75156 A Guide to ­Regulation and Supervision of ­Microfinance Consensus Guidelines October 2012 CGAP is an independent policy and research center dedicated to advancing financial access for the world’s poor. It is supported by over 30 development agencies and private foundations who share a common mission to improve the lives of poor people. Housed at the World Bank, CGAP pro- vides market intelligence, promotes standards, develops innovative solutions, and offers advisory services to governments, microfinance providers, donors, and investors. © 2012, CGAP/World Bank. All rights reserved. CGAP 1818 H St., N.W., MSN P3-300 Washington, DC 20433 +1 202 473 9594 www.cgap.org www.microfinancegateway.org Table of Contents Prefacevii Introduction1 Part I. Preliminary Issues 4 1a. Terminology: What Is “Microfinance�? What Is “Financial Inclusion�? 4 1b. Financial Inclusion as a Regulatory Objective; Regulation as Promotion 8 Special windows for microfinance 9 Create a new framework or amend an existing one? 10 Regulatory arbitrage 11 1c. Regulatory Definitions of “Microfinance� and “Microcredit� 11 Box 1. Crafting a regulatory definition of “microcredit� 12 1d. Prudential and Nonprudential Regulation: Objectives and Application 14 Box 2. A systemic rationale for applying prudential regulation to microlending-only institutions? 16 1e. Regulate Institutions or Activities? 18 Part II. Prudential Regulation of Deposit-Taking Microfinance 19 New Regulatory Windows for Depository Microfinance: 2a.  Timing and the State of the Industry 19 2b. Rationing Prudential Regulation and Minimum Capital 21 2c. Adjusted Prudential Standards for Microfinance 22 Permitted activities 23 Capital adequacy 23 Capital adequacy for financial cooperatives and their networks 25 Unsecured lending limits and loan-loss provisions 26 Governance28 Liquidity and foreign exchange risk 28 Loan documentation 30 Restrictions on co-signers as borrowers 30 Branching requirements 31 Reporting31 Reserves against deposits 32 iii iv Microfinance Consensus Guidelines Insider lending 32 To whom should special prudential standards apply? 33 Box 3. Basel core principles and depository microfinance 33 2d. Transformation of NGO MFIs into Licensed Intermediaries 34 Ownership suitability and diversification requirements 35 Board and management qualification requirements 36 Loan portfolio as part of minimum capital 36 2e. Deposit Insurance 37 Part III. Prudential Supervision Issues in Deposit-Taking Microfinance 39 3a. Supervisory Tools, Enforcement Mechanisms, and Limitations Regarding MFIs 40 Microcredit portfolio supervision 40 Stop-lending orders 40 Capital calls 41 Asset sales or mergers 41 3b. Costs of Supervision 41 3c.Where Should the Microfinance Supervisory Function Be Located? 42 Within the existing supervisory authority? 42 Delegated (or auxiliary) supervision 43 Self-regulation and supervision 44 Supervision of financial cooperatives 45 The case of small member-based intermediaries 47 Part IV. Nonprudential Regulatory Issues 49 4a. Permission to Lend 49 4b. Reporting and Institutional Transparency 51 4c. Consumer Protection 52 Box 4. Level playing fields, equal treatment 53 Adequacy and transparency of information 53 Discrimination54 Abusive lending and collection practices 55 Over-indebtedness55 Interest rate caps 56 Data privacy and security 58 Recourse59 An appropriate regulator 59 4d. Credit Reporting Systems 60 Box 5. The benefits and challenges of credit reporting in microfinance 61 4e. Limitations on Ownership, Management, and Capital Structure 63 Table of Contents v 4f. NGO Transformations into For-Profit Companies 64 4g. Secured Transactions 65 4h. Financial Crime 66 AML/CFT66 Fraud, pyramid investment schemes, and related financial crimes 68 Identity fraud 69 4i. Tax Treatment of Microfinance 70 Taxation of financial transactions and activities 70 Taxation of profits 71 Part V. Regulating the Use of Branchless Banking to Serve the Poor 72 Box 6. Bank-based and nonbank-based models 73 5a. Agents and Other Third-Party Arrangements 74 5b. AML/CFT in Branchless Banking 76 5c. Nonbank Issuers of E-Money and Other Stored-Value Instruments 76 5d. Consumer Protection in Branchless Banking 77 5e. Payment Systems: Regulation and Access 78 5f. Interagency Coordination 79 Part VI. Regulating Microfinance Providers in Microinsurance 80 6a. What Is Microinsurance? 80 Defining microinsurance 80 Roles in delivering microinsurance 82 Types of microinsurance products 82 6b. MFIs and Microlending Banks in Microinsurance 82 MFIs as insurance underwriters 83 MFIs as insurance intermediaries 84 Microlending banks in microinsurance 85 The special case of credit life insurance 85 6c. Regulation of Microinsurance Sales 87 Sales agents 88 Commissions89 Group sales 89 Disclosure, claims, consumer understanding 91 Bundling92 Part VII. Summary of Key Observations, Principles, and Recommendations 94 7a. General Issues 94 Regulatory definitions 94 Regulatory “windows� for depository MFIs 95 vi Microfinance Consensus Guidelines 7b. Prudential Regulation 95 Rationing prudential regulation and minimum capital 95 Adjusting prudential standards for microfinance 95 To whom should special prudential standards apply? 96 Transformation of NGO MFIs into licensed intermediaries 97 Deposit insurance 97 7c. Prudential Supervision 97 7d. Nonprudential Regulation 98 Permission to lend 98 Reporting98 Consumer protection 98 Credit reporting systems 99 Limitations on ownership, management, and capital structure 99 NGO transformations into for-profit companies 99 Secured transactions 100 Financial crime 100 Tax treatment of microfinance 100 7e. Branchless Banking 100 7f. Microinsurance 101 Appendix A. Authorship and Acknowldgments 103 Appendix B. Glossary 105 Appendix C. Microlending Institutions and Their Funding Sources 109 Appendix D. Bibliography 110 Preface This Guide updates CGAP’s 2003 Guiding Principles on Regulation and Supervision of Microfinance. The revisions reflect continuing developments in the global state of finan- cial access for poor and low-income customers, including the following: • Increased attention to financial services beyond microcredit • Proliferation of new providers and financial service delivery mechanisms • Rapid evolution of regulatory frontiers, such as the regulation of bank and nonbank agents and e-money issuers • Increased funding from the private sector and quasi-commercial public investors • More countries that have extended experience with microfinance and microfinance regulation and supervision • In some countries, increased transformation of microfinance institutions from non- profit to for-profit • Competitive saturation of microcredit markets in a growing number of countries, which can increase portfolio risk and heighten consumer protection issues • Integration of microfinance into mainstream finance institutions and markets • Emerging consensus about creating a level playing field, and the role of regulating “by activity� to achieve that goal • Focus of international financial standard-setting bodies on the need for proportionate regulation and supervision that does not result in the exclusion of low-income customers In contrast to the situation a decade ago, most policy makers, donors, and private investors involved in microfinance now appreciate that poor and low-income people, like the rest of us, need a variety of basic financial services, not just credit.1 The ability of the market to re- spond to this demand depends not only on providers developing sustainable, low-cost ways to provide such services, but also on having an enabling policy and regulatory environment. Appropriate regulation and supervision of financial service providers is therefore critically important in bringing to poor and low-income people the financial services they need. We hope that policy makers will find this updated Guide useful as they confront this challenge. 1 Even when low-income people have no access to a formal financial institution, they actively use a spectrum of informal providers for loans, savings, insurance, and fund transfers. See Rutherford (2000) and Collins, Murdoch, Rutherford, and Ruthven (2009). vii Introduction In the past decade, financial authorities in most developing and transitional economies have put more emphasis on bringing formal financial services to the large numbers of the world’s poor who currently lack adequate access.2 For many, the question has been whether and how to regulate “microfinance�—a term that evokes a particular set of services, providers, and customers. But the question is now being posed in broader terms: what kind of regula- tion and supervision will help to achieve full financial inclusion through the extension of fi- nancial services to the billions of poor and low-income people who are presently excluded? Even more broadly, how should we regulate and supervise the financial system as a whole in a way that balances effective access, financial stability, and financial integrity?3 ­ ction This effort requires policy makers to weigh the potential benefits of regulatory a against potential limitations on access due to the costs of compliance and enforcement. Regulation and supervision need to be proportionate: costs should not be excessive when measured against the risks being addressed (although both are difficult to measure, and there will very likely be differences of opinion among regulators, providers, and ­consumers).4 However, this balancing effort is particularly important to financial access, where cost reduction is crucial for expanded outreach. Scope. This Guide addresses regulatory and supervisory issues that are specifically relevant to formal financial services for poor people. While there is a need to think of microfinance in the context of the wider financial architecture, this Guide focuses on the specifics of microfinance regulation and supervision and addresses issues and principles applicable to financial sector regulation and supervision more generally only when neces- sary to understand the specifics of regulating and supervising microfinance. It focuses on private actors, both for-profit and nonprofit. A wide variety of state-owned ­ ddress the or state-controlled institutions also serve poor people, but this Guide does not a ­ iscussion of the issues that specific issues such institutions pose: they are too diverse to allow d would apply to each type. For the same reason, this Guide does not devote separate attention 2 To avoid repetition of the term “poor and low-income,� we often refer to clients as simply “poor.� Readers should understand that the term refers to a broader group that includes not only poor people but also low- income clients above the poverty line. 3 As described in a 2011 white paper prepared by CGAP on behalf of the G-20’s Global Partnership for Financial Inclusion (GPFI), “‘effective access’ involves convenient and responsible service delivery, at a cost affordable to the customer and sustainable for the provider� (p.1). 4 See Porteous (2006). 1 2 Microfinance Consensus Guidelines to savings banks, even though they are major providers of financial services for poor people in many countries.5 Nevertheless, most of the general principles in this Guide would apply to state institutions as well as private ones and to both state-owned and privately owned savings banks. Audiences and format. Financial regulators and supervisors are the primary intended audience for this Guide. But this Guide may also interest a wider audience, including not only other authorities whose decisions affect financial services, but also service providers and other local stakeholders who participate in the regulatory and supervisory decision- making process and live with the results, as well as staff of international agencies that support governments on financial inclusion policy. Some readers will use this Guide as a general introduction to the full range of topics covered; others will consult it as a reference on specific issues. Most sections begin with key points (in shaded boxes), followed by discussion and analysis. On some issues covered here, experience justifies clear conclusions that will be valid everywhere with few exceptions. On other points, the experience is not clear, or the an- swer depends on local factors, so that no straightforward general prescription is possible. On these latter points, the Guide suggests frameworks for thinking about the issue and identifies factors that need special consideration. Country-specific factors. Although the key points and analyses in this Guide are based country-specific on country experiences, we have intentionally avoided discussion of ­ ­ examples. Instead, the focus is on extracting from global experiences those general prin- ciples and recommendations that may be relevant across a range of country contexts.6 When considering the issues presented in this Guide, country context—including regu- latory framework; retail providers’ level of development; capacity and constraints of supervisors; and other political, economic, historical, and cultural factors—is critical. It is unwise to use another country’s regulatory regime as a template without thoroughly examining country-specific factors that might call for different treatment. In what sense is this Guide a consensus document? We developed the material in this Guide in consultation with a wide range of regulators, supervisors, and other experts. (Key contribu- tors are listed in Appendix A.)7 Although experts working on these topics do not agree on all points, there are wide areas of consensus. Based on our consultations, CGAP believes that the 5 For a discussion of regulation and supervision from the perspective of savings banks engaged in microfi- nance, see WSBI (2008). 6 See, e.g., Trigo Loubière, Devaney, and Rhyne (2004), p. 3: Arguing against “criticizing one country as having a deficient framework or another as having an exemplary one� and arguing for the premise that “each framework reflects the realm of possibilities that were present as microfinance entered the financial picture,� the paper advises “designers of new microfinance regulation, and the international advisors who propose microfinance frameworks [to] take these realities into account.� 7 Acknowledgment of these contributions does not imply that the individuals or their organizations have endorsed all the contents of the Guide. Introduction 3 main messages of this Guide reflect the views of most specialists with wide knowledge of past experience and current developments in regulation and supervision for financial inclusion. Other resources. Because this Guide focuses on regulatory and supervisory issues that are specific to poor people’s financial services, it does not address many broader principles of financial sector regulation and supervision. For these broader issues, readers should consult ­ the core principles and other publications of the relevant standard-setting bodies (SSBs), par- ticularly the Basel Committee for Banking Supervision (BCBS), the Committee on Payment and Settlement Systems, the International Association of Insurance Supervisors (IAIS), the International Association of Deposit Insurers (IADI), and the Financial Action Task Force (FATF).8 As noted in the 2011 white paper CGAP published on behalf of the G-20 Global Partnership for Financial Inclusion (GPFI), these five SSBs are paying increased attention to microfinance and financial inclusion.9 The contents of this Guide are generally consistent with the broad principles of the SSBs as well as their specific guidance relevant to financial inclusion. Recommendations on regulating and supervising to foster financial inclusion have also been published by various think tanks, civil society organizations, and industry groups.10 These sources are referenced in Appendix D. Organization. The Guide is structured as follows: • Part I discusses preliminary issues, such as definitions of microfinance and micro- credit, financial inclusion as a regulatory objective, and the difference between pru- dential and nonprudential regulation. • Part II discusses prudential regulation of deposit-taking institutions engaged in ­microfinance. • Part III looks at the challenges surrounding supervision of depository institutions engaged in microfinance. ­ • Part IV discusses nonprudential regulation of both depository and nondepository ­ institutions engaged in microfinance. • Part V addresses regulation of branchless banking. • Part VI discusses regulation of microinsurance, especially when it is sold or adminis- tered by microfinance providers. • Part VII summarizes the key observations, principles, and recommendations. 8 The relevant principles and publications of these bodies are listed in Appendix D. 9 See, e.g., BCBS (2010); G-20 Principles for Innovative Financial Inclusion (http://www.g20.utoronto. ca/2010/to-principles.html), and Multi-Year Action Plan on Development (http://www.g20.utoronto.ca/2010/ g20seoul-development.html). See also FATF (2012a); FATF (2011); IAIS and CGAP Working Group on Microinsurance (2007); and IAIS, Microinsurance Network, and Access to Insurance Initiative (2010). ­ 10 See, e.g., the Center for Global Development’s Policy Principles for Expanding Financial Access (http:// www.cgdev.org/content/publications/detail/1422882/); the Smart Campaign’s Client Protection Principles (http://www.smartcampaign.org/about-the-campaign/smart-microfinance-and-the-client-protection-principles); and WSBI (2008). Part I. Preliminary Issues This Part opens with a discussion of some basic terms used in this Guide. (Appendix B, Glossary, covers a much longer list of terms.) We turn then to enabling regulation—that is, regulation whose explicit purpose is to promote the development of providers and products that serve the financially excluded poor. Next we discuss regulatory definitions of microfinance and microcredit, which may be different from the definitions of those terms as used in general discussion. Then we note the crucial distinction between pruden- tial and nonprudential regulation of microfinance. This discussion of preliminary issues concludes by examining the question of whether to regulate by institution, by activity, or by some combination of the two. 1a. Terminology: What Is “Microfinance�? What Is “Financial Inclusion�? In most countries the modern microfinance movement started with a focus on micro- credit, and only later embraced the importance for poor people of savings, money trans- fers, and insurance services.11 Microcredit still looms large in the self-image of most providers who identify themselves as “microfinance institutions� (MFIs). However, the vision of full financial inclusion is evolving, and recent years have seen a vocabulary shift away from “microfinance� and toward “financial access,� “financial inclusion,� and similarly broad terms. Nevertheless, this Guide uses the terms “microfinance� and “microcredit,� mainly because many countries already use this terminology in their regu- lations, and such use appears likely to continue. In this Guide, “microfinance� refers to the provision of formal financial services to poor and low-income (and, for credit, in particular, nonsalaried) people, as well as others systematically excluded from the financial system.12 As noted, “microfinance� embraces not only a range of credit products (for business purposes, for consumption smooth- ing, to fund social obligations, for emergencies, etc.), but also savings, money transfers, and insurance. 11 Financial cooperatives have a different history in much of the world, and have been taking savings from low-income clients since long before the modern microfinance movement. 12 “Formal� indicates services provided by an institution that is legally registered with a government authority. 4 Preliminary Issues 5 What does “financial inclusion� mean, and how does it differ from “microfinance�? In this Guide, “financial inclusion� refers to “the state in which all working age adults have effective access to credit, savings, payments and insurance from formal service pro- viders.�13 In many developing countries and transitional economies, this includes large numbers of households that are not considered poor or even low income by local stan- dards, as well as most small- and medium-size enterprises (SMEs) (at least with respect to access to credit).14 Both “microfinance� and “financial inclusion� refer to formal financial services— those delivered by providers that are registered with or licensed by a government. Though informal providers are not discussed in this Guide, it is important to recognize that most poor people use a wide variety of informal providers, even when they also have access to formal services. As used in this Guide, the terms “microcredit� and “microloan� have four important dimensions:15 1. A microloan is typically much smaller than a conventional bank loan, although there is no universally agreed maximum. 2. The loan typically has either no collateral or unconventional collateral (which fre- quently is not sufficient to cover the lender’s loss in the case of a payment default). 3. The borrower is typically self-employed or informally employed (i.e., not salaried by a formal employer). 4. The lender typically uses the common microlending methodology described below. For various reasons, this broad working definition of microcredit would not be suit- able for a regulatory definition. See “1c. Regulatory Definitions of ‘Microfinance’ and ‘Microcredit’.� Although most microcredit clients are “microentrepreneurs� in the sense that they have their own income-producing activities, they use their loans not only for business purposes but also for nonbusiness purposes, such as consumption smoothing or financing social, medical, and educational expenses. Notwithstanding this, micro- credit is distinct from typical consumer credit (e.g., credit cards or deferred payment for purchases), which usually involves scored lending to salaried people. Consumer 13 GPFI white paper (2011, p. 1). See footnote 3 regarding the meaning of “effective access� (a term also used in the white paper), which incorporates concepts of responsible delivery and affordability. 14 Many regulatory issues in SME finance (such as borrower insolvency rules) are of limited relevance in microfinance and are therefore generally not explored in this Guide. 15 The terms “microcredit� and “microloan� are used synonymously in this Guide, even though the terms “credit� and “loan� have distinct legal definitions in some regulatory systems. 6 Microfinance Consensus Guidelines credit raises distinct regulatory and supervisory issues that are not discussed in this Guide. In this Guide, the term “common microlending methodology� refers to lending ­ approaches that have been applied over the past four decades and that involve most, but not necessarily all, of the following: • The lender’s personal contact with the borrower • Group lending, or individual lending based on an analysis of the borrower’s (and/or borrower’s household) cash flow as opposed to scoring • Low initial loan sizes, with gradually larger amounts available in subsequent loans16 • An understanding that borrowers who repay their loans faithfully will have prompt access to follow-on loans • A “compulsory savings� requirement that must be satisfied by the borrower before receiving the loan to demonstrate the borrower’s willingness and ability to make pay- ments and/or to provide a partial “cash collateral� for the loan This common microlending methodology is perhaps the most important distinguish- ing feature of microcredit from a regulatory and supervisory perspective and is critical to many discussions in this Guide. What is an MFI? In this Guide, the term refers to a formal institution whose primary business is providing financial services to the poor.17 The range of institutional types delivering one or more microfinance services includes a wide variety of nongovernmental ­ organizations (NGOs); commercial finance companies (sometimes referred to as non- bank finance companies); financial cooperatives of various types; savings banks; rural banks; state-owned agricultural, development, and postal banks; commercial banks; and a large array of state-backed loan funds. Many institutions offer financial services to the poor alongside products targeting more affluent clients, and not necessarily as a primary business activity. This distinction can be important: often the risks that regulation and supervision are intended to address will be different in the context of a diversified finan- cial service provider. This Guide focuses on the following institutional types: NGO MFIs, commercial mi- crolending companies, commercial banks, microfinance banks, and financial ­cooperatives. 16 This practice is not as universal as it once was. Also, it should be noted that even where this is a common practice, loan sizes do not continue to grow indefinitely. The lender (or regulation) may impose an upper loan limit for all borrowers or for an individual borrower, and borrowers often reach a voluntary plateau in the size of loan they want. 17 Many organizations engaged in microfinance—in particular, domestic and international NGOs—give equal or greater priority to nonfinancial services, such as business training, agricultural training and inputs, health services, and education. Preliminary Issues 7 In some cases, both NGOs and financial cooperatives are discussed separately from other types of MFIs owing to their distinctive attributes. Financial cooperatives take many forms, including credit unions, savings and credit cooperatives, cajas, caisses, cooperative banks, and others. In many countries they serve large numbers of poor people. However, this Guide does not address in depth the specific prudential issues raised by financial cooperatives. Although they are all member-based and have in common the “one member one vote� rule, there are wide variations in their structure (including governance) and attributes across countries and regions. This variation is due in part to the three distinct traditions underlying most models (German, French/Canadian, and Anglo-American) as well as differences in financial cooperatives’ operations and in the size and composition of their membership. Given their importance to financial inclusion as well as widespread concerns about the quality of their oversight, regulators, policy makers, and SSBs may all have a role to play in improving prudential regulation and supervision of coopera- tive intermediaries. As noted, this Guide does not discuss the regulatory issues specific to state-owned institutions or savings banks, but many of the issues discussed here apply equally to them as well. In many developing and transitional countries, state-owned financial ­ institutions and savings banks (which may be state-owned) provide a substantial por- tion of poor people’s financial services. In some countries, these are the only formal financial institutions serving the poor. The goal of a level playing field would argue for subjecting state-owned banks—in particular, state-owned commercial banks—to the same prudential and nonprudential requirements as privately owned banks engaged in the same activities. However, there is wide disparity in regulatory treatment of state- owned banks in both developed and developing countries. This may be largely due to ­ ssumption that the government will back all liabilities of its institutions. While the a the discussion, given its complexity, is beyond the scope of this document, it should be ­ overnments to ensure that state institutions complement rather than replace a goal of g private sector actors. Savings banks are often regulated differently from commercial banks. Although this Guide does not include a discussion of the issues of particular relevance to the regulation and supervision of savings banks, the regulation and supervision of microfinance activi- ties discussed herein generally apply to savings banks as well.18 18 The World Savings Bank Institute (WSBI 2008) has articulated general principles with respect to the regula- tion of microfinance as well as recommendations for specific regulatory measures, although these measures are not aimed toward the regulatory regime for savings banks in particular. 8 Microfinance Consensus Guidelines 1b. Financial Inclusion as a Regulatory Objective; Regulation as Promotion Key Points • To craft and enforce appropriate regulation with a financial inclusion objective, regulators need to understand the distinctive characteristics of microfinance, in- cluding clients and their needs, products and services, and institutions providing those products and services. • Problems often arise due to inadequate coordination among financial regulators and other government agencies whose responsibilities may affect institutions de- livering microfinance. • Regulation creates costs for both the regulated institutions and the supervisor. These costs should be proportionate to the risks involved. • To the extent possible, regulation should aim to be institution-neutral (support- ing an activity-based regulatory approach), both to create a level playing field that fosters competition and to reduce risk of regulatory arbitrage. • In creating new windows for microfinance, regulators need to be alert to the possibility of regulatory arbitrage. Some countries create special microfinance windows with one sort of activity in mind, and then are surprised to find that the window is also being used for other activities that the regulators might not have been so keen to promote. The prevailing view of regulation and supervision of financial institutions centers on protecting the financial system as a whole, and financial institutions’ ability to repay their depositors in a cost-effective manner.19 Adding a further objective of promoting financial inclusion introduces three new vectors of responsibility and risk: new service provid- ers, new customers (who may be unfamiliar with formal financial institutions), and new products and delivery methods, such as branchless channels. To craft appropriate regula- tion and to supervise effectively, regulators need to understand the particular character- istics and risks of microfinance, including the clients and their needs, the products and services, and the institutions providing them. The microfinance regulator’s challenge is how to balance financial access, financial stability, financial integrity, and consumer protection. This complex and constantly 19 This is accomplished through (i) defining ownership requirements and permitted activities, (ii) requiring institutions to have appropriate risk management policies and to meet certain performance norms, (iii) build- ing systems to keep the supervisor informed about the institutions’ operations, and (iv) equipping the supervi- sor with appropriate tools, including staff and power to intervene. Increasingly, regulators are focusing on financial providers’ relationships with each other (e.g., competition, cooperation) and with their customers (e.g., bank secrecy, consumer protection), as well as on preventing criminal use of financial markets, such as money laundering or financing of terrorism. Preliminary Issues 9 changing balance requires on-going cost–benefit analysis.20 It often involves not just the financial regulator but also other government agencies, such as the consumer protection agency, the competition agency, the social welfare agency, and law enforcement authori- ties. Unless there is robust communication and coordination among these bodies, sub- stantial problems are likely to result. Regulation now being enacted for microfinance seeks not only to protect the finan- cial system and depositors, but also to promote poor people’s access to formal finance. To begin with, regulation can be promotional simply by enabling or facilitating basic microlending. In some countries, reform is required to establish clear legal authority for nonbank entities to engage in lending (see “4a. Permission to Lend�). This is impor- tant because commercial actors often have been willing to enter microfinance only after experimentation by NGOs or other noncommercial lenders. ­ Regulation can also be promotional by adjusting norms so that existing institutions can reach new customers or expand the range of services offered (e.g., by removing interest rate caps that make small loans unprofitable, or by adjusting prudential norms to address specific ­nvestment in microfi- issues with deposit-taking MFIs). And regulatory change can make i nance more appealing (e.g., through favorable tax treatment). Finally, regulation can enable the formation of new kinds of MFIs. Regardless of its objective, regulation should aim where ­nstitution-neutral (see “1e. Regulate by Institution or Activities?�), both to cre- possible to be i ate a level playing field that fosters competition and to reduce risk of regulatory arbitrage.21 Special windows for microfinance Often, a new “special window�—that is, a distinct regulatory category—is created for micro- finance. In some countries, new regulation has created a series of special windows, with the possibility of graduating from one to the next. The approaches include enabling the following: • Nonbank microlending institutions • Nonbank deposit-taking institutions (both to offer poor people a savings alternative and to access deposit funding for lending operations) • Some combination of these two, which is sometimes referred to as a tiered approach. 20 Calculating costs and benefits is not easy. Costs include those of the regulated institutions and those of the regulator, neither of which may be easy to estimate ex ante. Some benefits (e.g., consumer protection) are hard to quantify. Stakeholders value risks and benefits differently, and the risks of nonregulation may not be fully apparent before a crisis. 21 Competition is touched on only briefly in this Guide. However, the central aim of competition policy (maxi- mization of consumer welfare) is likely to be different in countries with a high percentage of poor people. In such countries, the distributional aspects of maximizing consumer welfare (i.e., how wealth is distributed among the society, including to poor people) can be very important when crafting and enforcing competition policy. See Adam and Alder (2011). 10 Microfinance Consensus Guidelines In deciding whether to open a special window, the first step should be to determine what, if anything, in the existing legal framework is holding back the market. (That said, however, the decision to open a new window is sometimes driven more by political than by technical considerations.) Experiences around the world show that a new window that removes a barrier to nonbank microlending (e.g., ending an explicit or implicit prohibition of lending by non- profits) is likely to increase the number of customers served fairly quickly (although they always the customers envisioned by proponents of the regulatory change). By are not ­ contrast, where the regulatory objective of a new window is to enable deposit-taking, results have been mixed. Sometimes the binding constraint has been scarcity of investors,22 or a lack of managers who can competently motivated entrepreneurs and ­ handle the risks involved when lending from deposits. In such cases, opening a special window by itself may not do much to increase services, at least until investor interest and management skill are better developed. (See “2a. New Regulatory Windows for Deposi- tory Microfinance: Timing and State of the Industry.�) Create a new framework or amend an existing one? If a new special window is to be created, should this be done by amending existing ­ financial sector laws and/or regulations, or by creating new ones? Whichever approach is taken, the new institutional type(s) should be incorporated into the basket of rules that apply to similar institutions offering similar services (e.g., financial consumer protection and treatment in bankruptcy). Incorporation in the existing framework makes regulatory harmonization more likely, and inconsistent or unequal treatment less likely. This applies both to the ini- tial regulatory reform as well as to future amendments.23 However, local factors will determine the advisability of this integrated approach versus creating a separate new framework. Policy makers may be reluctant to open up the banking law for amend- ment because they don’t want to trigger a review of other issues that have nothing to do with microfinance.24 22 This problem can be exacerbated by regulation that is inappropriate or overly burdensome to new models. 23 When changes are implemented via regulations (as opposed to laws), future adjustments are typically much easier to implement, given the fewer procedural obstacles to adopting, amending, or repealing regulations as compared with legislative acts. 24 The decision has at times been influenced by a donor’s technical assistance package that includes a new draft “microfinance law� (often based on another country’s law). This approach is seldom, if ever, appropri- ately tailored to the local context. Preliminary Issues 11 Regulatory arbitrage If a new window involves lighter or more favorable regulation, existing institutions and new market entrants may contort to qualify as MFIs. Such speculation among regulatory alternatives, or “regulatory arbitrage,� can leave some institutions under- regulated. Also, entrants may use a new microfinance window for businesses that are quite different from what policy makers had in mind when creating it. For instance, consumer lenders (who generally target salaried borrowers) have used a licensing form intended for microfinance (where lending usually focuses on borrowers without formal employment) to benefit from a higher usury rate applicable to microlending. Com- mercial banks that cannot meet increases in the minimum capital requirement or other prudential requirements for banks have, in some instances, been relicensed under a microfinance window, without having the motivation or knowledge necessary to serve low-income customers effectively. In at least one case, many of the newly licensed MFIs that had previously been underperforming banks failed, tainting the entire concept of microfinance in the country. 1c. Regulatory Definitions of “Microfinance� and “Microcredit� Key Point Regulatory definitions of “microfinance� and “microcredit� should be tightly framed to meet specific regulatory objectives and should not simply be drawn from general literature on microfinance. Earlier, we gave broad definitions of “microfinance� and “microcredit� as those terms are used throughout this Guide. However, these definitions usually will not be suitable ‘Microfinance’? for use in a given country’s regulation (see “1a. Terminology: What Is ­ What is ‘Financial Inclusion’?�). An appropriate regulatory definition must be based on a clear articulation of the country- and situation-specific objective(s) the regulation is meant to serve. For instance, if the purpose is adjusting prudential norms for depository MFIs, it might be appropriate to define “microcredit� in terms of a specific maximum loan amount. However, the same definition might not be appropriate for determining whether an NGO MFI serves a sufficient public benefit to deserve a profit tax exemption. See Box 1. 