Leasing, Uganda — Leasing Development Finance Company Uganda (DFCU) is a leading commercial bank in Uganda. DFCU is listed on the Uganda Stock Exchange, and operates 29 branches throughout the country. Five percent of the bank’s credit portfolio is in the agricultural sector. The first three rural branches (in Mbarara, Mbale, and Hoima) were opened in 2000 as part of a project funded by the U.S. Agency for International Development (USAID). DFCU has since started leasing oper- ations in other towns like Lira and Arua, using its own funds. DFCU specializes in providing finance leases to SMEs for agricultural machinery — particularly tractors, milk equipment, harvesters, and agro-pro- cessing equipment. Typically, DFCU finances 60 percent of the asset purchase, while the client finances 41 the additional 40 percent. The client share, however, may range between 10–50 percent. DFCU retains full ownership during the life of the lease, though the asset is transferred to the client or sold to a third party after the lease terminates. Although DFCU’s interest rates are similar to those of banks, their leases are more attractive to SMEs because they typically offer longer payment periods (3–5 years as compared with around 2 years), provide flexible lease payment schedules that match enterprise cash flows, and recognize the leased asset as primary collateral. Additional security is only requested in specific circumstances. Cash flows are evaluated through documentation of income sources, 3–5 years of audited financial statements, and company history and business plan. It is the borrower’s obligation to select the equipment and submit an inspection report with an invoice. The asset must be insured during the entire life of the lease; the DFCU Insurance Premium Financing facility is available and may be tied to lease payments. DFCU staff engineers regularly monitor assets, and lease officers supervise clients in delinquency. Out of the 231 leases that DFCU facilitated in 2011 (valued at 18.3 million), 19 of these were agricultural (2.2 million). Roughly 20 percent of the bank’s agricultural credit occurs through their leasing opera- tions. DFCU reports 32 percent portfolio growth for leasing in the previous year, with non-performing assets (NPAs) limited to 1 percent and write-off at 0.4 percent. Key success factors include technical and agricultural expertise to offer leasing products that meet consumer needs and quick turnaround time to permit equipment use during the current season. 41 Agrifood Consulting International (2005) Innovati v e Agric u ltu ral S ME Finance Models 71 CASE 12  NMB, Tanzania — Warehouse Receipt Finance Cashew and Coffee Warehouse Receipt Finance NMB (National Microfinance Bank) has become Tanzania’s largest financial services provider, with a growing customer base of more than 1.4 million people. Rabobank acquired a 35 percent stake in the NMB in 2005, when the bank was partially privatized by the Tanzanian government. The warehouse receipt secured loans are given to registered farmer groups, individual farmers, com- modity traders, and businesspersons dealing with non-perishable commodities such as coffee, maize, cashews, and nuts. A warehouse receipt financing system was developed together with technical assis- tance from Rabobank in early 2007. Its funding is extended against a commodity stocked in the Bank’s controlled and authorized warehouse after submission of a warehouse receipt. The Bank holds the crops in the warehouse until buyers purchase and pay for the crops. Thus NMB can provide funds to farmers to enable them to continue preparing for the next crop while their goods are being stored. The bank provided a total of some USD 16 million in facilities to Primary Cooperative Societies (PCS) in the coffee, cashew, maize, sunflower, and sesame sectors; around 110,000 farmers benefited in 2010. Raw cashew nut prices for the farmers at the farm gate could be as low as TZS300 per kg. Thanks to the warehouse receipt system, farmers can sell their cashew nuts through their primary co-operative societ- ies, who in turn will auction the products in bulk. Cashew nut farmers can achieve an average price of up to TZS710 per kg. This scheme benefits from a 50 percent guarantee provided by the Government. To date, NMB has incurred no losses under the warehouse receipt financing and therefore has not had to call the Government guarantee. 72 Glob al Partn er ship for Financial I nc lu si on CASE 13  HDFC Bank, India — Warehouse Receipt Loans Facility Warehouse Receipt Finance HDFC was established in 1994 with a mission to be a world-class Indian bank. It has 2,201 branches and more than 5,000 ATMs in 1,174 Indian towns and cities. As of December 2011, the Bank had a balance sheet of USD 55 billion, of which approximately 14 percent finances agriculture. The warehouse receipt loans facility offers loans from Rs. 1 Lakh (USD 2,250) to farmers and small trad- ers against around 50 commodities to be stored in more than 3,500 approved private, central, and state warehouses. HDFC generally finances between 65–75 percent of the receipt value, and offers competi- tive interest rates of 8 to 10 percent. Farmers benefit from prompt loan disbursement upon delivery of the warehouse receipt. A top-up loan facility is available to existing asset relationship customers. Warehouse limitations include few inland operations and minimum lot requirements for stocking. Thus, the facility does not generally cater to small or rural farmers; it is larger farmers, traders, agro-proces- sors, and government departments that primarily occupy warehouse space. Groups of smaller farmers do participate when resources are pooled through a single representative farmer. A non-standardized grading system is identified as a key risk for banks. Operating profits range between 35–40 percent; risks and losses are low, around 1–2 percent. Losses have been managed by an efficient warehouse surveillance system set up by the Bank, consisting of well-trained warehouse inspectors and collateral managers, weekly mark to market valuations, and timely liquidation of stocks in case of defaults. In 2011, around 800 farmers participated, with market trends indicating increased participation. HDFC seems to have lower default risk and less distribution costs than with direct loans in this scheme. Eventually, with a more reliable warehouse system developing in India, banks will spend less on collateral management and supervision, thus increasing profit margins in this type of scheme. Innovati v e Agric u ltu ral S ME Finance Models 73 CASE 14  Ghanaian Financial Services, Ghana — Collateral Management Ghanaian Financial Services Ghanaian banks have a high level of confidence in local collateral management (CM) service providers, thus it is fairly easy for clients to access financing from the banks. The cost of collateral management ser- vices in and around the ports varies from USD 1,500–$2,300 per site per month; this includes security, but charges for insurance, tallying of goods, and fumigation of the warehouse are additional. The interest rates and charges may not be below commercial rates charged by banks for other kinds of working capital lend- ing. In an environment where banks otherwise insist on land and buildings as collateral, many clients’ only access to financing lies in using the services of collateral managers. Contractual disputes do not normally arise with collateral management agreements CMA in Ghana. Lending banks carry out regular monitoring involving periodic unannounced visits, and in some few cases (especially for high value goods) provide additional padlocks and require the collateral manager to send daily, rather than weekly, stock reports to the banks electronically. There has been no high profile fraud in CM in Ghana, which seems to reflect the professionalism of the ser- vice providers. Keys to the program include: strict inspection of the warehouse before leasing it and sign- ing the CMA; proper stacking inside the warehouse; supervised tallying in and out of the warehouse; appropriate and fair wage/salary structure for warehouse staff, and; sound knowledge of the commodities to be stored. The cost of the insurance is sometimes higher than for the CM service. For example, a ship- load of imported commodities worth $6–8 million might need to be warehoused for 2 months; an all-risks policy meeting the requirements of an international bank might cost around $3,000 per month. Clients seeking the service of CMs for smaller shipments often object to the high cost of the insurance cover demanded by banks, and this often leads to a process of negotiation mediated by the collateral manager. 74 Glob al Partn er ship for Financial I nc lu si on CASE 15  Dunavant Zambia Ltd. and Cargill Zambia Ltd. — Farmer Input Credit Tight Value Chain Finance Dunavant Zambia Ltd. is the largest cotton company in Zambia, with 100,000 contract farmers and a 60 percent market share. Cargill, which purchased Clark Cotton in 2006, has around 1,000 employees in Zambia. Together, both companies process around 90 percent of the country’s cotton. Dunavant and Cargill finance contract farmers through a structured loan package that provides inputs on credit. The growers participating in the scheme have no assets for collateral because land is commu- nal and held in a trust by a chief. To participate in the scheme, a grower must have at least 0.5 hectare of land. The input loan package includes: planting seed, which is disbursed at the beginning of the season; insecticide, which is disbursed after verification by field staff that the seed has been planted; fer- tilizer, which is provided at the same time as the insecticide; plastic knapsack sprayer for application of the pesticide for farmers or groups of farmers with 1 hectare of land; and wool bags for storage. The total value of the package without a sprayer is approximately 250,000 Zk (USD 47) per hectare and with a sprayer 520,000 Zk (USD 98) per hectare. The inputs are high quality, standardized products that would not be available to the farmer without the program. As such, more than 99 percent of Dunavant’s con- 42 tracted farmers participate. After harvest, farmers move the cotton by hired oxcart to one of the 1,440 buying points where they receive cash on delivery. The final payment received by the farmers at time of delivery is the net of the costs of the input package received. Although contracts are entered between the company and the growers, the system relies on trust and strong mutual commercial incentives, as contracts are generally not enforceable. Participating growers receive an identity card that establishes an account number, and the transaction is carefully tracked through a complex, paper-based monitoring system at the company’s main office in Chipata. In order to ensure the expected quality of production and promote grower loyalty, Dunavant and Cargill make training an essential component of the program. Training covers issues from proper pesticide application to care and maintenance of sprayers, and is sup- ported by an expansive network of permanent field staff. 