How Can Matching Grants in Agriculture Facilitate Access to Finance? Lessons Learned from World Bank Group’s Experience © 2019 The World Bank Group 1818 H Street NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org All rights reserved. This volume is a product of the staff and external authors of the World Bank Group. The World Bank Group refers to the member institutions of the World Bank Group: The World Bank (International Bank for Reconstruction and Development); International Finance Corporation (IFC); and Multilateral Investment Guarantee Agency (MIGA), which are separate and distinct legal entities each organized under its respective Articles of Agreement. We encourage use for educational and non- commercial purposes. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Directors or Executive Directors of the respective institutions of the World Bank Group or the governments they represent. The World Bank Group does not guarantee the accuracy of the data included in this work. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. All queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Photo Credits: World Bank Photo Library and Shutterstock.com WHEN AND HOW SHOULD AGRICULTURAL INSURANCE BE SUBSIDIZED? ISSUES AND GOOD PRACTICES B Rachel Sberro-Kessler | Finance, Competitiveness, and Innovation Global Practice World Bank Group Table of Contents Abbreviations and Acronyms................................................................................. III Acknowledgments..................................................................................................... V Introduction................................................................................................................1 1. Why Using Matching Grants and Why Engaging the Financial Sector May Matter................................................................................................ 3 1.1 The Rationale for Matching Grants and the Limited Evidence on Their Impact........................................................................................................... 3 1.2 How Matching Grants Can Help Address Constraints to Access to Finance................................................................................................................6 2. WBG Experience with Matching Grants in Agriculture and Drivers of Success................................................................................................ 11 2.1 Use of Matching Grants in Agriculture and Regional Distribution at the WBG............................................................................................................ 12 2.2 Characteristics of Agriculture Matching Grants Projects and Success Factors.................................................................................................... 16 3. Promoting Access to Finance Through Grants and the Four Roles of Financial Institutions......................................................................................... 25 3.1 Matching Grants Projects and Financial Inclusion.......................................25 3.2 Financial Institutions as Deposit Takers........................................................28 3.3 Financial Institutions as Funders.................................................................... 30 3.4 Financial Iinstitutions as Managers................................................................32 3.5 Financial Institutions as Advisors....................................................................33 Conclusion and Recommendations.......................................................................35 References................................................................................................................39 Annex 1. Financial Institutions as Deposit Takers...............................................41 Annex 2. Financial Institutions as Funders......................................................... 47 Annex 3. Financial Institutions as Managers...................................................... 49 Annex 4. Financial Institutions as Advisors........................................................ 51 Endnotes...................................................................................................................53 HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? I LIST OF BOXES Box 1. Experience in United States with Support for Asset Building Through Matched Savings................................................................................... 4 Box 2. The Experience with High-leverage Matching Grants in Europe and Central Asia (ECA).......................................................................... 27 Box 3. The Experience with Low-leverage Matching Grants in Africa............. 28 Box 4. Incentivizing Commercial Credit in Argentina for Climate-smart Agriculture Investments.................................................................................... 31 LIST OF FIGURES Figure 1. Demand-side and Supply-side Constraints That Can be Addressed with Matching Grants........................................................................7 Figure 2. Agriculture Finance Constraints That Can be Addressed with Alternative Instruments...................................................................................... 8 Figure 3. Matching Grants and Financial Services—The “Do No Harm” Approach................................................................................................................ 9 Figure 4. Approval of World Bank Projects with Matching Grant Component (Fiscal Years)....................................................................................12 Figure 5. Number and Volume of Projects by Region...........................................12 Figure 6. Success Rate by Year of Approval (Fiscal Year).................................. 13 Figure 7. Success Rate of Core Agriculture v. Nnon-agriculture Projects....... 16 Figure 8. Success Rate of Broad Agriculture v. Non-agriculture Projects.......17 Figure 9. Volume of Matching Grant Components by Sector.............................17 Figure 10. Success Rate by Availability of Various Levels of Matching..........20 Figure 11. Success Rate by Level of Matching......................................................21 Figure 12. Level of Matching by Penetration of Rural Financial Services........21 Figure 13. Success Rate by Provision of Diagnostic........................................... 22 Figure 14. Success Rate by Selection Mechanism.............................................. 23 Figure 15. Success Rate by Existence of Access to Finance Component........ 24 Figure 16. Financial Institutions Have Four Potential Roles as Part of Matching Grants Projects................................................................ 26 LIST OF TABLES Table 1. List of 15 Closed Matching Grants Projects in Agriculture..................14 Table 2. List and Frequency of Use of M&E Indicators for Agriculture Matching Grants Projects............................................................. 18 Table 3. Results on Competitiveness.................................................................... 43 TABLE OF CONTENTS II Abbreviations and Acronyms BDS Business Development Services CIG Common Interest Group CSA Climate-Smart Agriculture ECA Europe and Central Asia FFCA Fadama Farmers Community Association FUEF Fadama Users’ Equity Fund FUG Fadama User Group ICR Implementation Completion and Results Report  IDA Individual Development Account IE Impact Evaluation IEG Independent Evaluation Group LAC Latin America and the Caribbean MG Matching Grants PCG Partial Credit Guarantee PIU Project Implementation Unit RCT Randomized Controlled Trial SMEs Small and Medium Enterprises TA Technical Assistance WBG World Bank Group WTO World Trade Organization HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? III Acknowledgments This report was written by Rachel Sberro-Kessler. It has benefitted from inputs from Daniel Ortiz, Sandra Broka, Panos Varangis, Toshiaki Ono, Juan Buchenau, and Mazen Bouri. It has been peer reviewed by Diana Hristova, David Tuchschneider, Michael Goldberg and Ajai Nair. The author also wishes to thank Anne Himmelfarb for editing and Aichin Lim Jones for providing the design and layout for this report. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? V VI Introduction Matching grants are an instrument aimed at promoting private sector development that have been used extensively over the past years, in particular for agriculture development. A matching grant is defined as “a one- off, non-reimbursable transfer to project beneficiaries, for a specific purpose, based on the condition that the recipient makes a contribution for the same purpose.”1 These grants can be used for a variety of activities, including technical assistance, investment in assets, or financing of working capital. A recent review2 showed that the World Bank Group (WBG) had supported 106 private sector development matching grant projects over the past decades, including 21 in the agriculture sector. While agriculture projects account for a small portion of the total number of projects, total grant financing dedicated to agriculture reached US$650 million, or almost twice the volume of that outside of agriculture.3 In addition, the proportion of matching grants projects supporting agriculture has significantly increased in the 2000s. Recent interest for this instrument to support agriculture might be due both to growing concerns about forms of support which distort financial markets, such as interest rate subsidies, and to the compatibility of such agricultural subsidies with World Trade Organization (WTO) requirements.4 However, there is very limited rigorous evidence on the effectiveness of matching grants—specifically, on their additionality or sustainability impact. The issue of additionality can be summarized thus: “Do matching grants crowd out private investment by subsidizing investment that would have been made anyway?” On the other hand, the issue of sustainability addresses this question: “Can supported projects be self-sufficient after the matching grants project closes?” The conclusion from the recent WBG review5 is that “experience has shown that matching grants rarely yield the type of broad and durable economic benefits that would justify the subsidization of private enterprises with public funds.”6 While matching grants are often used as substitutes for well-functioning financial markets, literature suggests that matching grants do not sufficiently work as enablers of financial markets. Indeed, while the primary objective of matching grants is often to increase the income of beneficiaries in the absence HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 1 of well-functioning financial markets, matching Building on previous literature as well as a grants should also be designed in a way that helps detailed analysis of WBG agriculture matching beneficiaries build relationships with financial grants projects, this paper focuses on three institutions so that their future expenses and specific issues: (1) What is the rationale for using investments can be undertaken without the need matching grants in agriculture and why does the for grants.7 However, a recent report on matching financial sector matter? (2) What has been the grants for productive alliances in Latin America specific experience with WBG matching grants for and the Caribbean (LAC) indicates that “In their agriculture and what are the key drivers of success? design, almost all Productive Alliance projects (3) What are the various models of linkages with mention the goal of enhancing producers’ access to financial institutions and how can matching grants commercial financial services to complement project be used to promote financial inclusion? Based grant financing and beneficiary contributions, but in on this analysis, this paper suggests emerging practice few such linkages have materialized.”8 good practice on when to use matching grants for agriculture, and how to design them in a way that promotes sustainable impact and linkages with the financial sector. INTRODUCTION 2 1. Why Using Matching Grants and Why Engaging the Financial Sector May Matter Summary: Based on a literature review as well as an analysis of WBG matching grants projects, this section analyzes the rationale for matching grants projects and the role of matching grants in addressing agriculture and rural finance constraints. It shows that WBG matching grants generally lack proper identification of a market failure, thereby leading to suboptimal objective setting and limited long-term impact. This analysis suggests that when a market failure related to the lack of access to finance is identified, matching grants projects should include the improvement of access to financial services as a project objective. Matching grants may however not always be the most cost-effective instruments to help farmers invest in productive activities when rural financial markets are limited, and it is recommended that constraints to agriculture and rural finance are systematically assessed before setting up a matching grants project. Finally, in order to avoid misallocation and market distortions, matching grants should be designed to exclude bankable segments and bankable projects, and to offer tailored features by type of segment and type of project. 1.1 The Rationale for Matching Grants and the Limited Evidence on Their Impact Matching grants may stimulate market development and innovation and promote asset building among low-income segments. Matching grants may help foster private investments and push investors towards underserved markets by addressing specific barriers to market development. Matching grants may help farmers and agricultural small and medium enterprises (SMEs) invest in activities that have great potential to generate growth—activities that under other circumstances they would be unwilling or unable to finance due to various constraints (internal or external, financial or nonfinancial). In particular, matching grants are often used as a way to stimulate innovation or technology adoption, as investors might be reluctant to invest due to high risks. For instance, a large matching grant program managed by the Colombian innovation agency COLCIENCIAS provides evidence that, over the period 1995–2007, COLCIENCIAS funding had an average impact on labor productivity of 15%.9 In some cases matching grants mainly promote asset building as an objective in itself, as the experience HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 3 in the United States suggests (see box 1). Building when used to support public goods (e.g., agricultural assets can provide a vital financial cushion against research and development, agriculture extension) poverty when shocks happen, but can also serve as a or semi-public goods (irrigation schemes, climate- springboard for investments. smart agricultural investments, or market facilities benefiting several members of a community). In However, matching grants can also potentially particular, support to semi-public goods can be be misallocated—if public resources are used justified by positive externalities (e.g., job creation for projects that are nonviable investments in rural areas, increased food safety) and spillover or captured by elites—and can also distort effects (e.g., support to technology adoption among markets—if they substitute savings or commercial beneficiaries may lead to further adoption by non- credit. The use of grants can be justified to address beneficiaries, training of beneficiaries may then market failures or on poverty grounds. Literature benefit non-beneficiaries). suggests that matching grants are least controversial Box 1. Experience in United States with Support for Asset Building Through Matched Savings Background Since the 1990s, various U.S. states have promoted Individual Development Accounts (IDAs) as an asset-building strategy among low-income families. IDAs are savings accounts held by low-income individuals generally used to purchase a home, pay for post-secondary educational expenses, or start a business. IDA holders make monthly contributions to an account where funds are matched by public or private sources (or a combination of both) at a predetermined rate. Outcome and impact Over the last decade, more than 85,000 IDAs have been opened. The impact of this initiative has resulted in more than 9,400 new homeowners, 7,200 educational purchases, and 6,400 small business start-up and expansion purchases. Impact evaluation suggests statistically significant effects of the program on the three major forms of asset ownership.a IDA participants were 35% more likely to be homeowners, 84% more likely to own businesses, and 95% more likely to pursue post-secondary education than nonparticipants. Research also suggests that IDA participants not only are likely to become homebuyers earlier than other low-income persons, but also tend to be more successful homeowners and less vulnerable to shocks.b Compared to other low-income homebuyers who purchased homes in the same communities and over the same time period, IDA homebuyers obtained significantly preferable mortgage loan terms, with only 1.5% having high-interest mortgage rates, compared to 20% of the broader sample; and were two to three times less likely to lose their homes to foreclosure. This study provides the first evidence available on loan terms and foreclosure outcomes of IDA homebuyers. The findings suggest that participation in an IDA program with its related services and restrictions can improve homeownership outcomes for low-income households. a. Mills et al. (2008). b. Rademacher et al. (2010). 