12 Microfinance Consensus Guidelines Box 1. Crafting a regulatory definition of “microcredit� There is no standard regulatory definition of “microcredit� that would be suitable for global use. However, some general practical cautions emerge from the experience of countries that have crafted their own definitions for a variety of regulatory objectives: 1. Use of funds. Identifying the primary purpose of particular loans (e.g., microen- terprise, housing, education, other) may help MFIs and their clients manage risk. However, the regulatory definition should not require that the loan be used to fund a microenterprise. First, this would interfere with the other valid reasons for which poor people borrow. Second, money is fungible, and numerous analyses of the actual use of microcredit show that a significant portion of the money is indeed not invested in a microenterprise, even if this is the declared purpose of the loan and even if such a purpose is required by regulation. Most microloan customers are in fact microentrepreneurs—in the sense they are pursuing self-led, small-scale earning activities—but this does not mean that they always use their loan proceeds for microenterprise purposes. Financial services, including microloans, can be cru- cially important for poor households not only to finance income-producing activity but also to allow payment of regular consumption expenses despite irregular and unreliable income streams. In addition, microloans and other microfinance services enable families to accumulate sums of cash that are large enough to deal with emer- gencies, sporadic opportunities, or major social obligations. 2. Maximum amount. Regulators who set a maximum microloan amount intro- duce two problems for the sector. Microlenders (i) will not be able to accommo- date successful clients who eventually want loans that exceed the limit and (ii) will have less opportunity to balance the more costly small loans with less costly larger loans. On the other hand, a high maximum amount dilutes the low-income targeting and increases the risk of regulatory arbitrage. A two-pronged approach may help to strike the right balance: (i) a maximum average outstanding loan balance for the entire microcredit portfolio and (ii) a higher maximum initial amount for any microloan (aggregating multiple loans to a single borrower for purposes of this calculation).a The limit on average loan size preserves overall targeting, while the higher limit on individual loan size allows some flexibility. 3. Defining the customer. Defining the target customer in regulation is often tempting, but can pose practical challenges. For example, a definition that refers to “poor� customers could exclude low-income unbanked persons who are both needy and Preliminary Issues 13 Box 1. Continued potentially profitable borrowers—at least if there is any expectation that the limit on client income will actually be enforced. In practice, demonstrating or enforcing compliance with such a limit could also be very difficult and expensive. Microcredit is sometimes defined in terms of lending to microentrepreneurs (people who earn their income from work outside the formal sector). Even if the policy maker intends a tight focus on this clientele, it may be more practical at times to introduce a degree of flexibility, for instance, by including the house- holds of microentrepreneurs or by requiring that microentrepreneurs constitute the majority of borrowers. In some cases, the target clientele of microcredit has been defined quite narrowly, but in practice supervisors have allowed lenders considerable flexibility in choosing clients. The letter of the law can be invoked to deal with lenders who claim to be doing microcredit but are actually serving an entirely different type of customer. ­ 4. Requirements with respect to collateral. To differentiate microcredit from con- ventional retail bank loans, the regulator might want to require that microloans be uncollateralized. Any such regulatory requirement should build in flexibility. For instance, microborrowers who own no collateral and start off with small, un- secured loans occasionally succeed in growing their enterprises and raising their income to the point that they acquire substantial assets that could be pledged as normal collateral. Moreover, some MFIs accept collateral to enhance repayment incentives, even when the value of the collateral would not be enough to make the lender whole in the event of default. 5. Defining “microcredit� in law or in regulations. If economic circumstances change or the original definition proves problematic in practice, a definition of “microcredit� that has been enshrined in law requires the full legislative process to change. Definitions in regulations typically can be adjusted more easily. In some countries, legal protocol requires a certain level of specificity in the law, but still leaves room to permit important details to be determined in regulations. a Note that two different measuring rods are involved here: over time the average outstanding loan bal- ance of a microcredit portfolio tends to be a little more than half of the average initial disbursed amount of the loans in the portfolio. Thus, if the average outstanding balance (loan portfolio divided by number of active loans) is capped at 200 and the initial disbursed amount of any individual loan is capped at 2000, this would mean that the maximum for any single loan would be roughly five times the allowable average of loans across the portfolio. See Rosenberg (1999). 14 Microfinance Consensus Guidelines A regulatory definition of “microcredit� may be challenging, but defining “micro­ savings� is a much simpler matter, mainly because such a definition is rarely needed. ­ Restricting the income level or other characteristics of depositors usually would be coun- terproductive. Typically, the purpose of an institution awarded a depository microfinance license is to serve the needs of poor and low-income customers. Even on that assumption, deposits captured from more affluent customers serve that purpose as long as they are used to fund loans to the lower income customers. In fact, a common funding pattern in deposit-taking MFIs is that most of their depositors have very small account balances, but most of their deposit funding comes from a minority—often a small minority—of accounts with larger balances. However, deposit size limits still may be relevant, depending on the regulatory ob- jective. For example, increasing numbers of countries have capped savings balances and transaction sizes for certain accounts when defining categories of lower risk that will justify relaxed anti-money laundering and combating the financing of terrorism (AML/ CFT) requirements. 1d. Prudential and Nonprudential Regulation: Objectives and Application Key Point Absent extraordinary circumstances, nondepository MFIs should not be subjected to prudential regulation and supervision.25 When a deposit-taking institution becomes insolvent or lacks adequate liquidity, it can’t repay its depositors, and—if it is a large institution—its failure could undermine public confidence enough so that the financial system suffers a run on deposits or other system-wide damage. Prudential regulation involves the government in overseeing the financial soundness of these institutions and taking action when problems arise. In contrast, “nonprudential� regulation—which is also referred to as “conduct of business� regulation—does not involve monitoring or assessing the financial health of the regulated institution.26 Nonprudential regulation of microfinance tends to focus on 25 Some traditions of financial sector regulation—notably those with historical ties to France—apply pruden- tial regulation to all types of lending institutions, regardless of their potential to jeopardize systemic stability. 26 The use of the term “conduct of business� regulation may be more common. However, this term encom- passes a narrower range of regulation relating to the conduct of the provider regarding consumers and other providers and would not typically include topics such as taxes and ownership limitations. Preliminary Issues 15 three main objectives: (i) protecting consumers of financial services, (ii) enabling a range of institutions that provide a mix of appropriate products and services, and (iii) provid- ing governments with information to carry out economic, financial, and criminal law enforcement policy. Some nonprudential regulation is subject to general enforcement, in- cluding civil and criminal prosecution and private rights of action. Other nonprudential regulation may be enforced by specific regulatory bodies.27 Sometimes a rule serves both prudential and nonprudential objectives. Effective con- sumer protection-related regulation of lending, for example, can lead to better asset qual- ity, which in turn contributes to an intermediary’s overall financial health—even though this is not the primary regulatory objective. There is wide recognition that compliance with, and enforcement of, prudential regu- lation is usually more complex, difficult, and expensive than nonprudential regulation, for both the regulator and the regulated institution. This can be particularly problematic in some developing countries where regulators and supervisors are already stretched to capacity with their mainstream banking and insurance sectors. Thus, an important general principle is to avoid using burdensome prudential regu- lation for nonprudential purposes. For instance, if the concern is only to keep persons with bad records from owning or controlling nondepository MFIs, the bank regulator does not have to take on the task of monitoring and protecting the financial soundness of such institutions. It would be sufficient to require disclosure of the individuals owning or controlling them, and to submit proposed individuals to a “fit and proper� screening. Why do governments impose prudential regulation on banks and other financial intermediaries, but not on nonfinancial businesses? Banks, much more than other busi- nesses, fund their operations with “other people’s money,� mainly deposits from the pub- lic. This creates incentives for managers to take inappropriate risks. More importantly, it exposes banks to deposit runs. Loss of confidence by depositors in one bank can spread quickly to depositors in other banks, threatening to destabilize the entire banking and financial system, with serious consequences for all other sectors of the economy. There is wide agreement that the cost of prudential regulation is justified when the stability of the financial system or the safety of depositors is at risk. Lending-only MFIs obviously pose no risk to their depositors (they have none) and are not subject to depositor runs. However, as noted, they should be subject to appropri- ate nonprudential regulation. 27 For example, permission to lend and reporting requirements may be enforced by the regulator of the particular type of institution; financial consumer protection may be regulated by the financial regulator or a consumer protection body; AML/CFT will typically be regulated by a country’s financial intelligence unit. 16 Microfinance Consensus Guidelines A systemic rationale for applying prudential regulation to m Box 2.  ­ icrolending-only institutions? Are microlenders that are funded by sources of capital other than public depos- its engaged in financial intermediation that needs to be prudentially regulated? (See Appendix C for a discussion of nondepository MFIs’ sources of funding and their potential regulatory ramifications.) In general, our answer is a firm no. Lending-only MFIs can’t create depositor runs, but some observers have pointed out that they can occasionally create contagious “borrower runs.� During the global financial crisis of 2008–2009, more than one country experienced such a contagion ef- fect in its microcredit market. If an MFI is failing, then it is common for borrowers to stop repaying their microloans, because a key repayment motivation is the borrower’s expectation of a new loan when the old one is repaid. If a major MFI fails, borrow- ers at other institutions may have doubts about their own lender’s solidity and the reliability of its implicit promise to give future loans. If this happens, other MFIs can be contaminated by the original MFI’s repayment problems. In at least one country, the government (via a state-owned institution) funded the acquisition of one of the country’s largest MFIs by another MFI to prevent such a contagion effect and to avoid losses by the commercial banks that had large loans outstanding with the MFIs.a Does the possibility of borrower runs justify prudential regulation of lending- only MFIs? Four factors warrant examination: systemic risk, cost of prudential regu- lation and supervision, who should bear the risk of loss, and capacity to supervise effectively: 1. Systemic consequences. The failure of one or more microlending institutions would typically not imperil a country’s whole banking and financial system. Even when microlenders serve huge numbers of customers, their assets will seldom account for a large percentage of the country’s financial assets.b A rare exception ­ might be found in cases where wholesale loans to MFIs are a major part of the assets of a country’s banks, although the more direct and appropriate route is to improve supervision of the banks’ lending practices (as opposed to imposing prudential regulation on lending-only MFIs). 2. Cost. Prudential supervision is costly for both the supervisor and for the supervised. And the experience has been that it costs substantially more to ­ ­ supervise MFI assets than an equivalent volume of normal bank assets. Absent subsidy, sustainable institutions pass such costs on to their clients. In the case of depository institutions, the costs of prudential regulation and supervision are ­ Preliminary Issues 17 Box 2 Continued likely to be less than the costs of bailing out the financial system. In contrast, the failure of microlending institutions will typically not require a systemic bailout. 3. Risk of loss. The failure of a microlending institution results in the loss of inves- tors’ and donors’ funds. Investors and donors can mitigate this risk by supervis- ing the investee institutions. Where no deposits are involved, there is no loss of the public’s money. Clients of the failed institution may lose access to services, temporarily or permanently. The same is true of many nonfinancial businesses; however, this risk is not viewed as a reason to impose prudential regulation on those businesses. 4. Capacity to supervise. Retail depositors are generally not in a position to evaluate the management and condition of the banks that hold their funds. In contrast, owners, lenders, and donors can and should supervise the microlending institu- tions they invest in. In the great majority of circumstances, these factors do not add up to a good case for prudential regulation of nondepository microlenders.c (At the same time, nonprudential regulation—especially consumer protection regulation—is completely appropriate for such institutions.) a See Chen, Rasmussen, and Reille (2010). b In rare cases, MFI assets loom large in a country’s financial system, but in those cases the MFIs are typically already prudentially supervised because they take deposits. c Some regulators, particularly in countries with links to the French regulatory tradition, disagree with this position. As part of their lending methodology, some MFIs require compulsory savings from their borrowers before loan disbursement and/or during the life of the loan. This require- ment tests the client’s ability and willingness to make regular payments and provides “cash collateral� that protects a portion of the loan. Should MFIs that take compulsory savings, but not voluntary savings, be treated as deposit-takers and subjected to pruden- tial regulation? There are differing views on this question. Usually, clients of MFIs that take compulsory savings are net borrowers: the compul- sory savings amount is typically a (small) fraction of the customer’s loan size so the clients owe the MFI more money than the MFI owes them. If the MFI fails, these customers can protect themselves by the simple expedient of not paying their loans. 18 Microfinance Consensus Guidelines On the other hand, there are times when the customer has little or no loan balance outstanding, usually at the end of the term of an amortizing loan or between loans. At those times, the compulsory savings can exceed the loan balance, and the customer is in a net at-risk position. This fairly small degree of risk has to be weighed against the costs of prudential supervision. Several countries have taken a middle path on this issue by requiring prudential licensing for any MFI that intermediates clients’ compulsory savings (i.e., lends them to other borrowers), but not for MFIs that keep the savings in an account with a licensed bank or invested in low-risk securities. Regulators should consider protecting customers’ prior claim on the funds in the event the MFI fails, for instance, by requiring the compul- sory deposits to be held in a trust fund or escrow account.28 1e. Regulate Institutions or Activities? To advance financial inclusion while promoting a level playing field, similar activities institution should be subject as much as possible to similar regulation, regardless of the ­ being regulated. The ability to regulate activities similarly will depend on both the specific regulatory issue being addressed as well as how the different institutions are regulated (i.e., under one or many laws) and supervised (i.e., by one or many regulators). There are various approaches. Under a “functional� approach, supervision is determined by activ- ity. A unified or “integrated� approach combines prudential and nonprudential regula- tion and supervision under one roof. The “twin peaks� approach uses separate structures for prudential and nonprudential issues. In many countries that use a unified or twin- peaks approach, nondepository institutions fall outside of the supervisory regime. Some prudential rules should depend on institutional type: for instance, different kinds of institutions may call for different rules about permitted activities or capital adequacy. In contrast, most nonprudential standards applicable to microfinance services would generally be appropriate regardless of institutional type providing them. In par- ticular, there is a strong argument that financial consumer protection and AML/CFT are better achieved through a single set of rules applicable to all providers of a given service. Sometimes, regulation by activity (instead of by institutional type) makes good sense as a matter of principle, but is not feasible as a matter of political economy. For example, it may not be easy to get regulation adopted that puts upstart market entrants on a level playing field with established market incumbents. 28 The regulator will want the right to verify that the funds are treated in this way. Such verification could be viewed as prudential supervision, but this does not mean that other prudential requirements should be introduced. Part II. Prudential Regulation of Deposit-Taking Microfinance As microfinance moves beyond microcredit, prudential regulation of depository micro- finance comes into play.29 The primary reasons for prudential regulation of depository institutions are (1) to protect the country’s financial system by preventing the failure of one institution from leading to the failure of others, and (2) to protect small depositors who are not well-positioned to monitor the institution’s financial soundness themselves.30 If prudential regulation does not focus closely enough on these two objectives, scarce supervisory resources can be wasted, institutions can be saddled with unnecessary com- pliance burdens, and development of the financial sector can be constrained. 2a. New Regulatory Windows for Depository Microfinance: Timing and the State of the Industry As previously discussed, many countries have created new regulatory windows for depo­ sitory microfinance, while many others are considering such a step. Where new windows are under consideration, policy makers should begin by determining whether and how ­ ervices the existing financial sector regulation hinders institutions from providing savings s for the poor or from raising other deposit funding to expand their microcredit services.31 If existing regulation does not present a barrier, or if the real binding constraint lies else- where, then a new window will not necessarily improve access. And even if the existing regulation does inhibit the development of depository institutions serving poor people, it may be preferable to amend (or provide exemptions to) the existing regulation instead of embarking on the complex task of creating a new window. The likelihood that a new window for depository microfinance will substantially expand financial services for the poor depends on whether there will be a critical mass of qualifying institutions, which in turn will depend on the type of window opened and the regulations governing it. Thus, it is important for regulators to determine (i) which 29 See, e.g., BCBS (2010). 30 Insurance companies and securities firms are also subject to prudential regulation. The regulation of micro- insurance when delivered by MFIs and other microfinance providers is addressed in Part VI; the regulation of securities firms is beyond the scope of this Guide. 31 The analysis should include identifying obstacles to new delivery channels (such as agent or electronic banking) and new types of providers (such as nonbank e-money issuers). 19 20 Microfinance Consensus Guidelines actors—existing NGOs, domestic entrepreneurs, or foreign investors—are likely to ­ ­ respond to a new window and (ii) whether those actors are likely to have the manage- ment skill to run safe depository microfinance. In addition, the regulator needs to consider what entity would supervise the new institutions, and whether there will be any problems with supervisory capacity. If the expectation is that over the medium term the new depository window will be used mainly by existing NGO microlenders that “transform� into depository institu- tions, then regulators should first assess the practical capacity of the leading institutions to undertake such a transformation. In particular, have these microlenders demonstrated the ability to run a loan business that is stable and profitable enough so that it can be safely funded with public deposits? In this context, the actual financial performance and prospects of existing MFIs is a crucial element that often gets too little attention in dis- cussions of regulatory reform. In making such an assessment, regulatory staff who are inexperienced with the special dynamics of microcredit will need expert assistance. On the other hand, policy makers may see the new window as a means of attract- ing investment in start-up depository MFIs, more than as a path for transforming NGO MFIs. Here again, they need to think practically about who the likely takers will be. If they are hoping for “greenfield� (i.e., start-up) operations based on foreign investment and technical expertise, are there any identifiable actors on the scene who are likely to be interested? If local startups are expected, there are several factors worth considering: • Has the profitability of microfinance, or at least microcredit, been demonstrated in the country yet? Otherwise it may be hard to interest entrepreneurs and investors. • Does the country have experienced microcredit managers who can teach new staff how to make and collect microloans? • How entrepreneurial is the country’s private sector? The regulatory window that would draw dozens (or hundreds) of entrants in one country might draw relatively little interest in another country with a less entrepreneurial culture. Several countries have built technically sound new regulatory windows for microfi- nance but have seen little response. Conversely, most of the successful new depository windows are in countries that already had a strong microcredit industry before the win- dow was put in place. It would overstate the point to claim that a new window will never spark the emergence of new service providers when little had been going on previously. In a few countries, a new prudential licensing window for small rural banks (not neces- sarily doing microfinance) resulted in many new institutions providing service to areas previously without access. However, supervision proved much more difficult than antici- pated. In each case, many of the new banks failed, and the bank regulator had to devote Prudential Regulation of Deposit-Taking Microfinance 21 significant resources to cleaning up the situation. Nevertheless, many of the new banks ­ did not fail, and these continue to provide ongoing rural services. 2b. Rationing Prudential Regulation and Minimum Capital Key Points • Minimum capital should, in principle at least, be set high enough to ensure that the institution can cover the infrastructure, management information system (MIS), and start-up losses to reach a viable scale. Minimum capital should also provide incentives for adequate performance and continued operation. • In creating a new window for depository microfinance, it is important to assess supervisory capacity and set the minimum capital requirement high enough to avoid overburdening the supervisor. • Where possible, it is usually preferable to set minimum capital through regulation rather than legislation, but the regulator needs to be clear in its communications with the market to avoid a perception that the regulator is unpredictable. Pri- mary among other advantages, setting the requirement through regulation makes it easier for supervisors who are new to microfinance to start with a manageable number of new licensees, reserving the option of reducing minimum capital and licensing more institutions as experience is gained. Supervisors have limited resources, so there is a trade-off between the number of new institutions licensed and the likely effectiveness of the supervision they receive. And when measured as a percentage of assets supervised, the cost of effective prudential ­ supervision is higher for microcredit portfolios than for normal bank loans (even though there may eventually be some economies of scale).32 Finally, overstretched supervisory capacity can result in reputational issues for the supervisor as well as potential losses to depositors. Thus there are strong arguments for rationing the number of licenses, ­ especially at an early stage. The most common tool for this rationing is the minimum requirement—that is, the minimum absolute amount that owners must invest as capital ­ equity in an institution seeking a license to accept deposits.33 The lower the minimum capital, the greater the likely number of entities that will have to be supervised. 32 One way to cover such additional costs is to charge the supervised institutions a supervision fee although this can affect access. 33 Qualitative licensing requirements also serve to limit new entrants. 22 Microfinance Consensus Guidelines The minimum capital requirement should be high enough to fund appropriate infra- structure and systems and to cover start-up losses.34 (In some countries, there are different capital requirements depending on geographic scope, with the lowest requirement for a local [e.g., district-based] institution, followed by regional, provincial, and national.) Microfinance advocates who see regulation primarily as promotion usually want very low minimum capital requirements, making it easier to obtain new licenses. Supervisors who will have to oversee the financial soundness of new deposit-taking institutions know there are limits on the number of institutions they can supervise effectively and therefore usually favor higher minimum capital requirements. In addition and more importantly, supervisors are justifiably cautious about lowering minimum capital norms solely to enable more potential entrants to satisfy them. In principle, allowing more licenses tends to foster competition and access to services. But this doesn’t necessarily mean that the right policy is to enable the formation of many small deposit takers. In most microfinance markets, the vast majority of the clientele is being reached by five to 10 large MFIs.35 In some of these markets, licensing of many very small market entrants may not produce enough new outreach to justify the additional supervisory burden. Obviously, the appropriate number of microfinance providers will vary depending on the size and diversity of a country’s market. When policy makers are opening an MFI depository window for the first time, there is considerable uncertainty about how much burden it will put on supervisory resources, and how long it will take supervisory staff to learn the distinct dynamics of microfinance. Given this uncertainty, some regulators make a reasonable decision to err on the side of conservatism (higher minimum capital) at first, permitting the requirements to be ad- justed later when the supervisors have more experience with the demand for licenses and the practicalities of supervising microfinance. Such flexibility is easier if minimum capital is specified in regulations rather than in the law. 2c. Adjusted Prudential Standards for Microfinance Some prudential norms developed for conventional banking don’t fit well with the risks and requirements of microfinance, which involves different products and services. Whether microfinance is being developed through specialized stand-alone depository 34 In practice, the minimum required capital is often based not on calculations of operational costs but instead on a reduction in the capital required for a conventional bank license or on the requirements set in other countries. 35 For example, according to MIX Market data (www.themix.org), the top five MFIs in India accounted for 58 per- cent of the clients served by the 85 MFIs reporting data in 2009. This holds up in other markets across different regions, e.g., Bolivia (65 percent), Bosnia and Herzegovina (66 percent), and Ghana (53 percent). In some countries with hundreds of MFIs, the overall amount of services provided would drop only a little if the smallest 50 percent or even 75 percent of them closed down. In India the smaller 50 percent of MFIs reporting to MIX in 2009 accounted for only 3.6 percent of total borrowers. This is similar to 2009 data in Bolivia (9.1 percent) and Ghana (15 percent). Prudential Regulation of Deposit-Taking Microfinance 23 MFIs, through financial cooperatives, or as special product lines within retail banks, the regulatory requirements discussed next will need review and possible revision.36 The issues discussed below include the most common requirements, but other rules may need adjustment in some countries. Many of the adjustments relate to distinctive features of microlending, reflecting the fact that microfinance differs from conventional banking more on the credit side than on the deposit side. Permitted activities Key Point Regulation—including any proposed new regulation that provides for depository microfinance—should clearly define the types of permissible activities that a pruden- tially regulated institution may engage in. Regulation may permit certain institutions to engage only in lending and deposit- taking (or initially only lending, with deposit-taking being permitted later subject to super- visory approval). Other institutions may be allowed to provide money transfer or foreign exchange services. Limitations on permitted activities will significantly affect prudential requirements, including in particular capital and liquidity rules. Regulation may also limit an institution’s scope of activities by defining and restricting the concept of “microcredit.� Managers of newly licensed MFIs may not have much experience with managing the full range of banking activities and risks (e.g., retail savings delivery and asset and liability management). Permission to engage in sophisticated activities usually should be based on management capacity and institutional experience. For example, depository microfinance providers may be well-equipped to serve as microinsurance agents, but are unlikely to be well- positioned to underwrite insurance risk. (As noted in “6b. MFIs and Microlending Banks in Microinsurance,� such activities are often governed by a separate regulatory framework.) Capital adequacy Key Point There are strong arguments (and recent experiences) that support the imposition of higher capital adequacy standards for specialized depository MFIs than for banks. 36 See BCBS (2010). 24 Microfinance Consensus Guidelines Regulators require banks to hold capital that is adequate when measured against the size ratio and riskiness of the bank’s assets and its off-balance-sheet exposure. The capital adequacy ­ (CAR)—the ratio of equity to risk-weighted assets—is a primary focus of bank supervision. A higher CAR means less risk to depositors and the financial system. But a higher CAR also means less funding from deposits, which lowers profits and makes the inter- mediary less attractive for investors. New MFIs in general take longer than commercial banks to leverage their equity and build their loan portfolios, so a high CAR may not hamper their operations much in their initial years. Over the longer term, however, a higher CAR can reduce poor people’s access to loans and other financial services. The regulator needs to balance safety and access when setting capital adequacy norms. There have been years of debate about whether specialized MFIs should have a tighter capital adequacy requirement (i.e., a higher ratio) than diversified commercial banks. Five reasons have been advanced for imposing a higher CAR on MFIs: • Particular features of a microloan portfolio. Well-managed MFIs typically have main- tained excellent repayment performance, with delinquency often much lower than in conventional retail banks. However, the repayment performance of a microloan portfo- lio can deteriorate much more quickly than a conventional retail bank portfolio. First, microloans are usually unsecured (or secured by assets that are insufficient to cover the loan plus collection costs). Second, the borrower’s main incentive to repay a microloan is usually the expectation of access to future loans.37 When a borrower sees that others are not paying back their loans, his own incentive to continue paying declines because the outbreak of delinquency makes it less likely that the MFI will be able to reward the borrower’s faithfulness with a follow-on loan. And if a borrower has no collateral at risk, he may feel foolish repaying his loan when others are not. Thus, outbreaks of delin- quency in an MFI can be contagious. (This risk is lower in the case of a commercial bank with a diversified portfolio of which microloans are a part.) The conventional wisdom has been that as a general rule, if a microlender’s annualized loan loss rate rises above 5 percent, or over 10 percent of its portfolio is late by more than 30 days, management must correct the situation quickly or the situation is likely to spin out of control.38 • High administrative costs. MFIs’ administrative costs are high relative to the size of the individual loans. As a consequence, to stay afloat, MFIs need to charge interest rates that are higher than that of the typical commercial bank loan.39 When loans are 37 This is true not only for individual microloans but also for group-based microloans. 38 This general rule is widespread in the industry, but it is based on anecdotal experience, not statistical analysis. 39 This is because the costs of making loans don’t vary in direct proportion to the amount lent; a portfolio containing a large number of small loans will be more costly than a portfolio of the same face value but containing a smaller number of larger loans. However, many MFIs could increase efficiency, thus reducing administrative and operational costs. Prudential Regulation of Deposit-Taking Microfinance 25 not being repaid, the MFI is not receiving the cash it needs to cover the costs associ- ated with those loans. Because an MFI’s costs are usually much higher than a com- mercial bank’s costs per unit lent, a given level of delinquency will decapitalize an MFI much more quickly than it would decapitalize a typical bank. • Portfolio diversification issues. MFI loan portfolios are not concentrated in a few large loans. But they are often confined to one or two loan products with substantial covariant risks. In addition, some MFIs operate in a narrow geographic area. • Maturity of management and systems. In some countries, MFIs do not have a very long track record and tend to have less experienced management and staff than banks. This is an especially important issue in the case of a transforming NGO MFI: the pre- transformation staff has little experience with the particular challenges of asset and liability management of a deposit-taking institution. • Limitations of supervisory tools. In many countries, the supervisory agency has little experience with judging and controlling microfinance risk. Additionally, a few impor- tant supervisory tools work less well for specialized MFIs (see “Supervisory Tools and Their Limitations�). In response, those who argue that capital adequacy requirements should be no higher for MFIs than for conventional banks have pointed to long years of stability and solid loan collection in licensed MFIs. They also stress the importance of a level playing field between MFIs and banks. However, recent experience has suggested that the long-term risks in microlending may be higher than they appeared to be at earlier stages. In particular, more micro- credit markets are reaching saturation as competitors hungry for market share expand very quickly. Recent outbreaks of serious over-indebtedness and collection problems in some maturing markets strengthen the argument for a higher capital adequacy re- quirement for those institutions with assets comprised mainly of uncollateralized mi- crocredit. Capital adequacy for financial cooperatives and their networks Financial cooperatives present a particular issue when defining capital for CAR purposes. All members of a cooperative have to invest a minimum amount of “share capital� in the institution. But unlike an equity investment in a bank, a member’s share capital may usually be withdrawn when the member leaves the cooperative. From the vantage of institutional safety, such capital is not very satisfactory: it can easily be withdrawn at pre- cisely the point where it would be most needed—when the cooperative gets into trouble. 