94918 IFC ACCESS TO FINANCE GLOBAL MICROFINANCE ADVISORY IN PARTNERSHIP WITH DUTCH GOVErNMENT EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Based on Case Studies in LAC IN PARTNERSHIP WITH © International Finance Corporation 2014. All rights reserved. 2121 Pennsylvania Avenue, N.W. Washington, D.C. 20433 www.ifc.org The material in this work is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. IFC encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly, and when the reproduction is for educational and non-commercial purposes, without a fee, subject to such attributions and notices as we may reasonably require. IFC does not guarantee the accuracy, reliability or completeness of the content included in this work, or for the conclusions or judgments described herein, and accepts no responsibility or liability for any omissions or errors (including, without limitation, typographical errors and technical errors) in the content whatsoever or for reliance thereon. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The contents of this work are intended for general informational purposes only and are not intended to constitute legal, securities, or investment advice, an opinion regarding the appropriateness of any investment, or a solicitation of any type. IFC or its affiliates may have an investment in, provide other advice or services to, or otherwise have a financial interest in, certain of the companies and parties (including named herein). All other queries on rights and licenses, including subsidiary rights, should be addressed to IFC’s Corporate Relations Department, 2121 Pennsylvania Avenue, N.W., Washington, D.C. 20433. International Finance Corporation is an international organization established by Articles of Agreement among its member countries, and a member of the World Bank Group. All names, logos and trademarks are the property of IFC and you may not use any of such materials for any purpose without the express written consent of IFC. Additionally, “International Finance Corporation” and “IFC” are registered trademarks of IFC and are protected under international law. Cover photo: IFC Photo Library TABLE OF CONTENTS i Table of Contents ACKNOWLEDGEMENTS III EXECUTIVE SUMMARY V I. INTRODUCTION 1 II. MARKET OVERVIEW 3 A. DEFINITION OF VERY SMALL ENTERPRISE AND CHARACTERISTICS OF THE SEGMENT 3 B. OVERVIEW OF BANKS AND HOW THEY SERVE SMALL ENTERPRISES 5 C. OVERVIEW OF NON-BANKS AND HOW THEY SERVE VERY SMALL ENTERPRISES: 6 III. PREREQUISITES 7 IV. INSTITUTIONAL STRATEGY 11 A. STRATEGIC RATIONALE FOR ENTERING THIS NICHE 12 B. A “STEP-UP” OR A “LEAP-UP”?: WHO IS THE TARGET MARKET? 15 C. CONSIDERATIONS OF MARKET CONTEXT 16 D. CONSIDERATIONS OF LEGAL STRUCTURE 17 E. OTHER CONSIDERATIONS 18 V. DEFINING THE VALUE PROPOSITION 19 vi. COMMERCIAL STRATEGY 23 A. UNDERSTANDING YOUR MARKET 23 B. PRODUCT DEVELOPMENT 24 C. PROSPECTING AND MARKETING 29 D. OTHER CONSIDERATIONS 30 vii. CREDIT ANALYSIS 31 A. CAPACITY TO PAY UPGRADES 33 B. WILLINGNESS TO PAY ADJUSTMENTS 35 viii. CREDIT RISK MANAGEMENT 39 A. LOAN APPROVAL AND OVERSIGHT 39 B. MONITORING AND TRACKING STRUCTURES AND POLICIES 40 C. RECOVERY POLICIES 41 D. REGULATIONS 42 ii EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Ix. hUMAN RESOURCES 43 A. SALES AND LOAN ANALYSIS STAFF 43 B. COMPENSATION AND INCENTIVE ADAPTATIONS 46 C. SPECIALIZED CAPACITY AND QUALIFICATIONS 47 D. TRAINING AND SUPPORT 48 X. OPERATIONAL STRATEGY 51 A. DISCUSSION OF ORGANIZATIONAL STRUCTURE AND ADAPTATIONS NEEDED TO VARIOUS AREAS 51 B. IT CHANGES AND ADAPTATIONS 52 XI. FINANCIAL STRATEGY 55 A. COSTS AND PROFITABILITY 55 B. ASSET/LIABILITY MANAGEMENT AND FUNDING STRATEGY 57 XII. CONCLUSIONS 59 BIBLIOGRAPHY 763 ENDNOTES 67 ACKNOWLEDGEMENTS iii Acknowledgements This study was commissioned by IFC Access to Finance Microfinance team and jointly conducted by IFC LAC Access to Finance Microfinance/ SME Banking and Global Microfinance Advisory. The studies benefitted from peer review by Global and Regional Micro and SME Specialists (Girish Nair, Patricia Mwangi, Mohammed Khaled, Swati Sawhney, Neil Ramsden, Martin Hommes), as well as Financial Markets (Wendy Teleki) and CGAP (Minh Huy Lai). Special thanks go to the regional micro and SME specialists, Martin Spahr and Ignacio Estevez, and an industry expert, Andrew Pospeilovsky, for their technical leads. Barbara Magnoni, at EA Consultants, was hired to carry out the study. The team is grateful to the Netherland’s Government for the funding that made the study possible and to the IFC SME Finance Forum for the editorial support.. Project Team Task Leader: Gaamaa Hishigsuren, IFC Access to Finance Global, Microfinance Specialist Lead Technical Specialists: Martin Spahr, IFC LAC, Microfinance Specialist and Ignacio Estevez, IFC LAC SME Banking Specialist Consultant Researcher: Barbara Magnoni, President of EA Consultants Graphic Designer: Aichin Lim Jones, IFC Access to Finance, KM and Learning iv EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA EXECUTIVE SUMMARY v Executive Summary Very small enterprises (VSEs) in Latin America and the Caribbean (LAC) represent a broad and heterogeneous segment, often underserved by formal financial institutions. They are generally informal and often family businesses. The financial needs of these enterprises are typically overlooked by “downscaling” banks, which find larger and often more formal small and medium enterprises (SMEs) to be a more natural market for their products and services. Microfinance Institutions (MFIs) are starting to move upmarket to serve SMEs, and in particular, VSEs within this segment. However, they use varying definitions, methodologies and products to do so and to date there has been little research or documentation of their experiences. This report highlights the results of a recent study of the existing practices in Latin America of MFIs serving VSEs, where VSEs are considered to be those businesses with financing needs of between US$7,000 and US$30,000. It is a starting point for an institution considering entering the segment, or for one that finds itself having grown into the segment “organically” but with a view to strengthen its position. It includes several checklists for MFIs interested in expanding upmarket into the VSE space, as well as more detailed discussions and examples of the most relevant points. The research involved in producing this document and checklist included a desk review of literature, phone interviews with MFIs and other stakeholders and four field visits to MFIs in Bolivia and Peru. It is important to note that most of the institutions we spoke to have entered the VSE space relatively recently (the majority since 2009 or 2010, or in a few cases 2006), and that the macroeconomic context in most of their countries has been very favorable to microenterprise and VSE growth. This strong performance may buffer VSEs and the institutions that serve them from some of the volatility that is often characteristic of VSEs worldwide. As such, their models have not really been fully tested and the lessons drawn from their experiences are evolving. When an MFI is considering expanding upmarket to serve VSEs, it must first evaluate its readiness to expand into new business areas. This assessment should include a review of both the external opportunities (i.e. market gap, regulatory environment, credit bureau access) as well as the internal capacity across all areas. Balancing risk and growth is critical when expanding to VSEs, thus, an essential prerequisite for an MFI to successfully expand upmarket is that their existing credit risk management policies and procedures be strong, and that these can be adapted to a new segment. The next step is to clearly define its institutional strategy or the high level approach it plans to take to VSEs. At the core of this strategy is identifying how it would like to incorporate VSEs into its existing business. Generally speaking there are two types of business models that MFIs have adopted to serve the sector: an “organic model” that many MFIs who are expanding upmarket to follow their existing clients adopt; and a “Proactive Model” that involves greater structural changes and segmentation between micro and VSE clients, yet provides a stronger foundation for serving new VSEs and/ vi EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA or slightly larger VSEs. However, over time, even MFIs who enter the VSE segment somewhat organically often have to adopt a more Proactive approach to VSE lending as they move from purely serving “graduated” micro clients to working with new VSEs. Generally speaking, the key to entering the VSE market effectively, regardless of the model, appears to be in striking an appropriate balance between risk management and an institutions commercial goals. Microfinance requires similar trade-offs, but smaller loans are spread over a larger potential risk, naturally mitigating some of the problems of credit risk. With VSEs, loan sizes are larger, and portfolios may be more concentrated in fewer clients’ businesses. Thus getting the risk/growth balance right at the appropriate cost is more critical and requires a more nuanced strategy. A commercial strategy for MFIs to serve VSEs should identify a key market segment, or multiple segments and define the MFI value proposition for each, ideally one de-linked from interest rates as MFIs tend to be at a disadvantage compared to banks in terms of funding costs. While interest rates are important in some markets, having a greater appetite for risk and the willingness to lend larger amounts against a given collateral item can often be highly effective in attracting VSEs, as can an MFI’s convenience, flexibility, agility or product design. VSE loan products require some adaptation of traditional microenterprise loans including working capital and fixed investment loans with longer tenors, lower interest rates and larger loan sizes. Credit and non-credit products must be available to VSEs to serve their needs, but also to increase the total profitability of relationships with these businesses and the “stickiness” of the clients. A revision of credit analysis and risk management policies is essential to the success of expanding upmarket. Simply using “micro” technology to serve these informal VSEs can increase portfolio risk. However, the complexity of the credit analysis must be weighed against both the MFI’s cost of conducting it (in order to maintain profitability for the sector) as well as the impact of its agility and flexibility in serving its clients. Typical upgrades to an MFIs credit analysis and credit risk management may include: better validation of financial information; greater documentation of company information, references or collateral; formalized evaluation of external risks; more in- depth cash flow analysis and in some case pro-forma cash-flow analysis; more reliance on credit bureaus; greater collateral requirements; and revisions in loan approval and branch autonomy. The design of VSE loans and their collateral requirements must take into consideration not just the capacity of VSEs to provide real collateral, but also the costs of documenting and registering this collateral and their impact on the overall cost of the loan. VSE lending requires proactive monitoring, even in Organic Models. Proactive monitoring can help identify risky loans or sectors as soon as or even before they become problematic and help the MFI to avoid similar risks going forward. Centralized risk monitoring and tracking appears to be more effective in reducing portfolio at risk, but requires a sophisticated and agile MIS. It is absolutely essential that an MFI’s MIS allow for the micro and VSE loans to be tracked separately, ideally at the institutional, branch and loan officer level. This allows the risk department to identify not only EXECUTIVE SUMMARY vii problematic segments, but also the source of these segments, and to adjust policies accordingly. Key human resources considerations include deciding whether to hire specialized VSE loan officers (more common in Proactive models) or have existing loan officers work with both micro and VSE clients (more common in Organic models). Either model requires careful review of the incentive policies for loan officers to ensure the appropriate balance is struck between micro and VSE goals as well as portfolio growth versus risk management. Additional, an expansion upmarket requires investment in training or retraining not just loan officers but other branch and head office staff. The financial impact of serving VSEs for MFIs varies depending on the market. Generally VSE loans charge lower interest rates, but these lower interest rates can be offset by lower per dollar lent acquisition costs, lower portfolio at risk (at times) and cross-selling other products to the VSE owner or owner’s family members. Thus, when looking at a profitability for the VSE sector, MFIs should consider total client profitability versus total product profitability. Another key financial consideration of VSE lending is obtaining sufficent funding, generally longer term, in order to fund portfolio growth. viii EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA I. INTRODUCTION 1 I. Introduction Increasing access to finance for small and medium enterprises (SMEs) has become an important goal for many governments and development agencies as SMEs are seen as drivers of economic growth and creators of employment. Especially in developing countries, these firms have traditionally been ignored by banks because they are viewed as either too small, too risky or too costly to serve, yet their financial needs generally far exceed the product offerings of microfinance institutions (MFIs).1 In Latin America, the SME financing gap is estimated at 125-155 billion (Stein 2010). In recent years, many banks in Latin America and worldwide, have begun to recognize the market potential of the sector and started downscaling their operations to serve these enterprises. Despite their enthusiasm, however, only a fraction of SMEs have access to loans from a financial institution today. Smaller firms are at a particular disadvantage. Most banks have concentrated on the larger and more formal SMEs, excluding many very small enterprises (VSEs)2 due to their size, formality or inability to meet collateral or guarantee conditions. 3 In turn, MFIs and other specialized entities that serve low-income markets find their products, systems and operations inadequate to serve the small end of this sector. As a result, small enterprises have traditionally been one of the most underserved segments in the MSME sector. 4 A few MFIs in Latin America and worldwide have stepped in by adapting their products, systems and processes to the needs of this segment (IPC 2012, CGAP 2012). Some of these MFIs have expanded upmarket intentionally in recognition of an underserved niche, such as many of the ProCredit Banks, while others have done so more organically as they adapt to retain and meet the needs of their existing clients’ growth. However, to date, there has been little research done in Latin America on MFIs that are serving VSEs and few cases that are well documented. This document begins to fill that gap by providing a synthesis of information and lessons learned during a four-month research process. The research included a review of literature and data to understand what types of financing are currently available to VSEs in Latin America and to explore the challenges MFIs have encountered when entering the VSE segment in Latin America. It also included a series of phone interviews with industry practitioners and MFIs that have expanded upmarket in Latin America as well as in-depth visits with four MFIs in Bolivia and Peru. The resulting report represents an initial review of the current practices of MFIs in the region that have expanded upmarket to serve VSEs and some initial lessons that can be drawn from their experiences. One of the most important lessons is that the approach an MFI takes to expanding upmarket must be driven by the market and regulatory context in the country and their own institutional strategy and goals. Hence, the report is not meant to be prescriptive but rather to outline a spectrum of issues that must be considered when expanding, based on the experience and current practices observed in a variety of strong MFIs in the region. 2 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA It is also important to remember that most of these institutions have entered the VSE space relatively recently (the majority since 2009 or 2010, or in a few cases 2006). During this period, the macroeconomic context in most of their countries has been very favorable to microenterprise and VSE growth. The research focused on experiences in Colombia, Peru, Ecuador and Bolivia as these countries are considered to be “leading markets” for MFIs that expanded upmarket to serve small and medium enterprises, but in particular, the segment of VSEs within the more broadly defined SME sector. Each of these countries has a relatively well-developed microfinance sector as well as several MFIs that are serving both VSEs and microenterprises. The sophistication of the SME banking market varies, but in each country there is at least some penetration of the small enterprise sector by banks. Each of these countries has also benefitted from strong economic growth over the past decade, a liquid financial sector and growth of both microfinance institutions and the clients they serve. This strong performance may buffer VSEs and the institutions that serve them from some of the volatility that is often characteristic of very small enterprises worldwide. As such, their models have not really been fully tested and the lessons drawn from their experiences are evolving. Nonetheless, for those that are interested in this market segment, there is much that can be learned from these leading players.   II. MARKET OVERVIEW 3 II. Market Overview A. Definition of Very Small Enterprise and Characteristics of the Segment: Definition of “very small enterprise” Approaching the Very Small Enterprise (VSEs) segment must inevitably begin with a discussion of how this segment is characterized or defined as we have identified a lack of consensus around firm size definitions. The term “very small enterprise” has only recently started being used, and refers to smaller small enterprises, thus our discussion of definition starts with “small enterprise” (SEs). Governments and donor agencies often define micro, small, medium and large firms in terms of annual sales, asset size, equity, number of employees and/or loan size, although the range for each may vary depending on the entity, country or usage (See the box below). In Latin America, there seems to be a general consensus in definitions that SEs have between 6 and 50 employees, with some variation by sector. There is much greater variation in definitions that include annual sales or assets. Financial institutions, bank superintendencies and regulatory authorities seeking to monitor lending to SEs tend to use definitions based on loan size. However, definitions based on loan size also vary substantially. Moreover, even when loan size figures are comparable in absolute terms, they can mean very different things depending on the country. In Peru, for example, the definition captures the total formal indebtedness of a client while in Bolivia it only includes the loan from a given institution. While in post-crisis Bolivia, clients are far less likely to have multiple loans than before, this potentially allows for an underestimation of business size. In Colombia, clients are even more likely to have loans from multiple institutions as well as credit cards, which makes using loan size as a proxy for business size more complicated. Despite some of the potential problems with using loan size to define business size, loan size is the easiest way for most financial institutions to delineate their portfolio, loan size is very relevant to their operations and is the most frequently reported and available information in most cases. It is also typically used by financial institutions to make strategic decisions regarding the segment. Thus, for the purposes of our research, we used the IDB’s definition of SEs as those requiring US$10,000-US$150,000 in financing as a general benchmark to characterize SEs, while microenterprises are defined as those requiring less than $10,000 in financing.5 More specifically, our research looked at VSEs, which tends to be where most MFI’s SE clients are found. These enterprises need loans of between $10,000-$30,0006 in most cases and in some markets and instances, as much as $50,000 or $75,000. 