1. WHY USING MATCHING GRANTS AND WHY ENGAGING THE FINANCIAL SECTOR MAY MATTER 4 Matching grants should only be used when they would justify the subsidization of private enterprises are determined to be the most adapted and with public funds.”15 Ideally, matching grants for least-cost tool to achieve broad and durable agriculture should support investments that improve impact.10 The recent WBG review of matching product quality, reduce post-harvest losses, and/or grants indicates that without a rigorous economic enhance productivity on a long-term basis. analysis of the market failure, use of matching grants Overall, literature suggests that there is might lead to “limited additionality and spillovers, limited rigorous evidence on the effectiveness weak demand and disbursements, unintended of matching grants. Most studies only compare consequences on the business development services beneficiaries before and after matching grants, and market or nonsustainable impact if the project only a few randomized controlled trials (RCTs) have does not address binding constraints for SMEs been undertaken. In addition, most studies track (e.g. access to credit).”11 Matching grants are not a one specific indicator (increase in sales, increase in sustainable financing instrument, but their objective productivity) without analyzing broader measures should be to support sustainable investments. of impact (increase in income, profits, number of Indicators of sustainable investments suggested by people employed, etc.). One completed study is an the International Fund for Agricultural Development RCT of a government-led matching grant scheme (IFAD) include “improved access to financial in Mexico.16 The results on one-year impacts services by the beneficiary” and “replacement or show positive effects on return on assets and total expansion of the productive asset by the beneficiary factor productivity.17 A study in Yemen, which was over time.”12 interrupted due to eruption of civil conflict, showed However, WBG matching grants generally lack that in the first year after the program, the matching proper identification of a market failure, thereby grant was found to have led to more product leading to suboptimal objective setting and limited innovation, along with firms upgrading their long-term impact. More than a decade ago, a WBG accounting systems, marketing more, making more review of matching grants indicated: “Loose claims capital investments, and being more likely to report of market failure can easily result in misguided their sales grew.18 A meta-analysis of 20 individual interventions with grants (…) The fact that the assessments of matching grants suggests a positive private sector does not invest in certain fields is not impact on firms’ performance and employment.19 necessarily a sign of market failure.” 13 Unfortunately, Attempts were made to conduct seven RCTs in six a more recent review highlights the same weaknesses countries in Africa, but they proved not possible in WBG projects: “The justification for the use of this to implement for political and technical reasons.20 instrument is rarely well articulated in the reviewed While the RCT method could not be used, an impact projects and (…) a quarter of the projects in the evaluation (IE) in Mozambique suggested that sample had no meaningful indicator with which to matching grants had led to an increase in sales of gauge success.”14 One of the key conclusions of beneficiary firms above 20% but not to an increase this report is that “matching grants rarely yield the in profits. type of broad and durable economic benefits that HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 5 1.2 How Matching Grants Can Help Most WBG matching grants projects identify Address Constraints to Access the lack of rural finance as a sufficient rationale to Finance for matching grants, without fully identifying the specific market failure and whether other 1.2.1 Matching grants, when? Analyzing instruments might be more appropriate to unlock constraints to agriculture and rural rural and agriculture finance. As an example, finance and comparing matching grants the use of matching grants in the Nigeria Fadama to alternative instruments III Development Project (P096572) is justified in Matching grants for agriculture may be used the Project Appraisal Document as follows: “This to address a variety of market failures. These approach to financing is adopted due to the low include demand-side constraints both nonfinancial performance of rural financial markets in Nigeria, (e.g., lack of willingness to invest in business which are particularly deficient and limited in terms development services, or in technology which has of outreach in the rural areas.” unproven results) and financial (e.g., lack of trust in financial institutions). These also include supply- Matching grants are a temporary instrument side constraints both nonfinancial (e.g., lack of but can help address a variety of demand- supply of business development services providers) side and supply-side constraints to agriculture and financial (limited supply of rural finance). finance. Some of these benefits are only applicable Additionally, financial supply-side constraints may during the course of the project (e.g., reduction of themselves be due to a variety of factors, including risks and costs associated with financing farmers), lack of information, lack of know-how, lack of while others are sustainable after the project ends liquidity, or high risks and costs associated with (e.g., generation of trust, knowledge, and skills).21 rural finance. In figure 1 below, the first category of constraints and benefits is highlighted in light blue boxes, and the second category is highlighted in deep blue boxes. 1. WHY USING MATCHING GRANTS AND WHY ENGAGING THE FINANCIAL SECTOR MAY MATTER 6 Figure 1. Demand-side and Supply-side Constraints That Can be Addressed with Matching Grants Demand-side Constraints That Can be Addressed with Matching Grants • Matching grants can promote demand for credit by demonstrating to Lack of farmers the profitability of agricultural investments (e.g., see in Colombia, Willingness more than 11,000 rural households that were not beneficiaries of matching to Invest grants nevertheless adopted at their own expense improved practices promoted through the project). Lack of Skills • Matching grants can help farmers and SMEs develop skills to prepare to Invest business plans (both for current project and future activities). Lack of Trust • When matching grants projects require beneficiaries to save a specific Towards Financial amount at a specific frequency, matching grants projects can Institutions build financial capacity and trust towards financial institutions. Supply-side Constraints That Can be Addressed with Matching Grants Lack of Information • When matching grants projects require beneficiaries to save at a financial on Farmers and institution, financial institutions can capture information on farmers and Investments the cash flow patterns and profitability. Lack of know-how • When matching grants projects include financial institutions (in particular on Agriculture when they are required or incentivized to fund part of the investment), financial Finance institutions can gain know-how on agriculture credit methodologies. Lack of Collateral* • When matching grants support the acquisition of assets, these assets can serve as collateral for current and future projects of beneficiaries. • By increasing the repayment capacity of beneficiaries, matching grants Risks reduce the risks of investments that financial institutions co-finance. • By screening a variety of projects, matching grants can signal project Costs viability to financial institutions, therefore reducing the appraisal costs for financial institutions. Lack of Long-term • Matching grants reduce the amounts of financing required from financial Liquidity institutions. Constraint addressed in the long term (and also potentially for non-beneficiaries) Constraint addressed only during the course of the project and for project beneficiaries *In this example, the lack of collateral is classified as a supply-side constraint—although collateral is required by most financial institutions for access to credit—because some financial institutions manage risks in a different manner. Lack of collateral is a constraint that can sustainably be addressed through matching grants, but only for project beneficiaries. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 7 However, matching grants may not always matching grants for productive alliances in LAC,22 be the most cost-effective instruments to help almost all productive alliance projects have aimed farmers invest in productive activities when rural to enhance producers’ access to commercial finance, financial markets are limited, and it is therefore but these efforts have rarely been successful. Such recommended that constraints to agriculture difficulties might be due to constraints affecting rural and rural finance are systematically assessed financial markets that could not be addressed through before setting up a matching grants project. matching grants on their own (e.g., regulatory issues Indeed, some constraints to agriculture and rural that prevent financial institutions from making loans finance may be addressed on a more sustainable to groups of producers, etc.). The WBG has recently basis through other instruments (e.g., high risks, designed an agriculture finance diagnostic tool23 costs, and lack of long-term liquidity), while which aims at providing guidance to governments others cannot be addressed with matching grants on key constraints and opportunities for the at all (e.g., policy and regulatory environment). development of agriculture and rural finance. Such These alternative instruments are detailed in an approach may help policy makers make more figure 2 below. According to the WBG analysis of effective choices of public instruments to support Figure 2. Agriculture Finance Constraints that Can be Addressed with Alternative Instruments Constraints That Can be addressed with alternative instruments Lack of Know-how on • Offer technical assistance to develop lending methodologies with improved Agriculture finance risk management. • Develop alternative forms of collateral recognized by central bank regulations Lack of Collateral* including warehouse receipts systems (benefits all segments of the economy). • Establish a Partial Credit Guarantee (sustainable). Risks • Support the development of agricultural insurance (might require long-term subsidies). • Offer technical assistance to develop lending methodologies with improved risk management (sustainable). Costs • Promote the use of digital finance/agent banking (sustainable). • Subsidize start-up costs to open branches in rural areas (might require long-term subsidies). Lack of Long-term • Support the mobilization of long term savings (sustainable). Liquidity • Support credit lines for long-term funding (temporary). • Revise elements in the policy and regulatory environment that may Policy and Regulatory hinder the supply of financial services ( e.g., interest rate cap, loan Environment forgiveness programs etc.). Constraint that can sustainably be addressed by matching grants Constraint that can be temporarily addressed by matching grants Constraint that cannot be addressed by matching grants * Lack of collateral is a constraint that can sustainably be addressed through matching grants, but only for project beneficiaries. 1. WHY USING MATCHING GRANTS AND WHY ENGAGING THE FINANCIAL SECTOR MAY MATTER 8 agriculture finance. Similarly, key constraints to (sustainability). “Fully bankable” projects can be agricultural competitiveness may lie in the broader defined as projects that have sufficient collateral, policy environment related to land regulation, trade future cash flow, and high probability of success to policies (e.g., imports restrictions), and agricultural be acceptable to institutional lenders for financing. support policies (e.g., input subsidies only available “Potentially bankable” projects are projects that to a specific segment of the population creating have growth potential but do not fully present distortions for other market players). Such constraints these features.24 Projects that may be supported by cannot be addressed with matching grants. matching grants include bankable or “potentially bankable”25 projects undertaken by segments with 1.2.2 Matching grants for whom and for limited or no access to financial services. On the what? Balancing sustainability with other hand, segments that already have access to additionality financial services (e.g., SMEs) should only be Matching grants should be designed in a way supported for potentially bankable projects. This that excludes both fully bankable projects “do no harm” approach aiming at avoiding (additionality) and nonviable projects misallocation is represented in figure 3 below. Figure 3. Matching Grants and Financial Services—The “Do No Harm” Approach Who? Segment type Segments No Additionality with Access to = Financial Services No Intervention No Sustainability Screening-out mechanism: Clear identification of market failure, limited grant size Segments with = Limited Access to “High-leverage” Matching Grants Eg., Low level of matching, commercial credit incentivized Financial Services No Intervention or required. Segments with “Low-leverage” Matching Grants No Access to Eg., High level of matching, promotion of savings behavior, Screening-out mechanism: support to legal formalization, preparation of business plans Financial Services Professional MG committee, and financial accounts, support to acquisition of income- Financial institutions included generating assets. in MG committee. What? Project type Non-bankable Potentially Bankable Projects Bankable Projects Projects HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 9 Balancing additionality with sustainability also while segments with very limited access to financial requires introducing different design features services should receive “low-leverage grants,” which depending on segment type and project type. offer higher levels of matching but gradually pave Segments that already have access to financial the way for higher financial inclusion. Such design services should be granted “high-leverage grants,” features are described in more detail in section 2 (see which require high levels of financial discipline, boxes 2 and 3). 1. WHY USING MATCHING GRANTS AND WHY ENGAGING THE FINANCIAL SECTOR MAY MATTER 10 2. WBG Experience with Matching Grants in Agriculture and Drivers of Success Summary: The quantitative analysis below builds on the recent WBG analysis of 106 matching grants projects for private sector development26 by focusing on the 21 agriculture projects. It describes the specificities of agriculture matching grants compared to other sectors and analyzes whether specific features are associated with success. The referenced report analyzed all WBG active and closed lending operations including a matching grant component, and it assigned “implied ratings” for the matching grant component based on information provided in Implementation Completion and Results reports (ICRs) and Independent Evaluation Group (IEG) evaluations. Among the sample of 21 agriculture projects, most of the quantitative analysis addressing drivers of success focuses on the 15 closed projects which have implied ratings.27 Some of the analysis also includes an additional sample of seven projects which are not specifically focused on agriculture but include agriculture as one of their priority sectors. This analysis shows that the proportion of matching grants projects supporting agriculture has significantly increased in the 2000s, and that most agriculture projects have focused on the Africa region, although Latin America and the Caribbean is the first region by volume of matching grants components. Matching grants for agriculture are generally more successful and larger than outside of agriculture. Two notable specificities of agriculture projects are that all of them allow groups to benefit from matching grants,28 and that they allow the purchase of equipment.29 While the eligibility of groups can yield several advantages, allowing the purchase of equipment has been the subject of much debate. The provision of technical assistance,30 the availability of various levels of matching depending on beneficiary type or activity type, and the linkage of matching grants with an “access to finance” component seem to be important features for agriculture matching grants projects. Although no causal relationship is established, these modalities appear to be emerging good practices. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 11 2.1 Use of Matching Grants seems to have declined in more recent years (figure in Agriculture and Regional 4). Over the period 1996–2015, most agriculture Distribution at the WBG projects have focused on the Africa region, although The proportion of matching grants projects Latin America and the Caribbean is the first supporting agriculture has significantly increased region by volume of matching grants components in the 2000s, although the number of such projects (figure 5). Figure 4. Approval of World Bank Projects with Matching Grant Component (Fiscal Years) 35 30 Number of Projects 25 20 15 10 5 0 1991-1995 1996-2000 2001-2005 2006-2010 2011-2015 Agriculture Private Sector Development Figure 5. Number and Volume of Projects by Region 10 600 8 500 Number of Projects Volume of Projects 6 400 4 200 2 100 0 0 AFR LAC ECA SAR EAP Volume of Matching Grants (US$m) Number of Projects Note: AFR = Africa; LAC = Latin America and Caribbean; ECA = Europe and Central Asia; SAR = South Asia; EAP = East Asia and Pacific. 2. WBG EXPERIENCE WITH MATCHING GRANTS IN AGRICULTURE AND DRIVERS OF SUCCESS 12 There is no apparent learning curve in the design projects in the early 2000s. However, such a result of matching grants projects, as most recent might also be linked to the fact that indicators are projects do not have higher ratings on their more sophisticated over time and new variables get matching grants component than older projects. measured to assess success. The list of 15 closed Such a result is unexpected given that several good matching grants projects in agriculture is provided practices were established for matching grants in table 1 below. Figure 6. Success Rate by Year of Approval (Fiscal Year) 2006–2010 9 Number of Projects Year 2001–2005 5 1996–2000 1 0% 20% 40% 60% 80% 100% Share of Projects N/A Moderately Satisfactory Satisfactory Highly Satisfactory HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 13 Table 1. List of 15 Closed Matching Grants Projects in Agriculture Project Project Fiscal Year Country Matching Expected number Is there an Percent of match Implied MG ID Name Approved grant fund of beneficiaries access to (%) (fill in) component amount finance rating (US$m) component (number) (Y/N) P048505 MX- 1999 Mexico 343 “Irrigation: 33,000 No 50% for small Satisfactory Agricultural small individual farmers, 70% for Product producers poverty targeted Dairy: 10,000 rural development groups and 51,000 program producers Improed pasture: 110,000 producers 750,000 poor and small producers” P076467 IN: Chatt DRPP 2003 India 53 “20k community No Moderately 95% investment projects, Satisfactory 2k Panchayat (village) plans supported” P63622 NG-Fadama 2004 Nigeria 58.2 No “90% for rural Satisfactory SIL 2 (FY04) infrastructure development (beneficiaries to contribute 10% in cash or kind) 60% for Productive Asset Acquisition (increased to 70% during implementation)” P084792 IN-Assam Agric 2005 India 37.8 “80k groups of No 50% for irrigation Satisfactory Competitiveness 3-4 farmers for and mechanization irrigation projects (iniitally 30%0, 2.2k groups of 70% for drainage, 10-20 farmers for 50 to 90% for mechanization fisheries projects 15k farm families for micro-watershed drainage projects” P049721 Agriccompeti- 2005 Kazakhstan 26.69 800 subprojects No 40% for post- Moderately tiveness harvest infra Satisfactory projects, P104567 CO-Second Rural 2008 Colombia 24.8 300 PP with No 40 Satisfactory Productive 25,300 farmers Partnerships P064918 PA Rural 2007 Panama 19.8 70 business plans No 90% max Productivity of rural producer Satisfactory (former 2nd associations, (association Rur Po) representing provides minimum 5,000 small-scale 10% in cash or producers in kind) 2. WBG EXPERIENCE WITH MATCHING GRANTS IN AGRICULTURE AND DRIVERS OF SUCCESS 14 Table 1. List of 15 Closed Matching Grants Projects in Agriculture (continued) Project Project Fiscal Year Country Matching Expected number Is there an Percent of match Implied MG ID Name Approved grant fund of beneficiaries access to (%) (fill in) component amount finance rating (US$m) component (number) (Y/N) P108885 VN-Agriculture 2009 Vietnam 10.6 100 partnerships No 40% Moderately Competitiveness Satisfactory Project P081704 ML-Agr Compet & 2006 Mali 9.9 550 Yes 67% Highly Diversif (FY06)- Satisfactory (PCDA) P096105 SL-Rural Dev & 2007 Sierra 8 No 75% for Priv Sec Dev SIL Leone domestic market improvement component, 50% for agricultural export promotion, 90% for support to farmers associations P087925 BO Land for 2008 Bolivia 7.8 2,200 families Yes 80% Moderately Agricultural Dev Satisfactory P070063 ZM-Agr Dev 2006 Zambia 3 40k beneficiaries, Yes “50% (Extension Satisfactory Support Program 40 projects and technology (FY06) development) 60% (Studies and pilot) 75% (Support to smallholder producer organizations)” P049724 Agribusiness 2005 Kyrgyz 1.3 Yes 30% match to Satisfactory & Marketing Republic cooperatives, the other 70% loan from PFIs who administer program - match only paid after loan is repaid P110588 Sudan Gum Arabic 2010 Sudan 0.75 30 producer No 33% for private Satisfactory Export Marketing associations companies and Projec 67% for public agencies/ producer associations P083609 SN-Agr Markets & 2006 Senegal No Variable for small Agribus Dev producers and (FY06) SMEs. Business partnerships: 80% for smallholders, 50% for SMEs. Irrigation: 50% for family-farms, 20% for SMEs. Red meat 50% HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 15 2.2 Characteristics of Agriculture However, ratings reflect achievements of initial Matching Grants Projects and objectives rather than long-term impact. Indeed, Success Factors ratings strictly reflect the achievements of initial objectives, which often have a limited scope and do 2.2.1 Performance of matching grants not take into account the improvement of access to projects in agriculture compared to other financial services (see section 3.1). In addition, these sectors objectives are sometimes only partially achieved.31 Matching grant projects for agriculture generally have higher ratings than non-agriculture projects. 2.2.2 Size of matching grants projects in Indeed, among rated matching grant projects for agriculture compared to other sectors agriculture, 73% had ratings for the matching grant Agriculture projects have matching grants component of satisfactory and above, compared to components which, on average, are more than 47% for non-agriculture projects. five times larger than in other sectors. Indeed, the average fund amount is US$46 million for closed When including in the sample matching grant agriculture projects, compared to US$8 million for projects which target several specific sectors, closed non-agriculture projects. Such a difference including agriculture, the difference in ratings is linked to the fact that most agriculture matching becomes lower. “Broad agriculture” projects are grants allow the purchase of equipment, which is matching grant projects that are not solely focused described below in subsection 2.2.4 on eligible on agriculture but target a few specific sectors (e.g., expenses. As a result, although agriculture projects construction, tourism, etc.) which include agriculture. account for a small portion of total number of closed Among rated “broad agriculture” matching grant projects (25%), total grant financing dedicated to projects, 56% had ratings for the matching grant these projects reaches almost twice the volume of component of satisfactory and above, compared to grant financing outside of agriculture. 50% for other projects. Figure 7. Success Rate of Core Agriculture vs Non-agriculture Projects Non-agriculture 56 Number of Projects Agriculture 15 0% 20% 40% 60% 80% 100% Share of Projects N/A Unsatisfactory Moderately Unsatisfactory Moderately Satisfactory Unsatisfactory Highly Satisfactory 2. WBG EXPERIENCE WITH MATCHING GRANTS IN AGRICULTURE AND DRIVERS OF SUCCESS 16 Figure 8. Success Rate of Broad Agriculture vs Non-agriculture Projects Non-agriculture 51 Number of Projects Broad Agriculture 20 0% 20% 40% 60% 80% 100% Share of Projects N/A Unsatisfactory Moderately Unsatisfactory Moderately Satisfactory Unsatisfactory Highly Satisfactory Figure 9. Volume of Matching Grant Components by Sector Number of Projects 14 56 700 650 600 Grants in US$ Million 500 448 400 300 200 100 46 8 – Agriculture Non-agriculture Total Sum of Matching Grant Fund Amount (US$m) Average Sum of Matching Grant Fund Amount (US$m) HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 17 2.2.3 Rationale and objectives Using matching grants for equipment has often The 15 closed agriculture projects used a total been considered both unjustified and risky. of 31 indicators, 60% of which were output Indeed, as argued by Phillips,32 the fact that a market indicators such as number of funded projects or failure in financing of fixed assets is less likely than beneficiaries. Table 2 below presents the frequency in financing of know-how, the high appropriability of use of each of the indicators. of the return on physical assets, and the limited public good aspect would traditionally not justify 2.2.4 Eligible expenses a subsidy. The paper therefore recommends All projects in the sample allowed the purchase of “unbundling” equipment loans by offering grants for equipment through matching grants. Agriculture know-how with a public goods element combined matching grants include a variety of eligible expenses, with commercial credit for bankable equipment including fixed capital, working capital, and investments. Another concern commonly identified technical assistance (both to prepare and implement with equipment is the risk of abuse, as equipment business plans). Allowing the purchase of equipment could be resold for profit. is quite unusual (31%) for non-agriculture projects. However, agricultural equipment might often This feature is due to the fact that while investments be considered a semi-public good, and there are in equipment (e.g., irrigation infrastructure, storage effective ways to reduce the risk of grant misuse or processing facilities) are some of the investments and equipment resale as part of matching grants most needed by farmers and agricultural SMEs, projects. Indeed, it can be argued that investments there is generally no commercial funding available in fixed assets are often semi-public as they are for them due to their long-term and risky nature. For often allowed for village groups and cooperatives, instance, the Mali Agriculture Competitiveness and and also can have spillover effects on the economy Diversification Project (P081704) mainly supports (generating higher demand and higher supply of investments in irrigation equipment. credit for equipment). In addition, specific types Table 2. List and Frequency of Use of M&E Indicators for Agriculture Matching Grants Projects Output Outcome Beneficiary-level Financial Inclusion Impact Objective • Number of • Share of beneficiaries • Increased sales 4 • Beneficiaries funded projects/ satisfied with the grant 2 • Increase in the maintain a beneficiaries 3 • Increase in land quality of as measured system of 1 • Value of dollars under perennial crops by price premium accounts 1 disbursed 2 by organized producers of produce 1 • Revolving fund • Number of assets in the project area 1 • Increased productivity 1 in continuous purchased with • Number of productive • Increased income/ rotation 1 the grant 2 alliances created 1 profit 1 • Percentage of women benefiting from the matching grant 1 2. WBG EXPERIENCE WITH MATCHING GRANTS IN AGRICULTURE AND DRIVERS OF SUCCESS 18 of climate-smart investments can also be justified all crops. Indeed, working capital requirements may by their positive externalities on the environment. be much higher than investments for some crops. Moreover, there are effective ways to mitigate Matching grants projects in agriculture generally risks associated with equipment. These include (1) support two types of activities: investments in ensuring that the equipment is linked to a specific infrastructure and development of productive objective of the project and provides value to alliances. Productive alliances are defined as beneficiaries, (2) ensuring good practices within the partnerships between producers or between producers Project Implementation Unit (PIU) through tight and buyers which can help farmers upgrade their selection, training, and supervision of PIU members, production facilities and skills to strengthen their (3) facilitating supervision of beneficiaries through linkage to the market.35 Such projects may also an up-to-date database of beneficiaries, including support partnerships between farmers and input their names, address, photo ID, and GPS location, suppliers, so that farmers can obtain better prices (4) transferring funds to providers of goods and and more stable market relationships. Productive services rather than beneficiaries (once beneficiaries alliances have been widely promoted in LAC in have paid their share of the investment and show the past few years and are now being experimented receipts), and (5) creating accountability mechanisms with in other regions. For instance, the Panama among beneficiaries, for instance through a strong Rural Productivity Project (P064918) supported communication plan and visibility through local rural producer associations which had an alliance radio and television. with at least one agro-processor, wholesaler, or other Assets that are accepted as collateral by financial commercial partner. Eligible expenses included fixed institutions, such as land, are sometimes excluded capital (plant and equipment, minor infrastructure), from matching grants so as not to crowd out working capital, and technical assistance. commercial credit. For instance, the Bolivia Land for Agricultural Development Project (P087925) 2.2.5 Groups of beneficiaries excludes land purchase from eligible expenses While private sector development projects for matching grants.33 Similarly, the Angola Local sometimes restrict eligibility to single firms,36 Development Project (P105101) excludes from all of the agriculture projects in the sample eligible expenses all assets that commercial banks allow groups of farmers or SMEs to apply for a accept as collateral. common project or to benefit jointly from business development services (BDS). This modality has the Some projects cap the amount of working capital advantage of fostering linkages between groups, as which can be supported through matching well as reducing program administration costs when grants. For instance, the Angola Agriculture individual grants are small. Such an arrangement can Commercialization Project (P159052) considers sometimes take the form of a productive alliance capping working capital expenditures at 25% of the between a lead agribusiness and producers groups. grant amount. The rationale for such an approach is that, as one-off interventions, matching grants should 2.2.6 Levels of matching and availability promote investments that can improve agriculture of different levels of matching profitability in a sustainable manner rather than A majority of all projects in the sample (60%) support recurring expenditures. However, experience offered a level of matching of 50% or above, and from Niger,34 where working capital was capped at a majority of projects (60%) offered different 20%, shows that such practice may not be suited to levels of matching depending on beneficiary or HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 19 activity. While most projects that had various levels In general, literature suggests that the level of of matching offered levels of matching in the same matching should be as low as possible so as to range (above or below 50%), two projects in the encourage ownership and commitment from sample had a level of matching for small producers beneficiaries; however, “as low as possible” can above 50% and a level of matching for SMEs below vary widely depending on segments and country 50%. These two projects are displayed as “variable context. Technical guidance from IFAD, for instance, for small producers and SMEs” in the figures below. suggests that matching grants for private or semi- private benefit should be in the range of 10% to 60% Offering a variety of matching levels depending of the investment.37 The sample of WBG agriculture on beneficiary or activity type seems to be linked projects shows levels of matching going from 30% to positive outcomes. A majority of rated projects (Kyrgyz Republic) to 95% (India Chhattisgarh). which had various levels of matching (86%) have The impact evaluation of the matching grants in ratings of satisfactory and above, compared to 50% India, however, indicates