26 Microfinance Consensus Guidelines Capital built up from retained earnings—sometimes called “institutional� capital—is not subject to this problem. One approach to the issue is to limit members’ rights to withdraw share capital if the cooperative’s capital adequacy falls to a given level. Another approach is to give coop- eratives a few years to build up a required level of institutional capital, after which time capital adequacy is based solely on retained earnings. A federation of cooperatives may be integrated enough so that the member cooperatives can be treated as a single conglomerate for supervisory purposes (e.g., the federation man- ages the assets and liabilities of the network, has considerable control over the actions of the member cooperatives, or creates a mechanism to reimburse depositors if a network member fails). In such circumstances, a consolidated CAR for the whole federation (without applying a ratio to each member) might arguably be sufficient. However, a consolidated ratio risks inequitable distribution of capital (e.g., where one cooperative has riskier and possibly higher earning assets than its capital would support). There is at least one country that takes a mid- dle path by imposing a consolidated CAR on the network and a lighter ratio on each member. Unsecured lending limits and loan-loss provisions Key Points • A microloan portfolio should not be limited to a specified percentage of the lender’s equity nor burdened with a high general provision requirement simply because the loans are not conventionally collateralized. • Absent special circumstances, performing microloans should have the same provi- ­ owever, sion requirement as other loan categories that are not particularly risky. H the provisioning schedule for delinquent microloans that are uncollateralized should be more aggressive than the provisioning schedule for secured bank loans. Regulations often limit unsecured lending by a bank to a specified percentage of the institution’s equity. This kind of rule is inappropriate for unsecured microcredit port- folios that use the common microlending methodology (see “1a. Terminology: What Is Microfinance�? What Is “Financial Inclusion�?). Such portfolios have generally per- “­ formed well without traditional collateral. Limiting microlending to some fraction of an institution’s capital (as is common for banks’ unsecured conventional loans) would make microlending by specialized MFIs impossible. It would also make microcredit less appealing for diversified banks. This does not imply that microloan portfolios never ­ Prudential Regulation of Deposit-Taking Microfinance 27 deteriorate, but that well-implemented microlending methodologies can generally miti- ­ gate credit risk effectively. Bank regulations often require high (sometimes 100 percent) loan-loss provisions for all unsecured loans (except loans to other licensed intermediaries) at the time they are made, even before they become delinquent. This is obviously impractical for microcredit. Even if the MFI later recovers the provision expense when the loan is collected, the accumulated charge for current loans would produce a massive under-representation of the MFI’s real net worth. To manage these two problems, a few regulators have treated guarantees, particularly group guarantees, as “collateral� for purposes of applying such regulations to microcredit. This can be a convenient and pragmatic solution to the problem, cosmetically at least. However, group guarantees are less effective than is often supposed. Many MFIs do group- based lending but do not ask for guarantees. Many others ask for guarantees but do not enforce them.40 Still others do not use groups at all. The most powerful source of security in microcredit tends not to be the MFI’s use of group guarantees, but rather the strength of its lending, tracking, and collection procedures, as well as the credibility of the institution’s promise that clients who repay will have access to the services they want in the future.41 The more straightforward solution, at least where it is supported by the historical collection experience among microlenders in the market, is to waive the equity limi- tation and/or the prohibitively high general provisioning requirement for unsecured loans in qualifying microcredit portfolios (i.e., those using the common microlending methodology—see “1a. Terminology: What Is ‘Microfinance’? What Is ‘Financial Inclu- sion’?�). General provision requirements usually should be the same for uncollateralized microloans—provided that they are originated through adequate processes that take into ­ account the market’s characteristics and risks—as for normal secured bank loans. However, once a microloan falls delinquent, doubt is cast on the borrower’s willingness and ability to pay. The implicit contract between borrower and MFI—that faithful repay- ment will lead to future services—may be breaking down. This is especially true for shorter loan terms with more frequent repayments. After 60 days of delinquency, a three-month unsecured microloan with weekly scheduled payments presents a higher likelihood of loss than does a two-year loan secured by real estate and payable monthly. Since microlending is typically not backed by adequate collateral, the provisioning schedule for delinquent microloans should be more aggressive than the schedule for delinquent secured bank loans. 40 There is evidence that in some cases the impact of group lending versus individual lending on repayment rates is negligible. Gine and Karlan (2010), in a long-term field experiment in the Philippines, found that switching from group to individual lending did not impact repayment rates. Similarly, results of laboratory- based testing of this hypothesis found that, absent different peer monitoring systems, “performance is mostly similar across group and individual lending schemes� (Cason, Gangadharan, and Maitra 2008). 41 See Chen, Rasmussen, and Reille (2010). 28 Microfinance Consensus Guidelines Governance Key Point Boards of deposit-taking MFIs should be independent of management and should include members with experience in finance and banking, as well as members who understand the clients well. As stated in the Basel Core Principles, “sound corporate governance underpins ­ effective risk management and public confidence in individual banks and the banking sys- tem.�42 This emphasis is underscored by the proposed addition of a new core principle on corporate governance. (See “2d. Transformation of NGO MFIs into Licensed Interme- diaries� for a discussion of governance and ownership issues when an NGO transforms into a commercial, shareholder-owned entity.) Management should receive oversight from an independent board. This is especially important in microfinance, where NGOs have been historically dominant: NGOs have no owners, which sometimes leaves management without effective board oversight. ­ Additionally, a deposit-taking MFI should have a minimum number of board members and senior managers who have experience in finance or banking. It is also useful to have someone on the board who understands the clients and their use of financial services. Good corporate governance is of particular importance to the double-bottom line of many MFIs. Issues regarding how to ensure adherence to mission arise in various con- texts, including pretransformation (i.e., selecting owners who share a commitment to the mission) and exit (i.e., selling to individuals and institutions who will continue to pursue the mission). Liquidity and foreign exchange risk Key Points • Specialized MFIs may need higher, rather than lower, liquidity requirements. • MFIs should not borrow or transact in foreign currency without having the capacity to assess and manage currency risk. ­ 42 See BCBS (2012). Prudential Regulation of Deposit-Taking Microfinance 29 In some countries, regulators have imposed less stringent liquidity requirements on depository MFIs than on other depository institutions. This may make it easier for MFIs to operate, but it is not clear how lower liquidity requirements are justified from a safety point of view. Indeed, there is an argument that MFIs may need more, not less, liquidity than banks. Faced with liquidity problems, banks can often stop lending for a while to conserve cash. But MFIs cannot stop lending without undermining borrowers’ motivation to repay their outstanding loans (see “3a. Supervisory Tools, Enforcement Mechanisms, and Limitations Regarding MFIs�). And MFIs may not have access to emergency liquidity from the central bank or to the market sources of liquidity that banks rely on.43 Increasingly, MFIs are borrowing from foreign lenders. Even though much of this debt is denominated in foreign currency, many MFI managers are not familiar with foreign exchange risk and the tools for managing it.44 There should be limits on an MFI’s net open position in each currency in relation to the institution’s capital or earn- ings (BCBS 2010, p. 21). In addition, an institution that borrows or otherwise deals in foreign currency should be subject to relevant fit-and-proper criteria for its senior management. And in general, an MFI’s risk management policies and processes should ensure that it does not take on risks—such as foreign exchange risks—that it cannot manage.45 Even if an MFI’s management has the skills to assess and, in theory, to manage foreign exchange risk, there are no opportunities to hedge in many developing coun- tries. In those circumstances, the MFIs often make their microloans in foreign currency, or index their microloans to the foreign exchange rate.46 In both cases, the borrow- ers assume the risk, which may not be a satisfactory solution from a client welfare ­perspective. 43 Special liquidity rules that might be appropriate for a specialized MFI with a portfolio consisting largely of microloans may be unnecessary for a universal bank if microloans make up a small part of its portfolio. 44 See Abrams and Prieur (2011), finding that MFIs are overexposed to foreign exchange risk. See also Apgar and Reille (2010), regarding new projects focused on foreign exchange hedging for MFIs. 45 An applicant for a deposit-taking license should be required to submit proposed risk management policies and procedures as part of the licensing process. Such policies and procedures should demonstrate, among other things, how the institution will measure, monitor, and control currency mismatch between its portfolio of microloans (typically in local currency) and any foreign currency funding. The supervisor must be able to evaluate whether an MFI is complying with its policies. 46 Some countries prohibit indexing because microfinance clients are likely less well-equipped to deal with the consequences of a drop in the value of the local currency. If indexing is permitted, the MFI should be required to explain to its clients the risks that they are undertaking. See “4c. Consumer Protection� for a discussion of disclosure requirements. 30 Microfinance Consensus Guidelines Loan documentation Key Point Given the size of microloans and the nature of the borrowers, loan documentation requirements need to be lighter for microcredit than for conventional bank lending. With a typical microloan, there is no collateral appraisal or registration, no formal financial statements for the borrower’s business, and no evidence that the business is for- mally registered or in compliance with tax obligations. Such requirements must simply be waived for microloans. Typically, a microloan file should include the following: • The loan application • A copy of the customer’s ID (or acceptable substitute documentation) • The loan appraisal (at least for individual loans, where the appraisal should usually include an analysis of household cash flow) • Information regarding the customer’s previous repayment performance with the MFI (if applicable) • A credit report, if the MFI participates in a credit reporting service47 • The loan approval by the relevant committee or manager48 • The executed note • An amortization schedule, at least in cases where this schedule is not integrated into an automated loan management system However, when an MFI makes repeated short-term (for instance, three-month) loans to the same customer, it should not be required to repeat the cash-flow analysis and credit report for every single loan. Frequency and timing of updates to the credit analysis of a borrower should be based on written policies, not ad hoc determinations. Restrictions on co-signers as borrowers Key Point Prohibitions against co-signers as borrowers may need to be relaxed. 47 In many markets there may be regulatory and practical barriers to MFIs participating in credit reporting, or limiting the usefulness of the reports of existing credit reporting providers. See Lyman et al. (2011). See “4d. Credit Reporting Systems.� 48 It is usually not feasible to require microloan applications to be approved above the branch level. Prudential Regulation of Deposit-Taking Microfinance 31 Regulations sometimes prohibit a bank from lending to a person who has co-signed or otherwise guaranteed a loan from that same bank. This creates problems for institutions using group-lending mechanisms in which group members are liable for each others’ loans. Branching requirements Key Point Branching requirements should be re-examined, but not necessarily eliminated, for microfinance. Banks’ hours of business and location of branches are often strictly regulated in ways that could make it impossible to serve microfinance clients profitably—particularly those living in more remote and sparsely populated areas. For instance, client convenience might require operations outside normal business hours or the operation of a mobile branch, or cost considerations might require that staff rotate among branches that are open only one or two days a week. Other nonprudential requirements related to branches—security requirements (e.g., armed guards, vaults) or infrastructure rules—could also make it too costly for MFIs to open branches in poor, remote, or sparsely populated areas. Clients’ need for access to financial services has to be balanced against the security risks inher- ent in holding cash. To promote financial inclusion, some countries have revised branch requirements to accommodate different types of service points, including bank use of agents—see Part V, “Regulating the Use of Branchless Banking to Serve the Poor.� Reporting Key Point The content and frequency of reports should enable supervisors to conduct the anal- yses needed for effective supervision of a depository MFI. However, regulation must also consider the circumstances of its supervised institutions, which may not be able to comply with some requirements applicable to banks. Reporting to a supervisor can add substantially to the administrative costs of an intermediary, especially one that specializes in very small transactions. In addition, some requirements may not be feasible—for instance, transportation and communication con- ­eporting ditions can sometimes make daily reporting virtually impossible. Typically, r 32 Microfinance Consensus Guidelines requirements are simpler for depository MFIs and microfinance programs than for con- ventional retail banking operations. However, they should include periodic financial statements, reports of external and internal audits, and information on portfolio quality, leverage, prudential ratios, operating costs, funding structure, liquidity position, foreign exchange exposures, and interest rate repricing gaps.49 Reserves against deposits Many countries require banks to maintain reserves (held as cash in the bank’s vault or by the central bank) equal to a percentage of deposits or certain types of deposits (time deposits may be excluded. Some countries exempt institutions with total reservable deposit liabilities below a specified amount. These reserves may be a useful tool of monetary policy, but they increase the cost of deposit-raised capital. Given the below-market rate often paid on these reserves (if there is any return at all), such a requirement can squeeze out small depositors by raising the minimum deposit size that banks or MFIs can handle profitably. This latter drawback should be factored into decisions about reserve requirements. Insider lending Key Point With respect to MFIs that are not member-owned and are receiving favorable regu- latory treatment because of their focus on poor clients, it is hard to see a reason for allowing any insider lending except perhaps small welfare loans to employees. Regulators often limit the amount a bank can lend to insiders (e.g., board and management) and other related parties, because such loans can pose conflicts of interest: ­ a generous loan that is good for the inside borrower may not be good for the bank as a whole. In the case of NGO MFIs and other MFIs getting regulatory advantages because they are serving poor people, it is unclear why any insider lending should be allowed, except for perhaps small welfare loans to employees. Member-owned financial coopera- tives may need an exception: their members might be unwilling to assume management or board roles if that meant giving up their access to loans. Notwithstanding the social orientation of MFIs, fraudulent insider lending is not unknown. Where insider lending is allowed, there should be conflict-of-interest rules ­ 49 See BCBS (2010). Prudential Regulation of Deposit-Taking Microfinance 33 (e.g., requiring inside borrowers to recuse themselves from any decisions about their loans). Such rules ought to extend to financial cooperatives, many of which have also had problems with abuse of insider lending. Loans that help insiders purchase shares in an MFI raise both conflict of interest and capitalization issues. Some countries prohibit such lending, or allow it only if the capital/ asset ratio is above a specified percentage. To whom should special prudential standards apply? Key Point When creating new regimes for depository microfinance, regulators should take care enabled to that full-service banks and other financial institutions (not just MFIs) are ­ provide microfinance services. It is worth reiterating that several of the adjustments mentioned have to do with particular financial products and operational methodologies, not particular institutional types. These adjustments should apply not only to specialized MFIs, but also to micro- finance operations in financial cooperatives, commercial banks, and other prudentially licensed providers (BCBS 2010). If a conventional retail bank decides to offer microfinance products, or to partner with an MFI to offer those products, it should have a clear regulatory path to do so. Regulators and supervisors should encourage such developments. When microcredit is a small part of a diversified retail bank portfolio, the risk and cost of supervising the microfinance activity are lower. Moreover, a level playing field in terms of the prudential standards helps to stimulate competition. Box 3. Basel core principles and depository microfinance In 1997, BCBS, in cooperation with supervisors from member and nonmember coun- tries and other international SSBs, identified 25 Core Principles for Effective Bank- ing Supervision that set forth the de facto standard in regulation and supervision of banks. The principles were revised in 2006 to reflect important changes in banking regulation worldwide, and again in 2012 to address post-crisis lessons. The current proposed revisions highlight the principle of proportionality. Continued 34 Microfinance Consensus Guidelines Box 3 Basel core principles and depository microfinance Continued The 2010 BCBS guidance paper on applying the 25 principles to microfinance activities in depository institutions notes the following: • Licensing requirements should be tailored to the size and nature of the particular institutions’ activities and the systemic risk posed by these institutions. In connec- tion with licensing, the types of permissible microfinance activities (including mi- crocredit, in particular, to distinguish it from other loan types) should be clearly defined in the regulations. • Differences between microfinance and commercial banking should be understood and considered when assessing risk management processes and techniques. Spe- cifically regarding MFIs, the loan portfolio is their primary asset so supervisors should focus on credit risk in particular and should have specialized knowledge of the labor-intensive microlending methodology. • Provisioning and reserves for microloans should be tailored rather than grouped with other loan categories. • Liquidity requirements should reflect MFIs’ high rate of growth and fewer backup li- quidity sources as well as the particular behavior of microfinance assets and liabilities. • Assessment of operational risks as well as internal controls and audit procedures should consider, if applicable, the decentralized nature of the microlending meth- odology as well as the use of technology and agents (or other contractual and outsourcing arrangements). 2d. Transformation of NGO MFIs into Licensed Intermediaries50 Key Point To facilitate transformation of NGO MFIs into for-profit companies licensed to ­ accept retail deposits, regulators may want to consider temporary or permanent adjustment of certain prudential requirements. 50 The nonprudential regulatory issues that arise when NGO MFIs transform (including restrictions on NGO ownership of for-profit companies) are discussed in “4f. NGO Transformations into For-Profit Companies.� Some of the prudential issues discussed in this section also apply when a for-profit lender transforms into a licensed intermediary. Prudential Regulation of Deposit-Taking Microfinance 35 MFIs often start up as nonprofit NGOs, but later want to take deposits to fund their growth and to provide savings services to their customers.51 Regulators concerned about financial inclusion usually view this kind of evolution favorably (although there may be nonprudential concerns regarding the treatment of the NGO’s assets, particularly assets derived from subsidized funding, given that typical NGO transformations are in essence privatizations).52 There are several ways to transform an NGO MFI into a bank or other for-profit de- pository institution.53 Most commonly, the NGO transfers all or part of its loan portfolio and other assets, liabilities, and employees to a new or previously existing company.54 In return, the NGO receives an equity interest in the company or payment in the form of cash or debt. Transformations often face regulatory obstacles, and regulators may want to consider relaxing some of these. Ownership suitability and diversification requirements Often, an NGO undergoing transformation will want to have a significant ownership interest in the new company, but owner suitability and diversification requirements can pose serious problems. For instance, regulatory approval for significant owners of the new licensed company may depend on their ability to respond to a capital call. However, many NGOs would not be able to raise additional cash in a hurry if the new MFI gets in trouble. Some regulators have waived this requirement (at least temporarily) to enable the transaction to proceed; in at least a few cases, there has been a permanent waiver in recognition of the financial and institutional stature of the particular NGO. Prudential regulations often specify a minimum number of owners for depository institutions, as well as a maximum percentage interest for any owner or group of related 51 Typically, an NGO (whether an association or foundation or other type) is not permitted to engage in deposit- taking. Other reasons for transforming an NGO into a company with owners are discussed in Lauer (2008). The transformation of an NGO into a cooperative raises issues that are not dealt with in this Guide or in Lauer (2008). 52 When assets of a not-for-profit MFI are sold to private parties, either the banking regulator or the regulator responsible for not-for-profit associations should take reasonable steps to ensure that the price paid reflects fair market value. In many cases, however, it will be difficult to make precise valuations (O’Donohue et al. 2009). These issues can be particularly sensitive when NGO managers or owners become shareholders of a new for-profit MFI. If they receive shares (or other benefits) at below-market prices, ethical and legal ques- tions arise. See Rhyne, Leiberman, Busch, and Dolan (2010). 53 In most of these cases the term “transformation� is a misnomer. When an NGO MFI “transforms,� it usually does not change from one type of institution into another. Rather, it transfers its business to another institution. In a handful of countries, the law does in fact allow a nonprofit (usually a company limited by guaranty) to re-register as a company with owners, thus undergoing a true transformation. 54 In the absence of a clear path for transformation, NGOs in some countries have used more convoluted transactions to transform. See Lauer (2008). 36 Microfinance Consensus Guidelines owners. But a maximum ownership interest of 20 percent (for instance) would oblige the transforming NGO to find at least four additional owners. When the NGO is forced to find additional investors, it may be in a poor bargaining position when it comes to getting a fair piece of the new company in return for what it is contributing. Shareholder diversity requirements have also been waived (in some cases, temporarily) for NGO MFI transfor- mations. Whether such action should be taken will depend on the particulars of the situa- tion, including the objectives of the ownership requirements (and whether such objectives can be satisfied by other means) and the financials of the NGO and the other owners. Ownership limits can also pose an issue for an NGO that wants the social mission of the microfinance operation to be maintained. This can be harder to ensure when regula- tions force the NGO out of a controlling position. The NGO may not be able to recruit investors who share the same social vision. Continuation of the NGO’s control (or strong influence) over the business may or may not be desirable, depending on the local situation and government policy.55 Board and management qualification requirements The “fit and proper� requirements for board members and senior managers of licensed financial institutions may include professional experience in finance.56 If the regulation requires managers to have experience in a bank or other depository institution, it can pose a problem in the common case of managers whose work has been limited to the credit-only NGO that is transforming. The management team of a depository institution should include depository experience, but a temporary waiver of this requirement until the new institution starts taking deposits may smooth the transition. Loan portfolio as part of initial minimum capital Current rules may stand in the way of an NGO that has little else beyond its loan portfo- lio to satisfy the minimum capital requirement for the new depository institution. Some countries prohibit the exchange of shares for loan portfolio to satisfy the initial capital ­ erforming requirement; in other countries, the prohibition is absolute. Elsewhere only p 55 At least one country has required that the transforming NGO remain a majority owner of the for-profit company to which it has transferred its portfolio. The stated objective of the regulation was to protect against asset stripping after the transaction is completed. 56 Fit-and-proper rules for licensed deposit-taking institutions are typically based on both professional skill and moral character. Lending-only MFIs do not require prudential regulation, so the fit-and-proper test for board and management of such MFIs need not include the same professional experience as the test for deposi- tory institutions. Prudential Regulation of Deposit-Taking Microfinance 37 contributed loans can be transferred. Even if there is no limit on how much capital may be ­ in kind (whether a loan portfolio or other asset), the noncash contribution should be valued by an independent expert. ­ One approach to converting old portfolio into safe capital is to offer a provisional bank license that permits the new institution to engage in lending (and perhaps certain other limited activities) only, but not deposit-taking. During the provisional period, the NGO contributes the cash proceeds collected on its existing loans as share capital in the new company. The customers who repay the NGO can then receive their follow-on loans from the new licensed company. Once the contributed cash satisfies the minimum capital requirement, the supervisor issues an unrestricted license that allows deposit-taking. This approach effectively allows the NGO to convert its old loan portfolio into capital in the new company. At the same time, it avoids burdening the new company with collection risk on the old portfolio, so there is no need for any valuation of that portfolio. 2e. Deposit Insurance Key Point If a country requires commercial banks to participate in a deposit insurance scheme, then it may wish to consider imposing the same requirement on prudentially super- vised deposit-taking MFIs as well (including at least larger financial cooperatives). Deposit insurance is one of several mechanisms that are used, together with bank regulation and supervision, to create a sound banking system, reduce the likelihood of systemic crises, and limit their costs when they do occur.57 In countries with strong institu- supervisor,58 tional development, including effective bank regulation and an independent ­ well-designed deposit insurance can contribute to financial system stability, protect failures. the large majority of depositors, and assist in the orderly resolution of bank ­ Countries have increasingly been introducing explicit deposit insurance, not only for ­ ­ financial system stability but also to limit the use of implicit (and open-ended) govern- ment guarantees (www.iadi.org/deposit.html). An explicit or formal deposit insurance scheme has two advantages over implicit government guarantees: increased transparency about obligations to depositors and a limitation (at least in theory) of the government’s financial obligations in the event of bank failure. 57 But see Caprio, Demirgüç-Kunt, and Kane (2008, p. 35), which states that “financial crises have become more frequent and more expensive (in terms of losses per dollar of deposits) as safety nets have expanded.� 58 IADI includes as a precondition to deposit insurance that there be strong and independent supervision. 38 Microfinance Consensus Guidelines In most countries, all prudentially regulated commercial banks have to partici- pate in the deposit insurance scheme. In a limited but increasing number of countries, financial cooperatives and deposit-taking MFIs are also invited or required to partici- pate, provided they are subject to effective prudential regulation and supervision.59 However, before setting up such a system, it is essential to assess the funding of such a scheme. Who will fund it? How will the premiums be determined and assessed? What impact will these premiums have on the interest rates and fees charged to customers? And how will such premiums affect financial access (CGAP 2011)? A unitary insur- ance system makes depositor protection more consistent and levels the playing field among providers. Where nonbank deposit-takers are regulated and supervised differ- ently from banks, some countries have subjected them to separate deposit insurance schemes.60 Deposit insurance can create “moral hazard�—that is, the tendency of an insured institution to take on more risk than it would if it were not insured. Moral hazard may be mitigated by deposit insurance design features, including limits on coverage and types of accounts insured, differential or risk-based premiums, prompt corrective ­ action policies, and the insurer’s demonstrated willingness to take legal action against ­ irectors for malfeasance. Moral hazard can also be mitigated by strong managers and d institutional development (e.g., effective regulation and supervision) and market disci- pline (e.g., monitoring of banks by depositors and by other bank stakeholders), which are in turn positively affected by sound design features (e.g., caps on account coverage may encourage larger depositors to monitor their banks) (Demirgüç-Kunt, Kane, and Laeven (2007). 59 This may be a challenge in countries with lower levels of basic supervisory capacity. See discussion of deposit insurance in the CGFI white paper (2011). 60 In some countries, financial cooperatives have formed a stabilization fund—typically when there is no ­deposit insurance. All participants in the stabilization fund pay a fee based on a percentage of their ­ deposits; in the event of a liquidity or solvency problem, a participating cooperative can draw on the fund. When the fund is not managed by an independent body, problems can arise in determining which institutions to stabilize. ­ Part III. Prudential Supervision Issues in Deposit-Taking ­Microfinance Decades of experience around the world with many depository institutions that are not conventional retail banks—including financial cooperatives, mutual societies, rural banks, village banks, and depository MFIs—have demonstrated that there is a powerful temptation to underestimate the challenge and cost of supervising such institutions in a way that will keep the strong majority of them reasonably safe and stable. When stakeholders start discussing legal frameworks for depository microfinance country, it is relatively easy to craft regulations, but harder to do concrete planning in a ­ effective supervision. In relation to the assets being supervised, specialized MFIs for cost-­ are often more expensive to supervise than full-service banks although a proportionate approach should reduce costs of supervision. The result is that supervision sometimes gets ­ incorrect) too little attention in the process of regulatory reform—perhaps on the (often ­ assumption that supervisory challenges created by the new regulation can be addressed later, by pumping extra money and technical assistance into the supervisory agency for a while. The result may be regulation that is not enforced (which could be worse than no regulation at all) or very costly. Early and realistic attention to supervision issues is crucial because the govern- ­etail ment takes on a fiduciary responsibility when it grants licenses to mobilize r deposits. Depositors should be able to assume that the issuance of a license to a financial intermediary means that the government will effectively supervise the ­intermediary.61 Before deciding to a ­ uthorize the issuance of licenses to depository financial institutions, a government needs to be clear about its ability to fulfill this undertaking. 61 By “effective� supervision, we mean supervision that is sufficient to flag most problems before it is too late to fix them. No supervisor can, or should, prevent all failures. 39 40 Microfinance Consensus Guidelines 3a. Supervisory Tools, Enforcement Mechanisms, and Limitations Regarding MFIs Key Points • Assessing microcredit risk requires specialized examiner skills and techniques that differ substantially from the ones that supervisors use for conventional retail bank portfolios. ­ ergers— • Some supervisory tools—such as capital calls and forced asset sales and m work less well for MFIs than for conventional retail banks, and some, such as stop-lending orders, may actually be counter-productive. Microcredit portfolio supervision Conventional bank audit and inspection procedures do not provide enough assurance about microcredit portfolio quality, which is the predominant locus of risk in MFIs. Loan-file documentation is a weak indicator of microcredit risk. Sending out confir- mation letters to verify account balances is usually impractical, especially where client literacy is low. Inspectors can usually test the bulk of a commercial bank’s loan assets ­ by examining a relatively small number of its largest loans, with limited sampling of the remainder. In contrast, testing a microloan portfolio calls for more extensive sampling among thousands of tiny loans. In supervising microlending, the main reliance is on close review of the institution’s systems and policies for lending, collection, credit risk management, and internal con- ­ nowledge trols, as well as the actual performance of its portfolio. Such analysis requires k of microfinance methods and operations, along with experienced interpretation and judgment. Specialized techniques and expertise are essential.62 Supervisors new to micro- finance sometimes underestimate the need for additional examiner training. Stop-lending orders When a bank is in trouble, supervisors sometimes issue a stop-lending order to keep the bank from taking on further credit risk until its problems have been sorted out. A commercial bank’s loans are usually collateralized, and most of the bank’s customers do not necessarily expect an automatic follow-on loan when they pay off their exist- ing loan. Therefore, a commercial bank may be able to stop new lending for a period 62 See, e.g., Christen and Flaming (2009); MicroSave (2011); and Isern, Abrams, and Brown (2008). ­ icrofinance Prudential Supervision Issues in Deposit-Taking M 41 destroying its ability to collect its existing loans. The same is not true of most without ­ MFIs, where immediate follow-on loans are the norm. As discussed earlier, when an MFI stops issuing repeat loans for very long, customers lose their primary incentive to repay—i.e., their confidence that they will have timely access to future loans when they need them. When an MFI stops lending, many borrowers stop repaying, making a stop- lending order to an MFI counterproductive, at least if there is any hope of salvaging its uncollateralized portfolio. An alternative approach would be prohibiting lending to new customers. Capital calls When an MFI gets in trouble and the supervisor issues a capital call, NGO owners may not have enough liquid capital available to respond. Development-oriented investors often have plenty of money, but may have lengthy internal procedures for approving ­ and disbursing, and a low appetite for taking additional risk in a crisis. Thus, when problem surfaces in a supervised MFI, some of its shareholders may not be able to a­ respond promptly to a capital call (although this is becoming less of a problem with the ­ increasing commercialization of microfinance). Asset sales or mergers A typical MFI’s close relationship with its clients may mean that loan assets have little value collateralized in the hands of a different institution. When a commercial bank fails, its ­ loan assets can often be saved by transferring them to a solid bank, or by effecting the merger or acquisition of the bank in trouble. This will seldom work with microloans.63 The fact that some supervisory tools do not work very well for microfinance certainly does not mean that MFIs cannot be effectively supervised. However, regulators should weigh this fact when they decide how many new licenses to issue at first, how much dem- onstrated past performance to require for transforming MFIs, and how conservative to be in setting prudential standards such as capital adequacy. 