94160 IN PARTNERSHIP WITH OUR DONORS © International Finance Corporation 2012. All rights reserved. 2121 Pennsylvania Avenue, N.W. Washington, D.C. 20433 www.ifc.org The material in this work is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. IFC encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly, and when the reproduction is for educational and non-commercial purposes, without a fee, subject to such attributions and notices as we may reasonably require. IFC does not guarantee the accuracy, reliability or completeness of the content included in this work, or for the conclusions or judgments described herein, and accepts no responsibility or liability for any omissions or errors (including, without limitation, typographical errors and technical errors) in the content whatsoever or for reliance thereon. The boundaries, colors, denom- inations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The contents of this work are intended for general informational purposes only and are not intended to constitute legal, securities, or investment advice, an opinion regarding the appropriateness of any investment, or a solicitation of any type. IFC or its affiliates may have an investment in, provide other advice or services to, or otherwise have a financial interest in, certain of the companies and parties (including named herein. All other queries on rights and licenses, including subsidiary rights, should be addressed to IFC’s Corporate Relations Department, 2121 Pennsylvania Avenue, N.W., Washington, D.C. 20433. International Finance Corporation is an international organization established by Articles of Agreement among its member countries, and a member of the World Bank Group. All names, logos and trademarks are the property of IFC and you may not use any of such materials for any purpose without the express written consent of IFC. Additionally, “International Finance Corporation” and “IFC” are registered trademarks of IFC and are protected under international law. Credit Reporting Knowledge Guide Contents Preface ................................................................................................................................................................. v Figures and Tables............................................................................................................................................. vii Acronyms ........................................................................................................................................................... vii 1. Introduction and Overview............................................................................................................................ 1 1.1 Access to Finance and the Importance of Credit Reporting Systems ................................................... 3 1.2 Key Stakeholders in Credit Reporting Systems .................................................................................... 10 1.3 Benefits of Credit Information Sharing ................................................................................................ 12 1.4 Responsible Lending and Financial Education ..................................................................................... 14 1.5 The General Principles for Credit Reporting ........................................................................................ 16 2. Basics of Credit Reporting ............................................................................................................................ 17 2.1 The Taxonomy of Credit Reporting Service Providers ......................................................................... 17 2.2 Ownership Structures ............................................................................................................................ 24 2.3 Optimal Market Size.............................................................................................................................. 27 3. Credit Reporting for Micro, Small, and Medium Enterprises .................................................................... 29 3.1 Microfinance .......................................................................................................................................... 29 3.2 Regional Developments in Microfinance Credit Reporting ................................................................ 31 3.3 Small and Medium Enterprise Finance ................................................................................................. 34 3.4 Credit Reporting Systems and Secured Transactions ........................................................................... 35 4. Legal and Regulatory Framework ............................................................................................................... 37 4.1 Data Collection, Retention, Disclosure, and Security .......................................................................... 39 4.2 Consumer Rights .................................................................................................................................... 42 4.3 Licensing or Registration of Credit Reporting Service Providers ........................................................ 44 4.4 Oversight and Enforcement .................................................................................................................. 45 4.5 Governance and Risk Management ..................................................................................................... 45 4.6 Cross-Border Data Flows ....................................................................................................................... 46 iii Credit Reporting Knowledge Guide 5. Developing Credit Reporting Systems in Emerging Markets ................................................................... 49 5.1 Assessing Market Conditions ................................................................................................................ 49 5.2 Changing Perceptions and Building Awareness .................................................................................. 51 5.3 Ensuring Adequate Data Availability ................................................................................................... 53 5.4 Ensuring Financial Sustainability .......................................................................................................... 57 5.5 Creating an Appropriate Business Model ............................................................................................ 58 5.6 Identifying Appropriate Technology Needs ......................................................................................... 62 5.7 Operational and Practical Considerations ............................................................................................ 65 6. Developing Value-Added Services .............................................................................................................. 75 6.1 Automated Decision-Making Systems .................................................................................................. 75 6.2 International Industry Trends in Developing Value-Added Services .................................................. 76 6.3 Products.................................................................................................................................................. 78 6.4 The Use of Credit Information Data for Prudential Supervision ........................................................ 84 7. Case Studies .................................................................................................................................................. 87 7.1 Ecuador: Supportive Regulator and Pro-Active MFI Network Facilitate Credit Information Sharing between MFIs and a Private Credit Bureau ............................................................................ 88 7.2 Egypt: Strong Regulator Supports Establishment of Country’s First Private Credit Bureau.............. 90 7.3 Morocco: Establishing a Public-Private Partnership for the Credit Reporting System ...................... 91 7.4 India: Integration of MFIs into the Credit Reporting System ............................................................. 94 7.5 Cambodia: Developing a Private Credit Bureau with Strong Government and Stakeholder Support ............................................................................................................................. 96 References ......................................................................................................................................................... 99 Annexes ........................................................................................................................................................... 103 Annex 1: Regional Maps of Positive and Negative Credit Bureaus ......................................................... 103 Annex 1.1: Europe and Central Asia Region ....................................................................................... 104 Annex 1.2: Middle East and North Africa Region ............................................................................... 105 Annex 1.3: Africa Region...................................................................................................................... 106 Annex 1.4: South Asia Region and East Asia and Pacific Region ....................................................... 107 Annex 1.5: Latin America and Caribbean Region ............................................................................... 108 Annex 2: Overview of Credit Registries and Credit Bureaus around the World ............................... cover 3 iv Credit Reporting Knowledge Guide Preface In 2001, the International Finance Corporation (IFC) Program’s work and preparation of this Guide.2 We are launched the Global Credit Bureau Program, later renamed also grateful for the generous contributions of the credit the Global Credit Reporting Program, to better reflect bureaus and registries around the world that made possible the nature of its operations. The objective of this second the development and publication of this Guide. edition of the Credit Reporting Knowledge Guide is to disseminate best practices in credit reporting development, We would also like to acknowledge the support of our and to contribute to credit information sharing in emerging donors, without whom the Global Credit Reporting markets. Since the program was launched, it has helped Program’s activities would not be possible. We thank the develop favorable credit reporting environments in more Canadian Government for its support of our activities in than 60 countries, principally through technical assistance. Latin America and the Caribbean; the Government of This assistance has included support to the regional credit Switzerland for its support in Eastern Europe, Africa, and bureau in Central America and the first credit bureaus East Asia and the Pacific; the Government of the Netherlands established in the Arab Republic of Egypt, Cambodia, for its support of our activities in International Development Morocco, and Tajikistan; work on the legal and regulatory Association countries; the Austrian Government for support framework in Kenya and Panama; and ongoing assistance of our activities in Eastern Europe, Africa and Asia; the U.K. toward the development of credit reporting systems in Government for its support of our programs in Africa, Asia, Mongolia, the Solomon Islands, Sierra Leone, Tanzania, and the Middle East and North Africa; and the Japanese Liberia, Azerbaijan, and other countries. Since 2002, IFC Government for its support in essential global research. has also partnered with the World Bank to monitor credit Finally, we wish to acknowledge donors who have supported reporting environments in more than 180 countries. Annual our efforts in the past, namely the Italian government for its survey results are incorporated into the annual Doing support of our initial activities in Eastern Europe and Latin Business report and disseminated to governments, bureaus, America and the Caribbean; the Norwegian government for registries, creditors, and other interested stakeholders. overall program support and support of our initial activities Through this combination of analytical and operational in Africa; the Australian government for its support of our work, IFC and the World Bank have become recognized activities in Vietnam; the government of New Zealand for as leaders in credit reporting development in emerging its support of our activities in Pakistan and Indonesia; the markets. Government of Luxembourg for its support of our work in Africa and Asia, and in knowledge management and The Credit Reporting Knowledge Guide was prepared by research; and Visa International for global program support. members of the IFC Global Credit Reporting Program team under the direction of Tony Lythgoe.1 The authors We hope this guide will prove both informative and would like to thank colleagues in the World Bank Group useful to all those working in the area of credit reporting for their continuous support of the Global Credit Reporting development. 1 Contributors from IFC’s Global Credit Reporting Team include Oscar Madeddu, Colin Raymond, Shalini Sankaranarayanan, Peter Sheerin, Fabrizio Fraboni, Maria Pincetich, Alban Pruthi, and Jennifer Barsky. We also wish to acknowledge the feedback and inputs received from other IFC colleagues, including Peer Stein, Bassim Ahmed Sharafeldin, and Moyo Violet Ndonde. 2 We are grateful for the input received from IFC’s Secured Transactions and Collateral Registries team: Alejandro Alvarez de la Campa, and Elsa Rodriguez, as well as our World Bank Financial Infrastructure team colleagues, Massimo Cirasino, Fredes Montes and Cornelio Pimental (former World Bank). In addition we would like to acknowledge the excellent editing assistance of Dawn Mpati-Muchira, Mark Feige, and Mary Paden.On design, layout, and production, we would like to thank Amy Quach and Aichin Lim Jones. We also would like to acknowledge Jeffrey Lecksell and Bruno Bonansea for producing the global and regional maps. v Credit Reporting Knowledge Guide Figures and Tables Figures Figure 1.1 Growth of Credit Bureaus ............................................................................................................... 6 Figure 1.2 Credit Bureau Coverage by Region, 2011 ..................................................................................... 7 Figure 1.3 Growth of Credit Registries ............................................................................................................ 8 Figure 1.4 Credit Registry Coverage by Region, 2011 ..................................................................................... 8 Figure 1.5 Credit Information Index ................................................................................................................ 9 Figure 1.6 Key Stakeholders in Credit Reporting Systems ............................................................................ 10 Figure 1.7 Effect on Default Rates of Including Positive Information, United States ................................ 13 Figure 1.8 Effects on Default Rates of Including Positive Information, Argentina and Brazil ................... 13 Figure 1.9 Effect on Approvals of Including Positive Information ............................................................... 13 Figure 1.10 Effect of Including Positive Information on Approvals among Retailers and Other Lenders .............................................................................................................................. 14 Figure 1.11 Effect on Default Rates of Increasing Number of Information Sources ................................... 14 Figure 1.12 Effect of Types and Sources of Information on Predictive Power ............................................ 14 Figure 2.1 Markets Served by CRSPs............................................................................................................... 18 Figure 2.2 Sources of Information for Credit Bureaus .................................................................................. 19 Figure 2.3 Individual-Level Information Collected by Credit Bureaus ......................................................... 19 Figure 2.4 Sample History of Payments ......................................................................................................... 20 Figure 2.5 Firm-Level Information Collected by Credit Bureaus .................................................................. 20 Figure 2.6 Sources of Information for Credit Registries ............................................................................... 22 Figure 2.7 Individual-Level Information Collected by Credit Registries....................................................... 22 Figure 2.8 Firm-Level Information Collected by Credit Registries ................................................................ 23 Figure 2.9 Ownership Structures of Credit Bureaus ..................................................................................... 24 Figure 5.1 Average Time Between Request and Release of Data ................................................................ 56 Figure B5.2.1 Hub & Spoke Model in Central America ................................................................................. 59 Figure 5.2 Qualities of a Strong Technical Partner........................................................................................ 63 Figure 5.3 Common Delivery Modes for CRSPs ............................................................................................. 64 Figure 5.4 Key Items in Contracts/Agreements with Users and Data Providers .......................................... 66 Figure 5.5 Sample Organizational Structure of a CRSP ................................................................................ 67 Figure 5.6 Breakeven Point for a Newly Established Credit Reporting Service Provider ............................ 70 Figure 5.7 Key Performance Indicators of a Credit Reporting Service Provider .......................................... 72 Figure 6.1 Customer Life Cycle: Offering Value-Added Services .................................................................. 76 Figure 7.1 Information Sharing Flow, Morocco ............................................................................................ 93 vii Credit Reporting Knowledge Guide Tables Table 2.1 Comparison of Credit Bureau Ownership Structures .................................................................... 26 Table 5.1 Operational Phase Staffing ............................................................................................................ 68 Table 5.2 Hypothetical Pricing Matrix for Credit Reporting Service Providers ............................................ 69 Table 5.3 Hypothetical Profit & Loss Statement ............................................................................................ 71 Acronyms APC Asociación Panameña de Crédito, Panama BAM Central Bank of Morocco CRIB Credit Information Bureau, Sri Lanka CIBIL Credit Information Bureau (India) Ltd. CRB Africa Credit Reference Bureau Africa Ltd. CRSPs credit reporting service providers CII Doing Business Credit Information Index FBC Finanzas Bajo Control HDFC Housing Development Finance Corporation Ltd. IFC International Finance Corporation MSME micro, small, and medium enterprise MFI microfinance institution MFIN Microfinance Institutions Network, India NBC National Bank of Cambodia NGO nongovernmental organization RFR Red Financiera Rural, Ecuador SBS Superintendencia de Bancos y Seguros, Ecuador SME small and medium enterprise OECD The Organisation for Economic Co-operation and Development UEMOA Union Economique et Monétaire Ouest Africaine VAS value-added services XDS Xpert Decision Systems viii CHAPTER 1 Introduction and Overview I nadequate access to finance and credit represents In markets faced with these challenges, credit reporting one of the most critical constraints to economic service providers (CRSPs) can perform the crucial functions development, particularly for rural and self-employed of gathering and distributing reliable credit information, households and for micro, small and medium enterprises improving creditor protection, and strengthening credit (MSMEs). Much of the population in emerging markets is markets. In effect, the need for physical collateral can employed in the informal sector: many are self-employed be at least supplemented with reputational collateral. as farmers, household-based entrepreneurs with small retail Credit reporting service providers can reduce information shops, street vendors, artisanal manufacturers, or other asymmetry, thus reducing default rates, which in turn service providers. As such, they have no salary slips or should result in lower average interest rates, enhanced other traditional income statements for lenders to ascertain competition in the credit market, and ultimately increased whether a borrower has a steady source of income. Moreover, access to credit. poor households tend to lack collateral—or the right type of collateral, or the proper legal documentation—against This second edition of the Credit Reporting Knowledge which to secure credit. Guide aims to support the dissemination of knowledge on best practices in credit reporting development, based on Lenders often lack the necessary information to assess the IFC’s experience. The original Credit Bureau Knowledge creditworthiness of potential customers, including a lack Guide (2006) elaborated on the knowledge gained over of reliable and unique identification for individuals and several years of running the Global Credit Reporting businesses. In the absence of automated screening methods, Program3 and provided a variety of stakeholders, primarily the relative costs of personal screening and due diligence in emerging markets, with a comprehensive information are very high, while the loan amounts tend to be modest. resource to help them develop their own credit reporting Potential customers are often widely dispersed in rural systems. This second edition updates that information, areas, where it is not cost effective for lenders to operate a and covers changes in credit reporting services over the branch network. With limited access to inclusive and timely past six years. Among the new developments discussed data, lenders are also concerned that borrowers might are the first universal set of standards for credit reporting, accumulate many loans from multiple lenders—potentially credit reporting for micro, small, and medium enterprises, resulting in their overindebtedness and leaving lenders and the role credit reporting is playing in the evolving with an unacceptably large portfolio of nonperforming global responsible lending landscape. Supplementing the loans. Moreover, weak creditor protection and bankruptcy theoretical discussions is a set of case studies highlighting practices, coupled with shaky property rights, often make various aspects of developing credit reporting systems. collecting collateral an ineffective option. 3 Formerly known as the Global Credit Bureau Program. The program changed its name to the Global Credit Reporting Program in 2010. 1 Credit Reporting Knowledge Guide The intended audience for this Guide is IFC’s client long-term capital needs and to grow and expand their counterparts in emerging markets, which includes, but businesses. Providing access to credit to this market is is not limited to, government authorities, regulators and on the development agenda of most emerging markets. overseers, supervisors, financial and nonbank financial Microfinance, which serves an estimated 120 million to 190 institutions, other creditors, credit reporting service million clients,4 focuses on lower-income clients, who are providers, banking and microfinance institution (MFI) often self-employed, household-based entrepreneurs, with associations, and consumer organizations. only informal—if any—business records, little collateral, and no effective access to formal credit markets. While The remainder of Chapter 1 introduces key concepts in microfinance previously enjoyed a sterling reputation for credit reporting: Why is access to credit important? What low delinquency rates, average microfinance portfolios have are the factors limiting access to credit? How can credit witnessed rising “at-risk ratios” in recent years highlighting reporting systems improve access to credit? and, Who are the need for proper credit reporting to reduce the risks of the key actors in credit reporting systems? It examines the overborrowing. problem of asymmetric information—when borrowers know more about their ability and willingness to repay loans Chapter 4 outlines the legal and regulatory framework than do lenders. The chapter then presents a snapshot of options for credit reporting systems. The legal framework the evolution of the industry as it stands today, augmented for credit reporting differs from country to country, and with evidence from empirical research studies that validate may include a combination of credit reporting laws, the importance of credit reporting in the overall agenda for banking laws, data protection laws, consumer protection access to finance. It also discusses the development of the laws, fair credit granting and consumer credit regulations, General Principles for Credit Reporting, the first universal set and personal and corporate privacy and secrecy provisions. of standards for credit reporting. Credit reporting activities can take place in the absence of a legal and regulatory framework. However, in the long run, Chapter 2 introduces the different types of credit reporting best practice indicates credit reporting systems benefit from service providers, which collect information on a borrower’s a legal and regulatory framework that is clear, predictable, credit history from creditors and available public sources. nondiscriminatory, proportionate and supportive of data Unlike credit rating agencies, CRSPs focus on individuals subject and consumer rights.5 As recognition grows that and businesses. There are three basic types of CRSPS: credit credit reporting systems are vital to strengthening financial bureaus, credit registries, and commercial credit reporting infrastructure and ultimately access to finance, more and companies. Each type serves a different function, has its more countries are increasing efforts to create the ideal legal own strengths and weaknesses; no type is inherently better and regulatory environment for these activities. than another for any given market condition. Indeed, given adequate demand, the three types of service providers can Chapter 5 summarizes a decade of IFC experience in and do coexist in a market. Banks, technical partners, developing credit bureaus and credit registries around the government bodies, and private investors are all potential world. The chapter presents various approaches to the owners or investors in a CRSP; the various ownership development of the credit reporting environment and options are also discussed in the chapter. discusses the business, technology, financial, and other operational and practical considerations a developing credit Chapter 3 examines the role that credit reporting can reporting service provider must address. It also reflects on play in facilitating and expanding credit to MSMEs. IFC’s experience with establishing new credit reporting MSMEs require access to financing to meet short- and markets. 4 Lyman et al., 2011. 5 World Bank. 2011a. General Principle IV. 2 Credit Reporting Knowledge Guide Chapter 6 presents an overview of the value-added services corporations and conglomerates: however, the discussion typically offered by established credit bureaus through in this guide is restricted to the credit needs of individuals the repurposing of algorithms and data. The information and the micro, small, and medium businesses that stand to provided by both financial and nonfinancial institutions benefit the most from the development of credit reporting allows a credit bureau to provide comprehensive analysis systems. of borrower creditworthiness, information for portfolio monitoring, and fraud detection. The chapter also discusses Despite the tremendous need, a large proportion of the the use of credit reporting information for prudential world’s population does not have access to credit. In supervision and systemic risk monitoring of the economy developed economies, approximately 90 percent of adults as a whole by financial supervisors. have access to formal financial services compared with 41 percent in emerging markets.7 The total unmet need for Chapter 7 rounds out the theoretical discussions and credit by all formal and informal MSMEs in emerging practical guidelines with five case studies of recently markets today is in the range of $2.1 trillion to $2.5 established CRSPs in the Arab Republic of Egypt, India, trillion.8 Access to credit is largely hindered by the lack of Morocco, Ecuador, and Cambodia. The objective of the sufficient information on the ability of a potential borrower case studies is to provide practitioners with real examples to repay his or her debt and the lack of supporting financial of how credit reporting systems have developed over time infrastructure to make such information available.9 In most in various markets, along with the successes and challenges markets, commercial lending traditionally focused on large faced by each. companies and select retail clients. The credit needs of smaller entrepreneurs and communities are primarily met through informal financial services and nonbank credit.10 1.1 Access to Finance and the The basic approach to lending has remained traditional: Importance of Credit decisions are based on subjective judgments about a Reporting Systems borrower’s propensity to repay supported by alternative risk-mitigating mechanisms, including group guarantees Access to finance is an essential component to economic and the use of collateral. development and job creation. Many studies have shown a positive correlation between financial development and The development of financial infrastructure, broadly economic growth.6 Well-functioning financial systems offer speaking, helps address the issue of access to financial services a variety of financial products for savings, credit, and risk including credit. Financial Infrastructure constitutes the management to a wide range of people and enterprises. underlying foundation for the financial system—including Access to financial services enables rural and urban the institutions, information, technologies, rules, and households to smooth consumption curves and acquire standards that enable financial intermediation. Credit access to essential services including food, housing, health, bureaus, collateral registries, and payment, remittance, and and education. MSMEs require access to financing to meet securities settlement systems are all vital parts of a country’s short- and long-term capital needs and to grow and expand financial infrastructure. When financial infrastructure their businesses. Access to finance is also critical for larger is available, efficient, and reliable, the cost of financial 6 Beck et al., 2007, 27-49; Beck et al., 2005; Rajan et al., 1998; Beck et al., 2000. 7 Demirguc-Kunt et al., 2012. 8 Schiff et al., 2010, 5. 9 Demirguc-Kunt et al., 2012. 10 While this guide is limited to discussing issues with the supply side of providing access to formal finance, it must be noted that the demand side for formal finance also can limit greater financial inclusion. The informal sector is not always willing to be a part of the formal sector, which it may perceive as imposing greater tax burdens and regulatory burdens. 3 Credit Reporting Knowledge Guide intermediation falls; financial products and services become To investigate a borrower’s ability to repay, a lender might accessible to greater numbers of citizens; and lenders and hire investigators to check the borrower’s background, but investors have greater confidence in their ability to evaluate this is also expensive. Conducting in-depth background and guard against credit risk.11 This Guide focuses on the checks, while justifiable for larger loans, is not possible for development of credit reporting systems, one of the key small loans. The unavailability of information at a low cost elements of financial infrastructure, and discusses potential often restricts the ability of lenders to profitably lend to synergies with collateral registries. MSMEs. Monitoring and screening borrower behavior offers one 1.1.1 The Costs of Asymmetric Information way to minimize problems of asymmetric information. Past Credit markets are typically characterized by a fundamental behavior is seen as a reliable predictor of future behavior. problem: that of asymmetric information,12 where the For example, in many countries, banks commonly grant borrower knows the odds of repaying his or her debts credit to a firm only after the firm has had an account with much better than the lender does. The inability of the a bank for at least six months to a year, which allows the lender to accurately assess the credit worthiness of the creditor bank to observe the firm’s cash flow. Similarly, borrower contributes to higher default rates and smaller the group lending approach mostly used by microlenders, loan portfolios, which affect the profitability of the financial allows lenders to provide loans to individual borrowers institution. Differentiating between good and bad clients who, through participation in the group, have developed becomes very difficult or almost impossible when credit a credit history with the lending institution. In these reports are lacking. Without this information, the risk of examples, the credit history—sometimes referred to as lending is higher, which both raises the costs of borrowing “reputational collateral”—minimizes the perception of and reduces the availability of credit because lenders hesitate risk, thus enabling an individual or a firm to gain access to extend credit to unknown borrowers and seek to offset to financing. Nonetheless, the relevance of past behavior the costs of default through higher interest rates. should be considered in context, since it cannot explain all behavior, and could be irrelevant when adverse economic Lenders typically address these problems by requiring conditions change the circumstances. For example, a collateral to cover the loss in case of a default or by perfectly good borrower can default if faced with economic investigating a borrower’s ability to repay. Requiring hardship or other adverse circumstances. collateral is often problematic, especially in developing countries, particularly in the case of new firms and MSMEs, 1.1.2 The Development of Credit which often lack significant assets that are formally (legally) Reporting Systems recognized as useable collateral. In addition, the costs to lenders of seizing and liquidating assets used as collateral Credit reporting systems are a critical element of a country’s can be significant and the process lengthy. According to financial infrastructure, and are essential to facilitating the World Bank’s Doing Business survey data,13 in most access to financial services. They should effectively support emerging markets it takes one to two years to enforce a the sound and fair extension of credit in an economy as contract with costs reaching 20–40 percent of the debt. the foundation for robust and competitive credit markets. In extreme cases, for example in Timor-Leste, it takes on To this end, credit reporting systems should be safe and average more than three years to enforce a contract and may efficient, and fully supportive of data subject and consumer cost up to 164 percent of the cost of the claim. rights.14 11 World Bank et al., 2009 12 Stiglitz et al., 1981. 13 Doing Business Indicators (database), 2012, “Enforcing Contracts” indicator. 14 World Bank 2011a. 4 Credit Reporting Knowledge Guide Credit reporting systems help ensure financial stability by to granting credit to more automated lending processes enabling responsible access to finance and can also play assisted by inputs from quantitative models. As a result, an instrumental role in expanding access to credit and lenders are able to deliver financial services at significantly other services on credit to the underserved and unbanked. reduced costs and expand credit to broader segments of They facilitate lending processes by providing lenders with the economy, thus further democratizing credit services. In objective information that enables them to reduce their particular, the introduction of credit scoring in the 1950s portfolio risk, reduce transaction costs, and expand their in the United States—coupled with the automation of lending portfolios. By doing so, credit reporting systems, workflow and credit underwriting—played a key role in the enable lenders to expand access to credit to creditworthy rapid rise of consumer lending. borrowers including individuals with thin credit files, microentrepreneurs, and small and medium enterprises. Latin America has some of the oldest credit bureaus in the world, but it was not until the 1990s, that privately Credit reporting systems comprise the institutions, operated credit bureaus started to take off in most other individuals, rules, procedures, standards, and technology emerging markets. Between 1990 and 2011, the number that enable information flows relevant to making decisions of credit bureaus in the world almost tripled.17 In Asia, related to credit and loan agreements.15 At their core, credit many markets turned toward private credit reporting after reporting systems consist of databases of information on the financial crisis in the late 1990s. New developments in debtors, together with the institutional, technological, and credit reporting are underway in Central Asia, specifically legal framework supporting the efficient functioning of in Azerbaijan, the Kyrgyz Republic, Tajikistan, Uzbekistan, these systems. Whereas several entities collect information Nepal, and Mongolia. From the early 1990s to the late- on debtors for a variety of purposes, this Guide focuses on 2000s, a significant number of credit bureaus emerged in entities that collect such data with the aim of (1) improving Eastern Europe. Over the past few years, the Middle East and the quality and availability of data for financial and North African region has seen a growing interest in credit nonfinancial creditors to make better-informed decisions; reporting, with credit bureaus established in Morocco and and, (2) assisting banking supervision while improving Egypt, and new developments taking place in other North the quality and availability of data for supervised financial African countries. There are also many new developments intermediaries.16 These entities, also referred to as credit in Sub-Saharan Africa, with the launch of credit bureaus reporting service providers are broadly classified as credit in Ghana, Uganda, Kenya, Rwanda, Botswana, and other bureaus, credit registries, and commercial credit reporting countries. service providers (further discussed in Section 2.1). The rest of this section discusses the evolution and history of some The earliest record of a credit registry dates to 1934, when of these entities. the German credit registry was established. In 2011, 85 countries reported having a credit registry.18 Credit registries Although the first credit bureau may be traced to the early have generally focused on supporting the prudential 1800s in London, starting in the 1950s modern providers supervision and risk monitoring of regulated financial have evolved rapidly, fueled by improvements in technology institutions. and an expansion of credit. This revolution has made access to credit almost ubiquitous in developed markets by allowing IFC is working in partnership with several governments to banks to move from the traditional, subjective approach develop credit registries, enhance existing credit registries, 15 Ibid., see Glossary. 16 The General Principles for Credit Reporting provides a glossary of commonly used terms in reference to credit reporting systems. To be consistent, we are using terminology from the Principles. 17 IFC calculation based on survey data from Doing Business Indicators database for 2006 to 2011. 18 Doing Business Indicators (database), 2012, “Getting Credit” indicator. 5 Credit Reporting Knowledge Guide establish public-private partnerships in the development 1.1.3 Industry Overview and of credit reporting systems, and support the development International Trends of enabling legal and regulatory frameworks. Examples of IFC’s work with credit registries can be found in According to the World Bank’s Doing Business 2012,19 Ethiopia, Algeria, the Lao People’s Democratic Republic, approximately 134 countries out of 183 countries surveyed Bangladesh, China, and the Maldives (public credit had either a credit bureau or a credit registry at the end registries); and in Indonesia and Uzbekistan (joint public- of 2011. The credit reporting industry has experienced private partnerships). Many reform-oriented governments unprecedented growth since 2000, especially in emerging are supporting the development of credit reporting services markets (see Figures 1.1 and 1.3). This growth was driven in conjunction with broader reforms for greater access to by two factors: finance and the promotion of responsible lending practices. • High growth of retail credit in emerging markets: For more information, see map on “Overview of Credit Between 1985 and 1995, unfavorable macroeconomic Registries and Credit Bureaus Around the World” in the environments and structural restrictions in credit Annex. markets in emerging economies constrained credit Figure 1.1: Growth of Credit Bureaus 120 100 80 Number of Credit Bureaus 60 40 20 0 11 77 79 81 83 87 89 91 93 97 99 01 03 07 09 75 85 95 5 74 0 20 19 19 19 19 19 19 19 19 19 19 20 20 20 20 19 19 19 20 19 e- Pr OECD Sub-Saharan Africa East Asia & Pacific South Asia Europe & Central Asia Latin America & Caribbean Middle East & North Africa Source: IFC calculations based on Doing Business Indicators (database) 2004-2011. 19 World Bank 2011b. 6 Credit Reporting Knowledge Guide growth. During this period, the private credit–to–GDP According to Doing Business 2012, of 183 countries ratio for the emerging markets increased from 35 percent surveyed, 89 reported having one or more credit bureaus. to 45 percent.20 Financial liberalization and improved Figure 1.1 illustrates growth in credit bureaus from pre- macroeconomic stability in the period between 1996 and 1974 to 2011. 2011 saw a corresponding increase in credit growth with credit to the private sector increasing from 46 percent to Doing Business measures the quality of credit information 74 percent.21 With more lenders entering and expanding in a region or country based on coverage and the Credit the retail credit market, the need for credit information Information Index (CII). Coverage is defined as the number and for streamlining lending processes grew, leading to of records in the bureau or registry divided by the adult the establishment of credit reporting services. population in the country between ages 15 and 64. In terms of bureau coverage, the Latin America and the Caribbean • Developments in information technology: The credit region continues to lead among developing regions, with reporting industry is data driven. Recent improvements 34.2 percent adult coverage (see Figure 1.2). Yet, since in database management software and decreasing costs 2005, credit bureaus in Europe and Central Asia, East Asia of storing and processing data, as well as decreasing costs and Pacific, and the Middle East and North Africa have of hardware, and the ability of several markets to join demonstrated significant leaps in improving their coverage, and utilize the “Hub & Spokes” model22 have reduced with increases ranging from 7 to 22 percentage points. the start-up costs of a credit reporting service. Although Sub-Saharan Africa had the least-developed credit • Broader reforms stemming from recent crisis: The information infrastructure, with only 8 out of 46 countries recent 2007–2008 financial crisis, has given greater reportedly having credit bureaus, the region has made impetus for broad reform efforts at the national level as significant strides in recent years. Trends are encouraging authorities in developed and emerging markets realize in the Middle East and North Africa region, where 7 out the need for strengthening and improving financial of 19 countries had credit bureau coverage. The East Asia infrastructure, including credit reporting systems. and Pacific region also experienced somewhat positive Figure 1.2: Credit Bureau Coverage by Region OECD 63.9 Latin America & Caribbean 34.2 Europe & Central Asia 29.4 East Asia & Pacific 18.1 Middle East & North Africa 9.3 South Asia 5.9 Sub-Saharan Africa 5 0 10 20 30 40 50 60 70 Percent of adults Source: Doing Business 2012. 