that 30% of common for projects which had a single level of matching. interest groups supported with matching grants Though the sample size is small, 83% of rated failed to function effectively and were considered projects that had a level of matching above 50% unsustainable at the end of the project, which have ratings of satisfactory and above (compared raises the question of whether the level of required to 50% for projects that had a level of matching contribution (5% cash) was sufficient to ensure under 50%). This might be due to a higher level project ownership. of disbursements for matching grants projects with high levels of matching. Figure 10. Success Rate by Availability of Various Levels of Matching Yes 9 Number of Projects No 6 0% 20% 40% 60% 80% 100% Share of Projects n/a Moderately Unsatisfactory Unsatisfactory Highly Satisfactory 2. WBG EXPERIENCE WITH MATCHING GRANTS IN AGRICULTURE AND DRIVERS OF SUCCESS 20 There appears to be no connection between the account. One outlier seems to emerge in the case level of matching with the penetration of rural of the Kyrgyz Republic, where the penetration of financial services at country level. Penetration of rural financial services was very low (less than 2%) rural financial services is estimated by Findex data and where the level of matching offered to rural on the percentage of adults in rural areas with an cooperatives was significantly low as well (30%).38 Figure 11. Success Rate by Level of Matching Variable for Small Producers 2 and SMEs Number of Projects Under 50% 4 50% and Above 9 0% 20% 40% 60% 80% 100% Share of Projects n/a Moderately Satisfactory Satisfactory Highly Satisfactory Figure 12. Level of Matching by Penetration of Rural Financial Services 2 Variable for small producers and SMEs Number of Projects Under 50% 4 Over 50% 9 0% 20% 40% 60% 80% 100% Share of Projects Less Than 15% More Than 15% More Than 30% HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 21 2.2.7 Provision of diagnostics and (figure 14). Eighty percent of “first come, first technical assistance (TA) served” projects have ratings satisfactory and A majority of projects (73%) offered a form of above, but given the small number of “competitive” diagnostic or technical assistance which could projects, it is hard to determine whether the choice of take place before and/or after the matching grant selection mechanism brings any systematic benefits. application. Such support includes mandatory initial diagnostic to verify eligibility, TA to prepare 2.2.9 Access to finance component and sound business plans provided freely or for a fee, combining matching grants with other a complementary project component to create a instruments pipeline of applicants, and continuous provision Although matching grant components are of TA to support beneficiaries from application generally part of larger projects with other to implementation. The recent WBG review of components, a minority of agriculture matching 106 projects39 shows that this feature is the design grants projects include an access to finance modality which seems to be most often correlated to component. Only 27% of agriculture matching grants positive outcomes, and our sample also shows that projects included an access to finance component, 70% of rated agriculture matching grants projects compared to 34% for all WBG projects, including which include diagnostics and/or TA are rated non-agriculture projects. This specific component satisfactory and above. generally includes setting up a line of credit for financial institutions, establishing a Partial Credit 2.2.8 Selection mechanism Guarantee (PCG), and providing technical assistance A vast majority of projects (86%) used a “first to financial institutions, but it can also include come, first served” selection mechanism as support to an equity financing, or promoting of risk opposed to a “competitive” selection mechanism mitigating financial instruments.40 By addressing the Figure 13. Success Rate by Provision of Diagnostic Yes 11 Number of Projects No 4 0% 20% 40% 60% 80% 100% Share of Projects N/A Moderately Satisfactory Satisfactory Highly Satisfactory 2. WBG EXPERIENCE WITH MATCHING GRANTS IN AGRICULTURE AND DRIVERS OF SUCCESS 22 Figure 14. Success Rate by Selection Mechanism Yes 13 Number of Projects No 2 0% 20% 40% 60% 80% 100% Share of projects N/A Moderately Satisfactory Satisfactory Highly Satisfactory root cause of the lack of agriculture and rural finance, Such a feature also seems to be associated with such a feature can be very effective at ensuring the positive outcomes, as 75% of rated projects sustainability of matching grants projects. which included an access to finance component had ratings satisfactory and above (figure 15). The volume and scope of the access to finance However, as we will show in section 3, including an complementing matching grants can vary access to finance component is not necessarily the widely depending on projects. For instance, the only way to ensure linkages with the financial sector. Zambia Irrigation Development and Support Project (P102459) included a rather small access to finance Complementing matching grants with other component (4% of budget dedicated to matching instruments may indeed offer a holistic solution grants) aimed at improving access to long and short- to the various constraints that often hamper term financing. This component included a line of agriculture finance, and help combine short- credit for financial institutions, technical assistance term with long-term benefits. Agriculture and to financial institutions, and technical assistance rural finance is often constrained by a variety of to agricultural cooperatives to ensure linkages factors, which often need to be addressed in parallel with financial institutions. By contrast, the Zambia to effectively increase access to and usage of Agricultural Development Program (P070063) financial services by farmers and agricultural SMEs. included an access component which had twice As described in section 1, matching grants can be the budget of the matching grants component. This effective at addressing several of these constraints, component also included a line of credit as well as both on the demand side (e.g., lack of willingness technical assistance to financial institutions. to invest, lack of trust, etc.), and on the supply HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 23 side (e.g., lack of information on farmers, lack of Indeed, setting up a sustainable PCG or reforming collateral, etc.), but they fail to address other key the regulation of interest rates may generate profound constraints such as policy and regulatory constraints, changes in rural financial markets, but such projects risks, or lack of liquidity. In addition, complementing might take time. By contrast, matching grants are a matching grants with other instruments may allow temporary subsidy that can quickly help underserved combining short-term with long-term benefits. segments access resources for their projects. Figure 15. Success Rate by Existence of Access to Finance Component Yes 4 Number of Projects No 11 0% 20% 40% 60% 80% 100% Share of Projects N/A Moderately Satisfactory Satisfactory Highly Satisfactory 2. WBG EXPERIENCE WITH MATCHING GRANTS IN AGRICULTURE AND DRIVERS OF SUCCESS 24 3. Promoting Access to Finance Through Grants and the Four Roles of Financial Institutions Summary: This section is based on a qualitative analysis of six case studies stemming from projects reviewed in section 2, as well as three other relevant case studies from the literature review and interviews with experts. In particular, all Project Appraisal Documents, ICRs, and IEG reviews of agriculture matching grants projects have been reviewed to identify potential linkages with financial institutions. This analysis shows that none of the reviewed projects included the improvement of access to financial services as one of their objectives, which calls into question the sustainability of supported investments. However, several projects include some form of linkage with the financial sector, through the inclusion of an “access to finance” component and/or through specific design features. These specific design features have taken four major forms: (1) financial institutions are deposit takers, (2) financial institutions are funders, (3) financial institutions are managers of grants, or (4) financial institutions are advisors. This section analyses the advantages and challenges associated with each of these four roles. 3.1 Matching Grants Projects and Financial Inclusion None of the reviewed projects included the improvement of access to financial services as one of their objectives, which calls into question the sustainability of supported investments. Indeed, while most projects identify the lack of access to credit as the market failure justifying the use of matching grants, none of the projects include the improvement of rural financial markets as one of their project development objectives. The challenges associated with such an approach are highlighted in the ICR for the Vietnam Agriculture Competitiveness Project (P108885): “There are limits on the scalability of matching grants and it is important to build in an ‘exit’ strategy, in the form of improved farmer group access to commercial financial services. Even if the Partnership scheme does not involve distinctive activities to link the farmer organizations with commercial banks, a project should at least prepare farmer organizations to access credit once project support is completed.” Several projects, however, include some form of linkage with the financial sector, through the inclusion of a specific access to finance component and/ or through specific design features. The inclusion of an access to finance HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 25 component in the project refers to the setting up are required or incentivized to provide credit to of complementary instruments (e.g., PCG, lines finance part of the activities, (3) financial institutions of credit) and the introduction of legal reforms are involved in the management of grants, including (see section 2.2.9). the appraisal and disbursement of grants, and (4) financial institutions advise beneficiaries in the Some projects have promoted linkages with preparation of their business plans. financial institutions, through specific design features which can take four major forms. As These four models can also be summarized in two illustrated in figure 16 below, by order of frequency types of approaches: “high-leverage” matching among projects, the four major roles that financial grants blended with credit, for segments with institutions can play in such projects are as follows: some access to formal financial services; and (1) financial institutions are deposit takers, as “low-leverage” matching grants, incentivizing beneficiaries are encouraged to save (a specific savings for segments with no access to financial amount and/or at a specific frequency) from the services. These two approaches are described further proceeds of their activities, (2) financial institutions in boxes 2 and 3. Figure 16. Financial Institutions have Four Potential Roles as Part of Matching Grants Projects r viso Ma na Ad ge Core Light r Mexico Kyrgyz Republic Vietnam Burundi Incentivized Required Revolving fund Individual account Colombia Kyrgyz Republic India Angola (63% - grant conditional Chhattisgargh (required monthly on loan repayment) (required 10% for second deposits reaching tranche disbursement) 20% over the course Honduras of project) (30%) Colombia er ak (70% over the course Fu India Assam of project) T nd sit (requirement dropped) er Nigeria po De (10% annually) Note: Countries displayed in brown refer to projects that are not part of the initial project sample but stem from the literature review or interviews with experts. In some projects, financial institutions are in more than one of the four categories. Financial institutions may also move over time from one category to another. 3. PROMOTING ACCESS TO FINANCE THROUGH GRANTS AND THE FOUR ROLES OF FINANCIAL INSTITUTIONS 26 Box 2. The Experience with High-leverage Matching Grants in Europe and Central Asia (ECA) The model which blends matching grants with commercial credit for rural beneficiaries was implemented successfully in the mid-1990s and early 2000s in various countries in ECA. In this model, the level of matching is relatively low (about 20–30%), and grants are granted to beneficiaries who obtain access to a loan. In some projects, in case of non-repayment of the credit portion, the grant is canceled and becomes a loan. Countries that have experimented with high-leverage grants include the Kyrgyz Republic (see details in section 3.3), Moldova (matching grant for start-ups, for groups of farmers, or after agricultural shocks), Latvia (matching grant for start-ups), and Uzbekistan (matching grant for silk producers). This model is currently being replicated beyond ECA, for instance in India (project P157702 under preparation in Tamil Nadu). In this approach, financial institutions generally play the role of both funders and “core managers” of grants. In general, high-leverage matching grants are combined with other instruments to promote rural finance such as credit lines and Partial Credit Guarantees. Such a model can potentially help financial institutions serve rural markets. It is suitable to segments that have some access to financial services and have potentially bankable projects that banks are nonetheless reluctant to serve, as they are not aware of market opportunities. For instance, farmer groups may be composed of farmers who individually have access to finance, but still have difficulties accessing finance as a group. This model reduces collateral requirements for beneficiaries, but does not eliminate them completely. Indeed, due to the uncertainty on the value of rural assets, lenders generally have collateral requirements that are above the value of the asset. In addition, due to a timing issue (the asset needs to be registered before the loan is approved), beneficiaries need to provide collateral for the time period between loan approval and asset registration. Pros and cons for financial institutions: • Advantage: The matching grant design provides strong incentives for clients to repay on time, which acts as an implicit guarantee. In addition, such a model creates an opportunity for financial institutions to serve a new segment that has benefited from technical assistance prior to the loan application. • Disadvantage: The grant cannot be used to cover losses for the lender as the grant portion is returned to the project management unit (lender has to rely on collateral). In addition, interest income is lower if beneficiaries repay loan portion on time. Positive sustainable impact: This model increases access to finance for beneficiaries, as repayment rates are high, and most beneficiaries manage to access loans on their own after the project closes. In the case of Moldova Rural Investment and Services Project (P060434), the high-leverage matching grants led to over 700 enterprises created and financed, creating on average four jobs each, and an average increase of 55% in salary income. The six participating financial institutions increased their lending to rural clients by 39% as a share of their portfolio while maintaining an acceptable recovery rate of 96%. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 27 Box 3. The Experience with Low-leverage Matching Grants in Africa Low-leverage grants are matching grants with high levels of matching (above 50%) which are designed in a way that facilitates the entry of beneficiaries into the formal financial system. Such an approach includes business development services, incentives to save at a financial institution, (e.g., a specific amount, at a specific frequency), and support to business formalization. In this approach, financial institutions generally play the role of deposit takers. Financial institutions may also play the role of “light managers” or advisors. This approach has been successfully implemented with individual savings accounts in Angola (see details in section below). It has also been used to promote collective savings in a revolving fund (see Nigeria example below). Facilitating the path to financial inclusion may require additional support measures both to beneficiaries and partner financial institutions. Such support may include financial literacy trainings for beneficiaries, as well as technical assistance for partner financial institutions to ensure that processes are in place to deal with this new segment of clientele (e.g., simplified processes for deposits, trainings for beneficiaries on how to access loans, etc.). Partnerships with financial institutions which offer no-maintenance-fee transaction accounts, or temporary financial support to waive account maintenance fees, may also be considered. Stronger support measures include setting up credit lines or PCGs for financial institutions. 