42 Agrifood Consulting International (2005) Innovati v e Agric u ltu ral S ME Finance Models 75 Dunavant Zambia worked with more than 100,000 farmers in 2011, up from 70,000 contract farmers in 2009/2010. In 2007, Dunavant lent more than USD 10 million to farmers. Similarly, Cargill Zambia worked 43 with 65,000 farmers in 2011. Since divestment of the parastatal Lint Company of Zambia in 1994, Dunavant’s annual sales volumes have increased by approximately 30,000 metric tons. Yields have increased from 600 kilograms per hectare to an average of 1,200 kilograms per hectare, sometimes reaching 2,400 kilograms per hectare. The target for repayment of loans is 87.5 percent, but actual rates have been 94 percent in 2000, 95.35 percent in 2001, 95.5 percent in 2002, and 97.02 percent in 2003. In 2012, a database of all contract farmers was organized to prevent double contract farming and further reduce risk. The key to the successful growth and sustainability of the program is the enduring relation- ship between growers and the company — led by mutually beneficial commercial incentives, consistent input credit and extension/education services, careful farmer selection, strict controls on the quality and variety of seed, prompt payment systems with account monitoring for all contract farmers, and even HIV/AIDS workplace and family outreach programs. Primary challenges to sustainability include sharp drops in global cotton prices and opportunistic traders that source from and undermine the established supply chain. The business model used by Dunavant and Cargill is replicable for those physical traders and processers who see value in downward integration of a wide range of commercial intermediary functions. Financing, when it can be done with minimal risk, is therefore an important enhancement to 44 more traditional trading and manufacturing roles. 43 Peltzer (2011) 44 World Bank (2005) 76 Glob al Partn er ship for Financial I nc lu si on CASE 16  Palabana Dairy Cooperative Society & Parmalat, Zambia — Value Chain Finance Zanaco (Zambia National Commercial Bank Plc) is one of the leading banks in Zambia in terms of cus- tomer deposits, total assets, and points of representation. It was partially privatized in April 2007 when Rabobank acquired 49 percent of GRZ’s interest in the Bank. Zanaco was listed on the Lusaka Stock Exchange in 2008 and is considered “Citizen Owned” with over 50 percent of the shares owned by Zambians and the Zambia National Farmers’ Union (ZNFU). Palabana Dairy Cooperative Society was established in 1996. The cooperative has its own milk storage depot with a storage capacity of up to 3,000 liters of milk per day. The milk is collected directly by Parmalat, the off-taker and milk processor, from the Milk Collection Centre. Land O’ Lakes (the donor/consultancy arm of the largest U.S. dairy cooperative) provided initial capital by financing 22 cows. The quality-based payment system by Parmalat incentivizes the farmers to optimize quality; 100 percent of the milk is grade A. The members pay ZMK 100/liter commission to the cooperative (4 percent of the liter price) to pay for the cooperative’s workers and overhead. In 2006, Zanaco provided a USD 12,000 loan to the cooperative to finance 20 cows (Jersey and Friesian); a year later the loan was increased to USD 36,000 to finance another 30 cows. Both loans were repaid on time. In 2011, the Cooperative obtained a third loan of USD 120,000 to purchase 65 cows, repayable in 2015. Parmalat has signed a 5-year off-take guarantee with the cooperative and pays directly into the coopera- tive’s account with Zanaco on a monthly basis. Through this tri-partite agreement, Zanaco is able to underwrite predictable cash flow and collect repayment through deductions at the income source. Due to high repayment levels, the program has seen consistent expansion and participation: 50 farmers 45 participated in 2009/10, 120 in 2010/11, 200 in 2011/12, and 300 are anticipated in 2012/13. After repay- ment of the current loan, the cooperative would like to again double the loan to over USD 200,000 to finance new cows. In addition, the current milk tank capacity of 3,000 liters has to be increased, as daily production currently amounts to 2,000 liters. It is important to note that these loans benefited from a 46 larger Land O’Lakes development project financed by USAID. An IFPRI survey shows that the Land O’Lakes project achieved significant improvements in household income, food security, and dietary and 47 livelihood diversity for approximately 22,000 beneficiaries. Key risks of the project are price volatility and dependence on one large buyer: the cooperative sells milk to Parmalat, and Parmalat dictates prices. This risk is mitigated by the emergence of small milk processing companies like Nice Products and Kaposhi that may create competition for Parmalat. Another risk is weather: in the dry season, most small- scale farmers who depend on natural grazing are affected by a lack of grazing grass. This risk can be mitigated by supplementing with hay and molasses, and by acquiring multi-peril insurance covering drought, floods, and fire. 45 de Vries (2012) 46 Swanson (2009) 47 Hawkes and Ruel (2011); Christen and Pearce (2005) Innovati v e Agric u ltu ral S ME Finance Models 77 CASE 17  ECOM, Africa & Asia — Capital Improvement Loan Facility Value Chain Finance with Input Suppliers ECOM Agro-industrial Corporation Limited (“ECOM” or “the Company”) is a leading supply chain man- ager and integrated supplier of both raw and semi-processed agricultural commodities. ECOM, incorpo- rated in Switzerland, and its subsidiaries are commodity operators covering 30 countries in the United States, Central and South America, Europe, Asia, and Africa. ECOM’s principal activities consist of trading coffee, cotton, cocoa and, to a lesser extent, grain and other agricultural products, along with raising pork. The proposed IFC Asia-Africa Facility is aimed at enabling ECOM to initiate a program of medium-term funding to its coffee growers so that these farmers can undertake capital improvement projects that require longer-term funding than the short-term crop advances currently provided by ECOM and/or other traders. ECOM is willing to take the risk of financing the farmers with medium-term funding on its balance sheet. The project will consist of i) medium-term funding to be provided by ECOM subsidiaries in South and Central America to coffee farmers to fund capital improvements for ECOM producers in Kenya, Uganda, Tanzania, PNG, and Vietnam; and, ii) short-term crop advances. One reason for the additional working capital needs is the expansion of ECOM’s certification efforts and technical assistance from its original countries in Central and South America to producers in Africa and Asia. As a first step toward certification, ECOM’s field technicians do a baseline assessment of the farms, using the methodology of the relevant certification program, which includes assessment of product quality and economic, social, and environmental sustainability of farm practices. This baseline assessment identifies the aspects of the farm operation that need improvement in order for the farm to be eligible for certification. This initial assessment is audited by the verifier of the relevant certifica- tion program on a sample basis. Under ECOM’s guidance, those farmers that meet minimum sustain- ability and quality criteria implement mandatory improvements that mostly require short-term actions (up to 6 months) such as more documentation of the process, labeling, management of fertilizers and chemicals, minimum wage, no child labor, labor conditions, and worker housing improvement. More fundamental problems (soil conservation, watercourse and biodiversity protection) require medium- term (6 to 9 months) implementation periods. Once certified, the farmer must demonstrate continuous improvement in farm practices in order to remain in the program. This value-added coffee is then mar- keted under brand names created and owned by ECOM. While farmers are free to sell their coffee to any trader, ECOM is willing to underwrite the capital improvement loan facility for the following rea- sons: i) ECOM’s technical assistance and certification efforts creates loyalty with farmers; and, ii) the certification process facilitates a tight value chain in which farmers are financially motivated to accept certified coffee premiums from ECOM. ECOM’s on-going South American model works with 125,000 farmers, and provides USD 17.4 million in seasonal financing to over 14,000 farmers. ECOM’s technical assistance has enabled more than 10,000 48 farmers to become certified. 48 Wegner (2012) 78 Glob al Partn er ship for Financial I nc lu si on CASE 18  Ghana Grains Partnership, Ghana — Value Chain Finance The Ghana Grains Partnership involves a number of partners and sponsors. First is an international and national consortium of private sector sponsors: Yara International ASA, the world’s largest fertilizer com- pany, and Wienco (Ghana) Limited, a specialist in the importation and distribution of high quality agro- inputs. Wienco has developed some of the leading Ghanaian commodity associations, including the Cocoa Abrabopa Association (CAA), Wienco Fibres Ltd., and the Integrated Tamale Fruit Company. The Africa Enterprise Challenge Fund (AECF), farmers and farmer associations initially in Wienco’s out- grower scheme in the Northern Region, the Ministry of Food and Agriculture, the sector policy maker and regulator, Standard Bank and other commercial banks, and Technoserve, an NGO, also participate, along with output buyers (including processors) and traders. Prorustica, together with its local partner MCM Associates, provides advisory services to the Partnership. Yara initiated the Ghana Grains Partnership (GGP) in 2008, inviting a bottom-up dialogue with local farmers and developing a large-scale rollout model. Under this scheme, farmers form joint liability groups of 5 and 10 members for block farming, with each farmer cultivating an estimated 5 acres. Individual farmers with capacity, credibility, and commitment are also accepted with individual plot requirements of 5–10 acres. The pilot project rollout started in 2009 with the establishment of the growers’ associa- tion, Masara N’Arziki. Off-take contracts between the Association and maize farmers are entered, and hybrid seeds, chemicals, and fertilizers are provided on credit. In the absence of initial external financing for the Association, Yara and Wienco, the project’s input dealers, financed the first requirements of USD 1 million. The Association purchases the farmers’ total maize crop and compensates them for their product minus the cost of borrowed inputs. The Association is able to offer guaranteed prices as Wienco, the Association’s buyer, can guarantee a minimum price to the Association. Every participating farmer receives not only inputs on credit, but also benefits from the Good Agricultural Practices training and extension. Technoserve also provides governance and business management services. The program promotes conservation farming and no tilling to reduce moisture loss. Furthermore, crop rotation is introduced, particularly with crops such as soy beans that help improve soil fertility. This financing scheme manages side selling through a well-planned, integrative partnership that: (i) promotes account- ability through joint liability groups that have co-ownership in processing and profits down the value chain; (ii) develops farmer ownership in the planning stages; (iii) builds farmer-partner relationships and establishes trust through extensions services; and, (iv) selects farmers with exemplary commitment and capability. Currently, some 5,000 maize farmers have been reached through the program, which will be extended to 5,000 rice farmers. Production of the maize farmers participating in this program has jumped by as much as five times through fertilizer application optimization, the extension services, and the block farm- ing. Due to the program’s success, Standard Bank financed USD 8 million in 2010, through its Agra Guarantee Scheme, to Masara N’Arziki for program expansion. For the growing season in 2011, the Agro Development Bank refinanced Standard Bank. In the next 5–6 years, the target number of farmers under this program is 25,000 producing 250,000 metric tons of maize — exactly the volume that was imported by Ghana in 2011. Innovati v e Agric u ltu ral S ME Finance Models 79 CASE 19  CRDB, NMB, Kilombero Sugar & Mtibwa Sugar, Tanzania — Outgrower Finance Outgrower Scheme with Sugar Associations CRDB Bank Plc, is a leading, wholly-owned private commercial bank in Tanzania. NMB (National Microfinance Bank) has become Tanzania’s largest financial services provider, with a growing customer base of more than 1.4 million people. They, along with