4 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Defining SEs by Firm Size In Latin America, governments define small businesses differently, broadly ranging anywhere from US $40,000 to US$8,000,000 in annual sales and between 6-50 employees. Very small business are not typically a subsegment within these definitions. The World Bank defines small businesses as those with 10-49 employees, US$100,000-US$3,000,000 in assets and annual sales, and/or approximately US$10,000-US$100,000 in loan size (IFC 2009) Defining SEs and VSEs by Loan Size • The IFC uses an SE loan size proxy of US$10,000-100,000 and a VSE loan size proxy of US$10,000-$25,000 • The Inter-American Development Bank (IDB) defines SE loans as those between US$10,000 and US$150,000/ US$250,000 depending on the country • In Peru clients with between US$7,500 and US$110,000 in total debt (excluding housing loans) with regulated financial institutions are considered small business clients • In Ecuador SE loans start at US$20,000 • MFIs in Latin America in a 2012 survey by the Consultative Group to Assist the Poor (CGAP) reported making loans to SEs and VSEs varying between bottom ranges of US$7,500-25,000 and top ranges of around US$90,000- $100,000. Characteristics of SEs and VSEs A resonating theme in the literature and research on SEs is that although they are larger than micro businesses in terms of required loan size, employees, sales and assets, they have more in common with micro businesses than with medium or large businesses (IPC 2012, CGAP 2012). The VSEs that MFIs serve tend to be family businesses in which the owner still works, but may also employ non-family members (Shorebank, CGAP 2012, IPC 2012). They are generally informal or semi-formal (IPC 2012). Their book keeping, organization and management is more sophisticated than most micro enterprises, but still relatively weak. Where formal Characteristics of a accounts and documentation exists, it is often unreliable. Very Small Enterprise They are more vulnerable and their growth tends to be more volatile than large enterprises (OECD 2006). They A Compilation of Qualitative Comments from Field are also a heterogeneous group with divergent financial Visits in Peru and Bolivia: needs (CGAP 2012, IPC 2012, Beck 2008). Business Characteristics Because of this informality and poor documentation, • Semi-formal VSEs often lack access to financing from traditional • May have tax ID or pay minimal or partial taxes banks, however, their financial needs are generally larger • Typically hire an accountant to file taxes and more complex than micro enterprises and exceed • Have informal employees (can include family) what MFIs typically can offer (IPC 2012). Initially they • Few formal financial records may need short-term loans to manage cash flow and Owner Characteristics basic current accounts, but they also need longer-term • Typically older (over 35) investment capital, credit lines, overdraft facilities, transfer • More men than women and payments services and business credit cards (IPC 2012, CGAP 2012, IFC 2010). VSEs financial needs are II. MARKET OVERVIEW 5 not as complex as medium enterprises that may also want investment products or more fee based services such as letters of credit or foreign trade financing. SE financial product usage in Latin America Transactional and deposit services are indispensible to SMEs, while credit access and needs are more variable (FRS 2006 as reported in de la Torre 2009). All firms interviewed in this study used some kind of deposit product, primarily current and savings accounts (See Appendix 1, Figure 2), and more than 95% used some kind of transactional product (mostly internet banking; payment of taxes, wages or suppliers, insurance; and other payments and transfers). The use of credit services varied from 35% (Mexico) to 87% (Chile). A more recent study of emerging markets banks (McKinsey 2012) illustrates that all MSME clients of surveyed banks have transactional products, while only about a quarter have loans. The studies suggest that demand is not the only constraint to credit access. Supply of products to SMEs likely plays a role as well. For example, countries with greater flexibility and financial inclusiveness (such as Chile and Peru) show greater usage by SMEs of loan products than countries with more rigid financial sectors such as Mexico and Venezuela. B. Overview of banks and how they serve small enterprises According to recent research, banks continue to show a strong interest in financing the SME sector worldwide and in Latin America. Seventy-seven percent of the banks in the IDB’s latest SME Banking Survey had a favorable outlook on the sector and expected to grow their SME business (IDB 2012). The main motivating factors for banks to downscale to the SME sector are to increase their profitability and are also based on increased competition in their traditional markets (de la Torre 2009, Beck 2008, IDB 2012). Nonetheless, despite an increased interest in the SME sector, most banks are still not successfully serving the full range of SEs.7 Their lending requirements are generally geared towards larger more formal small businesses, and their policies and procedures are generally very rigid. Most banks still take a collateral based lending approach versus a cash-flow or character based lending approach, and many VSE’s either lack pledgable assets or have assets with insufficient documentation to meet bank’s requirements. Additionally, the cost of obtaining documentation and registering collateral for many VSEs may be prohibitive in relation to the size of loan being sought. Most banks have a separate area/business line to serve SMEs, a few also separate SEs from medium enterprises (Beck 2008). Eighty-seven percent of the banks from the IDB’s SME Banking Survey have a separate small business unit. Sales are generally decentralized, while loan approval, risk management and recovery functions are centralized – very different than the traditional MFI model. Banks tend to view the SME sector as riskier and more costly to serve than their traditional segments. To help manage risks, banks downscaling to serve SMEs have adopted a variety of techniques, including asset based lending (USAID 2009, de la Torre 2009) and leveraging value chains (FOMIN 2012). Some also use credit scoring and standardized risk tools to 6 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA control costs, especially for SEs, although most banks report that credit scoring is only one input in the credit analysis process (Beck 2008, de la Torre 2009). Banks may also boost their revenues from SME clients by cross-selling other fee-based products (de la Torre 2009), with up to 60% of revenues coming from noncredit products (IFC 2009). C. Overview of non-banks and how they serve very small enterprises In recent years many MFIs worldwide and in Latin America have started to expand upmarket to serve VSEs. According to a recent survey done by CGAP, 49% of LAC MFIs said small enterprises were part of their current business strategy, and an additional 25% of respondents indicated that they were considering expanding into the small enterprise sector. MFIs cited “business growth” and “growing with their clients” as the main reasons for expanding upmarket (CGAP 2012). Early on, there may be substantial room for an MFI to expand upmarket with its existing clients, generally by increasing loan sizes to help reduce a client’s borrowing from other institutions in order to meet his or her full funding needs. However, to grow a healthy VSE portfolio, MFIs also need to look externally for new clients, especially as most MFIs in Latin America find that only a small percentage of their clients are actually able to grow to become a VSE. MFIs may have several additional advantages over banks in serving VSEs. VSEs tend to resemble the micro enterprises that MFIs are familiar with more than they do the large enterprises with which banks are familiar. MFIs are accustomed to working with informal enterprises and ones with limited or poor record keeping. As a result, their policies and procedures tend to be flexible to accommodate this informality. They also tend to be more agile and able to more quickly disperse a loan, which is likely attractive to VSEs. One MFI in Peru notes that it can obtain approval for a VSE loan in 8-12 hours compared to up to a month by banks in the market. In addition, MFIs tend to have closer, more personal relationships with their clients than banks, which can help to mitigate the risks involved with serving VSEs. III. PREREQUISITES 7 III. Prerequisites Prerequisites • Sufficient institutional capacity at all areas of the MFI to expand into new businesses • Perception of market access gap in the VSE segment • Regulatory environment conducive to SME lending at market interest rates • Ideally, access to a credit bureau that combines bank and MFI loans • Access to technical assistance or resources to obtain specialized know-how • Strong risk management framework, tools, and systems in place • Access to sufficient funding for portfolio expansion • Access to funding sources with maturities over one year • IT system that offers flexibility for multiple products and maturities • Ability to take savings or offer “seamless” bank-MFI transfers Expanding upmarket is not an easy task. It requires resources, systems, and capacity in excess of what is typically available when lending only to microentrepreneurs. As such, MFIs must give cautious consideration to this task and be prepared to determine that it is, in fact, not yet prepared to take on the challenge. A microfinance institution must be have the capacity to expand into new business activities in general, and VSE lending in particular, prior to embarking on the task of expanding upmarket. Reviewing the checklist in this document can offer a tool for assessing this preparedness as well as areas that may need to be PR Tip #1: Be Realistic. A realistic strengthened. assessment of readiness should be Ideally, the institution should be working within a presented by Management to the Board regulatory context that is supportive, rather than deterrent. and discussed. In Ecuador, for example, entrants have been discouraged This may be revised over time to reflect evolutions in from the VSE market because of interest rate caps on the market or institution. larger loans. A regulatory framework that requires the segmentation of loans or clients between micro, small and Example: One MFI interviewed for this document medium, can be helpful for tracking and benchmarking. has gone through two rounds of strategic analysis The metrics used for these definitions are often rough. and planning for their VSE lending. During an initial Reviewing institutional capacity in this segment should round, its Management and Board were enthusiastic not be a static activity (see PR Tip #1) as external and about the segment, and planned for growth internal conditions may change over time. accordingly. After three years, the same institution has re-evaluated its strategy. Its risk management, An assessment of readiness to enter the VSE segment in particular, is challenged and constrains the success should include a review of both the external opportunities of the VSE portfolio. While perhaps not sufficient to and the internal health and risk management of the discourage entering the segment, at minimum, prior institution. Externally, some assessment of the potential strengthening of risk management processes and market will be needed (see Section IX). Institutions tools would have benefitted the institution. may have a hard time quantifying this market if limited 8 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA public data is available or if their own market is much “lower” than the VSE market. Analyzing whether there is potential demand in attractive scale is important. Determining whether there is a gap in access or room for a new player can be even more important. Sometimes institutions seek to expand into new business activities because the institution is struggling in its current business and thinks this is due to market issues, rather than institutional weaknesses. Before entering into any new business, the institution should take a critical approach to its own health and ensure its current risk management and organizational capacity is sound. Internally, an assessment of institutional capacity PR Tip #2: Benchmark with institutions should look not only at management and staff capacity, in country and outside to “test” but also its capacity in terms of funding, strategic risk whether you are VSE ready. management, IT/MIS, human resources, legal, as well as its future potential to serve the needs of VSEs for related This may be revised over time to reflect evolutions in financial services. This can be difficult in a vacuum. the market or institution. Taking institutions that are offering loans to VSEs as Example: Most institutions interviewed for this rough benchmarks can help an institution measure document that were actively lending to VSEs its own potential. The relative newness of this market were market “leaders”. Most have assets over suggests that this be done with some caution, nonetheless. US$500 million with over 200,000 clients. Most Balancing risk and growth is critical when expanding to were regulated institutions and a large number VSEs, and an assessment of risk management capacity is took deposits. There were important exceptions, critical. External factors such as regulatory requirements however. One relatively new, regulated deposit- and supervision, the effectiveness and use of credit bureaus taking institution has approximately 50,000 clients and market risk can be very supportive. In their absence, and US$200 million in assets. In Colombia, one credit policies may need to be more conservative, and institution was not regulated, rather it was an NGO. analysis may require more documentation. An essential Example: Most instituions we visited used core prerequisite for an MFI to successfully expand upmarket banking systems for their MIS, which were flexible is strong existing credit risk management policies and enough to manage VSE lending needs. However, procedures. Furthermore, these must be flexible enough not all used sophisticated data analysis to segment to be adapted to a new segment. An institution that is loans, track commercial goals and manage credit struggling with credit risk must correct critical institutional risk, often to their detriment. weaknesses before taking on the risks of a new segment. Those institutions whose analysis may not be VSE-ready should evaluate available sources of technical assistance prior to venturing into VSEs. Most institutions we interviewed were deposit-taking or used core banking systems to track their loans. As such, their IT was relatively well positioned for VSE lending. At some stage, most MFIs will require more sophisticated MIS capacity that segments loans by size and client to assess both commercial goals and risk (See PR Tip #2). VSE loans require larger sizes and often, longer tenors. Most of the MFIs interviewed for this document had been flush with liquidity, and did not need to establish a funding strategy to move gradually into VSEs right away. As the VSE portfolio grows, however, matching assets and liabilities in terms of both costs and tenors becomes more critical. Many MFIs who were taking deposits found they needed sources of longer- term funding to match their longer VSE or SME loan tenors, albeit at a higher cost. III. PREREQUISITES 9 Over time, VSE lending should transform into VSE client relationships, where MFIs understand their clients and offer them a suite of potentially useful products and services. These could be as simple as transactional deposit accounts and wires and as complex as leasing instruments. While not always available initially, an MFI with the capacity to continuously expand its product offering may be best positioned to serve larger enterprises, including VSEs, small and medium enterprises. 10 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA IV. INSTITUTIONAL STRATEGY 11 IV. Institutional Strategy Institutional Strategy Checklist • Understanding of legal and regulatory implicaitons of expanding upmarket • Strong strategic rationale for entering the market • Clearly defined definition of VSE and target market • Understanding of market dynamics and realistic expectations of segment’s potential scale • SWOT analysis and financial projections for the segment • Board of Directors engagement • Clear mapping of activities required to implement the expansion • Identification of responsibility and accountability for implementation When an MFI is considering expanding upmarket to serve small enterprises, it must recognize that VSEs differ from micro and SME businesses and require their own approach to service and promotion, credit analysis, collateral, risk management and operations. The extent to which these approaches differ from micro will depend on the institution and the market and should be defined in the institutional strategy or the high level approach the MFI plans to take to VSEs. At the core of this strategy is identifying how to incorporate VSEs into the MFIs existing business. Generally speaking there are two types of business models that MFIs have adopted to serve the sector. First, the “Proactive Model- (P)” borrows from the know-how and processes most typical in banks, including those that have downscaled into VSE lending. It is more complex and, in many ways, more deliberate in that it requires important structural decisions and adaptations from the start. As in a bank, units are segmented with clear distinctions between the micro and VSE businesses. Clients are segmented and directed to specific units, which determines who will serve them and how. At the headquarters level, it involves setting up a separate business line for VSEs and corresponding new policies and procedures, credit analysis and risk management. The VSE unit can leverage departments shared with the microfinance unit, such as Marketing, IT and Human Resources, but it is also likely to develop its own marketing and sales strategy. At the branch level, there are separate loan officers dedicated to VSEs and generally a separate VSE area of the branch. The model has been actively disseminated by Pro Credit (IPC 2012) and implemented regionally throughout the Pro Credit banks in countries such as Bolivia, Nicaragua, El Salvador and Ecuador. It generally targets a business segment that is outside of an MFI’s existing client base, and may also be underserved by the financial system. As such, it required significant prospecting to identify clients. If these clients have been excluded altogether from the formal financial sector, they may have a limited credit track record and require a high level of proactivity in terms of developing risk management models and controls. 12 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Second, the “Organic Model- (O)” follows a more organic process. It is often a reaction to a trend observed as an MFI’s microenterprise portfolio begins to grow in asset size by more than its client growth, suggesting that larger clients are requiring larger loans and in some cases may be leaving when these are not available. While it may initially represent a reaction to client trends, it does require strategic decision-making and adjustments to the traditional microfinance business model. However, these adjustments are less pronounced compared to the Proactive Model. They typically allow the MFI to target a new market segment without major changes to its operational structure. Loan products and requirements, most obviously loan amounts, are adjusted and tweaked, as is the credit analysis and risk management. At the headquarters level, there may be a project or product manager or small team in charge of driving the new VSE segment, but there is not a separate business unit and no major changes to reporting lines. At the branch level, loan officers may serve both micro and VSE clients or there may be some segmentation, though generally not as strongly delineated as in the Proactive Model (P). Revising risk management policies is important in this model, yet credit risk may be contained initially by focusing on existing clients with an established track record, and later similar types of clients. Notably, this model is easier and faster to implement and less costly to operate in the short term. Both models offer advantages and disadvantages, and while there are examples of both operating in LAC today, there are also indications of “hybrids” that borrow practices from both sides to help balance some of the trade-offs between efficiency and depth. When determining which of these two models to adopt (or a hybrid) an MFI must consider their strategic rationale for entering the market, its target clientele, the market dynamics (competition and potential size) and the legal implications of the expansion. A. Strategic rationale for entering this niche There are a variety of business motivations for an MFI to expand upmarket to serve VSEs, some opportunistic and proactive, others more reactive. Amongst the most common motivations mentioned in Latin America are: Client retention: As their best microfinance clients grow into VSEs, many MFIs have found it necessary to increase loan sizes in order to continue to meet their financing needs. O “We were losing our largest clients to the banks…. We have cut our desertion rate in half since we started offering VSE loans.” – Colombian MFI Blue ocean opportunity: Very small enterprises have access to few existing financing options and thus the MFI has the opportunity to be one of the first movers into the market. P “We started making bigger loans for our existing clients, but then we saw that the banks were not serving VSEs and realized it was a really good market opportunity.” – Bolivian MFI IV. INSTITUTIONAL STRATEGY 13 Competition in traditional markets: MFIs may expand upmarket to serve VSEs if its traditional microfinance markets are oversaturated. P O “We are interested in expanding upmarket where competition is lesser and we can carve out a niche.” – Panamanian MFI High levels of liquidity: Increased liquidity due to increased access to debt financing, expansion of liability products or less demand from its traditional clients may motivate an MFI to enter the VSE space in order to put these funds to work more quickly. P “There is excessive liquidity in the system. We have even put limits on deposit taking.” – Bolivian MFI Strategic long-term move: For some MFIs, an expansion to serve VSEs may be motivated by its long-term goals of transforming into a regulated financial institutions or a bank. P Many of Pro Credit’s NGOs began the transformation process early vis-à-vis competition in the microfinance sector. This positioned them well to offer a broader set of services to a higher market. Asset diversification: Expanding upmarket can help MFIs to diversify their asset base and thus their risks. When VSE clients have graduated from micro, they can represent a lower risk segment for an MFI. P O Growing with clients allowed a Peruvian MFI to keep its best customers and maintain a more diversified portfolio including VSE, micro and rural loans. Regulatory or public policy changes: Some MFIs may be enticed to enter the space because of incentives offered by governments or donors, such as guarantee funds or below-market financing. Regulatory constraints may keep actors out of the space. In Ecuador, interest rate caps on larger loans have kept some MFI players out of the market as they are not able to lend profitably. In Costa Rica, the lack of access by MFIs to the credit bureaus used by regulated banks increases the risk of lending to VSEs that may have had access to bank loans and performed poorly. P O Negative regulatory policies may also push MFIs into the space. For example an increase in the profit taxes in Bolivia has led some institutions to abandon their smallest clients. 