3b. Costs of Supervision In relation to the assets being supervised, specialized MFIs can—especially in the initial stages—be considerably more expensive to supervise than full-service banks, both for 63 See Rozas (2009). 42 Microfinance Consensus Guidelines the supervisor and the MFI.64 Governments and donors who promote the development of depository microfinance should consider providing transitional subsidies for supervis- ing the resulting institutions—particularly in the early stages when the supervisory staff is learning about microfinance and there are a small number of institutions to share the costs of supervision. (It would be preferable not to pass these start-up costs on to the clients.) Donors can also play an important role in funding supervisor training. However, over the longer term, the government must decide whether it will subsidize these costs or have MFIs pass the costs on to their customers. 3c. Where Should the Microfinance Supervisory Function Be Located? At first blush, placing supervision of deposit-taking MFIs in the same authority that ­supervises commercial banks would seem a natural choice. But banking supervisors do not resources, always think this responsibility is the best allocation of their scarce supervisory ­ especially in cases where they are not confident that they have the staff they need even to supervise the big commercial banks adequately. So alternative models, mainly “delegated supervision� and “self-supervision,� are sometimes proposed. The ultimate decision will depend on the specific country’s situation. Within the existing supervisory authority? Key Point In most cases, the best supervisor for depository microfinance will be the authority responsible for commercial banks. Having the bank supervisor supervise depository MFIs takes advantage of exist- ing skills (even though additional specialized skills need to be developed), lowers the ­ incentive for regulatory arbitrage, and may better preserve the political independence of supervision. In addition, there is the potential for improved communication and shar- ing of information with the supervisors of banks. Separate supervisory bodies have not always had the professionalism, the skills, the teeth, and the independence to do a good job. This has proved especially true in cases where supervision of financial cooperatives is left with the same agency that supervises all cooperatives. 64 To some extent this may be due to the typically low number of institutions being supervised. ­ icrofinance Prudential Supervision Issues in Deposit-Taking M 43 Whether there is a need to create a separate microfinance supervision department within separate depart- the banking authority varies from country to country. With or without a ­ ­ orking ment, supervisory staff often view jobs in nonbank supervision as less attractive than w on bank supervision. Work with nonbank institutions usually offers less prestige, less contact with top officials, and fewer chances for lucrative job offers from commercial banks. If the examiners, it needs to take these supervisor wants a strong and stable cadre of microfinance ­ ­ upervised by dedicated staff or, if there are dynamics into account. Ideally, MFIs would be s not enough MFIs to justify this, then general supervisory staff with specialized training. If a single regulatory scheme is going to embrace both lending-only MFIs and pru- dentially licensed deposit takers, the question of whether the banking supervisor should oversee both becomes more complicated. There are arguments for having the one agency supervise both. There is a (modest) degree of overlap in the functions—e.g., both micro- consumer protection lenders and deposit-taking MFIs will be subject to most of the same ­ regulation. The banking supervisor may have more political independence, and putting lending-only MFIs under the banking authority might smooth their transformation into licensed deposit-takers. Finally, there may be no other suitable agency.65 However, consolidating prudential and nonprudential regulation of microfinance appropriate within the same banking authority can create a risk of confusion about the ­ ­ treatment of nondepository institutions, possibly leading to over-regulation of them. As argued earlier, nondeposit-taking microlenders seldom pose major systemic risk and never put depositors at risk, so they don’t need prudential regulation and supervision, which is complex, costly, and intrusive. Delegated (or auxiliary) supervision Key Point It is not easy to delegate supervision effectively. The principal supervisor should closely monitor the work of the delegated entity. Delegated or “auxiliary� supervision refers to an arrangement where the government financial supervisor delegates direct supervision to an outside body, while monitoring and controlling that body’s work.66 Delegated supervision is perhaps most common for 65 This may be the case in particular in small countries, where it would not be cost-effective to set up a new agency. 66 In some countries the constitution prohibits outside delegation of supervisory powers and responsibilities. In those cases, it may be possible to frame some degree of de facto delegation as “contracting out support functions.� 44 Microfinance Consensus Guidelines financial cooperatives (see “Supervision of financial cooperatives�) but some countries have considered it for other nonbank financial institutions as well. Delegated responsibilities may include data collection and processing, off-site moni- toring, on-site inspections, recommendations for action, corrective enforcement, cease- and-desist orders, and (rarely) intervention and liquidation. For prudential supervision, agencies other than the bank supervisor may lack adequate resources and expertise, ­ operational independence, and remedial powers to fulfill their responsibilities. To date, there are not many successful examples of delegated supervision. Delegated supervision seems to have worked, for a time at least, when the principal supervisor closely monitored and had effective control of the quality of the delegated supervisor’s work. ­ upervisor. This requires a significant investment of time and resources of the delegating s Where this model is being considered, it is important to have clear answers to the following questions: 1. Who will pay the costs of the delegated supervision and the government supervisor’s oversight of it? 2. Will the delegated supervision cost less than direct supervision would? 3. If the delegated supervisor proves unreliable and its delegated authority must be with- drawn, is there a realistic fallback option available to the government supervisor? 4. What will be the delegated authority’s duties and powers? Will it be authorized to impose penalties, investigate breaches, conduct hearings, etc.? 5. When a supervised institution fails, which body will have the responsibility and ability to clean up the situation by intervention, liquidation, or merger? Self-regulation and supervision Key Point True self-regulation and self-supervision is almost always a gamble against very long odds. When policy makers decide that it is not cost-effective for the government financial supervisor to directly oversee large numbers of nonbank deposit-takers (especially small ­ ones), they may consider self-regulation as an alternative. Discussion of self-regulation tends to be confusing because people use the term to mean different things. In this Guide, “self-­ regulation� means prudential regulation (or supervision) by a body that is effectively con- trolled (whether in law or in fact) by the regulated entities, not by the government supervisor. ­ icrofinance Prudential Supervision Issues in Deposit-Taking M 45 Self-regulation of financial intermediaries in developing countries has been tried many times, and has rarely been effective in protecting the soundness of the regulated organizations. (Nonprudential self-regulation, including some consumer protection ­ issues, is a different matter and may occasionally be more successful.) One cannot assert ­ that effective self-regulation in these settings is impossible in principle. But it can be said that self-regulation, because of its inherent conflict of interest, is almost always an unwise gamble against very long odds, at least if the supervision is expected to enforce financial discipline and conservative risk management. Sometimes regulators have required certain small intermediaries to be self-regulated, not because they expect the supervision to be effective, but because this is politically more palatable than admitting publicly that these deposit-takers will be unsupervised. While self-regulation probably will not keep financial intermediaries healthy, it may have some benefits in getting institutions to begin a reporting process or in articulating basic standards of good practice. The question to be weighed in the balance is whether deposi- tors are being given a true picture of how effectively (or not) their deposits are being protected. Supervision of financial cooperatives Key Point Financial cooperatives—at least the larger ones—need prudential supervision by a specialized financial oversight agency that has the requisite skill, independence, ­ resources, and powers. Financial cooperatives of widely ranging types are the primary providers of finan- cial services to the poor in significant parts of the world. In many countries, financial cooperatives are licensed under a general law on cooperatives and supervised by a gov- ernment agency that oversees all kinds of cooperatives, including cooperatives focused production, marketing, and other nonfinancial activities. These agencies may have on ­ the legal responsibility for prudential supervision of the safety of deposits;67 however, as noted, they almost never have the resources, expertise, or independence to do the job ­ effectively. In some countries, the cooperative agency is not politically independent of the cooperatives it supervises. 67 Such responsibility may be implicit even if the regulations contain no specific prudential standards. 46 Microfinance Consensus Guidelines Financial cooperatives fund their lending mainly with members’ shared deposits and savings, so some argue that they don’t need to be prudentially supervised because they do not put the public’s deposits at risk.68 In fact, when the common bond among members is geographic (i.e., membership is open to anyone living in the given area), there is not much practical difference between members and the public: a nonmember can make a deposit simply by joining and paying a nominal membership contribution. Practically speaking, when it comes to deposit taking, such a cooperative isn’t very different from a bank. While the members have the right to elect the board, few may be interested in spending the time to stay abreast of the cooperative’s financial condition. Thus, aside from very small institutions, cooperative members may be in no better a position to supervise management than are the depositors of a conventional retail bank. In fact, many countries report that poor gover- nance is a major source of failure in financial cooperatives.69 Thus, the fact that financial ­ upervision— cooperatives serve “members only� does not mean they don’t need prudential s with the possible exception of very small cooperatives and those that include only employees of a given company where the company may be an implicit guarantor of deposits. Countries take different approaches to prudential supervision of cooperatives. A separate financial supervisor may oversee them. Alternatively, the bank regulator may establish a department for financial cooperatives, or at least those with assets or members ­ above some minimum threshold. Or the bank regulator may delegate authority to an independent body—typically a federated network of cooperatives.70 In the case of delegation to a network, the supervisor will supervise the federation and have the same enforcement powers as any other supervised entity.71 The f ­ederation will, in turn, supervise its members in accordance with prudential regulations devel- oped by the supervisor. Such federations should have, at a minimum, an independent audit department, an MIS for the network, consolidated data, and uniform operational ­procedures.72 To solve the supervision problem, a country may require cooperatives to federate, but this is unlikely to work well without the willing participation of the institu- tions involved and a significant initial investment of time and money. 68 In some countries, local law permits financial cooperatives to take deposits from the public. 69 See BCBS (2010), noting that the cooperative structure entails an inherent conflict of interest because the “owners are also borrowers and depositors,� which can result in “poor credit underwriting and management, inappropriate loans to related parties and frauds.� Id at 21, footnote 37. 70 Financial cooperatives sometimes create federations whose main purpose is advocacy and perhaps techni- cal assistance. In other cases (e.g., in Francophone settings), the federation structure will involve much more financial and operational integration between the federation and retail cooperatives. In tightly integrated structures, retail members operate more like branches than free-standing institutions. 71 This can be problematic for those (many) countries in which there are both federations and free-standing cooperatives that operate outside of any federated structure. 72 No federation should be permitted to conduct asset and liability management of the network if it does not have these technical tools. ­ icrofinance Prudential Supervision Issues in Deposit-Taking M 47 The case of small member-based intermediaries Key Points • In some cases, the best solution may be to allow formal but very small member- based deposit-takers to continue operating even though they cannot be effectively supervised. ­ xpect • Weak supervision can be worse than no supervision if it leads depositors to e levels of protection that cannot in fact be delivered. Ideally, all institutions taking deposits, regardless of size, should be subject to effective supervision. But some member-based deposit-taking intermediaries are so small, ­ and sometimes so geographically remote, that they cannot be supervised cost-effectively, no matter which body is doing the supervision. (Today, many countries have large num- bers of such small institutions operating without any supervision.) This poses a prac- tical problem for regulators. Should these institutions be allowed to operate without prudential supervision? Or should minimum capital and other requirements be enforced, ­ effectively closing down these institutions? Supervision required for all. Some regulators have initially inclined to the latter course. They argue that institutions that cannot be supervised are not safe, and therefore should not be allowed to take small depositors’ savings.73 After all, are not small and poor customers just as entitled to safety as large and better off customers? But this analysis is too simple if it does not consider the actual alternatives avail- able to the depositor. Poor people can and do save frequently, sometimes moving a larger portion of their income in and out of savings vehicles than better off people do.74 Where formal deposit accounts are not available, they use savings tools such as putting cash under the mattress, keeping livestock, buying building materials, join- ing informal rotating savings and credit clubs, or placing money with neighbors or relatives. All of these vehicles are risky, often riskier than a formal account in a small unsupervised intermediary.75 Closing down the local institution may in fact raise, not lower, the risk faced by local savers by forcing them back to less satisfactory forms of savings. 73 Regulators sometimes hope that, instead of shutting down, these small intermediaries will merge to form larger ones that can be supervised more easily. See discussion “Supervising merged or federated cooperatives.� 74 E.g., Collins, Murdoch, Rutherford, and Ruthven (2009). 75 E.g., Wright and Mutesasira (2001). 48 Microfinance Consensus Guidelines Supervising merged or federated cooperatives. In some countries, the supervisor has encouraged or required financial cooperatives to merge into a single cooperative large enough to allow cost-effective direct supervision or to group themselves into a federation that supervises its members under prudential norms developed by the supervisor. But merger or even meaningful federation of small cooperatives may be difficult due to the practical economics of branch operation, distance and geographic isolation, varying his- tory and institutional cultures of the merging cooperatives, and vested interests of local management. And as noted, delegated supervision has an uneven history and may not greatly reduce supervision costs. Approval to take deposits but no supervision. Other policy makers have sought an ­ eposits, intermediate course, giving small intermediaries some kind of approval to take d but without committing the supervisor to effective prudential supervision. But this risks misleading depositors about the effectiveness of the oversight the institution is actually getting. When a deposit-taking institution displays any kind of government authori- zation, many depositors assume that the government is overseeing the safety of their deposits. If the supervisor is not in a position to fulfill this expectation, it is better not to ­ allow it to be created. If very small intermediaries are allowed to take deposits without being subject to prudential supervision, then an argument can be made that their customers should be advised explicitly that no government agency is monitoring the health of the institution ­ and that they need to form their own conclusions based on their knowledge of the indi- viduals running the institution. Questions about the effectiveness or practical enforce- ability of such an approach are not easy to answer. Part IV. Nonprudential Regulatory Issues When CGAP published Guiding Principles on Regulation and Supervision of ­Microfinance in 2003, prudential regulation of deposit-taking MFIs was the main focus. More recently, nonprudential issues have become more prominent—notably consumer protection and prevention of financial crimes, including in particular money laundering and terrorist financing (ML/TF). Nonprudential issues—those that do not require the government to involve itself in protecting the financial health of providers—span a wide spectrum and generally apply to both depository and nondepository institutions. Nonprudential issues tend to be less complex and less costly to monitor than prudential issues, both for regulators and pro- viders. But nonprudential regulation is not cost-free and access can be restricted when the cost makes a given service or clientele unprofitable for providers. Therefore, just as with ­ onprudential prudential regulation, policy makers have to be cost-conscious in designing n measures, always balancing expected benefits against regulatory costs, and looking for less costly means to pursue a policy goal. Choosing the most appropriate regulator for nonprudential issues calls for care- ful consideration (see “3c. Where Should the Microfinance Supervisory Function Be ­everal Located? Within the existing supervisory authority?�). In addition, when s agencies have an interest in regulated institutions, effective communication and coordination is essential, especially for regulations that will apply to different types ­ of institutions. 4a. Permission to Lend Key Points • The regulatory framework should—absent particular local factors, such as ­ extreme corruption in the NGO sector—permit both NGOs and commercial companies to engage in microlending. • Issuance of a permit to engage in microlending should be straightforward, involving a public registry and a simple process, but not prudential regulation. ­ 49 50 Microfinance Consensus Guidelines In some legal systems, any activity that is not prohibited is implicitly permitted. In these countries, NGOs and other unlicensed entities are free to lend as long as there is no specific legal prohibition; no regulatory reform is needed to allow them to do so. In other legal systems, an institution’s power to lend—at least as a primary business—is at best ambiguous unless there is an explicit legal authorization for ­ it to conduct such a business. This ambiguity is particularly common in the case of NGO legal forms. In still other legal systems, only prudentially licensed and regulated institutions are permitted to lend, even if no deposit taking is involved. In all these cases, a simple way to expand poor people’s access to credit is to adopt ­ uthorizing nondepository MFIs to lend. These institutions would not regulation a ­ rudential regulation because there is no depositor risk and little, if any, require p systemic risk. Issuance of a permit to engage in microlending should be straightforward, involving a simple registration process with a public registry.76 Information require- ­ ments should be linked to specific regulatory objectives (e.g., identification of respon- sible parties, fit-and-proper screening of principals, etc.). Where the objective is to enable lending by NGOs, modification of the general legislation governing NGOs may be needed as well. To limit the number of credit-only MFIs, some countries may impose minimum capital requirements or other barriers to entry. This decision should be balanced with competition concerns as well as consumer protection and level-playing field issues (e.g., moneylenders not subject to such barriers may also not be subject to the consumer protection rules). As has been demonstrated in several countries, granting permission to engage in microlending presents the risk of regulatory arbitrage (see Box 1), which can introduce reputational issues for socially oriented MFIs if they are not distinguishable from payday lenders and consumer credit. 76 This is in contrast with the license application process for depository institutions, which should require submission of things such as operational policies and procedures; risk management policies and procedures; IT information; audit plans; contingency plans; detailed business plan for three years or more; several years of financial statements for previously operating companies; evidence that the board, significant owners, and senior management meet applicable fit-and-proper criteria; and evidence of contribution or accessibility of funds to satisfy the minimum capital requirement. Nonprudential Regulatory Issues 51 4b. Reporting and Institutional Transparency Key Point If regular reporting is required of lending-only MFIs, then the content and frequency of reports should be tailored to specific regulatory purposes and should be much addition, the lighter than what prudential reporting by deposit-takers would be. In ­ requirements should be harmonized as much as possible, with reporting require- ments imposed by other regulatory authorities (e.g., the regulator of NGOs). Nonprudential reporting by microlenders often includes (i) basic institutional infor- mation that is updated only when necessary (e.g., location, legal status, capital struc- ture, officers, and board) and (ii) periodically collected information about operations (e.g., financial statements and indicators of scale and portfolio quality) as well as in- formation on the characteristics, terms, and costs of the products being offered. These reports can serve several objectives: • Conducting “market reconnaissance,� so financial authorities know what is happen- ing in financial market areas that are not prudentially regulated • Promoting transparency, competition, and consumer protection • Combating ML/TF • Providing data for benchmarking purposes, so managers and stakeholders can com- pare their MFI’s performance with the performance of its peers. All of the nonprudential objectives listed above can be achieved without the more ­ eporting. burdensome (i.e., detailed, technical, and frequent) requirements of prudential r Indeed, except for consumer protection matters (e.g., abusive lending and collection practices) and preventing ML/TF, one can reasonably ask whether the government needs to be involved at all in collecting reports by nondepository microlenders. The agency collecting nonprudential reports may not need a deep understanding of the microlending business. The function might even be delegated to a private body, such as an MFI association, although in markets with diverse MFIs and methodologies, problems can arise if the body responsible for regulating the MFIs represents only part ­ eporting of the market. Whatever the body, it should be clear that its role is to monitor r compliance, not to address insolvency risk. Ideally, reporting would be harmonized with requirements of other agencies to whom microlenders report (e.g., the regulator of NGOs). 52 Microfinance Consensus Guidelines 4c. Consumer Protection Key Point As much as possible, all providers of a given financial service should be held to the same consumer protection standards. ­ roviders Financial services do not always help everyone who uses them, and financial p do not always treat customers fairly. In particular, rapid growth in credit markets that are approaching saturation can lead to over-lending and other behavior not in customers’ best interests. As a result, regulators are paying increased attention to financial consumer protection, focusing on three broad themes: • Adequacy and transparency of information—giving clients accurate and understand- able information about pricing and terms • Fair treatment—avoiding abusive lending and collection practices and other unethical treatment of clients • Recourse—providing clients with an effective mechanism for addressing complaints and resolving errors or disputes (Brix and McKee 2010). In addition, a focus is on identifying (or establishing) an appropriate regulator for financial consumer protection. Financial consumer protection deserves particular attention when applied to low- income clients who may have little education, little experience with formal financial services, and few formal providers to choose from.77 ­ The financial skills of customers themselves are an important complement to con- sumer protection regulation. Policy makers in many developed countries, and in an ncreasing number of developing and transition countries, are exploring efforts to ­ i­ increase the “financial capability� of customers. The term “financial capability� refers to the knowledge, understanding, skills, attitudes, and especially behaviors that people need to make sound personal finance decisions, suited to their social and financial circumstances. Regulation may play some role here, but the topic is not primarily a regulatory one, so it is not treated further in this Guide. 77 The Center for Financial Inclusion (CFI) has done significant work on client protection principles for microfinance (www.centerforfinancialinclusion.org). Its seven principles are appropriate product design and delivery, prevention of over-indebtedness, transparency, responsible pricing, fair and respectful treatment of clients, privacy of client data, and mechanisms for complaint resolution. The Smart Campaign has developed good practice guidelines to help MFIs follow these principles (www.smartcampaign.org). Nonprudential Regulatory Issues 53 Box 4. Level playing fields, equal treatment As much as possible, all providers of a given financial service should be held to the same consumer protection standards. For instance, consumers need the same protec- tion from misleading loan information whether the lender is a bank, an MFI, or a retailer selling appliances on credit. Applying the same rules to all providers builds consumer confidence in formal finance and reduces the risks of both regulatory arbi- trage and unethical players undermining responsible providers through unfair com- petition. Also, consumer protection rules should not favor one delivery technology over another, though this can be challenging to work out in practice. Similarly, all microfinance clients should be entitled to the same protection. There is little if any difference between an individual microentrepreneur and a small microenterprise in the use of financial services, especially in the case of microen- terprises with no employees outside the family. However, in many countries, con- sumer protection law safeguards only individuals, not businesses. Policy makers might consider extending consumer protection to microenterprises. If this is done, “microenterprise� should be defined. ­ Adequacy and transparency of information Key Point Microfinance providers should be required to give clients clear and complete information about services offered, including their terms and costs. However, with ­ respect to microloans, the standard annual percentage rate (APR) may not be the most effective way to communicate costs to low-income borrowers. For less experienced financial consumers in particular, disclosure rules should em- phasize plain language and simple presentation (e.g., font, format, timing, and location of disclosures; no complex formulas or calculations)78 and comparability of different 78 Some jurisdictions require financial service providers to given consumers a key facts information sheet ­ efore they purchase a financial product (e.g., the “Schumer box� in the United States is a summary of the b costs of a credit card and is required of all credit card companies). 54 Microfinance Consensus Guidelines products and different service providers.79 Where there are low levels of literacy, disclo- sure rules should ensure effective communication to those who cannot read. Transparency rules often seek to promote comparison shopping and price competi- tion by standardizing disclosure content (including a statement of fees and commissions), returns.80 forms, and wording, as well as formulas for calculating consumers’ costs and ­ including, in partic- It is not clear, however, whether complete and accurate information—­ ular, disclosure of APR on loans—will affect consumers’ behavior as much as expected.81 designed for more affluent consumers in developed Moreover, approaches to disclosure ­ ­ odification, when serving poor countries may not always work well, at least without m consumers in developing and transition countries.82 And more is not always better. At some point the volume of disclosure produces diminishing returns, so the focus should be on simplicity, quality, and clarity, not quantity. Discrimination Key Point Regulation should prohibit discrimination—whether against women or a particular race, caste, religion, or ethnic minority. Many of those who are discriminated against for their sex, race, caste, or otherwise are also poor. Prohibiting such discrimination can be an important part of regulating for financial inclusion. To address the problem of discrimination, financial service pro- viders should be required (i) to be transparent regarding the criteria for eligibility for a 79 Rules should, however, be sufficiently flexible to permit the development of new channels, such as branch- less banking. For instance, requiring that a bank or MFI employee provide all information to the customer in writing before the service is contracted would, without further clarification, prohibit the opening of accounts through agents. See “5d. Branchless Banking Consumer Protection.� 80 Careful consideration should be given to the type of standardized measure. For example, although APR may make comparison shopping easier, consumers may find the “total cost of credit� figure plus the repay- ment schedule easier to comprehend and prefer using it to assess affordability and choose from among loan offers (FSD-Kenya 2009). See Tiwari, Khandelwal, and Ramji (2008), effective lending rate may not be suit- able for microborrowers. 81 The evidence on the impact of disclosure on consumer decision-making and outcomes is mixed. Researchers have found low comprehension of tools, such as APR, that are intended to facilitate cross-­ product comparisons (see, e.g., Belsky and Essene [2008] for a review of recent evidence). However, some research indicates that behaviorally informed disclosure approaches based on consumers’ actual decision- making processes can improve consumer choices (Bertrand and Morse 2010) or impact providers’ pricing of products and disclosure of key terms in a way that benefits consumers (Stango and Zinman 2011). 82 See, e.g. Chien (2012); FSD-Kenya (2009, p. 21); and Tiwari, Khandelwal, and Ramji (2008, p. 4). Nonprudential Regulatory Issues 55 specific product and (ii) to provide a client who is denied a product the reasons for such denial. However, some regulators may allow providers to limit the provision of services to women only, or another group that is more likely to be the victim of discrimination. Abusive lending and collection practices Key Points • Regulation should address aggressive or coercive sales practices as well as ­ “predatory� lending designed to take advantage of borrowers’ lack of education or experience. • Restraints on abusive collection practices may be needed, but care is required in defining what is abusive. Regulatory tools to promote fair lending include standard loan forms and disclo- sure rules, recourse mechanisms for consumers, basic registration of all credit providers, required assessment of repayment capacity, and constraints on debt-to-income ratios. Regulations should also prohibit misleading advertising. Some microlenders’ have articulated a zero-tolerance policy toward delinquency, based on a concern that uncollateralized microlending is sustainable only if repayment ­ ver-aggressive rates are kept very high. But implementation of such a policy can lead to o collection practices that damage borrowers and their households. Possible regulatory tools include rules against intimidation and coercion, and due process for seizing and ­ auctioning pledged goods (Porteous and Helms 2005; Porteous 2009a; Brix and McKee 2010).83 Public and private dispute resolution channels can also help address abusive prac- problems. tices. Providers and supervisors can monitor complaints to identify recurring ­ Crafting clear and enforceable definitions of prohibited behavior is challenging. Especially in the area of collection practices, what is “abusive� will often depend on local ­ cultural norms. Over-indebtedness In light of the 2008–2009 global financial crisis and widespread over-indebtedness in some developed countries, policy makers have become more alert to the risk that too 83 For consumer lending to salaried employees, some countries have banned automatic collections, such as payroll deductions. 56 Microfinance Consensus Guidelines many clients of retail credit, including microcredit, are becoming over-indebted (Schicks and Rosenberg 2011).84 In fact, when fast growing, competitive retail credit markets reach a saturation point, a rise in over-indebtedness is to be expected. (Microcredit mar- kets can reach saturation—that is, supply can catch up with demand—at fairly low pen- etration levels; at any given point in time, large proportions of eligible borrowers simply do not want a microloan [Anand and Rosenberg 2008].) In the absence of reliable credit reporting systems that reach microcredit borrowers in a given country (and indicate a borrower’s total outstanding indebtedness to formal lenders), the only practical indica- tors of over-indebtedness are (i) the collection performance of the lending institutions and (ii) debt-to-income ratios (which may be collected by lenders), and (iii) increases in loan sizes, terms, or rescheduling. Collection statistics are trailing indicators that for the most part reveal problems after they have occurred. In some circumstances, clients’ inability to repay may take a long time to show up in a lender’s repayment statistics. Without adequate internal controls, loan officers may roll over unpayable loans, or clients may ­ repay one loan by taking out another from a different source. Assessment of a loan applicant’s cash flow and repayment capacity is a cornerstone of sound loan underwriting, at least for individual lending models. A reliable credit reporting system can tell lenders about an applicant’s outstanding debt and repayment ­ record with other formal lenders. (See “4d. Credit Reporting Systems.�) However, credit reporting systems don’t capture borrowing from informal sources, which may be sub- stantial. In addition to preventive measures, there can also be curative measures, such as debt counseling and legal remedies. Interest rate caps Key Point ­ emote Interest rate caps can restrict access by making it impossible to serve small or r borrowers. It may be politically difficult to set a cap that is high enough to cover the unavoidable costs of microlending and a profit margin high enough to attract capital to low-income financial services. 84 There is no simple, commonly accepted definition of the term “over-indebtedness.� Some associate it with borrowers who can’t repay their loans. Others would include borrowers who can repay, but only at the expense of sacrificing basic household consumption needs. Clients with multiple outstanding loans are not necessarily over-indebted, although multiple borrowing is statistically correlated with higher delinquency in most (but not all) of the empirical studies to date. Nonprudential Regulatory Issues 57 The frequent calls for regulatory caps on microloan rates are countered with the argument that caps tend to hurt poor people’s access to credit by making it unprofit- able to offer very small loans or loans to widely dispersed rural borrowers. (This is primarily due to administrative costs, which are a much higher percentage of loan assets for microlending than for conventional retail bank lending,85 although many MFIs could reduce their administrative costs through increased efficiency.) The limited empirical evidence available supports this concern, and also suggests that when inter- est rate caps are not carefully written or cannot be rigorously enforced, prices may become less transparent as lenders start using fees and commissions to increase their loan income.86 In theory, interest rate caps could be set at a level that permits sustainable microfi- nance operations while eliminating excessive profits. But achieving that balance can be politically difficult for the government agency that has to identify (and implicitly sanc- tion) a particular rate. Most people do not understand why tiny loans require high inter- est rates, so it tends to shock the public conscience when MFIs are allowed to charge very high rates to poor borrowers. Furthermore, there may be no one “sustainable� interest rate even within a given microcredit market, given the differences among suppliers, ­ market segments, and cost structures. Loan sizes and other cost factors can vary widely, and an interest rate that would produce large profits on $1,000 loans might not even cover costs on $100 loans. Even in the absence of interest rate controls, rates have been dropping in most microcredit markets. At the same time, there have been instances when MFIs have ­ ­ immediately reduced their rates in response to public criticism, demonstrating that rates in some markets can fall even lower than they have already. The interest rates of some MFIs are high in their market and produce profits that most people would regard as excessive, but worldwide the percentage of poor borrowers that are paying such rates appears to be quite small (Rosenberg, Gonzalez, and Narain 2009). A reasonable alternative to caps is effective disclosure combined with steps to help consumers understand the product and pricing. This—together with additional efforts to publish comparative prices among lenders—has served in some markets to bring down interest rates, spurring more effective competition among MFIs and other financial ­service providers. 