Kilombero Sugar Company Limited (KSCL) (owned by Illovo 55 percent, ED&F Man 20 percent, Government of Tanzania 25 percent) and Mtibwa Sugar Estates Limited (MSE), which are sugar cane outgrowers, and their trusts organized under four Outgrower Associations. CRDB’s and NMB’s outgrower loan scheme provides new opportunities to a Tanzanian outgrower model that dates back to the 1960s. The Kilombero/Mtibwa sugar estates are now vertically integrated and pro- cess outgrowers’ sugar supplies. Smallholder sugarcane farmers around Kilombero and Mtibwa estates currently supply about 50 percent of the total sugar cane used by the mills; 23,300 hectares are under cultivation by over 20,000 outgrower farmers. The loan scheme extends credit to sugar outgrowers for purchase of inputs for crop maintenance and other costs related to the development of sugar cane. Eligible growers are covered by a tri-partite agreement between the sugar estates (the buyer), the out- grower (borrower), and NMB/CRDB. The agreement ensures that all crops will be purchased by KSCL/ MSE, growers have experience in growing the crop, crops were harvested in the previous season with sufficient proceeds, the growers trust provides collateral to secure the loan, and there is a farming con- 49 tract between the sugar companies and the farmers. NMB and CRDB started financing outgrowers on the basis of guarantees or at least “comfort letters” from Mtibwa. Subsequently, the sugar company obtained a bulk loan from CRDB, which it retailed to farmers through savings and credit cooperatives (SACCOs). Mtibwa transfers all payments for sugar deliveries to SACCOs first, which then deducts the loan repayment from that amount and remits the net proceeds to the farmer. Strong working relationships persist between the sugar outgrowers and the estates. As revenues are shared between the estates and outgrowers — 55 percent to the Kilombero area farmers and 53 percent to the Mtibwa area farmers — outgrowers take ownership in the value chain and typically refrain from side selling. Recognizing that productive outgrowers are essential to their own profitability, KSCL has invested heavily in outgrower infrastructure, technical and business skills, local organization, input access, and com- munity development projects. In their outgrower loan scheme, NMB and CRDB are able to underwrite anticipated cash flows from strong, local relationships with established, secure, and long-term buyers. NMB has provided financing to thousands of outgrowers in Tanzania with loan amounts of around USD 4 million. Illovo’s outgrower rain-fed cane production produced an improved crop of 493,000 tons har- 50 vested in 2010/11. The Kilombero Cane Growers Association (KCGA) has 4,200 members with 5,836 hectares of sugarcane — allowing them to produce about 320,000 tons of sugar cane per season. 49 NMB (2010) 50 Illovo (2011) 80 Glob al Partn er ship for Financial I nc lu si on CASE 20  NMB, Tanzania — Agro-Dealer Scheme Value Chain Finance with Input Suppliers NMB (National Microfinance Bank) has become Tanzania’s largest financial services provider, with a growing customer base of more than 1.4 million people. Rabobank acquired a 35 percent stake in the NMB in 2005, when the bank was partially privatized by the Tanzanian government. NMB’s agro-dealer product is a credit facility for traders of agricultural inputs, which allows them to borrow working capital up to Tsh. 30 million (USD 19,000) in their NMB business account via a pre- 51 defined overdraft line. The product is delivered through facilities at NMB branches. These facilities ben- efit from a Risk Sharing facility from AGRA and the Financial Sector Deepening Trust that results in a 10 percent loss guarantee to NMB. NMB has agreed to provide some USD 6 million worth of financing to these Agro Dealers, and started in May 2008 in 11 districts. This product makes it easier for agro-dealers (and input suppliers such as Yara for which they retail) to increase sales of fertilizer and other inputs in Tanzania. The Ministry of Agriculture is now planning to roll it out to more districts, which will bring the financial need up to some USD 15.4million. NMB is in discussion with AGRA/FSDT to top-up the guarantee in order to continue supporting this Government initiative. The program helps farmers by making more inputs available during the start of the season and resulted in more stable prices of fertilizer among competing agro dealers. NMB approved facilities to over 148 agro dealers in all 11 eligible districts. 51 NMB (2012) Innovati v e Agric u ltu ral S ME Finance Models 81 CASE 21  Bayer, Raiffeisen Aval Bank, Ukraine — SME Farmer Input Credit Value Chain Finance with Input Suppliers Raiffeisen Bank Aval (RBA), established in 1992, is a wholly owned subsidiary of Raiffeisen Bank International, Austria. RBA is the fourth largest bank in the Ukraine and has a 30 percent share of the agri- lending sector. RBA has 930 outlets in the country, of which 280 are dedicated to agricultural clients. Bayer Ltd. is a Ukrainian entity and the main supplier of plant protection products in Ukraine, with reve- nues of USD138 million (2010). Bayer Ltd. is wholly owned by Bayer AG, a Germany-based pharmaceuti- cals, polymers, and agrochemicals conglomerate. Bayer CropScience manufactures herbicides, insecticides, fungicides, seed treatment, and seeds. Raiffeisen Aval provides a guarantee to farmers, allowing them to buy inputs on credit from Bayer. Raiffeisen Aval is comfortable taking on small farmer risk thanks to a risk sharing facility (“RSF”) with IFC on a portfolio of receivables generated by Bayer in connection with sales of crop-protection products to private sector farmers in Ukraine. The RSF will cover a portfolio of seasonal receivables (with maturities less than 1 year) in which IFC will share 50 percent of credit losses in the portfolio in the local currency. In addition, IFC enters into a first-loss compensation agreement whereby Bayer guarantees the first 10 percent of IFC’s losses. Bayer is willing to take the first loss risk, as it has both a business incentive to increase sales of inputs and purchase history information on the farmers to gauge their creditworthiness. The combination of the RSF, the first loss, and the Know Your Customer (KYC) link via Bayer together support RBA’s entry into financing smaller farmers than they would otherwise accept. In 2011 Bayer Ltd. reached about 750 farms, most of which are medium-sized farms by Ukrainian stan- dards, with an average of 4,000 hectares under operation. A task force of 24 company agronomists visits the farmers on a regular basis, advising on the use of Bayer products. IFC expects the facility to finance Bayer crop protection products for 27,750 farmers by the end of 2014. 82 Glob al Partn er ship for Financial I nc lu si on CASE 22  ITC & State Bank of India — Smallholder Input Finance Value Chain Finance with Input Suppliers ITC Limited is a leading private sector company in India, with a market capitalization of more than USD 33 billion and a turnover of USD 7 billion. ITC is a market leader in cigarettes, hotels, packaging, paper- board, and agricultural exports. State Bank of India (SBI) is the country’s largest and oldest bank, with 13,500 branches. The Reserve Bank of India (Central Bank) owns 60 percent of SBI. SBI partnered with ITC to make affordable loans available to farmers for input purchases. Under the arrangement, ITC facilitates all documentation and verification procedures, thereby reducing associated costs to the bank and allowing the bank to offer more favorable loan terms to more farmers. ITC also allows SBI to effectively manage and monitor credit risk through the local knowledge and support of 52 platform operators and ITC data on farmer transactions. ITC Limited was one of the first Indian compa- nies to enter into large-scale, direct procurement arrangements with smallholder farmers. Today, the company has the established capacity to source produce from more than 4 million farmers across India via an extensive network of 6,500 rural community platforms known as e-Choupals. Led by a host farmer, each e-Choupal is equipped with a computer and Internet connection that facilitates dissemina- tion of local and global price trends and provides direct connection to ITC services. Farmers may sell produce to ITC, order agricultural inputs, or receive valuable weather forecasting and market informa- tion, all while eliminating the inefficiencies of middle traders. ITC benefits from reduced procurement costs as farmers realize higher farm gate prices. ITC has the capacity to engage more than 50,000 vil- lages through their e-Choupal platform. For providing this service, the company receives a nominal com- mission at loan disbursement to help defray the administrative costs that it incurs. Since the program was launched in 2008, ITC has helped to facilitate nearly USD 65 million in credit to 53 more than 70,000 of its suppliers. 52 Annamalai and Rao (2003) 53 Technoserve-IFAD (2011) Innovati v e Agric u ltu ral S ME Finance Models 83 CASE 23  Centenary Bank & Technoserve, Uganda — Factoring Centenary Bank is the fourth largest commercial bank in Uganda, with approximately UGX 807 billion (USD 340 million) in assets, representing 7 percent of the country’s banking sector. As the largest indig- enous bank in Uganda, Centenary Bank services a largely rural clientele through its 40 bank branches in central, western, northern, and eastern Uganda. Technoserve is an internationally recognized leader in the field of economic development. With 900 employees, the NGO has affected millions of lives through 54 its activity in 40 countries. Centenary Bank offers financing at two levels: 1) factoring for brokers against their sales to universities and hospitals; and 2) microloans to individual farmers. TechnoServe began working with matoke (green banana) farmers in Uganda in 2005 to establish value chains linking them to urban institutions such as universities and hospitals. At the time, overproduction had driven prices so low that desperate farmers were on the verge of destroying their matoke plants and starting over with new crops. Several inter- related issues compounded the problem. Disaggregated farmers sold their matoke to brokers on a vola- tile open market. Unable to rely on a steady source of matoke, brokers had to rent trucks for three to four days to scope the countryside for sufficient product to fulfill orders. There were typically five or six inter- mediaries between the farmers and the end buyers, resulting in slim margins for farmers and brokers alike. Brokers that sold to urban institutions received good prices, but often had to wait several months for payment until students paid their semester bills or hospitals received their government allocations. As a result, many brokers were forced out of business as they lacked working capital to maintain high turnover. To address these issues and align incentives along the value chain, TechnoServe began setting up village-based groups of 30–50 farmers. They also identified brokers that were entrepreneurial and honest, and personally introduced them to staff who made purchasing decisions at urban universities and hospitals. Brokers could then buy at scale and minimize price impacts of middle traders, all while securing higher farm gate prices for matoke farmers. Most brokers, however, could still not afford to meet urban demand of three deliveries per week (each valued at USD 3,000–4000) due to the prohibi- tive payment schedules. Centenary Bank recognized both the creditworthiness of the large buyers and the impact factoring would have on the efficiency of the value chain. 