20 World Development Indicators, July 2012. Data based on domestic credit to private sector as a percent of GDP. 21 Ibid. Includes low- and middle-income countries. 22 For an explanation of the Hub & Spokes model, see Chapter 5, models for CRSPs. 7 Credit Reporting Knowledge Guide changes, where 8 out of 24 countries surveyed had credit growth in credit registries from pre-1964 to 2011. Europe bureau coverage. The situation was less promising in South and Central Asia led all regions with 16.2 percent coverage Asia region where only 4 out of 8 countries had any credit by credit registries, while South Asia lagged behind at 1.7 bureau coverage.23 percent coverage (see Figure 1.4).24 According to Doing Business 2012, of 183 countries, 85 A factor contributing to the low coverage ratios in some reported having a credit registry. Figure 1.3 illustrates the of these regions is that the percent of the population that Figure 1.3: Growth of Credit Registries 80 70 Number of credit registries 60 50 40 30 20 10 0 11 66 72 75 78 83 86 90 92 95 97 99 02 06 08 64 04 20 19 19 19 19 19 19 19 19 19 19 19 20 20 20 19 20 e- Pr Middle East & North Africa Latin America Europe & Central Asia Asia Africa OECD Source: IFC calculations based on Doing Business Indicators (database) 2004-2011. Figure 1.4: Credit Registry Coverage by Region Latin America & Caribbean 10.1 Europe & Central Asia 16. OECD 9.5 East Asia & Pacific 10.3 Middle East & North Africa 8.1 Sub-Saharan Africa 3.2 South Asia 1.7 0 2 4 6 8 10 12 14 16 Percent of Adults Source: Doing Business 2012. 23 Doing Business Indicators (database), 2012. “Getting Credit” indicator. 24 Ibid. 8 Credit Reporting Knowledge Guide uses credit constitutes only a small portion of the total ranked first among emerging market regions in 2005, moved population. As credit growth continues, the scope of credit to third place together with South Asia, while East Asia and reporting coverage can be expected to expand as well. the Pacific and Sub-Saharan Africa tied for last place. In addition to coverage, the Doing Business CII measures The development of credit reporting services in many credit information availability in a country based on six emerging markets often, but not always, involves key factors listed below (see Figure 1.5).25 For each of the partnerships with the major and well-established six factors, a country receives one point and the points are international CRSPs. As a result, several major players added to arrive at the total index score. dominate the credit information industry globally, namely Equifax, Experian, and TransUnion. Their main operations • Both positive and negative credit information (for are concentrated in the OECD countries, but all three have example on payment history number and kind of actively expanded into emerging markets. accounts, number and frequency of late payments, and any collections or bankruptcies) is distributed. Since the early 2000s, several new CRSPs with international • Data on both firms and individuals are distributed. operations have emerged as players; including CRIF, • Data from retailers and utility companies are distributed an Italian firm present in Europe, North America, Latin to financial institutions. America and the Caribbean, Africa, and Asia; Creditinfo, an Icelandic credit information provider with operations in • More than two years of historical data are distributed. Europe, Central Asia, the Middle East, and more recently Registries that erase data on defaults as soon as they are expanding into the Caribbean and Sub-Saharan Africa; repaid obtain a score of 0 for this indicator. CompuScan, Credit Reference Bureau Africa Ltd. (CRB • Data on loans below 1 percent of income per capita are Africa), and Xpert Decision Systems (XDS), which all distributed. A registry must have a minimum coverage operate in at least three or more African countries; Veda of 1 percent of the adult population to score a 1 for this Advantage, which operates in Australia and New Zealand; indicator. and Dun & Bradstreet South Asia Middle East Ltd., which operates in the Asia Pacific region, the Middle East, and • Regulations guarantee borrowers the right to access their Sub-Saharan Africa. The entry of new international CRSPs data in the largest registry in the economy. is a welcome development as more competition is likely to result in better product offerings and lower prices. Europe and Central Asia led among emerging market regions on Doing Business’ CII followed by the Middle East and North Although there is usually a sound commercial rationale Africa region. Latin America and the Caribbean, which for emerging market countries to seek partnerships with experienced international providers, the value of the locally Figure 1.5: Credit Information Index developed solution provider should not be overlooked. In 6.0 many emerging markets, for example Kenya and Barbados, Credit Information Index 5.0 5.0 5.0 4.0 the origins of credit information sharing can be found 4.0 3.0 3.0 with small businesses providing a localized service, often 3.0 2.0 2.0 with little or no support from policy makers or the central 2.0 bank. Creditinfo, CompuScan, XDS, and CRB Africa all 1.0 0.0 started out as small businesses in markets that the larger High Europe & Latin Middle East South East Asia Sub- international credit reporting companies had declined Income Central America & & North Asia & Pacific Saharan OECD Asia Caribbean Africa Africa to serve for a variety of reasons, and have ended up as Source: Doing Business 2012. international players in their own right. 25 Ibid. 9 Credit Reporting Knowledge Guide 1.2 Key Stakeholders in Credit Figure 1.6: Key Stakeholders in Credit Reporting Systems Reporting Systems Credit Reporting A credit reporting system comprises the institutions, Service Providers individuals, rules, procedures, standards, and technology that enable information flows relevant to making decisions related to credit agreements.26 The development of an effective credit reporting system is a lengthy process Authorities Data requiring a sustained commitment of all stakeholders. The Users Regulators Providers entire process of setting up a credit reporting system, from Supervisors initial discussions to public education and work on the legal and regulatory framework, to actual implementation of the systems, uploading data, and issuing the first credit report may take three to five years, if not longer. The credit Data information cycle of collecting, storing, and processing Subjects data, and distributing, and using information to support credit-granting decisions and financial supervision involves Source: IFC 2012. a number of actors: individuals, MSMEs, CRSPs, data Adapted from General Principles for Credit Reporting, 2011. providers, users, authorities, regulators, and overseers. Active participation by each of these stakeholders is The CRSP is also responsible for the sustainability of critical to ensuring the effectiveness of the credit reporting operations, reporting to shareholders (where applicable), system. Stakeholder participation is further enhanced by compliance with regulatory requirements, implementing government support of the system as a whole. These actors governance arrangements, personnel matters, and dealing and their roles are shown in Figure 1.6 and described below. with consumer complaints. Credit Reporting Service Providers: A CRSP is an Data providers: Data providers are creditors and other entity that administers a networked credit information entities that proactively and in a structured fashion supply exchange. A networked credit information exchange is information to credit reporting service providers.28 Data a mechanism enabling credit information collection, providers play a key role in the successful operation of a processing, and further disclosure to users of data, as well CRSP since the CRSP relies on their pro-active provision as value-added services based on such data.27 A CRSP of data. Traditional data providers include commercial collects data from creditors and available public sources on banks, other financial institutions, and credit card issuers. a borrower’s credit history. The CRSP runs and operates a Nontraditional data sources include retailers and utilities credit reporting service on a day-to-day basis. The CRSP’s providers. In addition, all private and public entities that duties are discharged by the on-site management team and collect information on consumers are potential data sources operational staff, whose responsibilities include collecting, for CRSPs. For instance, a CRSP may have agreements validating, and merging credit history and identification with administrators of databases on court judgments, data; and producing and dispersing credit information in information regarding unpaid debts, personal identity an organized format to users. The CRSP bears primary records, and registries of collateral such as vehicles, real responsibility for ensuring the system safety and efficiency. estate, and companies. 26 World Bank 2011a. 27 Ibid., see Glossary. 28 Ibid., see Glossary. 10 Credit Reporting Knowledge Guide Data providers and other data sources are usually separate third parties in a credit reporting system. 31 They are the legal entities and may be subject to other business and subjects on whom lenders wish to assess the risks of default legal requirements, especially requirements pertaining and nonpayment before approving new loans or advancing to the privacy of consumer information. Data providers further credit. should avoid furnishing erroneous information and should participate in the process of correcting errors identified by Regulators (other overseers): In jurisdictions in which they the CRSPs and data subjects. exist, the regulator is the authority with statutory powers of supervision over credit reporting activities and services. In some jurisdictions, data providers are restricted in the Statutory powers may include the power to issue licenses and manner they may share data about their customers and to create operational rules and regulations. The regulator clients. Accordingly, agreements entered into between may also have the power to stipulate compliance conditions CRSPs and data providers should make provision for for CRSPs, penalize them for violations or noncompliance, the basis of ownership of the data; the basis for consent or cancel their licenses. Once a CRSP is fully operational, to share the data with the CRSP; and how liability will the regulator’s role is to monitor compliance. In addition be proportioned in the event of harm stemming from to direct regulation, CRSPs may also be indirectly subject inaccuracies, security breaches, identity theft, or other to other laws, for example, business or company law, damaging events. consumer protection law, and information privacy law. As such, they may also have compliance obligations imposed Users: A user is an individual or business that requests credit by other regulators. reports, files or other related services from credit reporting service providers, typically under predefined conditions and A vast majority of countries assign regulation of, and rules.29 The information produced by CRSPs is of interest to authority over, credit reporting service providers to their a variety of users. These users “query” or submit an inquiry central banks. A few countries have a regulatory authority to the CRSP on a data subject that has approached them specifically dedicated to credit reporting, for example, the for credit. Users typically include financial institutions and National Credit Regulator in South Africa. nonbank creditors who contribute credit information about their customers’ accounts. Credit information might also be In other countries, a government agency assumes that role; of interest to other users, ranging from financial supervisors for example, the Federal Trade Commission in the United and central banks, to users in other sectors of the economy States has authority to enforce the Fair Credit Reporting Act such as employers, insurers or landlords (where this is legally (which applies to credit bureaus) as part of its mandate to permitted). In keeping with the principle of reciprocity,30 ensure consumer protection in credit and lending practices. only users that contribute information to the CRSP receive More recently, the Consumer Financial Protection Bureau credit information reports from the CRSP. Some CRSPs established by United States Congress to make markets charge users membership fees as well as a pay-per-use fee, in work for consumer financial products and services, has been which case users are also known as members or subscribers given the mandate to supervise credit bureaus (effective of the CRSP. September 30, 2012).32 Data subjects: A data subject is an individual or a business In some countries (e.g., China), the central bank acts as whose data could be collected, processed, and disclosed to both the regulator of the industry and the operator of a 29 Ibid. 30 Broadly speaking, according to the principle of reciprocity, users of a CRSP can obtain information from the CRSP only if they provide information to the CRSP. 31 World Bank 2011a, see Glossary. We occasionally use the term “consumers” in this Guide in place of “data subjects.” 32 http://www.consumerfinance.gov/pressreleases/consumer-financial-protection-bureau-to-superivse-credit-reporting/ 11 Credit Reporting Knowledge Guide CRSP. Despite the apparent conflict of interest, most of contains favorable information on an individual’s open and these systems operate reasonably well as long as the two closed credit accounts. Information sources could include: functions are undertaken by separate departments under debt ratios, on-time payments, credit limits, account type, different directorships: that is, the department issuing loan type, lending institution, detailed reports on the operating licenses and supervising credit bureaus is not the prospective borrower’s assets and liabilities, guarantees, debt same department that operates the credit registry.33 maturity structure, and pattern of repayments. According to Doing Business survey data, approximately 70 percent Since the core business of credit reporting involves the flow of all bureaus and registries surveyed provided both positive of information through a network of stakeholders, credit and negative data,35 often referred to as “comprehensive reporting activities touch on sensitive issues such as the credit reporting” or “full-file credit reporting.” individual privacy rights of consumers and the protection and security of the data subject’s data. Regulators and Negative-only databases were developed initially to help other overseers are tasked, therefore, with monitoring the lenders effectively screen and exclude high-risk borrowers activity of the entire credit reporting system along with its that had accumulated significant debt exposure. However, participants to establish a fair and competitive marketplace in the absence of positive credit reporting, a borrower could for credit reporting service providers while ensuring that remain excluded from credit access (for up to 5 years in some individual privacy rights are respected and protected. countries) based on a single negative event regardless of the current payment record and other information that reflects favorably on him or her. Furthermore, in negative-only 1.3 Benefits of Credit Information reporting systems, lenders do not have credit information Sharing on prospective borrowers who have never defaulted, since no information on them is reported or stored. Credit reporting service providers collect information from both financial institutions, such as banks and credit card In the late 1990s, Hong Kong SAR, China, and the companies, and a variety of nonfinancial institutions, such Republic of Korea experienced a major increase in retail as utility companies and retailers, as well as from public credit defaults as a result of the unfortunate combination records, and other data sources such as databases on of reckless lending practices, and unavailability of positive bounced cheques, promissory notes and protested bills of information. While both had negative credit bureaus, exchange, collateral registries, vehicle registries, real estate positive information was not being shared and lenders registries, personal identity records, company registries, tax were not aware of the level of indebtedness of existing and authority databases, and some court records.34 prospective borrowers. As competition in the credit card market increased and banks marketed credit cards more Credit history data can be broadly categorized as: negative aggressively, many consumers accumulated several credit data and positive data. Negative reporting includes only cards. Borrowers would typically open one credit card information pertaining to unfulfilled financial obligations account and then another to pay off the debt accumulated such as defaults, amounts in arrears, court judgments, and on the first credit card. This borrowing was unsustainable other adverse or negative information. Information on and resulted in a large number of credit card defaults. delinquent debts that are eventually paid off usually remains Following the crises, both countries moved to a system on file and forms part of the credit history for a defined of comprehensive credit reporting and started providing period of time. Historically, databases with negative-only both positive and negative information. Indeed, as shown data have been referred to as “black lists.” Positive credit data 33 The General Principles apply equally to all credit reporting service providers providing the same function. 34 World Bank 2011a. 35 Doing Business Indicators (database), 2012. “Getting Credit” indicator. 12 Credit Reporting Knowledge Guide in the regional maps in the Annex a number of countries Figure 1.8: Effect on Default Rates of Including Positive have adopted full-file credit reporting in their respective Information, Argentina and Brazil jurisdictions. 4 3.81 22% decline 3.37 3.5 2.98 45% Research has shown that comprehensive credit reporting 3 decline Default rates systems generate more accurate scores than negative-only 2.5 systems. A recent analysis of Chile’s credit reporting system, 2 1.84 a negative-only system with some positive data elements, 1.5 found that credit decisions based on comprehensive 1 information significantly outperformed those based on 0.5 negative-only information.36 Another study in the United 0 Argentina Brazil States simulated and compared default rates on loans Positive & negative information Negative information approved using a negative-only credit scoring model with default rates on loans based on a scoring model using Source: IFC using Powell, Mylenko, Miller, and Majnoni 2004 data. both negative and positive information. According to the study, the default rate on loans approved using negative- scoring models produces better predictions, and improves only system was 3.35 percent, whereas the default rate on the ability of lenders to separate good borrowers from loans approved using scores based on both positive and high-risk borrowers. For a bank with a $100 million loan negative information dropped to 1.9 percent, a 43 percent portfolio, this translates into an average savings of $830,000 decrease (see Figure 1.7).37 A similar exercise was conducted in Argentina and $1.5 million in Brazil. using data from Brazil and Argentina with similar results. Inclusion of positive information would have produced a 22 Figure 1.9 shows how including positive information percent decrease in the default rate for Argentinean banks increased approval rate by 88 percent in the simulation and a 45 percent decrease in default rates for Brazilian banks using data from the United States. The study also found (see Figure 1.8).38 Thus, including positive information in that sharing positive information derived from a broader Figure 1.7: Effect on Default Rates of Including Positive Figure 1.9: Effect on Approvals of Including Positive Information, United States Information 4 80 74.8 3.5 3.35 70 3 60 Default rates Approval rates 2.5 50 2 1.9 40 39.8 1.5 30 1 20 0.5 10 0 0 Negative information Positive & negative information Negative information Positive & negative information Source: IFC using Barron and Staten 2003 data. Source: IFC using Barron and Staten 2003 data. 36 Turner 2010. 37 Barron et al., 2003. Figures show the simulated credit defaults assuming an acceptance rate of 60 percent. The simulations were based on data in one of the largest U.S. credit bureaus. 38 Powell et al., 2004. 13 Credit Reporting Knowledge Guide category of sources would allow significant operational significantly reduce default rates and/or increase lending improvements by lowering defaults or increasing lending volumes (see Figure 1.11). volumes to new categories of borrowers (see Figure 1.10).39 In summary, credit reports that have the highest predictive Credit reporting brings benefits to both small and large power combine both positive and negative information institutions. The study using data from Argentina40 found from both banks and nonbank lenders. Bureaus or credit that while small lenders do benefit more than large lenders registries fragmented by industry that provide only negative from sharing information, large banks also benefit from a information deliver reports that have less predictive power significant drop in defaults if positive information is used. and often result in inaccurate credit risk assessment (see Although the results may vary from country to country, Figure 1.12). and from lender to lender, both anecdotal and available empirical evidence suggests that information sharing and Figure 1.12: Effect of Types and Sources of Information use of credit scoring allow both large and small banks to on Predictive Power Types of information Positive & negative Negative Figure 1.10: Effect of Including Positive Information on Approvals Among Retailers and Other Lenders Sources of information information information 84 83.4 “Full” (information High Lower predictiveness 82 shared by banks, predictiveness (e.g. Australia, 80 retailers, NBFIs) (e.g. U.S., U.K., India) Swaziland) Approval rates 78 “Fragmented” Lower Lowest 75.4 (e.g. information predictiveness predictiveness 76 shared among banks (e.g. Mexico, (e.g. Malaysia, 74 only or retail only) Kuwait) Botswana) 72 Source: IFC 2012. 70 Retail Retail & other lenders Source: IFC using Barron and Staten 2003 data. 1.4 Responsible Lending and Financial Education Figure 1.11: Effect on Default Rates of Increasing Number of Information The aftermath of the 2007–2008 financial crisis has brought 2.5 about a heightened awareness of the importance of financial 2.22 2.42 41% infrastructure, as well as its shortcomings. Although 2 decline helping to reduce risks and increase efficiency of financial 79% Default rates 1.5 decline intermediation, financial infrastructure can also sometimes 1.31 contribute to situations where excessive risks are taken. In the 1 case of credit reporting, credit scores were faulted for having 0.52 contributed to the subprime crisis in the United States as 0.5 mortgage lenders made suboptimal lending decisions based 0 solely on credit scores, the repercussions of which were felt Large bank Small bank Information from a bank & bureau Information from a bank throughout the global financial community. To provide a Source: IFC using Barron and Staten 2003 data. more balanced perspective, the crisis highlighted the need for 39 Ibid. The simulation shows that a lender with a target approval rate of 60 percent was able to reduce default rates by 38 percent. If the default rate is used as a target, the bank would be able to approve 11 percent more clients before reaching the targeted 3 percent default rate. 40 Ibid. 14 Credit Reporting Knowledge Guide more regulation and oversight over credit reporting systems credit terms and pricing options available to them, or as a whole, as well as over the use of the various products the implications of accepting multiple financial products and services that these systems provide. Nonetheless, one of when a credit reporting system is in existence. Also, the the key takeaways from the financial crisis and subsequent inability of customers to enforce their rights—whether it is regional crises, is that information sharing among creditors something as simple as challenging information on record is critical to ensuring more accurate portfolio management or questioning an adverse decision—is often a result of not and risk assessment, identifying potential over indebtedness knowing that such rights exist. issues, and allocating capital more efficiently. As global financial markets strive for more responsible Another key development since the crisis is the greater financial practices, it is now recognized that CRSPs have a level of effort globally to improve consumer awareness role to play in helping consumers gain the skills necessary and understanding of what a credit reporting system to take control of their personal finances and gain greater is, what it does, and how it impacts them—along with awareness of the role of credit information in the financial efforts to educate consumers on their related rights services sector. Many CRSPs have fully embraced the and obligations and generally enhance their financial concept of consumer education and have incorporated capability.41 Consumers may not fully understand the financial literacy programs into their corporate strategies (see Box 1.1). Box 1.1: Educating Borrowers on Credit Reporting In 2006, the credit bureau in Panama, Asociación Panameña de Crédito (APC), set up a program called Finanzas Bajo Control (FBC - Finances under Control) to educate Panamanian borrowers about managing their finances and the role of the credit bureau. Prior to setting up the program, APC conducted a detailed analysis of the consumer population to better understand its needs. Based on this analysis, APC developed a plan of action including a detailed budget, a set of key themes, and the channels of information delivery. The pilot program was launched in a controlled environment and tested with a limited number of consumers. Since its launch in 2006, the FBC program has leveraged a variety of channels in delivering its messages to borrowers including face-to-face counseling sessions; a website dedicated to providing information to consumers;42 webinars and seminars; mass-media outlets such as radio programs and TV spots; free press; and social media such as YouTube, Facebook, and Twitter. Over eight years, the program conducted 1.4 million face-to-face interactions with consumers. The website had over 174,000 hits in 2012, and 15 percent of consumers whose data was reported to the bureau claimed they viewed their report online. Through webinars, seminars, company events, and fairs, FBC trained over 5,000 people. The program developed over 50 TV programs and thousands of “tips” that are frequently broadcast on radio stations. The program publishes articles on credit reporting and tips on improving credit histories through free press channels at least once a month, and runs a regular column that provides financial advice to consumers in free press channels. More recently, the program has developed a presence on social media sites including Twitter, Facebook, and YouTube, and is slowly building up a regular follower base. 42 Asociación Panameña de Crédito website. www.miapc.com 41 The Organisation for Economic Co-operation and Development (OECD) defines financial capability as the process by which financial con- sumers improve their understanding of financial products and concepts and, through information, instruction, and objective advice, develop the skills and confidence to become more aware of financial risks and opportunities, make informed choices, know where to go for help, and take other effective actions to improve their financial well-being. See OECD 2005 and World Bank et al., 2009. 15 Credit Reporting Knowledge Guide Because CRSPs harbor personal information on individual institutions (“Tier 2” Group), although not formal members borrowers, they are responsible for protecting the privacy of the task force, were consulted during the preparation of rights of these individuals. Individuals should be aware the Principles. They included other industry associations, that their information is being collected, how it is being private sector operators, scholars, and practitioners. The collected, and what it is being used for or can be used for report was also released for public consultation.44 The role under their country’s legal and regulatory framework. They of the task force was to define a set of guiding principles should be given reassurance that their data is held safely and that could be used to promote best practice in any credit securely and will not fall into the wrong hands. In addition, information-sharing environment taking into account the they should be permitted access to their files, and should balance between the needs of the financial services industry be allowed to challenge information on their report they and financial supervisors to access data and the rights of the believe to be incorrect.43 individuals/businesses to whom that data pertains. In September 2011, the task force published the General 1.5 The General Principles for Principles for Credit Reporting, a set of principles to guide Credit Reporting the development of reporting systems intended for policy makers, regulators, overseers, credit reporting data providers, While credit reporting systems are developing rapidly CRSPs, and consumers.45 In addition to the five core general throughout the world, until recently there were no principles, the task force identified and developed a set of worldwide standards to systematically guide stakeholders in specific roles for each of the stakeholders involved in credit dealing with the challenges associated with the development reporting systems, as well as recommendations for effective and operation of these systems. In 2008, the World Bank, oversight of these systems. with support from the Bank for International Settlements, launched the Credit Reporting Standards Setting Task Although the General Principles provide guidance, they Force to develop guidelines and universal standards do not endorse any model of credit reporting development for credit reporting systems. The task force comprised over another. IFC experience also suggests that the most representatives from central banks and other financial and effective solutions are tailored by taking into consideration privacy regulators, from multilateral organizations involved the general principles along with a country’s existing credit in credit reporting, and from international credit reporting market environment. For additional information, please service providers. The task force also benefited from the refer to the General Principles for Credit Reporting.46 significant experience of the IFC Credit Bureau Team. Some 43 In case of CRSPs collecting data to support the prudential supervision function, data subjects do not have the same rights in terms of access and disputing errors. 44 A full list of task force members and their organizations can be found in Annex 6 of World Bank 2011a. 45 See World Bank 2011a. 46 Ibid. 16 CHAPTER 2 Basics of Credit Reporting A CRSP is an entity that administers a networked market preferences, level of financial development, and credit information exchange. A networked credit credit culture. No single solution is more appropriate than information exchange is a mechanism enabling another for any given market. credit information collection, processing, and further disclosure to users of data, as well as value-added services 2.1.1 Credit Bureaus based on such data.47 Credit bureaus, which are typically privately held and CRSPs differ from credit rating agencies, such as Standard operated, provide credit information on individual & Poors, Moody’s, and Fitch, which collect financial borrowers and MSMEs to a wide range of credit providers. information on large companies; conduct detailed analyses They collect information in a standardized format from of operations, finances, and governance of the companies; a variety of credit providers, including banks, credit card and then issue credit ratings. CRSPs focus on smaller companies, and other nonbank financial institutions. creditors; they concentrate on credit repayment records They also collect and distribute a wide range of publicly and rely on statistical analyses of large samples of borrowers available information such as court judgments, bankruptcy rather than on in-depth analyses of individual companies. notices, and telephone directory information, and/or facilitate access to third-party databases such as collateral registries. Information is also gathered from contributors 2.1 The Taxonomy of Credit of nontraditional data such as retail lenders and billing Reporting Service Providers data from gas, water, electricity, cable, telephone, internet, and other services, which enables credit bureaus to The global credit reporting industry can be roughly divided compile better and more comprehensive credit reports. into three homogeneous, but not exclusive, groupings: According to Doing Business survey data, over 40 percent credit bureaus, credit registries, and commercial credit of credit bureaus included information from utility reporting companies. The content of databases, clientele, providers, and more than 50 percent included information and associated products and services provided by these from microfinance institutions in their databases (see Figure three types of service providers vary from country to 2.2).48 This broadening of sources of data is beneficial to country. Figure 2.1 shows the different markets served by unbanked individual borrowers and MSMEs because it these entities and the degree of overlap among them. There enables them to build a credit history without necessarily are distinct differences among the three types in terms having had formal access to credit, thus overcoming the trap of strengths and weaknesses, operating models, and the of not being eligible for credit without having a previous markets they serve. All three types of service providers can credit history. coexist in a given market based on the size of the market, 47 Ibid., see Glossary. 48 Doing Business Indicators (database), 2012, “Getting Credit” indicator. 17 Credit Reporting Knowledge Guide Figure 2.1: Markets Served by CRSPs Credit Registries and Commercial Credit Reporting Companies Credit Bureaus, Credit Registries, and Commercial Credit Reporting Companies Credit Bureaus and Credit Registries Consumers Consumers, Micro, Small and Large Corporations Medium Businesses Source: IFC 2012. Once data is collected, it is cross checked to produce a credit reports normally also include repayment histories, noting report for each borrower, which is then sold to lenders. The payments over a 12 to 24 month period. 49 Figure 2.3 shows report constitutes a comprehensive profile of a borrower or the types of individual-level information collected by credit potential borrower’s personal information and information bureaus. on his or her credit accounts. The personal information section usually includes the borrower’s name, former names, Borrower credit history is often recorded in terms of the identification number (such as social security or other number of missed payments in a format similar to the one national identification number), date of birth, addresses, in Figure 2.4. The credit report also provides information on employment information, alerts (such as ID theft reported collections made on outstanding accounts and any available or security freezes), and date of information update. The public records, such as court judgments and bankruptcy credit summary section typically contains information on rulings. In many countries, credit reports include a credit all the borrowers’ credit accounts (both open and closed), an score: the statistical probability of a borrower making good assessment of whether such accounts are in good standing on his or her obligation, based on a number of characteristics (past due amounts and payment behavior history), and (see Section 6.3). inquiries made about the borrower in the recent past. The 49 Ibid. 18 Credit Reporting Knowledge Guide Figure 2.2: Sources of Information for Credit Bureaus 100 91 90 80 80 71 Percent of Credit Bureaus 69 70 61 60 55 54 47 47 47 48 50 40 31 30 28 21 19 20 10 0 Private Public Development Credit union/ Finance Credit card Firms prov Retailers Utilities Other credit Microfinance Employers Courts Statistical Other commercial commercial banks coops corps issuers loans/trade (45) providers bureaus institutions (14) (20) agencies (29) banks (90) banks (65) (49) (53) (78) (66) creditors (46) (40) (25) (49) (11) Sources of information Sources: IFC calculation based on Doing Business Indicators (database) 2012. Figure 2.3: Individual-Level Information Collected by Credit Bureaus 100 91 90 84 83 79 80 70 Percent of Credit Bureaus 59 60 53 50 48 45 46 43 41 40 30 26 25 20 10 2 0 Name of Address Taxpayer National Ownership of Tax Individual's Utility Bad check Bancruptcies Court Credit Individual's Other individual ID ID a business statements income & payment list joudgments enquiries gender other personal records from other financial information lenders Types of information Source: IFC calculation based on Doing Business Indicators (database) 2012. 19 Credit Reporting Knowledge Guide Figure 2.4: Sample History of Payments business lending and advances in information technology, more credit bureaus are able to collate and sell reports History of Payments–Observation Periods on small businesses. According to the Doing Business 2012 2010 2009 survey data,50 more than 80 percent of the 100 credit MAMFJ DNOSAJ JMAMFJ DNOSAJ J bureaus responding contained at least some information on firms (see Figure 2.5). Collecting information on both 42 111 111 11321 2121 11 23145 individuals and firms in one credit bureau has the benefit of allowing a combined assessment of a business and its Most recent record Read Histories Oldest record of payment from right to left available owner. The credit history of a small-business owner is an 1 Payment on time important predictor of the credit risk of the small business, 2 Payment delayed between 1 to 29 days since small business owners often mix personal and business 3 and above indicate more delays in payment finances. Many individuals personally guarantee their Source: IFC 2012. business loans. However, in such cases there is a need to Reports are usually available to lenders electronically and consider and respect all appropriate laws and regulations on most large modern creditors have credit reports fed directly privacy rights, and ensure that personal data is only used for into their loan processing systems and originating software. permissible purposes specified in the legal and regulatory Lenders pay the credit bureau for credit reports in the framework or only provided to users that are legally allowed form of a subscription fee, a fee-per-query with significant to access such data. volume discounts, or a combination of both. While credit bureaus have access to a broad range of data Historically, credit bureaus only collected information on and provide a wide range of services to assist lenders in individuals. In recent years, with the expansion of small making lending decisions, the business model is usually Figure 2.5: Firm-Level Information Collected by Private Credit Bureaus 90 82 80 77 72 70 70 60 58 54 Percent of Credit Bureaus 53 52 50 45 42 40 30 27 20 20 16 17 10 0 Name of Firm Address Taxpayer ID Name of Field of Assets and Tax and Owner's Utility Bad check Bankruptcies Court Credit Other owner business liabilities income personal payment list judgments enquiries from income records other lenders Types of Information Source: IFC calculation based on Doing Business Indicators (database) 2012. 50 Ibid. 20 Credit Reporting Knowledge Guide based on voluntary contribution of information by data Argentina, Spain, Peru, Italy, and Belgium), the credit providers (typically involving a reciprocity arrangement). registry now offers similar products and services to credit In some jurisdictions, usually in the formative stages of the bureaus. credit reporting environment, there can be resistance to the concept of sharing information from some potential data Generally, all regulated financial institutions are mandated providers, most commonly from larger institutions that to provide data to the registry (see Figure 2.6). In return, are unwilling to share customer data for fear of losing their the registry provides a credit report to all reporting market share. In these circumstances, the authority of the regulated financial institutions, which shows current central bank as overseer of the credit reporting system or aggregate exposures of regulated entities. Credit registry supervisory authority of regulated lenders, through its ability coverage tends to be limited by the scope of data providers to persuade participation in a data-sharing environment, (regulated lenders only). Because the operator of the can have a profound catalytic effect on establishing good registry (frequently a central bank or other authority) has practices. A trend seems to be developing for jurisdictions no regulatory purview over nonregulated sectors, such as to mandate regulated entities to share data and use the nonregulated MFIs, the information provided by a registry, services of credit bureaus. According to Doing Business while sufficient for prudential supervision and monitoring survey data,51 42 percent of the respondents said that the purposes, might not meet user needs for credit granting law required mandatory reporting to the credit bureau at processes. least by banks and 39 percent said financial institutions (including banks) were required to consult with a bureau. Credit registries collect information on both individuals Along with mandating participation, the regulatory body and firms. Individual-level information typically includes must also be empowered to enforce participation and identification data, loan type and characteristics data, monitor compliance. negative data, collateral and guarantee data, and payment history data (see Figure 2.7). 2.1.2 Credit Registries Firm-level information collected by registries typically Historically credit registries served a different purpose from includes identification data, business owner data, loan type credit bureaus. Most credit registries started out as internal and characteristics data, negative data, and payment history databases within a country’s central bank and were, and data (see Figure 2.8). in many cases still are, used as a supervision mechanism to identify systemic risk within the lending portfolios of Credit registries usually provide their credit reports at low regulated financial institutions. As such, these databases or no cost to the lenders. Of the 82 credit registries that focused primarily on large credit exposures, typically provided information to the Doing Business survey on their with a loan threshold value in excess of $5,000.52 Initially cost to inspect data, only 14 listed a fee.53 information in credit registries was used solely for internal purposes, but, in the absence of other credit reporting 2.1.3. Commercial Credit Reporting service providers in many countries (including China, France, Malaysia, and Indonesia), over time information Commercial credit reporting companies provide from credit registries has been made available to regulated information on companies, including sole proprietorships, lenders in the form of credit reports. With the growth partnerships, and corporations, available through public of consumer credit, the loan value thresholds have been sources, direct investigations, and payment behavior reduced or abolished, and, in some countries (e. g., France, reported by suppliers and trade creditors. Commercial 51 Based on Doing Business Indicators (database), 2012, “Getting Credit” indicator. 