85 It costs less to make and manage a single loan of $1 million versus 10,000 loans of $100 each (Rosenberg, Gonzalez, and Narain 2009). 86 E.g., Campion, Ekka, and Wenner (2010); Smart Campaign (2010); and Helms and Reille (2004). 58 Microfinance Consensus Guidelines Data privacy and security Key Points • Effective protection of microfinance clients’ privacy—so that information about them is collected, stored, viewed, and used only in proper ways by designated people—is an easy objective to articulate but difficult to implement in practice, especially in countries where diverse providers serve the same poor clients, but are subject to varying privacy related regulation. • Regulation on bank secrecy and general regulation on the use of personal data protect client privacy, but they can also hinder effective credit reporting for poor customers, making it harder to manage over-indebtedness risk. Protecting clients’ private financial information is usually addressed in bank secrecy laws. In an increasing number of countries, such privacy is also protected by general regulation on the use and processing of personal data. While bank secrecy laws rarely extend to MFIs that are not prudentially regulated, the general data privacy rules that protect individuals usually do. Breaches of privacy can result from client behavior, such as failing to protect a password, so consumer awareness is an important complement to regulation. Regulation can in turn reinforce consumer awareness, for example, by requiring providers to cover such basics in their disclosure (although many will do so without regulatory compulsion, simply as a matter of good business judgment, including as a means to combat fraud by employees or agents). Enforcing such a wide array of privacy-related regulation can be challenging, given the potential variety of regulators involved—especially for cross-border services and data. Civil penalties may be effective to enforce some privacy protections, but sanctions for others should be a matter of criminal law. However, there may be trade-offs between data privacy rules and the goal of greater financial access for the poor. For example, in a country without credit information ­ systems that reach microborrowers, informal information sharing among microlenders may be the only way to protect against dangerous levels of cross-borrowing, even though such data sharing would violate current bank secrecy and consumer data privacy protec- tions in many countries.87 Where specific regulatory barriers prevent information sharing among lenders serving poor clients, an argument can be made for relaxing these barriers to allow for credit reporting systems that mitigate over-indebtedness risk. 87 Informal information sharing among MFIs presents the risk of a consumer being unfairly blacklisted, especially as there is no mechanism (or incentive on the part of the MFIs) for correcting old or inaccurate information. Nonprudential Regulatory Issues 59 Recourse Key Point The consumer’s ability to lodge complaints and seek redress is an important part of financial consumer protection. For most microfinance consumers, judicial recourse will not be a viable option for many reasons (including expense and time), so the focus needs to be on alternatives. As a first step, providers should have, and can be required to have, a professional and easily accessible complaints-handling process. This internal dispute resolution process is more ­ rocedures. likely to be adequate if regulators oversee it as part of their normal supervision p Some regulators choose to require standardized reporting on complaints data and resolution status. If internal dispute resolution is unsuccessful, other avenues of recourse may be feasible, such as civil society organizations, industry associations, mediators, or ombuds schemes;88 regulators themselves may be in a position to offer an alternative recourse channel. Normal judicial processes are usually too expensive and lengthy for use by poor customers, but a simple, cheap, and easily accessible forum (e.g., a small claims court) can be effective. Recourse for microfinance consumers should feature simple, plain-language pro- cesses, little or no cost to the consumer, a trusted recourse provider, and convenient access points, such as call centers, walk-up complaint desks, or borrower group meetings. ­ Customers may need assistance in filling out and submitting written complaint forms. Regulators’ review of complaints data helps regulators to identify patterns of abuse and inform future consumer protection policy decisions. Absence of complaints is not always a good sign; consumers may not know how to file a grievance or may be intimidated by the process. Lack of trust in governments and financial institutions may lead consumers to doubt that their concerns will be heard so transparent processes are particularly important, and new recourse initiatives should include measures to educate customers about their options. An appropriate regulator Choosing an appropriate regulator for financial consumer protection is an issue in both developed and developing countries. The three most common options are (i) a financial institutions, regulator (whether the banking regulator, a regulator of nonbank financial ­ or a regulator of credit providers), (ii) a specialized financial consumer protection authority, or (iii) the general consumer protection body, if one exists. ­ 88 Ombuds schemes involve an independent body or individual that investigates and mediates consumers’ grievances. 60 Microfinance Consensus Guidelines Choosing a financial regulator has a few advantages over relying on a general con- sumer protection body. First, a financial regulator will be familiar with the products in question. Second, providers may be more responsive to a financial regulator, given their desire to keep in good standing. However, very few regulators around the world have responsibility for the full range of financial service providers that might be engaged in microfinance activities. The financial regulator is often reluctant to undertake additional responsibilities, including consumer protection enforcement. In contrast, a specialized ­ financial consumer protection regulator may create a more level playing field across dif- ferent providers offering similar products and services. (The recent global financial crisis has resulted in more than one country separating the prudential regulator from the mar- ket conduct regulator.) Whatever the decision, it is essential to clarify the scope of author- ity of all the relevant agencies as well as their relationships with other bodies regulating the same institutions. No matter which agency is selected to oversee financial consumer protection, it will likely face limited resources. To reduce strains on regulatory capacity, some authorities encourage industry associations to establish and enforce conduct-of-business standards on members, with the goal of compelling early resolution of problems before govern- ment intervention is needed. (In countries with multiple types of providers serving over- lapping poor clientele, such an approach will be of limited effectiveness unless all such lenders choose or are compelled to participate.) One of the main tools is a formal code of ­ conduct, where providers commit to standards covering disclosures, product suitability, marketing, collections, and complaints handling. But without a regulatory requirement or threat of regulation, such standards may not be very effective. 4d. Credit Reporting Systems89 Key Points • It is critical for the healthy development of microfinance to foster the develop- ment of broad and deep credit information databases that include current loan balances and negative and positive information on the past payment behavior of poor customers, particularly in markets approaching saturation. • Microlenders (of whatever legal form) and borrowers are better served by credit reporting that draws from comprehensive payment data rather than just micro- credit data. 89 See Lyman et al. (2011). Nonprudential Regulatory Issues 61 Box 5. The benefits and challenges of credit reporting in microfinance Credit reporting systems offer important benefits both for financial institutions and for their clients. By collecting information on clients’ status and history with a range of credit sources and possibly other parties to whom payments are regularly made, credit bureaus allow lenders to assess risk more accurately and less expensively. (Some argue that very small MFIs or those using a lending methodology that relies on community ­ owever, even knowledge, such as group lending, have less need of credit report data. H indebtedness and have these institutions have reason to be concerned about multiple ­ sometimes expressed interest in such reports.) At the same time, they allow borrowers to create “reputation collateral� based on their payment histories. Without a fairly comprehensive credit reporting database, a borrower’s faithful repayment of a loan is known only to the institution that made that loan; with comprehensive credit report- ing, repayment of a loan from one formal lender makes it easier for the borrower to borrower’s ob- get loans from other lenders, and gives lenders a fuller picture of the ­ ligations. Thus, credit reporting allows lenders to do more lending without physical collateral, and strengthens borrowers’ incentive to repay. In developed countries, the combination of credit reporting and statistical risk-­scoring techniques has massively expanded the availability of credit to lower income groups in recent decades. In developing countries, credit reporting has made important gains over the past decade. However, credit reporting has not yet been adopted (or remains far from comprehensive for low-income borrowers) in many of the poorest countries. Even in more advanced emerging markets, adoption and comprehensiveness remain uneven. In countries that have credit reporting, low-income consumers are often left out due to minimum loan size reporting thresholds (set by regulation or by the credit reporting services themselves), exclusion of MFIs from credit reporting databases (by regulation or customers, by practice), or MFIs’ lack of interest in participating. When MFIs compete for ­ particularly in more saturated markets, over-indebtedness and default rates have been seen to rise quickly in the absence of reliable credit reports on those applying for loans. Credit reporting entails risks as well as benefits. Corrupt database managers may sell information to unauthorized parties. Inaccurate information in the database can hurt borrowers, although guaranteeing them access to their own credit histories can lower this risk. Lenders may respond too conservatively to reports, automatically denying credit when there is any negative data in an applicant’s record. A database that ­ includes both negative information (about missed payments) and positive information (about good repayment) allows a more complete picture of creditworthiness to emerge. 62 Microfinance Consensus Guidelines Financial regulators often establish and house a public credit registry, which may require no specific legal or regulatory authorization, depending on the country. But the development of a private credit information market usually depends on some adjustment to the country’s legal framework. Private credit bureaus raise confidentiality issues, espe- cially when banks (which are typically subject to bank secrecy regulation) participate.90 Sometimes these issues can be handled simply by standard loan agreements in which the borrower authorizes the lender to obtain the borrower’s credit report and to share infor- mation with a credit bureau. Private credit bureaus tend to have some important advantages over public credit ­ncluding registries. Private bureaus generally collect data from a wide variety of sources, i utilities and retailers. They collect both positive and negative data. They tend to use modern information technology. And sometimes they offer value-added services, such as credit scoring. However, the large credit bureaus using conventional methods of operation and data collection often face obstacles to including MFI data in their reports. MFIs’ small loan volume can be cumbersome and costly for credit bureaus to handle, MFIs may not have the information systems or the data quality needed to meet credit bureau requirements, and sometimes MFI customers may not have the national identification documents that credit bureaus require. Further, the cost to MFIs of obtaining credit reports can be high in relation to loan size, especially if the MFIs don’t enjoy the volume discount price that large commercial banks get. If few of the participating lenders in a credit bureau serve microborrowers, then the data may not be very useful for MFIs. For this reason, it is important for private credit bureaus to include alternative sources of payment history, such as utilities or retailers. In countries where MFIs are not bound by bank secrecy or other data privacy rules, a private credit bureau formed solely for MFIs (e.g., by an MFI association) may have cost and access advantages. But as noted, all other things being equal, credit bureaus that draw from more comprehensive payment data sources are more helpful, both to micro- lenders and borrowers. If the private sector has not developed a credit bureau that adequately serves micro- lending, there may be a case for government involvement, such as (i) adopting regulatory incentives for credit reporting—e.g., lower provisioning for loans made using a credit report; (ii) requiring MFIs to organize a credit bureau for themselves or participate in a sufficiently broad-based existing scheme; or (iii) requiring the sharing of credit data 90 Public credit registries typically do not face the same legal hurdles regarding privacy and, more specifically, bank secrecy, because these issues are addressed in the relevant banking legislation. Nonprudential Regulatory Issues 63 among all licensed credit bureaus. The third option, which is typically accomplished by having the public credit registry share its data with the private credit bureaus on a low- cost or no-cost basis, is meaningful for microfinance only if the relevant credit providers participate. 4e. Limitations on Ownership, Management, and Capital Structure91 Key Points • Restraints on foreign investment or management can sometimes hinder the ­ development of financial services for the poor. • NGOs should be permitted to own shares in for-profit MFIs that specifically target the poor. Restraints on foreign investment or management can sometimes be a significant obstacle to the development of financial services for the poor. ­ “Greenfield� operations (i.e., start-ups) as well as NGO transformations can be hin- dered by rules about currency, citizenship, shareholder diversification, and foreign invest- ment. In quite a few countries, foreign financial and technical inputs have been important in the progress of microfinance. Thus, it can be problematic if MFIs face prohibitions or severe limitations on the participation of foreign equity holders, borrowing from foreign sources, or employment of noncitizens in management or technical positions. In some countries, the microfinance business may not attract many domestic commercial inves- tors for some years yet, so limitations on foreign investment or staff can impede the expansion of services. Nonprofit law may prevent a domestic or foreign NGO from holding an interest in a financial institution. In some countries the prohibition is rooted in the idea that nonprofit organizations should support themselves exclusively from donations, and not by earning returns from the sale of goods or services. However, this view is less common today than a decade ago.92 Concerns about NGO ownership based on this outmoded view should be addressed by clarifying the relevant NGO regulation. 91 This section focuses on nonprudential requirements, which are applicable to both depository and nondeposi­ tory institutions. Some prudential ownership limitations are discussed in “2d.Transformation of NGO MFIs into Licensed Intermediaries—Ownership suitability and diversification requirements.� 92 Today there is a growing consensus that an NGO MFI that generates net revenues does not violate its non- profit status if it uses the surplus to fund its work, provided it does not make distributions to private parties or otherwise use the surplus to benefit insiders (e.g., by paying above-market salaries). 64 Microfinance Consensus Guidelines 4f. NGO Transformations into For-Profit Companies Key Points • Creating a clear legal path for NGO MFI transformations can be an important enabling reform, and may involve changes in nonfinancial laws and ­regulations— particularly the legal framework for NGOs. • Regulators should address the risk of asset-stripping presented in the transfor- mation of a public-benefit NGO into a privately owned institution. A large number of MFIs in developing and transitional countries still operate as ­ locally formed NGOs or projects of international NGOs. Many NGO MFIs plan to transform into a for-profit company (either deposit-taking or nondeposit-taking) to access commercial funds, to offer services that an NGO is not permitted to offer, or to ­ comply with a new legal requirement. Some countries have required NGO MFIs to trans- form, usually in an effort to improve regulation and general organization of the sector.93 The most common transformation transactions involve the transfer by the NGO of its loan portfolio and other assets, liabilities, and employees to a new or previously existing company in exchange for shares in the company or payment in the form of cash or debt. An NGO is an ownerless legal entity that is generally required by law and its char- ter to serve a public benefit purpose rather than the private interests of its founders, managers, or other insiders. In contrast, for-profit companies are run for the benefit of their owners. Most transforming NGO MFIs will want an ownership interest in the new entity to ensure that it continues the mission (e.g., serving the poor)—although it should be noted that NGO ownership doesn’t guarantee against mission drift, nor are private shareholders necessarily worse at protecting social performance. At the same time, NGO transformations typically involve the entry of new investors. Who those investors are and how much of the new company they own will depend not only on practical consid- erations but also on applicable domestic regulation. If policy makers want to create a framework that helps NGO MFIs turn into for-profit companies, they often need to consider adjustments in several other areas of regulation, not just financial regulation.94 In particular, NGO law in many countries is not very well 93 While there can be benefits of this approach, forcing transformations in a tight timeframe can also be ­ disruptive for existing MFIs and their clients. 94 In most countries, existing regulation does not contemplate, and has never before been applied to, micro­ finance transformations. Nonprudential Regulatory Issues 65 developed, especially regarding complex questions at the boundary of what may be con- ­ ­ ountries, there may sidered appropriate for an entity serving the public interest. In some c be ambiguity regarding whether a public-benefit NGO can transfer its portfolio and other property to a for-profit company. There may also be a legal prohibition on an NGO having an ownership interest in a commercial company (especially a controlling interest), legal lim- its on foreign participation, or other ownership suitability requirements applicable to finan- cial institutions. Even where such restrictions do not prevent transformations, they can result in the use of artificial, inappropriate, or inefficient structures to capitalize the new company. Transformations present an opportunity for inappropriate transfer of the NGO’s assets to private pockets. Such asset-stripping is often a violation of NGO law, but the ­ legal provisions may be less than clear or not well-enforced. Insiders can strip assets through excessive compensation or the award of shares in the new company for less than their value. Asset-stripping is also an issue when outside investors are introduced. What percentage ownership does the NGO get in the transformed entity in exchange for its contribution of portfolio, systems, know-how, and intangibles such as customer relationships? And what percentage is appropriate for those investors who invest cash? As yet there is not a robust set of valuation benchmarks in terms of price/book value or price/earnings ratios, so it may not be obvious if a “sweetheart� deal is enriching private shareholders at the expense of the MFI (O’Donohoe et al. 2009).95 insiders Measures to prevent stripping of an NGO’s assets include (1) requiring NGO ­ to recuse themselves from the negotiations if they or related parties would be receiving shares in the transformed entity or be employees of the new private company96 and (2) bringing in a respected, independent person with valuation expertise to judge the fair- ness of the transaction. Such expert valuations are often costly. 4g. Secured Transactions Key Point Legal and judicial reform to support secured transactions—in particular, a collat- eral law and accessible collateral registries—may facilitate microfinance, although typical microcredit is effectively unsecured. 95 Also, Reille et al. (2010) and Glisovic, Gonzalez, Saltuk, and de Mariz (2012). 96 There is also a risk, if key NGO insiders recuse themselves, that the other NGO stakeholders who are left to make the decisions may be uninformed and may unintentionally drive too soft a bargain in negotiating with the new private shareholders. 66 Microfinance Consensus Guidelines Borrowers, lenders, and the national economy all benefit when not only real estate but also moveable and intangible assets can be pledged as collateral for loans. In many developing and transitional economies, it is expensive or even impossible to create and enforce a security interest in moveable or intangible collateral. Sometimes low-income people also face constraints on the use of their homes and land as collateral. Reform of secured transaction regulation is likely to benefit the middle class more than typical microborrowers. Many microloans are so small that registration of collateral would not be cost-effective. Many of the assets of the poor would not be suitable for pledging even under reformed regulation. And, of course, effective microlenders know how to recover loans without collateral. Nevertheless, in some countries the microloan clientele—especially at its upper end—includes borrowers who could provide worthwhile collateral if the regulatory framework is conducive. Where this is the case, MFIs take a pledge of whatever property the borrower can offer to increase repayment incentives— even if the loan amount is significantly higher than what the property would bring at a foreclosure sale. 4h. Financial Crime97 Three financial crime topics predominate in the regulation of microfinance: AML/CFT, fraud and related financial crimes (particularly pyramid investment schemes), and iden- tify fraud. In the past few years, AML/CFT has commanded the most international attention. But from the perspective of poor customers of financial services, both pyramid ­ investment schemes (also known as Ponzi schemes) and identity fraud loom large as well. AML/CFT Key Point Applying AML/CFT rules for conventional banking to tiny microfinance ­transactions can seriously limit access unless a risk-based approach is adopted. Almost all countries have committed themselves to implement the Forty Recom­ mendations of the Financial Action Task Force (FATF), the international SSB for 97 Financial crime can lead to an institution’s insolvency, so prevention of financial crime can serve both prudential and nonprudential purposes. ­ Nonprudential Regulatory Issues 67 AML/CFT.98 Key AML/CFT controls include (i) customer due diligence (CDD) mea- sures such as verifying customers’ identity (also referred to as “know your customer� [KYC] rules), (ii) record-keeping, and (iii) reporting of unusual and suspicious transac- tions to the authorities. The FATF Recommendations extend to a wide range of activi- ties, including taking deposits from the public, consumer lending, and providing formal or informal money transfers (although enforcing rules for informal transactions poses obvious challenges). A critical aspect of the newly revised Recommendations for financial inclusion efforts is the prominence of the risk-based approach (RBA), which is now the first Recommen- dation (R1).99 R1 instructs countries to identify, assess, and understand ML/TF risks and ensure that measures to prevent ML/TF are commensurate with the risks identified. R1 explicitly permits countries that “identify lower risks to allow simplified measures for some of the FATF Recommendations under certain conditions.� The Interpretive Note to the Recommendation further clarifies that “where there is a proven low risk of money laundering and terrorist financing, [the country may] decide not to apply certain Recom- mendations to a particular type of financial institution or activity� (FATF 2012). This new Recommendation—which incorporates the views previously expressed by FATF, including in a 2011 guidance note on financial inclusion (FATF, APG, and World Bank 2011)—reflects the understanding and acknowledgment that, when applied to hun- dreds of thousands of very small transactions involving poor customers, compliance with AML/CFT rules can increase operating costs significantly. Such higher costs can make it impossible to serve some poor customers viably. The Recommendation also reflects the recent awareness by FATF, regulators, and others that the financial integrity goals of AML/CFT regulation and the financial inclusion goals of microfinance can reinforce each other rather than conflict and, importantly, financial exclusion presents financial integrity risks.100 And although low value doesn’t necessarily mean low risk of money laundering or terrorist financing, many countries have already implemented special AML/CFT treat- ment for microfinance and other small-value financial transactions. 98 The original Forty FATF Recommendations were issued in 1990 specifically to combat money laundering). The Recommendations have been revised three times since: in 1996, 2003, and 2012. The 2012 revision inte- grated the original Forty Recommendations with nine Special Recommendations (issued in 2001 and 2003), which addressed FATF’s then expanded mandate to address terrorist financing. The 2012 revision also reflects an expanded mandate, which includes combatting the proliferation of weapons of mass destruction. 99 RBA is essentially the proportionate approach to regulation and supervision endorsed by several other SSBs. See CGPI white paper (2011). 100 adopting The guidance paper identified financial exclusion as a ML and TF risk. The Ministerial Decree ­ the revised Recommendations incorporates this concept. See FATF (2012b). See also Isern and de Koker (2009). 68 Microfinance Consensus Guidelines ­ emain. Notwithstanding these significant changes within FATF, important challenges r In many countries, the most significant challenge will be adjusting CDD rules to account for the limited formal identification available to poor people and to allow alternatives (e.g., verification of identity by a village leader). To address this challenge, some countries are trying to improve the availability and use of national identification cards or unique identification numbers. Once an identified customer has a relationship with service provider, AML/CFT risks in subsequent transactions can be mitigated by imposing caps on transaction sizes, monthly flow-through, and balances, as well as by implementing software and staff training to identify suspicious transactions. ­ The cost of producing and storing required records may also present a challenge. As noted, the inherently high administrative costs of large numbers of small transactions makes cost containment crucial to microfinance outreach. Regulation should permit elec- tronic rather than paper records of CDD verification, which can reduce costs dramati- cally. Other cost containment measures include allowing agents or other third parties to conduct CDD. Fraud, pyramid investment schemes, and related financial crimes Key Point In many countries, the existing anti-fraud and financial crime regulation may be adequate to address abuse connected to the provision of microfinance services, ­ or may need amendment only to add any new categories of institution to the regulatory landscape. Often the most pressing need is to improve enforcement of existing laws. Like other financial institutions, MFIs can fall victim to fraud and related financial crimes. For instance, loan officers may invent fictitious borrowers or embezzle customers’ payments; managers may misappropriate grant funds; or criminals may take deposits in a fake MFI or branch, and then flee. Pyramid investment schemes are prevalent in many developing and transition coun- tries; they occur in developed countries as well. In pyramid schemes, a high-yielding investment is promised; early investors are in fact paid the high returns promised, but out of the investments of subsequent victims. Nonprudential Regulatory Issues 69 Pyramid schemes are especially relevant to microfinance when they target victims who sometimes are less educated about and experienced with financial services. The schemes ­ assume, or mimic, a regulatory form that is also used by legitimate microfinance providers—e.g., pyramid schemes have often been structured as financial c ­ ­ ooperatives. yielding In the short term, this hurts legitimate institutions that lose customers to high-­ pyramid providers. Once the scheme collapses, it can have long-term consequences for poor customers’ trust in legitimate institutions, as well as in the regulator. Outrage at the government’s failure to prevent large-scale pyramid schemes (and even implicit complic- ity, in some cases) has had serious political consequences in quite a few countries, even to the extent of regime change. Unfortunately, there is no simple regulatory answer to pyramid schemes. Timely identification of a pyramid scheme is difficult, although this can be addressed through warnings to the public and through providing a specific channel for clients to deliver ­ information about perpetrators. Beyond these, the available tools are those used to attack financial fraud generally. However, to reduce the risk of poor enforcement (whether due to corruption or otherwise), it may be advisable to give a specified government agency explicit responsibility for pyramid schemes, including the authority to close them down and refer their perpetrators for prosecution. Often, more than one agency is involved in regulating and shutting down these schemes—such activities could involve bankruptcy investigations, criminal investigations, asset forfeiture, and the like. Where this is the case, it may be necessary to establish an intragovernmental body representing the key agencies that should act against pyramid schemes.101 Identity fraud Key Point In most countries, identity fraud and theft are recognized as crimes. The focus should be on improved enforcement of the existing criminal sanctions and of other data protection and privacy measures. Identity fraud in financial services involves assuming another person’s identity, or a fictitious identity, to procure financial services or to appropriate funds. Microfinance 101 The members of the task team would need to evaluate their capacities and powers to identify schemes promptly, and to ensure that their agencies jointly are able to close down the schemes, trace and repay investor funds, and take appropriate law enforcement action against the managers of the scheme. ­ 70 Microfinance Consensus Guidelines providers can be both victims and facilitators of identity fraud. For example, fraudsters may assume fictitious identities and falsify information about their cash flow to get loans they don’t intend to repay, or may assume the identity of a borrower with a solid credit rating for the same purpose. Identity theft is relatively easy when illiterate victims can be induced to sign documents they can’t understand. It can also result from a data security breach, or because the victim is induced to relinquish a password or personal identifica- tion number. Such activity is a crime in most countries. The focus should be on improved enforcement of the existing criminal sanctions and of other data protection and privacy measures. 4i. Tax Treatment of Microfinance102 Key Points • Favorable transaction tax treatment should be based on the type of activity or transaction, regardless of the nature of the institution and whether it is pruden- tially licensed. • It is reasonable to argue that NGO MFIs should be treated the same as other public-benefit NGOs, when the tax in question is a tax on net profits. • Profit-tax deductions for expenses that are available to banks should also be available to other providers of similar services. The wide disparity in tax systems and other local factors may call for differing results, but in most countries the discussion can at least be organized around the distinc- ­ tion between taxes on financial transactions and taxes on net profits arising from such transactions. Taxation of financial transactions and activities In principle at least, it is normally preferable for institutions of different types to pay the same transactions taxes if they are conducting the same activities. In some countries, only prudentially regulated institutions get favorable tax treatment with respect to financial transactions (e.g., no value-added tax on bank loans, no tax on bank interest revenue), 102 Taxation of insurance products involves many unique questions not applicable to providers of credit, savings, and money transfer services, and is not addressed in this section. ­ Nonprudential Regulatory Issues 71 even though the favorable tax treatment has little relationship to the particular objectives of prudential regulation. In other countries, financial transaction taxes affect financial cooperatives differently from banks. To avoid tax arbitrage, favorable transaction tax treatment should ideally be based on the type of activity or transaction, regardless of the nature of the institution and whether it is prudentially licensed. However, some countries have special transaction-tax allowances for nonprofits. Taxation of profits With respect to taxes on net profits, there is a respectable argument that NGO MFIs should be treated the same as other public-benefit NGOs. Exemption from profits tax is based on the principle that the NGO is rendering a recognized public benefit (in this case, serving the poor) and does not distribute its net surpluses into the pockets of private shareholders or other insiders; rather, it reinvests any surplus to finance further socially beneficial work. To be sure, there are always ways to evade the spirit of this nondistri- bution principle, such as excessive compensation and below-market loans to insiders. ­ However, the most sensible way to deal with this risk is usually better enforcement of NGO law, rather than taxation of NGO profits. In many countries, cooperatives are treated as noncommercial entities because of their mutual benefit orientation, and are therefore exempt from profit tax despite the fact that they may distribute their net surplus among the member owners. Opinions vary as to whether these institutions deserve such public subsidy when MFIs organized as companies that provide the same services to the same customers are ineligible. In some countries, it appears that MFIs that normally would have been set up as joint stock com- panies are instead set up as cooperatives solely to secure a profit tax exemption (i.e., tax arbitrage is taking place). Rules for tax deductibility of expenses (such as reasonable provisioning for bad loans) should apply consistently to all types of institutions subject to a profits tax, regard- less of whether they are prudentially licensed. Yet this is not the case in many countries, because the tax rules were drawn up with only banks in mind. Moreover, if prudential regulations require more aggressive provisioning of a delinquent microloan portfolio than a conventional loan portfolio, then the microlender’s profits tax deduction should also vary accordingly. Part V. Regulating the Use of Branchless Banking to Serve the Poor Key Points • Branchless banking—using technologies, such as mobile phones and smartcard readers, to transmit transaction details and using existing retail establishments to act as the principal customer interface—holds the promise of significantly expanding financial access by lowering transaction costs for the lender and ­ improving convenience for the customer. • A suitable regulatory framework for branchless banking should include (i) condi- tions for banks’ and nonbanks’ use of agents or other third parties as a customer interface; (ii) a flexible, risk-based AML/CFT regime; (iii) a clear regulatory regime for nonbanks to issue electronically stored value; (iv) consumer protec- tion tailored to the branchless context; and (v) payments system regulation that allows (at least in the long term) broad interoperability and interconnection. In developed and developing countries, more and more people are making and ­ receiving payments using technology, such as payment cards and mobile phones, instead of cash. These technologies are also being used for other financial services, including with- drawals, deposits, and savings. And in a few instances, these technologies are being used to disburse loans and collect repayments. This is branchless banking: the delivery of financial services outside of conventional bank branches; using third parties, such as retailers, as the principal interface with customers; and relying on technologies, such as card-reading point-of-sale (POS) terminals and mobile phones, to transmit transaction details. Because it uses existing infrastructure (e.g., mobile phones and retail stores) and agent networks, branchless banking potentially can reduce costs radically and improve convenience, and thus reach many customers who are presently unbanked or underbanked. 