54 Technoserve-IFAD (2011) 84 Glob al Partn er ship for Financial I nc lu si on Farmers responded to the profitability of matoke cultivation by demanding finance to increase production. Though initially hesitant to lend to smallholder farmers, Centenary Bank agreed to a 3-year risk-sharing program backed by a $500,000 credit guarantee from the Rockefeller Foundation to cover up to 50 percent of losses from loans within the matoke value chain. Although Centenary Bank applied traditional require- ments for fixed asset collateral in its microloans to farmers, the Bank would not have entered into the facility without an established and efficient value chain where purchase commitments from brokers and end buyers mitigate the risks of lending to smallholder farmers. Centenary Bank’s experience in purchas- ing brokers’ accounts receivable highlighted the profitability in the matoke value chain and provided confidence in lending against farmer cash flows. Within a year, farmer incomes had increased 70–100 percent, and by 2008, 12,000 matoke farmers were participating in the system via 300 village-based farmers’ groups. By the time the facility closed in 2008, Centenary Bank had lent out a total of $1.6 million and claimed losses of less than $21,000, far lower than the loss rate on its lending book overall. This example illustrates that lenders should be willing to shift their approach to risk assessment and away from collateral. Centenary Bank’s experience in factoring provided a degree of familiarity and comfort with buyers and borrowers. Understanding the value chain organization, timing of cash flows, and financing needs was critical to establishing both the factoring lines of credit and the individual smallholder farmer financing. Innovati v e Agric u ltu ral S ME Finance Models 85 CASE 24  Kenya Gatsby Trust — Factoring Kenya Gatsby Trust (KGT) 55 is a Nairobi-based nonprofit organization that aims to eradicate poverty and spur economic development by supporting micro and small enterprises (MSEs) in Kenya. The facility pays participating MSEs cash against delivery of product to customers in good standing whose payment terms would otherwise overextend the seller’s working capital. In 2002, KGT’s Financial Services Department established a factoring program to bridge the gap between commercial banks and microfinance. This facility enables MSEs that were previously selling their products to brokers for cash, to cut out these intermediaries and sell product directly into formal markets that pay higher prices but on 30, 60, or 90-day terms. MSEs register with KGT and pay a fee in order to utilize the factor- ing service. When an MSE delivers its product to its buyers, KGT immediately pays the seller 70–95 per- cent of the invoice value. After collecting payment from the buyer on the pre-agreed terms, KGT remits the remaining 5–30 percent to the seller. This system smoothes cash flow for small and growing busi- nesses and removes the uncertainty of having to collect accounts receivable from larger private compa- nies or institutions. In doing so, it enables MSEs to source from smallholder farmers, who typically require cash payment on delivery, without overextending their working capital. The arrangement also has ben- efits to buyers such as hospital and universities, many of which would like to support local MSEs and the smallholder farmers that supply them, but are unable to tie up their own working capital with advance payments or cash upon delivery. The program currently serves 25 MSEs sourcing from over 4,000 small-scale farmers and artisans. One challenge to growing the program has been the reluctance of bureaucratic government institutions to change their established practices and make payment to an entity other than the seller. As an NGO, it is important to note that KGT is not taking any interest or profit on the transactions, and is not protected in the event that the receivable is not paid in full. A similar model could be established to charge com- mercial rates on the financed amount that is eventually recovered as well as to apply appropriate factor- ing discount rates to cover any loss of face value of the A/R. 55 Milder (2008) 86 Glob al Partn er ship for Financial I nc lu si on CASE 25  Root Capital, Latin America & Africa — Export Trade Finance Root Capital is a nonprofit social investment fund that is pioneering finance for grassroots businesses in rural areas of developing countries. Root Capital makes loans in the range of $25,000–$1,000,000, targeting enterprises with environmen- tally sustainable practices that are exporting high-value products in the following sectors: agriculture, timber and non-timber forest products, fisheries, and handcrafts. The vast majority of these businesses had never received a loan prior to working with Root Capital. The most common type of loan is trade credit, which is available for up to 1 year and oriented around a production cycle such as a harvest. Trade credit loans are typically used by borrowers to purchase product from their farmer and artisan members or suppliers and to cover costs during the months between purchasing raw product and receiving pay- ment from buyers. Root Capital also offers long-term loans that extend up to 5 years and are used for investment in equipment and infrastructure and for general operations. Interest rates range from 9–10 percent per annum for loans up to one year and 10–12 percent per annum for long-term loans. All loans have a closing fee of up to 1 percent. To mitigate risk, Root Capital has developed a model that assesses collateral based on producers’ future sales rather than their existing assets. Under this three-way arrange- ment, Root Capital lends against signed purchase agreements between grassroots businesses and their buyers. Typically, the borrower is eligible for a loan of up to 60 percent of the value of the export con- tracts. The purchase agreement, in effect, becomes the collateral — a discrete, future revenue stream Typical Root Capital Financing Structure pledged by the borrower to repay Root Capital’s loan. When the bor- rower ships product to the buyer, 4 Pay for goods Root Buyer the buyer makes payment directly Capital to Root Capital, which, in turn, deducts the loan principal and  ake loan with 2M purchase order interest and remits the difference  hip 3S as collateral goods to the borrower. Root Capital’s due 1O  rder  emit payment 5R diligence and monitoring processes net of loan goods principle and are designed to identify any chal- interest lenges that might derail this trans- action, such as weather issues, a Borrower strike at port that prevents product from shipping, or the buyer going out of business. Innovati v e Agric u ltu ral S ME Finance Models 87 From 2000 to July 2008, Root Capital disbursed 506 loans totaling $100 million to 210 grassroots busi- nesses representing more than 340,000 small-scale producers in 30 countries across Latin America, Africa, and Asia. The repayment rate on Root Capital’s loans is over 99 percent, yet most of its clients continue to have few, if any, alternatives for affordable credit. Root Capital has applied this value chain finance model with 125 U.S. buyers, ranging from specialty importers, such as Equal Exchange and Sustainable Harvest, to large global buyers, including General Mills, Green Mountain Coffee Roasters, Pier 1 Imports, Starbucks Coffee Company, The Body Shop, The Home Depot, and Whole Foods Market. Ninety percent of Root Capital’s portfolio is made up of short-term (typically 5 to 9 months) trade credit loans to address the cash flow gap between the time an SME purchases raw goods from its farmer sup- pliers and when the business receives payment from its buyers several months later. Pre-harvest loans to SMEs to lend to individual farmers to purchase seeds and other inputs represent a growing portion of 56 Root Capital’s portfolio. 56 Milder (2008) 88 Glob al Partn er ship for Financial I nc lu si on CASE 26  Microensure & Kilimanjaro Native Coffee Union, Tanzania — Health Insurance Personal Insurance The Kilimanjaro Native Coffee Union (KNCU) is Africa’s oldest coffee cooperative union, comprising around 70,000 members in 92 cooperatives (“primary societies”) in the Kilimanjaro area of North Tanzania. PharmAccess Foundation, along with MicroEnsure, organized the health plan. Active in 31 African countries, PharmAccess Foundation is a Dutch not-for-profit organization that strengthens health systems in sub-Saharan Africa by facilitating participation of the economically disadvantaged in health insurance schemes. MicroEnsure, an Opportunity International affiliate, is a specialized micro- insurance broker that covers more than 3 million lives with micro-insurance products. The health plan serves those primary societies of KNCU where a majority of the members have voted for participation in the health plan. Access to primary healthcare is one of the major challenges faced by communities in rural areas of Tanzania. Most facilities lack the staff, equipment, or services required to provide an adequate standard of treatment. In order to provide an appropriate and sustainable level of healthcare, PharmAccess has pioneered comprehensive infrastructure improvement, which includes the redevelopment of facilities, the training of medical staff, and sufficient access to medication. PharmAccess funds a number of specialists to attend clinics and provide training to local medical staff. This allows an increasing number of conditions to be treated at a primary care facility — reducing health care delivery costs and increasing the level of care provided by the health plan. The KNCU Health Plan is a “per capita” scheme in which each healthcare provider is paid in advance for a projected number of patients. MicroEnsure’s role is to broker the insurance and reinsurance components, manage the capitation pro- gram, empanel the providers and ensure high quality care from them, and support sensitization and mar- keting efforts by leading member enrollment at the village/family level, utilizing iPod handheld programs. Payments to healthcare providers are monitored by analyzing claims data on an internal system to ensure providers have sufficient funds to meet patient needs and are following treatment protocols. Even in the pilot stage, treatment protocol improvements have occurred for the main conditions of malaria and acute respiratory infections, among others. Since its launch in April 2011, the KNCU plan has been rolled out to 5,000 people in three cooperatives. Eighteen more societies and 20,000 members are targeted for enrollment by the end of 2012. In 2011, members paid an annual premium of Tsh. 12,000 (USD 9) per person, per year, while a further payment of Tsh. 18,000 (USD 14) was funded by PharmAccess. It is expected that the KNCU Health Plan will cover over 250,000 lives by the end of 2013, moving the program to long-term sustainability. Farmer-level income impacts are not yet fully understood, but there is strong potential as the KNCU Health Plan ensures farmers and their family’s access to services earlier and at better quality. While the health plan should be distinguished from credit health insurance, and is thus not bundled with a loan, health insur- ance contributes to the lender’s ability to underwrite anticipated cash flow through a farmer-focused approach. Banks may lend with confidence, knowing that healthy farmers are improved investments, and that unexpected health expenses will not overburden farmers and their families, resulting in repay- ments that become more timely. Innovati v e Agric u ltu ral S ME Finance Models 89 CASE 27  Taytay Sa Kauswagan, Inc (TSKI) , Philippines — Index Insurance Production Risk Insurance for MFI Input Loans Taytay Sa Kauswagan, Inc (TSKI), a leading microfinance institution in the Philippines, has nearly 200,000 active borrowers. MicroEnsure, an Opportunity International subsidiary, is a specialized micro- insurance broker that covers more than 4 million lives with micro-insurance products (Malayan Insurance Co., Paris Re (reinsurer)). TSKI uses a typhoon index insurance product to protect smallholder rice farmers and their lenders from the financial risk of crop damage by typhoons. Farmers are to receive an automatic payout, triggered by satellite