52 Ibid. 53 Doing Business Indicators (database), 2012, “Getting Credit” indicator. 21 Credit Reporting Knowledge Guide Figure 2.6: Sources of Information for Credit Registries 120 99 100 80 Percent of Credit Registries 70 71 67 60 40 36 32 26 23 20 10 7 5 0 1 1 2 0 Private com Public com Dev banks Credit Union/ Finance Credit Card Firms prov Retailers Utilities Credit Microfinance Employers Courts Stat agencies Other banks Banks (64) coops corps issuers loans/trade (1) providers Bureaus institutions (2) (7) (5) (23) (96) (64) (34) (69) (22) creditors (0) (1) (9) (30) Sources of Information Source: IFC calculation based on Doing Business Indicators (database) 2012. Figure 2.7: Individual-Level Information Collected by Credit Registries 100 91 90 86 80 70 63 64 Percent of Credit Registries 60 50 42 40 37 38 33 30 30 20 16 16 11 10 2 2 0 Name of Address Taxpayer ID National ID Ownership of Tax Individual's Utility Bad check Bankruptcies Court Credit Individual's Other individual a business statements income & payment list joudgments enquiries from gender other personal records other lenders financial information Source: IFC calculation based on Doing Business Indicators (database) 2012. Types of Information 22 Credit Reporting Knowledge Guide Figure 2.8: Firm-Level Information Collected by Credit Registries 100 95 90 80 74 69 70 Percent of Credit Registries 60 50 50 42 40 29 29 30 27 23 20 17 10 6 6 3 4 0 Name of firm Address Taxpayer ID Name of Field of Assets and Tax and Owner's Utility Bad check Bankruptcies Court Credit Other owner business liabilities income personal payment list judgments enquiries from income records other lenders Types of Information Source: IFC calculation based on Doing Business Indicators (database) 2012. credit reporting companies cover companies that are Commercial credit reporting may include small businesses smaller in size and earnings than corporations covered by although the information is often limited because the credit rating agencies. The information compiled through reporting format is inappropriate for small companies. As these commercial credit reporting companies is typically discussed earlier, assessment of small businesses benefits from used for credit risk assessment or credit scoring, or for other the link to personal owner records, since small businesses purposes such as the extension of trade credit. tend not to produce publicly available financial statements. However, commercial credit reporting companies do not Commercial credit reporting differs from credit reporting collect personal data on individuals. In addition, the cost of on consumers in several ways. The information held by a report on a micro or small enterprise is likely to be high commercial credit reporting companies does not include in relation to the loan size. For this reason, micro and small sensitive personal information on individuals. The size businesses tend to be better served within the framework of of transactions reported to commercial credit reporting a credit bureau. companies is also significantly larger. Commercial credit reporting requires significantly more payment performance The international leader in commercial credit reporting and financial data than would be required of individual is Dun & Bradstreet, which traces its roots back to the borrowers. Whereas credit bureaus disclose the identity of Mercantile Exchange established in New York City in data providers to protect the rights of individual borrowers, 1841. Formerly, the company delivered its reference books commercial credit reporting companies do not disclose the to subscribers under lock and key. Today, it transmits identity of data sources or data recipients to their client credit information on more than 140 million businesses businesses. worldwide electronically. Coface, the second largest 23 Credit Reporting Knowledge Guide international credit risk insurer, entered the international • Credit bureaus that are partially owned by government market by building on its database of payment behavior of entities hundreds of thousands of medium-sized companies, which • Credit bureaus that are wholly owned by government it built through its credit risk insurance business. entities (this is very rare). Although the scope of the general principles for credit According to the Doing Business survey data,54 of 106 reporting covers commercial credit reporting companies, credit bureaus in 100 countries around the world, not all principles are applicable to them. IFC’s experience to approximately 44 percent had no ownership by banks, date has been largely focused on supporting credit reporting financial institutions, or credit card providers; 39 percent for consumers, micro, small, and medium sized businesses, were owned by banks, financial institutions, or credit card served through credit bureaus and credit registries. This providers; 12 percent were held by industry associations or Guide does not further discuss aspects of commercial credit chambers of commerce; and only 4 percent were partially reporting or of credit ratings agencies. held by governments (see Figure 2.9). 2.2 Ownership Structures Figure 2.9: Ownership Structures of Credit Bureaus Partially government Credit bureaus and credit registries normally serve separate owned 4% Government functions. Whereas the former generally focus on making owned Industry Associations/ information available to financial and nonfinancial Chambers of 1% creditors for credit-granting purposes, the latter typically Commerce 12% focus on assisting banking supervision while improving the quality and availability of data for supervised financial Private not owned by banks/ intermediaries. However, there are instances of bureaus Financial supporting banking supervision and instances of registries Institutions/ making data available to creditors in the market. Private owned by other credit banks/Financial providers Institutions/other 44% credit providers Based on the broad categorization described above, credit 39% registries are mostly owned and operated by public sector entities such as a central bank or other monetary/financial supervisory authority, as these entities are directly responsible Source: IFC calculation based on Doing Business Indicators (database) 2012. for prudential supervision and risk monitoring functions in an economy. Conversely, depending on its function and the 2.2.1 Owners and Shareholders range of stakeholders involved, the ownership structure of a credit bureau can fall into one of many categories: Owners and shareholders are generally the investors that provide seed capital for a CRSP, negotiate and prepare • Credit bureaus in which banks and/or other creditors are the pre-incorporation agreements, lease or acquire office either majority or minority shareholders premises, and contract for the initial needs of the CRSP, • Credit bureaus owned and operated by a separate entity such as acquiring technological assets and recruiting the with no ownership by creditors necessary personnel to manage the day-to-day operations. They may also be users of the service, for instance when • Credit bureaus formed on the basis of an association the CRSP is owned by member banks. Potential owners or chamber of commerce that mostly operate on and shareholders may include banks, technical partners, membership fees 54 Ibid. 24 Credit Reporting Knowledge Guide government bodies, and private investors, as described owned bureau should obtain commitment from as below. many lenders and data sources as possible before starting operations. Another potential challenge for independently • Banks and other financial institutions: It is not owned bureaus is limited capital (with no shareholder banks uncommon for banks to be shareholders in CRSPs, either to provide back-up capital injection if necessary). individually or as a group. For example, the Association of Banks in Singapore owns a share in Singapore’s credit In several countries, such as Argentina, Brazil, Croatia, bureau. Other countries in which a group of banks Germany, Romania, Turkey, Mexico, the Russian owns credit bureaus include Brazil, Croatia, Kazakhstan, Federation, Kazakhstan, and Ukraine, bureaus have included Poland, Romania, Serbia, Turkey, and Saudi Arabia. ownership by creditors. The advantage of this ownership • Private individuals, technical partners, or other structure is that it allows for a faster startup because entities: Individual entrepreneurs or organizations agreement among the banks to become shareholders brings that provide venture capital for the establishment of about a strong commitment to the principle of reciprocity CRSPs may also become shareholders. Examples are in information sharing. Furthermore, the commitment by DP Information in Singapore, Datacheck in Pakistan, existing lenders promises long-term financing. Participation CompuScan in South Africa, and CRB Africa in Kenya. by a government authority may also add credibility to the Many CRSPs have technical partners to manage the venture. Including a technical partner as a shareholder information technology requirements of the reporting allows the credit reporting service provider to better align system. It is not uncommon for technical partners to its incentives and focus on efficiency in operations. hold an ownership share in the company. • Government: In some countries, government entities Bureaus that include ownership by creditors face challenges (e.g. the central bank in Sri Lanka), or public sector as stakeholder differences can get in the way of maximizing financial institutions (e.g. India and Thailand), become business value. Creditor shareholders may be reluctant to shareholders in the credit bureau. In Dubai, the credit allow new creditors to participate because these newcomers, bureau is fully owned by government entities. while unable to contribute significant amounts of data, would benefit greatly from having access to information Credit bureaus in which creditors have no ownership, on existing clients. Furthermore, when creditors own the such as in the United States, Kenya, and New Zealand, credit bureau, they are less likely to use the services of are generally efficient structures because credit reporting any other bureau, thus increasing barriers to entry in the is their core business, and the shareholders’ main objective credit reporting market. If a few banks are shareholders, is to maximize business value by expanding operations whereas several other banks or nonbank creditors such as and providing new and innovative products and services. microfinance institutions are nonshareholding users of the Conflicts of interest are minimal because the bureau’s bureau, it is possible that shareholding banks may influence relationships with its members and users are driven by the pricing policy in a manner that penalizes nonshareholder commercial interests. users. Such unfair practices can be avoided if ownership by individual creditors is limited. Although this is the ideal structure, it is often not feasible in countries developing their first credit reporting service Partial government ownership in credit bureaus is rare, but because lenders are often reluctant to share information has been seen in some markets, when equity investment by unless they are shareholders (and therefore share control). government helps boost private investor confidence. For Often, lenders resist sharing their customer information instance, Sri Lanka’s Credit Information Bureau (CRIB) with a newly established bureau until they are certain that was established as a public-private partnership in which other lenders will do the same. Therefore, an independently the central bank originally held a 49 percent equity stake, now reduced to 19 percent.55 In India, the State Bank of 55 Ibid. 25 Credit Reporting Knowledge Guide India and the Housing Development Finance Corporation and other creditors has been a growing trend in emerging Ltd. (HDFC) were majority shareholders (40 percent each) markets. However, in IFC’s experience, as lenders gain of the Credit Information Bureau (India) Ltd. (CIBIL) more trust in the operations of a credit bureau, they tend to when it was first established in 2006. At the time, Dun & divest their shareholdings (e.g. in Hong Kong SAR, China, Bradstreet and TransUnion held the remaining 20 percent and the Dominican Republic). Table 2.1 summarizes the (10 percent each).56 Over the years, other banks joined different ownership structures and the advantages and as shareholders and State Bank of India and HDFC have disadvantages of each. reduced their holdings. Credit bureau ownership by banks Table 2.1: Comparison of Credit Bureau Ownership Structures Commercial, with Commercial, Non-commercial, Government ownership ownership by creditors no ownership by creditors creditor association (partial or full) Pros • Often the only feasible • No conflicts of interest • Boosts confidence of way to establish a credit in management private sector, creditors bureau and ensure buy-in and technical partners. from lenders • Lender support implies • Commercial outlook strong commitment and ensures innovation and ensures bureau sustain- high-quality service ability • Technical partners en- • Open for broad market hance the credit bureau’s coverage. creditworthiness • Commercial outlook ensures innovation and high-quality service. Cons • Conflicts of interest are • Banks generally are not • Limited incentives to • Not efficient use of possible, where existing willing to share data innovate government resources. shareholders resist the without taking ownership entry of new lenders in a bureau to the credit bureau or the introduction of new services • Slow decision process as • Lack of capital. • Usually lower quality • Government as diverging views of large of service than in a shareholder creates numbers of shareholders for-profit bureau conflict of interest need to be accommo- between supervisory and • Slow decision process. dated shareholder functions. • Barriers to entry for new providers as well as new members. Examples • CRIF (Italy) • Equifax (US, Spain) • Common in Latin • Sri Lanka • CIG (Iceland) • Experian (US, UK) America, where • Thailand Chambers of Commerce • SCHUFA (Germany) • TransUnion (US) operate lists of bad • United Arab Emirates. • Serasa (Brasil) • Compuscan (South Africa) debtors. • SIMAH (KSA) • Datacheck (Pakistan) Source: IFC 2012. 56 Credit Information Bureau (India) Ltd. (CIBIL), http://www.cibil.com/web/promoters.htm 26 Credit Reporting Knowledge Guide 2.3 Optimal Market Size information industry include the United Kingdom, Italy, India, and Chile. Germany, Austria, and most smaller Credit bureaus are characterized by network externalities European countries have only one major credit bureau. and economies of scale that could classify the market as a natural monopoly. Ongoing debate on the optimal number Whereas the number of bureaus differs based on each of CRSPs in a market has not produced consensus thus country’s needs, most countries have only one credit registry. far. On the one hand, a single credit reporting service In some instances, the credit registry is divided into separate combining aggregated information across the entire databases covering data on individuals and firms (e.g., system and including both bank and nonbank credit Tunisia and Belgium); separate databases for positive and information would provide lenders with the most complete negative information (Algeria); or separate databases that set of information, including comprehensive inquiry are used for monetary supervision and to provide reporting information. On the other hand, the lack of competition to regulated financial institutions (France and West Africa’s eliminates incentives for such a provider to improve data Banque Centrale des États de l’Afrique de l’Ouest). quality, provide value-added services, and lower prices. Indeed, bureaus and registries are by no means mutually Competitive credit information industries are more exclusive, and in several countries they exist side by side. common in large markets that can support more than one In those instances registries typically assist the financial CRSP. In the United States, for example, consolidation supervisors in prudential supervision and risk monitoring in the financial services industry over the past 20–30 and often provide comprehensive reporting back to years has resulted in three major credit bureaus operating regulated financial institutions. Bureaus largely support the concurrently and competing on the basis of product and credit reporting needs of financial and nonfinancial creditors service differentiation. In South Africa, the three major and often provide statistical information to the supervisor credit bureaus all contain information from the same banks, or credit registry to assist with the prudential supervision but compete on the quality of information and on value- function (e.g., Kenya, Uganda, and South Africa). added services. Other countries with a competitive credit 27 CHAPTER 3 Credit Reporting for Micro, Small, and Medium Enterprises E xpanding access to finance to low-income household as a development tool, and in part by promotion by national consumers and MSMEs is on the development governments, international development bodies, donors, agenda of most emerging markets. Microfinance is and socially oriented investors. It is estimated that the broadly defined as the provision of financial services to low- number of borrowers served by microfinance institutions income clients, including consumers and self-employed (MFIs) is as high as 120 million to 190 million.58 MFIs individuals, who traditionally lack access to banking and can be registered institutions such as nongovernmental related services. It is perceived as an important mechanism organizations (NGOs) and cooperatives, or small nonprofit in expanding access to finance as it focuses on lower- organizations, and are only one of the various types of entities income clients, who are often self-employed, household- that offer microfinance services. In addition to MFIs, many based entrepreneurs. Their diverse microenterprises include commercial lenders and consumer lending companies are small retail shops, street vending, artisanal manufacture, moving downstream and offering microfinance products, a and service provision. In rural areas, microentrepreneurs move that highlights the huge potential that this market often have small income-generating activities, such as food segment holds for commercial lenders. The microfinance processing and trade, and some are farmers.57 These clients business model is designed to fit the financing preferences usually have informal or no business records, no collateral, of low-income consumers and entrepreneurs through the and no access to formal credit markets. use of low-value, short-tenure loans with several installment payments. Small and medium enterprise (SME) finance can be distinguished from microfinance in two respects: first, During the years of rapid growth, from early 2000 to about it covers a wider range of entrepreneurial clientele, and 2008, microfinance enjoyed a reputation for strong asset second, SMEs are often larger and therefore represent a quality and low delinquency rates. Over the past three to bigger risk exposure to lenders than microfinance clients. four years, however, the quality of microfinance portfolios SMEs require a more in-depth credit review process than has deteriorated worldwide, with increasing portfolios-at- microfinance clients. Still, SMEs often fall into the middle risk values. Several factors contributed to this deterioration, market for lenders—too big for traditional microfinance such as inadequate risk management systems and controls, yet too small for mainstream banks. internal organization weaknesses, and excessive growth in narrow geographies, combined with unhealthy lending practices that affected borrower repayment incentives and 3.1 Microfinance behaviors. All these factors resulted in overindebtedness as witnessed in several markets, including Morocco, Bosnia Over the past five to ten years, microfinance has grown and Herzegovina, Pakistan, Egypt, Cambodia, and India.59 rapidly, driven in part by the global recognition of its value 57 Microfinance Gateway, http://www.microfinancegateway.org/p/site/m/template.rc/1.26.12263/#1 58 Lyman et al., 2011. 59 Ibid. 29 Credit Reporting Knowledge Guide The search for solutions to this crisis has highlighted the Microfinance has been brought into the ambit of credit critical role that credit reporting can play in averting reporting through different methods. In some cases, credit or reducing the risks of overindebtedness and borrower bureaus have expanded their databases to include payment defaults. Although credit reporting alone cannot create data on low-income individuals (and expanded their data credit discipline in a market, or compensate for inadequate subject and customer bases to include data from MFIs); underwriting standards, it can improve microlenders’ in other cases, credit registries have incorporated nonbank abilities to originate loans and manage credit risk and microlenders; in yet other instances, MFI-specific client it creates a powerful incentive for repayment among databases have been created. Each of these methods is borrowers. described briefly below: Traditionally, CRSPs served the mainstream banking and Credit bureaus: Credit bureaus collect a wide range of consumer lending sectors, leaving low-income consumers information on a borrower’s financial and credit situation, and MSMEs outside the credit reporting infrastructure. including information from nonfinancial institutions and Credit information products developed for more traditional MFIs. Whereas most low-income earners may not have lenders like commercial banks were designed to deal with formal salaries or other banking history information, they larger loan values—and with longer tenures and less frequent may have utility or telecommunications accounts and repayment schedules—than MSME loans. Microlenders payment histories from other credit providers including tended not to have the data required to populate credit retailers and MFIs. Credit bureaus have the ability to collect reporting databases, and collecting this data would require this information, collate it, and provide it to microlenders significant efforts and costs to create the right infrastructure for a range of purposes, for example, to indicate the relative and capacity. In countries where bureaus or registries creditworthiness of the borrower, assess risk, verify identity, reported collecting information from microlenders, they collect debt, and monitor fraud. Credit bureaus offer dealt with larger, regulated microlenders, which still left out the largest coverage and are in a position to offer broad a large portion of microlenders lending to those at the base information to MFIs. of the pyramid.60 In India, an MFI credit bureau project was launched in A credit reporting service provider that caters to the needs of June 2009 through IFC and local sponsors working closely microlenders offers the following benefits to these lenders: with the existing credit bureaus. As of the end of 2011, approximately 45 MFIs were reporting to credit bureaus and • Better methods for risk management as sharing credit approximately 55 million client records had been uploaded. information reduces a lender’s uncertainty about a Currently, more focus is being placed on increasing the borrower’s exposure, and at the same time protects the reach of MFI reporting to credit bureaus, greater use of borrower from becoming overindebted credit information, and client awareness regarding credit • Significant cost reductions as lenders attain the ability to reporting (see India case study, Chapter 7, section 7.4). In screen borrowers quickly Egypt, Morocco, and Nepal, discussions are underway to • Increased lending volumes over time due to automation link MFIs to existing credit bureaus. In Bangladesh, IFC of credit granting decisions is working with an association of microfinance institutions to support credit reporting for MFIs. The most likely • Promotion of responsible borrower behavior as the solution will be a bureau that caters specifically to MFIs lenders’ clients develop awareness of the use of credit and that would be supervised by the country’s Microcredit reporting and how maintaining a good credit history is Regulatory Authority. beneficial for them. Credit registries: Credit registries usually mandate reporting and compliance by regulated financial institutions, including 60 Ibid. 30 Credit Reporting Knowledge Guide regulated microlenders. Consequently, microlenders that In Mozambique, the central bank maintains a registry of are not regulated are usually excluded from credit reporting all loans, including microfinance loans, which is online requirements. In some countries, credit registries continue and available to MFIs for a fee. However, the quality of to maintain thresholds for loan reporting values, which information is weak and not suitable for rural MFIs with implies that microloans of low value would not be captured limited connectivity. In Rwanda, the credit bureau collects by these registries. However, in many countries oversight and distributes data to regulated MFIs for a fee. MFIs are and regulation of microlenders is growing, particularly in required to report loans above a certain threshold (over light of the recent spate of crises in this sector, thus bringing $300) to the bureau. In Uganda, the credit bureau serves the more microlenders under formal oversight of the regulatory regulated microfinance sector, however, it leaves out a large bodies and thus improving their chances of being a part of number of informal creditors that serve the microfinance the credit information sharing system. market. In other countries, such as Nigeria and Tanzania, the legal framework provides for the participation of regulated MFI-specific client database: In some situations, MFI- microfinance providers in credit reporting systems. In specific client databases are generated in the absence of Tanzania, no service providers are currently operational, and any formal framework or structure for sharing of credit in Nigeria, shortcomings in the legal framework impede the information among MFIs on their clientele. These databases proper functioning of bureaus. In Ghana, although bureaus are generally limited in scope and tend to cover only negative are operational, microfinance lenders are not mandated to information on MFI clients. Further they do not include participate and avoid participation because of the high costs information from other possible lenders to these clients, involved. Madagascar has a credit bureau for microfinance thus presenting a fragmented picture of borrower credit institutions, which can receive data for free. MFIs, particularly history. However, they are generated to fulfill a specific need smaller MFIs, face reporting constraints due to infrastructure at a specific point in time, and generally evolve over time and connectivity issues. Information sharing is fragmented, to become more comprehensive credit reporting service however, as a separate bureau exists for commercial lenders.63 providers, as was the case of Sin Riesgos in Nicaragua61 and the FINRURAL system in Bolivia (before it became a As can be seen, the microfinance sector in much of Africa part of the credit bureau, Infocred). still faces substantial capacity obstacles such as connectivity, lack of enabling frameworks, and fragmentation, among other things. There is also reluctance on the part of MFIs to 3.2 Regional Developments in share customer data, either with other MFIs or mainstream Microfinance Credit Reporting lenders because they fear their best customers would be poached by competitors. Because of these challenges, newly Sub-Saharan Africa (SSA) is a region with a rapidly established CRSPs do not see MFIs as an attractive market growing microfinance sector with over 700 MFIs, 5.3 segment compared with the formal banking sector and million active borrowers, and a gross loan portfolio of $7.2 other data providers and users such as telecommunication billion.62 However, credit reporting in the microfinance companies and utility providers. sector is still in its infancy because credit reporting systems in general, (even for mainstream lenders) are relatively East Asia and the Pacific (EAP) had over 400 MFIs with new and still maturing. South Africa is the only exception, 14.9 million active borrowers and a gross loan portfolio of with a relatively sophisticated MFI credit reporting system. $36.9 billion in 2011.64 In the Pacific Islands, only four Some of the challenges faced by lenders participating in countries (Fiji, Papua New Guinea, Tonga, and Vanuatu) credit reporting systems are described below. had credit bureaus, all established within the past 10 years. 61 Ibid. 62 Mix Market estimates as of September 2012. Based on 31 countries reporting data to MixMarket. http://www.mixmarket.org/mfi/region/Africa 63 Economist Intelligence Unit 2012. 64 Based on MixMarket data for 12 countries reporting in 2011, http://www.mixmarket.org/mfi/region/East%20Asia%20and%20the%20Pacific 31 Credit Reporting Knowledge Guide In Fiji and Papua New Guinea, some MFIs have joined the established a private credit bureau, which, once operational, credit bureaus; and, as the newly established credit bureaus will collect information from microfinance lenders. in Tonga and Vanuatu progress, it is expected that MFIs will soon become members. Microfinance credit reporting could benefit from further strengthening in the region with regards to explicit regulation In East Asia excluding the Pacific Islands, credit reporting of MFIs to enable their participation in the credit reporting for microfinance is still a novel concept and occurs only system, greater investment in technology platforms for in Cambodia, which launched a private bureau in March MFIs to enable credit reporting, capacity building of MFIs 2012. Regulated MFIs are required by law to contribute to use credit information data for underwriting and risk information on all their loans to the credit bureau; and they management, and encouraging service providers to provide must request a credit report for each new credit application more MFI specific products. or renewal of existing facilities, regardless of the amount of the loan. The very small nonregulated MFIs have not yet Middle East and North Africa (MENA) has a microfinance joined the credit bureau. market that is smaller than in other regions with over 80 MFIs serving 1.1 million borrowers and a gross loan Eastern Europe and Central Asia (ECA) had a microfinance portfolio of $901.3 million.66 Because of political changes in market that was served by over 400 MFIs, with 2.3 million the region, particularly in Egypt, Tunisia, and the Republic active borrowers and a gross loan portfolio of $8.8 billion of Yemen, there has been an increase in nonperforming in 2011.65 The region’s microfinance sector faced significant loans. Of the 20 countries in the MENA region,67 only challenges over the past five years, in particular in Central six have either a credit bureau or a registry. The majority Asia where overindebtedness became problematic, in part of these serve only regulated financial institutions, which due to lack of mechanisms for systemic risk management, are mainly banks. Recently, Egypt, Tunisia, Morocco, and insufficient information in credit reporting structures (credit Jordan, among others, have taken steps to integrate MFIs registries and credit bureaus), and ineffective supervision. into their formal credit reporting systems. Some of the challenges in credit reporting for the sector are described below. In Egypt, I-Score (Egypt’s credit bureau) and the Egyptian Microfinance Network (representing MFIs in Egypt) are In Armenia, MFIs report and obtain services from the working in partnership to integrate data from MFIs into private credit bureau. Although many credit reporting I-Score’s database, which already includes credit data of structures (for instance, in the Kyrgyz Republic, Bosnia regulated and nonregulated financial institutions. This and Herzegovina, and Moldova) also collect information agreement was reached after I-Score conducted a pilot test from the microfinance sector, coverage is low and data using samples of data from MFIs. The results showed a high quality continues to be an issue. In Azerbaijan, regulated level of cross lending between microlenders and banks, MFIs are mandated to report their loan information to the and to a smaller extent within the microlending sector credit registries and also share information with the credit itself. These results encouraged MFIs to agree to share their bureaus on a voluntary basis. In the Kyrgyz Republic, customers’ credit information. (For a detailed discussion of MFIs participate in the bureau, but participation is not the I-Score model, see the Egypt case study in Chapter 7.) mandatory and data quality remains an issue. Tajikistan has 65 Based on MixMarket data for 20 countries reporting in 2011, http://www.mixmarket.org/mfi/region/Eastern%20Europe%20and%20 Central%20Asia 66 Based on MixMarket data from 10 countries reporting for 2011, http://www.mixmarket.org/mfi/region/Middle%20East%20and%20 North%20Africa 67 The MENA region, as defined by IFC, includes: Afghanistan, Algeria, Bahrain, Egypt, the Islamic Republic of Iran, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Pakistan, Qatar, Saudi Arabia, Syria, Tunisia, United Arab Emirates, West Bank and Gaza, and the Republic of Yemen. 32 Credit Reporting Knowledge Guide A similar pilot program was conducted in Pakistan. Led in Morocco and Egypt) help to narrow the differences in by the Pakistan Microfinance Network (focusing on the objectives between credit reporting service providers on the Lahore microfinance market), 15 MFIs submitted data to one hand, and microlenders on the other hand. one of Pakistan’s private credit bureaus. The pilot proved successful, with the MFIs reporting benefits including South Asia (SAR) represents the largest market for reduced lending costs and a decrease in nonperforming microfinance borrowers with 75 percent of the world’s loans. Accordingly, it has been agreed that the program will microfinance borrowers, or 74 million borrowers. India and be replicated nationwide. It is expected that this initiative Bangladesh make up the biggest markets for microfinance will eventually help MFIs build capacity to make better credit in South Asia.68 The chief challenges in developing decisions leading to healthier portfolios and, consequently, microfinance credit reporting systems in South Asia are a more inclusive and risk-balanced microfinance market in similar to those in other regions, including lack of capacity Pakistan. on the MFI end (systems and technology), data quality issues (e.g., inadequate identification information and In Morocco, all MFIs are mandated to provide data every incomplete information), connectivity and infrastructure month to the private credit bureau. However, the cost of issues, affordability of credit reporting products for MFIs, accessing the bureau’s reports has been cited as a prohibitive lack of enabling legal and regulatory framework, and factor in stimulating MFI uptake of the system. Prior to the political stumbling blocks. development of the bureau, smaller MFIs formed a network, Réseau de la Microfinance Solidaire, to share information In India, the 2010 microfinance crisis in Andhra Pradesh, among themselves. The Réseau de la Microfinance Solidaire provided necessary impetus to the microfinance credit helped develop a common format for data sharing with the reporting agenda, which had already started to take shape bureau (once it was established) and retrieving information in mid-2010. In May 2011, two of the four licensed credit from the bureau through the central bank. The three major bureaus in India started providing credit reporting services MFIs already have CPU-to-CPU connections in place. for microfinance institutions and reported 67 million loan The central bank played an important role in brokering a accounts at the end of that year (see the India case study pricing negotiation between the bureaus and MFIs. in Chapter 7, section 7.4.) Work is underway to develop a credit reporting system for microfinance institutions Like the other regions, MENA faces its share of challenges in Bangladesh, but these efforts are still years away from to integrating MFIs into the credit reporting system. fruition. Common obstacles include enabling legal and regulatory frameworks that support data sharing by MFIs, MFI Latin America and the Caribbean (LAC) has a large resource constraints, telecommunications infrastructure, microfinance market with over 500 MFIs serving 18.1 and internet connectivity issues among others. MFIs million borrowers and a gross loan portfolio of $27.7 are reluctant to share data for fear of losing their best billion.69 Of all regions, Latin America and the Caribbean is customers. Pricing plays a big factor in the decision of MFIs the most advanced in terms of credit reporting systems for to participate in a credit reporting system. Bureaus tend the microfinance market. to charge microlenders prices that they would normally charge banks or other large lenders for products that are Ecuador, Bolivia, and Peru are widely known for their not entirely tailored to meet MFI needs, thus repelling progressive credit reporting systems incorporating potential microlender clients. Discussions between the information from the regulated and nonregulated CRSPs, microlenders, and regulators (as has taken place microfinance sectors. The Ecuador model, (see Ecuador case 68 Mix Market 2012. 69 Based on Mix Market data for 21 countries reporting for 2011, http://www.mixmarket.org/mfi/region/Latin%20America%20and%20 The%20Caribbean 33 Credit Reporting Knowledge Guide study in Chapter 7, section 7.1) has been followed in other addition to other obstacles), and called on authorities countries. Recent changes in the credit reporting market in to design policies to overcome these challenges.72 Ecuador, however, stand to undermine the effectiveness of the system. Historically, small business borrowers have represented a difficult market to serve because of the traditional high-cost In Mexico, MFI credit reporting services are provided of subjective credit evaluation. The SME business owner’s through two credit bureaus. Although the law does not personal finances are often comingled with those of the mandate it, funders of microfinance institutions increasingly business, and this distinction is not immediately apparent require these institutions to consult bureaus in an attempt to lenders. The difficulty in assessing the creditworthiness to control portfolio quality and indirectly check potential of SME businesses causes lenders to adopt protective overindebtedness. The market for microfinance credit measures, such as imposing higher interest rates, requesting reporting remains fragmented, however, as the two bureaus substantial collateral, or denying credit altogether to SME do not currently share data with each other, although recent borrowers. The financial services company Wells Fargo regulatory changes are attempting to address this issue. pioneered the adaptation of consumer lending technologies to small business lending in the 1990s in the United States. It established that the business owner’s consumer credit 3.3 Small and Medium history proved highly predictive of the credit performance Enterprise Finance70 of that business. This and other innovations in small- business lending have since been adopted widely in Access to finance is a key constraint to SME development developed countries and have also begun to find their way and growth, especially in emerging markets. In their early into emerging markets. stages, SMEs are often financed internally by the owner’s savings or earnings. Sustained growth, however, usually In the United States, an industry consortium launched requires external funding. A 2003 World Bank study71 small business credit reporting only in 2002. Several that looked at data from 5,000 firms across 51 countries emerging market credit bureaus in Thailand, India, Turkey, found that in countries without credit bureaus, 49 percent and Saudi Arabia incorporated provision of small business of small firms reported significant financing constraints, credit reporting into their business plans early on to avoid as opposed to 27 percent in countries that did have credit the mistakes of their more developed counterparts. In bureaus. The study also found that in countries with Singapore, the SME Credit Bureau was created in 2002 and credit bureaus, the probability of a small firm obtaining a became fully operational in 2005 as one of the first credit bank loan was 40 percent versus 28 percent in countries bureaus in the region that collates data with both consumers without credit bureaus. A more recent World Bank study and small and medium enterprises (SMEs). In India, the based on data from 99 developing countries found that country’s first rating agency focusing primarily on MSMEs small firms are large contributors to total employment was created in 2005 to improve credit flow to the sector.73 and job creation, but have lower productivity growth than larger firms because they are hampered by inadequate Drawing on lessons learned from 164 case studies, the financial infrastructure and regulatory environments (in SME Finance Sub-group of the G-20 and the Global 70 This Guide focuses on credit reporting systems. Accordingly, the discussion in this section focuses on best practice in establishing and develop- ing credit reporting systems to improve SME access to finance. For a comprehensive look at SME finance, please see IFC 2010, SME Banking Knowledge Guide. Also visit SME Finance Forum, an initiative launched by IFC as a knowledge-sharing platform for data, research, and SME best practices, at www.smefinanceforum.org. 71 Love et al., 2003. 72 Meghana et al., 2011. 73 For a more detailed discussion on Singapore’s SME credit bureau and India’s SMERA, see IFC 2010, SME Banking Knowledge Guide, 34–36. 