103 103 Branchless banking is not new: automatic teller machines (ATMs) and POS devices in retail outlets have been available for decades. But the use of branchless banking by populations who haven’t been served by traditional bank branches is new, as are some of the actors involved—e.g., mobile phone network operators. 72 Regulating the Use of Branchless Banking to Serve the Poor 73 Box 6. Bank-based and nonbank-based models ­ istinguish For purposes of discussing regulation and supervision, it can be useful to d two types of branchless banking. In bank-based models, the customer’s contract is customer’s with a bank or a similar licensed institution. In nonbank-based models, the ­ contract is with a nonbank service provider such as a mobile phone network operator, while a bank typically holds the e-money float. Both the bank-based and nonbank- based models rely on agents for the critical cash-in/cash-out functions. A key determinant in the development of branchless banking is the regulatory environment. Are retail outlets and other third parties allowed to be service providers ­ for customers of banks and other financial institutions, offering cash-in/cash-out services (e.g., deposits and withdrawals, although local law may define these transactions differ- ently from bank deposits and withdrawals), account opening, and other services? Is the retail payment system available (and affordable) to branchless banking providers that want to serve low-income clients? Do the AML/CFT regulations permit banks to use agents to conduct KYC and perform other compliance functions? Are there clear regu- lations and clear authority over the different actors involved in branchless banking, or are the regulations silent on who may issue electronically stored value? Are consumers protected from the particular risks posed by branchless delivery of financial services and its use of agents and other third-party actors? In many countries, industry innovation has been ahead of policy makers.104 In a few countries branchless banking has thrived because of the absence of regulation.105 But in the long run, the success and safety of branchless banking will depend on appropriate and coordinated regulation and supervision across the relevant regulatory domains, in- cluding banking, payment systems, telecommunications, and consumer protection. Some branchless banking providers (e.g., banks) will already be subject to financial regulation and supervision while others (e.g., mobile phone companies) will not. (See Box 6.) The balancing regulatory challenge is to create space for innovation and competition while ­ financial security and consumer protection. This process involves crafting regulation, 104 Some regulators have intentionally delayed the crafting of branchless banking regulations, preferring to watch the early development of the industry and to design the rules later, when the problems and benefits of the new channels are clearer. 105 In some countries without branchless banking regulation, progress in branchless banking may be partially due to avoidance of costs that regulations impose on bank transactions. In others, the main reason for branchless banking’s success is the shortcomings of other delivery channels. 74 Microfinance Consensus Guidelines monitoring its implementation and making necessary adjustments, and ensuring the ­ capacity of the various regulators as well as coordination among them. Diagnostic work during the past several years suggests that the following regulatory issues are key for developing branchless banking serving poor customers: • Using third parties to handle cash-in/cash-out functions and other customer interface tasks, such as account opening • Applying a risk-based approach to AML/CFT regulation that reflects the lower risks presented by small-value products and services • Applying proportionate regulation to nonbank payment providers and e-money issuers, with due attention to protecting customer funds • Meeting the particular consumer protection challenges raised by (i) inserting a third party between the customer and the provider of financial services and (ii) using remote communications • Having appropriate payment system regulations 5a. Agents and Other Third-Party Arrangements Key Points • Limitations on the nature and qualifications of agents and other third parties need to be crafted carefully to avoid limiting outreach to target clients. • Rules that prevent third parties from acting on behalf of a financial institution to open customers’ accounts or to handle cash typically need to be changed. • Regulation should be clear about the financial service provider’s liability for the acts of its third-party contractors. Branchless banking depends on the use of agents or other third parties to perform the direct customer interface functions—including, most importantly, taking in and ­ disbursing cash. While there must be some limits on who may act as a third party and which functions they can perform, regulators need to understand that overly tight restric- tions can seriously impede outreach to the unbanked and under-banked population. The use of pre-existing third-party actors to reduce costs and expand outreach is ­essential to any branchless banking model. Unbanked and under-banked customers need convenient means of reducing cash to electronically stored value and turning it back into cash, as well a convenient means of establishing a relationship with a trusted financial service provider. The third-party actor (e.g., a local merchant or mobile phone airtime reseller) may be an agent Regulating the Use of Branchless Banking to Serve the Poor 75 ­ypically involves the principal acting on behalf of the service provider as principal, which t being liable for all actions that the agent takes pursuant to the agency agreement. Or there may be some other type of contract: a services agreement, a partnership, a joint venture, or an alliance. The term “agent� is often used to refer to any third party that acts as the principal customer interface. (In some countries, the term “correspondent� or “facilitator� is used.) However, it is more accurate (even if cumbersome) to use the more inclusive term “third party,� because the term “agent� implies a legal relationship where the service pro- vider is more likely to be legally accountable to the customer for the acts of the third party. Third-party arrangements raise various regulatory issues.106 First, applicable rules may restrict the type of legal entity (e.g., commercial entity, nonprofit, individual, or other) permitted to act as an agent or other third party for a financial service provider. There may be criteria that a third party must meet (e.g., a business license or minimum capital). However, limitations on the nature and qualifications of agents and other third parties should be crafted carefully to allow providers some discretion on the appropriate criteria and to avoid limiting outreach to target clients. It is especially important to allow local retailers to act as third parties. Second, the rules should be clear about which services third parties can perform. Some countries prohibit banks from using third parties for core management functions, such as internal audit or loan approval; others prohibit the use of third parties to open accounts or handle cash. (Except with respect to AML/CFT compliance, there is rarely such a problem regarding agents for payment service providers and e-money issuers, given these institutions’ restricted list of permitted activities.) If regulators want to extend financial access by enabling branchless banking, rules that prevent third parties from opening accounts or handling cash typically need to be changed. This involves adopting a risk-based approach to AML/CFT rules and permitting agents (and other third parties subject to adequate bank oversight) to verify customer identity. Such third parties must be given appropriate training and technology. Third, regulation should be clear about the financial service provider’s liability for the acts of its third-party contractors. With respect to agents in particular, it is appropri- ate to hold a principal liable for its agent’s actions taken in connection with the agent’s specified responsibilities, including compliance with relevant regulations (e.g., AML/CFT requirements, consumer protection provisions). The principal—that is, the financial ser- vice provider—is often not well-positioned to prevent criminal acts by unrelated parties, and cannot be held responsible for certain risks, such as fake agents who take cash from the public without the principal’s knowledge. 106 For a discussion of regulatory issues regarding bank agents (including who can be an agent, agent qualifi- cations, and permitted agent services), see Tarazi and Breloff (2011). 76 Microfinance Consensus Guidelines 5b. AML/CFT in Branchless Banking Key Point In some countries, branchless banking cannot develop without risk-based adjust- ment of CDD and other AML/CFT rules. Branchless banking requires a risk-based approach to AML/CFT. There are two pri- mary obstacles. First, many low-income individuals cannot present the documentation normally required to establish identity. Second, many national AML/CFT regimes do not leave room for remote account opening (e.g., the customer submits data electronically, which is then verified with independent, third-party information) or for account opening by agents (e.g., CDD verification entrusted to third parties). Recordkeeping requirements can also be burdensome for agents, who often have limited storage capacity. FATF Recommendations adopt a risk-based approach, allowing for simplified mea- sures, such as lighter CDD in situations where small transactions pose a reduced risk of money laundering and terrorist financing.107 Several countries have already adapted their AML/CFT rules to the realities of low-income clients with limited access to formal documentation and allow remote transactions conducted through unsophisticated third- party retailers.108 5c. Nonbank Issuers of E-Money and Other Stored-Value Instruments Key Point Nonbank e-money issuers should be subject to appropriate regulation and supervi- sion, including liquidity and solvency-related requirements. Many countries have, or are considering, regulation of nonbank issuers of e-money and other stored-value instruments (e.g., prepaid cards). For purposes of advancing ­ financial inclusion, the basic proposition is to regulate nonbank e-money issuers propor- tionately. If these firms attract large amounts of repayable funds, it is arguable that they 107 FATF Interpretative Note to Recommendation 1. 108 See earlier discussion of risk-based approaches to AML/CFT and Isern and de Koker (2009). See also FATF (2011). Regulating the Use of Branchless Banking to Serve the Poor 77 pose systemic risk and should be prudentially regulated. But they typically have a much more limited scope of permitted activities than banks and need not be subject to the same level of prudential regulation and supervision. Most nonbank e-money issuers in developing countries are not very large, so regula- tors today are generally not concerned with systemic risk and are primarily concerned with protecting the “float� (i.e., the public’s funds held in the form of e-money by the e-money issuer).109 To mitigate the risk to customers and the financial system, regulators can restrict the use of the float so that the funds are not at substantial risk. Most commonly, the float has to be held in low-risk, liquid assets, such as government securities, or in deposit accounts at prudentially licensed institutions (referred to as “fund-safeguarding�).110 When the float is held in a bank account, it should be isolated from claims of the e-money issuer’s other creditors—e.g., by keeping the funds in a trust account for the benefit of customers, and prohibiting an issuer from using the float as collateral (referred to as “fund isolation�).111 For bank e-money issuers, there is no consensus on how they should handle float, including whether such float should be treated as “deposits,� in particular for purposes ­ of deposit insurance and reserve requirements. As noted in a 2010 consultation docu- ment of the U.K. Treasury, there are “practical difficulties and potential costs . . . of extending cover� including the costs of verifying—in the event of a bank failure—a very large number of small e-money claims. 5d. Consumer Protection in Branchless Banking Key Point Branchless banking raises a new layer of consumer protection concerns because of its use of third parties as the principal customer interface, the potential physi- cal distance between customer and provider, and often relatively inexperienced ­customers.112 109 See Lauer and Tarazi (2012). 110 There is, of course, the risk of failure of the prudentially licensed depository institutions, in which case customers’ funds could be lost. Even if there is deposit insurance coverage, the total float probably exceeds the coverage amount. One way to address such risk is for e-money issuers to place the float in different banks. In the United States, regulation permits “pass-through� insurance for prepaid card customers whose funds are commingled—i.e., the size limit for deposit insurance is measured against the size of the customer’s account rather than the size of the total float account. 111 For discussion of these measures, see Tarazi and Breloff (2010). 112 For discussion of consumer protection and branchless banking, including important issues of data security regulations, see Dias and McKee (2010). 78 Microfinance Consensus Guidelines The third party should be required to disclose that it is an intermediary so that customers know they are not dealing directly with the service provider. Clear display of pricing and fees should also be required, not only for transparency, but also to reduce the risk that customers are overcharged or steered to a particular provider if the third party represents several different providers. There should be a clear display of recourse mechanisms. Long-distance recourse can be challenging, especially for less sophisticated customers, but providers can be required to offer a simple complaint mechanism using the same technology that is used to transmit transaction details. 5e. Payment Systems: Regulation and Access Key Point competition Regulation of access to payment systems needs to balance promotion of ­ against the risk of discouraging innovation. Although regulatory attention has traditionally been on large value payment sys- tems, regulators have begun paying more attention to retail payment systems due to the large volume of transactions they handle.113 In some countries, regulators have mandated access (i.e., required payment systems to accept certain types of institutions) and have regulated maximum fees. Under certain circumstances, regulators have required partici- pants in a payment system to give competitors a specified time period to “catch up� with technological developments and join the system. While efficiency and competition may ultimately be advanced by interoperability of payment products, regulators have to ­ balance the risk of discouraging innovation. Access to some payment systems is determined by agreement among participants, who decide which institutions may have access and at what cost. Ideally, the partici- pants of a payment system will set requirements that are designed to ensure the system’s ­ integrity and not to limit competition. But the reality is that formal payment systems are typically dominated by large commercial banks that have little interest or incentive to serve the poor. Although payment service providers (e.g., money transferors) may also be able to participate indirectly in a payment system, the costs may be prohibitive. All of these access issues affect payment services for the poor. 113 For a country-by-country comparison of large value and retail transactions, see Tables II.3/a, III.1/c, and III.1/e, in World Bank (2008). Regulating the Use of Branchless Banking to Serve the Poor 79 5f. Interagency Coordination Key Point Establishing a coordination body among the various agencies regulating branchless banking can improve the chances that branchless banking will thrive. Branchless banking brings together actors from different industries—including in- dustries outside the financial sector, such as telecommunications and retail distribution— so regulation and supervision of branchless banking is likely to involve more than one agency. Ambiguities or overlaps in jurisdiction among the differing regulatory authorities can result in conflicting regulation or weak enforcement. For instance, the telecommuni- cation regulator and the relevant financial regulator might have inconsistent AML/CFT requirements, or the perspective of a central bank’s consumer protection division might differ from the perspective of a national consumer protection body with experience only in product liability. Effective coordination among these regulators can be crucial. Part VI. Regulating Microfinance Providers in Microinsurance ­ nsurers’ The primary purpose of insurance regulation historically has been to maintain the i financial solvency and soundness so they can carry out their long-term obligations to poli- cyholders and pay claims, and to guarantee the fair treatment of current and prospective policyholders and beneficiaries by both insurers and those who sell the insurers’ policies. Insurance, including microinsurance, differs from other financial services in terms of fund- ing, investment, core businesses, and risk exposures. Reflecting these differences, most countries have both dedicated laws on insurance and a dedicated insurance regulator. Insurance underwriters are normally prohibited from providing other financial services; conversely, banking and microfinance regulation typically prohibits banks and MFIs from underwriting insurance. At the same time, microfinance providers have potentially unique value in expanding access to insurance among poor households by aggregating customers into economically viable risk pools and by leveraging existing infrastructure. This Part does not attempt to capture the full range of regulatory and supervisory issues in microinsurance.114 Rather, it focuses mainly on topics specifically relevant to providers of other microfinance services when they become involved in delivering ­microinsurance. 6a. What Is Microinsurance? Defining microinsurance While legal definitions vary, insurance generally denotes a contract by which an insurer, in return for a premium, undertakes to provide specified benefits depending 114 In 2005, IAIS became the first of the global financial sectors SSBs to establish a formal mechanism to c ­ onsider financial inclusion issues, co-founding (together with the Regulation, Supervision, and Policy work- ing group of the Microinsurance Network, formerly the CGAP Working Group on Microinsurance) a joint working group on regulation and supervision of microinsurance. The Joint Working Group has published two issue papers covering a much broader range of topics than can be addressed in this Part. ­ Additionally, IAIS has formed a Financial Inclusion Subcommittee of its Implementation Committee and is expected to adopt an application framework on the topic in 2012. IAIS is also a founding partner in the Access to Insurance Initiative, a collaboration of international development agencies and insurance supervisors that ­ aims to (i) strengthen the understanding and capacity of insurance supervisors, regulators, and policy makers; (ii) facilitate their role as key drivers in expanding access to insurance markets; and (iii) support the imple- mentation of sound policy, regulatory, and supervisory frameworks consistent with international standards. 80 Regulating Microfinance Providers in Microinsurance 81 on the occurrence of specified contingencies (i.e., uncertain future events). Micro- insurance could be described as insurance with small benefits and premiums, or as risk-pooling instruments for the protection of lives and assets of low-income house- holds.115 This Guide adopts the definition used by IAIS and the Microinsurance Network, namely: “insurance that is accessed by [or accessible to] the low-income population, provided by a variety of different entities, but run in accordance with generally accepted insurance practices (which should include the [IAIS] Insurance Core Principles).�116 Does microinsurance thus defined warrant special regulatory and supervisory treat- ment? The revised Insurance Core Principles (IAIS 2011) emphasizes proportionality in assessing the observance of the principles in specific circumstances, depending on the nature, scale, and complexity of the insurer’s business. This does not imply that the supervision of microinsurance should be held to lower standards per se. Rather, similar risks should be subject to the same regulatory treatment, and the nature, scale, and complexity of risk in insurance-related activities should determine the level of supervi- sion. This approach supports a level playing field and limits opportunities for regulatory arbitrage. Regulators’ definitions of microinsurance vary, depending on local context and differing regulatory objectives.117 If the objective is to expand access to insurance by means of requirements that are easier to meet, the definition of microinsurance may specify the features of a lower risk product (e.g., short contract terms), thereby justi- 118 proportionate regulatory and supervisory response. fying a ­ A regulatory definition may also be used to designate products that qualify for subsidies or fulfill targets, or to permit some category of intermediaries to provide microinsurance subject to lower qualification or training requirements or different oversight rules. (See “6c. Regulation Microinsurance Sales.�) of ­ 115 See IAIS and the Microinsurance Network (2007). The paper analyzes the provision and regulation of microinsurance in light of the IAIS Insurance Core Principles. 116 Ibid. 117 Sometimes a definition may appear in a policy statement (and not in regulation) or may be informally adopted or applied by industry. 118 In addition to the term of the policy, other features that can impact the risk of an insurance product include (1) the size of the benefit payable under the policy (with lower benefits implying lower risk); (2) the nature of the risk event covered (some insured events happen with more predictability than others; e.g., mor- tality rates in a large population tend to be more predictable than disability or critical illness events, which tend to be less frequent and more subject to claims management and definitional uncertainties); and (3) the complexity of the product (insurance contracts with numerous options and complex features will be harder for the insurer to price correctly and it will be more difficult to set aside appropriate funds to meet future claims). 82 Microfinance Consensus Guidelines Roles in delivering microinsurance The underwriter is the entity responsible for paying benefits when the specified ­contingency happens—i.e., the party bearing the insurance risk. Insurance, including microinsurance, is generally required to be underwritten by a licensed, prudentially regulated insurer, al- though there are countries that permit institutions, including MFIs, to underwrite certain insurance without a specific license.119 customers. The underwriter of microinsurance often is not directly involved with the ­ Rather, an insurance intermediary (such as an MFI) typically manages the customer relationship, handling sales, premium collection, policy administration, claims assess- ­ ment, and settlement. Types of microinsurance products The range of microinsurance products includes credit life insurance, other life insur- ance (sometimes with savings features), funeral insurance, disability insurance, health insurance, and various kinds of property insurance, including specialized agriculture and livestock insurance. Sometimes more than one type of insurance may be bundled together as a package, such as life insurance and funeral insurance. Insurance can also be bundled with other types of financial products, as well as with nonfinancial products (e.g., agricultural inputs). (See “6c. Regulation of Microinsurance Sales— Bundling.�) 6b. MFIs and Microlending Banks in Microinsurance Because of their established relationship with the target customers of microinsurance, MFIs and banks already delivering other microfinance services to the poor often offer an attractive infrastructure for microinsurance as well. Many of these providers are moti- vated to offer microinsurance out of a desire to expand services to their clients, increase their revenues, and diversify their risks. 119 To diversify its own risk, an insurer typically “reinsures� the insurance risk it takes on. Reinsurance is available only for licensed insurers, so MFIs that underwrite without a license do not have access to it. This increases risk for both the MFI and its customers. Regulating Microfinance Providers in Microinsurance 83 MFIs as insurance underwriters Key Points • Other regulators of MFIs that underwrite or intermediate microinsurance should coordinate with the insurance regulator to avoid inconsistent regulations and regulatory arbitrage. • MFIs are generally not well-equipped to underwrite microinsurance—they may be too small to achieve sufficient pooling of risk and often lack the expertise to price the insurance products and manage the insurance risk. (Pure credit life insurance, as defined below, is an important exception.) As noted, insurance underwriting in most countries is subject to a separate body of insurance regulation and a separate insurance supervisor, and underwriters are normally prohibited from offering other financial services, so banks and depository MFIs cannot underwrite insurance. Such restrictions make policy sense. Granting loans and taking deposits require skills that are different from those needed to manage insurance risk. Moreover, regulators are wary of the possibility that income from premiums can be used to fund illiquid loans or to mask a poorly performing credit portfolio. Also, banking and insurance products can cannibalize each other—loan losses can erode the reserves needed to meet the insurance claims, and insurance losses can deplete an MFI’s capital depositors’ assets in the event of a large catastrophe. and ­ However, some MFIs (perhaps frustrated by the challenges of linking with mainstream insurance underwriters) are providing “self-underwritten� insurance products, particularly credit life. When this happens, the MFI acts as the final insurer as well as the agent to its clients. Although this model may allow for more customer-responsive ­ ­ products, it exposes the client and the typical MFI to high levels of risk. For example, most MFIs that underwrite mi- croinsurance products cannot do adequate actuarial analysis and consequently may not have adequate reserves to guarantee that claims can be paid when due. Further, MFIs are generally riskier as underwriters than insurance companies due to factors such as a small risk pool, an undiversified customer base that poses high covariant risk, and a lack of reinsurance options. The dilemma is that mainstream insurers in many countries—particularly poorer ones customers with under-developed insurance markets—are not yet interested in serving the ­ targeted by MFIs. To fill this gap in the market, policy makers in some countries allow MFIs to underwrite a special category of lower risk microinsurance with easier to meet compliance burdens. While this approach may be effective in attracting providers to 84 Microfinance Consensus Guidelines the microinsurance sphere, it can introduce opportunities for regulatory arbitrage and can distort evolving markets. To ensure separate underwriting and management of the ­ insurance risk involved, some regulators require an MFI to establish a separate legal entity and apply a proportionate approach for supervising its insurance operations. ­ MFIs as insurance intermediaries Key Point Established MFIs can be well-positioned to serve as intermediaries on behalf of mainstream insurance underwriters as the MFIs have an existing relationship and delivery infrastructure with target clients for microinsurance. In such cases, under- writers should be subject to the same regulatory treatment that they would receive without the involvement of a microfinance provider. Distribution is a major challenge for a mainstream underwriter that wants to offer microinsurance. MFIs can be attractive intermediaries because they have close contact with large numbers of low-income clients, as well as an existing delivery infrastructure (branches, loan officers, cash handling systems, and at least a simple MIS). Such advan- tages can reduce transaction costs for “reaching the last mile.� The regulatory framework can be tailored to encourage development of microinsur- ance products and to facilitate use of MFIs as insurance intermediaries. (In such cases, the underwriters should be subject to the same regulatory and supervisory treatment that they would receive without the involvement of a microlender.) MFIs can fulfill the following intermediation functions: • Sales. The MFI knows its clients and is, at least in theory, in a good position to ensure that the products being sold are appropriate to their clients’ needs and their house- hold’s capacity to pay. However, MFI staff still need proper training to sell insurance on behalf of an underwriter. • Collecting premiums and paying claims. Most MFIs have cash-in-and-out functions that can be used for microinsurance as well. Some MFIs may help smooth premium collections—e.g., a borrower group may temporarily cover the premium payment for an individual member who is short on cash.120 120 Financing of insurance premiums can be abused. Borrowing costs may be low initially, and then escalate as rates rise, with policyholders eventually owing more for the premiums than the policy is worth. Regulating Microfinance Providers in Microinsurance 85 • Policy administration. MFIs with strong processes and MIS can handle policy administration for insurers. On the other hand, MFIs without these strengths can ­ make things worse by taking on policy administration. • Claims assessment and settlement. The MFI’s close and regular interaction with its clients enables it to verify and assess claims quickly. (The MFI may have an incentive to act in its clients’ interest and help them manage the claims process, though this may create a conflict of interest when the MFI assesses claims.) However, taking on the role of intermediary requires new skills at all levels of an MFI, ­ egulation of and raises a variety of regulatory and supervisory issues for MFIs (see “6c. R Microinsurance Sales�). Microlending banks in microinsurance In all countries, regulation will place at least some conditions on the roles a conven- tional retail bank can play in either underwriting insurance or serving as an insurance intermediary. Banks, including microfinance banks, are often prohibited from any in- ­ volvement with underwriting insurance. They may also be precluded from intermediating insurance, or may be limited to intermediating only those products in which the bank has an insurable interest (e.g., only covering the outstanding value of the loan, with coverage terminating once the loan has been repaid). The special case of credit life insurance Key Points • Pure credit life insurance (as defined below) poses minor prudential risk to an MFI that underwrites it and very little risk to the insured borrower. There is an argument that such insurance—which is underwritten by the microlender itself—could be viewed as a feature of the loan contract. • For many reasons—including the inherent potential for abusive practices in credit life insurance—the insurance supervisor has a role to play, even in the case of pure credit life insurance underwritten by a microlender. • Credit life insurance on microborrowers can also be underwritten by insurance companies and can be a “gateway� microinsurance product for them. 86 Microfinance Consensus Guidelines “Pure credit life insurance,� as the term is used here, is most typically issued by a microlender and cancels a borrower’s loan balance of principal and interest in the case of death. (This is the only insurance benefit; when other benefits are combined with pure credit life insurance, the additional concerns about MFIs underwriting insurance discussed earlier are triggered.) With pure credit life insurance the MFI doesn’t have to disburse any cash, so that “insured� borrowers face no risk that their “claim� won’t be honored: the family of the deceased borrower can enforce the claim simply by not paying off the rest of the loan, as long as the family is aware of the existence of the policy. There is an argument, therefore, that pure credit life insurance should be viewed as simply one of the terms of the loan contract that should be clearly disclosed (both the pricing and the policy’s terms, including the identity and rights of the beneficiaries). Since the risk of loss is to the MFI (and the loss is minor with respect to any one borrower), it seems reasonable to consider whether the underwriting of pure credit life insurance could be supervised in a limited manner by the banking or other authority responsible for oversight over MFIs. This issue should be worked out mutually between the banking or other relevant authority and the insurance authority. As noted, these observations apply to pure credit life insurance, where the only benefit is cancellation of outstanding debt. If other benefits (e.g., funeral expenses or other cash payments) are included, then normal insurance risk is created and normal principles of insurance regu- lation should be brought to bear. Pure credit life insurance does not add significant insurance risk to microlenders’ ex- isting credit risk. But this does not mean there are no risks involved. In particular, some practices relating to credit life insurance can be problematic for consumers. Examples include, among others, the following: • Pricing obligatory credit life insurance far above the actuarial and administrative cost to the MFI, effectively using the insurance as a disguised means to raise the interest rate. • Requiring credit life insurance in circumstances where the benefit to the borrower’s household is minimal (because the MFI would not collect a dead borrower’s loan in any case). • Requiring the borrower to purchase a policy for an amount in excess of the total amount of original indebtedness (which means the insurance no longer meets the definition of “pure credit life insurance� as used here). • Providing a policy of level term insurance in an amount equal to the original indebted- ness, whereas risk of loss is measured by the amount of outstanding loan plus interest on the principal. • Extending insurance coverage beyond the term of the debt (which again means the insurance no longer meets the definition of “pure credit life insurance� as used here). Regulating Microfinance Providers in Microinsurance 87 • Lack of transparency in pricing of credit life insurance or information to borrowers accepting and their families about the existence of the insurance benefit (coupled with ­ ­ nsurance). repayment from the borrower’s family notwithstanding the existence of the i Consumer protection challenges of credit life insurance for microborrowers are present regardless of whether the microlender or a mainstream insurer underwrites the ­ insurance and holds the insurance risk. The microborrower is typically not motivated to select a microlender by comparing insurance charges. Moreover, since the typical small borrower needs the loan proceeds much more than the microlender needs the individual borrower’s business, the microlender has virtually a captive market (of course, the same is true with microloans as well, wherever the market is not competitive). Consequently, the microlender is in a position to dictate the choice of insurance coverage, premium rates, insurer, and agent.121 The microlender has little interest in setting or securing low premium rates because such premiums are the source of its own agent commissions (or underwriting profits if the MFI is also the underwriter). Thus there is a risk that the MFI will be more focused on fee generation than on value creation for its clients. There is disagreement about whether credit life insurance for microborrowers, in- cluding pure credit life insurance, should be encouraged. Some question whether the client benefit is great enough to justify the consumer protection risks. Others note the empirical experience in some countries, where credit life insurance underwritten by main- stream insurance companies (rather than MFIs) has served as a “gateway� microinsur- ance product for them, leading them ultimately to offer a range of types of coverage of clearer benefit to the beneficiaries. 6c. Regulation of Microinsurance Sales The nature of insurance (a promise to pay on the occurrence of an uncertain future event) means that insurance sales are vulnerable to abuse. Regulation of insurance sales nor- mally takes one of three forms: (1) control of the persons who can sell insurance and the extent to which they can be involved in the process of selling; (2) control of the commis- sions payable for insurance sales; and (3) in some countries, control of the sales process itself, including the information to be disclosed to customers, whether advice is provided, and if so what form it takes. 121 Some jurisdictions aim to mitigate this problem by requiring that consumers be given the option of choos- ing another credit life underwriter. Even so, consumers faced with a default option or having to source their own insurance—thereby delaying the loan—may rarely be in a practical position to exercise the right to choose an alternative provider. 