tracking of a typhoon’s path and wind speed. GPS coordinates of policyholders’ farms are recorded, and actual payouts are based on covered farms being within 140 km of the typhoon track with wind speeds in excess of 59 mph. Policyholders do not need to file a claim, the insurer does not need to perform a loss assessment, and payouts can be made within 10 days. The model was designed to reduce transaction costs and facilitate efficient payouts during a sensitive time for farmers and their families. Data are provided by the Japanese Meteorological Authority in real time and are freely available on the Internet. This product secures TSKI crop loans, thus supporting bank lending under the Philippines Agri Agra law. An education process targeting farmers included comic books in English and Tagalog. During the pilot project, the Philippines Insurance Commission approved weather index insurance products for the first time. Starting in May 2009, the insurance pilot covered 446 farmers on Panay Island, the Philippines, for one crop- ping period. While one typhoon entered the coverage diameter during this period, its wind speeds were below the payout level. After a countrywide assessment of the index, it was concluded that rainfall level, in addition to wind speed, is an integral factor in accurately capturing storm damage. In 2010, a new product based upon cumulative rainfall in specific intervals and consecutive wet and dry days was developed. Using local weather stations that were recently installed by the Philippine Government, TSKI/MicroEnsure Philippines launched the product in 2012 and anticipated covering 8,000 farmers by the end of year. 90 Glob al Partn er ship for Financial I nc lu si on CASE 28  Syngenta Foundation & UAP Insurance, Kenya — “Kilimo Salama” Index Insurance Input-linked Weather Index Insurance UAP Insurance is a leading insurance and financial services company in East Africa, with operations in Kenya, Uganda, and Southern Sudan. Syngenta Foundation for Sustainable Agriculture is a non-profit organiza- tion founded by Syngenta to focus on “pre-commercial” growers. The weather insurance product is branded Kilimo Salama, “safe farming” in Kiswahili, and is meant to be simple, affordable, and relevant to small farmers. Farmers purchase the product through local agro-dealers, who use a camera phone to scan a special bar code that sends the policy to UAP over Safaricom’s mobile data network. This mobile phone application then sends a text (SMS) message to the farmer’s mobile phone confirming the insurance policy. Kilimo Salama allows smallholders to insure selected farm inputs at their local retailer and pay only half the premium. Payouts are determined by data collected through 30 weather stations in the targeted regions that have been renovated with automated, solar-powered systems capable of broadcasting regular updates on weather conditions and rainfall quantities occurring near individual farms. When data from a particular station, which is transmitted over Safaricom’s 3G data network, indicates that drought or other extreme conditions (including excessive rains) have reduced yields, all farmers regis- tered with that station automatically receive payouts through Safaricom’s popular M-Pesa mobile money transfer service. Index-based payouts administered through M-Pesa substantially reduce transactions costs and ensure immediate payment. To make the insurance affordable, Kilimo Salama’s agribusiness partners pay the other half of the premium. In the 2009 pilot phase, partners included Syngenta East Africa Limited and the fertilizer company MEA. Their involvement enabled the scheme to get off the ground quickly, in time for the next growing season. In November 2010, the IFC-led Global Index Insurance Facility (GIIF) entered into an agreement to support the Syngenta Foundation to further develop the technology of the SMS-based mobile platform and assist 57 scaling up the product in the country. During the 2012 season, the enhanced product was to cover around 47,000 farmers. In 2009, the product pilot was tested by one of the worst droughts in recent history. Covering 200 maize farmers through two weather stations, the product offered payouts to all farmers totaling either 30 or 80 percent of their insured maize seeds. Syngenta had paid the entire premium. In the following season, 12,000 farmers were covered through 25 additional weather stations in five regions. Two-thirds of these clients adopted Kilimo Salama as part of a bundled microfinance package, and half of the 10 percent premium was paid by Syngenta. About 1,200 farmers received payouts ranging from 10–50 percent of their insured inputs. In February 2011, Syngenta launched Kilimo Salama Plus, expanding the insurable sum per farmer to the expected harvest value of a wider array of crops including maize, wheat, beans, potatoes, and sorghum. The product was also expanded to insure farmers growing under contract farm arrangements for agribusinesses — reaching farm- ers with as little as 1/4 an acre up to 1,000 acres. Kilimo Salama and Kilimo Salama Plus are available to farm- ers in the productive breadbasket regions of Southern Nyanza, covering Oyugis and Homa Bay, Busia, and 58 Northern Rift, including Kitale and Eldoret, as well as the semi-arid areas of Embu and Nanyuki. 57 The GIIF program is a World Bank Group initiative that seeks to address the scarcity of affordable insurance protection against weather and natural disaster risks in developing countries. 58 Syngenta Foundation for Sustainable Agriculture (2011) Innovati v e Agric u ltu ral S ME Finance Models 91 CASE 29  PepsiCo, ICICI Lombard & WRL, India — Index Insurance Weather Insurance for Contract Farming PepsiCo is a global leader in convenient snacks, foods, and beverages, with over 285,000 employees and revenue exceeding USD 60 billion. ICICI Lombard General Insurance Company is the largest private sector general insurance company in India with a gross written premium of Rs. 4,734.89 crore in 2011. It has over 4,000 employees at 315 branches across the country. Weather Risk Management Services Limited (WRL) is a private broker and weather station operator. To protect the farmers in its supply chain from weather events, PepsiCo offers index insurance as part of its contract farming program. The insurance is sold through the ICICI Lombard General Insurance Company, an international insurer, and managed by Weather Risk Management Services (WRMS). PepsiCo added index insurance to its contract farming package not only to limit farmers’ weather risk, but also to establish long-term relationships with farmers and limit the risk in its supply chain. In its contract farming arrange- ment, PepsiCo offers an extensive package of services: high quality potato seed; access to fertilizers, pes- ticides, and other chemicals; technical advice on production practices; fixed purchase price and incentives from the beginning of the season; weather information and advisories via mobile phone Short Message Service (SMS); and the weather index insurance. PepsiCo sets a base buy-back price for its contract farm- ers at the beginning of the season and offers incremental price incentives according to: (i) quality of the potatoes (+Rs 0.30/kg); (ii) use of fertilizers and pesticides (+Rs 0.25/kg); and (iii) purchase of index insur- ance (+Rs 0.15/kg). In PepsiCo’s experience, the main drivers that influence a farmer to purchase index insurance include: assured buy-back price from PepsiCo, ability to finance the premium and other produc- tion costs through a loan, trust in the various actors involved (e.g., corporation, processor, insurer, local representatives), demonstration of timely payouts in previous seasons, and perceived need to mitigate the risk of losing the significant upfront costs of production, in part to cover the production costs for the fol- lowing season. PepsiCo also encourages the purchase of index insurance through client education, as it finds index insurance simpler, more transparent, and faster to settle than conventional insurance. Overall, in the PepsiCo collaborative farming program, index insurance plays an important role in a wider package of services and information that links smallholders to markets. The index insurance option was initially offered in the Indian province of Punjab in 2008. Gradually, it was expanded to the provinces of Maharashtra and West Bengal. Among the 24,000 PepsiCo contract farm- ers across the nine state locations, around 50–60 percent elected to purchase index insurance — a high proportion that is driven in part by price incentives and conditions on state bank loans that require insur- ance. By 2013, the contract farming program is expected to reach 30,000 farmers. The program has pro- vided claim payouts in almost all state locations over the last 5 years, with farmer retention rates in excess of 90 percent. 92 Glob al Partn er ship for Financial I nc lu si on CASE 30  Bagsa Agricultural Commodity Exchange, Nicaragua — Commodity Price Risk Management Bagsa has 180 shareholders and 36 brokers, and its net worth exceeds USD 1.25 million, with operational volumes of around USD 600 million. In 2008, it had 30 active business and individual members, includ- ing stockbrokers and financial institutions, farmers (21 percent), independent investors related or unre- lated to agricultural activities, agro industry (19 percent), commerce, and agribusiness services. Bagsa facilitates bilateral cash and forward contracts, and holds a small number of auctions. It thereby provides SMEs access to contracts for their production and thus helps them deal with persistent market volatility while limiting depletive coping strategies. Bagsa works in a context of soaring domestic food inflation rates that have only slightly recovered from the 34 percent measure of 2008. The coffee crisis of 1998–2001 serves as a stark reminder that these global price shocks can have enormous effects on the well-being and productivity of small farmers and their families. Given Nicaragua’s strong export market and the increased trade liberalization of the region, SME farmers in the country are increasingly inte- grated with global markets and are poised to benefit from expanding risk management opportunities. From 80 shareholders and USD 80,000 in capital (1993) to 180 shareholders and a net worth of USD 1.25 million (2008), Bagsa has seen consistent growth in its member profile, net worth, and operational volume. Exchange commissions are low at 0.125 percent. It is estimated that Bagsa represents 40 per- cent of the total market participation. In recent years, the number of estimated transactions in Bagsa was around 50,000 per year, involving about 50,000 producers and 50 industry members and representing contract values of more than USD 500 million (2007–2008). Innovati v e Agric u ltu ral S ME Finance Models 93 CASE 31  M-Pesa and M-Kesho, Kenya — Mobile Banking Commercial Bank of Africa is one of East Africa’s largest privately owned banks, operating in Kenya and Tanzania. Equity Bank is the largest bank in the region, with 5.7 million accounts, over 57 percent of all bank accounts in Kenya, and operations in Uganda and Southern Sudan. Safaricom is Kenya’s largest mobile service provider, employing over 1,500 people with over 14 million customers. M-Pesa is a mobile phone-based service for sending and storing money offered by Safaricom. Safaricom customers can register for M-Pesa by visiting one of more than 27,000 merchants who act as “agents” for account opening, handling of deposits and withdrawals into the customer’s virtual “wallet,” and cus- tomer support. Customers can then use an application on their mobile phones to check their balance, send money to other people, pay bills, and purchase mobile phone airtime. Customer funds are held in a special trust account at the Commercial Bank of Africa. While a primary function of M-Pesa is low-cost money transfer, especially in the form of remittances to rural areas, the service is increasingly used to store value. Those with M-Pesa accounts are 32 percent more likely to report having some savings than those without accounts. Thus, M-Pesa encourages savings by providing a secure and widespread mech- anism to facilitate formal savings across socioeconomic boundaries. M-Pesa accounts may be used by lenders in tandem with other innovative financing tools to establish secondary collateral sources through an integrated and farmer focused approach. M-Kesho is a bank-integrated mobile savings product that was released in 2010 through a partnership with Safaricom and Equity Bank. M-Kesho combines the advantages of M-Pesa with those of banking services beyond simple money storage and transfer, including the ability to earn interest and secure small loans remotely through the extensive agent network. While M-Kesho is a promising opportunity to extend high-value banking services to the rural poor, it has met with limited popularity to date. Over 14 million customers have registered with the M-Pesa service. Since its commercial launch in March 2007, M-Pesa has achieved substantial scale along several key metrics. After 2 years of operation, 40 percent of the adult population had registered. To date, over 70 percent are registered. An average of 150 million Ksh (USD 1.96 million) is transferred through M-Pesa each day, mostly in small amounts averaging just over 1,500 Ksh (USD 20) per transaction. So far, the system has handled over 130 billion Ksh (USD 1.7 billion). 