34 Credit Reporting Knowledge Guide Partnership for Financial Inclusion issued a 2011 report74 collateral than movable collateral, they are unable to meet that identified a number of policy related issues collateral requirements to secure a loan. Lenders lose out constraining SMEs’ access to finance, not the least of which as well, as they are unable to tap into the huge borrowing was weak credit reporting systems. Among the report’s base of SME and microborrowers with movable assets. recommendations for scaling up SME financing and for • Weak legal and regulatory frameworks surrounding establishing an enabling environment for SMEs were: the use of collateral can present a challenge to lenders • Develop country specific diagnostics and strategies in collecting debts. If legal enforcement mechanisms • Develop a supportive legal and regulatory framework are weak or ineffective, the costs to lenders of pursuing delinquent debtors are increased. Faced with the • Strengthen the financial infrastructure (including credit potential of higher costs incurred in obtaining a legal information systems, secured transactions, and payment remedy, either through the judicial system or extra- systems) judicial processes, lenders may choose to grant credit at • Design effective government support mechanisms unfavorable terms to SME borrowers and microclients, or deny them credit altogether. • Build consistent and reliable data sources on SME finance The first challenge can be addressed by the creation of • Build capacity of financial institutions to cater to collateral registries, public databases that register interests SMEs. in or ownership of assets. Lenders can consult a registry to ascertain ownership of assets, and whether or not a particular piece of collateral has any potential claims against 3.4 Credit Reporting Systems and it. The registry enables potential borrowers to establish Secured Transactions the legitimacy of their collateral in securing a loan. The second challenge can be addressed by developing strong As identified in Section 3.3, the challenges of lending to legal and regulatory frameworks to facilitate these secured the SME segment prompts lenders to impose high collateral or “collateralized” transactions. This aspect is beyond the requirements on SME businesses to secure their loans. Both scope of this Guide and is not discussed further. lenders and SME borrowers are faced with challenges, however, when it comes to granting and taking credit Generally, collateral registries collect information only against collateral. The two main challenges are: on certain classes of movable or immovable property of • In most jurisdictions, the definition of collateral borrowers. Meanwhile, information collected by CRSPs generally implies fixed/immovable assets such as land includes borrowers’ credit histories and past payment and property, and ignores the more common moveable behaviors. Theoretically, information on one borrower assets of SMEs. Because moveable assets such as (credit history, past payment behavior, mortgage, vehicles, equipment, and inventory are not considered immovable property, and assets encumbered by security formal collateral, lenders are not willing to grant credit interests) could be collected by, and be available from, against them. In emerging economies, 78 percent one location. Accordingly, there are potential synergies of the capital stock of business enterprises is typically between CRSPs and collateral registries. More mature movable assets such as machinery, equipment, or credit reporting service providers with developed databases receivables, and only 22 percent is immovable property.75 and sophisticated technology platforms have the capacity Because most SME borrowers have more immovable to incorporate information from collateral registries. These 74 Global Partnership for Financial Inclusion 2011. 75 Safavian et al., 2006. 35 Credit Reporting Knowledge Guide providers may also have the potential to develop their own appropriate legislative regime, create the collateral registry, collateral database and perform the function normally and develop the appropriate business model to support the performed by collateral registries. operation. CRSPs can provide access to data in a collateral registry A version of the public-private partnership model can either by establishing and hosting a collateral registry as be found in some countries in Latin America, such as part of their value-added services, or by joining an existing Colombia and El Salvador, where the government has collateral registry database and sharing the technology delegated public functions, such as the establishment and resources. In developing markets, where technical management of the collateral registry to private-sector infrastructure and local capacity are inadequate to support institutions (e.g., the chambers of commerce). An example the development of a separate credit reporting service of the third option can be found in China, where both the provider and a collateral registry, joint solutions are likely credit registry and the collateral registry are managed under to gain acceptance. the Credit Reference Center, which is a public service unit under the People’s Bank of China. Three models can be considered in setting up a joint credit reporting service and collateral registry: Benefits of a joint infrastructure are that it enables a more • Create a CRSP and collateral registry within the same efficient utilization of scarce technical and human resources, private-sector institution and allows the sharing of common disaster-recovery facilities and business continuity plans. However, differences • Create a public-private sector partnership between the two types of services need to be taken into • Establish the function of both CRSP and collateral account when setting up a joint infrastructure. Whereas the registry under one government agency, such as the data contained in CRSPs are private and individual, data central bank. held in collateral registries are publicly available. Thus any type of joint infrastructure should involve an appropriate A version of the first model is being undertaken in Sri governance arrangement that ensures the two databases are Lanka, where the credit bureau, Credit Information Bureau kept separate while being hosted in the same infrastructure. (CRIB), has been mandated by law to create and operate There should be transparent service-level agreements the movable property registry. IFC is providing technical between the government entity and the CRSP hosting the assistance to CRIB and the government to help develop the collateral registry. 36 CHAPTER 4 Legal and Regulatory Framework T he overall legal and regulatory framework for the United States.77 Other countries have adopted credit credit reporting should be clear, predictable, reporting regulations, usually issued by the ministries of nondiscriminatory, proportionate, and supportive finance or central banks based on powers bestowed on of consumer rights. The legal and regulatory framework them through banking legislation.78 The European Union should include effective judicial or extrajudicial dispute and several countries regulate credit reporting activities resolution mechanisms.76 Ideally, the legal and regulatory under broad data protection laws that cover not only framework should enable and promote the development of credit reporting activities but also other relationships and secure, efficient, and reliable credit reporting systems, while transactions involving data management and exchange.79 fostering competition in the credit market and protecting the rights of consumers with respect to their personal Because credit registries generally cover the regulated information. As recognition grows that credit reporting lending sectors (banking), they derive their mandate to systems are vital to strengthening financial infrastructure operate through a country’s banking laws. Credit registries and ultimately access to finance, more and more countries are typically overseen by central banks that are entrusted this are increasing efforts to create an optimal legal and role through banking laws. Credit bureaus, conversely, are regulatory environment for these activities. usually covered under specific credit reporting laws and/or laws relating to data and consumer protection and are often The legal framework for credit reporting differs from regulated by central banks or other financial supervisory country to country, and may include a combination authorities. Because credit reporting markets are expanding of credit reporting laws, banking laws, data protection to include several different types of data providers (such as laws, consumer protection laws, fair credit granting and microfinance institutions, telecommunications providers, consumer credit regulations, and personal and corporate and utility providers), the scope of applicable legislation is privacy and secrecy provisions. In some countries, specific also expanding, since these nontraditional data providers credit reporting laws have been enacted. Most of these are covered by separate legislation. Although the central laws have been developed over the past decade and were bank usually regulates microfinance institutions, in several modeled after the Fair Credit Reporting Act (1971) in countries, the microfinance sector has a different regulator 76 World Bank 2011a, General Principle IV. 77 Specific credit reporting laws can be found worldwide, for example; Ley de Buros de Credito in Ecuador; Ley de Sociedades de Informacion Crediticia in Mexico; Law on Credit History Bureaus in Moldova; Credit Bureau Act in Sweden; Credit Information Bureau of Sri Lanka, Act No. 18 of 1990; Credit Information Companies Regulation in India; and Credit Reporting Bill in Guyana. 78 Some examples include the recently passed Decree on Credit Information Activities in Vietnam, regulations on a credit risk center in Spain (Circular 3/1995 of Bank of Spain), regulations on a credit risk center in Italy (Circolare N 139, 1991 de la Centrale dei Rischi, Bank of Italy), regulations on credit reporting and scoring companies issued by Central Bank of Egypt, and regulations CN/27/G/2007and CN/28/G/2007 on credit information, issued by Morocco’s Bank Al-Maghrib. 79 Examples can also be observed in emerging markets particularly in Latin America and Eastern and Central Europe (e.g., Argentina, Chile, Colombia, Uruguay, Bulgaria, Moldova and Romania). 37 Credit Reporting Knowledge Guide or is not regulated. Telecommunications and utility Regulations enable authorities to implement the specific providers are regulated by a different set of nonfinancial provisions of legislation. Regulations are easier to change regulators. Because a variety of authorities are involved in than legislation and tend to be more prescriptive. Central regulating credit reporting systems and the participants in banks, regulators, overseers, and other authorities should, these systems, a key challenge in creating an enabling legal in the initial phases of developing a credit reporting and regulatory environment for credit reporting systems system, consult the General Principles for Credit Reporting is ensuring alignment in the objectives of these regulatory as the framework from which to draft specific operational bodies and increased dialogue and collaboration across the regulations.80 The involvement of all stakeholders in different regulators and overseers. creating regulations promotes transparency and facilitates better compliance with the eventual standards. Lawmakers Whichever approach is followed, the legal framework and policymakers are assisted in implementing legislation should support the key concepts in credit reporting, reflect through detailed regulations that provide more specific the full scope of credit reporting functions and operations, guidance on how each aspect of legislation should be carried and accommodate evolving trends. In practice, the legal out. Regulations are enforced through various government framework surrounding credit reporting generally should: agencies. Regulations on credit reporting systems generally • Establish the rules for a fair, competitive, and efficient cover the following aspects: market in the provision of credit reporting services • Establishing licensing or registration processes to ensure • Establish the rights and obligations of the different that service providers have the financial, business, and participants in the credit reporting system, namely the technological capacity to provide an efficient credit CRSP(s), data providers, and users. reporting service • Provide clear guidelines on the kinds of data that can be • Ensuring that service providers adhere to minimum levels collected and the permissible purposes for which it may of maintaining data accuracy (minimum information be shared inputs should be clearly defined, and other permissible methods of validating information should be prescribed) • Provide guidance on data security obligations, data retention periods, and other compliance matters • Specifying permissible data sources • Establish consumer rights and provide a framework for • Ensuring service providers adhere to minimum levels of consumer concerns with credit reporting data maintaining data security • Establish rules for compliance and actions in the event of • Ensuring service providers adhere to consumer privacy noncompliance. safeguards (instances when consent is required, disclosure, and “permissible purposes” should be clearly Because legislation can be difficult to change once put in defined in the rules) place, the legal framework for credit reporting should be • Prescribing a process for consumer rights’ safeguards (the broad and flexible enough to accommodate evolving trends redress mechanisms and process to be followed in the in the credit reporting market. For instance, legislation event of a complaint must be clearly set out in the rules) may require CRSPs to “take reasonable steps” to verify the accuracy of consumer information reported to it, or to “have • Prescribing permissible purposes for data collection and in place policies and procedures” that deal with data privacy use and security. Such provisions are not prescriptive and allow • Establishing power of the authority to handle escalated for interpretation during the process of implementation. or unresolved consumer complaints 80 See World Bank 2011a for more details. 38 Credit Reporting Knowledge Guide • Establishing power of the authority to conduct 4.1 Data Collection, Retention, compliance inspections Disclosure, and Security • Establishing power of authority to take appropriate action in the event of noncompliance (including Defining data scope and data sources: Section 1.3 reviewing and conducting hearings and issuing has already discussed negative and positive data, and penalties and fines) comprehensive or full-file credit reporting. Generally, the • Establishing power of the authority to conduct audit scope of data that can be collected and distributed by a checks credit reporting system is defined by the legal framework. In some countries, the scope is wide, whereas in others, • Providing consumer education and outreach. the legal framework is set up to permit reporting of only negative data and prohibit the collection and sharing of In line with the General Principles for Credit Reporting,81 positive credit data. the overall legal and regulatory framework should be clear and predictable, ensuring that the various participants in A database with negative-only data, while excluding highly the system (CRSPs, data providers, and users) are aware exposed borrowers that have defaulted in the past, continues of the consequences of their actions. Rules should be to exclude them from access to finance for long periods nondiscriminatory and apply equally to participants in the following their defaults regardless of their current financial system with few exceptions. Laws and regulations should performance and other favorable information. CRSPs that apply proportionally to the various participants to ensure collect a wide range of information are able to generate some participants are not unfairly penalized over others. more comprehensive credit reports. They are more reliable While considering protection of data subject and consumer and more efficient than CRSPs that operate on a limited rights, laws and regulations should strive to balance data scope of data. Ideally the legal framework should allow for protection needs against the practicalities of achieving such inclusion of positive and negative data in credit reporting to protection levels. facilitate better credit granting decisions. Sections 4.1 and 4.2 discuss key issues for credit reporting Essentially, all data that is relevant for an analysis of legal and regulatory frameworks surrounding data collection, creditworthiness, including data in public records, should retention, disclosure and security, and data subject rights. be collected, while the collection of irrelevant data may be Section 4.3 discusses licensing and registration of CRSPs as prohibited.82 Data is considered relevant in relation to the a mechanism used by regulators to control who can provide purpose for which it is collected. So for instance, in some credit reporting services in a market. In some countries, an countries, CRSPs are prohibited from collecting information entity is entrusted with oversight of the different parties about a consumer’s race, medical status or history, religion, in a credit reporting system to ensure compliance with or other information that is deemed irrelevant for purposes the respective legal and regulatory framework. Section 4.4 of analyzing creditworthiness and making credit decisions.83 discusses the oversight function including objectives and In other countries, notably the United States, a broader roles of the overseer. range of information—including, employment, judgments, tax liens, and other information in public records— may be collected by CRSPs and information from CRSPs can be used for purposes beyond the granting of credit, such as employment reference checks or for the collection of debt. 81 Ibid. 82 Ibid., General Principle I. 83 See, for example, South Africa. 39 Credit Reporting Knowledge Guide In addition to permitting positive and negative data, reporting service providers can collect. For instance, public the legal framework for credit reporting should permit records like identification databases, civil status records, comprehensive credit reporting which allows for the and court proceedings can enable better identification of collection of data from a wide variety of sources and sectors, a borrower and give a more holistic picture of his or her including retail, small business, microfinance, credit cards, credit history. There is no worldwide standard on access insurance, telecommunication companies, utilities, and to public information and the jurisprudence differs from others. Ideally, the legal framework would permit the region to region. Some countries have adopted laws on following data sources: access to information that classify data and establish levels of accessibility based on a need-to-know basis.84 Ideally, • Banks operating in the same country the legal framework should make provision for access to • Mortgage finance companies relevant public information by credit reporting service • Finance leasing companies providers. • Microfinance institutions Retention periods: Legislation typically stipulates a • Insurance companies specific length of time that information can be stored and disclosed. Although historical information enables • Institutions that offer credit to MSMEs lenders to assess a borrower’s credit quality over a period • Asset management companies of time, the legislation should specify a cut-off period for information disclosure, after which time information is no • Suppliers of goods and providers of services on a post- longer distributed to users to give borrowers a fresh start. paid or installment payment basis (telecommunications Doing Business survey data show that payment history and utility providers, retailers, and health providers) information is usually maintained for a minimum of five • Other credit reporting services (CRSPs and collateral years. Public records relating to bankruptcy are usually registries) retained for seven or more years. In some countries, such • Identification databases and other private or public as in Brazil, information is never deleted, although it may records not be distributed beyond a certain number of years.85 In some countries with negative-only reporting systems, • Other sources of relevant information provided the once a bad debt is paid off, all negative data related to it express consent of the data subject is obtained and is deleted from databases, either because it is mandated by confidentiality of the information is maintained. law or simply because it is common practice in the market place. Such practice is detrimental to the ability of creditors The last provision is particularly important as it allows CRSPs to make informed credit granting decisions. Rather than to obtain other relevant information from nontraditional erasing information on defaults once loans have been data sources such as organs of the state and courts, entities repaid, this information should be stored with the rest of involved in fraud and corruption investigations, educational the borrower’s file for an assigned period of time. According institutions, and debt collectors. to Doing Business survey data, out of 84 credit bureaus that provided information, only 15 preserved historical Access to public information is relevant for credit reporting information for less than five years while 69 preserved data purposes because information available through public for five or more years. For credit registries, out of 88 that records can enhance the quality of the data that credit provided information, 75 preserved information five to ten years or longer.86 84 For example, in Guatemala, Nicaragua, and Ecuador in Latin America, and is embedded in the European Directive 2003/98/EC on the re-use of public sector information in the European context. 85 Doing Business Indicators (database), 2012, “Getting Credit” indicator. 86 Ibid. 40 Credit Reporting Knowledge Guide Data disclosure and permissible purposes: The ability to • An investigation into fraud, corruption, or theft collect and analyze a wide scope of data from a wide range of • Considering a candidate for employment (in some sources does not necessarily permit CRSPs to use or disclose countries, this is permitted with the express consent of the information put together using such data. To safeguard the subject) consumer privacy, some legal frameworks set up a finite list of “permissible purposes” for which collected data may be • Tenancy contracts (in some countries, the lessor is used. Permissible purposes vary from country to country, but permitted to conduct a credit check of the lessee in most cases include “assessing an application for credit.” applicant). The list of permissible purposes can require separate express consent, for example, when considering a candidate for Data security: In addition to defining the scope and employment.87 Conversely, some countries expressly prohibit sources of data, and purposes for which data may be credit reference checks for purposes of employment.88 collected and used, the legal and regulatory framework may impose standards to ensure accuracy, confidentiality, and Generally, the more value-added services the CRSP wishes security of information in databases used to generate credit to provide, the more extensive the permissible purposes reports. Since consumer protection is the motivation for need to be, and the more the issue of consent for disclosure such requirements, responsibility for accuracy and security will come into play. Accordingly, the regulation listing is taken out of the prerogative of credit reporting service permissible purposes should, in addition to listing specific providers and data providers and made a legal obligation. purposes, make provision for other purposes provided the Some common threats to data security include hacking, consent of the consumer is obtained prior to the credit improper use by CRSPs or their employees, and tampering. report being issued. Ideally, the legal and regulatory framework defining permissible purposes would include As such the laws and regulations governing the operations the following purposes: of CRSPs require that credit reporting service providers take active steps to ensure the protection of data against • Assessing an application for credit, insurance, or a loss, corruption, misuse, or theft. This legal requirement is mortgage usually drafted as a general obligation requiring the operator • Reviewing existing credit facilities to take reasonable steps and establish processes to cope with the logical, physical, and organizational aspects of data • Developing a credit scoring system security. For example, see the regulations in South Africa • Acceptance of guarantees and the OECD Guidelines on the Protection of Privacy and • Application for services (for example, when a person Transborder Flows of Personal Data.89 The level and detail applies for a mobile phone service contract in the of security arrangements necessary for each credit reporting United States, the telecommunications company may service is not usually specified by the regulator. For more conduct a credit check of the applicant) details on the specific measures that CRSPs can undertake to ensure data security, see Section 5.6. • Verifying personal credentials • Payment history in respect of continuing credit services with retailers 87 For example, disclosing information for employment purposes requires a separate express consent from customers in South Africa and the United States. 88 For example, Chile. 89 For South Africa, see the National Credit Regulations, GG 28864, May 2006 (s.18). Principle 11 of the OECD Guidelines provides that, personal data should be protected by reasonable security safeguards against such risks as loss or unauthorized access, destruction, use, modification or disclo- sure of data. http://www.oecd.org/document/ 41 Credit Reporting Knowledge Guide 4.2 Consumer Rights purposes. For example, banks may share information with the banking industry supervisor or with other financial Consumer rights within the context of credit reporting institutions as long as they are regulated by the same systems refer to privacy of the data subject’s information and supervisory authority. of the accuracy of products and services developed using this data. There is no definitive approach to the protection In many countries, including Thailand, Kazakhstan, Mexico, of data subject rights within credit reporting systems. For Peru, and Panama, as well as in the European Union, instance, in the United States, no specific legislation protects laws require explicit borrower consent for a data provider data subject privacy rights, but the Fair Credit Reporting to provide information to the CRSP. If data providers do Act defines specific permissible purposes for which data not have consent to share their customer information with can be used and disclosed, which offers some protection CRSPs, the CRSPs may be required to secure consent of the data subject’s privacy. In the European Union, directly from data subjects. In the absence of specific legal directives establish a broad range of consumer protections and regulatory frameworks, or in case of nonregulated that go beyond the credit reporting systems. The objective, lenders, an agreement between lenders and the CRSP to regardless of the approach taken, is establishing consumer collect consent and share information is advisable, but is confidence and trust in the credit reporting systems. rarely followed in practice. Consent is not applicable for all types of CRSPs. For instance, a registry collecting data from Consumer protection and privacy considerations are closely regulated entities under a banking law mandate would not linked to the purposes of data collection and disclosure. Legal require consent to collect this data or provide it back to the and regulatory frameworks can use consumer consent90 regulated entities. and permissible purposes as mechanisms for protecting the rights of consumers with respect to their data. Data In the interest of maintaining operational efficiency, collection, disclosure, and permissible purposes have been the legal framework should place the onus of obtaining discussed in Section 4.1. and maintaining a record of borrower consent for data submission on data providers and sources. In the event of Depending on the jurisdiction, the regulator may require a dispute, the data provider must be able to demonstrate a data provider to collect explicit or implicit individual that it had obtained borrower consent in accordance with borrower consent to provide data to a CRSP and to access a the law. For example, a typical bank consent appears in credit report prepared by a CRSP. The objective of consent its privacy policy, a copy of which is usually signed by the is to enable the data subject to control the flow and use customer at account opening, or when he or she applies of his or her personal information. Typically, banking for credit. Privacy policies outline how the bank or creditor secrecy laws restrict disclosure of customer accounts and manages its customers’ personal information and it describes transactions information without the customer’s consent. generally the sorts of personal information held and for Such provisions are often cited as an impediment to the what purposes. Customers should know up-front for what development of a comprehensive credit reporting system. purposes their information is collected, and to what uses However, in the banking industry, obtaining consent to such information may be put. In countries with developed collect personal information usually makes provision for credit reporting systems, the consent given to banks by their sharing such information with third parties for specific customers usually makes provision for consent to share the customers’ data with credit reporting service providers.91 90 Consent is defined as “a data subject’s freely informed and specific agreement, written or verbal, to the collection, processing and disclosure of personal data. World Bank 2011a, see Glossary. 91 Usually a privacy policy informs the customer of the bank’s intention to collect personal information, and also informs the customer about the purposes for and circumstances under which the information can be disclosed to third parties. The privacy form, the signing of which usually amounts to consent to share information with credit bureaus, may typically state, “We may collect and share your information with third parties to offer you other products and services for marketing purposes or to assess credit applications.” 42 Credit Reporting Knowledge Guide Data accuracy and redress mechanism: Data accuracy of an error and failure to take corrective measures should is critical to the subject of consumer rights, because liability for noncompliance arise. inaccuracies in data can lead to negative consequences for a consumer. Errors in credit decisions may result from Consumers also have a role in ensuring their information incorrect or inadequate information supplied to the CRSP, is correctly reported. The legal framework usually grants problems with assignment of information to the wrong consumers the right to access their own credit reports, and consumer file (for instance where there are similarities of the ability to challenge incorrect or incomplete information names and addresses), or if the CRSP sends the wrong file in their files. Modern credit reporting systems provide to the requesting creditor. consumers with the right to access their credit reports free of charge on a periodic basis (e.g., once per year) or in specific To protect consumers, laws governing credit reporting circumstances (e.g., if the consumer is the victim of fraud).93 may require that specific minimum information inputs be In a groundbreaking move, Callcredit, a credit bureau in the captured in each consumer file. This requirement must be United Kingdom, recently introduced its Noddle service complied with by both the CRSP and the data providers that provides consumers with free credit reports for life. and sources. For instance, it may be a legal requirement that As discussed in Section 1.4, such rights are only effective if the information submitted to a CRSP contain a consumer’s consumers are aware of them and kept informed of changes identifying information such as his or her full name(s), that affect these rights. date of birth (where available), identification number or passport number (where available), address and contact When a data subject challenges the information on his or information (where available), and details regarding current her record with the CRSP, the legal framework generally employment status (where available). The rule should requires the CRSP or the data provider to investigate the allow the CRSP to use other methods of identification and claim, identify the source of error if the claim is valid, and matching when traditional methods are not available. For take corrective steps to rectify the error. Responsibility for an example of nontraditional identification methods such correcting the error lies with the source of the error. If a as biometric identification, see Box 5.1. borrower disagrees with the final decision with respect to his claim of data error or omission, the borrower should Imposing strict standards for data accuracy by imposing be entitled to obtain resolution through a judicial (court excessive penalties in the event of erroneous reports based system) or extra-judicial process. Depending on the on incorrect information could impede the free flow of jurisdiction, this process might be conducted through the information and affect the efficiency of the reporting data protection agency as in most European Union member system.92 Ideally, regulation should place responsibility countries, a consumer protection body, a unit within the without imposing strict liability. Legal provision should central bank, or other oversight body. require that CRSPs, data providers, and other data sources take all reasonable steps to ensure that the information In addition to providing data subjects with the right to collected and reported is accurate, up-to-date, relevant, and access, challenge, and correct information in their files, valid. Imposing responsibility without strict liability also the legal framework may require transparency of credit means that when the CRSP identifies incorrect information, decisions. Transparency means that data subjects should it should notify the data provider, who is responsible for be notified of adverse credit decisions that have been taken correcting the information. Only in the event of knowledge against them based on a credit report. Accordingly, the rule 92 For example, as has happened in Thailand when the restrictive Credit Information Business Act, B.E. 2545 (2002) law was passed in 2002. 93 For example, the U.S. Fair Credit Reporting Act requires credit reporting companies to provide a consumer with a free copy of his or her credit report at the consumer’s request once every 12 months. In addition, when a consumer notifies a credit reporting services provider about an error in the file, the service provider must send the dispute back to the creditor/data provider. The creditor/data provider must investigate the dispute and report back to the service provider, which must then correct its records and notify the consumer of the outcome of the dispute. 43 Credit Reporting Knowledge Guide usually provides that any person who uses a credit report to can collect data on consumers and MSMEs for the purposes deny an application for credit, insurance, employment, or of generating credit reports. Licensing is also a method of to take other adverse action against a data subject, should governing the operations of CRSPs by stipulating observance notify the data subject of their decision and inform him of minimum business standards. The licensing process is or her of where the report was obtained. This knowledge usually an evaluation of the proposed operator’s business, is an incentive to data subjects to protect their credit financial, and technological capacity to provide a secure and reputation or improve their credit profile, especially if an efficient credit reporting service, and the operator’s ability adverse decision has been taken based on such information. to observe obligations respecting privacy laws and consumer The legal framework may also provide for consumers to rights. (See Box 4.1.) Where licensing is a requirement, the claim compensation or damages in case of adverse events legal framework must provide clear and precise guidance stemming from the use of erroneous data. on the qualities and abilities an operator must demonstrate. The legal framework also makes provision for the unlikely event that a service provider goes out of business, exits 4.3 Licensing or Registration the market, or has its license revoked. In such instances, of CRSPs provisions are made for the transfer of data to the regulator until an alternative provider is identified. Some jurisdictions have adopted a scheme of entry and exit requirements for CRSPs, which serves to mitigate Many countries require credit reporting service providers risks associated with consumer rights, competition within to register with the regulator. If the process of registration the credit reporting market, and business sustainability. A is mandatory and entails filing information about the licensing process can be used to place restrictions on who CRSP’s business, financial, and technological capacity, it is Box 4.1 Licensing Credit Bureaus in Kenya The Central Bank of Kenya recently licensed two credit bureaus under its 2008 Banking (Credit Reference Bureau) Regulations, which began operation in 2009.94 Under these regulations, a person may not establish or operate a credit bureau unless he or she is incorporated as a limited liability company under Kenya’s company laws, and is licensed to operate a credit bureau by the central bank. The regulations detail the requirements and process that must be met by a proposed operator. An application must be filed accompanied by specified supporting documents showing the nature of the planned business and its organizational structure, internal control systems, and monitoring procedures. The supporting documents must include a market analysis; ownership, management and governance structure; operation manuals pertaining to databases (methods of uploading, processing, and updating data); proposed security and control measures; and the proposed fee structure.95 In addition to provisions for granting a credit bureau permission to operate, the regulations list specific activities in which the bureau is allowed to engage: to obtain and receive customer information; store, evaluate, and update customer information; compile and generate reports from customer information; assess the creditworthiness of customers; and sell reports to institutions. The new system also requires banks to disclose and share the credit details of their borrowers with other money-lending institutions. 94 IFC advised the Kenyan government and other stakeholders over a two-year period on the complicated process of developing the new credit reference regulations. The Banking (Credit Reference Bureau) Regulations 2008 govern the licensing, operation and supervision of credit bureaus by the Central Bank of Kenya. The two new credit bureaus went live in 2010. 95 For more details, please see these regulations at http://www.cgap.org/p/site/c/template.rc/1.9.44955/ 44 Credit Reporting Knowledge Guide similar to a licensing process.96 Even if there is no licensing As with legislation, oversight and regulation should provide or registration requirement, the operations of a CRSP are for appropriate enforcement measures that encourage usually subject to some oversight, especially with regard to compliance by all parties, but are not so stringent as to data collection, security of data, data privacy, and consumer discourage the operation of credit reporting services. For rights. These provisions may be contained in a country’s instance, the regulatory framework could make provision banking laws, company laws, or other laws touching on for issuing notices of noncompliance in the event of alleged consumer protection. or real noncompliance with safeguard obligations. Under this process, service providers are given the opportunity to remedy violations without adverse action by the authority. 4.4 Oversight and Enforcement Penalties and damages should be imposed in the event of willful or negligent noncompliance with regulations (for The primary objective of overseers of credit reporting instance, inaction despite notices) and with respect to systems is to ensure the safety and efficiency of these noncompliance with consumer rights provisions. systems.97 Authorities engaged in oversight typically include central banks, financial supervisory bodies, data Provisions in credit reporting regulations that deal with protection authorities, ministries of finance and commerce, specific matters as opposed to processes, are not usually or consumer protection authorities. Oversight is exercised enforced through the “notice” system. For instance, if a over CRSPs, traditional data providers, as well as users of report is disclosed for a nonpermitted purpose, a violation credit reporting products and services. has per se occurred and the notice process would be useless. Accordingly, the oversight role should combine Given that oversight over different aspects of credit enforcement provisions that follow a compliance notice reporting systems can be entrusted to different overseers, process with enforcement provisions for outright violations. the oversight function requires collaboration among the Finally, while the industry may be regulated by an authority different overseers. In case of cross-border flows of data, with powers to review complaints, issue specific compliance this collaboration should extend to overseers located in measures, and impose penalties, recourse to the traditional different markets. Overseers should clearly communicate court system should not be excluded. their objectives to the market to promote transparency and accountability of the different oversight and regulatory bodies. 4.5 Governance and Risk Management The overseer of the credit reporting system must have the necessary human and financial resources to actually The governance arrangements of credit reporting service undertake oversight and enforcement activities – a key providers and data providers should ensure accountability, challenge in many markets, where oversight and regulatory transparency, and effectiveness in managing the risks functions are defined, but the institutions are not provided associated with the business and fair access to the with the capacity to perform their functions. The role of information by users.98 CRSPs are usually created as entities oversight is evolving in most emerging markets. with separate legal status, thus are subject to corporate laws 96 For example, the National Credit Regulator in South Africa is tasked with the registration of credit providers, credit bureaus, and debt counselors. Registration of credit bureaus entails the filing of supporting documents about the operator’s business information and structure including human resources, financial statements, operational resources (procedures to safeguard databases), and procedures for handling consumer complaints. 