14 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Alternative revenue source: Although interest rates on VSE loans are generally lower than micro loans, compressing financial margins, serving smaller VSEs can still produce attractive returns as operational costs per dollar lent tend to be lower and for clients with a good credit history loan losses are smaller. Over the long-term, VSEs may also offer “total client profitability” versus just “product profitability” if MFIs develop models that offer a range of products to VSE clients. Furthermore, in some segments and market contexts, where economic growth is dynamic and VSEs are benefitting from growth themselves, VSEs may be less interest rate sensitive. MFIs can offer these high growth enterprises more appropriate loans of larger sizes, while reducing the need for enterprises to borrow from a variety of institutions, which is a cost savings in itself. PO One MFI in Peru has identified the small enterprise “sweet spot” as those businesses with funding needs of around $20,000. Their prospecting is geared to clients needing $20,000 as they are not as interest rate sensitive as larger clients, are not on as many banks’ radars, and most importantly, are posed to grow and represent a long-term investment for the MFI. Figure 1. Drivers of Proactive and Organic VSE Models for MFIs Proactive Model Organic Model Blue Ocean Client Opportunity Retention Long-Term High Competition Strategy Liquidity in “Micro” Asset “Leap-up” Diversi cation “Step-up” Large SE Regulatory Limited SE Market Potential Incentives Market Regulated Institution NGO IV. INSTITUTIONAL STRATEGY 15 B. “Step-up” or a “Leap-up”?: Who is the target market? The second consideration when choosing a business model for serving very small enterprises is a thorough understanding of who the target market is. Our research has found that MFIs tend to serve an VSE segment with some common characteristics (with variations by country and sector) which include a relatively low level of formalization, record keeping that is somewhat informal and “in house”, and a larger scale and volume of sales than traditional microenterprises (see Section V and VIII). For MFIs whose primary motivation for entering the VSE space is to retain their existing clients, the move likely Example of a Step-up constitutes a “step-up” from their micro clients. They “We have all been offering large loans to micro need larger loans with potentially longer tenors, but are enterprises without having to develop a different likely still informal and have poor record keeping, thus technology for lending. These are enterprises that are credit analysis based on microenterprise lending including growing, but still behave like microenterprises. They character references and cash flows “created” by loan are informal, pay no or low taxes, employees don’t officers are still very relevant. Even some MFIs whose have fixed salaries….Slowly, there is a development expansion motives may be more proactive may find that of real small businesses. They are registering in their target market is not very different from their existing FUNDEMPRESA and separating family from business. client base and still just a “step-up”. In both these cases, They require a different analysis. “ the Organic Model may prove sufficient to successfully expand to serve VSEs. - Director of a Bolivian MFI However, in some cases, an MFI may find that the VSEs it hopes to serve have significantly different financial needs, require much larger loan amounts and/or that the sales, analysis or risk management of these clients is wholly differ than their micro clients. Thus, the expansion constitutes a “leap-up.” This may be because their existing market niche is quite low. For example, when they may be using village banking or group lending to serve this niche; hence the expansion requires not just tweaks in their lending methodology but the development from scratch of an individual lending methodology for VSEs. The expansion may require a “leap-up” if an MFI decides to target a slightly larger and more formal niche than that of their faster-growing micro clients. In these cases, especially if there is also sufficient scale (see below), an MFI is likely to find that the Proactive Model allows it to better organize itself internally to serve VSEs and microenterprises. It is important to note, that even though an MFI may initially enter the VSE space in order to continue serving their best clients, most that are successful at VSE lending do eventually expand to serve other VSEs as well. For example, the VSE portfolio of one Bolivian MFI that began VSE lending in 2006 is now comprised of 50% new clients and 50% old clients. New clients already make up approximately 26% another MFI’s VSE portfolio in Peru and they have been offering VSE loans for less than two years. So long as the new clients are similar to the original market niche identified by the MFI this should not dramatically impact the MFI’s choice of VSE business model. 16 MICROFINANCE AND THE GLOBAL FINANCIAL CRISIS – A CALL FOR BASEL C. Considerations of Market Context The market context of the VSE business is another major factor in deciding which business model to adopt. When evaluating the market A SWOT analysis is critical to refining the approach context, an MFI must look at the size of the target market, for entering the VSE market. MFIs can use this the existing competition and its competitive advantages in document as a framework for thinking about their serving VSEs in order to determine the potential scale of own competitive strengths and weaknesses for the VSE business for them. This scale must justify both serving this market, comparing these with their entry into the market as well as support the operational planned services with both banks and informal costs of the business model. The larger the scale, in number lenders in the market. of clients, loan sizes and breadth of product offering, the greater the argument to adopt the Proactive Model. Size of the target market: When determining market size, it is not sufficient to simply look at the number of VSEs in the country. An MFI must also consider how many are located in the areas in which the MFI operates and what their financial needs are, both credit and non-credit. An MFI should try to quantify how many “high-potential” microenterprises there are, especially those where financing is a major constraint to growth. To do so, it must take into consideration the overall health of the market. Is the potential market limited by a small country or weak economy? Are there other non-financial constraints to the development of VSEs in the market that would affect the MFI’s business? Competitive environment: In addition to determining the size of the market, the MFI should consider the competitive environment and how this may impact how they choose to serve VSEs. Do the VSEs in the market already have access to financial services or is this a blue ocean opportunity? Is the market slightly, moderately or very competitive? Who is the competition (banks, other MFIs, other types of financial institutions)? It is also important that an MFI investigate the types of products and services others are offering to VSEs. Although there may be some financial institutions already in the market, their products may be inappropriate for VSEs but VSEs are bending products tailored for smaller or larger firms to their needs as there are no appropriate alternatives. For example, they may be cobbling together multiple loans to meet their full credit needs or using personal credit cards for business purposes. They may also be accessing bank loans based on personal credentials and/or collateral as opposed to business assets, and thus taking on great personal risk. Competitive advantages: Given the size of the market and the competitive environment, the MFI must then determine how much of the potential market it can reasonably capture. • External considerations: The MFI should consider how it can compete with the existing competition (See Section V below). Is there a niche currently not being served? Is it able to offer more appropriate products specifically geared towards VSEs that perhaps do not already exist? Will it be able to provide better, faster customer service? Will its collateral and/or lending requirements be more flexible, leveraging is microfinance background, and thus be able to serve VSEs excluded by existing providers? Will it compete on price, and if so, is this sustainable? IV. INSTITUTIONAL STRATEGY 17 • Internal considerations: An MFI must also consider where there are internal factors that may limit the growth of its future VSE business. For example, does its mission limit the resources it is willing to dedicate to VSEs or are there legal (see below) or IT constraints (see Section XIII) to offering a breadth of VSE products? Additionally, expanding to VSE requires the involvement of all areas of the MFI, not only the commercial areas. Does the MFI have the capacity and resources to coordinate and mobilize managers across business lines or is it under-resourced or too busy with other projects? Another key consideration is funding (see Section XIV). VSE loans turn over more slowly and use up available funds more quickly, so this may limit an MFI’s ability to grow quickly. D. Considerations of Legal Structure The legal structure of an MFI is another influential factor in determining the type of business model to adopt for VSE How Interest Rate Caps Affect VSE lending. The more formal and regulated an institution, Lending –Ecuador generally the more it makes sense for the institution to adopt the Proactive Model. These institutions are more Although interested in moving into the small likely to be able to offer a broader range of products to enterprise space, one Ecuadorean MFI notes that it VSEs (short and long term loans, savings and current has not made the move because of the government’s accounts, payments and transfer service). It may make size based interest rate caps. The current caps sense for the development and provision of these services are 28% for loans under $10,000, 21% for loans to be more centralized as in the Proactive Model. The between $10,000 and $20,000 but only 11.8% for ability to cross-sell a variety of products and increase the loans above $20,000 for businesses with annual revenue generated per customer can also help to offset the sales over $100,000, range in which most VSEs fall. higher operating costs of the Proactive Model. The spread over their financing costs, 7-8% is just not sufficient. Although the MFI is regulated and On the other hand, an NGO that is only able to offer takes deposits, they have to offer a high interest rate basic credit products may be better served by the Organic of these deposits in order to attract deposits from Model as it is less costly to put in place and to operate on the more traditional banks and the repatriation fee of an ongoing basis. For these MFIs there is less of a need to foreign funds also raises their financing cost. have a separate business unit for VSEs as the breadth of product offering will be limited and less complex. (See section IV below). The legal structure of an MFI also has other implications on their entry into the VSE space. Most notably, it may affect their ability to raise funds and the cost of the funds, with deposits generally a far cheaper source of funding that external loans. Thus non- deposit taking institutions may find it more difficult to compete for VSE clients if the market is very interest rate sensitive. The legal structure can also affect risk management. Regulated institutions are obliged to comply with provisioning requirements, which may vary for micro and VSE loans, guarantee requirements and client information requirements. For example, in Bolivia clients borrowing over $15,000 must have real guarantees and MFIs must be prepared to evaluate and register these guarantees. Non-regulated institutions may be able to side step some of these regulations and serve VSEs regulated institutions cannot. 18 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA However, they may face other constraints because of their legal structure, just as not having access to credit bureau information, which negatively impacts their ability to assess willingness to pay. E. Other Considerations Regardless of the business model chosen, getting board of directors and senior management buy-in to the VSE expansion plan is essential to an institution’s success. Once the big picture strategy has been agreed on and the basic business model defined, the institution can start mapping out the project components and how the necessary changes and/or developments in each area of the institution. A Steering Committee than includes members from all relevant areas of the institution can work to define how and in what order these changes need to occur. Management should also clearly define who has responsibility and accountability for the implementation of the project. Some MFI’s have found it incredibly useful to have a project coordinator in charge of overseeing the implementation process and ensuring that each relevant area is on track. For example, an MFI in Colombia encountered many delays and setbacks in the first six months of their implementation because they did not have a project coordinator to ensure work was being done on all fronts. It is also important to remember that the VSE business model may change over time as motivations, market context or other the characteristics of clients change. For example, when one Bolivian MFI first starting serving VSEs in 2006, they employed the Organic Model as they were mostly growing with their clients. However, they observed that at the time there were very few other institutions serving this market niche and became more proactive in seeking out new clients. Since then they have transformed into a bank (in part to increase funding sources for VSE lending), and migrated to a more Proactive Model to lending to VSEs including specialized loan officers in branches where the size of the market allows for it and more in-depth loan analysis of VSE loans, but not a separate business unit for VSEs. V. DEFINING THE VALUE PROPOSITION 19 V. Defining the Value Proposition Defining the value proposition checklist • Define specific small enterprise segments or subsegments to target • Identify a market gap in access to appropriate small enterprise loans • Define the institutional value proposition to VSEs – Is it compelling? Above we discuss two basic models (or a combination thereof) that MFIs can consider when moving into the VSE market. A critical consideration in both the Proactive and Organic Models is the definition of the target market and a keen identification of the value proposition that the MFI is offering this market. When determining a market strategy, identifying and articulating a value proposition is the first step in determining a market strategy. Additionally, an MFI should assess its value proposition vis-à-vis the alternative sources of financial services, both formal and informal used by VSEs. Keeping this in mind will assure that the MFI’s analysis of the potential market does not “over-shoot” and can help define a more precise commercial strategy based on serving segments rather than offering products (See section VI). “Ants Dressed as Elephants” Because of the broad variation in regulatory definitions and firm sizes of MSMEs, common practice has moved into accepting a definition based on loan sizes for financial institutions lending to MSMEs. This simplification can distort results, however, potentially either over- or under-estimating business size. One Bolivian MFI noted that some VSEs are “ants dressed as elephants” referring to the likelihood that many microenterprises are being characterized as VSEs because of their loan size rather than their own characteristics of organization, employees, or asset or revenue size. In other cases, “elephants” might also be dressed as “ants” when there is no information about the total outstanding loans of an enterprise. One Bolivian MFI we visited for this publication noted that based on loan size, some 40% of clients were VSEs, however, using the government’s definition of VSE that takes into account sector, sales volume, assets and number of employees, only 17% of clients were defines at VSEs. Different VSE Segments Obtain Different Value The value proposition of MFIs to serve VSEs may vary depending on how VSEs are defined. VSEs are not homogeneous but cover a broad range of characteristics including employees, size, formality, tax status, and their role in a commercial and service ecosystem. As discussed above, Organic models tend to target largely informal businesses that have outgrown “micro” loans but still require a credit technology that draws from microcredit where financial statements, invoices and receipts are incomplete or perhaps do not exist. MFIs’ value added compared to banks in this 20 MICROFINANCE AND THE GLOBAL FINANCIAL CRISIS – A CALL FOR BASEL segment is high, as they can retain good clients without Figure 2: Example of Peruvian MFI SE requiring onerous new processes. The value proposition Sub-Segmentation by Loan Size begins to weaken when the VSEs targeted have greater levels of formality, or may be large enough to require new technologies and larger capital allocations that may be beyond the scope of many MFIs. To the extent that MFIs Lower Value of MFIs US$ 50,000+ can begin to segment their clients more granularly, taking into account that those that receive US$10,000 have different needs than those who borrow US$100,000 and that their formality, size and financial and management US$ 25,000 - 50,000 capacity can vary substantially, they can also differentiate Higher Value the types of service offered. US$ 7,000 - 25,000 of MFIs To better distinguish and serve the various segments within “small enterprise”, one MFI in Peru has developed three sub-segments of SEs that are defined by loan size but also help to differentiate between three types of business: 1) the very small partly informal enterprise; 2) the formal small enterprise; and 3) the “corporate” small enterprise (See Figure 2). As a rule of thumb, the MFI considers that those enterprises in the smaller loan ranges are more likely to lack formal employees or records while larger businesses may be more formal, paying taxes, being registered in public records of chambers of commerce, or do business with large government or corporate entities. By sub-segmenting SEs into these categories, the MFI is better able to specifically tailor its products and services and policies and procedures to each segment, maximizing its value proposition. However, this method does not avoid the problem of potentially dressing “ants” as “elephants” where microenterprises may be receiving loans that are considered “Very Small Enterprise” products. Considering the MFI Value Proposition: Banks vs. MFIs Defining VSEs and sub-segments within the VSE category can be useful in determining the specific value proposition of an MFI. Both banks and MFIs can be well positioned to serve VSEs. The value proposition of banks typically lies in that they can offer lower interest rates on loans, larger loans sizes, transactional accounts, and slightly more complex products and services that some growing small VP Tip #1: Consider the value businesses may need. For VSEs with less complex financial proposition of an MFI vis-a-vis both needs, there is still value in working with a bank as it will formal and informal alternatives when be able to meet the VSE’s more robust financial needs as defining the target market. it continues to grow and develop. This creates a potential incentive for clients to build a relationship with a specific Example: One MFI in Peru has determined that banks bank to obtain access to services in the future. Nonetheless, rarely lend less than US$30,000. Thus, they have MFIs can offer VSEs access to finance when banks may not determined that lending up to $30,000 is the most be comfortable taking a risk. This MFI value proposition appropriate market niche for their VSE strategy since appears to be most compelling when MFIs’ VSE clients are customers in this nche do not have access to lower- informal. Unregistered and lacking financial statements, cost bank loans. these VSEs are often unable to access bank loans (in some limited examples of countries with extensive downscaling V. DEFINING THE VALUE PROPOSITION 21 of banks). When MFIs move into a more formal VSE or SME segments, the scale tips against them in terms of their value proposition and they can be pushed to reduce interest rates to levels that may no longer support the business case for VSE expansion. It is critical for MFIs interested in this segment to consider three main commercial issues: • Are VSE customers able to access loans from alternative sources and what are these? • What type or sub-segment of VSEs can they offer the most attractive value proposition to? • How much flexibility will the MFI have in reducing interest rates and increasing loan sizes, while still remaining agile in their loan methodology? A Personal Touch One value proposition often referred in the literature is the “personal” service that MFIs can offer. Low-income clients might feel uncomfortable or intimidated walking into a traditional bank, and are more at ease with MFI services and personnel that are designed to be more approachable What are the comparative advantages and respectful regardless of clients’ education, financial of one MFI in Peru status or experience. One MFI in Bangladesh explained: Vis-a-vis banks?* “For us, VSEs are our corporate clients, our priority clients. We treat them very well, while for the banks they are their Flexibility: Some VSE borrowers who have been with least important client and they often look at them with this MFI for some time noted that at the beginning, disdain.” While anecdotally this “personal touch” is often had to provide a lot of documentation for a loan, but cited as an advantage, it is difficult to substantiate. now could skip this step to get a loan more easily with the MFI. Interviews with clients and loan officers at four leading Speed: The disbursement of loans in one or two days institutions in Peru and Bolivia offered some additional was noted as an advantage, however, in one case, a nuance on this topic. Indeed, MFIs are seen as offering more customer complained of a three day turnaround. personal service. But probing more deeply into the meaning of “personal” is key to understanding the relevance this may * From interviews with 5 small business clients have in terms of value added for small business clients. Some clients interviewed for this report meant personal when they said “personal”. They were strongly attached to their loan officer, and perhaps the MFI that helped them get on their feet when they were much smaller businesses and much riskier clients. However, other clients seemed to associate “personal” service with MFIs’ processes and products than with their personal relationship with the MFI or loan officer (See box above). These may include more flexible requirements for larger loans, speedier loan disbursements or less cumbersome documentation. One General Manager of an MFI in Bolivia notes: “clients have grown more sophisticated, they compare loan terms and they are not uncomfortable walking into a bank as they were once”. The box above illustrates some of the specific characteristics clients of one MFI in Peru perceived to be the MFIs comparative advantage. While loan officers and clients alike may refer to this as “personal” service, it is not limited to the relationship between the loan officer or the MFI and the client, but refers more to the MFI’s more flexible approach to informal lending to VSEs. A personal touch is a component, which is perhaps more relevant where banks are not reaching VSEs, but should not be overestimated if MFIs want to reach a broad range of VSEs. 22 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Considering non-Bank Competition for What are the comparative advantages MFIs of one MFI in Peru Banks are of course not always the only alternative source vis-a-vis non bank alternatives?