94 Glob al Partn er ship for Financial I nc lu si on CASE 32  Banque Populaire du Rwanda — Mobile Banking Banque Populaire du Rwanda (BPR) S.A. has 1.3 million customers (out of a total population of 12 mil- lion), 1,400 employees and 189 locations in Rwanda. Rabobank has a share of 35 percent in BPR. BPR Mobile banking offers balance inquiry, mini-statements, money transfers between BPR accounts, pre-paid airtime, bill payments, electricity, newspaper subscriptions, and a help function. The service was launched in September 2010 and is based on USSD II mobile banking and SMS alerts. The service uses MTN and Tigo as carriers. Customers can sign up at any connected branch and receive a SMS with set-up instructions. A survey of 170 customers revealed that users would trust the mobile phone for transactions. Thus in the next phase, BPR plans to add a “modern savings account,” which will be a trans- 59 actional savings account that people can sign up for in their village or on their farm. 59 Armstrong (2011) Innovati v e Agric u ltu ral S ME Finance Models 95 CASE 33  Refresh Mobile WING, Cambodia — Mobile Banking Refresh Mobile provides electronic payment and cell phone top-up services. With its recent purchase of WING from ANZ, it is Cambodia’s leading mobile banking service provider. WING is a mobile-phone-enabled payment service that allows customers to transfer, deposit, and with- draw money via any mobile phone in Cambodia, at low cost. With a WING account, customers can cash in and cash out their accounts at any of the 850 WING CashXpress outlets, and cash out from all ANZ-Royal ATMs using the accompanying WING ATM card. Transactions such as sending and receiving money to WING and non-WING users, phone top-up, and bill payments are done from any mobile phone, and are secured by a 4-digit pin code. All active mobile phone operators in the Cambodian market are now connected to WIING, providing full geographical coverage. There is no monthly fee charged for holding an 60 m-wallet with WING, and funds are stored in a regulated bank. In line with the company’s commitment to provide banking services to the rural poor, 64 percent of WING subscribers have household incomes below USD 5,000, and 48 percent live outside the capital. With outlets in all 24 provinces, WING provides mobile banking services to approximately 250,000 sub- scribers. Since its commercial launch in 2009, the WING platform has grown consistently with around 600,000 processed transactions to date. WING maintains the largest mobile money infrastructure to support MFIs in Cambodia, offering a transactions platform for secure and safe deposit management, loan disbursement, and loan collection. 60 WING (2012) 96 Glob al Partn er ship for Financial I nc lu si on CASE 34  United Bank Ltd., Pakistan — “Omni” Branchless Banking United Bank Ltd. (UBL) is Pakistan’s second largest private bank, with approximately 3 million clients, 1,121 branches, and over 500 ATMs with 9 percent market share. Recognizing the small percentage of banked adults in Pakistan (10 percent have an account at a formal 61 financial institution ) and the limited reach of banks in rural areas, UBL launched the “Omni” mobile account and agent network in April 2010. Full-service kiosks are located at retail partner locations, known as Omni Durkaans, in over 500 cities and towns across Pakistan. In this “bank based” model, UBL accounts are created at Durkaan locations, and linked to Omni customers’ mobile number. Clients may transfer money to any Omni customer over any carrier, as well as send money to any bank account in Pakistan through the bank ATM switch One-Link. Optional debit cards are also available, and may be used at any of the 500 ATM network locations. Customers without mobile phones may also elect to send money or pay bills through the popular over-the-counter transfer service. Durkaan agents are furnished with a Blue Tooth printer for customer receipts. Account holders must maintain a minimum balance of Rs 100 (USD 1.15), and may choose from pay-as-you-go, weekly, monthly, or annul payment options. In the next 3–5 years, UBL anticipates gaining between 15–20 million Omni customers. UBL has also partnered with several governmental and non-governmental organizations to facilitate payment for relief and support programs. Two million Pakistanis have received payment from the Pakistani Government’s flood relief pro- 62 gram, the Benazir Income Support Program, and the World Food Programme through 5,000 Durkaan agents. UBL has also partnered with microfinance institutions to accept loan repayments. As Omni’s over-the-counter services are favored by many, a key challenge for UBL is marketing full customer accounts. 61 According to the Global Financial Inclusion (Global Findex) Database. World Bank (2012) 62 Bold (2011) Innovati v e Agric u ltu ral S ME Finance Models 97 CASE 35  Opportunity International Bank, Malawi — Mobile Banking Opportunity International Bank of Malawi (OIBM) is a commercial bank that focuses on serving eco- nomically disadvantaged Malawians. From 2006–2010, it increased its depositing clientele by well over 300 percent and its gross loan portfolio by approximately 25 million. The Mobile Bank reaches clients and individuals in deprived communities without adequate access to formal banking services. The Mobile Bank is a custom-equipped vehicle that travels to six trading centers in central Malawi along two scheduled routes. Each route is serviced one to two times per week. While two of the trading centers had formal banking services before its operation, the other four were previously unbanked. The facility operates like any fixed branch and offers the full range of banking services, includ- ing depositing, withdrawals, balance confirmation, and customer services. Transactions occur in real time through the Mobile Bank’s built-in ATM machine. Although loan services are offered, loan approvals must occur at the head office in Lilongwe. The Mobile Bank was designed to meet the unique needs of its rural clientele — offering relaxed identification requirements, reduced minimum balances and, most impor- tantly, reduced transactions costs and increased access and convenience. It is also fully networked to serve the entire client base of the company. The Mobile Bank is manned by well-trained tellers and has a cus- tomer service desk on board to advise and assist clients in their transactions. The Mobile Bank is secured by a security guard on board and has the highest level of security monitoring and control. After 2 years of operation (2008–2010), the number of savings accounts increased along the route by 80 percent, largely due to the Mobile Bank. OI’s market share of savings accounts increased by 10 percent at baseline, and 23 percent at the end line. Opportunity Malawi expects to increase its 305,000 savings accounts to 1 million within 3 years, banking with approximately 7 percent of the Malawian population. 98 Glob al Partn er ship for Financial I nc lu si on CASE 36  Dunavant Zambia Ltd. — Mobile Payment Systems Dunavant Zambia Ltd. is the largest cotton company in Zambia, with 100,000 contract farmers and a 60 percent market share. Mobile Transactions Zambia Limited (MTZL) is a Zambian mobile money com- pany that specializes in electronic transactions for unbanked and rural end users. In 2009, Mobile Transactions and Dunavant began to develop a system that interfaces with the out- grower management system to pay farmers electronically into accounts on their mobile phone “m-wal- lets.” Mobile Transactions first developed Dunavant’s online outgrower management system, which serves as the core of the information system to support the outgrower agricultural operations. Within one season, this system moved from a decentralized database at each of the nine agricultural offices to a centralized web-based platform hosted within Dunavant’s head office. Agricultural offices and rural sheds, equipped with laptops powered by solar-charged car batteries that can connect to the Internet via mobile GPRS modems, were able to capture real-time data into a centralized system and, most importantly, facilitate payment within 3 days. Recognizing that this delay still encourages a degree of side selling, MTZL built an online interface that facilitates payment through farmer’s MaKwacha Account m-wallets as soon as the Crop Voucher Receipt is processed, usually within 1 day. Farmers need only to visit their local MaKwacha agent to withdraw their cash. In order to serve the large number of farmers without a mobile device, MTZL offers same day payment at local agent locations through agent mobile phones. This system has been approved by the Bank of Zambia and is currently being piloted in Dunavant’s Eastern and Southern regions. Farmers can also use their mobile phones as an interface to store and transfer money, purchase airtime, and make retail purchases, such as for agricultural inputs. Several MaKwacha Agents also act as agricultural retailers and accept electronic payment for seeds, fer- tilizer, chemicals, and farming implements. Farmers can even contract local, small-scale tillage and spray service providers by making person-to-person money transfers between their mobile phones. Another option is that farmers can pay school fees directly to local schools that accept MTZL transactions. By partnering with schools, cooperatives, and input dealers to add value to their services, MTZL is able to build relationships with farmers and rapidly expand its operations. During the 2009/10 season, Dunavant farmers in four districts had the option of being paid into accounts on their mobile phones. MTZL and Dunavant are planning to offer this service to all 100,000 contract 63 farmers. By April 2011, Dunavant employee payments valued 1.7 million were made through the plat- form. There are formidable challenges, however. Though Zambia has followed the African trend of rapid mobile phone growth (there are now three million mobile phone users on the two largest networks, Zain and MTN), coverage is still limited to mainly urban areas. Only a small percentage of farmers have mobile phones, and many do not have an incentive to buy one because of low network coverage. 