97 World Bank 2011a. 98 World Bank 2011a, General Principle III. 45 Credit Reporting Knowledge Guide and business practices in their countries. In most modern including data providers and users. The legal framework economies, corporate governance mechanisms and controls for credit reporting may cover the requirements for these for corporations are mandatory.99 Governance arrangements participants, but more frequently, they are covered by the capture the relationships among the CRSP’s management, legal frameworks governing their activities. its shareholders, its clients, and external stakeholders. Governance and risk management measures are important in the context of credit reporting service providers because 4.6 Cross-Border Data Flows these entities are entrusted with sensitive data pertaining to consumers. The success and continuity of a CRSP’s As consumers and businesses increasingly migrate from operations are considered to be of broad public interest. one jurisdiction to another, financial markets are becoming regionalized and globalized, generating increasing demand The legal framework may include broad provisions to ensure for credit reporting on data subjects outside of their home adequate governance arrangements for credit reporting markets. Cross-border data flow is a useful mechanism service providers. Some such provisions include: through which a data subject’s credit can be monitored from multiple markets. With cross-border data flow • Laying out minimum criteria for qualifying shareholders, models, a borrower applying for credit in a country where directors, and other CRSP management officials, who he or she has no credit history, but who has a credit history are collectively responsible for the overall operation of in his or her country of origin, can be assessed easily since the CRSP the information is available to potential creditors in both • Holding management and board members accountable countries. for compliance with the legal framework • Requiring the CRSP to appoint independent external Although in principle this credit reporting model would auditors and undertake regular audit and compliance work well in the regional context where several countries are reviews in close proximity, and whose citizens have free movement from country to country, there are several potential • Setting rules related to the fair and equal access to challenges in the exchange of such data. All the challenges information by the users.100 of providing credit reporting services in a domestic market apply to cross border data flows, with some challenges Regulations may specify the reporting requirements for being more prominent, including: the existence of multiple CRSPs by which regulators and oversight authorities national legal and regulatory frameworks that may not be can ensure compliance with the legal framework. Such aligned to facilitate such data flows; issues in matching requirements may entail disclosing key financial results, data subjects correctly; issues concerning standardization materials changes or proposed changes in ownership of data formats, inputs and data quality across markets; structure, and other key information that could affect the identification and mitigation of risks arising from cross- governance arrangements of the CRSP. border data flows; and a heightened need for protection of data subject rights with respect to privacy, to name a few. In addition to these legal controls, CRSPs should have Moreover, putting in place infrastructure to facilitate such internal controls and policies to ensure that the risks to credit data flows can be expensive. which they are exposed are adequately managed or mitigated. Similar requirements on governance and risk management Given the relevance of cross-border data flows for several apply to other participants in the credit reporting system, markets worldwide, the General Principles for Credit 99 OECD 2004 and, in the United States, the Sarbanes Oxley Act of 2002. 100 OECD 2004; World Bank 2011a, General Principle III describes the ideal governance arrangements for credit reporting services and data providers: The governance arrangements of CRSPs and data providers should ensure accountability, transparency, and effectiveness in managing the risks associated with the business and fair access to the information by users. 46 Credit Reporting Knowledge Guide Reporting devotes a principle to it, which states that Several models of cross-border data flows exist. For instance “Cross-border credit data transfers should be facilitated, in the European Union, several credit registries have where appropriate, provided that adequate requirements signed a memorandum of understanding to facilitate the are in place.”101 In order for cross border data flows to be exchange of credit data among the registries, for supervisory facilitated, certain preconditions should be fulfilled, such purposes.102 Cross-border data flows can exist between as a demonstrated need for such data flows based on the different bureaus in different markets, as well as through a existence of strong financial and economic integration of bureau serving several markets through one location. This the relevant markets, national-level policies for financial latter arrangement, called the “Hub & Spokes” model, integration, small size of markets, and the economic is discussed in more detail in Chapter 5. Discussions are viability of setting up systems that enable such cross-border under way to consider cross-border data flows in several data flows. The general principles highlight the importance regional blocs such the East African financial community, of a cooperative framework between the multiple regulators Central African countries, and the Union Economique et and overseers in markets with credit data flows and careful Monétaire Ouest Africaine (UEMOA) region. assessment and mitigation of all risks arising from such data flows, in addition to providing guidance on the challenges noted above. 101 World Bank 2011a. General Principle V. 102 World Bank 2011a. 47 CHAPTER 5 Developing Credit Reporting Systems in Emerging Markets D 5.1 Assessing Market Conditions eveloping a credit bureau or credit registry is a time- and resource-intensive project involving the commitment of many stakeholders such A market assessment can help determine whether a CRSP as government, supervisory authorities, regulators, credit is financially sustainable in a particular market and, if so, reporting service providers, data providers, users, and in what form. Different stakeholders can play a role in consumers. This chapter, drawing on the General Principles assessing the market conditions. Development institutions for Credit Reporting introduced in Chapter 1 together with like IFC can work with government authorities or creditor IFC experience and expertise in the process of setting up associations to undertake an assessment. The components credit reporting systems in client countries, outlines key of this in-depth analysis may include the following aspects, practical aspects of that process, in particular: which are discussed below: • Assessing market conditions • Market analysis • Changing perceptions and building awareness • Stakeholder analysis • Ensuring adequate data availability • Technical scoping study • Ensuring financial sustainability • Legal and regulatory environment assessment • Creating an appropriate business model • Specifying staffing requirements and identifying • Identifying appropriate technology needs available skills in the labor force. • Identifying operational and practical considerations 5.1.1 Market Analysis • Establishing an appropriate legal and regulatory framework.103 A market analysis projects demand and costs to enable the credit reporting service provider to price its products and These activities can be carried out simultaneously or in services. Pricing is one of the key factors in sustainability, sequence depending on the availability, capacity, and needs and crucial investment decisions such as software of the stakeholders involved. The following sections provide acquisitions and disaster recovery plans should be aligned additional guidance on the objective of each activity, who with the pricing strategy to avoid potential losses. A typical should be engaged, and how it can be carried out. market analysis focuses on the following: • Population size, which indicates potential customer base for lenders • Size of existing retail and SME credit market and potential for growth 103 This aspect has been discussed in Chapter 4. 49 Credit Reporting Knowledge Guide • Level of sophistication of the credit market in terms of to all potential participants (lenders) and following up with products and services meetings to discuss the survey results. The focus includes issues such as: • Size of the existing CRSP(s) in terms of borrowers covered • Types of consumer and MSME credit products offered • Capacity and scope of data in the CRSP’s database • Level and growth rates of retail and MSME credit, by product • Potential demand for credit information • Current and expected number of credits issued to • Existing and potential data sources and public inform projections about the potential volume of information sources inquiries • Extent to which the demand for credit information is • Availability of electronically stored historical satisfied by existing providers information • Risk of competition from other CRSPs • Borrower consent to disclose information to a CRSP • Credit market trends • Availability of unique ID numbers for individuals and • Legislative or regulatory limitations. MSMEs, or other identification methods • Level of sophistication of lenders’ internal information 5.1.2 Stakeholder Analysis management systems The stakeholder analysis assesses the potential stakeholders • Technology and infrastructure constraints of lenders (e.g., lenders, nontraditional data providers, authorities, and potential necessary upgrades policy makers) of the credit reporting system and their • Level of awareness among lenders on issues related to commitment to the project by asking the following credit reporting questions: • Level of technical and communication infrastructure in • Is there a broad consensus among lenders on the the country, whether it will be able to support the needs usefulness of credit information sharing? of the proposed CRSP, and potential necessary upgrades • Who are the potential members or users of the proposed that would require significant investment. CRSP(s)? Comprehensive analysis of the technical capacity is needed • Are lenders willing to share positive and negative data? to determine whether a technical partner is needed, develop • Do lenders have the technological capacity to share the the technical specifications for the proposed credit reporting data? service, and help lenders make any changes needed in their • Are the regulatory authorities supportive? technology platforms to enable them to join the credit reporting service. Technology needs and the qualities of a • What is the potential business model for the CRSP? strong technical partner are further discussed in Section 5.6. 5.1.3 Technical Scoping Study 5.1.4 Legal and Regulatory Environment The objective of a technical scoping study is to assess the Assessment104 technical capacity and readiness of the lenders to participate This component entails consultations with regulators and in the credit reporting system. It involves sending detailed qualified legal experts to assess the country’s legal landscape. questionnaires on the nature and formats of available data 104 See also Chapter 4. 50 Credit Reporting Knowledge Guide The main issues to be addressed with regulatory agencies in the market and to estimate what skills training will be are: needed. Section 5.7.1 discusses the organizational structure and staffing requirements of a newly established CRSP. • Is information sharing permitted or limited? • What is the existing legislation relevant to information sharing and the proposed credit reporting service? 5.2 Changing Perceptions and • Who are the oversight and enforcement authorities Building Awareness relevant to information sharing and credit reporting services? A critical step in developing a credit reporting market, is to change perceptions and build awareness on credit reporting • Is an operating license or registration required to establish within the sector and community. Bank secrecy and stiff a CRSP? competition typically characterize the lending environment. • What are the implications of the legal framework for the Lenders are generally resistant to sharing positive data on service provider’s operations? their clients for fear that competitors will steal their good customers. For political reasons, authorities unfamiliar or • If the regulatory environment is limiting or not enabling, uncomfortable with sharing financial information may what regulatory reforms need to take place to achieve also be resistant to this concept. In markets where cash an environment conducive to information sharing and is still predominantly used for daily transactions and the credit reporting? credit culture is weak, the public is unlikely to understand • What new rules or regulations are being proposed? the importance of providing their data to credit reporting service providers. In markets where credit is more prevalent, • How organized are consumer groups, and How likely are borrowers may be hesitant to share their personal data out they to oppose information-sharing plans? of privacy concerns. The proposed service provider should ascertain (as part of Consequently, the initial phase of building a CRSP should its market assessment) that it is allowed to legally operate focus on building awareness among lenders and their before finalizing any aspects of its operations. If the market clients, the public, government officials, policy makers, assessment reveals that the legal and regulatory environment regulators, and other potential participants on the benefits is not enabling, further efforts to engage legislators and of the credit reporting system. The market analysis and oversight authorities should be made promptly, as the the stakeholder analysis discussed in Section 5.1 will give process of introducing amendments or creating new the key stakeholders driving the reform process an insight laws takes between one and five years. Depending on into the issues that need to be addressed through awareness the complexity of a country’s rule-making processes, raising efforts. Tools that can be used to change perceptions government authorities and regulators who are supportive and build awareness include the following: of the development of the credit reporting system may tackle the necessary regulatory changes simultaneously with Roundtables and conferences. Consensus and buy-in of the project’s design or set-up phase. stakeholders is achieved through building awareness of the benefits of information sharing. In 2012, IFC facilitated the 5.1.5 Specifying Staffing Requirements first regional conference on credit reporting for countries and Identifying Available Skills in the of the UEMOA. The event was instrumental in laying the Labor Force groundwork in developing a credit reporting system for the region. Similar consultations instrumental in promoting the A CRSP relies on information technology skills, which in establishment of credit bureaus were followed in Tajikistan, many countries may be in short supply. In this final part of Morocco, Kenya, Egypt, Vietnam, Russia, and several other the market conditions assessment, the aim is to match the countries. The APC in Panama regularly holds seminars to skills required for the operations with the skills available 51 Credit Reporting Knowledge Guide educate SMEs and consumers to understand their credit consumer credit and credit reporting. The site must also reports and how it impacts their ability to get credit. direct consumers on how to access their credit reports and explain the channels available to challenge and rectify A range of stakeholders can be involved in consultations, inaccuracies identified in their credit reports. Credit conferences, and roundtables, including: reporting service providers can take advantage of advances in social networking tools such as Twitter and Facebook. • Supervisory and regulatory bodies such as the central (See Chapter 1, Box 1.1 for a description of Panama’s bank and other financial supervisory authorities “Finances under Control” awareness program.) • Other government bodies, for example, ministries of finance or commerce Awareness-raising activities should deliver different, targeted • Policy makers and lawmakers messages to different stakeholders. Each stakeholder will, at some stage, require support from various government • Credit reporting service providers (existing and/or bodies, supervisory and regulatory bodies, policymakers potential) and lawmakers. Awareness raising targeted at government • Lenders, including banking and nonbanking financial officials, policy makers, and regulators should address the institutions, microfinance institutions, leasing following issues: companies, insurance providers, and other creditors • The importance of input from government officials, such as utilities and retailers policy makers, lawmakers, overseers, and regulators in • Other potential data providers (public data sources) creating a safe and efficient credit reporting system • Consumer representative organizations and the public. • The role of government and the need of government leadership in developing a legal and regulatory These events can be organized by the key stakeholder driving framework that is conducive to credit information the process of credit reporting development, typically a sharing central bank or a banking association, depending on the • The importance of information sharing for financial country context. Development partners like IFC are also stability and expansion of credit (different products, regularly involved in arranging and facilitating such events. more borrowers, different choice of providers) Media. Media coverage of conferences and roundtables, • The benefits of improved oversight of the financial sector as well as articles on the role of credit information with • The role of authorities in: expert opinions and reflections on the local debate can be useful in promoting credit bureau development. For – encouraging data providers to participate in and use example, conferences on the role of credit information the credit reporting service providers held in Azerbaijan, the Kyrgyz Republic, Tajikistan, and – overseeing activities of credit reporting service Russia were well covered in the local press. As a result, providers and ensuring compliance public awareness of the need to build a credit history and to submit one’s credit records to a credit bureau improved – enabling credit reporting service providers’ access to significantly. Initially, media coverage can be facilitated public records through the key stakeholder driving the process. Once a – ensuring consumer privacy rights are upheld. system is developed, credit reporting service providers may choose to provide press releases or attract media coverage to For an audience of financial and nonbank creditors and promote the concept of credit reporting. other data providers, awareness-raising efforts should focus on explaining what they can gain from being a part Internet. A CRSP’s website should be user friendly and of the credit reporting system. Efforts should be made to contain consumer-oriented information on aspects of educate these participants about their rights, roles, and 52 Credit Reporting Knowledge Guide responsibilities in the credit reporting system. Specifically, their roles and overall support for the development of awareness raising should: the credit reporting system. Credit reporting service providers or data providers can establish links with • Address concerns about sharing information and dispel consumers and explain how consumer data is handled fears of losing market share due to such information and treated to allay fears about data privacy and security. sharing Such awareness raising efforts should: • Highlight and explain different roles of a credit registry • Explain the role of a credit reporting service provider and/or credit bureaus and the benefits it offers • Explain the different measures that could be enforced • Discuss the types and nature of data that will be to prevent competitor institutions from poaching collected and the purposes for which this data will be customers shared or disclosed • Emphasize the need for cooperation among a country’s • Discuss the obligation of CRSPs to respect the privacy banking, financial, and nonbank financial institutions of personal information, and their duty to treat all such for the credit reporting service to succeed information as confidential • Assure lenders of the confidentiality of all information • Discuss conditions under which consumers can access provided and discuss the obligations of lenders to treat their own data confidential information appropriately • Discuss the redress mechanisms that will be available • Explain the importance of sharing full-file information to consumers to challenge and correct erroneous sharing and positive data information on CRSP databases • Encourage broad participation by bank and nonbank • Emphasize the importance of consumer consent to lenders in the credit reporting service enable data sharing • Encourage timely and accurate data submission and • Emphasize the role of the consumer in providing the emphasize the importance of compliance most accurate information. • Emphasize the benefit of improved risk evaluation throughout the account lifecycle In addition to educating the public about credit reporting, • Emphasize improved transparency in risk management campaigns should educate the public on using credit responsibly and reducing the risk of becoming overindebted. • Promote the introduction of updated credit control policies and procedures taking into account the information in the credit reporting service provider’s 5.3 Ensuring Adequate database Data Availability • Highlight the need to educate staff about credit reporting Data refers to all the information collected, processed, and used to generate reports and value-added services by a • Address how an adequate legal and regulatory CRSP. The different types of data providers to a CRSP were environment provides for an efficient and smooth credit defined and discussed in Chapter 1, section 1.2. The market reporting environment. analysis discussed in Section 5.1 gives the CRSP a sense of the challenges it will face in collecting data to populate At different stages in the process of credit reporting its database. To ensure adequate data availability, the CRSP development, the key stakeholder driving the process should pay attention to the characteristics of data and data may organize outreach to the public. Government collection described in the following sections. authorities, overseers, and regulators may want to explain 53 Credit Reporting Knowledge Guide 5.3.1 Data Quality in a format that allows the credit reporting service provider to easily extract the information and upload it onto its own Data quality is the most important element in successful system to further match and merge with other data. credit reporting. CRSPs must take steps to ensure that the data they use is accurate, complete, and up-to-date.105 A challenge to data accuracy and validation is the lack of To ensure a high quality of credit reporting, data should be: uniform identification schemes. Issuing national unique • Accurate identity numbers is usually within the prerogative of the • Sufficient, relevant, and collected on a systematic basis government. Adopting an identification system at the initial from all reliable, appropriate, and available sources phases of establishing a credit reporting system would be ideal, but is not always realistic. Therefore, in jurisdictions • Timely (updated on a continuous basis and available to without national identification numbers or where the users promptly) use of such identification numbers is prohibited by law, • Retained safely for a sufficient amount of time. CRSPs may have to develop their own system to identify data subjects using matching algorithms that traditionally The role of ensuring data quality and constantly working to combine name, address, and date of birth. In New Zealand improve it falls in various degrees on data providers, credit and Germany, for example, CRSPs use sophisticated reporting service providers, and data subjects. matching solutions because legislation prevents the recording of unique identifiers or specific unique IDs do According to the General Principles for Credit Reporting,106 not exist. In Australia, the one unique identifier available accurate data is free of error, truthful, complete, and up- (tax file number) can by law be used only for tax purposes. to-date. Inaccuracies in data can result in adverse events such as the inadvertent refusal of a good consumer’s credit The ability to use algorithms to match pieces of data is application or the extension of credit to a bad borrower. restricted in emerging markets where crucial basic data such Credit reporting service providers rely largely on data as names, addresses, and dates of birth are often unreliable providers for accuracy of data content. Responsibility for or missing. In Uganda, CompuScan, a credit bureau the input of information, and therefore the accuracy of based in South Africa, sought to overcome this challenge information supplied, should remain with the data provider. by developing a biometric-based identification system for However, the CRSP is responsible for validating the data financial institutions (see Box 5.1). before uploading it onto its database. The data-capturing system of the service provider should not allow alteration of the records supplied by the lender. Although a service 5.3.2 Data Sufficiency provider may accept or reject a file supplied by the lender, As discussed in Chapter 1, section 1.3, several studies it cannot make changes to the file, thus limiting the service have shown that inclusion of data from nonbank lenders provider’s liability in the event of information errors. into a credit scoring model generates scores with a higher predictive power, whereas credit reporting fragmented by A credit reporting service provider should have a method industry has less predictive power. Credit reporting service for consolidating data into uniform formats. If information providers should collect both negative and positive data to is incomplete, it should have a method for matching and provide lenders with the most comprehensive picture of merging separate pieces of data to construct a complete file their portfolios. Broadly speaking, all data that is relevant on a data subject. Ideally, the credit reporting service provider for an analysis of creditworthiness, including data in and the data providers should agree on minimum data public records and private nonfinancial sources, should be inputs, and on methods to store data subject information collected. In many countries the collection of irrelevant data 105 World Bank 2011a, General Principle I. 106 Ibid. 54 Credit Reporting Knowledge Guide Box 5.1 Adoption of Biometrics in Uganda A feasibility report commissioned by the Central Bank of Uganda in the mid 1990s precipitated the development of a credit bureau to serve the banking community and its clients. In establishing the credit bureau, the lack of an adequate borrower identification system in Uganda, (i.e. an official ID), prompted the central bank and its technical partner, CompuScan of South Arica, to consider the option of developing a biometric-based finance card to uniquely identify borrowers. The biometric fingerprint system, known as the Financial Card System, allows licensed financial lenders, with the help of the credit bureau, to link borrowers’ fingerprints to their loan repayment information across any institution in Uganda. After enrolling in the system, each borrower receives a financial card. A national mandate that all banks and financial institutions in Uganda issue financial cards to their borrowers to enable proper identification necessitated the deployment of biometric hardware throughout the banking sector to take fingerprints and photographs of potential borrowers. In 2009, all bank branches in Uganda had the software and hardware set up allowing for enrollment to commence. The solution was developed to work both online and offline with direct hook up to the fingerprint database. The credit bureau was rolled out in stages. Data collection from participating institutions is at an advanced stage with 95 percent of the institutions supplying monthly data loads to the bureau. For almost a year, CompuScan spent significant effort and resources enabling financial institutions to cleanse date and develop systems and processes so that data can be shared with the bureau. Even for the most advanced lenders, these requirements were extensive, and detailed project planning was required to ensure that the project started correctly. As for financial cards, a number of challenges emerged. For example, the costs of compliance have proven prohibitive for some institutions as they had to purchase additional hardware to satisfy customer service levels, establish or increase internet access, and invest in new credit control processes and methodologies. Also, the expansion of financial card to a wider lending audience such as microfinance lenders, retailers, and telecommunication companies has been limited. Uganda, much like any other economy, is subject to risks of fraud, especially impersonation fraud. The financial card solution significantly reduces banks’ (and individuals’) exposure to this type of fraud. By June 2012, monthly registrations had declined because a majority of borrowers had been identified on the system. Twenty-nine institutions with more than 550 enrollment / registration outlets were established to assist in new client enrollment or customer verification. As a result of the ability to properly identify the customers using a common platform, lenders have been able to reap the intended benefits of the nation’s new credit bureau. is explicitly prohibited. Irrelevant data includes data about some countries, such as Australia, positive credit reporting a consumer’s race, medical status or history, religion, or is prohibited by law, although discussions are underway other information that is deemed immaterial for purposes to move to a positive reporting system. In other instances, of analyzing creditworthiness. nontraditional providers of data, such as utilities and telecommunications companies may fall under a different Whereas, in principle, a CRSP would choose to collect data regulatory purview than traditional data providers like from as many different sources as possible, in reality, legal banks and financial institutions. The respective legal and and regulatory frameworks surrounding credit reporting regulatory frameworks may not permit data sharing with may not support data collection from some sources. In credit reporting service providers. 55 Credit Reporting Knowledge Guide Some public records, such as identity registries for individuals 78 credit bureaus) reported that data requests were met and businesses, might not be available to the public or access instantaneously. All but 3 percent of the credit bureaus may be restricted. CRSPs should seek to negotiate special filled requests within seven days (see Figure 5.1). The key agreements with public records agencies to ensure a smooth indicators for the timeliness of service include: and systematic flow of information crucial for validating • Time between obtaining the query and issuing the the identity of the data subject. In some cases it may be report: In many countries, the process is automated. necessary to define a cost-recovery scheme to alleviate the Depending on the search capacity of the software, it may financial burden on the public agency. Depending on the take just a few seconds. In many developing countries legal and regulatory environment facilitating access, CRSPs where the reports are not provided online, the process may also enter into agreements with private data sources to may take hours or, in some cases, days. Minimizing the collect data. To ensure that all CRSPs in the market have delivery time is an important objective for the CRSP. access to a wide range of data sources, it is recommended that data providers and other data sources do not enter into • Time to assimilate data and update records: This exclusivity contracts with any specific CRSP. refers to the time between receiving data or updates from the data providers and its integration into the In addition, the technology platform of the service provider CRSP’s database. Validating and merging data received must be designed to receive data in different formats. In from lenders may take anywhere from one day to one some markets small banks and nonbanking financial month108 depending on the quality of the data supplied institutions may be unable to provide data electronically. by the lenders, the reliability of identifiers, or the The service provider should have the capacity to accept data merging algorithm. This parameter is critical to ensure on DVDs, CDs, diskettes, magnetic tape, or other portable that the data available to lenders is up to date. data storage devices as long as they are secured by encryption • Time to correct errors: Of the 78 credit bureaus or another appropriate method. It should have the ability to surveyed in the Doing Business survey, approximately systematically upload new data onto its platform. 76 percent reported taking less than two weeks to rectify errors. Another 8 percent reported taking between two 5.3.3 Data Timeliness Data should be made available in a timely manner because Figure 5.1: Average Time Between Request and creditors make critical credit-granting decisions based on Release of Data the information they receive from credit reporting service 7 days – 1 month 3% providers. This timeliness requirement requires data providers and other data sources to update their databases 1 – 7 days frequently (i.e. within a specified number of days after the 37% occurrence of a specified relevant event, or at end of each billing cycle). Updated data must be provided to the CRSP systematically, usually on a predefined schedule as agreed by the CRSP and data providers. Updated data should be Information is released instantly incorporated into credit reports, which should be accessible 58% to subscribers as soon as practical. The World Bank’s Doing Business survey data107 Within the same day 2% indicates that 58 percent of credit bureaus (in a survey of Source: IFC calculation based on Doing Business 2004 data. 107 Doing Business Indicators (database), 2004, “Getting Credit” indicator. 108 Ibid. Based on information from 62 credit bureaus. 56 Credit Reporting Knowledge Guide weeks and one month to correct errors.109 Generally, when In some countries with a negative-only reporting system, a CRSP finds errors in a file, it sends a correction to the once a bad debt is paid off, all negative data related to it is data provider, who has the responsibility of correcting deleted from databases, either because it is mandated by law errors. or because it is common practice in the marketplace. Such a practice may paint a false picture of a borrower who may be a recalcitrant debtor who pays off an old loan only to get 5.3.4 Data Retention110 a fresh loan, which he or she then fails to repay. Conversely, Data should be retained safely for a sufficient amount of disclosing data, especially negative data, for excessively time. Most credit reporting service providers retain data long periods (more than five years) can unduly penalize a from five to seven years. In some markets, the length of borrower who has otherwise reformed his or her payment time that data may be stored is restricted by legislation. habits. Most countries opt to limit the number of years The retention period of data is determined by the purpose that negative information may be shared to give previously for which the data is to be used. On the one hand, data delinquent borrowers a second chance at accessing credit. should be kept for a sufficient amount of time to allow for debt collection and reducing the risk of overindebtedness, The agreement between the CRSP and the data providers which, based on global experience, ranges between five will usually stipulate how long information will be shared, and seven years. In jurisdictions where credit scoring and when it will be archived (and for what purposes archives other value-added products have been developed, data may be used), and when it will ultimately be deleted. In should be retained for at least three years to allow sufficient practice, data is archived rather than deleted so that it is observations to build predictive scores. Data for supervision always available, but is no longer distributed after a defined and statistical purposes may need to be retained for a longer time period. period. For example, in the United States, data remains on credit bureau records for two, seven, or nine years, depending on the type of credit or debt. In some countries 5.4 Ensuring Financial (e.g., Brazil), data is never deleted from the database. Sustainability A distinction should be made between the length of time A CRSP needs to be financially sustainable, regardless of the that data is retained, and the length of time data is included market in which it operates, and regardless of the primary in a credit report or disclosed. Typically, legislation provides function it performs. The size of the credit-active population guidance on how many years data can be disclosed, which, dictates the level of sophistication and complexity of the based on information from Doing Business surveys, ranges credit reporting system to be implemented. In emerging from three to five years. Different types of data are subject to economies, a very large proportion of the population is different distribution time limits. For example, data relating often unbanked, with the result that existing credit accounts to previous inquiries (the data footprint that is left on the reflect only a small percentage of the potential market. credit bureau each time an institution requests a credit Credit bureaus depend on volume (number of inquiries report on a data subject) is of little value to lenders beyond or consultations by their users) to be sustainable and to 12 months, and is, therefore, usually masked from credit generate profits. Although credit registries are not focused reports after a year. In Brazil, while data is never deleted, it on profits, they should have access to a consistent source of may not be distributed beyond a certain number of years; for funds to maintain registry operations. instance, negative information is distributed for five years and positive information is distributed up to 15 years.111 109 Ibid., information not available for 13 credit bureaus. 110 See also Section 4.2 of this Guide. 111 Doing Business Indicators (database), 2012, “Getting Credit” indicator. 57 Credit Reporting Knowledge Guide CRSPs (mostly credit bureaus) make their profits by selling 5.5.1 Model 1. Credit Bureaus reports in response to queries from their users/members. Chapter 2 covers the basics of credit bureaus and discusses Without a large borrower base, CRSPs would have to charge the potential range of shareholders or owners of bureaus. high fees for their credit reports, which could reduce the Some markets demonstrate a willingness, as determined by demand from lenders. In countries where the use of credit stakeholder interest and readiness, to allow credit bureaus is not widely prevalent, CRSPs might face this challenge to provide credit reporting services. The most common in their initial years of operation. Developed countries ownership structures are bureaus in which creditors/ with small populations, such as Iceland (population lenders are shareholders and bureaus that are independently 320,000) and New Zealand (population 4.5 million) are owned and operated (see Section 2.2.1 for benefits and able to operate small but profitable credit reporting services disadvantages of each structure). because their populations, though very small, use credit markets actively. For example, in New Zealand, where Regardless of the ownership structure, a key consideration the economically active population is estimated at slightly in determining the optimal model for a credit bureau is more than 2 million, one of the three credit bureaus receives whether to host the bureau on shore or off shore. Credit about 4.5 million queries a year. reporting is a capital-intensive business in which significant In emerging markets where the economically active investments are required for start up, and for the continuous population is too small to generate sufficient demand technological updates required (e.g., quality, security, from lenders, a regional solution may be the viable option. integrity of data, value-added services development, TransUnion Central America, for example, operates a compliance with the legislation). Countries with large regional credit reporting service covering five countries. populations (such as China, India, Brazil, Indonesia, Russia, and Mexico), solid consumer credit industries, and credit culture normally represent an attractive business case for 5.5 Creating an Appropriate international credit bureau operators. Significant volumes Business Model of inquiries dramatically shorten the break-even period for a bureau to attain financial sustainability, thus allowing the After conducting its market assessment, the proposed credit business to generate earnings and profits. reporting service provider should have an overview of the market environment and be ready to move into the “design Conversely, markets with smaller credit-active populations and build” phase. Since conditions differ from country to lack this appeal, and may face more difficulty attracting country, the best design is one suitable to a country’s market large international bureau operators. Fortunately, an environment, taking into account global best practices. alternative option, the Hub & Spokes model, which has Accordingly, the results of the market assessment will direct been successfully tested in small markets (Latin America, the next steps: deciding on the best model for the credit Europe, Pacific Islands, and Africa), offers a viable solution. reporting service, developing a business plan, and creating Box 5.2 shows the example of a Hub & Spokes arrangement an enabling legal framework. in Central America. The market assessment influences the model and business The Hub & Spokes model is optimal for smaller markets structure of the proposed credit reporting service. The most where establishing individual CRSPs would not be common models112 are: economically viable. Under the Hub & Spokes structure, • Credit bureaus a single, internationally operating CRSP is set up to serve multiple small markets. As the name suggests, the “hub” • Credit registries houses data in silos from each country while each “spoke” • Public-private credit reporting service provider. receives and delivers secure data to the respective country 112 See also Chapter 2 of this Guide. 58 Credit Reporting Knowledge Guide Box 5.2 Hub & Spokes Model in Central America A successful Hub & Spokes model for credit reporting – TransUnion Central America – operates in Central America. Established in 1999, TransUnion Central America now has a hub in Guatemala and regional spokes in Honduras, El Salvador, Costa Rica, and Nicaragua, covering a population of over 38 million.113 Historically, lack of full-file credit bureaus in Central America was a constraint to credit access for consumers and micro, small, and medium sized businesses. Developing individual credit bureaus for each of these countries would have required investment disproportionate to the scale of the individual markets. A credit reporting system that could serve all five countries provided an optimum solution. The individual country service providers (spokes) share and leverage the modern and sophisticated technological system that has been developed in the hub, allowing for improved efficiency. Furthermore, the creation of a single cross-border credit reporting service facilitates the design of standardized products and services across all five countries, which greatly benefits lenders with cross-border business operations.114 Figure B5.2.1: Hub & Spokes Model in Central America Guatemala Honduras Nicaragua El Salvador Costa Rica Source: IFC 2012. 113 The hub was originally established in Costa Rica. In 2007, it was moved to Guatemala because of higher quality of telecommunica- tions networks, and lower operational costs. This shift was done rapidly and without any disruption. 114 Guatemala does not have a comprehensive legal framework covering credit reporting or data protection to facilitate the institution of this Hub & Spokes model. Any changes in the legal and regulatory framework in the hub or any of the spokes could have implications for this model and the bureau in terms of being compliant with the framework. 