* of financing for VSEs. Although Peru’s banks have been scaling down to reach some of the relatively smaller urban Convenience: A client reported that in the past businesses, in many countries in the region, this is yet to be he had managed four loans at once. However, the norm. This does not protect MFIs from competition this was a logistical headache and said that he in the VSE segment, however. Alternative funding sources learned that it is not worth your time. Instead, can include: 1) multiple MFIs, combining various “micro” he preferred a single larger loan, even if the loans to meet their needs; 2) supplier loans and leasing interest rate is slightly higher. contracts; or 3) informal sector loans. These alternatives Flexibility: Suppliers may be able to lend VSEs are likely to be less desirable to VSE clients than one larger money for specific inputs but cannot help fund loan from an MFI or bank. Loans from multiple MFIs the growth and expansion of a business. may be difficult to manage and organize while supplier loans can be restrictive, as they do not offer cash in hand. Supplier loans are quite common but typically used only to buy equipment or merchandise, thus limiting their flexibility, especially when loans are aimed at investing for expansion. Informal loans from friends or family can be attractive, often offering lower interest rates, but may not reach the amounts required by customers. Additionally, they may come with the social burden of expected reciprocity. In summary, the value proposition of an MFI can be twofold: it can offer VSE clients a personal touch, where clients and their loan officer or other staff build loyalty and trust over time, and perhaps more importantly, flexible terms that are customized to the constraints of informal VSEs where clients may not be able to meet the documentation, collateral and other requirements of bank loans; or, it can offer VSEs the opportunity to consolidate their financing with one institution, reducing the non-financial costs of borrowing This section and sections VII and VIII on Credit Methodology and Risk touch on some of the repercussions that this comparative advantage may have on an institution’s credit risk management practices, where flexibility must be balanced with a higher risk of making larger loans. VI. COMMERCIAL STRATEGY 23 VI. Commercial Strategy Commercial strategy checklist • Target the market (s) where the MFI’s value proposition is strongest • Mine existing data on large loans to assess initial potential demand and needs • Implement additional market studies or leverage existing studies • Define products to meet the needs of these target segments • Collaborate with Finance, Risk, and Credit departments to ensure products can be offered sustainably • Collaborate with HR department to identify HR requirements • Identify an appropriate prospecting strategy for reaching existing and new clients in SE • Identify a marketing strategy that includes VSEs without excluding “micro” segment A. Understanding Your Market Proactive and Organic models may approach product development differently. A Proactive model is more likely CS Tip #1: Market studies can start with to define a target market that includes a “blue ocean” a relatively low-cost analysis of existing market that is largely unserved by the financial sector. Its clients focusing on “larger”client product development process should be more deliberate, performance. including a hypothesis of potential needs, client surveys or qualitative market research and a pilot of a new product Example: A Peruvian MFI we visited identified that tailored to the needs of this market. One institution using its larger micro clients were borrowing from multiple a Proactive model interviewed for this publication noted sources, likely because their investment needs were that it had implemented both market studies and pilots higher than what typical microenterprise working prior to offering VSE loans, while most of the institutions capital loans could offer. By offering larger loans, using Organic models had not. Organic models tend they were able to better serve and retain these to instead make small adaptations to existing products clients. to meet the needs of existing clients. This “learn as you go” approach, can be less expensive and require fewer human resources to oversee the process. However, it may limit the potential for innovation or out of the box thinking, in particular in developing products that can help retain good clients. VSE market studies are often not prioritized within institutions because they can be costly, time consuming, and ultimately offer insights that an MFI may feel it already intuitively knows about its clients. It is important to consider various types of studies and approaches that suit an institution’s budget, time and existing knowledge. It is also important to think about studies that may offer insight about a range of market segments spanning from micro to medium enterprises to leverage the investment. It may also help identify potential products and services that can be offered to various segments rather than only a narrow slice of VSEs (See CS Tip #2). 24 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Where little is known about a VSE market, institutions CS Tip #2: Market studies can help fear that the cost of being a first mover outweighs the identify specific VSE segments and benefits. One institution in Central America interviewed their product needs. Some of these for this study suggested that market studies should new products may also be useful for be a public good, offered through technical assistance growing “micro” clients or larger grants by donors and made accessible to all institutions “small/medium” size firms. in the market. When one MFI invests in developing such know-how and appropriate products, it is likely Example: One Proactive model MFI implemented that others will soon copy these products thus saving an in-depth market study of its MSME clients to the investment and potentially threatening the market identify VSE needs. It identified a process by which share gained by a first mover. enterprises’ needs evolve from working capital to investment loans, then current accounts, money transfers, credit cards and bank guarantees. By B. Product Development offering this broad range of products, the institution felt that it could capture a range of VSEs from those Characteristics of Predominant Credit that were small and slowly starting to need new Products products to those that already required a broad As microenterprises grow into VSEs, their needs for range of products. financial products and services may change. In Latin America and the Caribbean, we assume that these microenterpreneurs will have had some access to informal or formal loans, and begin with an analysis of credit CS Tip #3: Market studies in institutions needs. While this may oversimplify the needs of a very adapting a more Organic model for small business, it also reflects the revenue drivers that entering the VSE market can start small, may compel MFIs to engage this segment. VSE loans, beginning with an analysis of existing at minimum, will have larger loan sizes. Often this is client data and behavior of larger driven by the needs of clients as they grow. Throughout clients. this publication we discuss some of the consequences of making larger loans on credit methodology, credit risk, Example: While few institutions interviewed for financing and operations. In particular, we note that this study used data mining techniques to analyze pressures in competitive microfinance markets may lead client behavior and needs, most had access to quite institutions to increase loan sizes for microenterprises that detailed data about clients’ loan size, business size, may threaten to leave the institution unless they obtain and growth. This information can be a first step in a larger VSE loan. Throughout our interviews we noted segmenting ranges of VSEs and identifying which the tension between an institution’s need to manage its might be more able to handle larger loans. credit risk as loans become larger, and the pressure on loan officers to retain good clients are often at odds. This tension is especially critical in VSE lending because as loan sizes increase, risk cannot be spread out as easily among multiple loans, and loan officers are compensated on their success securing a smaller number of larger loans. The additional credit risk implied by VSE lending can be offset through greater guarantees, where more significant collateral is required. This comes with its own set of complexities as we discuss in Section VIII, including the lack of documentation or difficulty in assessing properties. The design of VSE loans and their collateral requirements must take into consideration not just the capacity of VSEs to provide real collateral, but also the costs of documenting and registering this collateral and their impact on the overall cost of the loan. VI. COMMERCIAL STRATEGY 25 With size, loan tenors may also increase, especially if loans are being used for investment purposes. Many MFIs CS Tip #4: Credit and savings products offering small business loans have been offering loans for that can increase the “stickiness” of up to four years rather than traditional 1-year or lesser a relationship with an MFI can help terms. Nevertheless, this is not always the case. MFIs retain good clients and improve their interviewed for this study often noted that VSE clients profitability. One example is credit can have a hard time with the discipline of paying back lines, which are often well-regarded by loans over a longer time period, additionally, longer loans these customers because of the ease of require a more disciplined engagement between the loan requirements, speed of disbursement, officer or MFI and the client, since contact can be more limited. We have observed that VSE loan tenors are more and flexible payment terms. commonly of two years and under in this segment. Additionally, credit lines strengthen a relationship of trust between a client Additionally, to avoid this risk of losing the client and an institution that can foster relationship, some MFIs have worked on increasing the retention. “stickiness” of their client relationships (See CS Tip #4) that can be looser when there is a less frequent interaction Example: A leading Costa Rican MFI believes that between a loan officer and a client with longer-term loan credit lines have helped to ensure the retention of maturity. some of their better small business clients who make frequent withdrawals on credit lines vs. Much less Most importantly, interest rates are typically lower for frequent business loan withdrawals. In Peru, an VSE loans than for microfinance loans. In most MFIs MFI notes that offering credit lines can drive small interviewed, these rates are not set as a result of a costing business client loyalty. exercise but based on market trends and client demands. In more competitive markets VSE loans are offered at much lower rates than micro loans. Table 1 offers an example of one MFI’s interest rate schedule for larger USD Table 1. Example of interest rates by loan loans, illustrating the relationship between loan size and size of a Bolivian MFI interest rates. Loan amount USD Annual Interest Rate For MFIs that offer VSE loans to existing clients, reducing 2,000-5,000 26% + interest rates does not always mean switching clients into a 5,000-10,000 23%+ different product as they grow out of microcredit and into 10,000- 20,000 20% + VSE loans. In practice some MFIs using Organic models, offer their best clients a “preferential ” interest rate rather 20,000-30,000 18% + than VSE Product. Where countries are regulated with 30,000- 40,000 17.5% + interest rate caps on VSE loans, this type of pressure is 40,000- 50,000 17% + exacerbated. An Ecuadorean MFI manager interviewed 50,000- 100,000 16% + for this study noted that they do not make loans over US$ 100,000 + 15% + 20,000 because the government caps interest rates on those loans at 11%, which is not sustainable for the MFI. 26 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA One MFI visited for this study offers an interesting example of how interest rate pressures can be detrimental to the business case. In Peru, where the VSE segment is highly competitive, managers complained of pressure from loan officers to reduce interest rates for their best clients for fear of losing their business. As this segment likely already has lower margins than the microenterprise segment, this type of pressure can ultimately lead to a decline in profitability and potentially a decision to exit or reduce exposure to VSEs. Characteristics of other Credit Products MFIs interviewed for this study note that credit product needs for VSEs are not significantly different from those of microenterprises. They may be more inclined to borrow for investment (and thus take larger loans at lower interest rates) than microenterprises. In general both fixed investment loans and working capital loans are attractive for VSEs (See CS CS Tip #5: VSEs are more likely to Tip#5). Additionally, VSE owners will take on mortgages, borrow for investment and working either for the business or the owner’s own home. These can capital equally. Both products should be be used to build or expand workshops or sales outlets or made available. improve the living standards of VSE owners. Credit lines are also attractive, offering VSEs the opportunity to borrow Examples: One MFI in Peru noted that about half only what they need and pay these down as soon as they the loans in the VSE segment were for investments are able. Other products such as factoring and leasing may vs. Working capital. be interesting (See Box), but are still rare in MFIs, partly because of the informality of the clients in this segment. Additionally, regulatory restrictions can be a limitation to offering these products. In Peru, MFIs were only authorized to use factoring in 2008, for example. Factoring can be attractive to those VSEs working with small contracts from government or municipal offices or larger companies. There may be demand limitations that constrain MFIs from developing these products, however. Firstly, VSEs are often concentrated in commercial or service sectors, where clients are retail or individuals, even some producers sell their products in the retail market. Additionally, even when VSEs are in productive sectors and sell to larger buyers, these may not always be willing to accept an invoice and tax registration, as they themselves may be informal. Leasing tends to be offered in even fewer cases, and opportunities are perhaps more limited. The same reasons that limit factoring may limit demand for leasing products. Additionally, those VSEs that buy equipment or vehicles often buy these second-hand. One MFI client interviewed for this study bought a complex machine for cleaning and recalibrating car engines directly from a producer in China instead of in Peru through a distributor. This reduced the overall cost of the equipment, but also limited financing options for this US$15,000 machine. VI. COMMERCIAL STRATEGY 27 Non- Credit Products VSEs, even more so than microenterprises, may have financial needs beyond typical credit products. For MFIs hoping to remain in the segment for the long-term the identification of these needs and long-term strategies on how to meet this needs, either internally or through CS Tip #6: Banks earn money from VSE partnerships, is essential. Moreover, taking a broader, relationships beyond loans and MFIs longer-term approach to VSEs can increase the profitability can begin to look at profitability in of the relationship with a VSE, as it incorporates not only terms of total customer profitability the VSE loans, but also the relationship with the enterprise, its owners, and family members and employees. Banks over time and across products. often approach this segment with a strategy that relies on Examples: Letters of Credit. A Bolivian MFI cross selling vehicle loans and mortgages to owners, family interviewed for this study is offering its SME clients members and staff. Additionally, banks can earn revenues Letters of Credit, a fee-based service targeted at a and ensure a longer, “stickier” relationship through small segment of the market and typically offered by fee for service products. For MFIs, which often have a banks, it has helped VSEs that were otherwise unable more limited product offering, this holistic approach to to participate in public procurement opportunities. customer profitability can be tougher to realize. “Leading” These can be based on existing CDs, savings MFIs, which typically have a broad product offering have accounts, credit lines or fixed asset guarantees, mimicked some of these strategies with credit products, many of these offered by the MFI to its VSE clients. offering vehicle and mortgage loans to VSE owners and their families. For smaller MFIs with fewer product offerings, VSE profitability will rely primarily on the revenue stream of one credit product. While cross-selling credit products can often be a way to achieve greater profitability from VSE clients, in low-income VSE markets in Latin America, there is less scope to cross sell investments, deposit services and fee for services as with larger businesses. Nevertheless, non-credit services at no fee or low fees have been popular with VSEs in some of the cases interviewed for this study and may offer a way for MFIs to boost profitability and create “sticker” relationships. One example is a Bolivian MFI that we interviewed, which offers letters of credit for clients seeking to work with local governments or other contractors that require these (See CS Tip #6). Few MFIs, including banks and regulated entities have developed current accounts for VSE clients, but customers are using other types of products as a transactional alternative. For example, one MFI in Peru offers savings accounts that look much like current accounts. These have no limits or charges on deposits and withdrawals, low minimum balances and debit cards linked to an ATM network. While savings balances are low, this is not necessarily a reflection of the lack of interest in the accounts, but instead of the transactional purpose these are used for. Additionally, VSEs with savings accounts can use these to make payroll disbursements to employees directly at the MFI’s cashiers without issuing checks. They can also make transfers between cities to suppliers or clients at no cost. In Bolivia, one MFI has introduced international wire transfers to meet the needs of VSE clients that buy products directly from China or other countries. While such services are not typically used to diversify revenue, they can improve customer satisfaction and “stickiness” as well as generate demand for future fee-based products. 28 CLOSING THE CREDIT GAP FOR FORMAL AND INFORMAL MSMEs MFIs that are not regulated to take deposits can be CS Tip #7: Smooth backoffice systems especially disadvantaged in the face of VSE needs for and partnerships with banks that are non-credit products and services as transactional accounts not competing for the segment can such can be a critical and attractive product for this help overcome the challenge of limited segment. Technology offers the potential to overcome product offerings by non-deposit taking some of these challenges by linking MFIs to banks, non- institutions. bank correspondent networks and other service providers seamlessly in countries where these are not in direct Examples: One MFI we spoke to in Costa Rica is competition with MFIs for the segment (See CS Tip #7). not authorized to take deposits. Clients receive loan disbursements directly through partner banks and are able to deposit payments into these accounts. Convenience and customer service Customer service, in particular convenience is a critical need for VSEs and can be defined in a variety of ways. Interviews with 12 VSE clients of MFIs in Peru and Bolivia for this document revealed that there are a variety of conveniences and time saving services that VSE customers either used or asked for. Not all were products, and in many ways, services were more relevant to these clients. These included: Save travel time: • ATMs or non-bank correspondents in peripheral areas of large cities where branch presence was low • Branch offices near businesses when available • Frequent loan officer visits to businesses to offer new products, check in on business performance Save money and time hiring bookkeeper or tracking accounts • Lack of requirements for formal receipts, invoices, and tax documents for obtaining a loan Save travel time and/or cost of wiring funds CS Tip #8: Balancing convenience and • Inter-city wire transfers for no cost or low cost to send or “touch” is critical when automating receive payment for merchandise client relationships. • International wire transfers to facilitate imports of machinery, inventory and other goods Examples: A specialized MSME Bank in Bolivia spends time training small business clients to use “Personal service” on-line banking to reduce transaction costs and • Friendliness of staff at branches and in the field branch time but their competitors note that this • Ability to address problems and offer solutions financial institution does not scrimp on relationship • Understanding of the clients’ business needs banking, offering an extensive period of time with • Personal relationships between the loan officer and client customers, understanding their business and providing customer support on commercial issues. The section above discusses the preference for “personal service” of some VSE clients, suggesting that this can play an important role in strengthening client relationships and ensuring “stickiness” and longevity in a client relationship. This is a critical consideration when determining to what extent low-touch conveniences such as online services have to be balanced with effective, and efficient personal service. Customer service encompasses both high and low touch points and can be an important differentiator. VI. COMMERCIAL STRATEGY 29 C. Prospecting and Marketing Identifying the appropriate VSE customers can be a challenge. Typically, an institution implementing an Organic model for reaching VSEs will begin by servicing its own “graduating” clients. Over time, as they build experience with these clients, they seek out new VSEs. We visited two intuitions in Peru that have been serving VSEs actively since 2009. In both cases, VSE loans represent approximately 40% of their total portfolio volume. In one institution, this represented year on year growth of 69% in volume and 40% in number of CS Tip #9: To poach or not to poach. loans above US$7,000. This type of growth and volume Many MFIs in fast growing countries in cannot typically be achieved solely through graduation of the region have aggressively increased microenterprise clients in such a short time. Conservative their VSE portfolios, serving their estimates from the literature suggest that 6-10% of microenterprises graduate to become small enterprises. existing growing microenterprise While MFIs don’t typically track this rate, even in high clients while recruiting new clients. In growth countries such as Peru, these are perceived by these countries where microenterprises are MFIs to be no higher than 50%. One Peruvian institution not growing as successfully, existing noted that 50% of its VSE borrowers are new clients to clients may be more limited and the institution, which were prospected through references seeking new VSE clients a necessity. by existing clients, poaching of loan staff from other This adds to the risk of an VSE strategy, institutions, and agreements with employers and other as many MFIs note that “known” clients organizations. In Bolivia, one institution echoed that offer lower portfolio risk. Poaching aggressively growing an VSE portfolio involves poaching, loan officers of competing institutions since it is difficult to recruit new clients and not enough to gain know-how can be useful, but existing clients graduate (See CS Tip #9). Nonetheless, obtaining new VSE clients through another MFI in Bolivia noted that their rapid growth in poaching can add to the risk of over the VSE segment was almost completely a result of client references and not poaching strategies. indebting VSEs by offering larger and faster loans to gain their business. Marketing to VSEs is most effective through word of Example: An MFI in Bolivia highlights that most of mouth and face-to-face prospecting. As such, offering its in-house expertise in VSE lending was obtained good service and convenient terms is critical to secure by hiring middle managers and credit offers from new business. Additional