63 M’Grath, (2009) Innovati v e Agric u ltu ral S ME Finance Models 99 CASE 37  Africa Agriculture and Trade Investment Fund (AATIF) The Africa Agriculture and Trade Investment Fund (AATIF) is a debt investment fund that focuses on investments into the agricultural sector. It targets small, medium, and large-scale agricultural farms as well as agricultural businesses along the entire agricultural value chain, which will be financed indirectly or directly. Indirect Investments relate to investments into local financial institutions or other intermedi- aries (such as large agribusinesses) tant on-lend to the agricultural sector, to fund smallholders or SMEs, for example. Direct Investments comprise cooperatives, commercial farms, and processing companies, among others. The Fund is able to provide tailored financing solutions for agricultural investments, offer- ing refined and structured financing packages on market-oriented terms. Established as a closed-ended investment company, AATIF is an innovative public-private partnership dedicated to uplifting Africa’s agricultural potential for the benefit of the poor. AATIF was initiated by KfW on behalf of the German Federal Ministry for Economic Corporation and Development (BMZ) and was initially capitalized by BMZ, KfW, and Deutsche Bank, with the BMZ contribution being used as a first-loss layer. The current Fund volume amounts to USD 120 million. Since its establishment in August 2011, three investments have been financed by the Fund, with more in the last phase of closing. AATIF’s approach is based on close cooperation with partners who have investment experience in Africa and a profound knowledge of the respective value chains. These partners could be financial institutions, agri- businesses, off-takers, processing enterprises, traders, or others. They help to assure the quality of investments while supporting the fund in sourcing investments. In addition, the Fund’s partners will share the risks involved in investments in an adequate manner. As the first investments show, AATIF is able to provide early stage funding, thereby giving a positive signal for further capital raising (private investors, local financial institutions) from the market. The first investment finances the expansion and intensification of wheat, soy, and maize production through state-of-the-art irrigation systems in Zambia. The expected impact of the investment includes: ƒƒ Long-term employment generation ensuring adequate wages and social benefits; ƒƒ Productivity increase and innovation through modern irrigation system with efficient water use; ƒƒ Knowledge transfer by special training programs for neighboring or contracted smallholders; and ƒƒ Food security by enhancing production for the Zambian market and for DR Congo and Zimbabwe. The second investment comprises the financing of a rice mill in Ghana, allowing the investee and its smallholders to create an integrated value chain in rice production. The investment facilitates local impact through: ƒƒ Roll-out of a smallholder scheme targeting to impact up to 15,000 smallholders; ƒƒ Job creation, including higher skilled labor; and ƒƒ An employment scheme promoting the development of workers, from lower to higher skilled labor. The third investment is a wholesale refinancing line to a regional development bank. It provides funding for project finance to the agricultural sector along the entire agricultural value chain, covering the full agricultural spectrum. 100 Glob al Partn er ship for Financial I nc lu si on ANNEX A Case Overview # of # of Farmers Cases Covered (million) TOTAL 98 37.02 Geographic Distribution Africa 61 2.90 Asia 21 14.20 Europe 1 0.01 Latin America 14 19.90 Middle East 1 0.01 Distribution by type of Environment I Weak Business Environment, low Ag Productivity 20 6.61 II Strong Business Environment, low Ag Productivity 65 27.94 III High Ag Productivity 13 2.50 Distribution by secondary repayment source Farmer 37 31.36 Movable Asset 16 0.9 Buyer 45 4.78 Distribution by type of case Financing model 86 21.8 Risk Model 9 1.00 Distribution models 3 14.24 Innovati v e Agric u ltu ral S ME Finance Models 101 # of # of Farmers Cases Covered (million) Distribution by ownership Private 84 18.38 Public 14 18.64 Distribution of ALL cases by type of environment and risk I Weak Business Environment*, low Ag Productivity** Farmer 8 6.10 Movable Asset 2 0.00 Buyer 10 0.51 II Strong Business Environment***, low Ag Productivity** Farmer 22 22.82 Movable Asset 10 0.84 Buyer 33 4.27 III High Ag Productivity**** Farmer 9 2.44 Movable Asset 2 0.06 Buyer 2 0.00 General Note: information depth and quality varies significantly among cases, and large gaps exist, in particular with regard to # of farmers covered. Cases cover mostly Africa, to a lesser extent Asia, only very little of the Latin American reality, so they are by no means representative. * Rank in Doing Business table of less than 91 ** Productivity per aagricultural worker of less than 2000 *** Rank in Doing business table equal or higher than 91 **** Productivity per agricultural worker higher than 2000 USD (constant 2000 USD) 102 Glob al Partn er ship for Financial I nc lu si on # of # of Farmers In % of Cases Covered (million) all cases Distribution of PRIVATE cases by type of environment and risk I Weak Business Environment*, low Ag Productivity** Farmer 6 0.42 7% Movable Asset 2 0.00 10% Buyer 10 0.50 99% II Strong Business Environment***, low Ag Productivity** Farmer 17 15.82 69% Movable Asset 10 0.84 100% Buyer 29 0.45 10% III High Ag Productivity**** Farmer 6 0.28 11% Movable Asset 1 0.06 100% Buyer 2 0.00 100% General Note: information depth and quality varies significantly among cases, and large gaps exist, in particular with regard to # of farmers covered. Cases cover mostly Africa, to a lesser extent Asia, only very little of the Latin American reality, so they are by no means representative. * Rank in Doing Business table of less than 91 ** Productivity per aagricultural worker of less than 2000 *** Rank in Doing business table equal or higher than 91 **** Productivity per agricultural worker higher than 2000 USD (constant 2000 USD) Innovati v e Agric u ltu ral S ME Finance Models 103 ANNEX B NPLs and GDP Dependency on Agriculture Agricultural portfolio NPLs tend to be independent from financial markets. They are affected more by prices for inputs and outputs and, most importantly, by weather and other production risks, all of which are uncorrelated with financial markets. This is an opportunity to diversify and deploy capital more effi- ciently for a larger portfolio. As identified in this document, there is a vast array of innovative financing, risk mitigation, and distribu- tion models around the world that demonstrate that risks and costs can be reduced significantly, result- ing in relatively high profits in the under-banked sector. With regard to the performance of agricultural lending, we conducted a two-step analysis of country- level data. Given that data on agricultural lending portfolios are hard to come by, the team first tested whether a high agricultural contribution to the economy is somehow associated with the level of stability (NPLs) and efficiency (lending spreads) of bank lending in general. Given that data on agricultural lend- ing as a share of total bank lending are not uniformly reported, we have conducted an analysis using agriculture as a share of GDP as a proxy for lending to agriculture as a share of total lending. Figure 15  Agricultural lending share and NPLs in 5 countries 20 18 BGL 2004 16 BGL 2007 MOZ 2003 14 NPLS/GROSS LOANS IN % BGL 2005 12 BGL 2006 BGL 2008 10 MAU 2003 8 MAU 2004 MAU 2002 TZA 2008 MOZ 2004 6 TZA 2009 4 MAU 2005 MOZ 2005 UGA 2002 MAU 2006 MOZ 2006 MOZ 2007 2 MOZ 2008 0 0 2 4 6 8 10 12 14 LENDING TO AGRICULTURE/TOTAL COMMERCIAL LENDING IN % 104 Glob al Partn er ship for Financial I nc lu si on Of course, a high proportion of agriculture in a country’s GDP does not necessarily correspond to a simi- lar proportion of lending to the agriculture sector. Therefore, the following analysis is limited by the assumption that there is some degree of correlation between agricultural GDP and agricultural lending. We analyzed data from 93 countries and found that higher agricultural shares of GDP are positively correlated with lending spreads (non-weighted correlation 47 percent), but there is no correlation between agricultural GDP and NPLs (see Annex A for details). Given the significant association between agricultural GDP and lend- ing spreads, we assume that higher agricultural shares are associated with relatively larger agricultural lending portfolios. This is, of course, a strong assumption to make, but some anecdotal data seems to confirm that 64 lending to agriculture and associated activities seems to increase with the size of the sector in the economy. In the chart below, we plot agricultural lending shares and NPLs of five countries (Mozambique, Uganda, Mauritius, Bangladesh, and Tanzania) for those with available data. The chart illustrates a trend towards lower NPLs regardless of agricultural shares in commercial lending generally, and quite significant agricultural 65 lending shares that tend to be higher in higher agricultural GDP countries (Mozambique, Tanzania). We therefore grouped 63 developing and developed countries into three categories (high, medium, and low agricultural GDP shares), and found that NPL levels of high agricultural GDP countries tended to be significantly higher as recently as 2002, but converged with NPL levels of medium and low agricultural GDP countries over the course of the decade. Furthermore, the financial crisis at the end of 2008 seems to have had a smaller effect on the high agricultural GDP countries, as the chart below demonstrates. Figure 16  NPLs of high, medium and low agricultural GDP share countries 20 18 16 NPLS/GROSS LOANS 14 12 10 Financial Crisis 8 6 4 2 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 High Ag GDP share Medium Ag GDP share Low Ag GDP share 64 For the high agricultural GDP share category, an example is Uganda’s commercial banks, which have almost 7 percent of their portfolios in agriculture and 6 percent in food, beverages, and tobacco manufacturing [Bank of Uganda (2012)], whereas agriculture represents 24 percent of GDP. Tanzania’s commercial banks had more than 10 percent of their portfolios in agriculture which generated around 28 percent of GDP in 2010 [IMF (2010)]. In the medium agricultural share of GDP category, China’s rural credit cooperatives hold 6 percent of all financial assets whereas the agricultural GDP has been 10 percent of the total in 2010 [IMF (2011)]. 65 The exception seems to be Mauritius, where higher NPLs in the sugar sector led to banks diversifying into tourism and other new sectors, resulting in lower agricultural lending shares. Innovati v e Agric u ltu ral S ME Finance Models 105 Three phenomena may partially explain this trend: changes in the way governments intervene in agricul- tural finance markets, the degree of correlation of agricultural lending with financial markets, and the commodity booms since 2007. There is significant evidence for less direct government intervention in agricultural finance in the last decade, which lowers the high NPLs of direct state lending institutions or 66 programs such as “revolving” funds. Currently, governments and donors tend to work more through second-tier lending schemes (i.e., Mexico’s FIRA or Brazil’s BNDES), guarantees, and risk sharing mecha- nisms. The second phenomenon is the substantial disconnect between agricultural lending portfolio per- formance and international financial markets. Agricultural markets tend to be more stable — they are essentially “decoupled” from financial markets and therefore less affected by the financial crisis fallout. 