59 Credit Reporting Knowledge Guide in which it is based. This configuration centralizes many 5.5.2 Model 2. Credit Registries of the common, repetitive, and time-consuming tasks such The market assessment might indicate a market preference, as data cleansing, security, customer-support, and system particularly by the central bank, or other financial sector maintenance. This model leverages highly sophisticated supervisory authorities and regulators, to develop a credit security systems that are already in place for the hub, and registry to meet the credit reporting demands. Chapter provides high-security facilities and systems to store data 2 touched on the purpose, features, and organization of from the spokes at a fraction of the cost of creating such credit registries. Registries are generally operated by central secure facilities from scratch. Because of the sensitive banks or other authorities charged with a supervision nature of the data housed by a CRSP, the selected service function in an economy. Given that these registries house provider must have extensive experience in managing a data that enable authorities to monitor the systemic risk credit reporting service to international standards and be levels in a market and maintain financial stability, credit able to ensure that no data is shared across the silos without registries are typically hosted in the country in which they a data-sharing agreement. The Hub & Spokes approach are established. The principles of data quality, integrity, not only offers top service quality for users/data providers, security, and financial sustainability apply to the operations but is also a way for small emerging markets to overcome of a credit registry. innumerable challenges linked to the time and cost of developing credit reporting services. Other advantages In some instances, the entity housing the credit registry include reduced staffing needs and personnel training costs, may also be responsible for overseeing its operations and and the ability to leverage products, technical experience, ensuring that it is in compliance with the legal framework. and the sophisticated value-added services used in advanced In Bangladesh, for instance, the central bank is charged markets. The web-based technology used by most bureau with operating the credit registry, as well as overseeing its operators allows easy inclusion of other countries / lending operations. Given the inherent conflict in this situation, it sectors regardless of size. would be advisable for the central bank to entrust the two functions – operation and supervision – to two separate Another example of the Hub & Spokes model is found in departments to ensure the integrity of the system. For South Africa, where TransUnion runs a credit bureau that example, in China, the credit registry, the Credit Reference services Namibia and Botswana. In Europe, some of the Center is operated by the People’s Bank of China, with its large credit bureaus operate the Hub & Spokes model or supervision falling under the Credit Information Services offer business continuity coverage to off-shore operations Bureau, a separate department of the People’s Bank. through a similar configuration, including an outsourced arrangement for the Czech Republic and the Slovak Republic operated from Italy by CRIF (see Box 5.3). 5.5.3 Model 3. Public-Private Credit Several emerging markets, such as the UEMOA region Reporting Service Provider in West Africa, and some island nations in the Caribbean In some instances, market stakeholders may indicate a (Belize, Bahamas, Barbados, Guyana, Suriname and the preference for a hybrid model, which involves both the eight OECS countries - Anguilla, Antigua & Barbuda, private sector and the public sector. This model is based on Dominica, Grenada, Montserrat, St. Kitts & Nevis, St. a strong and significant partnership between the public and Lucia, and St. Vincent and the Grenadines, are considering private sectors, in which the public sector plays a significant a Hub & Spokes approach. In the Pacific, Fiji, Papua New role in developing the infrastructure and process for credit Guinea, Tonga, and Vanuatu are currently participating in information collection and sharing. Central banks, in a Hub & Spokes arrangement, with the hub hosted in New their capacity as regulatory and monitoring bodies for Zealand. Two other island nations, Samoa and the Solomon financial institutions, are well placed to steer required legal islands, intend to join the Hub & Spokes arrangements as reforms, and also to build awareness about the benefits of soon as they are available, tentatively in 2012 or 2013. information sharing among financial institutions. In many 60 Credit Reporting Knowledge Guide Box 5.3: Outsourcing from the Czech Republic Banks in the Czech Republic were eager to get the credit reports that credit bureaus would supply, and they were willing to pay for them on a per-transaction basis. But they were not willing to invest in the development of a costly data security infrastructure. The solution was to outsource the operations of the credit bureau to CRIF, a leading Italian credit bureau. In partnership with CRIF, two credit bureaus were set up in the Czech Republic, a banking bureau in 2001 and a nonbanking bureau in 2004, both using CRIF’s facility in Italy. In 2006, the two credit bureaus began sharing credit information with each other (based on consumer consent), thereby allowing financial institutions to have access to reliable cross-industry credit information. Using CRIF’s platform, banks were not required to invest in the development of the credit bureau infrastructure such as a new local data center and related security infrastructure, or hardware and software. They were able leverage the shared data center of CRIF Italy, and were thus able to benefit from a higher level of data security than would have been conceivable in an in-country bureau. Fortunately, the legal environment posed no problems: Czech law states that personal data can be processed abroad, provided that the hosting country abides by data protection laws that are the same or stricter than those in the Czech Republic. The business model based on outsourcing was designed to achieve the most cost-effective solution, along with the highest level of security. It also generated positive impacts on the overall operations of the two credit bureaus. With a local staff fundamentally focused on clients instead of IT issues, the bureaus achieved a much faster start- up both in terms of data collection and data dissemination. Best practice internal processes were put in place to fully integrate the two cross-border technical structures. Last but not least, the technical and process environment facilitated development of value-added products, reducing cost and time to market. The bureaus have reached almost full penetration in the retail banking market, with 26 member banks (nearly 100 percent market share) and 27 nonbanking financial institutions (over 80 percent of leasing and consumer finance lenders). The banking bureau match rate or hit rate115 is 90 percent, which is comparable to the hit rate in the most developed bureau markets. Nearly 14 million records are in the banking credit register, covering over 5 million people. The nonbanking bureau now has an additional 3.2 million credit files on 2.2 million people. Inspired by the success of the Czech credit bureaus, the Slovak Republic chose to develop an outsourced credit bureau as well. In partnership with CRIF, a banking bureau was set up in 2004, and a nonbanking bureau in 2008. Both bureaus operate out of Italy. 115 This is the ratio of the number of reports issued to the number of queries received and is an important indicator of the ability of the bureau to satisfy lenders’ demand for information. 61 Credit Reporting Knowledge Guide countries, financial institutions have a high degree of trust 5.6 Identifying Appropriate in their central bank’s role as an independent third party. In Technology Needs the absence of a data protection authority, central banks are often in position to leverage this “capital trust” to establish CRSPs require adequate technical infrastructure and credit reporting services in partnership with the private communications networks to process and manage data and sector. databases, as well as to offer effective and secure delivery of credit reports to their clients. CRSP technical infrastructure This model offers several advantages, notably providing the systems are not off-the-shelf solutions that can be acquired central bank with a wealth of free information to enable it to and installed into a computer hardware system. CRSPs must perform its primary function of monitoring and managing develop or acquire locally adapted and customized systems systemic credit risk. This model also: that will enable data collection from existing and new data • Prevents the creation of a monopoly on information sources. The development process may take 6 months to sharing by allowing as many local and international 18 months, and involves an analysis of available data from private entities as possible to enter the market where the data sources, preparation of functional specifications, actual size of the market supports competition system development, and acceptance testing. The process of lenders extracting data from their core systems is one • Lays the groundwork for the creation of a solid, of the most challenging and potentially time-consuming competitive, and dynamic information-sharing market, elements that must be addressed as a CRSP is established which will allow for competition in terms of prices and and should not be underestimated. quality of services, with the obvious resulting advantages for lenders and consumers Cost should not be the only driver in the decision to • Establishes a complete and seamless credit information develop or purchase a technology platform for a CRSP’s system that is accessible to all lenders operations. In addition to a solid technical infrastructure, • Facilitates the inclusion of data provided by entities not a CRSP requires unique knowledge and experience because regulated by the central bank. of the complexity of its technical infrastructure and high sensitivity of the data held. In some emerging markets, One key disadvantage of this model is the duplication of newly established CRSPs face a shortage of specialized effort involved in setting up the technical infrastructure. As information technology and business skills. Accordingly, the collector of data from the entities that it supervises, and in such markets, the participation of an established and the distributor of data to CRSPs, the central bank must experienced CRSP, either as a shareholder or technical establish a basic technical infrastructure (a data warehouse). partner, can benefit a new CRSP in terms of technical Furthermore, the central bank as aggregator of data must expertise, reputation, crucial business know-how, and the have and maintain the capacity to continue to provide expertise to develop value-added services as the CRSP this service, which may be costly. Some public-private grows and the market matures. partnerships are discussed in detail in the case studies on Ecuador, Egypt, and Morocco in Chapter 7. Technological advancements in the last decade have dramatically alleviated the cost of developing information The models described above are not exclusive. A country technology systems for the credit reporting industry. Until may have a registry and one or more bureaus operating side a few years ago, the industry mostly operated on heavy by side. The models used are determined by the market and costly mainframes. It is now possible for new CRSPs assessment, in particular, by the stakeholder assessments, in emerging markets to acquire information technology which reveal which structure is preferable for the market as platforms from external sources—usually internationally well as the optimal number of bureaus as determined by the reputable technology providers—rather than “build” them size of the credit market. in-house. Figure 5.2 lists the qualities of a strong technical partner. 62 Credit Reporting Knowledge Guide Figure 5.2: Qualities of a Strong Technical Partner Technical Strategic • Experience (years) • Willingness to add value to business plan and financial model • Track record / success with setting up credit reporting services in developing and • Willingness to take equity positions developed economies • Financial strength of company • Expertise of personnel / management team • Management profile • Ability to provide comprehensive solutions • Availability of office / skilled resources in or near (products, software and value-added services). project country • Understanding of domestic banking / credit market and related issues • Direct relationship (no 3rd party) • Willingness and proposal for know-how transfer. Source: IFC 2012. When selecting a technical partner, the CRSP should These functions are described in more detail in the sections evaluate a potential partner according to the following below. criteria: • Technical: Does the potential partner have the capability 5.6.1 Collect, Validate, and Merge Data to implement the system in accordance with the local The success of a CRSP’s operations depends on its ability to technical specifications? Does it have a track record extract credit performance data from financial institutions in implementing credit reporting services in similar and other lenders, and deliver credit reports in an easy- markets? to-use format. In countries as diverse as Russia, India, • Strategic: Is the potential partner able to commit to the and Egypt, extracting data in a format acceptable to the CRSP over the long term? respective CRSPs was a major challenge that required • Financial: Is the cost of the system in line with the substantial investment in information technology resources demand for services? to upgrade old legacy systems. It has proved easier to extract credit card records, which tend to be hosted on modern The CRSP’s technology system must perform the following systems that store data in a logical format. Legacy banks, basic functions: often state-owned or recently privatized banks, and MFIs with large branch networks face a major challenge because • Collect, validate, and merge data often their records are paper-based and their credit functions • Generate and distribute reports decentralized. For CRSPs operating in these markets, the practical solution is to start collecting credit portfolios that • Provide data security and backup. have better-quality data from banks that are able to provide such data easily, and then gradually start collecting data from more lenders and more portfolio types. 63 Credit Reporting Knowledge Guide The CRSP is responsible for validating all data it receives 5.6.2 Generate and Distribute Reports before uploading it. The initial phase may be labor When enough data has been uploaded and the CRSP’s intensive. The CRSP’s system must include automated process for validating and merging data is in place, the processes to check for completion of all mandatory fields CRSP is ready to generate reports. The reports remain and conformity to the standard format. The system must available on the CRSP’s database for use by users. Figure also be able to reject files that have critical errors or missing 5.3 shows several common delivery modes used by CRSPs. information and return them to the data provider to resend The typical modes of access for users are: a corrected file. • Online access: The user’s system is connected to the After the data have been validated, the CRSP must merge CRSP’s interactive system, from where the user extracts the new data into its database. The system must be able to reports as required. The interaction is system-to-system locate the respective subject, be it an individual or a legal that is, performed entirely through the user’s system with entity, using national unique identifiers, such as passport no human interaction. Host-to-host connectivity may or identity card numbers or tax IDs or other match and be a good solution for a newly established CRSP, since merge techniques discussed in Section 5.3.1. The objective some data providers with large volumes of customer data of the CRSP is to be able to match the incoming data with could integrate their database system with the CRSP’s the single best possible match from all the files held on the system, thereby eliminating data duplication and bureau database. streamlining work flow. • Dialup or Web: The user accesses the CRSP’s system Once the correct subject file has been identified, the system via traditional internet browsers and PC software. Once will update the existing record or, if the information relates connected to the CRSP’s system, the user provides to a new borrower, create a new credit file in the database. authentication information (user name, password,) to Figure 5.3: Common Delivery Modes for CRSPs 160 140 Number of Credit Registries and Credit Bureaus 120 100 80 60 40 20 0 Internet Modems/Dedicated Computer Telephone Facsimile Mobile In person Written paper phone lines disks/CDs consultations documents Delivery Modes Credit Registries Credit Bureaus Source: IFC calculation based on Doing Business 2012 data. 64 Credit Reporting Knowledge Guide validate access. This mode of access is less expensive • Periodic testing of backup hardware and recovery plans and is preferred by users who are either not technically • Delineating authority among network administrators capable of permanent system-to-system connection, or and staff who submit limited inquiries to the CRSP. • Ensuring physical security of the facility, the systems, • Batch access: Data providers deliver information to the and the data CRSP electronically or via portable storage devices. The batch access method provides users with a cost-effective • Organizational security policies and procedures for means of processing large volumes of inquiries. It is handling different data security breaches. usually recommended for processing of risk monitoring for large client portfolios. The CRSP should create a plan for responding to different threats and assign specific accountability to different • Consumer access: Consumers seeking copies of their personnel (e.g., network administrators, IT directors) for own reports must be able to either approach the credit ensuring compliance with security policies and procedures. bureau in person, via an approved agent network, or, as CRSPs should develop and routinely test business continuity in Singapore, via a sophisticated web-based solution. plans. CRSP management should provide for regular audit checks to ensure adherence to and enforcement of security 5.6.3 Provide Data Security and Backup policies and procedures. Staff should be aware of the security policies and procedures, changes to these policies Data security116 is a high priority for CRSPs and data and procedures, and consequences of violating the policies providers because they manage highly confidential and procedures. Extensive background checks should consumer information. Secure systems protect the data be conducted on new hires. In addition, management and reports and in doing so protect the CRSP’s integrity should review and update security policies and procedures and reputation. The enormous amount of data collected is periodically to ensure that they are consistent with several stored in database systems that are subject to concerns such factors, such as changing standards for data security, as loss, tampering, destruction, theft, or misuse. Specific changing regulations, and system upgrades. measures and safeguards should be adopted to cope with the logical, physical, and organizational aspects of data security; with the objective of containing, limiting, and 5.7 Operational and Practical responding to data security breaches. Ensuring data security Considerations is an ongoing obligation and safeguard measures should be regularly reviewed and updated to be effective against newly The first operational task of a CRSP is collecting data from emerging threats. Security policies might include: data providers and uploading the data onto its own database for further processing. Data sharing between CRSPs and • Limiting access to the database via mechanisms for data providers/sources is usually governed by agreements identifying and authenticating users (including staff and between the parties. Since the principle of reciprocity is one contractors) of the bases for exchanging information, data providers are • Maintaining and monitoring logs to track each access to generally also the users of data. In some exceptional cases, the database a data provider (e.g., a public data source) may agree to • Protecting the database against cyber breaches (hackers) only supply information and not make inquiries of the CRSP. Figure 5.4 summarizes the key issues that should be • Maintaining a database back-up addressed in agreements between CRSPs and its users and • Continually updating all items stored in the offsite data providers. recovery database 116 According to General Principle II, “Credit reporting systems should have rigorous standards of security and reliability, and be efficient.” World Bank 2011a. 65 Credit Reporting Knowledge Guide Figure 5.4: Key Items in Contracts / Agreements with Users and Data Providers Contracts / Agreements Data Users providers • Principles of reciprocity • Protection of the CRSP, users and consumers • Rules of data sharing • Commercial conditions on external data provisions and usage • Data ownership • Restrictions on the use of the data • Usage protocol • Stipulate frequency of updates • Confidentiality • Notification of errors in the information • Cost of services and availability • Specify type, media and format of updates • Adherence to data protection and / or consumer credit legislations • Notification of changes to type, media and / or format of the data • User obligations to provide accurate, timely data • Process to ensure high data quality standards • CRSP responsibility to process data, maintain integrity and security • Data protection implications with the data supplied • Other clauses concerning claims, costs, damages • In case of outsourcing, a Service Level Agreement and penalties for inaccurate data. defining, obligations, availability of data, access to the database, backup, etc. • Purge rule. Source: IFC 2012. In the case of registries, the legal mandate to provide data Operational phase: Once the CRSP becomes operational will overrule the need for agreements, however, the registry (i.e., the system starts selling its first reports) several factors and data providers still need to agree on data formats, data affect the decision on staffing. A CRSP’s function and its inputs, reporting frequency, mode of reporting, and other employees’ duties are to obtain credit history data from data details. providers and to sort and aggregate the data into personal credit histories. The CRSP’s system then generates reports based on the captured data. Among the factors to consider 5.7.1 Organizational Structure in determining workloads are the following: Pre-operational phase: Initially, staff members should cover • Number of existing and potential subscribers more than one role, whenever possible. The early phase of a CRSP can essentially be run by a general manager/project • Number of branches/workstations connected to the CRSP manager, an office/communication manager, and a technical • Inquiry volumes coordinator. Ideally, employing a general/project manager • Competitors’ strength who is knowledgeable, experienced, and well connected in the financial sector is a critical factor for success. In • Consumer awareness and education needs addition to providing technical assistance, a reputable • Projected and actual database size international technical partner can also provide strategic and • Growth plans for the CRSP business development support to management. Finance, administrative and legal functions can be outsourced in the • Complexity of operations (e.g., need for off-line checks/ beginning. updates overnight or on weekends). 66 Credit Reporting Knowledge Guide The main divisions of the operational CRSP are: IT and who have problems connecting to the system, uploading Operations, Compliance, Business Development and data, and modifying some of their data. They may also assist Marketing/Sales, and Finance/Administration. Divisional new lenders that require additional help in enabling their heads in each area report directly to the chief executive internal systems to interconnect to the CRSP system. officer (CEO)/managing director, who manages the company’s activities and, in turn, reports to the board of The customer service department deals with consumers directors. The board, whose members are appointed by and firms that have queries regarding credit reports or the shareholders/owners of the bureau, is responsible for their information on the CRSP’s database. Staff in this the overall corporate governance. Ideally, the board should department should be knowledgeable on the CRSP’s include an independent director, one or two members redress mechanisms such as registering customer complaints of the executive team (the CEO /managing director and and providing educational information to customers in operations director/representative of the technical partner). accordance with the CRSP’s operation policies. The board of directors nominates one of its members as chairman of the board. Figure 5.5 shows a sample To accommodate the needs of growing numbers of users and organizational structure of a CRSP. borrowers and their respective requests, most of the growth in staff will occur in the customer service department. The Staffing requirements and responsibilities for an operational sales and marketing group also need to grow to promote CRSP are outlined in Table 5.1. the CRSP’s products and services as it seeks to expand into new markets. The database team is responsible for validating all data received from data providers before it is uploaded onto the Last but not least, it is recommended that the CRSP appoint system. The CRSP should operate a help desk to assist users a compliance officer(s) early on in the process of setting up Figure 5.5: Sample Organizational Structure of a CRSP CEO/MD Assistant to CEO Compliance Unit Head of Finance / Head of Business Head of IT & Operations Administration Development & Marketing Administration Officers Sales & Marketing Officers Customer Service Officer Database Officer Network Administrator IT/Support Source: IFC 2012. Services 67 Credit Reporting Knowledge Guide Table 5.1: Operational Phase Staffing Role Key Tasks CEO/Managing Director • Overall bureau strategy • Marketing / business development activities Head of Finance and • Finance and administrative operations Administration • Human resources functions (recruitment, compensation, performance management, career development). Finance / Administration • Day-to-day administrative and bookkeeping operations. Legal Counsel • Overall legal support • Internal legal training. Head of Business • Market segmentation Development & Marketing • Product development • Branding • Advertising • Sales and promotion. Sales & Marketing Officers • Maintain relationship with existing clients and enroll new client • Implement sales & marketing plan and achieve business objectives • Advertising, conferences/exhibitions • Sales and promotion • Market research • Media affairs • Identify new data sources. Head of IT & Operations • Vendor relations • Data management • Technology management • Network and security operations • Customer service. Customer Service Officer • Consumer Help Desk. Database Officers/Analysts • Data validation and quality checking • Data uploading • Emergency updates. Network Administrator • Network administration • Subscriber communications interfaces • Network security. IT Support Service • Housekeeping • System administration • Subscriber and internal Help Desk Compliance Unit • Internal process audit • External compliance • Oversee data quality and dispute resolution process. Source: IFC 2012. 68 Credit Reporting Knowledge Guide operations. CRSPs, which are regulated in most countries, of inquiries generated by these users. In general, lenders in need a compliance officer to ensure that the CRSP is in countries where the size of the credit-active population compliance with its regulatory obligations, its internal is small, will be faced with higher prices. Within a given operational policies, and industry codes of conduct. The market, lenders that generate smaller volumes of inquiries compliance unit should closely and consistently coordinate (based on the size of their lending portfolio), including its function with relevant CRSP departments (including smaller microfinance lenders will be faced with higher database officers and management) to ensure that the CRSP prices. The issue of pricing for microfinance lenders has is in compliance with its internal and external obligations. been noted earlier in the various microcredit reporting regional developments, and continues to be a hotly debated agenda item. 5.7.2 Financial Projections Forecasting financial outcomes of a newly established CRSP The inquiry-demand estimate should be based on a survey requires an assessment of potential revenue and costs, and of potential users. The financial projections for revenue an identification of the drivers in each of these categories. should allow for time between the launch of a CRSP’s operations and the breakeven point at which it achieves Revenue Projections: The main revenue driver for the CRSP its targeted inquiry volume. Technical issues related to is the number of credit reports or value-added services sold. connecting a lender to the CRSP and integrating the CRSP’s Revenue projections are based on the estimated demand information into the billing cycle of the lender institution for credit reports and the pricing of reports. In most cases, are common and may take at least three to six months to the CRSP charges a flat membership fee plus a charge per resolve. The growth rate for the volume of inquiries is based inquiry (per click). Volume discounts usually apply, and on the projected credit growth rate for the economy and the it is common to have a pricing matrix depending on the expected number of new users joining the bureau. Growth volume of inquiries and the type of user. Table 5.2 provides rates of 50 percent and above are feasible in the first three a hypothetical pricing matrix based on the annual inquiry to five years of a CRSP’s operations in a country with stable volume per user. The cutoff points for volume discounts credit growth and new users joining the CRSP. are determined by projected demand and average expected inquiries. Cost Projections. In large part, costs are driven by the choice between acquiring the CRSP’s technology platform It is important to note that the pricing matrix in Table and developing the technology platform in-house. With 5.2 is purely hypothetical and is not intended to provide either choice, the possible cost range is wide and depends a benchmark for any market. Pricing in each market will on the level of sophistication of the system and the types of ultimately be determined by the size of the market in terms products it is expected to provide. of credit active population, the number of records in the database, the number of users of the system, and the volume Cost projections based on the assumption that an existing platform will be acquired should include the following cost Table 5.2: Hypothetical Pricing Matrix for Credit elements: Reporting Service Providers • Development/customization/installation fee for the Inquiry volume Price per Inquiry (in $) technology platform (usually paid in installments) <25000 1.75 • Maintenance fee, usually a flat fee paid monthly, 25001 – 50000 1.00 quarterly, or annually 50001 - 100000 0.95 100001 - 250000 0.85 • License and royalty fees paid to the technical partner 250001 - 500000 0.8 based on the number of inquiries received by the >500,000 0.7 system in addition to fees to cover ongoing updates and Source: IFC 2012. enhancements to the system, usually at an agreed-upon rate 69 Credit Reporting Knowledge Guide • Consultancy fees charged by the technical partner for Typically new CRSPs in emerging markets find it difficult any service over and above the services specified in the to justify the costs of implementing more than rudimentary development and maintenance agreement. disaster recovery solutions. As the CRSP services become more embedded in the business- critical processes of its Other elements to be addressed in the cost projections users—for example, incorporated within scoring solutions include hardware such as database and network servers, or automated application processing systems—service network equipment and workstations, system software availability becomes increasingly important and more applications, office furniture and equipment, utilities and sophisticated and expensive disaster recovery solutions are telecommunications expenses, and labor costs, which can required. be substantial. In some cases, an important cost component is the cost of the data that the CRSP may acquire from Table 5.3 provides a hypothetical profit and loss statement external data sources; for example, a source that contracts over the first five years of a CRSP. to provide data only to the CRSP. Based on this hypothetical financial plan, the CRSP would Over and above the basic operating costs, there will be a break even in the third year of operations. In most cases, variety of costs associated with business continuity and risk CRSPs reach the breakeven point over a three-to-five year mitigation, key areas covered in the general principles.117 Of period (see Figure 5.6). these, the largest component is typically the cost of operating a disaster recovery site—a business contingency in the event In preparing a business plan for the CRSP, it is important that the primary data center becomes inoperable. A variety to assess high and low scenarios for profitability because of disaster recovery solutions are possible depending on the the successful operationalization of the CRSP depends on urgency required to reinstate services, which have a direct many external factors. For example, the CRSP often faces bearing on cost. The quicker the required backup the more start-up delays caused by banks’ inability to upload data. costly the solution. The options fall broadly into three In many countries, historical data is simply not available to categories: populate the database. The first few years may be dedicated • Hot stand-by – effectively a mirror copy of the live to building a database from scratch. Underestimation of database operating in parallel to the main data centre at a different location. If the primary site fails, the back up Figure 5.6: Breakeven Point for a Newly Established system kicks in seamlessly with almost no loss of service Credit Reporting Service Provider to the user. millions 3 • Warm stand-by – a duplicate of the hardware environment that has been pretested but typically does 2.5 not contain live data. In a disaster the most recent data 2 back ups would have to be loaded before services can be resumed, which may take several hours. 1.5 • Cold data back up – a process of taking regular copies 1 of the database and storing them off site to protect against the destruction of the primary site. Although this 0.5 method protects the raw data, which is the real value of the CRSP, in a disaster it may be several days before the 0   Year 1 Year 2   Year 3     ear 4 Y Year   5 service can be reintroduced for users. Total revenue Source: IFC 2012. Total cost 117 World Bank 2011a, General Principles II and III. 70 Credit Reporting Knowledge Guide Table 5.3: Hypothetical Profit & Loss Statement YEAR 1 YEAR 2 YEAR 3 YEAR 4 YEAR 5 Total revenue (in US$) 0 500,000 1,000,000 1,750,000 2,625,000 % change in revenue 0 100 75 50 Costs Operating cost Labor 315,000 346,500 450,450 585,585 761,261 Rent 50,000 52,500 55,125 57,881 60,775 Utilities 1,500 1,800 2,160 2,592 3,110 Office equipment, supplies 7,000 8,000 8,000 8,000 8,000 Telecommunications 14,400 17,280 20,736 24,883 29,860 Audit, legal and other fees 12,000 12,000 12,000 12,000 12,000 Insurance 13,000 13,000 13,000 13,000 13,000 External data, marketing 20,000 25,000 30,000 37,500 46,250 Total operating costs ($) 432,900 476,080 591,471 741,441 934,256 % of total cost 52 55 54 53 53 Fixed costs Rent, furniture, other fixed costs 20,000 20,000 20,000 20,000 20,000 System hardware & software 75,000 75,000 75,000 75,000 75,000 Technology platform 300,000 300,000 400,000 550,000 725,000 % of total cost 36 34 37 40 41 Total fixed cost 395,000 395,000 495,000 645,000 820,000 Total cost (S) 827,900 871,080 1,086,471 1,386,441 1,754,256 % change in cost 5 25 28 27 Net income before (827,900) (371,080) (86,471) 363,559 870,744 interest & taxes ($) Tax 0 0 0 109,068 261,223 Net income after taxes ($) (827,900) (371,080) (86,471) 254,491 609,521 Source: IFC 2012. costs or time required to customize and implement the monitoring of the financial system. Traditionally, the system is common. Usually, this means the CRSP has to experience has been to not charge users (regulated financial pay high consulting fees to the technology provider to institutions) for reports. This was the case for most registries finalize the system implementation, which is likely to delay operating in Latin America and the Caribbean. Thus, the timing of the breakeven point. revenues for a registry would be zero. In some countries (e.g., Lebanon, China, Bangladesh, and soon the Maldives), the Although the need to generate revenue is obvious for credit law empowers the registry to recover operating costs of its bureaus, which generally operate for profit, it is not as clear services. Pricing policies that enable the registry to recover for credit registries. Most credit registries are established costs seem prudent in light of the objective of maintaining under the mandate of a banking law, on a not-for-profit financial sustainability of operations. basis, to enable prudential supervision and systemic risk 71 Credit Reporting Knowledge Guide 5.7.3 Measuring Effectiveness of a Credit one credit line and the history on each credit line is Reporting Service Provider stored separately. The effectiveness of a CRSP, like that of any other business, • Hit ratio. This is the ratio of the number of reports issued can be measured in many ways. A good performance to the number of queries received. It is an important measurement system includes multiple dimensions of indicator of the ability of the CRSP to satisfy lenders’ performance, including financial, operational, and behavioral demand for information. The hit ratio is indicative of characteristics. The key categories for measurement include: the depth of data available in the CRSP. quality, quantity, timeliness of products and services • Number of products offered. This measure could include delivered, financial performance, and customer satisfaction basic reports, detailed reports, credit scores, portfolio (see Figure 5.7). monitoring, and fraud detection. Figure 5.7: Key Performance Indicators of a Credit A CRSP’s objective is to simultaneously increase its Reporting Service Provider coverage ratio, defined as the number of borrowers in the system divided by the economically active population, and its hit ratio. Considering only one of these two measures does not provide an adequate understanding of the CRSP’s Quantity Quality performance. For example, a CRSP could have a high hit ratio but a very low coverage ratio, a situation often found in markets with underdeveloped credit markets. This Customer assessment indicates that the formal financial system serves Timeliness satisfactipon a small group of individuals and most lenders continue targeting the same group for new lending. Financial performance Quality. This category refers to the accuracy, completeness, Source: IFC 2012. consistency, and updated nature of the CRSP’s data. Information, the main asset of the CRSP, only has value if it is accurate and current. Relevant indicators of quality Quantity. This category is a measure of the volume of include: goods and services delivered. Relevant indicators include: • Number of complaints: The CRSP must have a • Number of queries received by the system over the mechanism to receive and log complaints from reporting period. This measure is the key measure of the consumers/borrowers about the accuracy of information demand for the CRSP’s services. in their credit reports. • Number of credit reports sold. This measure is the key • The percentage of complaints with inaccuracies due output measure for the CRSP. It can also be tracked at to actions of the CRSP: Many complaints that a the product level, for example how many basic reports CRSP receives may be unjustified or result from errors are sold, how many reports with credit scores are sold. stemming from the data provider. Tracking the number • Number of borrowers with credit records in the system of complaints that can be attributed to the CRSP’s at the end of the reporting period. This measure can be actions allows the CRSP to improve the quality of its tracked for different categories of borrowers, such as processes. firms and individuals. • Data quality reports: The CRSP should run data quality • Number of records in the system at the end of the reports to analyze the completeness and consistency of reporting period. Each borrower may have more than the data. Such reports produce tabulations of fields such 72 Credit Reporting Knowledge Guide as IDs and addresses, dates of birth, and other identifying • Profit margin per client. Bureaus aim to attract large information, and allow the CRSP to determine whether creditors by providing significant volume discounts. there are duplicate or incomplete files in the system. On the flip side, smaller creditors such as microfinance institutions are less likely to pay the same prices for • Number of rejected files: When accepting a data file from credit reporting products as their banking counterparts. the data provider, the CRSP runs simple consistency The bureau would stand to gain by offering lower prices checks on the data (e.g., checking for minimum inputs). to small creditors to attract greater numbers of them If the file does not pass this test, the system rejects it and to enroll as bureau users. Analysis of profit margins by sends it back to the data provider. Tracking the number client allows a bureau to better tailor its pricing strategy. of rejected files allows the CRSP to monitor the quality of data available in the market. • Registries that do not operate for profit will nonetheless want to closely monitor the sustainability of their Timeliness. CRSPs should monitor their performance operations year after year. based on how quickly they can respond to inquiries/requests from users, how quickly they can turn around requests to Customer satisfaction. Methods used to measure this rectify errors, and how quickly they can update, assimilate, category include customer surveys or actions taken by and merge records.118 customers, including: • Number of complaints. By tracking complaints from Financial Performance. Whereas return on equity, profit lenders and data subjects separately, the bureau can margins, and operational costs are standard indicators of identify areas for improvement. financial performance, the CRSP may also track more specific indicators, such as: • Average time to resolve complaint. Providing fast responses to complaints is one way of improving client • Profit margin per product line. The services that CRSPs satisfaction. One approach is to operate a help desk (mostly bureaus) provide vary greatly and are bound to with staff available to answer questions and complaints have different levels of profitability and cost structure. promptly. For example, while the CRSP may sell raw data at a relatively low cost, it may sell analytical products, such • Systematically tracking a set of key indicators enables the as credit scoring and portfolio monitoring, at higher CRSP to monitor its performance and formulate a clear margins. strategy to improve service. 118 See also Section 5.2.2 of this Guide. 