88 Microfinance Consensus Guidelines Sales agents Key Point To facilitate the selling of microinsurance by MFIs, insurance regulation should permit MFIs and their employees (acting as representatives of their MFI) to be sales agents. Regulation of insurance sales typically allows payment of commissions (or other remuneration) only to registered brokers and agents. Given the limited involvement of brokers in the intermediation of microinsurance, this discussion is limited to agents.122 If regulations allow legal entities to register as insurance agents, then each type of institution found in the microfinance sector should be permitted to register as an agent.123 Regulation typically sets minimum education and training requirements for persons wishing to register as agents. In situations where MFIs or their employees can register as agents, these requirements should not be set so high that they exclude the typical profile of persons in the microfinance sales force. Lower requirements for micro- insurance agents will usually be reasonable given the simpler products and lower risks involved. There is an opportunity to leverage new technology and delivery channels, such as payments by mobile phone and other forms of branchless delivery, to expand the ­ outreach of microinsurance through microfinance providers. For large-scale adoption of new technology-driven delivery of microinsurance, regulations need to provide adequate consumer protection as well as the certainty and clarity that technology providers, such as mobile network operators, are likely to demand. 122 Agents act on behalf of an insurer, selling its products to clients and sometimes engaging in the other insurance intermediation activities noted earlier. Brokers act on behalf of the insured, advising them on the best coverage and underwriter for their particular needs. This normally involves analyzing the needs of the potential client. The process of analyzing needs and providing advice is generally too expensive to undertake for low-premium, low-margin microinsurance policies. Thus, microinsurance brokers are at most involved in brokering a master policy between an MFI and the insurer. 123 An exception occurs where banks and microfinance banks are precluded from intermediating insurance, or are limited to intermediating only those products in which the bank or MFI has an insurable interest (e.g., credit insurance that covers only the outstanding value of the loan, terminates once the loan has been ­ repaid, and includes no other benefits for the borrower’s family). See “Microlending banks in microinsur- ance.� Some such limitations may be an unnecessary barrier to microinsurance distribution, and warrant a discussion between the banking and insurance supervisors to determine which can be justified from a ­prudential perspective. Regulating Microfinance Providers in Microinsurance 89 Commissions Key Point If adequate competition exists in the microinsurance market, then commission caps should be imposed with great caution, if at all. Preferably, both the level and struc- ture of commissions should be left to the market to decide. High commissions can add to client costs and create an incentive for agents to sell peo- ple inappropriate insurance.124 Many countries impose caps or price controls on the level of commission on a commission payable on different categories of insurance policies (e.g., the ­ life policy may not exceed a certain low percent of the value of the premium). They may even go further and regulate how the commission is to be paid (e.g., on the signing of the policy contract or when the individual premiums are paid). However, commission caps—especially ­ icroinsurance, since even a small nominal com- percentage-based caps—can limit access to m mission can be a high percentage of a small premium. Insurance will reach poor customers only when compensation structures are adequate to attract agents. Also, commission caps may be evaded (e.g., by disguising commission as administrative expenses) unless regula- tions effectively cover all forms of compensation to the agent and enforcement is strong. Group sales Key Point confirms Sales of group products should be permitted, provided that each individual ­ receipt of the policy certificate and the insurer receives the names and other relevant information of each individual policyholder. To take advantage of the aggregation provided by MFIs, insurance is normally sold to MFI clients on a group basis; the risk of the group, as opposed to individual client risk, is assessed to determine the risk premium. This can leverage the MFI’s existing 124 In the insurance business, there is a common practice of rebating whereby something of value is given to sell the policy that is not provided for in the policy itself. Rebates can be made in the form of cash, gifts, services, payment of premiums, or almost anything of value. 90 Microfinance Consensus Guidelines infrastructure and client base to bring useful new insurance products to poor people. Some countries permit selling to groups only when membership in the group is closed (e.g., only employees of a particular company can become policyholders) and prohibit selling to voluntary or open groups where one can join the group solely for purposes of obtaining insurance. A requirement that the agent analyze each microfinance client before providing advice or selling a product could easily price microinsurance sales out of the low-income market.125 Fortunately, such requirements are unusual. They should be avoided unless they are essential to address particular abuses that cannot be corrected through the control of agents or appropriate disclosure. In some cases, the insurance policy is a group policy sold to an MFI, which then issues certificates to individual borrowers. The MFI is the group policyholder on behalf ­ of the individual clients, and the insurer may not have the names and contact details of individual clients. In this case, the insurer will be underwriting an agreed number of ­ persons for agreed risks, yet the individual clients will have recourse only against the MFI (and not against the insurer) unless the policy provides otherwise. A challenge arises when a group policy for credit life insurance is sold to an MFI, which then issues certificates to individual borrowers. Because the microlender purchases the policy, insurers market the product to the lender and not to the borrowers—the ­ ultimate consumers who pay for the product. Insurers may bid for the lender’s business by providing higher commissions and other compensation for the lender. Greater compe- tition for the lender’s business can lead to higher premiums for borrowers. ­ ncludes Another problem occurs when group insurance coverage is sold to a group that i ­ ither know- members who are ineligible for benefits. The lender sells the insurance policy, e ing the consumer is ineligible for benefits or not checking on this. The insurer is happy to take the premium from consumers ineligible for benefits, but when the consumer files a claim, the insurer denies the claim based on eligibility. The result of this arrangement is that MFIs and insurers keep the premiums paid by ineligible consumer who never file an insurance claim, while refusing to pay on the same policies if claims are ever filed. protection Group polices intermediated by an MFI pose challenges from a consumer ­ point of view. Therefore, regulation should require at a minimum that the MFI (i) ­produce confirmation that it has issued policy certificates to the individual policyholders, indicat- ing the name of the insurer and evidencing its liability to pay claims and (ii) communicate to the insurer the names of the policyholders, their contact details, and the names of beneficiaries. 125 Likewise, requiring individual assessment of risk (e.g., required medical check-ups) can increase costs of a simple, standardized group-priced policy. Regulating Microfinance Providers in Microinsurance 91 Disclosure, claims, consumer understanding Key Points • MFIs involved in selling microinsurance policies should be subject to minimum disclosure requirements. • Supervisors should monitor complaints, claim ratios, and lapse ratios for mi- croinsurance products sold by MFIs and take measures if there are too many complaints, the claims ratios are too low, or the lapse ratios are too high. Minimum disclosure should include the premium amount, who is underwriting the risk, the duration of coverage (and with respect to credit life insurance, any options for extending coverage beyond the loan period), commission payable, any exclusions, how to make a claim, and recourse mechanisms. However, even if regulation requires disclosure, protection of clients may be limited because they may not understand the terminology, may have trouble assessing the premium in light of the value of the covered ­ risk, and may not know enough to file a claim when a risk event occurs. As discussed, the sale of microinsurance through the credit channel is open to particular abuse, due in part to clients’ eagerness to obtain the loan (and not risk losing it while shopping for insurance). It is therefore important for supervisors to monitor compliance. In the case of MFIs that are not otherwise supervised, this can be a practical challenge. As with other financial products, microinsurance clients need protection from ­ complex and unintelligible policies. This may include requiring simplified language and standardized terms in policies, limiting the length of policy documents, or requiring a simple one-page summary of policy terms. It may also include prohibiting certain types of exclusions that clients do not normally understand. Even if products comply with such regulation, they may not necessarily provide good ­ lients’ value to the client. This could be evidenced in low claims ratios and may be due to c limited understanding of product benefits (even with simplification requirements) or a complicated claims processes. Supervisors should therefore monitor complaint levels, claims ratios, and lapse and nonrenewal ratios. If there are too many complaints, claims ratios are too low, or lapse ratios are too high, supervisors should take additional mea- sures to deal with abusive insurers. Poor households may not understand basic insurance concepts and products and be unaware of their rights (such as recourse mechanisms) and responsibilities (such as paying renewal premiums on time or properly reporting claims). Limited consumer 92 Microfinance Consensus Guidelines sophistication is a general challenge in microfinance, but it can be especially significant ­ for insurance, where the basic concepts are more abstract and further removed from the informal sector practices with which poor households may be familiar. As noted, this challenge can be addressed by mandating standardized microinsurance products with simple terms and conditions. Standardization may help consumer’s comprehension when choosing an insurance policy, as well as their capacity for comparison shopping among simplified policies. (However, excessive standardization can reduce product innovations that could benefit consumers.) It is also possible that broad-based consumer awareness campaigns may be effective. Bundling Key Points • Bundling different insurance products and bundling insurance with other finan- cial and even nonfinancial products can be in the best interest of poor clients, especially when bundling is the only way to make products financially viable. • However, when MFIs sell bundled insurance, clients may get insurance or other products that they don’t want, or pay excessive prices because the MFI has a captive clientele. • Consumer protection rules should aim to help ensure that clients clearly under- stand the bundled products and their pricing. Bundled microinsurance products combine multiple kinds of benefits in a single ­ policy. For example, a credit life policy may also include components, such as funeral insurance on the borrower or insurance on a product bought on credit, bundled into the same policy. Insurance can also be bundled with other financial products beyond just credit, as well as with nonfinancial products, such as agricultural inputs.126 Letting microfinance providers offer bundled microinsurance products may be the result, only way to make the business financially viable for a mainstream underwriter. The ­ of course, can be a broader range of useful financial services for poor people. 126 Bundling microinsurance with nonfinancial products and services raises some specific regulatory issues that are not discussed in this Guide. However, some of the issues with the bundling of microinsurance and financial services will also be relevant to bundling with nonfinancial products and services. On the other ­ hand, bundled microinsurance (especially when noninsurance products are part of the package) may lead to some poor customers buying coverage or other products they don’t need, or paying excessive prices. Regulating Microfinance Providers in Microinsurance 93 Bundling insurance and credit does raise particular consumer protection issues, however, especially when the client can’t get a loan without buying the insurance. The more types of insurance that are bundled, the more challenging these issues become. Clients may not realize that they have bought insurance or the coverage may be stated so opaquely that clients don’t really understand what they are signing up for (e.g., funeral or disability insurance when the client already has such coverage). Comparison shopping may be unlikely, but it can be impossible if clients don’t know the cost of the bundled insurance. The credit provider may have a captive insurance market and the opportunity to charge premiums that do not reflect the risks being insured. In addition, insurance policies bundled with credit typically expire when the credit is repaid. While this makes sense in the case of credit life insurance, it does not make sense for other types of insurance coverage (e.g., funeral insurance) if the client cannot continue it after repaying the debt. When such insurance is tied to having an active loan, it could encourage clients to borrow too often—in theory at least. If bundling is permitted, consumer protection rules should aim to help ensure that clients clearly understand bundled products and their pricing and benefits. Outright pro- hibition of bundling could be another option, though it is a very blunt tool and, as observed, may result in the clients of the MFI or bank microlender not having access to useful insurance coverage. Part VII. Summary of Key Observations, Principles, and ­Recommendations 7a. General Issues • To craft and enforce appropriate regulation with a financial inclusion objective, regu- lators need to understand the distinctive characteristics of microfinance, including clients and their needs, products and services, and the institutions providing them. (See p. 8.) • Problems often arise due to inadequate coordination among financial regulators and other government agencies whose responsibilities may affect institutions delivering microfinance. (See pp. 8–9.) • Regulation creates costs for both the regulated institutions and the supervisor. These costs should be proportionate to the risks involved. (See p. 8.) • To the extent possible, regulation should aim to be institution-neutral (support- ing an activity-based regulatory approach), both to create a level playing field that fos- ters competition and to reduce risk of regulatory arbitrage. (See p. 9.) possibility • In creating new windows for microfinance, regulators need to be alert to the ­ of regulatory arbitrage. Some countries create special microfinance windows with one sort of activity in mind, and then they are surprised to find that the window is also being used for other activities that the regulators might not have been so keen to promote. (See p. 11.) ­ Regulatory definitions • Regulatory definitions of “microfinance� and “microcredit� should be tightly framed to meet specific regulatory objectives and should not simply be drawn from general literature on microfinance. (See pp. 11–14.) • Absent extraordinary circumstances, nondepository MFIs should not be subjected to prudential regulation and supervision. (See pp. 14–18.) 94 Summary of Key Observations, Principles, and Recommendations 95 Regulatory “windows� for depository MFIs window • Policy makers and regulators should assess the market before creating a new ­ for depository MFIs. In most countries where a new microfinance window has been successful in expanding outreach, there was already a critical mass of profitable credit- only MFIs before the special window was opened. (See pp. 19–21.) • If existing regulation does not present a barrier to the formation or operation of an institution providing savings or other financial services to the poor, or if the binding ­ constraint lies elsewhere (such as a shortage of skilled managers and interested inves- tors), then a new window will not necessarily improve access. (See pp. 19–21.) 7b. Prudential Regulation Rationing prudential regulation and minimum capital • Minimum capital should, in principle at least, be set high enough to ensure that the institution can cover the infrastructure, MIS, and start-up losses to reach a viable scale. Minimum capital should also provide incentives for adequate performance and continued operation. (See p. 22.) • In creating a new window for depository microfinance, it is important to assess supervisory capacity and set the minimum capital requirement high enough to avoid ­ overburdening the supervisor. (See pp. 21–22.) • Where possible, it is usually preferable to set minimum capital through regulation rather than legislation, but the regulator needs to be clear in its communications with the market to avoid a perception that the regulator is unpredictable. Primary among other advantages, setting the requirement through regulation makes it easier supervisors who are new to microfinance to start with a manageable number of for ­ new licensees, reserving the option of reducing minimum capital and licensing more institutions as experience is gained. (See p. 22.) Adjusting prudential standards for microfinance • Regulation—including any proposed new regulation that provides for depository microfinance—should clearly define the types of permissible activities that a pruden- ­ tially regulated institution may engage in. (See p. 23.) 96 Microfinance Consensus Guidelines • There are strong arguments (and recent experiences) that support the imposition of higher capital adequacy standards for specialized depository MFIs than for banks. (See pp. 23–25.) • A microloan portfolio should not be limited to a specified percentage of the lender’s equity nor burdened with a high general provision requirement simply because the loans are not conventionally collateralized. (See pp. 26–27.) • Absent special circumstances, performing microloans should have the same provision requirement as other loan categories that are not particularly risky. However, the provisioning schedule for delinquent microloans that are uncollateralized should be ­ more aggressive than the provisioning schedule for secured bank loans. (See p. 27.) • Boards of deposit-taking MFIs should be independent of management and should include members with experience in finance and banking, as well as members who understand the clients well. (See p. 28.) • Specialized MFIs may need higher, rather than lower, liquidity requirements. (See pp. 28–29.) • MFIs should not borrow or transact in foreign currency without having the capacity to assess and manage currency risk. (See p. 29.) • Given the size of microloans and the nature of the borrowers, loan documentation requirements need to be lighter for microcredit than for conventional bank lending. (See p. 30.) • Prohibitions against co-signers as borrowers may need to be relaxed. (See p. 31.) • Branching requirements should be re-examined, but not necessarily eliminated, for microfinance. (See p. 31.) • The content and frequency of reports should enable supervisors to conduct the analyses needed for effective supervision of a depository MFI. However, regulation ­ must also consider the circumstances of its supervised institutions, which may not be able to comply with some requirements applicable to banks. (See pp. 31–32.) • With respect to MFIs that are not member-owned and that receive favorable ­regulatory reason for any in- treatment because of their focus on poor clients, it is hard to see a ­ sider lending except perhaps small welfare loans to employees. (See pp. 32–33.) To whom should special prudential standards apply? • When creating new regimes for depository microfinance, regulators should take care that full-service banks and other financial institutions (not just MFIs) are enabled to provide microfinance services. (See pp. 33–34.) Summary of Key Observations, Principles, and Recommendations 97 Transformation of NGO MFIs into licensed intermediaries accept • To facilitate transformation of NGO MFIs into for-profit companies licensed to ­ retail deposits, regulators may want to consider temporary or permanent adjustment of certain prudential requirements. (See pp. 34–37.) Deposit insurance • If a country requires commercial banks to participate in a deposit insurance scheme, then it may wish to consider imposing the same requirement on prudentially super- vised deposit-taking MFIs as well (including at least larger financial cooperatives). (See pp. 37–38.) 7c. Prudential Supervision • Assessing microcredit risk requires specialized examiner skills and techniques that differ substantially from the ones that supervisors use for conventional retail bank portfolios. (See pp. 39–40.) • Some supervisory tools—such as capital calls and forced asset sales and mergers— work less well for MFIs than for conventional retail banks, and some, such as stop- lending orders, may be counter-productive. (See pp. 40–41.) • In most cases the best supervisor for depository microfinance will be the authority responsible for commercial banks. (See pp. 42–43.) • It is not easy to delegate supervision effectively. The principal supervisor should closely monitor the work of the delegated entity. (See pp. 43–44.) • True self-regulation and self-supervision is almost always a gamble against very long odds. (See pp. 44–45.) • Financial cooperatives—at least the larger ones—need prudential supervision by a specialized financial oversight agency that has the requisite skill, independence, resources, and powers. (See pp. 45–46.) ­ • In some cases, the best solution may be to allow formal but very small member-based deposit-takers to continue operating even though they cannot be effectively super- vised. (See pp. 47–48.) • Weak supervision can be worse than no supervision if it leads depositors to expect levels of protection that cannot in fact be delivered. (See pp. 47–48.) 98 Microfinance Consensus Guidelines 7d. Nonprudential Regulation Permission to lend • The regulatory framework should—absent particular local factors, such as extreme corruption in the NGO sector—permit both NGOs and commercial companies to engage in microlending. (See p. 50.) • Issuance of a permit to engage in microlending should be straightforward, involving a public registry and a simple process, but not prudential regulation. (See p. 50.) Reporting • If regular reporting is required of lending-only MFIs, then the content and frequency of reports should be tailored to specific regulatory purposes and should be much lighter requirements than prudential reporting by deposit-takers would be. In addition, the ­ should be harmonized as much as possible with reporting requirements imposed by other regulatory authorities (e.g., the regulator of NGOs). (See p. 51.) Consumer protection • As much as possible, all providers of a given financial service should be held to the same consumer protection standards. (See pp. 52–53.) • Microfinance providers should be required to give clients clear and complete informa- tion about services offered, including their terms and costs. However, with respect to microloans, the standard APR may not be the most effective way to communicate costs to low-income borrowers. (See pp. 53–54.) • Regulation should prohibit discrimination—whether against women or a particular race, caste, religion, or ethnic minority. (See pp. 54–55.) predatory� • Regulation should address aggressive or coercive sales practices as well as “­ lending designed to take advantage of borrowers’ lack of education or experience. (See p. 55.) • Restraints on abusive collection practice may be needed, but care is required in defin- ing what is abusive. (See p. 55.) • Interest rate caps can restrict access by making it impossible to serve small or remote borrowers. It may be politically difficult to set a cap that is high enough to cover the unavoidable costs of microlending and a profit margin high enough to attract capital to low-income financial services. (See p. 57.) Summary of Key Observations, Principles, and Recommendations 99 • Effective protection of microfinance clients’ privacy—so that information about them is collected, stored, viewed, and used only in proper ways by designated people— is an easy objective to articulate, but difficult to implement in practice, especially ­ ountries where diverse providers serve the same poor clients, but are subject to in c ­ varying privacy related regulation. (See p. 58.) • Regulation on bank secrecy and general regulation on the use of personal data protect client privacy, but they can also hinder effective credit reporting for poor customers, making it harder to manage over-indebtedness risk. (See p. 58.) • The consumer’s ability to lodge complaints and seek redress is an important part of financial consumer protection. For most microfinance consumers, judicial recourse will not be a viable option for many reasons (including expense and time), so the focus needs to be on alternatives. (See pp. 59–60.) Credit reporting systems • It is critical for the healthy development of microfinance to foster the development of broad and deep credit information databases that include current loan balances and negative and positive information on the past payment behavior of poor customers, particularly in markets approaching saturation. (See pp. 61–63.) • Microlenders (of whatever legal form) and borrowers are better served by credit reporting that draws from comprehensive payment data rather than just microcredit ­ data. (See p. 62.) Limitations on ownership, management, and capital structure • Restraints on foreign investment or management can sometimes hinder the develop- ment of financial services for the poor. (See p. 63.) • NGOs should be permitted to own shares in for-profit MFIs that specifically target the poor. (See p. 63.) NGO transformations into for-profit companies • Creating a clear legal path for NGO MFI transformations can be an important enabling reform, and may involve changes in nonfinancial laws and regulations—­ particularly the legal framework for NGOs. (See pp. 64–65.) 100 Microfinance Consensus Guidelines • Regulators should address the risk of asset-stripping presented in the transformation of a public-benefit NGO into a privately owned institution. (See p. 65.) Secured transactions • Legal and judicial reform to support secured transactions—in particular, a collateral law and accessible collateral registries—may facilitate microfinance, although typical microcredit is effectively unsecured. (See p. 66.) Financial crime • Applying the AML/CFT rules for conventional banking to tiny microfinance trans- actions can seriously limit access unless a risk-based approach is adopted. (See pp. 66–68.) • In many countries, the existing anti-fraud and financial crime regulation may be ade­ quate to address abuse connected to the provision of microfinance services, or may need amendment only to add any new categories of institution to the regulatory landscape. Often the most pressing need is to improve enforcement of existing laws. (See pp. 68–69.) • In most countries, identity fraud and theft are recognized as crimes. The focus should be on improved enforcement of the existing criminal sanctions and of other data pro- tection and privacy measures. (See pp. 69–70.) Tax treatment of microfinance • Favorable transaction tax treatment should be based on the type of activity or transaction, regardless of the nature of the institution and whether it is prudentially ­ licensed. (See pp. 70–71.) • It is reasonable to argue that NGO MFIs should be treated the same as other public- benefit NGOs when the tax in question is a tax on net profits. (See p. 71.) • Profit-tax deductions for expenses that are available to banks should also be available to other providers of similar services. (See p. 71.) 7e. Branchless Banking • Branchless banking—using technologies such as mobile phones and smartcard readers to transmit transaction details, and using existing retail establishments to act as the Summary of Key Observations, Principles, and Recommendations 101 principal customer interface—holds the promise of significantly expanding financial access by lowering transaction costs for the lender and improving convenience for the customer. (See pp. 72–73.) • A suitable regulatory framework for branchless banking should include (i) condi- tions for banks’ and nonbanks’ use of agents or other third parties as a customer regime interface; (ii) a flexible, risk-based AML/CFT regime; (iii) a clear regulatory ­ for nonbanks to issue electronically stored value; (iv) consumer protection tailored to the branchless context; and (v) payments system regulation that allows (at least in the long term) broad interoperability and interconnection. (See pp. 73–74.) • Limitations on the nature and qualifications of agents and other third parties need to be crafted carefully to avoid limiting outreach to target clients. (See pp. 74–75.) • Rules that prevent third parties from acting on behalf of a financial institution to open customers’ accounts or to handle cash typically need to be changed. (See p. 75.) • Regulation should be clear about the financial service provider’s liability for the acts of its third-party contractors. (See p. 75.) • In some countries, branchless banking cannot develop without risk-based adjustment of CDD and other AML/CFT rules. (See p. 76.) • Nonbank e-money issuers should be subject to appropriate regulation and supervi- sion, including liquidity and solvency-related requirements. (See pp. 76–77.) • Branchless banking raises a new layer of consumer protection concerns because of its use of third parties as the principal customer interface, the potential physical distance between customer and provider, and often relatively inexperienced customers. (See p. 78.) • Regulation of access to payment systems needs to balance promotion of competition against the risk of discouraging innovation. (See p. 78.) • Establishing a coordination body among the various agencies regulating branchless banking can improve the chances that branchless banking will thrive. (See p. 79.) 7f. Microinsurance • Other regulators of MFIs that underwrite or intermediate microinsurance should coordinate with the insurance regulator to avoid inconsistent regulations and regula- ­ tory arbitrage. (See pp. 83, 86.) • MFIs are generally not well equipped to underwrite microinsurance—they may be too small to achieve sufficient pooling of risk and often lack the expertise to price the insurance products and manage the insurance risk. (Pure credit life insurance, which cancels a borrower’s loan balance in the case of death, but provides no other insurance benefit, is an important exception.) (See pp. 83–84.) ­ 102 Microfinance Consensus Guidelines • Established MFIs can be well-positioned to serve as intermediaries on behalf of main- delivery stream insurance underwriters as the MFIs have an existing relationship and ­ infrastructure with target clients for microinsurance. In such cases, underwriters should be subject to the same regulatory treatment that they would receive without the involvement of a microfinance provider. (See pp. 84–85.) • Pure credit life insurance—which cancels a borrower’s loan balance in the case of death, but provides no other insurance benefit—poses minor prudential risk to an MFI that argument that underwrites it and very little risk to the insured borrower. There is an ­ such insurance—which is underwritten by the microlender itself—could be viewed as a feature of the loan contract. (See p. 86.) • For many reasons—including the inherent potential for abusive practices in credit life insurance—the insurance supervisor has a role to play, even in the case of pure credit life insurance underwritten by a microlender. (See pp. 86–87.) • Credit life insurance on microborrowers can also be underwritten by insurance com- panies and can be a “gateway� microinsurance product for them. (See p. 87.) ­ ermit • To facilitate the selling of microinsurance by MFIs, insurance regulation should p MFIs and their employees (acting as representatives of their MFI) to be sales agents. (See pp. 87–88.) • If adequate competition exists in the microinsurance market, then commission caps should be imposed with great caution, if at all. Preferably, both the level and structure of commissions should be left to the market to decide. (See p. 89.) • Sales of group products should be permitted, provided that each individual confirms receipt of the policy certificate and the insurer receives the names and other relevant information of each individual policyholder. (See pp. 89–90.) • MFIs involved in selling microinsurance policies should be subject to minimum disclo- sure requirements. (See p. 91.) • Supervisors should monitor complaints, claim ratios, and lapse ratios for microinsur- ance products sold by MFIs and take measures if there are too many complaints, the claims ratios are too low, or the lapse ratios are too high. (See pp. 91–92) • Bundling different insurance products and bundling insurance with other financial and even nonfinancial products can be in the best interest of poor clients, especially when bundling is the only way to make products financially viable. (See p. 92.) • However, when MFIs sell bundled insurance, clients may get insurance or other ­ products that they don’t want, or pay excessive prices because the MFI has a captive clientele. (See p. 93.) ­ • Consumer protection rules should aim to help ensure that clients clearly understand the bundled products and their pricing. (See p. 93.) Appendix A. Authorship and Acknowldgments Boulder The principal writers of this Guide are Robert Peck Christen, president of the ­ Institute of Microfinance; Kate Lauer, policy advisory consultant to CGAP; Timothy ­ R. Lyman, senior policy adviser at CGAP and chief of CGAP’s Government and Policy Team; and Richard Rosenberg, senior adviser at CGAP. The consultative process for this publication included (i) a review by an expert panel of financial regulators and supervisors with considerable experience in the regu- lation and supervision of microfinance, (ii) exchanges with others regulators during the programs of the Boulder Institute of Microfinance, and (iii) a public consultation. The expert panel of financial regulators and supervisors included Carlos Gerardo Ace- ­ hatterjee; Nestor A. Espenilla, Jr.; Abdul Haseeb; Oleg Ivanov; A. vedo Flores; Arup C ­ ­ arlos Lopez Moctezuma; Matu Mugo; Narda Sotomayor Valenzuela; Kanagalingam; C Abdybaly Suerkul; and other central bankers and regulators. Neither the regulators consulted nor their institutions have endorsed all the contents of this Guide. However, based on our consultations with these regulators and supervisors, we are confident that the positions taken in the Guide are consistent with the views of the large majority of those consulted. This Guide draws on research and writings from many institutions, including the Alliance for Financial Inclusion, the Asian Development Bank, the Basel Committee on Banking Supervision at the Bank for International Settlements, the Center for Global Development, the Financial Access Initiative, the G-20 Global Partnership for Financial Inclusion and its predecessor, the G-20 Financial Institutions Expert Group, the Inter- American Development Bank, the International Association of Insurance Supervisors, the World Bank, the World Council of Credit Unions, World Savings Bank Institute, our colleagues at CGAP, of course, and many others. We would also like to recognize and thank the experts who produced or contributed substantially to initial draft text on specific topics: • Microinsurance—Hennie Bester (Centre for Financial Regulation and Inclusion [CENFRI]), Doubell Chamberlain (CENFRI), and Christine Hougaard (CENFRI) • Consumer Protection and Box 3: Basel Core Principles and Depository Microfinance— Laura Brix (at the time of drafting, consultant to CGAP; currently with FDIC) • Deposit Insurance—Laura Brix and Jasmina Glisovic (CGAP) 103 104 Microfinance Consensus Guidelines • Data Privacy and Security—Denise Dias (CGAP consultant) • Financial Crimes—Louis de Koker (University of Deakin) • CGAP colleagues and consultants who provided major critical review of specific ­sections—Branchless Banking, Michael Tarazi; Branchless Banking—Consumer Pro- tection, Denise Dias; Supervision of Financial Cooperatives, Denise Dias, and Laurent Lhèriau; Reporting and Institutional Transparency, Djibril Maguette Mbengue; and Credit Reporting Systems, Margaret Miller. We sought and received important insight, opinions, and feedback from Karla Brom (consultant to CGAP), Greg Chen (CGAP), Carlos Cuevas (Bill & Melinda Foundation), Frederique Dahan (EBRD), Xavier Faz (CGAP), Mark Flaming Gates ­ ­ (consultant to CGAP), Adrian Gonzalez (MIX), David Grace (independent consultant, formerly with WOCCU), Kathryn Imboden (consultant to CGAP), Dean Karlan (Yale University, ­ ­ Innovations for Poverty Action), Kate McKee (CGAP), Rafe Mazer (CGAP), Rasmussen (CGAP), Xavier Reille (IFC, formerly of CGAP), and Stefan Staschen Steve ­ (consultant to CGAP). ­ No formal endorsement of this Guide was sought from any party, nor should any be inferred from participation in the consultative process by which it was developed. CGAP is solely responsible for its content. Appendix B. Glossary Varying terminology used in the discussion of microfinance regulation sometimes leads to confusion. This Guide uses the following general definitions: Anti-money laundering and combating the financing of terrorism (AML/CFT). Legal re- quirements, controls, and practices designed to detect and prevent money-laundering, the financing of terrorism, and other illicit activities. Branchless banking. The delivery of financial services outside conventional bank branches, often using third parties (such as small retailers) and relying on information and commu- ­ obile phones). nications technologies (such as card-reading, point-of-sale terminals and m Common microlending methodology. Lending approaches applied over the past four decades involving most, but not necessarily all, of the following: • The lender’s personal contact with the borrower • Group lending or individual lending based on an analysis of the borrower’s (or ­ borrower’s household) cash flow as opposed to scoring • Low initial loan sizes, with gradually larger amounts available in subsequent loans • An understanding that borrowers who repay their loans faithfully will have prompt access to follow-on loans • A “compulsory savings� requirement that must be satisfied by the borrower before receiving the loan to demonstrate the borrower’s willingness and ability to make payments and/or to provide a partial “cash collateral� for the loan Compulsory savings (also referred to as forced savings, obligatory savings, or compen- sating balances). Savings that many MFIs (often lending-only institutions) require of their borrowers, both to demonstrate the borrower’s ability to make payments and to serve as partial security for the repayment of the loan. The cash is posted by the borrower with the MFI and sometimes deposited by the MFI at a commercial bank in an account (sometimes a trust account). If the funds are held in trust, they cannot be intermediated by the MFI. If the savings are intermingled with the MFIs’ funds, then the MFI effectively uses the funds for its lending operations. Credit bureau. A private agency or firm, established either as a profit-making venture by en- ­ ssociation trepreneurs (with or without financial institution owners) or as a cooperative a information. This by a group of lenders, that gathers and provides consumer credit ­ 105 106 Microfinance Consensus Guidelines information can be used to assess an individual’s creditworthiness and other factors important to a lender when determining whether to grant a loan. The term “credit ­ bureau� can also be used to refer to a public credit registry (defined below). Credit registry. A database maintained by a government agency (e.g., the central bank) to which regulated financial institutions are typically required to submit loan and repay- ­ ccess ment information. In many countries, only regulated financial institutions can a information from a public credit registry. Customer due diligence (CDD). Requirements imposed on banks and other financial insti- tutions by regulation. FATF has a specific Recommendation on CDD setting forth what risk-based financial institutions should be required by regulation to do (subject to the ­ approach), including (i) identifying the customer and verifying that customer’s identity, (ii) identifying the beneficial owner, (iii) understanding the nature of the business relation- ship, and (iv) conducting ongoing due diligence on the business relationship. Similar (and sometimes identical) to “know-your-customer� requirements (see definition below). Delegated regulation/supervision. Regulation or supervision that is outsourced by a pri- federation mary prudential regulatory and supervisory body to another body, such as a ­ of retail institutions. Typically, the delegating body retains responsibility for the per- formance of the body to which regulation or supervision is delegated. E-money. Monetary value represented by a claim on the issuer that is (i) stored on an electronic device, (ii) issued on receipt of funds of an amount not less in value than the monetary value issued, (iii) accepted as a means of payment by parties other than issuer, and (iv) convertible into cash. In practice, the customer exchanges cash at the ­ a retail agent in return for an electronic record of value. Financial cooperative. A member-owned financial intermediary, such as a savings and credit cooperative, credit union, or cooperative bank. Members share an economic stake in the principle— outcome of cooperative’s operations and govern by a “one member, one vote� ­ that is, each member of a financial co-op has one vote regardless of the amount of money that he or she has invested. Financial cooperatives typically engage in both lending and deposit-taking, with members’ money (from membership shares and deposits) typically funding all or most of the co-op’s lending activity. Financial cooperatives can be “stand alone� financial institutions or can be organized into federations, with the federation of- ten exercising critical functions, such as liquidity management on behalf of its members. Financial intermediation. The process of accepting repayable funds (such as funds from deposits or other borrowing) and using these to make loans or similar investments. Fit and proper (specifically in the context of financial regulation). A minimum set of requirements and/or competencies applicable to those individuals with a controlling interest in a financial institution, as well as members of its senior management and governing board. Requirements often include the absence of a criminal record and ­ Glossary 107 personal bankruptcy and (particular with respect to senior management and board members) prior professional experience with a depository institution. Greenfield institution. A newly established institution. Interoperability. The ability of firms to use electronic communications systems to share and ­ essaging system. exchange information and services, normally by means of a common m Know your customer (KYC). Due diligence (sometimes referred to as customer due diligence [CDD]) that banks are typically required (pursuant to prudential requirements, AML/CFT requirements, and also internal guidelines) to perform on potential customers to ascertain and verify the identity of a client. Common KYC requirements include the provision of national identification cards and documentary proof of home address and employment. Microcredit. Small-scale credit typically provided to self-employed or informally ­ employed poor and low-income individuals and microenterprises. Other common features of microcredit include lending methodology characterized by familiarity with the borrower, lack of collateral, expectation of a follow-on loan, and very small loan amounts (although the size of microcredit loans varies from country to country.) See “Common microlending methodology.