strategies used by some MFIs specialized financial institutions that serve VSEs. This to reach a broader business have included sponsoring has sped up the institution’s know-how significantly. sports teams or community activities as well as modifying brochures to reflect larger businesses. Most of the MFIs Example: An MFI in Central America notes that interviewed for this research did not dramatically attempt those VSE clients that are not served by banks are to alter their brand image to serve the VSE market, in the key “blue ocean” segment for MFIs. However, large part because their first VSE clients tended to be their they can be the riskiest to serve. Without transparent existing clients and they capitalized on their strong brand information and efficient credit bureaus, it can be with these clients. There is a tension, for MFIs that focus difficult to distinguish good VSEs from bad ones. on the microenterprise sector to avoid alienating their key microenterprise customer by changing their image too radically. At best, these changes are described as “tweaks”. In the case of some MFIs, small changes at the branch level such as shorter lines, waiting areas and more senior staff for VSE clients can attract the attention of existing clients on a path to growth. In other cases, the costs are not 30 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA justified. In Bolivia, the regulatory authorities do not allow preferential lines for specific groups, and this is not a possibility. Working with supply chains is relatively common for banks downscaling into SME, financial institutions that are more active in the SME and even larger corporate sectors can be best positioned to lend to value chains as a commercial strategy. MFIs with a large microenterprise portfolio that are moving into VSE are less likely to take a supply chain approach, particularly in urban markets where VSE loans are often concentrated. Unlike banks, they do not have access to larger companies that can link them to their suppliers, limiting their ability to go “down” the chain to link into new businesses. Nevertheless, supply chains can be used to identify larger businesses within a “micro” market, where MFIs go “up” from their traditional market, tapping into existing micro clients, or visit areas where there is a large concentration of micro clients and begin to scout the area for their suppliers. D. Other Considerations A commercial strategy for MFIs to serve VSEs should identify a key market segment, or multiple segments and define the MFI value proposition for each. Few institutions have implemented market studies of these segments, and even fewer have shared these so as to glean possible new opportunities to serve VSEs with financial products and services. At minimum, VSEs require some adaptation of traditional microenterprise loans including working capital and fixed investment loans with longer tenors, lower interest rates and larger loan sizes. All of these changes will imply adaptations to an MFIs business model including in Credit, Risk, Finance, Operations, IT and other areas as discussed in the sections below. These will also require adaptations to a commercial strategy. Credit and non-credit products must be available to VSEs to serve their needs, but also to increase the total profitability of relationships with these businesses. Aggressive prospecting may not be necessary, but can be depending on the market context. In highly competitive markets, MFIs will have to assess what their key competitive advantage is when serving VSEs: product, interest rate, speed of service, risk appetite, etc. While interest rates are important, in some markets, having a greater appetite for risk and the willingness to lend larger amounts against a given collateral item can often be highly effective in attracting VSEs. However, this in turn places additional pressures on the management of risk of these loans. Potentially, non-financial services are also needed, in particular training and support services that can help firms grow and reach a slightly larger and more stable segment of “small” entrepreneurship. Experiments to link these into existing credit offerings are extremely limited in the region, and a greater understanding of these linkages may help further expand the demand, growth and “stickiness” of VSE customers of MFIs. VII. CREDIT ANALYSIS 31 VII. Credit Analysis Credit analysis checklist • Reengineer credit analysis process to ensure the different risk profile of VSEs is addressed • Upgrade analysis of client’s “capacity to pay” to include historical monthly P&L, horizontal analysis, supplier references, cross-referencing and documentation of financial information • Potentially include projected monthly P&L as part of capacity to pay analysis • Revise loan guarantee policies, requirements and procedures for larger loans. Ensure guarantee policies balance the risk appetite of the MFI with the collateral guarantee capacities of the clients and the costs of documentation. • Include requirements of the regulatory authorities in all policies • Work with HR to incorporate new analysis and policies in HR training Regardless of whether an MFI chooses to adopt the Proactive or Organic Model when expanding upmarket to serve VSEs, it needs to revise and upgrade its credit analysis policies in light of VSEs different risk profile. For MFIs expanding upmarket mainly to retain their existing clients, it is easy for an MFI to fall into the trap of only making minimal adjustments to its credit analysis as clients already have a track record with the institution and are “low-hanging fruit.” However, there are significant challenges to making loans to new VSEs, and CA Tip #1: Don’t fall into the “trap” of depending on the market, they may be one of the riskiest thinking that knowing your existing segments of the market. Simply using “micro” technology clients is sufficient for analyzing to serve these informal VSEs can increase portfolio risk. their risk when offering Very Small Strong internal controls and auditing functions are also Enterprise loans. essential to contain loan officers and branches when there is strong pressure to offer larger loans and not lose clients. Example: Various MFIs we spoke to streamlined VSE loan processes for existing clients, where very In addition to differentiating between micro and VSE little financial and business data were collected for credit analysis and risk management policies, it may be existing clients, but more extensive data and analysis necessary for an MFI to do further sub-segmentation was required of new clients. While in the short-term, within VSEs (See section V). For example, in Peru, this does not appear to lead to delinquencies in where VSEs are defined by the superintendence as those existing clients, the model has not been “tested” in with loans of between $7,500-$100,000, there are huge LAC countries during a macroeconomic downturn, differences in small enterprise sophistication, formality where historical performance may not be sufficient and financial needs. Thus, blindly using the government’s to predict future performance. definitions could result in inappropriate credit analysis and increased risk. In one MFI in Peru, we found that VSEs have the highest PAR of any segment. The MFI has PAR30 of 8-9% for its Very Small portfolio vs. 4% for its Micro portfolio and 5% Small. This likely reflects the pressure to compete and disburse large amounts more quickly. 32 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Most MFIs in Latin America that have expanded upmarket to serve VSEs have adopted three basic credit analysis methodologies. It is important to note that these methodologies may not have been fully tested as the macroeconomic conditions in many of these markets have been quite favorable to micro enterprise and very small enterprise growth. However, these represent some of the more common approaches used by MFIs to date. Additional approaches exist, including credit scoring used by some banks or consumer lending institutions. In particular, anecdotal evidence from Peru suggested that both specialized MFIs as well as downscaled banks used scoring models to determine all of part of the willingness to pay analysis. However, limited evidence is available on the performance of VSEs using credit scoring. • A “Micro Plus” methodology with some added prudence and cross-referencing of financial information. • A “Specialized VSE” hybrid that adds to “Micro Plus” further cross-referencing, some additional analysis and collateral guarantees and an annualized cash flow analysis. It may include a SWOT and other external risk analysis of the business and sector. • A “Pro-Forma VSE” which examines historical cash slows as well as an assessment of future cash flows based on an investment plan, though loan amounts are not always linked to the results of this analysis. MFIs having adopted the Organic Model are more likely CA Tip #2: Outside support may be to use one of the first two credit methodologies, especially necessary when adopting a more in very competitive markets where clients are resistant to rigorous methodology (Specialized VSE/ delays in loan disbursement caused by a more thorough Pro-forma VSE) for analyzing VSE loans. analysis of a business. MFIs having adopted the Proactive These borrow from SME methodologies Model are more like to have adopted a “Specialized VSE” with some special techniques for hybrid methodology to serve VSEs, or even a “Pro-Forma estimating financial information that is VSE” model that borrows from the SME methodology of institutions such as ProCredit. Regardless of the model, not recorded. however, credit analysis needs to be adapted to take into Example: One Bolivian MFI hired loan officers with account the greater risk. Micro loan methodologies are VSE experience and worked with external consultants likely insufficient to evaluate the risk of larger loans (often to initiate a credit methodology that could analyze used for investment or leaps into a new stage of growth), VSE risk in greater depth than a simple “Micro Plus” where VSEs themselves may not be as well prepared to method. analyze their own investment and risk. However, the complexity of the credit analysis must be weighed against both the MFI’s cost of conducting it (in order to maintain profitability for the sector) as well as the impact of its agility and flexibility in serving its clients. VII. CREDIT ANALYSIS 33 A. Capacity to Pay Upgrades There are five main ways the evaluation of a SE’s capacity to pay may vary from the evaluation of a micro client’s capacity to pay. These differences can be viewed as a spectrum: Organic Models will have made the first one or two adjustments to their capacity to pay analysis and potentially the third, while Proactive Models are likely to have made the first four adjustments and maybe the last to their analysis. Better validation of financial information: For VSE lending MFIs tend to require additional validation of the financial information reported by clients and crosschecking the information from multiple Sample “Validation” Check or “Cruce” directions. For example, revenues might be checked against sales receipts, register books or work orders. Margins A cross-check seeks to verify either the sales or the might be asked for directly and checked against financial margins…the analyst will look at the inventory, ask statements (if available), but also calculated by looking at about sales, ask about margins, ask about purchases raw sales and purchases. Although similar checks are done and then also calculate the margin. for micro loans, they tend to be simpler and more trust is placed in the hands of the micro client to accurately estimate sales and profit figures. Greater documentation: Both the documentation required from clients as well as the documentation loan officers are required to complete tends to be greater for VSE loans than micro loans. For VSE loans, MFIs tend to require greater proof of financial information through sales receipts, tax payments, etc., and this proof must be documented in the loan files. It may also include more documentation from references (providing letters of reference rather than just an oral validation) from suppliers and other creditors (landlords, etc.). Additionally, greater documentation of collateral may be required. For example, in Bolivia, some MFIs do not require land or property titles for smaller loans, but will require these for VSE loans. Loan officers might visit these homes and take photos of them to include in loan files. In one MFI, an architect accompanies the loan officer on the visit to assess its value independently. Loan files are longer and more detailed and tend to take three to four times longer for a loan officer to complete. Although, it is prudent to have more stringent documentation requirements for larger loans, these requirements must be tailored to what an VSE can reasonably be expected to have, not so costly to obtain as to make the loan unviable, and not too burdensome on loan officers, or an MFI risks losing one of its key competitive advantages over the banks to serving this market – its agility and flexibility. Audited financial statements, for example, are rarely available and cost-prohibitive for some VSEs. Formalized evaluation of external risks: Inherent in most microfinance lending is an “understanding” of the external risks a client’s business faces. Loan officers tend to know their markets and customers well and because most businesses are simple, can quickly get a sense of how a business is positioned in the market and what external pressures it might face. However, there is rarely a formal analysis or documentation of these risks and this “understanding” is not institutionalized. Because of the size of the loans and the relative flexibility of microenterprises, should a negative external event occur, microentrepreneurs are generally able to continue to meet their loan 34 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA obligations, often through other household income. The CA Tip #3: A“horizontal analysis” of larger the business and the loan, the less likely that this financial data can help MFIs see if the will be possible and the more complex the risks faced trends in the business are positive, by the VSE. Thus, for VSE lending it is recommended negative or static and if its growth that a more formal analysis of external risks is conducted warrants a larger loan. and that these risks be documented as part of the loan file. This analysis can include: a description of suppliers, Example: One Bolivian MFI looks at historic financial a description of clients, a SWOT or additional market data for existing clients in order to see if the business analysis. Developing in-house expertise or working in has grown and if they can prudently increase the conjunction with local universities and think tanks can loan size. be useful if exposure to certain economic segments is significant. Additionally, implementing ex-post analysis of sectorial risk concentration and imposing limits on this can help to cap this exposure (See Risk tip #4). More in-depth cash flow analysis: Most MFIs only look at a static “snapshot” of cash flow when making micro loans, whereas a more nuanced moving “video” of cash flow works better for VSE lending. Usually this is a 12-month cash flow for VSE loans that will capture the seasonality of the business and checks that the business will have the funds needed to make repayments during this cycle. Given the informality of the businesses, the loan officer must generally work with the client to create this cash flow. Thus the cash flow serves as a tool for evaluating capacity to pay, but also a medium through which the loan officer gets to know the client and business and an initial sense of willingness to pay. (See Figure 3 below). Figure 3. Comparison of VSE cash flow analysis of various credit models Micro Plus Specialized SE Pro-Forma SE • Weekly or monthly • Calculate weekly or monthly P&L • Calculate annual/seasonal P&L (See P&L (See Micro Plus) Specialized SE) • Business Sales • Ask if this month’s sales are average, • Calculate the business margin (Sales- Cost • Costs of goods sold poor or better than usual of Goods Sold) • Family income/ • Assign a sales revenue figure for each • Estimate additional purchases (inventory) expenses (informal type of month (i.e. “what does a bad from the use of the upcoming loan businesses) month look like in terms of sales? What • Estimate the growth in sales (using margin • Less loan payments does a good month look like?) calculations) that this will produce and other obligations • Review each month of the year to • Based on the increase in sales figures, determine whether it is an average, construct a projected annual P&L [where • Horizontal analysis good or bad month.(i.e. Is May usually revenue = margin x new inventory]. of prior loans and a good month? What about June,?) • Subtract the cost of operations growth in cash flow • Calculate excess cash profit monthly • Calculate the remainder to ensure that it from earlier stages and annually to ensure that it is is sufficient to cover loan payment (for existing clients) sufficient to cover loan payments VII. CREDIT ANALYSIS 35 Pro forma cash flow analysis: Some MFIs may also take into consideration the pro forma impact of the loan on a business’s cash flow and capacity to pay. As many VSEs may be using the loan for investment purposes, this is very reasonable, especially as many fast growing VSEs might not qualify for the funds needed for investment based strictly on historical performance. Most of the Pro Credit Banks conduct this type of analysis, as do some traditional banks, but this is not standard practice for most MFIs moving upmarket in Latin America. Many MFIs consider it too risky to lend based on growth assumptions, especially given the degree of specialized analytical skills needed to do so. For example, how much a loan increases the bottom line varies if it is used to upgrade an existing facility, purchase a new building, move to a better location or increase inventory. When selecting an approach to measuring capacity to pay, MFIs should consider the rigor they need and balance this with their internal capacity to implement a methodology, incorporate it into training and control processes and balance it with ex-post risk analysis. In addition, the cost of the analysis must be taken into consideration and justified for the size of the loan and the margins they offer. Figure 3 offers a summary of key approaches to assessing capacity to pay. B. Willingness to Pay Adjustments In microfinance, willingness to pay has traditionally been judged based on informal character analysis and references and “ensured” through a variety of mechanisms including: group guarantees, individual guarantors or the pledging of small assets as collateral. Clients are generally offered larger loans as they build a credit history with the institution. For VSE clients that are graduating from micro clients, willingness to pay may be less of an issue because the client already has a strong track record with the MFI. Willingness to pay is otherwise measured with some relatively traditional bank- like tools, adjusted to the realities of informal sector enterprises. For example, credit bureaus are consulted where they are available, but where they are not, or if they only publish information about delinquent loans rather than overall indebtedness, their utility is reduced. Character references are another common requirement. Measuring character by asking regular business colleagues such as suppliers to vouch for a client can be an important validation of their responsibility and good judgment. Often formal letters are asked of suppliers or larger customers of a business. Finally, while collateral tends to play a small role in microcredit, it typically plays a larger role in VSE loans. Collateral is mostly taken in the form of property or land titles (and to a lesser extent vehicle or machinery) but there is varying flexibility in terms of the required documentation and assessment of property. Credit bureaus The MFIs interviewed for this research have generally placed a large emphasis on credit history for determining a VSE’s willingness to pay. However, most are located in countries with strong credit bureaus and information. In both Peru and Bolivia, credit bureaus capture information not just from banks, but also regulated microfinance 36 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA institutions (as well as NGOs that volunteer information). CA Tip #4: Credit bureaus have helped take This makes it very easy to check not just repayment some of the “guesswork” out of evaluating history, but also overall indebtedness. We were informed willingness to pay, but may encourage that one institution in Peru, uses a credit bureau (Equifax) lending larger amounts to stay ahead of for an initial screening and also to determine loan size. While credit bureau information is useful and should be the competition and meet disbursement one factor in determining willingness to pay, care must be targets, leading to over indebtedness and taken not to rely too heavily on it. problems with capacity to pay. In countries without strong credit bureaus or where MFIs Example: In Peru, it is not uncommon for VSE clients do not have access to the information in the credit bureau, to have 3 or 4 loans with different institutions. One MFIs face additional risks when lending to VSEs as MFI discussed with us a critical risk noting that when compared to microenterprises, as clients are more likely to they decide to reduce the requested amount of a loan have had prior loans with other institutions. For example for a client because their capacity to pay analysis in Costa Rica, many of the VSEs target by a leading MFI cannot justify it, they face a greater risk that other are large enough that they may have had access to bank MFIs will make these additional loans, placing the loans in the past, but because the MFI is an NGO they client’s capacity to pay even more at risk. Analysis can’t check the repayment records for these clients. This of credit bureau data for indebtedness of clients increases the risk that they will simply get the bank’s worst suggests this may be taking place. performing clients. Guarantees and Collateral In addition to assessing a VSE’s willingness to pay, MFIs can “ensure” willingness to pay through collateral guarantees and guarantors. While it is prudent for MFIs expanding upmarket to consider requiring guarantees for VSE loans, when determining their guarantee policy, they must balance their tolerance for risk with the guarantee capacities of their clients and the costs of documenting collateral guarantees. One of the key competitive advantages that MFIs have in expanding upmarket is their traditional flexibility in terms of guarantees. If they are too strict with guarantees for VSE lending, they risk becoming too bank-like and losing their competitive advantage. Often it is not that the VSE doesn’t have assets, but rather that the documentation on these assets is incomplete and insufficient for many banks or that the cost and effort of obtaining the necessary documentation/registration of collateral does not justify the loan size. Additionally, assets may be difficult to value as they might be homes in neighborhoods where property values are low, or in poorly constructed buildings, for example. Most of the MFIs interviewed for this research do not take real guarantees for micro loans. Most only required a guarantor or a pledge of machinery, appliance or other small assets, and for many micro clients, the MFIs require no guarantee at all. For VSEs there is greater divergence in guarantee policies. Most of the MFIs interviewed require “real” guarantees for larger loans. Acceptable “real” guarantees tend to include property, land, vehicles, deposits and some machinery. However, what is considered a larger loan varies. At one Peruvian MFI the standard practice for a normal risk clients is to require real guarantees for loans over $11,000, but a low risk, preferred client can borrow up to $44,000 without a guarantee. At another VII. CREDIT ANALYSIS 37 Peruvian MFI, clients can borrow up to $35,000 without a guarantee. Meanwhile, in Bolivia the regulator requires Credit Scoring real collateral for loans of over $15,000. The later may help control systemic risk, but also limits many businesses’ MFIs can also utilize credit scoring to help in the credit capacity to borrow, especially those growing quickly and analysis process. None of the MFIs we interviewed in search of funding for investments. for this research used credit scoring, however, it is used by some institutions in Latin America working When MFIs do decide to take real collateral on a loan, it with VSEs. It is best suited for those institutions is important that this collateral is valued correctly and is already familiar with credit scoring and/or those sufficient to meet the institution’s collateral requirements. with rich data bases from which to develop a credit Unfortunately many loan officers do not have the skills scoring model. It also requires external or internal or training required to correctly estimate the value of expertise to develop such a model. In addition, the collateral, such as a house. This puts both the institution MFI should have a sophisticated MIS and a team able and the client at risk should there be a problem with to track and analysis the efficacy of the model. An ill repayment as the MFI’s exposure may not be covered by designed model can quickly increase an MFIs credit the real value of the collateral and the client may be left risk exposure. Even a well designed model must still with nothing, or worse yet, still indebted, even after the be fine-tuned on a regular basis (every 6 months in a sale of the house. One MFI in Bolivia has dealt with this challenging credit environment). problem by having a third party, usually an architect; provide an estimate of the real value of the collateral. Another MFI reduces the stated value by 20% to ensure some caution. 