73 CHAPTER 6 Developing Value-Added Services V alue-added services (VAS) comprise a broad class efficiency. However, raw data in the form of a credit of products that more sophisticated credit bureaus report can be extremely difficult to integrate into such can offer.119 Such services entail the processing and systems. Fortunately, many types of VAS (e.g., application analysis of raw credit and financial data to produce tools processing systems and behavioral risk assessment), lend that can be easily integrated into other financial products well to inclusion within automated systems. and tools. The range of potential value-added services is extensive and includes, but is not limited to: The major benefit of automated decision-making systems is • Marketing services that they allow users to manage many customer decisions on an exceptions basis, rather than having to review each • Credit scoring case. This ability reduces the need for employing highly • Application processing experienced, and often very expensive, individuals to make mundane or rudimentary decisions, and allows lenders to • Portfolio monitoring channel employees into more productive tasks. • Fraud detection Larger financial institutions that operate in developed • Collections. markets typically develop customized value-adding tools, either using in-house analytical teams or contracting Raw credit data can be useful in each of these areas, with one of the numerous specialized companies that however, significant time, resources, and expertise is have emerged to service this market. Smaller financial required for proper analysis and interpretation. A variety institutions, particularly in emerging markets, may have of techniques, ranging from simple data aggregation and customer databases that are too small for such solutions cross-referencing to complex statistical algorithms, can be to be statistically reliable, or find it difficult to justify the employed to provide the lender with a simple interpretation capital cost of development. of the information (e.g., a risk score). In emerging markets, therefore, the credit bureau can 6.1 Automated Decision-Making play an important role in making these services available to a broader audience, by pooling data across a range of Systems customers, and by spreading the cost of development across Given the volume of decisions often required to manage a its user base. typical retail portfolio (e.g., grant/reject facility, overlimit authorization, cross sell/up sell, past due action required), Although users still have to pay for these services, typically many lenders have turned to automation to maintain on a “pay as you go” or “per click” basis, they get immediate 119 Based on the functional differentiation between bureaus and registries, discussed in Chapter 2, value-added services generally fall under the domain of bureaus, although some registries, such as in France and Palestine, do offer credit scoring products. 75 Credit Reporting Knowledge Guide access to the benefits of improved lending methodologies, acquisition (loan processing), customer relationship more cost-effective processes, and increased operational management, and collections. efficiency that, under other circumstances, would be available only to larger institutions. The credit bureau, typically, builds products or solutions that help its customers in each of these business functions make better or faster decisions by using the predictive nature 6.2 International Industry Trends of bureau data. In effect, the bureau is recycling its databases in Developing Value-Added so users access files more than just at the point of an initial Services loan inquiry. For example, a behavioral scoring system may access a customer’s credit file monthly to identify updates The range of value-added services offered by credit bureaus rather than only at the point of application. has broadened significantly over the past 20 years fueled both from the demand side—users wanting increasingly Some VAS may be no more than enhanced bureau reports, sophisticated products—and the supply side—bureaus such as an alert service that pro-actively advises a lender of trying to increase/maintain income margins in an a change to a customer’s file, and requires little in the way environment of downward pressure on commodity prices of analytical expertise. Having introduced these services at (the cost of the raw data). a relatively early stage, most credit bureaus aim to move up the value chain to add increasingly more sophisticated The scope of products offered is a function of the tools, such as scoring and credit information management environment in which the credit bureau operates; that is, software. These more complex solutions have the dual the extent to which the raw data can be used. The trend benefit of generating greater revenues for the bureau, and in developed markets has been to create a suite of value- also of locking clients in to bureau services, (i.e., making added products aligned to what is sometimes referred to as users more reliant on the supplying bureau and thus less the “customer life cycle” (see Figure 6.1, which mirrors the likely to turn to competitive sources of information). core business functions most lenders apply when managing customers: prospecting and marketing, new business More mature bureaus tend to use specialized internal analytical teams to develop and maintain these value-added services. More frequently, however, bureaus outsource their Figure 6.1: Customer Life Cycle: Offering Value-Added Services development, often to the same specialized vendors that supply custom services directly to the lenders. The critical issue, however, is not who develops the services, but when Collections Prospecting and they can be offered. Skip Tracing Marketing Debt Management Customer Profiling / Acquisition Geodemographics In most developed countries, credit bureau databases have File Access Prospect Lists had many years to develop, are rich in information, and (Consumers) Mail Screening usually offer high-quality data, thus providing an ideal base Workout Prospecting for data mining and data modeling. In many emerging Customer Customer Customer New markets, however, credit bureau databases are considerably Relationship Management Acquisition Business Management Acquisition less rich: they may have information only from banks and Portfolio Application may not have been operational long enough to house historic Management Processing Behavioural Scoring Bureau Scores information and build the diversity of information sources Monitoring & ID Verification required for value-added products. In these circumstances, Evaluation Fraud Detection it may be difficult, or indeed impossible, to build some of the more sophisticated solutions, such as credit scoring. Source: IFC 2012. 76 Credit Reporting Knowledge Guide Planning for the development of VAS requires an telecommunications payments may be indicative of future understanding of the stages required for a credit bureau to defaults on bank credits. “mature.” Adding new bureau members also has implications in terms Stage 1: Initial deployment. At inception, a new credit of reciprocity, namely access to the information on the basis bureau must work to build its database of records. In some of their level of data contribution. The rules of reciprocity instances, no data may be available and the bureau must extend to the design and delivery of value-added products. start from scratch. In other instances, the regulator can step A bureau score that incorporates positive credit history in and mandate that all regulated entities collect consent information, for example, should not be made available to from their borrowers to share historical and new credit data a member that provides only negative information, even if with the bureau. This process, which should occur prior the member never actually sees the positive data. to the bureau development process, enables the bureau to populate its database with historical records. Stage 5: Database maturity. Credit bureau databases change over time as the availability of data sources and Stage 2: User acquisition. Although not necessarily the the number and type of users change. Databases tend to case in all countries, the trend in many emerging markets is grow in both depth and breadth, but not always. Privacy for the initial development of credit bureaus to take place restrictions can result in changes to the availability of certain within the banking community. The main driver of this types of information as was seen in the United Kingdom in approach is that the banks are the major providers of credit 2000 when restrictions were placed on using electoral roll and have one clearly defined supervisory entity. The first information. step is to upload the data from the initial members, that is, the lenders. In general, the core bureau database needs a period of time to mature through the above stages of development in order Stage 3: Data diversification. In parallel with Stage 2, the for its data to be predictive of a future outcome (see Section bureau attempts to augment the basic credit history data 6.3.1). The ever-changing nature of the database explains with other forms of information that may be beneficial why value-added products and services require continuous to users, such as electoral rolls, identity records, court monitoring and fine tuning. Estimates based on today’s judgments, telephone numbers, and company registration data may not apply 12 months from now as the overall records. This type of data can be particularly useful to economic environment may change. members: it may be predictive of future borrower behavior or it may make their processes simpler by providing a portal Stage 6: Service expansion. There are no rules as to when to a “one-stop data shop.” The data also provide a valuable VAS can be introduced. Simple services, such as expanded source of information for data mining and modeling. credit reports, can be introduced at low cost at a relatively early stage, even during stages 2 and 3. Bureaus typically Stage 4: User diversification. Even if banks take a pro-active develop more sophisticated products, such as credit scoring, role in establishing the credit bureau, it is often clear from which are usually more expensive to build and maintain, the outset that, at some point, the user base should expand when the database and to some extent the user base have to include nonbank creditors, such as telecommunications reached a level of maturity where the resulting products will companies and microfinance lenders. The introduction of be both robust and have a reasonable shelf life. This level is new users can have a profound effect on the composition of most likely to occur once the bureau has reached stages 3 or the bureau databases and, therefore, the predictive nature 4. It is only when the bureau has reached stage 5, however, of the data. For example, in several countries, expanding that a broad suite of products, as described in Figure 6.1, to include telecommunications providers has improved the can be contemplated. predictive power of the inquiry database as the pattern of 77 Credit Reporting Knowledge Guide Two other key factors that a bureau would typically take 6.3 Products into consideration when developing VAS are return on investment and users’ capacity to adopt the service. The following list, although not inclusive of all of the value- added products credit bureaus provide, serves as a guide Return on investment: A clear business case must exist to the key services typically available. The accompanying for the development of a VAS. The projected revenue from examples indicate how these products are deployed in certain the sales of the services must cover the investment cost and markets and may not be applicable to all circumstances. produce positive return. The pricing and marketing strategy often includes bundling VAS with the sale of core data. 6.3.1 Bureau Scores The capacity of users to adopt the service: Members will A credit score is a number assigned to a borrower based only demand a service if they have the capacity to use the on his or her ability and capacity to repay debt. This service to improve some element of their own processes. A number falls within a range with a higher score indicating bureau score, for example, adds no value unless the lender a higher probability that a borrower will repay. This score is is able to integrate it into its credit underwriting process computed from available credit history information using a to lower the costs of credit approval. User-side constraints statistical model or mathematical algorithm. Credit scores have a significant bearing, especially in emerging markets, can be used in the loan approval process for simple accept/ on who will use the services and in what quantities. reject rules or for more sophisticated risk-based pricing rules and credit limits. Even in developed markets, the uptake of new bureau products and services is not guaranteed and typically requires “Bureau score” refers to credit scores developed on the a highly pro-active sales and marketing department/staff to basis of the credit bureau data and are different from the promote the product. In emerging markets, the problem credit scores developed on the basis of the data supplied of acceptance is even more pronounced. Except for the by an individual lender. Bureau scores are based on the international banks, many lenders in emerging markets lack information pooled across many creditors as well as public an understanding of the lending methodologies that can be information sources and thus include characteristics implemented using VAS and of the information technology otherwise unavailable to the individual lender, such as infrastructure needed to deploy them. total exposure, number of outstanding loans, and previous defaults within the system. All of these are highly predictive Credit bureaus in emerging markets should not measures of future repayment. Credit bureaus typically underestimate the need for in-house outreach training, build scores using three historical data files that are unique market development, and sales functions. As products to the credit bureau: become more sophisticated and more analytical, bureaus • Defaults on previous credit transactions should also recognize the need to have internal specialist resources to monitor and maintain the products and, • Positive payment behavior (trade line data) perhaps more importantly, communicate the benefits to • Previous searches/inquiries. potential users. In certain circumstances, the models may include other Developing VAS can benefit both the bureaus and their types of data, such as: customers and ultimately may improve access to finance for the broader community. The opportunities, challenges, • Third-party data (e.g., court judgments and and ensuing benefits, however, will vary depending on a bankruptcies) bureau’s individual circumstances and the market. • Demographic data (e.g., applicants’ personal attributes, such as age) 78 Credit Reporting Knowledge Guide • Geo-demographic data, aggregated information at the used for any other type of customized model development. geographic level. However, several unique challenges can complicate the process of building/deploying bureau models, as described Each of these components could potentially add predictive below. power to a bureau score, but care must be taken to ensure that the resulting models do not conflict with a lender’s existing Retrospective data: A key requirement of the analysis is decision-making process. For example, a bureau score that the ability to observe the transition of a credit file from incorporates the customer’s age may be incompatible with the point at which an application was made, through the a lender’s custom scorecard that also includes age. Typically, observation period, to the outcome point. This requires the therefore, a credit bureau may choose to develop a suite bureau be capable of retrospectively reconstructing a credit of models rather than just one model to accommodate file at various points in time. With adequate archiving of as many different customer requirements as possible. the database, reconstruction may not be a significant issue. Examples include: However, changes in customer name, address, ID numbers, and the like can cause tracking problems if not appropriately • Positive bureau score for closed user group members addressed. providing both positive and negative data and typically used as a plug-in or addition to in-house custom scores Thin file: The data files may range from extremely detailed, • Enhanced bureau score incorporating additional as when a data subject has a variety of pre-existing credit customer demographic data and typically used on a facilities with various outcomes, to very thin, as when the stand-alone basis by lenders with no other scoring bureau has no pre-existing information on the applicant. In models cases when a bureau has only a limited amount of data on • Industry-specific bureau scores using data derived borrower performance and outcomes, standard statistical from specific industry sectors, such as banking or multivariate analysis may not apply and other methods telecommunications should be used. • Public domain bureau score using data available in the Scoring model calibration: The bureau builds the credit public domain and, therefore, available to all customers. scores from a broad spectrum of customer histories found in its database. The derived scores are typically calibrated Because different users can use the scores for different for an average portfolio; that is, the distribution of purposes, the credit bureau typically uses a variety of customers across the range of scores reflects what is seen different distribution channels. In its simplest form, the across the whole spectrum of customers at the bureau. credit score can be incorporated into a credit report, usually While probability of default at any given score should with some explanation as to its meaning. Alternatively, remain constant for all users, the cumulative good-to- the bureau may supply the score to the users electronically bad odds will vary from portfolio to portfolio depending so that it can be incorporated into customized scoring on the risk profile of the applicant base. This can have a solutions or automated software applications. A third and profound effect on the way lenders manage their cut-off increasingly popular service is a regular batch service that strategies (the scores at which the lender chooses to accept rescores complete portfolios periodically. The charging or decline applicants). It is highly recommended, therefore, structure for each of these services also varies although most that individual portfolios be retrospectively tested before bureaus charge users on a per-score or per-click basis. the models are implemented. When adequate quantities of reliable information are In emerging markets where either the market is too small or available, bureau scores can be statistically derived, typically the credit bureau is insufficiently mature to have confidence by using some form of multivariate regression analysis. The in the data, the bureau may consider offering models that techniques used to develop the models are similar to those rely more heavily on customer demographic characteristics 79 Credit Reporting Knowledge Guide than on credit performance data. Although less predictive, criteria, and scoring algorithms, including score cut-off these models often provide a useful introduction to criteria. the methodology for lenders with little or no previous • Decision output: An automated application processing experience in credit scoring. system assimilates all of the input data, including any available online information from the credit bureau; 6.3.2 Software Applications applies the rules and scoring models from the decision engine; and presents the operator with a recommended A key advantage of credit scoring is the bureau’s ability to course of action, such as accept, refer, or reject. This establish a quantifiable measure of risk in what is otherwise output is then queued so that the final decision is a highly subjective process. Having a numeric value (a presented to an individual with the appropriate level measure of probability of default) for risk is valuable in of underwriting authority. The degree of complexity of its own right but becomes increasingly powerful when such software solutions varies depending on the technical integrated into automated processes and used to pro- sophistication of the user. Advanced decision systems are actively manage strategy and a lender’s appetite for risk. capable of managing almost all aspects of the decision- making process, including customer segmentation and To help facilitate this process, many credit bureaus in strategy allocation (e.g., terms, limits, and product mature economies have developed a range of software features) and even champion/challenger strategy setting solutions that complement both the raw bureau data to test the lender’s appetite for risk. and the scoring process adopted by sophisticated lenders. These solutions are commonly provided either as software • Behavioral scoring (card management solutions): applications—customized to specific user requirements and For a variety of credit products, such as credit cards, maintained within the client’s own systems environment— charge cards, and overdrafts, the initial decision whether or as bureau solutions, more generic in nature and hosted or not to lend is only the first of many decisions taken at the bureau. The available solutions are many and varied, during the life of the lender-borrower relationship. These but the following represents a summary of the more popular dynamic products require a greater degree of monitoring applications. than term loan products because the exposure to risk increases over time. Additional credit decisions must be Application Processing: A key driver of profitability in mass made on issues, such as limit management, overlimit market lending environments, such as consumer loans and authorizations, and card reissue. credit cards, is the ability to keep the cost of new business Behavioral credit scoring is an adaptation of traditional acquisition to a minimum. Many financial institutions scoring techniques designed to observe and evaluate the have turned to automated application processing systems as payment behavior patterns of borrowers. The output a means of streamlining the credit-granting process. Many score changes to reflect the changing risk profile over examples of such systems exist, but the common design time and can be used either to automate routine decisions incorporates several fundamental features: or provide operators with an immediate assessment of • Electronic data capture: Typically an application current risk. processing system has a series of standardized data A range of powerful software solutions has been capture screens. These screens allow the operator to designed to host card management solutions and capture the information necessary to process the decision provide strategic control over practically all aspects of and, perhaps more importantly, store the customer data customer relationship management. Although complex in a format that can later be used for analysis. and expensive, these systems have become almost an • Rule/scoring engine: The system captures the application integral part of mass market credit management. data electronically, then the software automatically • Model Tracking and Performance Monitoring: applies policy rules, such as minimum required lending An overlooked benefit of introducing credit scoring 80 Credit Reporting Knowledge Guide methodology into the lending process is the ability to another application for credit, the previous lender can be monitor customer risk in an objective and quantifiable informed. manner. Undertaking this analysis requires an in-depth • Debt management: Debt collection is an expensive understanding of the way the models are performing. and time-consuming function and typically requires Several credit bureaus provide score diagnostic tools that specially trained and dedicated personnel. Some lenders, monitor and report on the performance of scorecard therefore, opt to outsource this function, sometimes to characteristics in terms of their continuing ability credit bureaus. These services are usually performed on to discriminate and the way shifts in the applicant a fixed-fee basis or on a performance basis, under which population may create misalignments that would affect the collector gets to keep a proportion of any monies the quality of the decisions. recovered. • Collections Scoring: Collections scoring systems • Debt purchase: Credit bureaus that specialize in help lenders identify and differentiate between clients receivables management may choose to take the ultimate that have a high probability of payment despite late risk and buy distressed or nonperforming accounts from payments, and those that have a high probability of the credit provider. In these circumstances, the bureau nonpayment. Based on these scoring systems, lenders purchases the outstanding balances from the lender at a can apply tailored strategies/collections actions that discount, assumes responsibility for collecting the debt, more accurately reflect the risk of a client, as opposed to and keeps the proceeds once the debt has been collected. relying on traditional strategies such as past due times (for instance, all clients that are 30 days late receive the same call/letter). Lenders stand to benefit because a tailored 6.3.4 Collateral Registries strategy helps reduce delinquencies and losses, provides a more pro-active collection strategy, and enables them For secured loans, a lender needs to establish that the collateral to use their resources more efficiently. used for the loan actually exists and is unencumbered. Developed credit bureaus often attempt to become more than just a source of credit data by providing customers 6.3.3 Collections Services (Receivables with access to associated lending information, such as Management) collateral registries. Bureaus can provide this service either A long and often successful association has existed between by building an automated link to a third-party database or credit bureaus and debt collection companies. In several by building and hosting the service themselves (see section instances, negative information in credit bureaus has 3.4). Whether dealing with fixed assets, such as land and been derived directly from information gathered by debt buildings, or movable assets, such as motor vehicles, these collection companies (as was done by Baycorp in New services typically provide two basic functions: Zealand, Credit Reference Bureau in East Africa, and • Inquiry: This function allows users to ascertain the bone InfoScore in Germany). fide nature of the asset and whether or not there are any encumbrances prior to purchase or acceptance of the Many collections products and services are available, with asset as collateral. the following three among the most common. • Registration of interest: This function allows the lender • Tracing: Tracing products use the credit bureau data to or individual to register a notice of a charge or lien on identify the whereabouts of a customer with whom a the asset. lender has lost contact (“skips”). These products either trawl bureau databases to identify contact information of which the lender may be unaware (e.g., telephone 6.3.5 Marketing Services numbers or a new address) or place a marker on the The use of credit bureau data, especially closed-user-group customer file so that if the customer subsequently makes data, for marketing purposes is often a highly contentious 81 Credit Reporting Knowledge Guide issue. In many countries, including Australia, laws either what a customer may contribute over the lifetime of the prohibit the use of such data or restrict it to specific relationship. applications. In many other countries, especially in emerging markets where lenders are already nervous about sharing Geo-demographic analysis: Geo-demographic modeling credit information, marketing applications are intentionally looks at the relationship between geographical areas, excluded from the definitions of permissible purpose in indicated by zip codes or postal codes, and the types of either the industry code of conduct or the membership individuals/businesses that live/work in a given area. agreement between the bureau and its customers. The technique creates similar customer profiles to those described above but does so using aggregated rather than There are, however, several value-added marketing services individual data. that the bureau can provide that do not necessarily involve the use of credit bureau data. The range of potential List services: In countries that have a mature direct products/services that can be offered is extensive. The most marketing industry, many credit bureaus have developed common examples are described below. products and services to assist with customer prospecting. These services range from providing prospect lists (e.g., Customer profiling: Historically, many financial the names and contact details of potential customers) organizations have suffered from poor knowledge augmented with credit bureau data or geo-demographic management systems (e.g., paper-based customer records). data, to the outsourced management of a client’s customer Consequently, these organizations have relied heavily on relationship management database. branch distribution channels to obtain comprehensive information about their customers. Customer profiling Mail screening: Again, in countries that use direct mail attempts to bridge this knowledge gap by providing extensively as a means of acquiring customers, the credit analytical services that profile the attributes of particular bureau can be useful in helping ensure efficient targeting of types of customers. This service may include augmenting potential customers. Mail screening removes from a mailing the lender’s existing customer information with additional list those applicants who are most likely to be rejected for data from the credit bureau. The subsequent analysis an offer of credit if they were to apply. This screening saves identifies homogeneous customer clusters or segments the lender time and effort. This service also has positive that have similar profiles—such as young, credit-active customer benefits in countries that operate a do-not-mail high achievers—that can then be used to help the financial database—a screening facility for consumers who prefer not institution either provide a more tailored relationship or to receive unsolicited marketing offers. better target promotions to cross-sell and up-sell. Where marketing services are permissible (e.g., the United Modeling: As with credit scoring, the number of applications States and United Kingdom) and are used extensively, they for modeling services is extensive. Among the more popular have proven to be a highly lucrative form of added value for are propensity modeling and response modeling. Propensity the credit bureau and a significant value-added proposition modeling tries to predict the likelihood that a particular for the user. These services also have a positive effect on the prospect will take up a marketing offer; response modeling risk management process of the bank by allowing the bank measures the effectiveness of particular marketing campaign to prescreen the offers. to increase the responsiveness of customers in the future and thereby optimize the cost of new business acquisition. 6.3.6 Portfolio Monitoring More complex forms of modeling include applications such as customer worth or customer life-time value. These Monitoring and maintaining credit quality is a task that techniques analyze customer potential not only in terms all lenders undertake but one that has taken on more of actual, current contribution/profit but also in terms of prominence in recent years with the introduction of Basel 82 Credit Reporting Knowledge Guide III reforms. Some credit bureaus have been providing 6.3.7 Fraud Detection services in this field for many years, using a range of As an economy’s retail credit market grows, so will the standard reporting and bureau scoring products. incidence of fraudulent financial transactions. Fraudulent activity can range in severity from what is sometimes referred Portfolio monitoring services: These services advise a to as soft fraud—embellishing application information lender of any significant change to a customer’s credit file, to obtain credit—to hard forms of fraud, such as identity such as a default registered by another lender. theft. A variety of products and services can be developed on the back of the bureau platform to help lenders identify Batch screening: This service allows lenders to periodically and prevent fraud. These products include, but are not update the risk profile of entire portfolios by reviewing the limited to, the following: current credit scores of its clients. • File cross-referencing: These relatively simple products Monitoring and reporting: These services typically help cross-reference various data files to identify anomalies. smaller lenders with limited internal analytical capacity to • Known/suspect fraud closed user groups: These produce the management information required to track industry initiatives, such as the Credit Industry Fraud credit quality. Avoidance Scheme in the United Kingdom, pool information about known or suspected fraudulent Implementing the Basel Committee on Banking activity. Supervision’s120 Basel II,121 advanced internal ratings-based approach requires all lenders to be capable of calculating not • Fraud scoring: This product category includes custom only “probability of default” but also “loss given default and built models for individual institutions or generic models exposure at default.” With the recent financial crisis and the developed by the credit bureau. introduction of Basel III122 reforms, the need for lenders to • Fraud detection systems: These sophisticated software comply with the best practice risk-management guidelines solutions use a combination of rules logic, scoring, have created an increased focus on the ability of lenders to and enhanced databases to identify application fraud. monitor portfolio quality. Credit bureaus with developed A range of software solutions has also been developed analytical capabilities have seized this opportunity to use specifically to track card fraud by means of payment advanced modeling, software solutions, and consultancy to behavior analysis. help their clients with these compliance issues. 120 The Basel Committee on Banking Supervision consists of senior representatives of bank supervisory authorities and central banks from Argen- tina, Australia, Belgium, Brazil, Canada, China, France, Germany, Hong Kong SAR, India, Indonesia, Italy, Japan, Korea, Luxembourg, Mexico, the Netherlands, Russia, Saudi Arabia, Singapore, South Africa, Spain, Sweden, Switzerland, Turkey, the United Kingdom, and the United States. It provides a forum for regular cooperation on banking supervisory matters, and usually meets at the Bank for International Settlements (BIS) in Basel, Switzerland, where its permanent secretariat is located. 121 Basel II: International Convergence of Capital Measurement and Capital Standards: A Revised Framework- Comprehensive Version sets out standard guidance for banks in measuring capital adequacy and the minimum standard to be achieved. The framework was developed by the Basel Committee for Banking Supervision with the objective of being adopted in their respective countries. This framework and the standard it contains were endorsed by the Central Bank Governors and Heads of Banking Supervision of the Group of Ten countries. Revisions to the framework were made in 2009. Relevant publications on Basel II can be found http://www.bis.org/publ/bcbs128.htm, http://www.bis.org/publ/ bcbs157.pdf, http://www.bis.org/publ/bcbs158.pdf, and http://www.bis.org/publ/bcbs159.pdf 122 Basel III: A Global Regulatory Framework for more Resilient Banks and Banking Systems and Basel III: International Framework for Liquidity Risk Measurement, Standards and Monitoring together form the Basel III reforms proposed by the Basel Committee on Banking Supervision. The reforms aim to strengthen global capital and liquidity rules with the goal of promoting a more resilient banking sector. The objective of the reforms is to improve the banking sector’s ability to absorb shocks arising from financial and economic stress, whatever the source, thus reducing the risk of spillover from the financial sector to the real economy. More details can be found at http://www.bis.org/publ/bcbs189.pdf. and http:// www.bis.org/publ/bcbs188.htm. 83 Credit Reporting Knowledge Guide 6.4 The Use of Credit Information includes information on consumers and firms, account history information, and payment history information. Data for Prudential To be effective, however, credit registries and bureaus must Supervision123 contain accurate, complete, and up-to-date records and supervisors must be able to access credit information data Because banks and other financial institutions are highly from a comprehensive range of data sources. leveraged, several international guidelines have been set to control the systemic risks that these institutions pose to the Supervisors can use the information contained in credit economy. These standards are captured by the Basel reforms. registries complemented with data from bureaus to monitor the credit risk undertaken by an individual institution, by The 2007–2008 financial crisis has shown that both, the a peer group of institutions, or by the financial system market on one hand, and supervisors (financial institutions as whole. The information contained in registries allows supervisors) on the other hand, were poorly equipped to supervisors to assess the quality of credit assets and get a deal with systemic risk issues stemming from widespread holistic picture of the concentration of risk exposures and concentrated exposure to credit risks in the financial (e.g., by sector, geographic distribution, type of borrower, markets. Supervisory authorities did not have access to or type of credit). Thus supervisors can assess whether broad, timely, and reliable information, especially about off- financial institutions meet capital adequacy requirements balance-sheet exposures that tend not to be regulated, and as stipulated by their country’s relevant legislation or the also were not adequately prepared to deal with complex and Basel framework, which in turn is an indicator of the innovative financial instruments (e.g., derivatives, options, systemic risk level in the economy. Systemic risk levels rise asset-backed securities) to assess all the risks assumed by when a large number of financial institutions are exposed financial market players. The tools that supervisors used to the same risks. Supervisors can keep track over time of in conducting on-site inspections and off-site monitoring the losses incurred in every single credit, compare the level of regulated institutions—econometric models, stress of risk and credit classification for a particular borrower testing, accounting criteria—were dated and unable to pre- across the financial system, compare levels of provisions emptively identify the potential risks assumed by the system and, consequently, capital allocation according to the risk as a whole and make recommendations for appropriate level.124 preventive action. Supervisors typically use off-site surveillance and on-site Credit registries play an important role in supporting the inspections to monitor the overall health of the financial prudential supervision and risk monitoring function of institutions they supervise. Data in credit registries and supervisory bodies. Data from credit registries that would bureaus can serve as important inputs into the various tools be useful for prudential supervision, include but is not be that supervisors use in these inspections and surveillance. limited to: borrower type and identification data, credit On-site supervisions can be costly and time consuming. information data, current risk classifications and collateral Moreover, supervisors are unlikely to be able to analyze and guarantee information, all of which would enable every credit record in the portfolio of the financial supervisors to model the probability of default of different institution that it is inspecting. Credit registry or bureau borrowers and calculate and monitor potential loss given data can provide useful “sample data” that highlight defaults of the various creditors. Credit bureaus have also key trends and characteristics in a financial institution’s started complementing the information collected through portfolio, including changes in portfolio quality due to the registries for prudential supervision purposes. Such data introduction of new financial products. Supervisors can 123 Whereas the Section 6.2 on value-added services speaks specifically to products developed by bureaus, this section on prudential supervision relates directly to a function performed by regulators using credit registry databases, occasionally complemented with data provided by credit bureaus. 124 Girault et al., 2010. 84 Credit Reporting Knowledge Guide use the information from the sample data to determine allocation of supervisory resources. Early warning systems what areas of an institution’s portfolio needs closer review can prompt early action on the part of the supervisory and thus allocate their time and resources more effectively. bodies with minimal disruptions to the financial markets. Samples obtained from credit reporting service providers can also flag discrepancies in the financial institution’s Stress testing: Supervisors use stress testing models to risk classifications and ratings of borrowers and whether understand the impact of different economic shocks on adequate loan-loss provisions have been made. financial market players. Based on the various scenarios developed and the results of the stress testing, supervisors Whereas frequent on-site inspections cannot feasibly be can recommend adequate capital levels to absorb losses conducted, off-site surveillance tools can support supervisors associated with large, and often unpredictable, shocks. in continuous supervision and monitoring. Once again, For instance, supervisors can stress test the impact of a data from credit reporting service providers can provide a downgrade of one, two, or more levels of borrower risk valuable input into some of the tools that supervisors use in classification in a portfolio and compare the effects of this conducting off-site surveillance. These tools are: downgrade given by a financial institution or the system as a whole. The results would demonstrate potential impact Indicators: Supervisors can use the data obtained from on capital requirements and profitability due to additional service providers to create regular reports containing provisioning requirements. Supervisors can also stress test different indicators that summarize the exposure to credit the actual level of provisioning against different economic risk of different financial institutions. Some such indicators conditions, as well as the consistency and robustness of are: concentration expressed as a percentage of total risk rating systems and credit classification models used by exposure, concentration expressed as a percentage and financial institutions over a period of time. origin of funds, exposure by economic sector, volume of nonperforming loans, credit classification, level and Transition matrices: Another tool used by supervisors is evolution of credit provisioning, growth of credit portfolio, the transition matrix. Banks and other creditors generally growth by credit lines, historical loss for each line of credit develop their own internal borrower ratings systems, which (eventual adjustment of regulation and capital requirement), classify borrowers by their risk profiles. Supervisors are both at an individual level and at an institutional and system increasingly developing such rating systems to validate the level. The indicators can help supervisors verify whether the systems developed by the financial institutions that they financial institutions are in compliance with prudential regulate. Transition matrices track movement of borrower regulation for borrower risk classification and also indicate ratings, based on individual credit operations, from one level the level of interlinkages among different financial to another (upgrade or downgrade) over different periods of institutions (which raises the level of systemic risk). These time such as three months, six months, one year, or five indicators can also provide a framework for comparison of years. Data from credit reporting service providers can borrower ratings across different financial institutions in an provide valuable inputs into transition matrices. Supervisors economy and flag outliers or aberrations to the authorities. can analyze differences in ratings across different time It may also provide valuable confirmation that regulated periods, geographical areas, economic sectors, volatility in entities are complying with any mandatory requirements ratings, average default rates for borrowers subgrouped by to submit data to credit reporting service providers, and similarities in credit type, financial institution type, or other consulting this data before extending credit. factors. Over a period of time, series of observations of behaviors across a transition matrix can provide supervisors Early warning systems: The indicators developed using with greater insights into the probabilities of default and data from credit reporting service providers can be modeled the level of risk in the system. into early warning system models that enable supervisors to focus on vulnerabilities and critical levels of exposures Although the possibilities of using credit reporting data to in the market. This modeling enables them to focus their support the prudential supervision function are limitless, surveillance and inspection efforts and thus optimize the 85 Credit Reporting Knowledge Guide challenges remain. Whereas supervisors oversee only to the system as a whole from this interconnectivity of regulated financial institutions, financial markets comprise regulated and unregulated lenders. Credit registries should other types of creditors that are unregulated and yet may aim to collect data from a broad range of financial market be interconnected with the formal banking system as participants to ensure that significant exposures across major customers of the banking sector or as having the the financial system are adequately captured. Since credit same exposures as the banking sector. Using credit registry bureaus generally collect a wider range of information, data that only provides information on regulated lenders incorporating data from credit bureaus can complement limits the ability of the supervisor to assess the risks posed the data from credit registries. 