67 The third phenomenon is the commodity price boom since around 2007 that might tend to improve agricultural business performance and lower NPLs. Again, these country cross-comparisons do not directly compare agricultural lending portfolio quality and performance; they only infer those agricul- tural lending portfolio qualities from national data on aggregate lending portfolios. Nonetheless, the results at least indicate that agricultural lending does not appear to be necessarily riskier than lending to other sectors as there is a clear trend towards lower NPLs in high agricultural GDP countries. Additionally, data on lending spreads from the same countries suggest that transaction costs in high agricultural share and high agricultural lending countries are higher, and therefore lending spreads are higher. These high lending spreads and profits in high agricultural GDP countries might also be explained by less bank- ing competition in these countries. 66 GTZ-BMZ (2005) 67 For example, oilseed prices at N.W. European ports hovered around 350 USD/ton in early 2007, reached 950 USD in the case of sunflower seeds in early 2008, returned to a higher level of 450 USD in 2009, and again climbed to 650 USD in early 2011 well into a second commodity boom period. LMC-International (2011) 106 Glob al Partn er ship for Financial I nc lu si on ANNEX C Cases with Color Coded Ratings Risk Name of model Finance approach Model Focus Continent Country Environment Ghana Grains Value Chain finance External VCF Buyer Africa Ghana 1 Partnership for open market crops Ecom Trading Input finance for Internal Value Buyer Africa Ghana 1 coffee farmers chain finance PepsiCo Contract farming External VCF Buyer Asia India 2 sweet potato for closed market crops Kenya Gatsby Factoring Factoring Buyer Africa Kenya 2 Trust Equity Bank Agri Value Chain finance External VCF Buyer Africa Kenya 2 SME financing for open market crops Agro Dealer Agro Dealer Internal Value Buyer Africa Tanzania 2 financing NMB financing chain finance NMB Outgrower Value Chain finance External VCF Buyer Africa Tanzania 2 Scheme Sugar for closed Cane market crops Gulu Agricultural Supply chain Internal Value Buyer Africa Uganda 2 Development financing -inputs chain finance Company for cotton farmers (GADC) Centenary Bank/ Factoring Factoring Buyer Africa Uganda 2 Technoserve Bayer Input finance Internal Value Buyer Europe Ukraine 2 chain finance Innovati v e Agric u ltu ral S ME Finance Models 107 Risk Name of model Finance approach Model Focus Continent Country Environment Dunavant Mobile payments External VCF Buyer Africa Zambia 1 Cotton/Mobile for closed Transactions market crops Zambia Ltd. Dunavant Cotton Contract farming External VCF Buyer Africa Zambia 1 contract farming for closed market crops Zanaco Dairy Value Chain finance External VCF Buyer Africa Zambia 1 Cooperative for closed Financing market crops M-PESA mobile banking Mobile Farmer Africa Kenya 2 banking M-KESHO microsavings, micro Direct Farmer Africa Kenya 2 insurance, Smallholder microcredit lending Kilimo Salama Weather insurance Crop/Weather Farmer Africa Kenya 2 Insurance Finterra (Private Emerging Farm Emerging Farmer South Mexico 3 Agri bank) business Finance Farm business America Finance TSKI MFI Weather insurance Weather / Farmer Asia Philippines 2 Typhoon index Crop insurance Insurance Wizzit Bank Branchless banking branchless Farmer Africa South Africa 3 banking Kilimanjaro Health Insurance Health Farmer Africa Tanzania 2 Native supports coopera- Insurance Cooperative tive coffee lending Union Mtibwa Sugar Nucleus/outgrower Nucleus/ Farmer Africa Tanzania 2 cane outgrower outgrower scheme NMB Kilimo sav- Cash Collateral Savings Farmer Africa Tanzania 2 ings account & account linked input financing Finance 108 Glob al Partn er ship for Financial I nc lu si on Risk Name of model Finance approach Model Focus Continent Country Environment Munda (or LIMA) Cash Collateral Direct Farmer Africa Zambia 1 Scheme Smallholder lending Zanaco Emerging Farm Emerging Farmer Africa Zambia 1 Emerging Farmer business Finance Farm business Project Finance De Lage Landen Asset and Leasing Term Loans/ movable South Brazil 3 Brazil finance Equipment collateral America Finance EcoSafe Ghana Collateral Collateral movable Africa Ghana 1 Ltd. management management collateral agreement Term Loans/ Value chain movable Jain Irrigation Equipment Asia India 2 financing collateral Finance NMB Warehouse Warehouse Receipt Ware House movable Receipt Africa Tanzania 2 Financing Receipt collateral Financing Cashew Innovati v e Agric u ltu ral S ME Finance Models 109 Bibliography AGRA. 2009. AGRA Stories from the Field. http://www.agra-alliance.org/content/story/detail/1002/ Agrifood Consulting International. 2005. Smallholder Agricultural Commercialization Strategy for Zambia. Project Brief Series. Anathakrishnan, P. V. 2007. Structured Finance through Collateral Management. Andrade, M. 2011. Correspondent Banking to Increase Outreach to Smallholder Farmers: The HDFC Bank Experience. Hyderabad. Annamalai, K., and S. Rao. 2003. What Works Case Study: ITC’s e-Choupal and Profitable Rural Transformation. World Resources Institute. University of Michigan. Armstrong, D. 2011. Mobile Banking Is Mainstream. Rabo Development: Case Studies in Mobile Banking. http://www.slideshare.net/danarmstrong/mobile-banking-rabo-development-partner-banks Bank of Uganda, 2012. Sectoral Analysis of Commercial Banks’ Credit to the Private Sector. Kampala: Bank of Uganda. Bold, C. 2011. Branchless Banking in Pakistan: A Laboratory for Innovation. CGAP Brief. Carletto, C., S. Savastano, and A. Zezza, A. 2011. Fact or Artefact:The Impact of Measurement Errors on the Farm Size-Productivity Relationship. Washington DC: World Bank DECRG. Cash, D., W. Clark, F. Alcock, N. Dickson, and N. E. Jäger. 2002. Salience, Credibility, Legitimacy and Boundaries: Linking Research, Assessment and Decision Making. John F. Kennedy School of Government Harvard University Faculty Research Working Papers Series. Christen, R. P., and D. Pearce. 2005. Managing Risks and Designing Products for Agricultural MicroFinance: Features of an Emerging Model. Washington DC: CGAP. Dalberg, Citi Foundation, and Skoll Foundation. 2012. Catalyzing Smallholder Agriculture Finance, Dalberg Global Development Advisors. de Vries, K. 2012. Dairy Key Fact Sheet: From Islands of Success to Seas of Change Scaling Inclusive Agri-food Market Development — International Learning Initiative and Workshop. Center for Development Innovation, Seas of Change. Center for Development Innovation. Deininger, K., and D. Byerlee. 2011. The Rise of Large Farms in Land Abundant Countries: Do They Have A Future? World Bank, Development Research Group, Agriculture and Rural Development Team. DFCU. 2012. http://www.dfcugroup.com. FAO. 2009. How to Feed the World in 2050. Rome: FAO GTZ-BMZ. 2005. Reforming agricultural development banks. Bonn. Hawkes, C., and M. T. Ruel. 2011. Value Chains for Nutrition. Leveraging Agriculture for Improving Nutrition and Health. New Delhi: International Food Policy Research Institute. 110 Glob al Partn er ship for Financial I nc lu si on Hazell, P. B., and U. Hess. 2010. Drought Insurance for Agricultural Development and Food Security in Dryland Areas. Food Security, 395–405. Hess, U., and P. Hazell. 2011. Sustainability and Scalability of Index-based Insurance for Agriculture. Innovations in Insuring the Poor. Washington DC: IFPRI. Holliger, F. 2004. Financing Agricultural Term Investments. Rome: FAO-GTZ. IFAD. 2010. ESA Portfolio Performance Report — Annual Review July 2009–June 2010. Rome: IFAD. IFC. 2011. Scaling Up Access to Finance forAgricultural SMEs: Policy Review and Recommendations. Washington DC: IFC. Illovo. 2011. Annual Report 2011. South Africa: Illovo. IMF. 2011. People’s Republic of China: Financial System Stability Assessment. Washington DC: IMF. IMF. 2010. United Republic of Tanzania: Financial System Stability Assessment Update. Washington DC: IMF. IMF. 2003. Uganda Financial System Stability Assessment. Washington DC: IMF. Kisaame, J. 2003. Case Study of DFCU Leasing Company — Uganda. Paving the Way Forward for Rural Finance: An International Conference on Best Practices. LMC-International. 2011. Oilseeds, Oil and Meal analysis. LMC International. Luna-Martínez, J. D., and C. L. Vicente, C. L. 2012. Global Survey of Development Banks. World Bank Working Paper. Matango, R. 2006. Mtibwa Outgrowers Scheme: A Model for Smallholder Cane Production in Tanzania. Paper presented at the UNCTAD Expert Meeting: “Enabling Small Commodity Producers in Developing Countries to Reach Global Markets.” McConnell, M., E. Dohlman, and S. Haley. 2010. World Sugar Price Volatility Intensified by Market and Policy Factors. Amber Waves. M’Grath, B. 2009. DUNAVANT and Mobile Transactions: The Future of Rural Agricultural Payments. Compaci News Competitive African Cotton Initiative. MicroEnsure. 2010. MicroEnsure Philippines Storm Insurance. http://www.microensure.com/news.asp?id=69 Milder, B. 2008. Closing the Gap: Reaching the Missing Middle and Rural Poor through Value Chain Finance. Enterprise Development and Microfinance December 2008. NMB. 2012. NMB Tanzania AgroDealers. http://www.nmbtz.com/index.php?option=com_content&view=article&id=151&Itemid=224 NMB. 2010. NMB Kilimo Account. http://www.nmbtz.com/index.php?option=com_content&view=article&id=264&Itemid=248 OMNI. 2012. UBL. https://www.ubl.com.pk/omni/ Oxfam. 2009. The Missing Middle in Agricultural Finance — Relieving the Capital Constraint on Smallholder Groups and Other Agricultural SMEs. Oxfam. Peltzer, R. 2011. African Cotton Is Making a Comeback. Outlook for COMPACI beyond 2012. COMPACI. Philippines Telecomresearch. 2012. Philippines — Telecoms, Mobile, Broadband and Forecasts. http://www.budde.com.au/Research/Philippines-Telecoms-Mobile-Broadband-and-Forecasts.html Sulle, E., and T. N. Forum, editors Lorenzo Cotula and Rebeca Leonard. 2007. A Hybrid Business Model: The Case of Sugarcane Producers in Tanzania. Alternatives to Land Acquisitions: Agricultural Investment and Collaborative Business Models. Maputo, Mozambique: IFAD. Innovati v e Agric u ltu ral S ME Finance Models 111 Swanson, R. 2009. Final Evaluation of Land O’Lakes Zambia Title II Development Assistance Program. Washington DC: USAID. Syngenta Foundation for Sustainable Agriculture. 2011. Syngenta Foundation Agricultural Index Insurance Initiative. http://www.syngentafoundation.org/index.cfm?pageID=562 Technoserve-IFAD. 2011. Outgrower Schemes — Enhancing Profitability. Rome: IFAD. Telecom Regulatory Authority of India. 2011. Highlights of Telecom Subscription Data as of 31st December, 2011. New Delhi: Government of India. TigoGhana. 2010. Tigo Family Care Insurance. http://www.tigo.com.gh/Innovations/Tigo-Family-Care-Insurance.aspx Vargas-Hill, R., and M. Robles. 2011. Flexible Insurance for Heterogeneous Farmers Results from a Small-Scale Pilot in Ethiopia. Washington DC: IFPRI. Vargas-Hill, R., J. Hoddinott, and N. Kumar. 201. Adoption of Weather Index Insurance. ESSP II Working Paper 27. Wegner, L. 2012. Coffee Fact Sheet: From Islands of Success to Seas of Change Scaling Inclusive Agri-food Market Development — International Learning Initiative and Workshop. Centre for Development Innovation. Wiggins, S., J. Kirsten, and L. Llambi. 2010. The Future of Small Farms. World Development, 38 (10), 1341–1348. WING. 2012. http://www.wingmoney.com/en/products-services/what-is-wing/ WIZZIT. 2012. http://www.wizzit.co.za/ World Bank. 2005. Rural Finance Innovations. Washington DC: ARD World Bank. Worldbank. 2006. Buffalo, Bakeries, and Tractors: Cases in Rural Leasing From Pakistan, Uganda, and Mexico. Washington DC: ARD World Bank. World Bank. 2007. World Development Report 2008: Agriculture for Development. Washington, DC: The World Bank. World Bank. 2010a. Doing Business 2010. Washington DC: World Bank. World Bank. 2010b. World Development Indicators. Washington DC: World Bank. World Bank. 2012. Global Financial Inclusion (Global Findex) Database. http://www.worldbank.org/globalfindex World Factbook. 2011. Number of Mobile Phones in Use. https://www.cia.gov/library/publications/ the-world-factbook/ Zanaco. 2012. Munda Facility. http://www.zanaco.co.zm/Corporate%20Banking%20-%20munda.htm Innovati v e Agric u ltu ral S ME Finance Models Cover3 International Finance Corporation 2121 Pennsylvania Avenue, NW Washington, DC 20433 USA www.ifc.org