38 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA VIII. CREDIT RISK MANAGEMENT 39 VIII. Credit Risk Management Credit risk management checklist • Revise loan approval authority based on segments and/or loan size, balancing branch autonomy and agility with risk management • Revise recovery procedures for VSE loans taking into consideration potentially new guarantees and collateral • Set limits on sector concentrations at institutional and branch level • Set up a proactive portfolio monitoring system, tracking micro and VSE loan portfolios separately at an institutional, branch and loan office level • Strengthen internal controls to prevent fraud and ensure compliance with new policies and procedures • Establish periodic reviews of collateral valuations to ensure risk coverage metrics are met One prerequisite for an MFI to successfully expand upmarket is that their existing credit risk management policies and procedures are already strong. An institution that is struggling with credit risk is in no position to take on the risks of a new segment before it gets the rest of its house in order. Decentralization in microfinance is often considered essential because it allows an MFI to control costs and quickly disburse loans to clients, a quality valued by microentrepreneurs. However, it places a lot of authority in the hands of the branches. With clear standardized credit analysis and risk management policies and procedures, as well as good internal controls, risks arising from this concentration of power at the branch level can be mitigated when making larger VSE loans. Overall in Latin America, successful expansion upmarket to serve VSEs appears to involve some level of centralization (or perhaps re-centralization) of risk management, be it at the loan approval stage or the monitoring and tracking stage. For VSE loans, if MFIs seek to maintain this degree of decentralization it should be offset with greater rigor in loan analysis and requirements. A. Loan Approval and Oversight Speed of service is key for microloans and loan approval is generally decentralized. For very small micro loans, senior loan officers or credit committees comprised of several loan officers may have approval authority. For larger micro loans, loan approval often requires branch Risk Tip #1: Loan approval for VSE loans manager signoff. However, for VSEs, full autonomy for should not be delegated completely to loan approval should generally not rest at the branch the branches. level, despite potential reductions in the agility of loan disbursement. If they don’t already exist, an MFI should Example: At a Peruvian MFI we visited, loans below establish different levels of loan approval based on loan $25,000 can be approved by the branch manager, size and segment served. For sectors that are considered but loans above $25,000 must be approved by a more risky, regional or national loan approval may be regional manager, and loans above $50,000 must be required at a lower loan size. approved at a national level 40 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA It may be possible to maintain greater authority at the branch level for VSE loans if risks are mitigated in other ways, such as through more in-depth credit analysis or the involvement of more branch staff. For example, for larger loans, one Peruvian MFI requires loan officers to present loans to a credit committee that includes branch staff whose incentives are not linked to approval amounts. As we discuss in the Human Resources section below, in the Proactive Model, the sales and credit analysis functions are sometimes separated in order to better control risk and allow for greater staff specialization. However, this is atypical for most MFI’s micro loans and also uncommon for MFIs using Risk Tip #2: Credit risk can be reduced a more Organic Model. Nonetheless, an effective risk by involving more branch staff in the management strategy for VSE loans, even under the Organic Model, is to require additional oversight prior loan approval process. to loan approval. For example, at another Peruvian MFI, Example: A Peruvian MFI we visited for this study larger loans require an additional visit to the client by a requires loan officers to present larger loans to a non-analyst or a regional supervisor. On the one hand subgroup of branch staff including non loan officers. this can help avoid fraud, and on the other it’s a second This presentation and discussion can identify new opinion on a client’s capacity and willingness to repay a business risks for the VSE, and allows for input from loan from someone whose incentives are not linked to staff whose incentives are not linked to disbursement loan approval. In certain economic sectors, it may also be targets.. useful for MFIs to consider hiring industry specialists to contribute to the credit analysis. B. Monitoring and tracking structures and policies VSE lending requires proactive monitoring, even in Organic Models. Proactive monitoring can help identify risky loans or sectors as soon as or even before they become problematic and help the MFI to avoid similar risks going forward. Centralized risk monitoring and tracking appears to be more effective in reducing portfolio at risk, but requires a sophisticated and agile MIS. It is absolutely essential that an MFI’s MIS allow for the micro and VSE loans to be tracked separately, ideally at the institutional, branch and loan officer level. This allows the risk department to identify not only problematic segments, but also the source of these segments, and to adjust policies accordingly. Setting portfolio concentrations limits also helps in risk Risk Tip #3: Centralized, proactive mitigation. Concentration limits based on loan size can monitoring of VSE loans allows risks to help MFIs limit VSE risk, especially when first entering the sector. Regulators may also mandate these limits for be identified and mitigated early. MFIs, such as in Bolivia. Sectorial concentration limits Example: One MFI in Peru has automated alerts and are also useful when lending to VSEs, especially if MFIs controls and also does in-depth portfolio analysis are using new promotion strategies, such as value-chains, daily. It then adjusts a branch’s autonomy if its that might result in sectorial concentrations. More overall portfolio at risk or a sector’s portfolio at risk defined limits to subsector concentration (i.e. restaurants, is too high. transport, clothing manufacturing, grocery, etc.) are more effective than broad limits such as services, commerce, VIII. CREDIT RISK MANAGEMENT 41 agriculture. Applying limits at the institutional and the branch level can further reduce risk, but limits may need to be adjusted in certain markets. For example, an MFI with an institutional limit of 10% for livestock loans, may need to have a branch limit of 25% in rural regions. Why are delinquency rates so much lower in Bolivia than in Peru? • Regulatory focus • Regulators concerned with over indebtedness and repayment mandate processes, collateral requirements and authorization (soon also interest rates) • Lending methodology: Capacity to Pay Analysis • Monthly/seasonal cash flow projections • Loan officers “build” a monthly P&L based on seasonal estimates • Greater validation of P&L and balance sheets through invoices, receipts, tax documentation • In Peru, documentation showing the intended use of loan is often required • Lending methodology: Willingness to Pay/Collateral • Bolivia: Regulator mandates real guarantees (property) for loans over US$15,000 • Peru: MFIs usually ask for real guarantees for loans above US$30,000 • In Bolivia, more emphasis is given on the analysis of collateral to ensure it is fairly valued • Branch Autonomy • Peru offers more branch autonomy for larger loans for faster loan disbursements • Bolivia centralizes authorization of larger loans at the expense of slower disbursements • Ex: Bolivia up to $7,000 in Peru up to $20-30,000 C. Recovery policies There are three major adaptions to recovery policies that MFIs must consider when expanding upmarket to serve VSEs. This first is who is responsible for initial recovery efforts. Generally this responsibility falls on loan officers for micro loans, however, for VSE loans, which represent a greater risk for the MFI should they not be recovered, a team approach or support from the branch manager or recovery specialist may be required. The second is the timing of recovery efforts. If a micro client falls 1 or 2 months behind in paying their loan, it is hard for them to regularize the loan. However, business cycles for VSEs are more volatile and they may miss a payment, which can negatively impact an MFI’s cash flow, but are more likely to be able to regularize their payments. The third major difference in recovery policies is that for VSE loans that have real collateral, the case can be addressed to the legal system for recovery and because the loan amounts are greater the cost of servicing them legally is more worthwhile. However, in many countries in Latin America, this can be a long process and requires the support of lawyers, either internal or external. 42 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Risk Tip #4: Sectorial risk matters, especially in VSE loans. MFIs can either have sectorial expertise and analyze risk ex-ante, limit sectorial exposure and analyze risk ex-post, or both. Example: One MFI noted that in 2009, during a financial crisis, businesses in the tourism sector suffered, affecting the institution’s portfolio. At the time, loans were relatively small and the risk was less concentrated but the event alerted the MFI to the risk of sectorial concentration. Rather than having sectorial expertise and analysis, however, the MFI has chosen to limit exposure to any one sector to 10% of its portfolio. Example: Another MFI in Nicaragua developed an in-house VSE lending methodology with few sectorial limits. Their portfolio was also highly concentrated in cattle raising and when the financial crisis hit in 2008, the poor performance in this segment of its business was one of the main contributors to its downfall. D. Regulations Regulators can impact VSE lending in a variety of ways. Regulations designed to limit systemic risk help MFIs to mitigate risks, but can also impact a MFIs ability to serve VSEs. Bolivia’s regulator, for example, has several restrictions to VSE lending, such as limiting loan size to twice equity, requiring that VSEs pay taxes in order to qualify for higher loan amounts, and limiting SME lending to 30% for MFIs. The later forces MFIs that want to make significant moves into the VSE market to also consider transformation, while the tax payment requirement may encourage smaller VSEs to seek out multiples lenders. Asset backed financing as a way to mitigate risk: can mfis do it? One non-traditional way for MFIs to increase VSEs access to finance while mitigating risks is to develop asset backed financing products. For example, MFIs can offer leasing services, where the MFI maintains ownership of the equipment being leased for the duration of the “loan.” Thus far there have been very few instances of MFIs offering these types of services in Latin America, in large part because the methodologies are so different from the traditional microfinance model and require more advance legal operations to implement and in part because VSEs often purchase equipment second hand. Some additional considerations regarding VSE asset-backed lending: • Using accounts receivable may not work for very many VSEs as they tend to be more informal and their accounting less reliable. • Warehousing might work for some types of VSEs. However, it requires a sophisticated MIS system that many MFIs may not have and can be costly to implement if there is not significant scale. • Factoring may work for VSEs that work with a few larger clients, including municipalities and public agencies. However, this is not the case for most VSEs in Latin America that tend to work with retail or individual clients. It also requires access to reliable credit information on an SE’s clients. • Purchase order finance may be the easiest asset backed product for MFIs to develop, however, it requires more oversight than some of the other options as the product is not yet produced. It has already been tried by some MFIs, but has not reached scale. Source: USAID 2009. IX. HUMAN RESOURCES 43 IX. Human Resources Human resources checklist • Determine whether the target market (per branch) justifies exclusive VSE loan officers or can be managed by loan officers who cover various segments • Decide whether loan officers should be in charge of sales, loan analysis and relationship management or if some specialization of functions is needed • Adjust compensation and incentives structures to motivate VSE staff and avoid potential conflicts • Ensure loan officers have adequate experience and finanical analysis capacity for VSE sector • Assess internal capacity and training curriculum to identify gaps in training for VSE loan analysis • Ensure that training (classroom and field) offers sufficient opportunities to work on cases and “practice” lending • Assess whether training curriculum and delivery gaps can be filled internally or require external resources • Incentivize tutoring and mentoring by senior staff to new staff • Incorporate refresher trainings into core training process • Update training through lower-cost technologies A. Sales and loan analysis staff Having adequately qualified and well trained field staff is key to the success of any MFI’s expansion upmarket. Who staff is made up of depends on the model and methodology adopted by the MFI, but there are two key decisions, which need to be made: • Should the MFI have specialized VSE loan officers or can loan officers serve both micro enterprises and VSEs concurrently? • Should loan officers be in charge of sales, analysis and relationship management or should there be additional specialization of functions for VSEs? Specialized VSE Loan Officers – the Proactive Model In the Proactive Model, there are specialized loan officers to serve VSEs. In addition to serving a different clientele, these loan officers are likely to have different targets, compensation, and training, and potentially different education and work experience than micro loan officers. These loan officers may report to the branch manager or directly to a head quarter’s based VSE manager. Under this model, VSEs that approach the MFI are directed to a VSE loan officer for service. Likewise, a micro enterprise client that grows into a small enterprise client is transferred from their micro loan officer to a new VSE loan officer. There are several advantages to this model. • Loan officers can specialize in high growth VSE sectors and perhaps identify opportunities for new clients in the market. 44 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA • Loan officers can evaluate a business in an in-depth, holistic manner analyzing the market sector, the business’ investment plans and the details of business performance with greater care. • Risk management is likely to be stricter, having distinct criteria and greater centralization of approvals for loans made by VSE loan officers. • “Micro” loan officers can focus on smaller loans and meeting growth targets without the distraction of volume targets. Nonetheless, there are also several potential disadvantages to this model: • The model is costly, relying on more highly paid, senior loan officers to specialize in VSE loans, even where volume may not justify a dedicated resource. • VSE loan officers can lose enthusiasm and leave if an MFI cannot retain clients with their existing services. The key benefit of this model is that it allows loan officers HR Tip #1: Having dedicated VSE or SME to be selected specifically based on their ability to work staff makes sense only when branch with VSEs, and to hone their promotion, credit analysis volume merits it. and relationship manager skills to the sector. This Example: One MFI in Bolivia has three levels of specialization should allow an MFI to both grow its VSE loan officers (micro, micro/small and SME), yet only portfolio quickly and maintain a higher quality portfolio. assigns a SME loan officer to a branch if they have The key disadvantage to this model, however, is that it can over $2 MM in portfolio. be more costly to start-up and operate as there are now two (or more) loan officers covering each geographic market. In markets with a relatively high concentration of VSEs or when the characteristics and needs of VSEs are quite distinct from micro enterprises, the benefits of the model are more likely to outweigh its costs. The figure of a specialized loan officer can also lead to some HR Tip #2: Team incentives at the internal conflicts between micro and VSE loan officers at branch level can reduce moral hazard a branch as micro loan officers may resent “giving their and tension between micro and VSE best clients away” to other loan officers simply because loan officers over client “ownership.” the client graduated. Likewise, it can lead to inefficiencies and potential loss of clients if not properly implemented. Example: Senior loan officers from one MFI in Peru For example, when an MFI in Peru hired specialized are the primary “VSE” lenders at the branch and loan officers in the past, they actually saw a reduction have incentives that help meet their goals. However, in their portfolio, because micro loan officers were not the team shares branch incentives and colleagues are passing on VSE leads to the VSE loan officers. One way asked to occasionally visit each other’s applicants or to mitigate such conflict is to ensure that strong branch jointly visit delinquent clients to foster teamwork and level performance benefits all loan officers; to compensate ensure that protocols are being enforced. microloan officers for lost clients and to incentivize them to share leads with their VSE colleagues. IX. HUMAN RESOURCES 45 Multipurpose Loan Officers – the Organic Model It is relatively uncommon to see MFIs operating the Organic Model that have specialized VSE loan officers. This is in large part due to the fact that one of the main reasons they expanded upmarket was to retain their best clients, an incentive for MFIs and loan officers alike. Thus, in most purely Organic Models, loan officers work with both micro and VSE clients. There are several advantages to this model. • Loan officers can keep their clients as they grow, strengthening the relationship that has already been formed. • There can be lesser potential conflict between micro and VSE loan officers including when referring leads for potential new clients • Businesses with greater growth potential can obtain larger loan sizes without hesitation from loan officers who may fear losing their clients when they outgrow them. • Loan officers know their clients and can evaluate businesses with greater speed of service. HR Tip #3: Allowing only “senior” • The model is less costly, especially in markets with loan officers with a certain amount relatively few VSEs or where clients are widely dispersed. of experience to serve VSEs can help Nonetheless, there are also several potential disadvantages mitigate risk. to this model: Example: According to management of one leading • Limits the focus a loan officer can place on recruiting Peruvian MFI, some degree of specialization is VSEs. required to serve VSEs, but this specialization can • Sales and promotion strategies of existing loan officers be acquired through experience and at most of may be lacking or inappropriate for SEs. their branches, the senior loan officers have this • Credit analysis skills of existing loan officers may be experience. insufficient to serve VSEs. • Lack of specialization may limit an analyst’s ability to perceive risks in new loans or additional client needs. • Getting the compensation and incentive structures HR Tip #4: Conflict between junior and right can be complicated. senior loan officers can be mitigated MFIs can compensate for the potentially weaknesses of by clearing defining the career path not having specialized VSE staff, by only allowing more for junior loan officers and by building senior loan officers that have already proved themselves mentoring into the targets and incen- to serve VSEs. However, care must be taken to ensure tives of senior loan officers. this does not lead to poor “micro” service as senior loan officers have to spend most of their time building more One Bolivian MFI offers loan officers financial rigorous VSE loan files. Moreover, just as with specialized incentives, certification, and promotion opportunities VSE loan officers, prohibiting junior loan officers from for senior loan officers that contribute to training serving VSEs may create conflicts between junior and of more junior staff. This ensured that the more senior analysts. complex skills required of VSE loans are transferred down, not only in terms of technical requirements but qualitative analysis. 