3.2. Financial Institutions as The revolving fund model—used in Colombia, Deposit Takers Nigeria, and India—can help producer groups get Under this scheme, beneficiaries are required or better access to finance as well as their members.41 incentivized over the course of the project to set For instance, the Colombia Second Rural Productive aside—either individually or collectively—part Partnerships Project (P104567) required producer of revenues generated from the matching grant– organizations to set aside 70% of the grant received financed activities. The “deposit-taking” role for in a revolving fund. The objective of the revolving financial institutions is the most common role for funds was to allow producer organizations to financial institutions when engaged in matching continue their operations after the project, as well grants projects. Two different approaches have as build up their creditworthiness, in order to enable been experimented with: (1) each producer group them to obtain commercial financial credit. The sets aside money in a revolving fund, or (2) each impact evaluation shows that—compared to control individual sets aside money in a savings account. group producers—more beneficiary producers have This section builds on case studies from Colombia, obtained credit for productive investments, and Nigeria, India, and Angola (described in detail in more of them reinvest part of their net revenues in annex 1). their agricultural production. Similarly, the Nigeria 3. PROMOTING ACCESS TO FINANCE THROUGH GRANTS AND THE FOUR ROLES OF FINANCIAL INSTITUTIONS 28 Fadama III Development Project (P096572) However, setting up revolving funds among introduced a savings mechanism called Fadama producer groups requires caution and clear Users’ Equity Fund (FUEF). Indeed, the project communication. Indeed, such a design feature can required that 10% of the replacement value of the only work in contexts where producer groups are common assets used for income-generating activities well-managed and members trust that the revolving of Fadama User Groups (FUG) be saved annually fund will be used in a way that benefits them. It also (with effect from year 2). This feature served as a requires that groups have sound governance as well sustainability provision of the project and aimed as some expertise in managing loans and financial at facilitating the observed desire of participants accounts. When this is the case, farmer groups to continue investment after completion of the can grow into viable financial institutions, as the matching grant. With the project support, a total experience of Raiffeisen in Germany shows.42 In the of 37 Fadama Farmers Community Association case of Nigeria, one institution obtained a banking (FFCA) groups were created, one for each state and license; however, less than half of the state-level the Federal Capital Territory, with the objective of institutions created as part of the project were still in transforming some of the vibrant FFCAs into self- operation by the end of the project, which confirms sustaining institutions. Of them, seven institutions challenges related to financial capability and have generated enough capital and expertise to have governance of producer groups. In addition, in cases applied for a license to operate as a microfinance where most of the grant portion needs to be repaid in bank, and one of them obtained a banking license in a collective revolving fund, it might be confusing to 2015. Finally, the India Chhattisgarh District Rural label the support as a grant.43 Poverty Project (P076467) required community The individual savings account approach—used members to place 10% of the matching grant amount in Angola—can be an effective instrument to pave into a village fund (Apna Kosh) as a condition for the way for farmers and agricultural processors the release of the second tranche of the subprojects. to enter the financial system. For instance, the The fund was aimed at covering operation and Angola Local Development Project incentivizes maintenance costs and further village development each individual within beneficiary producer groups beyond the lifetime of the project. The IE showed to save 20% of the grant amount in a savings that project beneficiaries had more confidence in account over the course of the project, with monthly dealings with banks, had access to bigger loans, deposits.44 Beneficiaries receive small rewards for and displayed stronger savings discipline. Indeed, each monthly deposit made on time (phone cards beneficiaries, especially women, highlighted a with airtime). On the one hand, such requirements considerable increase in their confidence in dealing and incentives can foster savings discipline among with banks. While the IE did not show a significant beneficiaries. On the other hand, as these segments difference in the share of Common Interest Group build a transaction history, this requirement can foster (CIG) beneficiaries with bank accounts compared better knowledge of farmers and agricultural SMEs to control areas, it did find that more CIG members among financial institutions. The impact evaluation with bank accounts succeeded in taking loans shows that beneficiaries gradually fulfilled financial compared to the situation in control villages (48% institutions’ requirements through the project (e.g., v. 37%), and that in control areas more still used had an active account, had collateral, were formal money lenders. Also, the total amount of loans taken enterprises), and some of them managed to access by CIG households was 30% higher than the amount loans from financial institutions. borrowed by non-CIG households. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 29 Regardless of the choice of an individual or 3.3 Financial Institutions as collective approach, requiring matching grant Funders beneficiaries to save money can be an effective Under this scheme, matching grants beneficiaries instrument to pave the way for financial are either required or incentivized to secure a inclusion, but it also requires caution related to loan from a financial institution to cover part of feasibility and additionality. Requiring high levels the investments. The “funding” role for financial of savings might not be feasible, in particular for institutions is the second-most common role for farmers involved in long production cycles and/ financial institutions when engaged in matching or facing significant production risks. For instance, grants projects. This section builds on case studies in Colombia, only 50% of producer organizations from the Kyrgyz Republic, Honduras, India, and managed to reach the required recovery rate of Colombia (more detail is in annex 2). 70%. In particular, farmers involved in perennial crops (e.g., cacao, forestry, mango), which have In some cases, such as in Colombia, blending long gestation periods, had limited capacity to repay grants with commercial credit is only incentivized. quickly. In addition, unusually adverse weather Indeed, in the Colombia Second Rural Productive significantly affected a large number of beneficiary Partnerships Project (P104567), the levels of producers, who suffered significant losses during matching and ceiling grant amounts were higher the project and therefore were unable to repay for beneficiaries who managed to obtain credit.45 into their revolving funds. Such difficulties in For example, if the partnership obtained a credit of accumulating savings show that such a requirement Col$100,000 per household, the government financial might have the advantage of strengthening financial incentive could be increased by Col$100,000 per discipline without exposing project beneficiaries to household. This 1:1 relationship would be respected challenges associated with non-repayment of formal up to a maximum of Col$2 million per household. credit. However, by contrast, the fact that some Similarly, in all matching grants projects where beneficiaries manage to “repay” a large portion of the level of matching is low, beneficiaries have an their grant into a savings account or a revolving fund implicit incentive to seek commercial credit, as they raises the question of the additionality of matching might not be able to fund their required contributions grants. Indeed, it is possible that the 50% of producer from their own funds only. Furthermore, another way organizations that managed to obtain a recovery to incentivize commercial credit is to restrict grants rate of 70% could have financed their project from to expenses that financial institutions are reluctant to commercial credit. Such a model illustrates the finance (assets, technical assistance) and limit the use trade-off between the need to ensure additionality of matching grants for working capital so as to crowd (channeling grants only to segments who are not able in commercial credit.46 A recent matching grants to finance their investments through credit) and the project in Argentina also incentivized commercial need to promote sustainability (channeling grants in credit through (1) referrals to credit institutions when such a way that beneficiaries do not need grants for semi-capitalized family farmers present a subproject their future investments). to the PIU, and (2) varying levels of matching depending of the level of capitalization of farmers (see box 4). 3. PROMOTING ACCESS TO FINANCE THROUGH GRANTS AND THE FOUR ROLES OF FINANCIAL INSTITUTIONS 30 Box 4. Incentivizing Commercial Credit in Argentina for Climate-smart Agriculture Investments47 Context: Most vulnerable family farmers in Argentina have limited access to formal financing through banks or other financial institutions. In addition, some of the climate-smart agriculture (CSA) technologies are innovative and/or have significant positive externalities, so farmers require an incentive to try them. Targeted investments and segments: Matching grants are offered to encourage farm-level adoption of validated CSA technologies and risk management instruments. The project also includes special incentives for matching grants to groups that include women farmers (additional scores for prioritizing matching grant proposals and higher percentage of matching grants). Linkage with financial institutions: Matching grants are intended to help vulnerable family farmers integrate into the formal financial system. Key features to facilitate financial inclusion are as follows: (1) depending on the level of capitalization of farmers, the level of matching varies and the process for obtaining matching grants may include banks; and (2) commercial banks, which may not be familiar with complex CSA technologies, receive support from the PIU for such complex investments. The mechanisms for obtaining and disbursing grants for family farmers to invest in CSA technologies differ depending on whether the technology is simple or complex: • Some CSA and risk management practices and technologies consist of specific, simple solutions that are well known and relatively easy to evaluate, such as irrigation systems, direct sowing, and nets to prevent hail from damaging crops. Farmers can apply for such grants directly at a bank, or they may be referred to commercial banks if they have applied to the PIU and lack sufficient resources to provide the counterpart funding for these investments. If banks cannot finance the counterpart resources requested by these farmers, they may approach local or provincial funds or other non-bank sources of credit. • For complex CSA technologies and practices, which rely on multiple complementary practices and inputs to produce farm-level changes, banks and other formal financial institutions generally lack the capacity or experience to advise or assess them. These interventions are prepared by farmers supported by expert consultants, and submitted for evaluation and approval to the PIU with tranches of the grant disbursed through the beneficiaries’ bank accounts. In piloting, local banks may opt to provide financing if beneficiaries require a loan to pay the counterpart portion of the investment. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 31 In other cases, such as in the Kyrgyz Republic commercial credit can potentially help beneficiaries and Honduras, blending grants with commercial access new types of loans (e.g., longer maturity, credit is required. larger volume, etc.). However, if beneficiaries’ financial capacity is overestimated, such an approach The Kyrgyz Republic Agribusiness & Marketing might jeopardize project disbursements and/or create Project (P049724) had the highest requirements challenges for borrowers. Indeed, the India Assam for commercial credit, as it required that 63% Agricultural Competitiveness Project (P084792) of the investment is provided as a loan from a initially required that commercial banks cover 50% financial institutio48 and that the matching grant of the investments; but it had to be restructured due component is converted into a loan if the loan to slow disbursements, and it dropped the mandatory portion is not repaid in full and on time. In this case commercial bank linkage. Moreover, requiring the matching grant instrument was combined with segments that are excluded from financial services three other instruments aimed at increasing the to use commercial credit could lead to defaults and supply of agriculture finance services: a credit line exclusion from the financial system.50 In addition, to participating financial institutions for agribusiness while such a challenge has not been documented lending, technical assistance to loan officers to in the impact evaluation of these projects, a model broaden the base of eligible borrowers, and training whereby financial institutions channel both grants of agricultural cooperatives to strengthen their and loans can potentially create confusion among management skills and allow them to apply for beneficiaries. In such an arrangement, it is important financing. While the project was small in scale,49 to raise awareness among beneficiaries about the it had a positive impact on the economic activities difference between grants and loans. It is also of beneficiaries (such as increases in sales, profits, important to provide financial institutions with and market share), and all beneficiaries were able technical assistance on credit risk management to reimburse their loans. In a similar approach, the and loan collection procedures. Indeed, experience Honduras COMRURAL Project (P101209) required with matching grants in Niger,51 where financial that subproject beneficiaries secure a loan from a institutions were required to provide 50% of financial institution, covering at least 30% of total investments, shows that financial institutions did subproject investment costs. The matching grant not have the financial and technical capacity to grant instrument was combined with a Partial Credit such loans, and that the quality of the agriculture Guarantee fund to increase the supply of agriculture credit portfolio declined as a result. finance services. As a result of their participation in the project, various producer organizations have received further loans. 