� Microfinance. The provision of formal financial services to poor and low-income people and those systemically excluded from the formal financial system. Microfinance institution (MFI). A formal (i.e., legally registered) entity whose primary activity is microfinance. National payments system. A country’s institutional and infrastructure arrangements and processes for making payments (specifically by commercial banks and the central bank). Nongovernmental organization (NGO). An institution that does not have “owners� (in the sense of parties with an economic stake in the outcome of the entity’s operations) and has one or more enumerated public benefit purposes, as stated in its constituent documents and often as required by law. Because there are no owners to elect it, an NGO’s governing body may be self-perpetuating (i.e., the body chooses its own suc- cessors) or it may be chosen by third parties, such as a general assembly of members or founders. The capital structure of an NGO is distinguishable from other institu- tional types because an NGO’s initial equity base is typically grant-funded, and it can’t raise additional equity by issuing shares or otherwise bringing in new owners. The only means of raising funds is through borrowings, grants and other donations, and retained earnings. In most regulatory systems, NGO MFIs are not permitted to mobilize voluntary savings from retail customers, so an overwhelming majority are microlending-only organizations. Over-indebtedness. There is no single, commonly agreed definition for over-indebted- ness. Some of the more widely accepted indicators of over-indebtedness include con- sistently poor repayment rates over a period of time (generally a lagging indicator), 108 Microfinance Consensus Guidelines high ratios of debt-service-to-income or debt-to-assets, and inability to make loan payments without extreme family or personal hardship. Payment system. A funds transfer system with formal standardized arrangements and common rules for processing, clearing, and/or settling payment transactions. Proportionate approach. An approach to regulation and supervision in which the costs should not be excessive when measured against the risks being addressed and the benefits that should result. Prudential (regulation or supervision). Regulation or supervision that governs the fi- nancial soundness of licensed intermediaries’ businesses, to prevent financial-system instability and losses to small, unsophisticated depositors. ­ Regulation. Binding rules governing the conduct of legal entities and individuals, whether they are adopted by a legislative body (laws) or an executive body (regulations). Regulations. The subset of regulation adopted by an executive body, such as a ministry or a central bank. Retail payment system. A payment system, such as an automated clearing house or a payment card scheme, that processes retail payment instruments. Risk-based approach (RBA). An approach to implementing AML/CFT measures—as specified in FATF Recommendation 1—that requires countries to identify, assess, and understand the money laundering and terrorist financing risks for the country and ensure that measures to prevent money laundering and terrorist financing are propor- tionate to the risks identified. Self-regulation/supervision. Regulation or supervision by a body that is effectively con- trolled by the entities being regulated or supervised (whether in law or in fact). Supervision. External oversight and engagement aimed at determining and enforcing compliance with regulation. Transformation. A change of an MFI’s business from one organizational type to another. The most common type of transformation is from an NGO MFI into a new or previ- ously existing shareholder- or member-owned company (Newco). Such a transforma- tion is typically effected via the transfer by the NGO of all or part of its loan portfolio and other assets, liabilities, and employees to Newco in exchange for shares in Newco or payment by Newco or its other shareholders or founders in the form of cash, debt, or a combination thereof. In some instances, Newco may be a bank or other form of depository MFI. Other types of transformation include (i) a for-profit lender becom- ing a deposit-taking institution, (ii) a member-based organization transferring its as- sets to a licensed financial institution (with a similar exchange as noted), and (iii) an NGO transforming into a member-based organization. Appendix C. Microlending Institutions and Their Funding Sources Nondepository microlending institutions fund their lending from various sources other than public deposits. Donor funds. Historically, many MFIs have been supported, at least in their start-up phase, by grant funding from donors, including bilateral and multilateral development agencies, international NGOs, foundations, and private benefactors. Some donors sup- port MFIs through loans with below-market interest rates. Compulsory savings and other cash collateral. Many MFIs require their borrowers to deposit cash with the institution (both before and during the period of a loan) to demon- strate the borrower’s ability to make payments and to serve as security for the repayment of the loan. This cash collateral is sometimes held by a third party (such as a commercial bank) in a trust account; in those situations it is not intermediated by the MFI. In many countries, a nondepository MFI that intermediates compulsory savings would be in viola- tion of the banking law. Members’ savings and other repayable funds. In some countries, most microfinance is provided by financial cooperatives that typically fund their lending largely or entirely from their own members’ redeemable share capital and savings. Semi-commercial and commercial loans and equity. Many MFIs now get a substantial portion of their funding from loans and (in the case of MFIs formed as companies) equity investments by investors whose interest rates and other conditions are (or are close to) microfinance purely commercial investment terms. Such investors include international ­ investment vehicles, other social-purpose institutional investors, commercial banks, and ordinary investors with no social objectives. In some countries, MFIs can issue commer- cial paper, bonds, or similar instruments in the local securities markets. 109 Appendix D. Bibliography Abrams, Julie, and Jerome Prieur. 2011. “Risky Business: An Empirical Analysis of Foreign Exchange Risk Exposure in Microfinance.� Microfinance Currency Risk ­ Solutions and FMO (Finance for Development). Access through Innovation Sub-Group of the G-20 Financial Inclusion Experts Group. 2010. “Principles and Report on Innovative Financial Inclusion.� Seoul: G-20, September. Economic Adam, Michael, and S. Alder. 2011. “Abuse of Dominance and Its Effects on ­ Development.� In Eleanor M. Fox and Abel M. Mateus, eds., Economic Development: The Critical Role of Competition Law and Policy. Northampton, ­ Mass.: Edward Elgar Publishing. Alexandre, Claire, Ignacio Mas, and Dan Radcliffe. 2010. “Regulating New Banking ­ elinda Models That Can Bring Financial Services to All.� Seattle, Wash.: Bill & M Gates Foundation. Anand, Malika, and Richard Rosenberg. 2008. “Are We Overestimating Demand for Microloans?� Brief. Washington, D.C.: CGAP, April. Apgar, David, and Xavier Reille. 2010. “Microfinance Foreign Exchange Facilities: Per- formance and Prospects.� Occasional Paper 17. Washington, D.C.: CGAP, April. Ariely, Dan. 2009. Predictably Irrational. New York: Harper Collins. Armendáriz de Aghion, Beatriz, and J. Morduch. 2005. The Economics of Microfinance. Cambridge: MIT Press. Barr, Michael S., Sendhil Mullainathan, and Eldar Shafir. 2008. “Behaviorally Informed Financial Services Regulation.� Washington, D.C.: New America Foundation, October. Barth, J. R., Gerald Caprio, and R. Levine. 2006. Rethinking Bank Regulations: Till Angels Govern. Cambridge: Cambridge University Press. BCBS (Basel Committee on Banking Supervision). 1997. “Core Principles for Effective Banking Supervision.� Basel: Bank for International Settlements (BIS), September. ———. 1998. “International Convergence of Capital Measurement and Capital Stan- dards.� Basel: BIS. ———. 2004. “International Convergence of Capital Measurement and Capital Standards: A Revised Framework.� Basel: BIS. 110 Bibliography 111 ———. 2006a. “Core Principles for Effective Banking Supervision.� Basel: BIS, October. ———. 2006b. “Core Principles Methodology.� Basel: BIS, October. ———. 2010. “Microfinance Activities and the Core Principles for Effective Banking Supervision.� Basel: BIS, August. ———. 2011a. “Operational Risk—Supervisory Guidelines for the Advanced Measure- ment Approaches.� Basel: BIS, June. ———. 2011b. “Principles for the Sound Management of Operational Risk.� Basel: BIS, June. ­ eptember. ———. 2012. “Core Principles for Effective Banking Supervision.� Basel: BIS, S BCBS and IADI (International Association of Deposit Insurers). June 2009. “Core Principles for Effective Deposit Insurance Systems.� Basel: BIS. ­ ———. 2005. “Outsourcing in Financial Services.� The Joint Forum. Basel: BIS, October. Beck, Thorsten, Samuel Munzele Maimbo, Issa Faye, and Thouraya Triki. 2011. Financing Africa: Through the Crisis and Beyond. Tunis: Making Finance Work ­ for Africa Secretariat, African Development Bank. Belsky, Eric, and Ren Essene. 2008. “Consumer and Mortgage Credit at a Crossroads: ­Preserving Expanded Access while Informing Choices and Protecting ­Consumers.� Cambridge, Mass.: Joint Center for Housing Studies. Harvard University. Bertrand, Marianne, and Adair Morse. 2010. “Information Disclosure, Cognitive Biases and Payday Borrowing.� Working Paper. Chicago: University of Chicago. Bester, H., D. Chamberlain, L. de Koker, C. Hougaard, R. Short, A. Smith, and R. Walker. 2008. Implementing FATF Standards in Developing Countries and F ­ inancial Inclusion: Findings and Guidelines. The FIRST Initiative. Washington, D.C.: World Bank. Bester, H., D. Chamberlain, and C. Hougaard. 2008. “Making Insurance Markets Work for the Poor: Policy, Regulation and Supervision: Evidence from Five Country Case Studies.� Cape Town, South Africa: Cenfri. Bradford, Terri, Matt Davies, and Stuart E. Weiner. 2003. “Nonbanks in the Payments System.� Kansas City, Mo.: Federal Reserve Bank of Kansas City. Brix, Laura, and Katharine McKee. 2010. “Consumer Protection Regulation in Low- Access Environments: Opportunities to Promote Responsible Finance.� Focus Note 60. Washington, D.C.: CGAP, February. Brom, Karla. 2009. “Asset and Liability Management for Deposit-Taking Microfinance Institutions.� Focus Note 55. Washington, D.C.: CGAP, June. Campion, Anita, Rashmi Kiran Ekka, and Mark Wenner. 2010. “Interest Rates and Implications for Microfinance in Latin America and the Caribbean.� IDB Work- ­ ing Paper 177. Washington, D.C.: Inter-American Development Bank. 112 Microfinance Consensus Guidelines Caprio, Gerard, Jr., P. Honohan, and D. Vittas. 2002. Financial Sector Policy for Devel- oping Countries: A Reader. Washington, D.C.: World Bank. Caprio, Gerard, Jr., Jonathan L. Fiechter, Robert E. Litan, and Michael Pomerleano, eds. 2004. The Future of State-Owned Financial Institutions. Washington, D.C.: Brookings Institution Press. Caprio, Gerard, Jr., Asli Demirgüç-Kunt, and Edward J. Kane. 2008. “The 2007 Melt- down in Structured Securitizations: Searching for Lessons, Not Scapegoats.� Policy Research Working Paper 4756. Washington, D.C.: World Bank, Develop- ment Research Group, Finance and Private Sector Team, October. Carmichael, Jeffrey. 2009. “Regulation by Objective—The Australian Approach to Regulation: Statement to the US Senate Committee on State Homeland Security ­ and Governmental Affairs.� Singapore: Promontory Financial Group, May. Cason, Timothy N., Luta Gangadharan, and Pushkar Maitra. 2008. “Moral Hazard and Peer Monitoring in a Laboratory Microfinance Experiment.� Institute for ­Research in the Behavioral, Economic, and Management Sciences: Paper No. 1208. West Lafayette, Ind.: Purdue University. CGAP (Consultative Group to Assist the Poor). 2011. Global Standard-Setting Bodies and Financial Inclusion for the Poor: Toward Proportionate Standards and Guidance. A white paper written on behalf of the G-20’s Global Partnership for Financial Inclusion. Washington, D.C.: CGAP, October. CGAP and World Bank. 2009. Financial Access 2009. Washington, D.C.: CGAP and World Bank. ———. 2010. Financial Access 2010: The State of Financial Inclusion through the Crisis. Washington, D.C.: CGAP and World Bank. Chatain, Pierre-Laurent, John McDowell, Cédric Mousset, Paul Allen Schott, and Emile van der Does de Willebois. 2009. “Preventing Money Laundering and Terrorist Financing: A Practical Guide for Bank Supervisors.� Washington, D.C.: World Bank. ­ Chatain, Pierre-Laurent, Andrew Zerzan, Wameek Noor, Najah Dannaoui, and Louis de Koker. 2011. Protecting Mobile Money against Financial Crimes: Global Policy Challenges and Solutions. Washington, D.C.: World Bank. Chen, Greg, Stephen Rasmussen, and Xavier Reille. 2010. “Growth and Vulnerabilities in Microfinance.� Focus Note 61. Washington, D.C.: CGAP, February. Chien, Jennifer. 2012. “Designing Disclosure Regimes for Responsible Financial Inclu- sion.� Focus Note 78. Washington, D.C.: CGAP, March. Christen, Robert Peck, Richard Rosenberg, and Veena Jayadeva. 2004. “Financial Insti­ tutions with a ‘Double Bottom Line’: Implications for the Future of Microfi- nance.� Occasional Paper 8. Washington, D.C.: CGAP, July. Bibliography 113 Christen, Robert Peck, and Mark Flaming. 2009. “Due Diligence Guidelines for the ­ Review of Microcredit Loan Portfolios.� Washington, D.C.: CGAP. Cirasino, Massimo, and Jose Antonio Garcia. 2007. “Retail Payment Systems to Support Financial Access: Infrastructure and Policy.� Washington, D.C.: CEMLA and World Bank. ———. 2008. “Measuring Payment System Development.� Financial Infrastructure ­ Series: Payment Systems Policy and Research. Working Paper. Washington, D.C.: World Bank. Claessens, Stijn, Patrick Honohan, and Liliana Rojas-Suarez. 2009. “Policy Principles for Financial Access.� Report of the CGD Task Force on Access to Financial Services. Washington, D.C.: Center for Global Development, October. Collins, Daryl, Jonathan Morduch, Stuart Rutherford, and Orlanda Ruthven. 2009. Portfolios of the Poor: How the World’s Poor Live on $2 a Day. Princeton, N.J.: Princeton University Press. Collins, Daryl, David Porteous, and Jeff Abrams. 2010a. “Prudential Regulation in Microfinance.� Policy Framing Note 3. Cambridge: Financial Access Initiative. ­ ———. 2010.b “Interest Rate Policy. Cambridge: Financial Access Initiative.� Policy Framing Note 4. Cambridge: Financial Access Initiative. Committee on Payment and Settlement Systems. 2001. “Core Principles for Systemati- cally Important Payment Systems.� Basel: BIS, January. ———. 2003. “Policy Issues for Central Banks in Retail Payments.� Basel: BIS, March. ———. 2005. “Central Bank Oversight of Payment and Settlement Systems.� Basel: BIS, May. ———. 2006. “General guidance for national payment system development.� Basel: BIS, January. Cuevas, Carlos E., and Klaus P. Fischer. 2006. “Cooperative Financial Institutions: Issues in Governance, Regulation and Supervision.� World Bank Working Paper ­ No. 82. Washington, D.C.: World Bank. Cull, Robert, Asli Demirgüç-Kunt, and Jonathan Morduch. 2006. “Financial ­Performance and Outreach: A Global Analysis of Leading Microbanks.� World Bank Policy Research Working Paper 3827. Washington, D.C.: World Bank, February. ———. 2008. “Does Regulatory Supervision Curtail Microfinance Profitability and Outreach?� New York: Financial Access Initiative, October. ­ ———. 2009. “Banks and Microbanks.� New York: Financial Access Initiative, September. Cull, Robert, Asli Demirgüç-Kunt, and Timothy Lyman. 2012. “Financial Inclusion and Stability: What Does Research Show?� Brief. Washington, D.C.: CGAP, May. Cull, Robert, Lemma Senbet, and Marco Sorge. 2004. “Deposit Insurance and Bank Intermediation in the Long Run.� BIS Working Papers, No. 156. Basel: BIS, July. 114 Microfinance Consensus Guidelines Dawney, Emma, and Hetan Shah. 2005. “Behavioural Economics: Seven Tips for Policy- makers.� London: The New Economics Foundation. De Koker, L. 2009. “The Money Laundering Risk Posed by Low Risk Financial Products in South Africa: Findings and Guidelines.� Journal of Money Laundering Con- trol, Vol. 12 No. 4. De Ville, Alex, Quentin Lecuyer, and Sophie Weisner. 2009. “Does Foreign Ownership in Microfinance Interfere with Local Development?� ADA Discussion Paper. Luxembourg: ADA, December. ­ Demirgüç-Kunt, Asli, Edward J. Kane, Baybars Karacaovali, and Luc Laeven. 2005. “Deposit Insurance around the World: A Comprehensive Database.� Washington, D.C.: World Bank. Demirgüç-Kunt, Asli, Edward J. Kane, and Luc Laeven. 2006. “Determinants of Deposit- Insurance Adoption and Design.� Policy Research Working Paper No. 3849. Washington, D.C.: World Bank, January. ———. 2006. “Deposit Insurance Design and Implementation: Policy Lessons from Research and Practice.� Policy Research Working Paper No. 3969. Washington, ­ D.C.: World Bank, July. Demirgüç-Kunt, Asli, Thorsten Beck, and Patrick Honohan. 2008. Finance for All? Poli- cies and Pitfalls in Expanding Access. Washington, D.C.: World Bank. Dias, Denise, and Katharine McKee. 2010. “Protecting Branchless Banking Customers: Policy Objectives and Regulatory Options.� Focus Note 64. Washington, D.C.: CGAP, September. Drake, Deborah, and Elisabeth Rhyne, eds. 2002. The Commercialization of Microfinance: Balancing Business and Development. Bloomfield, Conn.: Kumarian Press, Inc. EBRD (European Bank for Reconstruction and Development). 2007. “Mortgages in Transition Economies: The Legal Framework for Mortgages and Mortgage Securities.� London: EBRD. ­ Emran, M. Shahe, and Joseph E. Stiglitz. 2009. “Financial Liberalization, Financial ­ Restraint and Entrepreneurial Development.� Washington, D.C.: Institute for International Economic Policy, January. European Union Payment Service Directive. 2007. 2007/64/EC of 13 November. FATF (Financial Action Task Force). 2012a. “International Standards on Combatting Money Laundering and the Financing of Terrorism and Proliferation, February 2012.� Paris: FATF/OECD. ———. 2012b. “Declaration of the Ministers and Representatives of the Financial Action Task on April 20, 2012.� http://www.fatf-gafi.org/documents/repository/ ­ ministersrenewthemandateofthefinancialactiontaskforceuntil2020.html Bibliography 115 FATF, APG (Asia Pacific Group on Money Laundering), and World Bank. 2011. “Anti- Money Laundering and Terrorist Financing Measures and Financial Inclusion.� Paris: FATF. Fiebig, Michael. 2001. “Prudential Regulation and Supervision for Agricultural Finance, Agricultural Finance Revised No. 5.� Rome: Food and Agriculture Organization of the United Nations (FAO)/Deutsche Gesellschaft für Technische Zusammen- arbeit (GTZ). Flaming, Mark. 2007. “Guaranteed Loans to Microfinance Institutions: How Do They Add Value?� Focus Note 40. Washington, D.C.: CGAP, January. FSD-Kenya. 2009. “Definition of a Standard Measure for Consumer Interest Rates in Kenya: A Scoping Study.� Nairobi: FSD-Kenya. G-20. 2008. “Declaration: Summit on Financial Markets and the World Econ- omy.� Washington, D.C.: G-20. http://www.g20.org/Documents/g20_summit_ declaration.pdf. Gine, Xavier, and Dean Karlan. 2010. “Group versus Individual Liability: Long-Term Evi- dence from Philippine Microlending Groups.� Washington, D.C.: World Bank. Glisovic, Jasmina, Henry Gonzalez, Yasemin Saltuk, and Frederic Rozeira de Mariz. 2012. “Volume Growth and Valuation Contraction: Global Microfinance Equity Valuation Survey 2012.� Forum 4. Washington, D.C.: CGAP and J.P. Morgan. Gonzalez, Adrian. 2007. “Resilience of Microfinance Institutions to National Macro­ economic Events: An Econometric Analysis of MFI Asset Quality.� MIX Discussion Paper No. 1. Washington, D.C.: Microfinance Information Exchange (MIX), July. Goodhart, Charles, Philipp Hartmann, David Llewellyn, Liliana Rojas-Suarez, and Steven Weisbrod. 1998. Financial Regulation: Why, How and Where Now? London: Bank of England. Goodhart, C. A. E. 2000. “The Organizational Structure of Banking Supervision.� Basel: Financial Stability Institute, BIS, November. Hannig, Alfred, and Edward Katimbo-Mugwana, eds. “How to Regulate and Supervise Microfinance? Key Issues in an International Perspective.� Financial Systems Devel- opment (FSD) Project 2000, FSD Series No. 1. Kampala, Uganda: Bank of Uganda and German Technical Co-operation Financial Systems Development (FSD) project. Helms, Brigit. 2006. Access for All: Building Inclusive Financial Systems. Washington, D.C.: CGAP. Helms, Brigit, and Xavier Reille. 2004. “Interest Rate Ceilings and Microfinance: The Story So Far.� Occasional Paper 9. Washington, D.C.: CGAP, September. HM (UK) Treasury. 2010. Laying of Regulations to Implement the New E-Money Directive—A Consultation Document. October. ­ 116 Microfinance Consensus Guidelines ———. 2011. The Electronic Money Directive Consultation: A Summary of Responses. February. IAIS (International Association of Insurance Supervisors). 2011. “Insurance Core Prin- ciples, Standards, Guidance and Assessment Methodology.� Basel: BIS, October. IAIS and Microinsurance Network. 2007. “Issues in Regulation and Supervision of Microinsurance.� Basel: IAIS, June. ­ IAIS, Microinsurance Network, and Access to Insurance Initiative. 2010. “Issues ­ Paper on the Regulation and Supervision of Mutuals, Cooperatives and Other Community-based Organisations in Increasing Access to Insurance Markets.� ­ Basel: IAIS, October. InterAmerican Development Bank. 2004. “Principles and Practices for Regulating and Supervising Microfinance.� Washington, D.C.: InterAmerican Development Bank. International Bank for Reconstruction and Development and World Bank. 2008. “Payment Systems Policy and Research. Balancing Cooperation and Competi- ­ tion in Retail Payment Systems: Lessons from Latin American Case Studies.� Financial Infrastructure Series. Washington, D.C.: International Bank for Recon- struction and Development/World Bank, November. Isern, Jennifer, David Porteous, Rául Hernández-Coss, and Chinyere Egwuagu. 2005. “AML/CFT Regulation: Implications for Financial Service Providers That Serve Low-Income People.� Focus Note 29. Washington, D.C.: CGAP, July. Isern, Jennifer, and Louis de Koker. 2009. “AML/CFT: Strengthening Financial Inclusion and Integrity.� Focus Note 56. Washington, D.C.: CGAP, August. Isern, Jennifer, William Donges, and Jeremy Smith. 2008. “Making Money Transfers Work Delivering for Microfinance Institutions: A Technical Guide to Developing and ­ Money Transfers.� Technical Guide. Washington, D.C.: CGAP, March. Isern, Jennifer, Julie Abrams, and Matthew Brown. 2008. “Appraisal Guide for Microfi- nance Institutions.� Technical Guide. Washington, D.C.: CGAP. Karim, Nimrah, Michael Tarazi, and Xavier Reille. 2008. “Islamic Microfinance: An Emerging Market Niche.� Focus Note 49. Washington, D.C.: CGAP, August. Kumar, Kabir, Claudia McKay, and Sarah Rotman. 2010. “Microfinance and Mobile Banking: The Story So Far.� Focus Note 63. Washington, D.C.: CGAP. Lauer, Kate. 2008. “Transforming NGO MFIs: Critical Ownership Issues to Consider.� Occasional Paper 13. Washington, D.C.: CGAP, May. Lauer, Kate, Denise Dias, and Michael Tarazi. 2011. “Bank Agents: Risk Management, Mitigation, and Supervision.� Focus Note 75. Washington, D.C.: CGAP, December. Lauer, Kate, and Michael Tarazi. 2012. Supervising Nonbank E-Money Issuers. Brief. Washington, D.C.: CGAP. Bibliography 117 Ledgerwood, Joanna, and Victoria White. 2006. Transforming Microfinance Institutions: Providing Full Financial Services to the Poor. Washington, D.C.: International Bank for Reconstruction and Development and World Bank. Lhériau, Laurent. 2009. Précis de réglementation de la microfinance. Agence Française de Développement, Département de la Récherche. Paris: Agence Française de Développement. ———. 2002. “La microfinance commerciale en zone urbaine: quelles possibilités et quelles perspectives en zones franc?� Techniques Financières et Développe- ment, n° 68, October–November. http://esf.asso.fr/portail/IMG/TFD68_Article_ LHERIAU.pdf. ———. 1998. “La réglementation sur les réseaux mutualistes bancaires dans l’UMOA (loi Parmec).� Cahiers Juridiques et Fiscaux de l’Exportation, n° 5. Llewellyn, David T. 2006. “Institutional Structure of Financial Regulation and Super- vision: The Basic Issues.� Paper presented at “Aligning Supervisory Structures with Country Needs,� World Bank, Washington, D.C., June. Lusardi, Anna Maria, and Peter Tufano. 2009. “Debt Literacy, Financial Experiences, and Overindebtedness.� NBER Working Paper 14808. Boston: National Bureau of Economic Research, March. Lyman, Timothy R., Mark Pickens, and David Porteous. 2008. “Regulating Transfor- mational Branchless Banking: Mobile Phones and Other Technology to Increase Access to Finance.� Focus Note 43. Washington, D.C.: CGAP, January. Lyman, Timothy, Tony Lythgoe, Margaret Millier, Xavier Reille, and Shalini Sankaranarayan. ­ uccess Factors.� 2011. “Credit Reporting at the Base of the Pyramid: Key Issues and S Forum 1. Washington, D.C.: CGAP and IFC. Männle, Manfred. 2003. “Interoperable Mobile Payment—A Requirements-Based ­Architecture.� Wirtschaftinformatik 2003. Heidelberg, Germany: Physica-­ Verlag, September. McKee, Katharine, Estelle Lahaye, and Antonique Koning. 2011. “Responsible Finance: Put- ting Principles to Work.� Focus Note 73. Washington, D.C.: CGAP, September. MicroSave. 2011. “Internal Audit and Control Toolkit.� Utta Pradesh: MicroSave. http:// www.microsave.org/toolkit/internal-audit-and-control-toolkit Mullainathan, Sendhil, and Sudha Krishnan. 2008. “Psychology and Economics: What It Means for Microfinance.� Cambridge: Financial Access Initiative, April. Mwenda, Kenneth K., ed. 2002. Banking and Microfinance Regulation and Supervision: Lessons from Zambia. Parkland, Fla.: Brown Walker Press. Napier, Mark. 2011. Including Africa—Beyond Microfinance. London: Centre for the Study of Financial Innovation. 118 Microfinance Consensus Guidelines O’Donohoe, Nicholas P., Frederic Rozeira de Mariz, Elizabeth Littlefield, Xavier Reille, and Christoph Kneiding. 2009. “Shedding Light on Microfinance Equity Valua- tion: Past and Present.� Occasional Paper 14. Washington, D.C.: CGAP, February. Organisation for Economic Co-operation and Development. 2009. “Financial Consumer Protection (Progress Report).� Paris: Organisation for Economic Co-operation and Development, 24 March. Pickens, Mark, David Porteous, and Sarah Rotman. 2009. “Banking the Poor via G2P Payments.� Focus Note 58. Washington, D.C.: CGAP, December. Porteous, David. 2004. “Making Financial Markets Work for the Poor.� Johannesburg: FinMark Trust. ———. 2005. “The Access Frontier as an Approach and Tool in Making Markets Work for the Poor.� London: DFID. ———. 2006. “The Regulator’s Dilemma.� Cambridge: FinMark Trust. www.finmark trust.org.za. ———. 2009a. “Consumer Protection in Credit Markets.� Policy Focus Note 1. Cambridge: Financial Access Initiative, July. ———. 2009b. “Competition Policy in Microcredit Markets.� Policy Framing Note 2. Cambridge: Financial Access Initiative, December. Porteous, David, and Bridget Helms. 2005. “Protecting Microfinance Borrowers.� Focus Note 27. Washington, D.C.: CGAP. Porteous, David, Daryl Collins, and Jeff Abrams. 2010a. “Prudential Regulation in Micro- finance.� Policy Framing Note 3. Cambridge: Financial Access Initiative, January. ———. 2010b. “Interest Rate Policy.� Policy Framing Note 4. Cambridge: Financial ­ Access Initiative, January. Reille, Xavier, Christoph Kneiding, Daniel Rozas, Nick O’Donohoe, and Frederic Rozeira de Mariz. 2010. “All Eyes on Asset Quality: Microfinance Global Value Survey 2010.� Occasional Paper 16. Washington, D.C.: CGAP. http://www.cgap.org/ gm/document-1.9.51444/OP16_SP.pdf Rhyne, Elisabeth. 2002. “The Experience of Microfinance Institutions with Regulation and Supervision.� Presented at the 5th International Forum on Microenterprise, Inter-American Development Bank, Rio de Janeiro, 20 September. ———. 2009. “Microfinance for Bankers and Investors: Understanding the Opportunities and Challenges of the Market at the Bottom of the Pyramid.� New York: McGraw Hill. Rhyne, Elisabeth, Ira Leiberman, Brian Busch, and Stephanie Dolan. 2010. “Aligning Interests: Addressing Management and Stakeholder Incentives during Micro- finance Institution Transformations.� Washington, D.C.: Calmeadow and the Center for Financial Inclusion. Bibliography 119 Rosenberg, Richard. 1999. “Measuring Microcredit Delinquency.� Occasional Paper 3. Washington, D.C.: CGAP, June. ———. 2010. “Does Microcredit Really Help Poor People?� Focus Note 59. Washington, D.C.: CGAP, January. Rosenberg, Richard, Adrian Gonzalez, and Sushma Narain. 2009. “The New Money- lenders: Are the Poor Being Exploited by High Microcredit Interest Rates?� ­ Occasional Paper 15. Washington, D.C.: CGAP. Roth, J., M. J. McCordand, and D. Liber. 2007. “The Landscape of Microinsurance in the World’s 100 Poorest Countries.� Appleton, Wis.: Microinsurance Centre. www.microinsurancecentre.org Rozas, Daniel. 2009. “Throwing in the Towel: Lessons from MFI Liquidations.� September. http://www.microfinancegateway.org/gm/document-1.9.38716/Throwing%20 in%20the%20Towel_%20%20Lessons%20from%20MFI%20Liquidations.pdf Rutherford, Stuart. 2000. The Poor and Their Money. Oxford: Oxford University Press. Rutherford, Stuart, and Sukhwinder Singh Arora. 2009. The Poor and Their Money. Second Edition. United Kingdom: Practical Action Publishing. Schicks, Jessica, and Richard Rosenberg. 2011. “Too Much Microcredit? A Survey of the Evidence on Over-Indebtedness.� Occasional Paper 19. Washington, D.C.: CGAP. http://www.cgap.org/gm/document-1.9.55377/OP19.pdf Seibel, Hans Dieter. 2003. “History Matters in Microfinance.� International Journal of Microfinance and Business Development. Small Enterprise Development. Vol. 14, no. 2 (June), pp. 10–12. Stango, Victor, and Jonathan Zinman. 2011. “Fuzzy Math, Disclosure Regulation and Credit Market Outcomes.� Review of Financial Studies 24(2): 506–534, February. Staschen, Stefan. 1999. “Regulatory and Supervision of Microfinance Institutions: State of Knowledge.� Eschborn: GTZ. ———. 2003. “Regulatory Requirements for Microfinance: A Comparison of Legal Frameworks in 11 Countries Worldwide.� Division 41, Economic Development and Employment Promotion. Eschborn: GTZ. Smart Campaign. 2011. “Client Protection Principles.� Washington, D.C.: Smart Campaign, July. http://www.smartcampaign.org/about-the-campaign/smart-microfinance-and- the-client-protection-principles. ———. 2010. “Responsible Pricing: The State of the Practice.� Washington, D.C.: Smart Campaign, July. Steele, William F., and David O. Andah. 2003. “Rural and Micro Finance Regulation in Ghana: Implications for Development and Performance of the Industry.� Africa Region Working Paper Series No. 49. Washington, D.C.: World Bank, June. 120 Microfinance Consensus Guidelines Tarazi, Michael, and Paul Breloff. 2010. “Nonbank E-Money Issuers: Regulatory Approaches to Protecting Customer Funds.� Focus Note 63. Washington, D.C.: ­ CGAP, July. ———. 2011. “Regulating Bank Agents.� Focus Note 68. Washington, D.C.: CGAP, March. Tiwari, Akhand, Anvesha Khandelwal, and Minakashi Ramji. 2008. “How Do Microfi- nance Clients Understand Their Loans?� Chennai, India: Centre for Microfinance. Trigo Loubière, Jacques, Patricia Lee Devaney, and Elisabeth Rhyne. 2004. Supervising & Regulating Microfinance in the Context of Financial Sector Liberalization: Lessons from Bolivia, Colombia and Mexico. Somerville: Accion. United Nations. 2006. “Building Inclusive Financial Sectors for Development.� New York: United Nations, May. (Also referred to as the Blue Book.) Van Greunig, Hennie, Joselito Gallardo, and Bikki Randhawa. 1999. A Framework for Regulating Microfinance Institutions, Vol. 1. Washington, D.C.: World Bank. Villacorta, Omar, Begoña Gutiérrez-Nieto, and Carlos Serrano-Cinca. 2010. “The New Regu- lation Framework of Bolivian NGO Microfinance Institutions. Perceptions on Risk Management, Solvency and Social Performance.� Saragossa: University of Saragossa. Walker, David, and Joshua Lattimore. 2008. “CDIC-IADI: International Deposit Insur- ance System Survey: Current Designs and Trends.� Presentation prepared for World Bank seminar on deposit insurance, in Washington, D.C., 21 October. Weiner, Stuart E. 2008. “The Federal Reserve’s Role in Retail Payments: Adapting to a New Environment.� Kansas City: Federal Reserve Bank of Kansas City. Weiner, Stuart E., Terri Bradford, Fumiko Hayashi, Richard J. Sullivan, Zhu Wang, and Simonetta Rosati. 2007. “Nonbanks and Risk in Retail Payments.� Paper pre- sented at the Joint ECB-Bank of England Conference on Payment Systems and Financial Stability, in Frankfurt, 12–13 November. World Bank. 2008. “Payment Systems Worldwide: A Snapshot (Appendix).� Payment Sys- tems Development Group. Washington, D.C.: World Bank. http://siteresources. worldbank.org/INTPAYMENTREMMITTANCE/Resources/Global_Survey_ Appendix.pdf Wright, Graham, and Leonar Mutesasira. 2001. “Relative Risk to the Savings of Poor People.� Utta Pradesh: Microsave. http://www.microsave.org. ­ avings WSBI (World Savings Bank Institute). 2009. “Beyond Microcredit: The Role of S Banks in Microfinance—Experiences from Latin America, Africa and Asia.� Perspectives 59, July. ­ ———. 2008. “Position Paper on the Regulation of Microfinance Services.� Brussels: World Savings Bank Institute, June.