46 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA Specialization of sales and credit functions: Is relationship banking needed? Given their larger size, in some markets MFIs may find that VSEs require more sophisticated sales and relationship management than their micro clients. This is more likely to be the case when MFIs have made a “leap-up” in their expansion. In such markets it may make sense for MFIs to separate the sales/relationship management functions from the credit analysis functions. This separation makes it easier for an MFI to find staff with the higher level of sales and relationship management skills needed, without compromising the financial and credit analysis skills that may be necessary. Most Pro Credits, which epitomizes the Proactive Model, have separated their sales and credit analysis functions for their small enterprise clients. The relationship manager/ loan officer is in charge of selling the product and making sure the client meets basic criteria. Then a back-office credit officer collects the financial information and does the credit analysis. Together the two draft the credit memo. An MFI using an Organic Model is unlikely to separate the sales and credit analysis functions for its VSE loans unless these functions are already separated for its micro clients. For example, this may be the case for MFIs with centralized credit analysis functions. B. Compensation and incentive adaptations In both the Proactive and the Organic Models, it is HR Tip #5: VSE incentive schemes essential to adapt loan officer incentives to take VSE should focus on portfolio quality. Some loans into consideration. Loan officers working with schemes have eliminated variable VSE clients will quickly see a jump in their portfolios, incentives for loan officers alltogether, but slower growth and potentially even a reduction in while others incentivized more on vol- the number of clients served. Additionally, VSE loan ume rather than number of clients. officers concentrate much more institutional risk in their portfolios, and delinquency may have to play a stronger Example: One Costa Rican MFI that follows a Pro role in their incentive schemes. Thus the MFI must adjust Credit model, compensates VSE sales executives on the incentive systems to compensate for this as well as to a commission basis, while credit analysts, who are accommodate their goals for the sector. responsible for analyzing loans are paid on a fixed salary. For MFIs with specialized VSE loan officers, developing a compensation and incentive system can be more Example: At One MFI in Colombia, VSE variable straightforward as the incentives for micro loan officers pay is 70% related to portfolio quality and 30% and VSE officers are largely independent. Typically a larger dependent on portfolio size. portion of a VSE loan officer’s compensation is fixed as compared to micro loan officers. The variable portion of her/his salary is weighted more towards size of the portfolio, while for micro loan officers, number of clients has greater weight. There may be exceptions to this under certain market conditions. For example, in order to break into a market, an MFI may choose to put greater emphasize on the number of VSE clients rather than portfolio size, especially if the hope is to identify high potential micro or VSE clients and grow with these clients. IX. HUMAN RESOURCES 47 For MFIs that do not have specialized VSE loan officers, getting incentives right can be more complicated. Variable HR Tip #6: Linking non-specialized loan pay weighted too much towards portfolio size can cause officer variable pay to specific VSE and loan officers to place too much focus on VSEs at the micro targets can ensure neither sector expense of their micro clients. Likewise, pay weighted too is neglected. much towards number of clients will not motivate loan officers to seek out new VSE loans, which are generally Example: At one MFI in Peru, loan officers have more time-consuming to analyze and process. Additionally specific targets for both micro and small enterprises low incentives for delinquency can place the portfolio at and their incentives are linked to meeting both of risk, especially when VSE loans are prioritized. these targets. Targets vary on the seniority of the loan officer as well as the branch/market served. Compensation structures for MFIs that separate the sales Example of targets: and credit analysis functions must also appropriately incentivize both staff (See HR Tip #5). One approach is • Senior loan officer: 35 micro loans and 5-6 small to have relationship managers that make a commission if loans a month loans are approved, while credit officers are paid on a fixed • Junior loan officer: 20 micro loans and 1-2 small basis. If credit officers have variable pay linked to portfolio loans a month quality, it may create too strong of an incentive not to take enough risk. Following the financial crisis, most Pro Credit banks moved to a model of fixed rather than variable compensation, to counter some of these pressures. While this was well received in many of these institutions, few others have followed suit, fearing that fixed salaries might reduce staff incentives to grow portfolios. C. Specialized Capacity and Qualifications Internal vs. External Hires Ensuring staff is adequately qualified and trained to work with VSEs is essential to the successful expansion upmarket. In the Organic Model, as staff are generally not specialized in VSEs, the emphasis is on training rather than staff selection (see Training section below). However, even in the experiences that we studied for this document of more Proactive Models, the preference seems to be to first look internally for staff able to serve VSEs and then to expand the search externally if necessary. Convincing the best loan officers that they should shift from micro to VSE clients can be complicated as it requires Quotes from micro loan officers at one more work for the loan officer and can also be more risky. MFI in Peru: For example, at a leading Bolivian MFI, SME loan officers have monthly goals to disburse $100,000 and it takes two “The problem is that if I am a VSE loan officer weeks to a month to disburse a SME loan, while micro and I don’t find VSE clients, I don’t get paid.” loan officers have a goal of only $30,000 and disburse loans in 1-4 days. It may require initial pay guarantees “The VSE clients sometimes [leave us and] and assurance that in the long-term a VSE loan officer can go to the banks and your portfolio can fall earn more than a micro loan officer and/or that it will lead dramatically with the loss of just one client.” to greater chance of promotion. 48 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA An VSE loan Officer Profile Microfinance institutions that offer individual loans typically find that the ideal profile for a VSE loan officer is someone with a college degree, ideally in finance, economics, or engineering, but also some experience in micro or HR Tip #7: Experience trumps education VSE lending. This experience is more important for VSE for loan officers lending to small loan officers than for micro loan officers, who may be green and learn on the job. Experience helps ascertain enterprise. qualitative characteristics of a business owner that are • Example: One MFI in Colombia created a new better perceived with experience (trust-worthiness, values, profile for VSE loan officers and recruited them persistence) and that are critical to repayment capacity. internally. Now, every 3 months they offer an One MFI branch manager notes: the “sense of smell of internal course of VSE lending in each of the big a loan officer is important. If there is a doubt [about the cities in Colombia and based on how participants willingness to pay of the client], we reject the loan.” Even do, they select new VSE loan officers. for non-specialized VSE loan officers, it is a good idea for an MFI to require a minimum amount of experience. A leading Peruvian MFI requires a loan officer to have at least 1 year of experience with the institution before they can serve VSEs. If there is no staff with sufficient experience in-house, MFIs can also consider poaching this experience from other institutions (see section VI). We interviewed one MFI in Peru that sought to grow its VSE portfolio quickly and could not wait for loan officers to “graduate;” thus, it hired VSE loan officers with not only experience in VSE but also with an existing portfolio that it could move to the institution. D. Training and Support A strategy of internal promotion might lead to a loan officer staff with a good “sense of smell” but a weak preparation in formal analysis of larger business. This is one of the greatest weaknesses of the Organic models and one that must be actively addressed when entering the VSE market. Even when the staff working with VSEs is specialized, without the appropriate training staff can reach a limit in in terms of understanding business size/investment complexity. Staff shifting from micro to VSEs or expanding to VSEs typically need additional training in: • How to conduct additional more in-depth credit analysis: creating simple balance sheets and P&L statements; cross referencing strategies; stress-testing existing balance sheets and income statements; detailing and projecting monthly cash flow analysis; ascertaining the business’ SWOT, including an understanding of market trends and opportunities. • Processes: Ensuring compliance with processes and protocols for approvals of larger loans, including interfacing with IT systems. • Guarantees: what are the guarantee policies for VSE loans; how to appropriately value the guarantees; paperwork/legal requirements of recording guarantees • Prospecting: new prospecting techniques for VSEs; sales skills; relationship management techniques. IX. HUMAN RESOURCES 49 Training generally includes both classroom training and on-the-job training. In the case of VSEs, there may HR Tip #8: Testing staff in the field in be a greater need for classroom training in order to advance of classroom training can weed fully prepare loan officers for more complicated credit analysis. One MFI visited for this study only offered out those prospective candidates that three classroom days of training and only three hours of are not suited for field work prior to technical training on financial math and analysis with investing costly classroom time in them. no case study work and found that this was insufficient • Example: An MFI in Peru “tests” new hires in for staff. Another MFI struggled with the content of the field for three - four weeks prior to starting financial and case study training and noted that external classroom training to save on training costs. consultants with experience were hired to develop this material in coordination with the credit area of the MFI. In-depth classroom training with a case study approach offers a guarantee of quality and care of small business loan analysis. This classroom training is reinforced with on- the-job training and mentorship. Example of a Training Process for a hybrid (Proactive/Organic MFI Model) One MFI offers three full months of training for new loan officers, where all prospects receive three months of training including: • One month of classroom training (including institutional mission, vision and values; credit methodology and analysis), • One month of combined classroom and field training, • A final month of primarily field training. More experienced loan officers spend more time in the field earlier on but are still required to participate in the first month’s classroom training. While the training process is standardized for micro and VSE loan officers, VSE loan officers are trained in more in-depth analysis with case studies including mechanisms for cross-referencing financial information, which is critical to the formation of a balance sheet and cash flow statement in informal businesses. Many MFIs are resistant to providing extensive classroom training, as it can be very costly. During HR Tip #9: On-the-job training can be the initial expansion to VSEs, training costs can be more valuable than a classroom, but mitigated by utilizing online and video training and/or MFIs cannot assume that junior staff by decentralizing trainings through an initial training of will learn these lessons quickly and trainers. Once an MFI has started offering loans to VSEs, well unless senior staff is appropriately it may prove less costly to pay VSE/senior loan officers to incentivized to teach junior staff . train new officers and also help weed out weak candidates early on. A personnel policy that fosters low turnover and Example: One MFI pays its trainers, who are also high employee satisfaction is also key to avoid losing an loan officers, an additional fee for serving as trainers MFI’s investment in training. and also offers them one Saturday off a month and points towards future promotions. 50 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA HR Tip #10: Costly training must be balanced with a personnel policy that fosters low turnover and high employee satisfaction to avoid losing investment in HR. Example: One MFI notes that while their salary is perhaps slightly below market rates in the MFI sector in Bolivia, it offers attrractive pension, health and insurance benefits as well as a flexible work schedules for employees that prefer a more balanced work/life schedule. The HR manager notes that this keeps their loan officer turnover rate lower than the rest of the market. X. OPERATIONAL STRATEGY 51 X. Operational Strategy Operational strategy checklist • Evaluate and strengthen operational controls in anticipation of expansions • Conduct a full mapping of processes and procedures for new products/segment with Operations department to ensure each step is accounted for • Ensure that internal controls are strengthened at the branch level through branch operations team • Consider potential changes in cash needs of branches and review insurance, transportation and cash management policies • Include IT in discussions to address any need to add cells or information requirements for new products • Ensure IT reports segment loans by MSME segment and sector and highlight VSE A. Discussion of Organizational Structure and adaptations needed to various areas Operations are critical to all aspects of an MFI and will need to adapt to any changes when loans or other products are developed or risk management processes and controls are changed. These changes are not especially relevant to VSE lending but come up at any point when MFIs develop new products. Thus while operations should not be forgotten, they will not likely require strategic decisions or broad changes. It will be important to review the Op Tip #1: Consider changes in cash policies, processes and protocols for any new VSE loan management needs at branches. with operations to ensure that these are communicated to branches through branch managers and that protocols Example: One MFI in Peru notes that larger loans are followed. Where credit analysis is centralized for VSE let to a need to become more efficient with the loans, for example, operations staff will be made aware of institution’s cash management. Armored trucks, new protocols for loan approvals. If these involve greater for example, were costly and charged fixed rates attention or new processes for operations staff at the per mobilized vehicle and number of stops but also branch level, these will need to be communicated, and variable fees depending on how much money was potentially trained. being moved. By increasing loan amounts, the volume of cash at each branch increased and so did Additionally, in some cases, where loan sizes and turn over the costs of moving this cash. The MFI addressed this are much larger and more frequent, cash management and by centralizing the decision of when to move money insurance needs must be reviewed and potentially revised rather than allowing branches to call in for trucks (See Tip #1). This includes the maximum allowable cash (they were typically more conservative and preferred held and insured at branches and transportation of cash. to stay “light” on cash. Additionally, they increase Because this might add costs to existing processes, it is the amount of cash that each truck carried, ensuring important to evaluate this and optimize costs. to stay below the insured limit by sending trucks out with greater frequency. 52 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA B. IT changes and adaptations The Bare Essentials An expansion upmarket to serve VSEs with ease may require adaptations and even an upgrade of an MFI’s information technology. An appropriate MIS is essential to successful expansions in that in can help an MFI manage its risk and also control costs. Some of the key adaptations and requirements for the IT system include: • Adding the VSE loan product(s) to the MIS: As discussed above, the terms of the basic loan products for microenterprises and VSEs are different, most notably in their sizes, but also in their tenors and possibly their repayments cycles. The MIS must be flexible enough to accommodate these differences. • Adding other VSE products to the MIS: VSEs are OP Tip #2: Launching new products also interested in other credit products, such as credit lines in conjuntion with IT upgrades can lines, in which the balance outstanding can fluctuate cause delays in both and also increase daily. Calculating the interest due, managing liquidity an institution’s risk. and monitoring the risk of such products require more robust and sophisticated MIS than typical microcredit Examples: We visited an MFI that had recently products. An MIS that is not transactionally strong upgraded its IT system. The new system is robust and enough to incorporate these products may constrain the allows for a wide range of VSE products, however, institution’s ability to offer such products to VSEs. the adjustment period has been difficult. The IT staff have been very busy with migrating the data to the • Ability to track the VSE and micro portfolios new system and unable to respond quickly to the separately: Monitoring and tracking the MFIs VSE and data requests from the business units. ME portfolios is essential for risk management purposes. It allows management to monitor the expansion upmarket from an early stage and to identify risks and make adjustments quickly to policies and procedures. At a minimum the system needs to have a field that allows the institution to identify a VSE client/product from its micro clients and the ability to produce reports based on this. The more automated these reports are the better. Most of the institutions that were interviewed for this document are bank-like institutions and already had core-banking software in place prior to expanding upmarket. This software is typically flexible and adaptable enough to accommodate VSE loans with longer tenors for example. Thus, the expansion upmarket required tweaks to the existing software (i.e. adding the new VSE loans to the system, activating new fields in the system, developing new reports and interfaces to extract data from the system, etc.), but not a major change to the MIS or an upgrade. However, MFIs that do not already have robust core-banking systems in place, likely the non-regulated institutions, may encounter greater challenges in getting their MIS ready to accommodate the new sector. In some cases the MIS may be so inflexible that the only way incorporate the VSE business will be to upgrade to a new system. This can be very costly and time consuming and may impact an MFI’s decision to enter the market all together. X. OPERATIONAL STRATEGY 53 The Bells and Whistles In addition to the basic IT requirements above, there are many other ways technology can be used to facilitate an expansion upmarket. Generally, the more robust the system, the better. An ideal IT system would typically address: • The ability to track the income and expenses of the VSE and micro portfolios separately in order to monitor the profitability of each segment and to facilitate business line management. • The need for segmentation and analysis for market strategy and product development/refinement. This may require an additional software that can be linked to the core MIS. It may also be useful for an MFI to have statistical analysis software to help in its segmentation. • The need for data for risk management purposes. As above, this might require using additional software to OP Tip #3: Technology can be used to manipulate data more flexibly. Down the line, the MFI help reduce costs and streamline pro- may begin to use its data to develop risk management cesses and procedures for VSE loans. alerts, or a scoring system for VSEs. Examples: A Peruvian MFI has digitalized a portion • Flexibility in designing new reports. The system of their loan files in order to reduce the paperwork should allow for flexibility in designing reports and burden of VSE loans. This digitalization has helped allow middle office users to customize their reports, it control costs, streamlined loan approval for especially for monitoring and tracking purposes. The larger loans not approved at the branch level and business and risk units should be conducting the bulk strengthened its internal controls and auditing. of the data analysis for the MFI, and should not be beholden to the IT department. In one Peruvian MFI, for example, the IT system offers the possibility to download cross tabs for some specific information but does not allow for an analysis of VSE loans by gender, for example, or by branch. This can limit the sophistication of both market and risk analysis. • Paperwork for small business loans is more voluminous and adds costs – digitalizing parts of the loan file can add efficiency. Digitalization can also help to streamline processes and reduce transaction times when MFIs decide to adopt more centralized loan approval and risk management. For example, branches can upload loan applications and key paperwork and credit committees at the regional and/or headquarters level can easily review the files and make credit decisions. 