3.4 Financial Institutions as Managers Involving financial institutions as funders can be Under this scheme, financial institutions an effective way to deepen financial inclusion of play a role—either “light” or “core”—in the segments who already have some access to formal management of matching grants. The “managing” financial institutions, but it requires a rigorous role for financial institutions is the third-most initial assessment of beneficiaries’ financial common role for financial institutions when engaged capacity as well as a transparent communication in matching grants projects. This section builds on plan. For segments who already have some access cases studies from the Kyrgyz Republic, Vietnam, to formal financial institutions (e.g., have an account and Burundi (more detail is in annex 3). and access to small loans), blending grants with 3. PROMOTING ACCESS TO FINANCE THROUGH GRANTS AND THE FOUR ROLES OF FINANCIAL INSTITUTIONS 32 Under the “light management” approach, by region, etc.) could also be an effective way to financial institutions play a role in the selection or capitalize on lessons learned and promote better pre-identification of matching grants beneficiaries, knowledge of agricultural investments among which can ensure that only financially viable financial institutions. projects are selected while building agriculture Under the “core management” approach, such finance knowledge among financial institutions as in the Kyrgyz Republic, financial institutions at the same time. For instance, in the Vietnam both select beneficiaries and channel grants, Agriculture Competitiveness Project (P108885) as which can simplify the investment process but well as in the Burundi Keeping Good Firms Alive also potentially create confusion for beneficiaries. & Well: Phased E-matching Grants to Tackle Debt Such a model makes sense in the Kyrgyz Republic Overhang and Recreate Credit Histories project,52 Agribusiness and Marketing Project (P049724), the evaluation committee for the matching grants where financial institutions bear most of the financial business proposals includes representation of the risk (63% of the investment). This approach can also commercial banking sector.53 In the Burundi case, speed up and simplify the investment process, as financial institutions are also involved at a more matching grants applicants only need to have their upstream level, during the pre-identification phase investment project approved by a financial institution of SMEs. Indeed, partner financial institutions are rather than applying both for a grant and for a loan. asked to identify some of their client SMEs that Such an approach may however generate confusion have overdue loans but that could potentially have among borrowers as regards the nature of the support their loans restructured and access new loans with they receive. matching grant support. Another design element is partial debt forgiveness from outstanding arrears. While this project is still in the early stage of 3.5 Financial Institutions as implementation, such an approach might be an Advisors innovative way to reintegrate in the formal financial Under this scheme, financial institutions advise system segments that have lost access to it due to matching grants applicants on the preparation exogenous price fluctuations and a general economic of their business plans. The “advising” role for crisis. In addition, including financial institutions financial institutions is extremely rare, and this in the matching grants committee offers the double section only builds on the example of one ongoing advantage of ensuring that only financially viable project in Mexico (detailed in annex 4). The Mexico projects are selected, while building agriculture Sustainable Production Systems and Biodiversity finance knowledge among financial institutions at Project (P121116) establishes commercial financial the same time. Indeed, instead of sharing knowledge institutions as key technical service providers which of agricultural investments and profitability only are hired to support the management of financial among members of a PIU, which is time-bound, services for each producer association as well as for this approach allows financial institutions to gain the preparation of business plans. valuable knowledge about the relative profitability of Such an arrangement might offer advantages various agricultural investments. While none of the both for financial institutions and matching grant projects include such a feature, publishing a briefing beneficiaries. Indeed, while this project is still under note to all financial institutions summarizing the implementation, such an approach might be an relative performance of matching grants–supported effective way to build knowledge among financial agricultural investments (by type, by value chain, institutions about farmers and agricultural SMEs, HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 33 while at the same time strengthening financial skills advice provided is perceived as suboptimal. It can among matching grants beneficiaries. This approach, also lead to a potential conflict of interest if financial however, exposes financial institutions to a reputation institutions both help prepare business plans and risk if financed projects are not successful and the then provide funding for the project. 3. PROMOTING ACCESS TO FINANCE THROUGH GRANTS AND THE FOUR ROLES OF FINANCIAL INSTITUTIONS 34 Conclusion and Recommendations WBG matching grants for agriculture are generally more successful than those outside of agriculture, but generally are not focused on long-term sustainability, both on their rationale and on their design: • Indeed, most projects identify the lack of rural finance as a sufficient rationale for matching grants, without fully identifying the market failure or whether other instruments might be more appropriate to unlock rural and agriculture finance. Low penetration of rural financial services might be due to a wide variety of constraints, and only a subset of these can be effectively addressed with matching grants. • Most projects do not include an access to finance component or a design feature promoting financial inclusion. Among 21 agriculture matching grants projects, only 6 engaged financial institutions through specific design features, and 6 included an access to finance component.54 • Several projects do not capture results effectively. The 15 closed agriculture projects used a total of 31 M&E indicators, 60% of which were output indicators such as number of funded projects or beneficiaries. Two main approaches have been used in matching grants projects to promote financial inclusion: high-leverage matching grants blended with credit for segments with some access to formal financial services, and low-leverage matching grants incentivizing savings for segments with no access to financial services. These two approaches can themselves be broken down into four roles for financial institutions, and each of them is associated with specific advantages and challenges: (1) deposit takers, (2) funders, (3) core or light managers, and (4) advisors. Available impact evaluations show that such projects are successful at improving agricultural income, but also improve access to finance in a sustainable way. For instance, a project in Kyrgyz Republic showed that among matching grants beneficiaries, 41% saw an increase in profit, 37% saw an increase in total sales, and 47% saw an increase in market share compared to before the investments. Furthermore, impact evaluation of a project in India, which required beneficiaries to save part of their proceeds at a financial institution, showed HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 35 that saving discipline among beneficiaries led to 2. If one of the identified market failures is the increased trust towards financial institutions and lack of access to finance for farmers and more access to loans, and that beneficiaries’ loans agricultural SMEs, improving access to were larger than non-beneficiaries’55 In addition, agriculture and rural finance should be one some projects have generated large spillover of the objectives of the project, and financial effects, such as a project (Colombia Second Rural sector experts should be involved in project Productive Partnerships Project, P104567) in which design. 11,000 rural households that were not matching 3. In order to assess whether matching grants are grants beneficiaries nevertheless adopted (at their the most cost-effective instrument to improve own expense) improved practices promoted through access to agriculture and rural finance, matching grants.56 constraints to agriculture and rural finance should Combining matching grants with other be systematically assessed through an agriculture instruments that support commercial credit can finance diagnostic, and various alternative prove impactful. A few matching grants projects instruments should be considered to replace or are combined with other instruments aimed at complement matching grants. addressing specific constraints to agriculture finance, 4. Matching grants design features should be such as lines of credit (to address lack of liquidity), determined carefully to foster linkages with Partial Credit Guarantees (to address exposure the financial sector and long-term impact: to agricultural risks), or technical assistance to financial institutions (to address lack of knowledge • Size of the grant and level of grant matching of agriculture finance). Such combinations can help should differ by type of beneficiary address structural constraints to agriculture finance (micro-enterprises and farmer groups, small while offering immediate investment opportunities to enterprises, medium enterprises) and by type of farmers through matching grants. However, impact investment (technical assistance, fixed assets, evaluations generally do not disaggregate the grant working capital) so as to ensure additionality impact from the impact of additional instruments, and sustainability. making it difficult to assess the respective role of • Beneficiaries’ contribution must be set high each instrument. Such an analysis may be an area for enough to ensure ownership and crowd in further research. commercial credit. Emerging good practices to improve the financial • For beneficiaries who have no relationships sustainability of agriculture matching grants with financial institutions, a path towards projects are as follows: financial inclusion through low-leverage 1. Before designing a matching grant project, a matching grants should be promoted. Such strong economic rationale must be established an approach would incentivize beneficiaries and market failures must be properly described to save part of the proceeds in an account at (e.g., lack of demand for or supply of business a financial institution, but also support legal development services, limited supply of financial formalization, preparation of business plans services, limited bankable demand of financial and financial accounts, acquisition of income- services, etc.). generating assets, etc. This model can be an effective instrument to pave the way for financial CONCLUSION AND RECOMMENDATIONS 36 inclusion but also requires caution related to or “light managers” should be considered. feasibility (in particular for farmers involved In addition, sharing with financial institutions in long production cycles or farmers facing a database of all agricultural investments and significant production risks) and additionality. their relative profitability could help address Moreover, additional caution is required when their knowledge gaps. savings are set aside in a collective revolving • Suggested M&E indicators related to fund. financial sustainability include the following: • For beneficiaries who have limited % of beneficiaries who have saved more than relationships with financial institutions, X% in their account by the end of the project, high-leverage grants should be considered. % of beneficiaries who have saved for the first Such an approach requires stronger financial time at a financial institution, % of beneficiaries discipline from beneficiaries because it requires who have maintained an active account by the or incentivizes blending of matching grants with end of the project, % of beneficiaries who have commercial credit. Opting for high-leverage established a track record with a value chain grants may be effective at deepening financial player, % of beneficiaries who have accessed inclusion but requires a solid assessment of loans for working capital/further equipment beneficiaries’ financial capacity and financial from a financial institution during the project institutions’ appetite. In particular, requiring period, % of beneficiaries who have accessed a mandatory share of credit might jeopardize loans on commercial terms after the project, project disbursements and/or create challenges % increase in share of agriculture in lending for borrowers. When financial institutions portfolio of participating financial institutions. are involved as funders, other instruments Other emerging practices for successful design of to facilitate credit should also be considered matching grants include the following: (technical assistance, PCGs, etc.) to ensure financial institutions’ participation. Moreover, • Matching grants projects should include involving financial institutions as “core technical assistance to beneficiaries, both for managers” of matching may make sense when preparing and implementing business plans. financial institutions bear most of the financial • Involving the matching grants PIU in the risks of supported projects. Such arrangements drafting and adjustment of the matching grants may simplify and speed up the investment manual is important to strengthen the capacity process but require strong communication so of the PIU, ensure project ownership, and as to avoid confusion among beneficiaries on ensure that processes are flexible. Throughout the nature of the support. the project, the matching grants manual should be a • For beneficiaries who have lost access to working document that can be adjusted according finance, financial institutions could play a to circumstances. For instance, in a context of low leading role in the identification and selection demand for matching grants, it might be useful for of matching grants beneficiaries. the PIU to revise the selection process and switch to a “first come, first serve” selection mechanism. • When financial institutions’ lack of Similarly, in case the composition of the matching information and know-how on agriculture is grants selection committee leads to delays in identified as one of the key market failures, the selection process, it should be possible to involving financial institutions as advisors HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 37 quickly change the composition of the committee • Mechanisms should be put in place to avoid so as to maintain the reputation of the project. the risk of companies overcharging for eligible Alternatively, switching to a virtual committee equipment. Such mechanisms include the which does not require physical meetings should following: clear product specifications vetted by be possible. an independent qualified consultant; standard bid document to allow for price comparisons and • Having a strong communication plan about promote price transparency to help farmers choose matching grants from the beginning of the between different products; lists of pre-approved implementation is key to ensure equal access qualified vendors who have a track record where to grants, promote accountability, and foster applicable; independent verification and audit spillovers. For instance, showcasing matching of prices charged with and without subsidy, as grants beneficiaries on local television, radio, well as sanctions (such as debarring from future and social media increases project ownership and bids) in case of overcharging; and (depending decreases the risk of grant misuse. Additionally, on amounts and complexity) an auction where it can foster innovation and technology adoption several companies compete on price and quality, among non-beneficiaries, which is a key expected and project beneficiaries are only allowed to impact of matching grants projects. Task team purchase equipment from approved vendors.  leads should work in coordination with social development specialists to ensure communication • Suggested M&E indicators related to cost- materials and information reach indigenous efficiency include % operative costs/total amount populations. of matching grant, increase in beneficiaries’ income linked to the subproject/total amount of • Contracts with BDS providers should be matching grant, % operative costs/ increase in designed to ensure quality and results. For beneficiaries’ income linked to the subproject. instance, TORs may include a payment schedule where most of the payment is made at the end of • Suggested M&E indicators to track spillovers the contract based on the achievement of specific may include % of nearby farmers adopting objectives (e.g., productivity improved, website promoted technology/equipment compared to a built, etc.) control group. CONCLUSION AND RECOMMENDATIONS 38 FINANCIAL COOPERATIVES — ISSUES IN REGULATION, SUPERVISION, AND INSTITUTIONAL STRENGTHENING References For the sample of 23 agriculture matching grants projects Project Appraisal Documents, Implementation Completion and Results Reports, and reviews by IEG Others Bruhn, Miriam, Dean Karlan, and Antoinette Schoar. 2013. “The Impact of Consulting Services on Small and Medium Enterprises: Evidence from a Randomized Trial in Mexico.” Policy Research Working Paper 6508, World Bank, Washington, DC. Campos, Francisco, Aidan Coville, Ana M. Fernandes, Markus Goldstein, and David McKenzie. 2012. “Learning from the Experiments That Never Happened: Lessons from Trying to Conduct Randomized Evaluations of Matching Grant Programs in Africa.” Policy Research Working Paper 6296, World Bank, Washington, DC. CTA (Technical Centre for Agricultural and Rural Cooperation). 2016. “Blending 4AG.” Innovative Partnerships for Agricultural Finance, Conference Proceedings, Brussels, Belgium, November 7–8. Dercon, S., and L. Christiaensen. 2011. “Consumption Risk, Technology Adoption and Poverty Traps: Evidence from Ethiopia.” Journal of Development Economics 96 (2): 159–73. Goldberg, Michael, and Daniel Ortiz del Salto. 2009. “Matching Grants: A Review of Matching Grants Systems in Private Sector Development Projects.” Unpublished memo. World Bank, Washington, DC. IEG (Independent Evaluation Group). 2014. The Big Business of Small Enterprises: Evaluation of the World Bank Group Experience with Targeted Support to Small and Medium-Size Enterprises, 2006–2012. Washington, DC: World Bank. IFAD (International Fund for Agricultural Development). 2012. “Matching Grants: Technical Note.” International Fund for Agricultural Development. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 39 Inter-American Development Bank. 2011. and Medium Enterprises on Firm Performance in “Evaluación del Impacto de los Fondos de Desarrollo Low- and Middle-Income Countries: A Systematic Tecnológico (FDT): El caso de COLCIENCIAS.” Review.” Campbell Systematic Reviews 2016.1. McKenzie, David, Nabila Assaf, and Ana Paula Vander Meer, K., and M. Noordam. 2004. “The Use Cusolito. 2015. “The Additionality Impact of a of Grants to Address Market Failures.” Agriculture Matching Grant Program for Small Firms.” Policy and Rural Development Discussion Paper 27, World Research Working Paper 7462, World Bank, Bank, Washington, DC. Washington, DC. WBG (World Bank Group.) 2016. Linking Farmers Phillips, David. 2001. “Implementing the Market- to Markets through Productive Alliances: An Based Approach to Enterprise Support: An Evaluation Assessment of the World Bank Experience in Latin of Ten Matching Grant Schemes.” Policy Research America. Washington, DC: World Bank. Working Paper 2589, World Bank, Washington, DC. World Bank. 2016. How to Make Grants a Better Piza, Caio, Tulio Cravo, Linnet Taylor, Lauro Match for Private Sector Development: Review of Gonzalez, Isabel Musse, Isabela Furtado, Ana World Bank Matching Grants Projects. Competitive Cristina Sierra, and Samer Abdelnour. 2016. “The Industries and Innovation Program. Washington, Impact of Business Support Services for Small DC: World Bank. REFERENCES 40 Annex 1. Financial Institutions as Deposit Takers Angola Local Development Project Description The Angola Local Development Project aims at improving business development skills and participation in markets of selected producer groups and SMEs in a few targeted sectors, including agriculture. This project includes a matching grants component for selected producer groups, SMEs, and business development service providers. Linkage with financial institutions In Angola, the matching grants agreement incentivizes each individual within the producer group to save 20% of the grant amount in a savings account over the course of the project, with mandatory monthly deposits.57 Beneficiaries receive small rewards (phone cards with airtime) for each monthly deposit made on time. Positive impact The impact evaluation shows that beneficiaries gradually fulfilled financial institutions’ requirements through the project (e.g., had an active account, had collateral, were formal enterprises). In addition, there are a few examples of beneficiaries who were able to access a loan after benefiting from matching grants. From a sample of 49 subprojects, two beneficiaries (4%) got access to a loan from a financial institution by the end of the project. The latest evaluation, from 2017, indicated that each US$1 spent in grants generated on average US$23 in revenues for producer groups and SMEs. Colombia Second Rural Productive Partnerships Project Description The Second Rural Productive Partnerships Project (2007–2013) included a matching grants component (known as the Incentivo Modular) of US$327.13 million (of which IBRD contributed US$24.42 million), which was awarded to beneficiary producer organizations. Grants were available for improving on- farm infrastructure, purchasing machinery and equipment, financing consumable HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 41 inputs (such as seed, fertilizer, and veterinary supplies), or paying for hired labor. Matching grants also paid for technical advisory services, marketing studies, and training activities designed to increase the productivity and entrepreneurial capacity of the beneficiary producer organizations. The Incentivo Modular could not exceed 40% of the total partnership investment cost, and it was capped at Col$5 million per beneficiary producer household (less than US$2,000). Linkage with financial institutions This project promoted the participation of financial institutions both as funders, through the matching grant design, and as deposit takers, through the setting up of a revolving fund. Matching grants were designed in a way that promoted commercial credit. Indeed, the level of matching and ceiling grant amounts were increased for beneficiaries who managed to obtain credit. The grant amount could be increased to a maximum of Col$6 million per beneficiary producer household during the course of subproject implementation if the partnership could demonstrate that it had mobilized an additional Col$2 million per household in commercial credit. For example, if the partnership obtained a credit of Col$100,000 per household, the government financial incentive could be increased by Col$100,000 per household. This 1:1 relationship would be respected up to a maximum of Col$2 million per household. Additionally, beneficiaries were required to set aside 70% of the grant received in a revolving fund. Under an earlier matching grants project, many participating producer organizations had experienced difficulty accessing commercial credit. The design for the Second Rural Productive Partnerships Project therefore included stronger measures to entice financial institutions to provide credit to the selected partnerships. Producer organizations were encouraged to use and manage their mandatory revolving fund as a tool to finance their working and investment capital requirements. The objective of the revolving funds was to allow producer organizations to continue their operations after the project, as well as build up their creditworthiness, in order to enable them to obtain commercial financial credit. Positive impact This project generated considerable spillover benefits that have accrued to non-beneficiary producers. More than 11,000 rural households that were not members of a producer organization enrolled in a subproject nevertheless; adopted (at their own expense) improved practices promoted through the project; or were able to benefit from collective goods paid for by the project. One significant spillover effect is the 24.4% higher gross income found for “nearby producers” compared to “distant producers,” showing that the impact of the project on Colombia’s rural sector was amplified beyond direct project beneficiaries. There were significant improvements in beneficiaries’ savings and access to credit. The revolving funds were mainly used for (1) giving out loans to non-beneficiary producers of the same producer organization (42%), (2) financing technical assistance (25.3%), and (3) purchasing specialized machinery (20.3%). The impact evaluation shows that—compared to control group producers—beneficiary producers obtained credit for productive investments (table 3), and more of them reinvested part of their net revenues in their agricultural production compared to the control group. ANNEX 1: FINANCIAL INSTITUTIONS AS DEPOSIT TAKERS 42 Table 3. Results on Competitiveness Project Beneficiaries Non-beneficiaries Differences in Producers (Treatment) (Distant Controls) Means With access to improved, 7.1.7 44.1 *** certified inputs With access to collection 3.5 3.3 ** center(s) Reinvesting from net revenues in 66.4 57.0 * their agricultural production Went from not having to having a transport system for the 19.8 8.4 *** marketing of products Went from not having to having contracts or agreements for the 49.3 11.3 *** sale of products Went from not having to having credit for investments in 25.6 16.1 *** productive activities Note: Significance level: * = 10%; ** = 5%; *** = 1%. Challenges Requiring matching grants beneficiaries to save 70% of the grant in the revolving fund proved challenging,58 in particular for farmers involved in long production cycles and farmers facing significant production risks: • Of producer organizations, 50% had a recovery rate below 70%, in particular organizations involved in producing perennial crops (e.g., cacao, forestry, mango), which have long gestation periods, limiting producers’ ability to repay quickly. • In addition, production losses due to adverse weather had adverse impact on producers’ ability to repay into their revolving funds. From 2009 to 2011, unusually adverse weather significantly affected a large number of beneficiary producers, who suffered significant losses during project implementation from fire (97%), flooding (87%), landslides (82%), drought (42%), and outbreaks of diseases (34%). These weather conditions had serious adverse impacts on production and sales, and on the producers’ ability to repay into their revolving funds. Such difficulties in accumulating savings show that many project beneficiaries would not have been ready to reimburse credit, and therefore that the project helped pave the way for financial inclusion in a gradual manner. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 43 While the design of the matching grants promoted financial institutions’ participation through favoring beneficiaries who had managed to obtain credit, such a model raises the question of the additionality of matching grants. Indeed, while this is not indicated in the ICR, it is possible that some of the matching grants beneficiaries could have financed their project solely from credit. Such a model illustrates the trade- off between the need to ensure additionality (channeling grants only to segments who are not able to finance their investments through credit) and the wish to promote sustainability (channeling grants in a way that beneficiaries do not need grants for their future investments). Nigeria Third National Fadama Development Project Description The Third National Fadama59 Development Project (2008–2013) included three matching grants components, which financed (1) advisory services for farmers/pastoralists (called Fadama users); (2) input support (mainly seeds, fertilizers, and agrochemicals) with a matching grant of 50% of the input purchase price; and (3) capital assets to undertake a broad range of small-scale income-generating activities and facilitate access to markets Linkage with financial institutions This project introduced an innovative savings mechanism called Fadama Users’ Equity Fund (FUEF). Indeed, the project required that 10% of the replacement value of the common assets used for income- generating activities of Fadama User Groups (FUGs) was saved annually (with effect from year 2). This feature served as a sustainability provision on the project and aimed at facilitating the observed desire of participants to continue investment after completion of the matching grant. Fadama Farmers Community Associations (FFCAs): Under South-South Cooperation, the Fadama project undertook a study tour of India and Sri Lanka for learning experiences on intensification of federation of community groups, use of agricultural technological innovations and ICT, and promotion of a rural savings- credit revolving scheme. Positive impact The project promoted the formation of FUGs, which were federated into associations at state level as the FFCA. More than 490,000 farmers had access to agricultural inputs thanks to the matching grants program. Beneficiaries also acquired agricultural productive assets worth N11.6 billion, equivalent to about US$72 million. Livestock assets (e.g., poultry production) and crop assets (e.g., water pumps, sprayers) constituted the largest number of productive assets. Across the 36 states, about 7.32% of the value of these assets was saved in FUEF accounts, slightly below the 10% target. A total of 37 FFCAs were created, one for each state and the Federal Capital Territory, with the objective of transforming some of the vibrant FFCAs into self-sustaining institutions. The ICR indicates that a total of 15 FFCAs were still functional and providing services to their members. Of them, seven institutions have generated enough capital and expertise to apply for a license to operate as a microfinance bank. They are waiting for the Central Bank’s approval before commencing operations as a microfinance bank. In the state of Plateau, the Central Bank of Nigeria awarded the FFCA its banking license in 2015, and the Plateau State ANNEX 1: FINANCIAL INSTITUTIONS AS DEPOSIT TAKERS 44 Fadama Farmers Microfinance Bank has been operational since January 2016. It is expected that the other seven FFCAs will also get banking licenses and will contribute to providing requisite financing. Challenges International experience suggests that rural financial institutions stemming from farmers’ groups can face issues related to lack of financial expertise and governance. The fact that less than half of the state- level institutions created as part of the project were still in operation by the end of the project confirms these challenges. India-Chhattisgarh District Rural Poverty Project: Requirement of 10% of Investments in Savings for the Second Matching Grants Tranche to be Disbursed Description This project included two matching grants components: (1) matching grants for community investment subprojects from Common Interest Groups (CIGs) to finance collective income-generating activities, including agriculture as well as traditional activities such as trading; and (2) matching grants for Panchayat (village groups) plans to finance investment subprojects in village infrastructure. Linkage with financial institutions Community members had to contribute 5% in cash upfront towards subproject costs and place 10% into a village fund (Apna Kosh) as a condition for the release of the second tranche of the subprojects. The fund was aimed at covering operation and maintenance costs and further village development beyond the lifetime of the project. Towards the end of the project, a decision was made to revise the guideline on the village fund and use the fund as working capital/revolving fund for the federations of the CIGs. Outcome The project fostered investments in collective income-generating activities as well as village infrastructure: • Income-generating activities: A total of about US$39.32 million in matching grants for collective income- generating activities was provided to 20,446 completed CIG subprojects identified by community members. Communities contributed about US$2.06 million for these activities. These subprojects were mainly for agriculture, livestock, and traditional local activities. • Village infrastructure: A total of about US$13.57 million was provided as matching grants to finance 3,314 completed investments in village infrastructure; 60% of these investments were for paved cement roads. Members also consistently raised the 10% of savings, although it led to some implementation delays for poorer individuals. At completion, around US$3.56 million, or 10.1% of actual subproject costs, had been placed in fixed term deposits. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 45 Positive impact Project beneficiaries had more confidence in dealings with banks, had access to bigger loans, and displayed stronger savings discipline. The IE shows that beneficiaries, especially women, considerably increased their confidence in dealing with banks. While the IE did not show a significant difference in the share of CIG beneficiaries with bank accounts compared to control areas, it did find that more CIG members with bank accounts succeeded in taking loans compared to the situation in control villages (48% v. 37%), and that in control areas more still used money lenders. The CIG members’ loans were for productive purposes, primarily agricultural investments, and were double the amount of productive loans taken out in control areas. Overall, CIG households saved more than non-CIG households, in particular most excluded beneficiaries. This saving discipline contributed to CIG households becoming more creditworthy. According to the IE, the total amount of loans taken by CIG households was 30% higher than the amount borrowed by non-CIG households. Project households borrowed 30% more from banks and 40% less from money lenders, compared to control households. Challenges The impact evaluation indicates that 70% of the groups are expected to sustain their improved farm operations during the project period, while 30% of the CIGs failed to function effectively and are considered unsustainable. This raises the question of whether sufficient screening was undertaken and whether the level of required contribution (5% cash) was sufficient to ensure the sustainability of CIG projects. ANNEX 1: FINANCIAL INSTITUTIONS AS DEPOSIT TAKERS 46 Annex 2. Financial Institutions as Funders India Assam Agricultural Competitiveness Project Linkage with financial institutions In its initial design, the project promoted strong participation of financial institutions, as opposed to an earlier World Bank project in the region.60 Started in 2004, the Assam Agricultural Competitiveness Project in India included a matching grants component for irrigation investments for farmers. This program initially required that farmers contribute 20% of the investment and that commercial banks cover 50% of the investment. This design aimed at promoting private sector financing, compared to the previous project providing grants covering 70% of the investment costs. Challenges The project had to be restructured due to slow disbursements. After 18 months the project was able to provide only 470 irrigation pumps against the target of 6,170. The project was therefore restructured to raise the grant to 50%, with the balance of 50% contributed by the beneficiary as cash, while also dispensing with the mandatory commercial bank linkage. This experience highlights the trade-off between promoting a design which promotes sustainability and ensuring quick disbursements. It also illustrates the difficulty of changing matching grants patterns after an earlier project with a low level of beneficiary contribution. Honduras COMRURAL Project Linkage with financial institutions The Honduras COMRURAL project required that a subproject secure a loan from a financial partner, covering at least 30% of total subproject investment costs, in addition to the 10% in cash or in kind required by the Produce Organization. From the beginning of the project, a broad range of financial partners were identified as eligible to participate in co-financing the subproject: (1) commercial banks, finance associations, and private finance development HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 47 institutions regulated by the Banking and National Insurance Commission; (2) credit and savings cooperatives affiliated with the Honduran Federation of Credit and Savings Cooperatives; and (3) other microfinance institutions, and other buyers such as input providers. The matching grant instrument was combined with a Partial Credit Guarantee fund to increase the supply of agriculture finance services. Positive impact During implementation, the producer organizations’ contribution was often higher than required, in some cases almost 50% of subproject costs. As a result of their participation in the project, various producer organizations have received further loans. ANNEX 2: FINANCIAL INSTITUTIONS AS FUNDERS 48 Annex 3. Financial Institutions as Managers Kyrgyz Republic Agribusiness and Marketing Project Financial institutions as primary providers of funds (70% required) and managers of grants, with grants conditional on loan repayment Description The program allowed eligible cooperatives that had participated in the specialized training, and required investments in productive assets for agricultural production and small-scale processing, to obtain a portion of necessary financing to procure the asset on a conditional grant basis. Linkage with financial institutions In this project, financial institutions played a key role, both in the appraisal of grant requests and in the funding of the investments. The participating financial institutions appraised requests for grant financing to ensure the cooperatives were financially viable, and filled in any financing gap (between the total amount of the subproject and the grant financing) from their own credit line resources. The grant could be obtained in an amount of up to 30% of the loan amount. For example, if a cooperative wanted to purchase a tractor, it had to invest 10% of own funds (as borrower’s contribution). Of the remaining 90%, up to 30% could be obtained as matching grant, and the remaining 70% was provided by the participating financial institution as a loan from own resources or from the credit line. The grant represents the last 30% of the provided financing—the loan portion (70%) had to be repaid in full and on time in order for the 30% to become a grant. In case of non-repayment of the loan portion, the entire matching grant amount became a loan, repayable with interest. The matching grant instrument was combined with three other instruments aimed at increasing the supply of agriculture finance services: (1) a credit line to participating financial institutions for agribusiness lending, (2) technical assistance to loan officers to broaden the base of eligible borrowers, and (3) training of agricultural cooperatives to strengthen their management skills and allow them to apply for financing. 49 Positive impact While the project was small in scale, it had positive impact on the economic activities of beneficiaries, who were all able to reimburse their loans. Fifty-eight matching grants for a total of US$800,060 were provided to cooperatives to co-finance investments in agricultural machinery and other productive assets. The repayment rate on the subloans financed through the credit line (amounting to about US$16.6 million) was 100%. According to the impact survey at the end of the project, 32% of the respondents saw an increase in their production output, 41% saw an increase in profit, 37% saw an increase in total sales, and 47% saw an increase in market share compared to before the investments. Challenges While such a challenge is not indicated in the ICR, a model whereby financial institutions channel both grants and loans could potentially create confusion among beneficiaries. In such an arrangement, it is important to raise awareness among beneficiaries about the difference between grants and loans. Burundi Project Keeping Good Firms Alive & Well: Matching Grants to Tackle Debt Overhang and Recreate Credit Histories Description The overall objective of this SME Launchpad project is to reintegrate in the financial sector SMEs that have lost access to finance due largely to exogenous price fluctuations and a general economic crisis. Linkage with financial institutions Commercial banks are involved at two important stages of the selection process: (1) identification of eligible SMEs based on their credit history; and (2) selection of SMEs, as the selection committee includes a representative of the National Association of Banks and Financial Institutions of Burundi. Preliminary results after four months of implementation: • Five commercial banks were involved in the pilot project. • One-third of beneficiary SMEs were involved in the production, storage, and/or distribution of agricultural products. • Of commercial banks participating in the pilot, 80% acknowledged that the project would lead to partial and/or full SME debt write-offs. • Of participating SMEs, 50% reported being satisfied or very satisfied by the loan restructuring proposed by their bank on the basis of the investment plan. ANNEX 3: FINANCIAL INSTITUTIONS AS MANAGERS 50 Annex 4. Financial Institutions as Advisors Mexico Sustainable Production Systems and Biodiversity Project Description The project makes matching grant financing available to existing producer associations and to producer groups that have developed a business plan for the production, processing, and marketing of biodiversity-friendly products. Linkage with financial institutions This project establishes commercial financial institutions as key technical service providers which are hired to support the management of financial services for each producer association. The project includes three types of technical service providers for producer groups, including financial institutions: (1) local technical groups provide training and technical assistance to producer groups and producer associations; (2) Technology Transfer Units provide services for research, technology development, and innovation; and (3) financial agents support the management of financial services. Additionally, this project paves the way for better access to finance through the requirement that every producer group needs to be formalized in order to receive financing. HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 51 52 Endnotes 1. IFAD (2012). 2. World Bank (2016). 3. These numbers capture only closed projects. 4. In some cases matching grants might distort financial markets and violate Article 6.2 of the WTO Agreement on Agriculture. In at least one case, the World Bank discontinued a matching grant fund on the belief that the fund violated WTO rules, which had come into existence since the project was approved. To satisfy the requirements of the framework agreement of the WTO, grants are defined as nonactionable subsidies because they are not specific. Specificity under WTO rules typically involves “targeting of geographic regions or economic sectors.” 5. World Bank (2016). 6. Campos et al. (2012). 7. In this paper, “access to finance” for farmers and agricultural small and medium enterprises is defined as access to suitable formal financial services such as savings, credit, insurance, and payments. 8. WBG (2016). 9. Inter-American Development Bank (2011). 10. See World Bank (2016). 11. bid. 12. IFAD (2012). 13. Vander Meer and Noordam (2004). 14. World Bank (2016). 15. Ibid. 16. Bruhn, Karlan, and Schoar (2014). HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 53 Ibid. Owners also had improvements in 17. expected loss for the organization and “entrepreneurial spirit” when compared with inducing a producer group to engage in a the control group. The analysis finds a large profitable activity it would have not have increase in the number of employees and the engaged in otherwise. total wage bill several years after the program. 22. WBG (2016). The paper documents that there is no singular mechanism for all firms. See the upcoming WBG Agriculture Finance 23. Diagnostic for Côte d’Ivoire and Senegal. 18. McKenzie, Assaf, and Cusolito ( 2015). 24. Each financial institution defines “bankability” 19. Piza et al. (2016). differently depending on its business model. The reasons include ethical concerns about 20. This section therefore uses the term “fully government “randomizing out” eligible bankable projects” for projects that can be applicants; program application rates too low perceived as bankable by at least one financial to enable the planned selection of a random institution. sample of eligible applicants; and continued 25. “Potentially bankable projects” include implementation delays that prevented the start investments which are perceived as particularly of the impact evaluation. risky or innovative by financial institutions The literature points to several ways that 21. (e.g., investments in new drought-resistant matching grants can help increase access to crop varieties with high return potential, large finance: projects with long return periods, etc.). • In addition to the matching grants providing 26. World Bank (2016). needed funds, the approval of a business 27. Of the 15 closed projects, some did not have proposal by the government might serve to an ICR when the review was undertaken and signal the quality of the proposed investment therefore do not have ratings. These unrated and increase access to credit to finance the projects have an “n/a” success rate in the remainder of the investment. See Campos et analysis below. al. (2012). 28. Compared to 72% for all 106 projects. • Risk sharing may induce higher technology adoption. Risks are widespread in rural 29. Compared to only 22% for non-agriculture productive activities, and small farmers projects. have limited access to credit and insurance, 30. This feature was highlighted in the World Bank in particular to smooth common shocks. (2016) review of 106 projects as the design The high variability of the climate and the modality which is most often correlated to associated risk of poor harvests heighten this positive outcomes. risk, thus lowering farmers’ propensity to invest in new technologies; see Dercon and 31. For instance, the IEG review of the Zambia Christiaensen (2011). This is even more the Agricultural Development Program (P070063) case where farmers produce similar crops, indicated “inadequate evidence on the extent increasing covariant risk. A matching grant to which” the key objective of the matching scheme can potentially provide the necessary grants—namely improvements in agricultural risk-sharing arrangement by lowering the productivity—had been achieved. ENDNOTES 54 32. Phillips (2001). 40. See Niger Agro-Pastoral Export Promotion (P095210). 33. Indeed, the project involves two components: the first one consists of a revolving line of credit 41. Information included in the ICR of the to finance land purchases, and the second one Sudan Gum Arabic Export Marketing Project provides matching grants for infrastructure and (P110588) seems to suggest that a similar productive investments on the purchased lands. approach was adopted for this project; however, The acquired land can serve as collateral for the Project Appraisal Document for this project long-term credit. was not available. 34. “Niger: Note politique sur le credit agricole,” 42. Raiffeisen banks came from multipurpose June 2018. cooperatives with a strong agricultural focus, which spun off their financial activities into 35. According to WBG (2016): “Productive dedicated financial institutions. Alliance involves three core agents: a group of smallholder producers, one or more buyers, 43. One last area for caution is compliance with and the public sector. These three agents are WBG Operational Policy 10 related to revolving connected through a business proposition, or funds. ‘business plan,’ which describes the capital and 44. There was also an option for producer groups to services needs of the producers and proposes have a collective savings account; however, all improvements that would allow them to producer groups preferred setting up individual upgrade their production capacities and skills savings accounts. to strengthen their linkage with the market, i.e. the buyer(s). The implementation of such a The grant amount could be increased to a 45. business plan through a subproject is typically maximum of Col$6 million per beneficiary supported through three core inputs and/ producer household during the course of or activities directed towards the producers’ subproject implementation if the partnership needs: productive investments, technical could demonstrate that it had mobilized an assistance, and business development. These additional Col$2 million per household in core inputs are financed through public grants commercial credit. provided by the project, which are matched by 46. For instance, the Angola Agriculture the beneficiary producers and in some cases Commercialization Project (P159052) under also by the buyer(s).” preparation considers capping working capital 36. This was true of 28% of 106 projects reviewed expenditures to 25% of the grant amount. as part of World Bank (2016). 47. This refers to an ongoing project in Argentina, 37. IFAD (2012). This source further notes: “The Integrated Risk Management in the Rural lower the contribution of the recipients, the Agroindustrial System (P162316). lower their ownership, the higher the interest of 48. The grant could be obtained in the amount of local politicians and potential beneficiaries, and up to 30% of the loan amount. For example, the faster the disbursement rate.” if a cooperative wanted to purchase a tractor, 38. 2011 Findex data. they would have to invest 10% of own funds (as borrower’s contribution). Of the remaining 90%, 39. World Bank (2016). up to 30% could be obtained as matching grant, HOW CAN MATCHING GRANTS IN AGRICULTURE FACILITATE ACCESS TO FINANCE? 55 and the remaining 70% would be provided by in control villages (48% v. 37%), and that in the participating financial institution as a loan control areas more still used money lenders. from own resources or from the credit line. According to the IE, the total amount of loans taken by beneficiaries was 30% higher than the 49. It provided 58 matching grants for a total of amount borrowed by non-beneficiaries. Project US$800,060 to cooperatives. households borrowed 30% more from banks 50. The Tamil Nadu Rural Transformation Project and 40% less from money lenders, compared to (P157702) under preparation considers a control households. slightly softer approach, whereby matching 56. One significant spillover effect is the 24.4% grants beneficiaries are required to obtain higher gross income found for “nearby commercial credit, but the credit is converted producers” compared to “distant producers,” into a grant if the loan is repaid in full and on showing that the impact of the project on time. Colombia’s rural sector was amplified beyond 51. “Niger: note politique sur le crédit agricole.” direct project beneficiaries. Juillet 2018. 57. There was also an option for producer groups to 52. This is an SME Launchpad project currently have a collective savings account; however, all under implementation. producer groups preferred setting up individual savings accounts. 53. Such a feature is also being considered for the Republic of Congo Commercial Agriculture 58. By the end of the project, the average recovery Project (P159979). rate across all producer organizations was 64%, below the 70% target. 54. Some projects both have an access to finance component and a design feature promoting 59. “Fadama” refers to rural lands and water financial inclusion. resources within Nigeria. 55. The impact evaluation of the Chhattisgarh 60. The earlier project was Rural Infrastructure and District Rural Poverty Project showed that Agricultural Services Project (ARIASP) during beneficiaries with bank accounts succeeded the period 1995–2004. better in taking loans compared to the situation ENDNOTES 56