54 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA XI. FINANCIAL STRATEGY 55 XI. Financial Strategy Financial strategy checklist • Estimate the upfront costs of VSE expansion and changes in cost structures • Conduct cost-benefit analysis and/or financial pro forma model of VSE expansion • Develop a method to measure and monitor the profitability of VSE loans and/or VSE clients over time • Modify ALM models to include longer tenor funding and lower interest rates • Revise funding strategy to guarantee adequate liquidity for VSE portfolio growth • Seek out lower cost funding, as VSEs are more interest rate sensitive than micro enterprises • Ensure there are longer-term funding sources available to match the longer maturities of VSE loans A. Costs and Profitability An expansion upmarket to serve VSEs implies a variety of upfront costs as well as a shift in the overall cost structure of the institution. The upfront costs are greater in the Proactive Model as there are more major changes to the organizational structure of the institution; however, FS Tip #1: Prior to making the move upmarket, an there are also upfront costs in the Organic Model related MFI should attempt to estimate upfront costs and to planning and research; policy, procedures and system ongoing shifts in cost structure and incorporate them tweaks; product piloting; etc. Overtime, it appears that into a cost-benefit analysis and/or financial proforma some of the upfront costs can offer long-term gains, model. specifically in terms of portfolio risk (and reducing the cost of provisioning). While little data exists, upgrading credit processes and controls from microfinance practices may play an important role in keeping delinquency down in this segment. Measuring Costs Depending on the market context, the ongoing cost structure of an MFI can also shift dramatically with an expansion upmarket. On the one hand, interest rates on FS Tip #2: Early movers into the VSE VSE loan are generally lower than micro loans, compressing market, may be better positioned financial margins. On the other hand an MFI is able to long-term as the market gets more disburse a larger amount to a single client and (ideally) competitive. reduce its operating costs per dollar lent. Cross selling Example: One Bolivian MFI was able to invest and may help boost revenues but for many MFIs this may be grow its VSE business prior to the market becoming limited. How loan losses affect cost structures depends on very competitive and interest rate senstive. Now, it the market, the exact segment and the strength of an MFI’s continues to be profitable. Interest rates are 11-12%, credit and risk management systems. compared to 20% for its micro loan portfolio, but MFIs expanding upmarket should ideally be able to delinquency rates are lower and costs contained. monitor the costs and profitability of their VSE business 56 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA and micro business separately. In the Proactive Model this FS Tip #3: Operational costs per client is likely easier to do as the business lines are already largely are higher for VSE loans, however, separated and are likely to have their own cost centers. operational costs per dollar lent should At the branch level, VSE and micro cost centers can be be lower. established and front-line staff assigned accordingly. Back- office and support staff and branch overhead costs will Example: Loan officers at two Peruvian MFIs need to be distributed based on usage. At a headquarters reported spending 3 to 4 times as long preparing a level, the VSE unit and micro unit should also be assigned VSE loan file compared to a micro loan file. In Bolivia their own cost centers and the costs of shared units (i.e. disbursement times for VSE loans ranged from two marketing, IT, etc.) should be distributed based on usage. to four weeks as loan officers prepared files and In the Organic Model it may be much more difficult to credit committees analized loans. separate out the costs of the VSE and micro enterprise businesses as the same staff are likely to work on both and it requires an MFI to be able to estimate their time allocations. Nonetheless, developing a product costing model or cost/revenue models can be very useful as it allows an MFI to make informed decisions about its VSE business; to identify inefficiencies in its processes and develop cost saving solutions. Per loan vs. per client profitability Product profitability can be analyzed per operation (loan) or per client. Given that the business case for VSE clients is intrinsically long-term, benefiting from the historical performance of a firm, an MFI’s relationship with the firm and its potential growth, there is an argument for attempting to analyze the profitability on a per client versus per product basis for VSEs. A VSE borrower is a potentially “sticky” client, which is able to acquire more than one product at any given moment and over time. As such, analyzing total client profitability over time may be more relevant than understanding only the profitability of a specific loan product. While important, few institutions interviewed for this document had analyzed the cost of these loans, and none had determined the best approach to costing (per product or per client). This is largely because some of the institutions interviewed did not have product costing capabilities prior to expanding upmarket, nor did they develop them alongside the development of the VSE lending. Cross-selling Institutions able to offer a wider range of products may be able to offset costs and increase the profitability per VSE client through cross selling. However, it is important that MFIs not overestimate revenue generation through cross selling. Many of the VSEs served by MFIs have very similar profiles as their micro clients and cross-selling opportunities may be limited. Cross-selling strategies to keep client relationships “sticky” may be more appropriate, where MFIs generate greater loyalty while maintaining VSE loans, as the core revenue generator is the loan. XI. FINANCIAL STRATEGY 57 B. Asset/Liability Management and Funding Strategy Offering larger VSE loans can quickly work through an MFI’s funding, especially as VSE loans tend to be longer- FS Tip #4: Deposit taking institutions term loans. Thus, when expanding upmarket to serve have an advantage over non-deposit VSEs an MFI must have a strong risk management system taking institutions in terms of reducing and framework in place. Asset/liability management, fuding costs. on particular becomes critical. One MFI interviewed for this study demonstrated sophisticated monitoring of Example: One of the reasons that One MFI we visited their financial risks including asset/liability and liquidity in Bolivia transformed from an NGO to a Bank was to management. The institution integrated this process increase their funding options to support the growth into general management and commercial processes and of their VSE and SME portfolios. It is now financing planning to ensure risk was both monitored and managed. over 60% of its portfolio through deposits at a An institution moving into VSE should carefully consider financial cost of only 2.7%, making its average cost its funding strategy and ensure adequate funding is of funding only 5.3%. available for VSE portfolio growth. Portfolio returns are rarely sufficient to fund portfolio growth and even less so VSE portfolio growth, thus MFIs must seek other sources of funding, be it from local depositors, domestic or lenders, international investors/microfinance investment vehicles (“MIVs”), or domestic capital markets. The ability to lower funding costs is key to economically serving VSEs, who are more interest rate sensitive than micro loan clients. Thus, institutions able to capture low- cost deposits or offer no-cost current accounts have an advantage over NGOs or other non-deposit taking institutions when serving VSEs. However, many MFIs expanding upmarket may not legally be able to take deposits, or may fully have developed these services yet. The increase in the number of Microfinance Investment Vehicles (MIVs), as well as funds from multilateral investors, has made it easier for these MFIs to consider expanding upmarket. Larger and stronger MFIs, in particular, have some bargaining power and are able to get larger loans, for longer tenors and at somewhat preferential interest rates from these sources. VSE lending may require an MFI to review its asset and liability management (ALM) and make adjustments to FS Tip #5: Care must be taken to match its funding sources to mitigate mismatches in currencies maturities and currencies of VSE loans and maturities. Deposit taking MFIs with ample liquidity and funding sources, however interest may find it beneficial to access more costly international rates on such loans are typically high. investments in order to match the maturities of their assets Example: We interviewed a Peruvian Cooperative and liabilities, especially if their deposits tend to be very MFI that funds the majority of its lending through short term or on demand (See FS Tip #5). Local currency member deposits, but finds it necessary to bonds in local markets can address the problem of currency supplement these funds with external longer-term mismatches that international investors often don’t. In financing. This financing is relatively small, however both Peru and Bolivia, the local bond/commercial paper and its overall funding cost is not largely affected by markets are growing and some MFIs are able to access the higher interest rates on such loans. funds there. The terms are still relatively short, generally 58 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA one-year, but the intention is to try to build out a longer-term yield curves, where a promising, local currency, longer-term source of funding in the future. XII. CONCLUSIONS 59 XII. Conclusions This document aims to explore some of the key strategic and commercial drivers for MFIs expanding upmarket into VSE lending, and to outline many of the decisions and adaptations that these institutions need to make when moving into the segment. It assumes that MFIs expanding upmarket are of sufficient size and resource capacity to consider entering new markets or deepening their presence in existing market. This document is written from the general perspective of MFIs stepping “up” or perhaps even leaping “up” into the VSE sector, where microcredit is the initial starting point and VSEs are a key next step. However, this document can also be useful for financial institutions seeking to ‘down-scale’ into VSE lending – the checklists are still valid, even if the starting point for the institution is at the opposite end of the axis from MFIs. This document is based on the Latin America and Caribbean (LAC) VSE lending experience, specifically based on interviews and field visits with institutions from the region. Our desk review revealed that within LAC, many of the MFIs with the most experience in VSE lending come from countries that have “leading” microfinance markets. Often these have strong regulatory regimes, widespread use of credit bureaus, and a competitive market environment. Many of these countries have also experienced a period of strong economic growth over the last years, which has trickled down to the micro and small businesses that are served by the sector. As such, the experiences referenced in this document are to some extent “untested” as they have not been challenged by severe downturns in growth or financial sector strains. • Many LAC MFIs are well positioned to move into VSE lending, if they haven’t already begun to “organically”. MFIs in LAC represent a broad range of sizes, expertise and markets for small loans. According to the IDB/MIF,8 the microcredit portfolio in LAC includes over US$40 billion offered through over 1,000 institutions and serving at least 20 million clients. While loans across the region’s MFIs average US$2,000, many can be much larger. Regulated institutions, which represent 86% of all MFIs, in particular average US$2,500 loans (compared to an average US$800 for unregulated institutions). These may be best positioned to have the structure, systems and capital to branch out into larger loans within our definition of VSEs (US$7,000-30,000).9 • However, most MFIs in LAC have difficulty offering low interest rates to price- sensitive VSEs. Most MFIs have trouble competing with banks, which have lower funding cost structures. As such, they often charge higher average interest rates on their loans and presenting a potential competitive disadvantage when it comes to VSE lending. Unlike banks, most MFIs obtain almost all of their revenues from lending. They are generally unable to diversify loan revenues with fee-based income because of their limited product offering as well as the liquidity constraints of the low-income segment they serve. Additionally, the operational cost of lending to microentrepreneurs exceeds that of banks, especially for smaller microfinance loans. 60 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA • MFI’s must compete by offering convenient, flexible access to loans, and targeting those segments that are least served by banks. MFIs’ typically higher cost of funding and resultant difficulty in competing with banks on interest rates pose a challenge for these institutions who may otherwise offer an attractive value proposition to VSEs in terms of convenient, flexible and agile services. In markets where formal alternative VSE funding exists (primarily from downscaled banks), this pressure can be brutal, pushing interest rates down and forcing some institutions to relax credit policies excessively to maintain an advantage over their less flexible bank competition. In the case of one institution interviewed for this study, this pressure has made the VSE sector look increasingly unprofitable as interest rates have come down to stay competitive while credit quality has declined as a result of excessively lax loan approval policies for VSEs. In other cases, where MFIs fill an identified gap, the VSE segment can be an attractive source of both growth and profitability. Efficiency and effective risk management and controls are key. • We identify two primary ways that MFI’s enter the VSE segment: Organically or Proactively. These drive many of the subsequent choices about commercial, operational, human resource, credit and risk management strategies. We analyze institutions that have grown organically into lending to VSEs (Organic models) and find that strong and efficient processes can help overcome some of the additional costs of lending to VSEs, which should typically involve Proactive vs. Organic Models for MFIs more in-depth loan analysis implemented through more serving VSEs experienced and higher-paid staff. While investing in more Proactive models (See Box), which include more in- The “Proactive Model- (P)” borrows from the depth (and lengthy) credit analysis, human resources and know-how and processes most typical in banks. It capacity building, risk management and controls can be is more complex and/or deliberate in that it requires costly, it may eventually be necessary for all MFIs in the important structural decisions and adaptations from VSE segment. These MFIs can start lending to VSEs that the start. The model has been actively disseminated represent their own “graduated” clients early on. Their by Pro Credit (IPC 2012). It generally targets a relationship with the VSEs and familiarity with their business segment that is outside of an MFI’s existing business and character can help overcome some of the client base, and may also be underserved by the weaknesses inherent in upscaling credit methodologies financial system. designed for microenterprise to assess VSE risk. The “Organic Model-(O)” follows a more organic • Regardless of whether MFIs enter the VSE market process, often in reaction as an MFI’s microenterprise by lending to existing clients, they typically start portfolio begins to grow in asset size by more than seeking new VSE clients as they grow. Only a small its client growth. While it may initially represent a percentage of microenterprises tend to graduate to reaction to client trends, it does require strategic VSEs, limiting the MFIs growth in the sector, and over decision-making and adjustments to the traditional time, existing clients may also hit ceilings, unable to microfinance business model. However, these grow or expand because their own business capacity may adjustments are less pronounced compared to the be challenged. Thus, MFIs will also likely stray from Proactive Model. They typically allow the MFI to purely serving graduated clients and move into new target a new market segment without major costs or markets. Some will be clients of other institutions and changes to its operational structure early on. others might be “blue ocean” opportunities. XII. CONCLUSIONS 61 • Growth and expansion to serving new VSE clients will require more effort in balancing an institution’s commercial strategy and growth with appropriate credit risk management. VSE lending requires a more in depth effort to understand the business, its risks and opportunities and potential benefits of investment. Growth must be carefully balanced with a more cautious approach to larger loans. Even MFIs who enter the VSE segment somewhat organically, must eventually adopt a more Proactive approach to VSE lending. They must manage their risk more proactively as well as measure and analyze it in more depth. Often, this process requires an investment in resources to strengthen processes and systems as well as an increase in funding for VSE lending. 62 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA BIBLIOGRAPHY 63 Bibliography General SME and SME Financial Background Beck, Thorsten, et al (2008): “Banking Financing for SMEs around the World: Drivers, Obstacles, Business Models, and Lending Practices.” World Bank Policy Research Working Paper, No. 4785, November 2008. De la Torre, Augusto, et al (2008): “Banking Involvement with SMEs: Beyond Relationship Lending.” World Bank. Ernst & Young (2013). “Technical Note on the Verification of IFC’s SME Loan Size Proxies.” DRAFT. Prepared for the IFC. Gibson, Tom and H.J. van der Vaart (2008): “A Less Imperfect Way of Defining Small and Medium Enterprises in Developing Countries.” Brookings Institution. Hinton, Peter (2010): “Developing SMEs: a case study in transformational development.” SDG. Powerpoint presentation. IFC (2009). “SME Banking and Knowledge Guide.” Khrystyna, K., M. Mirmulstein, and R. Ramalho. 2010. “Micro, Small, and Medium Enterprises around the World: How Many Are There, and What Affects the Count?” IFC and World Bank. Leeds, Roger (2003). “Financing Small SMEs in Developing Countries: Learning from Experience”. Transnational Publishers, 2003. Mckinsey (2012) Micro-, small and medium-sized enterprises in emerging markets: how banks can grasp a $350 billion opportunity. Banking Practice Report. OECD (2006). “Financing SMEs and Entrepreneurs.” Policy Brief. Organization of Economic Cooperation and Development, SME and Entrepreneurship Division (2006): The SME Financing Gap, Volume I, Theory and Evidence SMEs. Sia, Manuel & Nails, Donna (2008). “Winning Strategies for Successful Small Business Lending.” Exchanging Views Series #6, ShoreCap Exchange. Stein, Peer, Tony Goland, Robert Schiff (2010). “Two trillion and counting: Assessing the credit gap for MSME in the developing world.” UNCTAD (n.a.): “Growing Micro and Small Enterprises in LDCs. The Missing Middle in LDCs. Why micro and small enterprises are not growing.” ITE/TEB/5. 64 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA UNDP (2004): “Unleashing Entrepreneurship: Making Business Work for the Poor”. Commission on the Private Sector and Development. UNDP (2008): “Creating Value for All: Strategies for Doing Business With the Poor”. USAID (2009). Enabling SME Access to Finance. FS Series #4. World Business Council for Sustainable Development (2007): “Promoting SME for Sustainable Development”. LAC Specific Asociacion de Supervisores Bancarios de las Americas (2012): “How to Contribute to Financial Inclusion: Inventory of Products and Services Available in the Region. Banco Central de Reserva del Peru (2012). “Reporte de Estabilidad Financiera”. Noviembre 2012. IDB 2012. “Las PYMES de America Latina y el Caribe: Un Negocio Estrategico para los Bancos de la Region.” 5th Regional Survey in Latin America and the Caribbean. IDB. Rojas-Suarez, Liliana, Federacion Latinoamerica de Bancos (2007). “Promoting Access to Financial Services: What does Data about Bancarization in Latin America Tell us?”. FELEBAN. Titelman, Daniel (2011): “Financiamiento en Costa Rica: miro, pequeña y mediana empresa. Qué nos dice la demanda?” CEPAL. Upscaling Specific CGAP (2012). “Financing Small Enterprises: What Role for MFIs?” Focus Note No. 81. CGAP. CGAP (2011). Role of MFIs in Serving Micro and Small Enterprises, Industry Survey 2011. CGAP. Ferranti, David and Anthony Ody (2007). “Beyond Microfinance: Getting Capital to SMEs to Fuel Faster Development.” Policy Brief 159. Brookings Institution. Gonzales, Adrian (2012). “Are Non-Microenterprise Loans Less Resilient to Domestic Shocks? Lessons from the 2009 Economic Recession.” MicroBanking Bulletin, March 2012. Microfinance Information Exchange. GTZ. “Upscaling: Enhancing Access to SME Credit.” SMEDSEP Information Brief 5. BIBLIOGRAPHY 65 Databases and Resource Banks www.smetoolkit.org – in particular local counterparts for regional toolkits rru.worldbank.org/Documents/Other/SMEDatabase.xls MSME-CI-Data.xls. “MSME country indicators.” http://smefinanceforum.org www.gmsbizforum.com -> SME Finance Conference www.enterprisesurveys.or 66 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA ENDNOTES 67 Endnotes 1. Stein (2010) estimates that the total unmet need for credit by all MSMEs in emerging markets today is in the range of $2.1-2.5 trillion, and that of the estimated 365-445 million MSMEs in the developing world, approximately 70 percent do not use external financing from financial institutions. The global SME financing gap is estimated at $700-850 billion. 2. Very small enterprise is a term recently introduced by the IFC that refers to smaller small enterprises. See the next section for a discussion of firm size definitions. 3. For example, a study in Costa Rica study showed that more SEs reported there to be less than sufficient access to finance than micro or large enterprises, where SEs were defined as those with 6-30 employees. 4. De la Torre (2008) found that small firms finance, on average, 13 percentage points less of their investments with external finance when compared to large firms. Stein (2010) estimates that the total unmet need for credit by all MSMEs in emerging markets today is in the range of $2.1-2.5 trillion, and that of the estimated 365-445 million MSMEs in the developing world, approximately 70 percent do not use external financing from financial institutions. The global SME financing gap is estimated at $700-850 billion. 5. We used the IDB’s definition in place of the IFC definition for 2 reasons: 1) it was designed to be regionally specific and 2) an Ernst and Young study noted that the IFC’s loan size proxy for medium enterprises often captured “small” enterprises according tot the IFCs’ other criteria (employees, sales and assets). Although it is perhaps a bit wide, it encompasses the bulk of the different loan size definitions we encountered in our phone interviews and secondary research. 6. Or in the case of Peru, beginning from US$7,000 7. According to the results of FELABAN’s survey, the use of specialized risk assessment techniques in assessing risks, in micro, small, and medium size enterprises and clients from medium-low income levels is still very poor. As a result, a large part of these clients are considered high-risk and low-profitability clients; therefore, non- desirable from a financial point of view. 8. Trujillo, Veronica, Microfinance in Latin America and the Caribbean: The Sector in Numbers 2013, FOMIN, 2013 9. Trujillo, Veronica, Microfinance in Latin America and the Caribbean: The Sector in Numbers 2013, FOMIN, 2013 68 EXPERIENCES OF MICROFINANCE INSTITUTIONS SERVING VERY SMALL TO SMALL ENTERPRISES IN LATIN AMERICA 2121 Pennsylvania Avenue, NW Washington, DC 20433, USA www.ifc.org JULY 2014 The original had problem with text extraction. pdftotext Unable to extract text.