86 CHAPTER 7 Case Studies T his chapter examines five case studies that provide Egypt, the regulator played a strong role by encouraging examples of setting up credit reporting systems. the development of a legal and regulatory framework that The examples of Ecuador, Egypt, Morocco, and mandates information sharing with a privately operated Cambodia illustrate the pro-active support of a strong credit bureau that is licensed and supervised by the regulator. regulator, with a clear vision of defining the most optimal The regulator maintains an internal database of information credit reporting solution for the country. The case study from regulated lenders for prudential supervision and risk- of India presents a more complex political environment, monitoring purposes, but is not involved in the activities with a divergent set of views and priorities expressed by the of the private credit bureau. Under this model, the role of involved stakeholders. Each of the case studies highlights the regulator in the actual process of credit information challenges faced in collecting data and operationalizing generation is limited, and instead the regulator can focus the credit bureau,125 and analyzes how these challenges its efforts on areas core to its functions, such as supervision, were overcome. The IFC was directly involved in the licensing, and regulation. The case study on Egypt discusses establishment of the credit reporting systems in four of the new developments with the bureau and efforts to improve five countries examined; the study on Ecuador was based on coverage by including other sectors such as microfinance. IFC discussions with local stakeholders. In Morocco, where the regulator also played an important In Ecuador, the regulator sanctioned the development of role in establishing the credit reporting system, the regulator private credit bureaus following a banking crisis and, in opted to act as interface between the regulated lenders and a unique approach to ensuring uptake of credit reporting private credit bureau, similar to what was done in Ecuador. requirements mandated by the law, decided to act as an In addition, the regulator also licenses and supervises interface between the regulated lenders and the privately the credit bureau. The authority of the regulator and the licensed credit bureaus. The flexibility of this private-sector trust and reputation it had built for itself in the lending solution provided an opportunity for nonregulated lenders, community were critical in the development of the system. including microfinance institutions, to participate in the The Morocco model is effective in an environment where credit reporting system, thus enhancing the overall coverage lenders are hesitant to share data, and the involvement of of the credit reporting system. the regulator helps mitigate the risk of lenders refusing to share data. However, this approach works best when there is As discussed in Chapter 4, legislation and regulations are more than one licensed bureau to ensure that users receive increasingly mandating lenders to share data with credit suitable products and services at competitive prices. bureaus. Once mandatory sharing is implemented, either by legislation or through a central bank regulation (as In Cambodia, the regulator once again took a pro-active in Morocco), regulators need to decide what role they approach to supporting the development of all facets of the wish to play in the credit information-sharing system. In credit reporting system from development of the legal and 125 See Chapters 1 and 5. 87 Credit Reporting Knowledge Guide regulatory framework to supporting the development of a 7.1 Ecuador: Supportive Regulator private bureau. In this case, stakeholders benefited from a and Pro-Active MFI Network prior experience of working with a regulator-run registry that had failed to meet the needs of the regulated lenders Facilitate Credit Information and the broader lending community. This experience Sharing between MFIs and a created a strong consensus among lenders to support a Private Credit Bureau private-sector solution serving all lenders. Overall, the project benefitted hugely from strong support not only of In the late 1990s, Ecuador suffered a banking crisis that the regulator but also of the lending community, including resulted in a huge drop in GDP, widespread unemployment, banks and microlenders. and harsh social conditions as reflected in increased poverty across the country. To fight growing poverty, the government The India case study demonstrates IFC’s efforts to promoted measures and tools to enable easy and broad access integrate microfinance institutions into the formal credit to finance by the lower-income population, primarily living reporting system in a volatile political environment with a in rural areas. One tool was enabling regulations introduced multitude of stakeholders holding different viewpoints. A by the Superintendencia de Bancos y Seguros (SBS) in 2003 key takeaway was that systematic outreach and awareness that permitted the establishment of private credit bureaus, raising throughout the process is critical to ensuring project and provided the SBS a regulatory and oversight capacity success. Although lenders understood (in varying degrees) over these bureaus. Until then, the SBS had operated a credit the importance of credit reporting as a risk-management registry, which was the only provider of credit reporting in tool in identifying high risk individuals with little to no the country. The credit registry collected information only credit experience, potentially larger benefits of credit from regulated lenders and there was no provision for credit reporting were not perceived: namely, the ability of the reporting for microlenders, which were largely unregulated. credit reporting service provider to help identify good borrowers (who typically outnumber bad borrowers by a Following introduction of the regulation, the SBS licensed factor of 10 to 1). These good borrowers are individuals six private credit bureaus. The public credit registry of the and small businesses that have demonstrated they can meet SBS was closed to lender inquiries. Instead, the SBS handed financial commitments and can build a track record that over all the data collected from regulated entities through will enable them to obtain further funding to grow their the registry to each of the licensed credit bureaus. Each businesses at lower costs in future. A key argument for the bureau was then free to complement the data with other participation of MFIs in credit bureaus should be focused sources of information, develop and negotiate agreements on these positive benefits rather than whether they will fix with different data providers and users, and compete with an overindebtedness problem that is more likely to result each other on the basis of differentiated product and value- from irresponsible lending than from information opacity. added service offerings. The lesson learned is the need to keep a balanced view and promote the holistic benefits that such participation Bureaus that sought to include coverage of the rural would deliver and an environment that encourages such population (in line with the government’s financial involvement. inclusion agenda) were initially faced with difficulties. Historically, these segments had been excluded from the Each case study points out the challenges faced in collecting traditional banking sector (26 institutions), thus little to data (as discussed in Chapter 1) and in operationalizing a no data existed in terms of payment histories. MFIs, both new credit bureau, as well as how these challenges were regulated and nonregulated, had stepped in to fill this overcome. lending gap and were serving an estimated market of over 1.7 million customers. Because the majority of these MFIs were small nonregulated entities focused on the rural areas, 88 Credit Reporting Knowledge Guide they were not mandated to share information with the Nonregulated MFIs had an avenue for credit information credit registry, nor did they voluntarily share information sharing through the private credit bureau, Credit Report, among themselves. which provided tailored services and differentiated pricing to this segment of lenders. Recognizing the need for credit information sharing among microlenders, an association of MFIs, Red Financiera Rural The results of project SERVIR were positive, enabling the (RFR), launched a pilot project, SERVIR, in 2005 to share project to expand to include MFIs in the rest of the country. information between the nonregulated MFIs and one of According to statistics from SBS, from December 2004 the licensed credit bureaus. Initially the pilot was limited to June 2006, MFIs’ portfolio volumes grew 53 percent. to two provinces in the central part of the country. These Credit default rates (1 day) decreased from 41 percent to provinces were characterized by a large number of MFIs 10 percent, the number of clients increased by 33 percent, (at least 250), high rates of poverty, a predominantly rural while the average loan amount increased from $1,800 to (four-fifths), indigenous population, and many small-scale $2,400. producers of goods and services. Ten microlenders (mainly NGOs, cooperatives, and credit unions) initially agreed to The legal and regulatory framework in Ecuador provided share and exchange data with the bureau. The intention for key roles for the public central credit information was to monitor the results and ascertain the benefits, with entity and private credit bureaus. Today, Ecuador has the goal of eventually extending coverage to MFIs in other a fairly advanced, full-file and nonfragmented credit provinces. reporting system. Of the country’s 500 NGOs, over 330 share data with the credit bureau. RFR continues to work RFR played an important role in facilitating collaboration on promoting credit reporting among microlenders not between the MFIs and the bureau. It helped the MFIs already a part of the bureau and on negotiating prices, while select Credit Report as the best bureau to partner with providing basic training to associates on using credit reports based on the strength of its technology platform, financial and scoring systems. In addition, RFR uses credit bureau soundness, ownership by the international credit reporting data to supply financial analyses in standardized format to firm Equifax (representing experience), and price per report its members on a monthly basis. guaranteed to RFR members.126 RFR employees worked with the bureau and the MFIs to undertake diagnostics of Despite the strength of the framework, in 2011, after years the microlenders’ lending processes, data, and technological of intense competition and bare bones pricing (reports cost systems, and then helped define the structures for credit less than $0.10 each), only Credit Report remained of the information data collection including minimum data six initial credit bureaus. The unique access it had to data inputs, data quality rules, data collection, validation and on borrowers at the base of the pyramid through RFR’s processing rules, method of data delivery, and the types of members was a key factor contributing to its success, giving products and services that would be affordable and useful. it superior market coverage compared with its competitors. For RFR and the MFIs it serves, working with a privately The supportive role of the regulator must be noted here. owned credit bureau provided access to data from other parts When the SERVIR project was launched, Ecuador displayed of the credit market; access to related tools, such as credit a strong legal and regulatory environment that supported scoring; and access to quality data—all at a very attractive credit information sharing and protected borrower rights. price. The competitiveness of the market is questionable Regulated MFIs were obliged to contribute data to the however as Credit Report is the only remaining bureau in SBS, which supplied all licensed bureaus with the same the market. information base. Lenders were mandated to inquire with at least one credit bureau before providing credit. 126 Lyman et al., 2011. 89 Credit Reporting Knowledge Guide 7.2 Egypt: Strong Regulator a competitive selection process and with advisory support from IFC, contracted Dun & Bradstreet South Asia Middle Supports Establishment of East Ltd., as a technical partner to provide software solutions Country’s First Private and operational know-how in the creation of its database. Credit Bureau By working closely with the Central Bank of Egypt, IFC, I-Score, formerly Estealam, is the first credit bureau in Egypt. the financial community, and an experienced technical It was established in 2005 and became fully operational in partner, I-Score successfully established a transparent and 2008. The founders and shareholders consist of 25 banks advanced credit reporting service that offers services in both and the Social Fund for Development. The role of I-Score is Arabic and English. Since its inception, I-Score’s data center to provide Egyptian banks and other creditors with accurate, has vastly expanded to include 8.8 million data records— factual information relevant to the history and payment an almost tenfold increase from its baseline of 0.9 million habits of their clients or prospective clients, thus enabling borrower records initially held by the Central Bank of them to better assess their clients’ creditworthiness. I-Score Egypt’s credit registry. I-Score currently services the credit also educates the public on the benefit of having good credit information needs of 55 institutional subscribers, which histories, and the consequences of negative credit events. include 41 banks, 8 mortgage finance companies, 4 leasing I-Score’s financial literacy role has been a major contributor companies, the Egyptian Social Fund for Development, and in changing and modifying the financial behavior and 1 retailer. All banking institutions and the Egyptian Social culture among borrowers. Fund for Development have completed the credit data migration process to I-Score. In addition, several mortgage The Central Bank of Egypt was highly instrumental in finance and leasing companies have submitted their data creating a legislative framework conducive to the operations to I-Score. Currently, I-Score is offering a range of services of a bureau. In November 2005, I-Score (then Estealam) to its customers including credit reporting, bureau scoring, requested IFC’s technical advisory support to develop the account monitoring, and data analytics. bureau. In addition to assisting with developing the I-Score business plan, IFC provided technical support throughout As of December 2011, I-Score has issued over 8 million I-Score’s implementation process, which commenced in credit reports, signed up 55 subscribers and improved on November 2005 and culminated in I-Score’s launch in Egypt’s ranking in the Doing Business indicator “getting 2008. credit.” The total number of records in the bureau (consumers and firms) rose from 3.9 million at the end The first step in the process involved extensive consultations 2008 to 8.7 million as of December 2011, with over 23 between stakeholders and the regulatory authority to million credit facilities. Consumer loan portfolios grew amend existing laws and introduce new rules that would from $14.6 billion in June 2008 to over $17.6 billion at simplify information sharing. Legislative amendments were the end of 2011. Loan loss provisions had declined from introduced to allow the exchange of information among 15.3 percent in March 2009 to 10.4 percent in March 2011 banks and mortgage finance and financial leasing companies while percent of nonperforming loans had declined from and to allow the sharing of such information with private 13.4 percent in March 2009 to 11 percent in September credit bureaus and the central bank supervisory system 2011. According to Doing Business 2012, Egypt advanced without obtaining borrower consent. The new legislation 78 places to rank 78 in 2012, up from 156 in 2007. The also mandates all users, (i.e. subscribers of I-Score) to bureau’s coverage increased from 0 percent in 2007, to obtain credit reports and utilize the services provided by 13.7 percent. Egypt’s score on the Doing Business 2012 CII the private credit bureau prior to making a decision on a increased from 2 out of 6 in 2007 to 6 out of 6 in 2012.127 credit application. In September 2006, I-Score, through 127 World Bank 2011b. 90 Credit Reporting Knowledge Guide Recently, Egyptian MFIs have decided to join I-Score as • In addition, roughly 13,000 MFI clients had been members and data providers. In the beginning of 2011, after granted credit lines (one, two, or more) by other MFIs, nearly three years of operations, I-Score was still trying to with an outstanding balance of LE 14 million. add MFI data to its database, hitherto fed by contributions from banks and mortgage and leasing companies. Mistrust The incidence of multiple lending as demonstrated by the of the intentions of the bureau and traditional lenders cross-tabulation analysis sufficed to dissuade the MFIs from had caused MFIs to develop plans for building a separate keeping their data separate from other lending sectors. “microfinance credit bureau,” which would have led to Vertical information silos would not provide the necessary fragmentation of information in the market. quality and completeness of information needed to conduct responsible and reliable lending. At the same time, the Egyptian microfinance market was beginning to face a crisis of repayments in the microfinance Following the cross-tabulation analysis, in early 2012, IFC sector, similar to Morocco’s experience in 2008–2009.128 entered into an agreement with the Egyptian Microfinance Rising portfolio volumes were accompanied by a Network, an umbrella organization of the main MFIs, to corresponding increase in portfolio at-risk ratios (portfolios support the integration of four of the largest MFIs into the more than 30 days overdue). Discussions were underway bureau over a year. The objective was to ensure a smoothly to strengthen microlending practices in Egypt and spare it functioning system of data submission and uptake of the from a similar crisis. bureau by MFIs, as well as to help MFIs obtain the capacity to make better credit decisions using credit reports, thus Extraordinary situations require unusual solutions. In 2011, leading to healthier credit portfolios. I-Score and IFC proposed a cross-tabulation analysis to three of the largest Egyptian MFIs. The aim was to compare I-Score has been flexible in supporting the needs of a sample of their loan portfolios with the data in the I-Score microfinance institutions: special prices have been agreed database to see if there was evidence of cross-lending upon for MFI lenders, technical support is being offered, (borrowers being granted lines of credit by other lending a free trial period for newcomers is granted, and the sectors, such as banks). The results were both unexpected development of ad-hoc services is part of the package and alarming for the MFIs: tailored for Egyptian MFIs. • Over 14 percent (nearly 50,000) of the MFIs’ clients in the sample had been granted credit lines (one, two, or 7.3 Morocco: Establishing a more) by the banks, corresponding to an outstanding balance of nearly LE 500 million (thrice as much as the Public-Private Partnership for MFIs’ total outstanding balances). the Credit Reporting System • 6,000 of these MFI/bank customers were 90 plus days In 2005, the Central Bank of Morocco (BAM) in partnership past due with IFC’s Global Credit Reporting Program and the IFC • 460 of them were undergoing legal actions regional office for the Middle East and North Africa129 embarked on a project to create a national credit reporting • 100 (approximately) showed dishonored checks records 128 For more details on the microfinance crisis, please refer to Reille 2010. 129 Recognizing the critical role that credit bureaus and registries play in helping financial institutions assess and manage credit risk, IFC’s Private Enterprise Partnership (PEP)-MENA launched the MENA Credit Bureau Program in July 2005, building on the experience of the IFC Global Credit Bureau Program. The program aims to promote the creation and use of credit information and reporting services throughout MENA, thus helping retail consumers and MSMEs create “reputational collateral.” Credit history information is essential for financial institutions to be able to increase financing to these segments, particularly MSMEs, a priority market segment for IFC. The program’s clients include central banks, banking associations, and private credit bureaus with global and regional expertise in the areas of legislative reform, business planning, partner identification, and new product development. 91 Credit Reporting Knowledge Guide infrastructure that would support the development of the registry would contain serious information deficiencies if it first private credit bureau in Morocco. An initial diagnostic were used to store and retrieve a high volume of information. at the outset of the project showed that the Moroccan In its diagnostic report, IFC recommended to BAM that market suffered from fragmentation of credit-information- the design, construction, and operation of the database(s) sharing initiatives and the reluctance of key lenders to share be undertaken by the private sector. In particular, IFC information. BAM had a credit registry that maintained recommended the approach followed in Ecuador, whereby records on banks’ lending operations. As the credit market BAM could maintain a leadership role among financial grew, the registry struggled to meet the market demand. To institutions. improve information availability, BAM was preparing for a major upgrade of its credit registry. IFC recommended that BAM develop the technical infrastructure of its credit registry to receive all types of According to the banking law of 1993, BAM had the right borrowers’ information (positive and negative) from the to centralize credit information from regulated entities. universe of lenders. The registry would then consolidate Just before the project started, and after IFC presented this information and make it available to any private its preliminary diagnostic, the banking law was amended credit bureau licensed by BAM. Once the first private to give BAM the right to decide whether to handle credit credit bureau was fully operational, lenders regulated by reporting in house or delegate it to the private sector. BAM (banks, microfinance institutions, and nonbanking financial institutions) would no longer be allowed to access In 2005, at BAM’s request, IFC experts conducted a the registry, (but would continue to provide information) technical and market diagnostic of a select sample of and would be required to consult at least one credit bureau financial institutions’ retail and SME lending portfolios prior to making any credit decision. Initially, only regulated to assess their capability to provide data as well as to entities would be able to consult the bureau, but eventually receive and integrate credit bureau information into their nonregulated entities would be able to provide data directly credit underwriting and portfolio management processes. to any private credit bureau based on consumer consent. Characteristics of the credit market in Morocco included Under principles of reciprocity, this would allow them to a high incidence of collateral in lending, a high rate of consult the bureau as well. Licensed credit bureaus would nonperforming loans, and a lack of sufficient physical or all receive the same set of information from BAM and then useable collateral. In the absence of collateral and any other compete on the basis of differentiated product and service reliable information about potential borrowers, lenders offerings and prices. Figure 7.1 illustrates the information were restricted to lending to a small subset of the potential sharing flow of the model. borrower population. Projections suggested that consumer credit would be the primary driver of growth for lenders. IFC also provided recommendations on the legal and This prediction further confirmed the need for a credit regulatory framework for credit reporting. BAM would reporting system to support, nurture, and control this license and regulate credit bureaus, and have access to growth in the Moroccan market. information to assist its supervision of the financial system. IFC recommended that the most suitable way to create an A review of available credit information on banks in BAM’s enabling private credit reporting legal framework would be credit registry revealed (1) limited scope and questionable through circulars issued by the BAM. Hence, BAM issued quality of data, (2) limited information technology circulars outlining the scope and type of information to connectivity, (3) inaccurate identity data, and (4) limited be collected from all regulated lenders, including positive database and reporting design. It was also noted that the and negative information, and the terms and conditions for lenders to access such information.130 The circulars 130 The Banking law in Morocco was modified to allow BAM to become the licensor and supervisor of private credit bureaus. BAM issued several regulations in 2007 (2/G, 27/G, 28/G), to regulate credit reporting including consumers’ rights, mandating sharing and inquiries, clarifying share- holding pattern limits for users, and so forth. 92 Credit Reporting Knowledge Guide Figure 7.1: Information Sharing Flow, Morocco Regulated entities (banks, microfinance institutions, nonbank financial institutions) Customers Nonregulated entities (second phase) Regulated Entities (Banks, MFIs, and NBFIs) Credit Report Central Bank of Morocco Credit Report PCBs Source: IFC 2012. made it mandatory for regulated entities to supply updated an information repository to assist in its supervisory role as information to BAM on a periodic basis as well as to inquire a regulator of the financial system. about a customer before granting credit. In addition they gave the customer the right to view his or her credit report To pre-empt the creation of a monopoly, the legal and and dispute false information, among others. BAM also regulatory framework allows BAM to license more than one issued a code of conduct to cover operational aspects and to bureau. Supplying all private credit bureaus with the same govern the relationship between the lender, the private credit set of information from BAM further reduces the danger bureau, BAM, and the customer.131 With the circulars and of having a monopoly. Licensed private credit bureaus are the code of conduct, a solid legal and regulatory framework then free to compete based on prices and quality of service. was created to govern the private credit bureau industry in Although, BAM has not yet licensed a second bureau, it Morocco. intends to do so. In September 2007, BAM issued the first private credit Experian Morocco became operational in November 2009 bureau license to Experian Morocco, which provides both and regulated lenders started to provide data to BAM positive and negative credit reporting information. By periodically. As of February 2012, the database had collected implementing the model recommended by IFC, Morocco over 3.7 million borrower records (both individuals and has established a transparent, competitive, and advanced firms) from lenders representing over 4.7 million active credit information-sharing infrastructure. After Experian credit facilities from 15 banks, 12 finance societies, and Morocco was established, BAM created a separate credit 4 MFIs. The quality of the information collected by the bureau supervision unit that oversees activities of the database has constantly improved and the credit bureau is credit bureau, and monitors data exchanges between data working on expanding its sources of information. providers, the bureau, and users. BAM has also been building 131 Users and data providers cannot hold more than 30 percent of the credit bureau’s shares as a group, and not more than 5 percent individually. This requirement overcomes conflict of interest situations arising out of majority data provider/user participation in the shareholding structure. 93 Credit Reporting Knowledge Guide Morocco now has a clear vision as to the positive role an organization of 46 of the leading nonbank financial that private credit bureaus can play in improving access companies and MFIs whose combined business constitutes to finance. This vision, coupled with an effective enabling over 80 percent of the Indian microfinance sector,132 on environment, attracted considerable attention from creating credit information-sharing mechanisms for the international credit bureau operators interested in investing microfinance sector. The industry responded to the Andhra in a new credit bureau in Morocco. The model selected Pradesh crisis in a proactive and unified manner by putting for Morocco is potentially a viable solution for many in place some self-regulation and representation via MFIN. countries and has significant advantages for improving data About the same time, three new credit reporting entities133 consistency and establishing a competitive environment for were going through a process of applying for an operating credit bureau operators. license from the Reserve Bank of India. Although no industry consensus was emerging on the best type of credit reporting system, it was becoming clear to stakeholders that 7.4 India: Integration of MFIs into credit reporting in the microfinance sector had become the Credit Reporting System necessary134 to enable microfinance providers to obtain credible information on clients, provide better and more The Indian microfinance sector is potentially the largest in efficient services, improve risk management processes, the world, but also the least well served. In the past few reduce transaction and operational costs, and increase years, microfinance has grown at a tremendous rate in client outreach. IFC’s intervention was timely in bringing India. However, fast growth has led to a number of complex key players together to explore tangible solutions to the issues, many of which culminated in the recent repayment problems facing the industry with the development of a crisis in Andhra Pradesh. The Indian MFI sector was credit information system. impacted by the global financial crisis, which was brought to a crescendo in Andhra Pradesh in 2009. The impact of IFC provided advisory services to its client, MFIN. IFC’s multiple borrowings and financial overindebtedness led to assistance included bringing together existing and newly reported suicides of poor farmers, significantly affecting licensed credit bureaus and MFIN’s members as well as other the industry and leading to excessive intervention by the non-MFIN members, to agree on the practical steps needed Andhra Pradesh government. Key factors leading up to the to share credit information among the MFIs and with the crisis were: concentration and proliferation of MFIs in a few credit bureaus. The project’s first objective was to complete states resulting in multiple lending and overindebtedness a scoping/feasibility study and design an implementation issues and information asymmetry or lack of any credit roadmap for MFIN clients to gauge their readiness and information sharing among MFIs. capacity to contribute borrower and loan information, and also to retrieve credit reports for use in their loan assessment In June 2009, several months prior to the beginning of process. IFC’s role was to enable MFIs to start reporting the Andhra Pradesh crisis, IFC was in discussions with the data to credit bureaus. As the project evolved through newly formed Microfinance Institutions Network (MFIN), regular interactions and workshops with stakeholders, it achieved much more beyond the original objectives. 132 Microfinance Institutions Network (MFIN). See: http://www.mfinindia.org/ 133 India has four licensed credit bureaus including CIBIL, Equifax, High Mark, and Experian. 134 In due course, both the Malegram Committee Report of January 2011 to the Reserve Bank of India (http://rbidocs.rbi.org.in/rdocs/Publi- cationReport/Pdfs/YHMR190111.pdf ) and the subsequent Microfinance Institutions Bill of June 2011. (http://finmin.nic.in/the_ministry/ dept_fin_services/micro_finance_institution_bill_2011.pdf ) came out strongly for the participation of MFIs in credit bureaus. In fact, the new bill introduced in parliament in May 2012 requires microfinance institutions to become members of Credit Information Bureaus. Furthermore, the two MFI associations, MFIN and SaDhan, launched a joint code of conduct (http://www.sa-dhan.net/Inner.aspx?Others/CodeConduct. htm) in December 2011, recommending that their members contribute to at least two credit bureaus and utilize credit reports for identifying overindebtedness. 94 Credit Reporting Knowledge Guide The project was divided into three phases. In the first on a regular frequency, and develop and test sophisticated phase, which lasted six months starting in June 2009, matching algorithms to adapt them to the Indian setting. IFC conducted a scoping study to gauge the readiness of the microfinance sector to participate in a credit bureau, One of the main challenges was the potential market the viability of MFI credit reporting, and the level of fragmentation among the many parties seeking a systemic stakeholder interest and commitment. Simultaneously, data-sharing solution for the sector. There was a need IFC, in collaboration with MFIN, conducted 11 workshops to galvanize the various parties steering initiatives with in cities across the country to raise awareness of credit similar intentions to link the microfinance sector to credit reporting among MFIs. IFC facilitated discussions among reporting systems (e.g., stand-alone MFIs, bureau for key stakeholders and licensed credit bureaus to agree on a nonbank financial corporations, association-run credit common data format that MFIs could use to submit data bureaus, FINO credit bureau, and other possible solutions). to the credit bureaus, as well as a list of requirements for a IFC recognized their efforts, however considered them common credit information report. noncomprehensive because some promoted use of negative- only data, sectorial sharing of data, or closed user groups. In the second six-month phase, IFC assisted and advised IFC’s role was to prevent market fragmentation from MFIN on the practicalities of designing the most optimal a combination of uncoordinated side initiatives and to business model, including how best to integrate the promote comprehensive sharing of credit information on MFIs into a multibureau environment while maintaining all accounts from all MFIs nationally. IFC achieved this competitiveness among bureaus, identification of a software goal by participating in several roundtables, conferences, extraction tool to support MFIs that lacked capacity with and high-profile events with stakeholders to discuss the data extraction tasks, and providing advice on pricing for benefits of having a comprehensive credit reporting system. credit information reports based on global benchmarks. At the time, several ongoing political, legal, and regulatory By May 2011, two credit bureaus, High Mark and Equifax, changes were affecting the sector as the federal government had started providing credit reporting services for MFI and Andhra Pradesh sought to impose restrictions and clients. Approximately two-thirds of the 300 MFIs in India regulations on the sector in the aftermath of the local crisis. were members of at least one bureau and either submitted data regularly or were in the process of setting up their data In the last phase of the project (12–15 months), IFC helped submission. Coverage of the existing MFI borrowers by the all stakeholders with implementation, specifically that bureaus was 80 to 90 percent, as almost all of the major of collecting and submitting data to the bureaus. There nonbank financial corporation MFIs was using a bureau. was significant political pressure to have a working credit As of May 2012, High Mark operated India’s largest rural bureau as soon as possible that was capable of providing and microfinance database of 75 million accounts and 40 credit reports on MFI borrowers. The challenge of obtaining million borrowers. Close to 7.5 million credit inquiries good-quality data from the MFIs for the bureaus to upload were conducted in the 12 months since its launch. was undertaken with zeal by both sides. The MFIs’ loan account data was generally found to be of reasonable quality. The impact of the project is yet to be seen given that two However, the demographic details of each borrower, (e.g., bureaus have just started including MFI data. IFC, in names, addresses, dates of birth, voter ID, ration card) conjunction with MFIN, identified its remaining objectives caused the most problems. Because India does not yet have as bringing on board nonparticipating MFIN members as a national ID for all citizens, the bureaus had to rely on well as non-MFIN members; continuing to address data other characteristics of the borrower’s demographic profile quality issues; and incorporating systematic outreach and to match files. Considerable effort was expended by both awareness-raising efforts to MFIs and MFI borrowers on MFIs and bureaus to collect additional data elements (such the benefits of credit reporting, the importance of on-time as father’s name or spouse’s name) to correct erroneous payments, and other matters. IFC also recognizes the need data, build capacity of MFIs to extract and submit data to leverage the opportunity to research the impact of credit 95 Credit Reporting Knowledge Guide reporting on borrower behavior. Such research would be a utilities, and telecommunications companies. Based on critical part of long-term outreach and awareness-raising recommendations provided by IFC, the NBC decided efforts in continuing to integrate the sector into the credit to pursue the option of setting up a private sector credit reporting system. bureau, while continuing to play a key role in the entire process. 7.5 Cambodia: Developing a IFC worked primarily with two Cambodian associations Private Credit Bureau with (the banking association and the microfinance association) Strong Government and to establish a private credit bureau. This process entailed identifying the optimal business model and related business Stakeholder Support rules (expressed as a memorandum of association) including equity participation and subsequent payment, selection In 2006, the National Bank of Cambodia (NBC) of directors, and dividend payments through numerous established a pilot credit information sharing system within stakeholder discussions and with local legal support. the national bank. The system, had some structural and functional limitations, but provided valuable information Banks and MFIs collaborated to establish a working group to and experience to the NBC. A pilot of the system, in which plan for credit bureau development and to select a technical voluntary participation was limited to banks, confirmed vendor and a strategic partner. IFC supported stakeholders that there was a latent demand for credit information and the NBC in designing a transparent vendor selection sharing that extended beyond the banking sector to the process that reflected the needs of the market. MFI sector and the wider lending community. The NBC decided to undertake efforts to expand credit information Once the strategic partner was selected, the lending sharing to the broader market and enlisted the support of community established an implementation committee IFC and the World Bank. with key representatives nominated from local banks, microfinance institutions, and the NBC to determine IFC assisted NBC in developing a robust regulatory business rules and specifications for the credit bureau. framework that met the specific needs of Cambodia. Being able to openly discuss and agree on these rules with This advice was provided by the World Bank Financial the technical partner was seen as critical. IFC organized Infrastructure unit, supported by a prominent local lawyer. a forum that allowed for opinions and different lenders IFC also provided inputs into defining the components requirements to be addressed. of the regulation, undertaking wide-ranging consultation on the draft regulations, incorporating feedback and There were numerous problems with the availability and input from stakeholders, and securing the passage of the quality of data in the market, due to the different technical regulations and promoting their implementation. and resource capacities of the various lenders, records being maintained in English and the local language, irrelevant At the request of the NBC, IFC conducted an in-depth data being captured, lack of a unique identifier, and other analysis of the Cambodian banking and MFI markets factors. All stakeholders were required to invest substantial to gauge the ability and willingness of the local lending time and resources into cleaning up their databases. Some community to share sensitive personal information. used the establishment of the credit bureaus as a reason to Results showed universal support from both sectors for invest heavily in new core banking systems to better manage sharing of credit information to be made compulsory, and their businesses. IFC, in its honest broker capacity, worked for the use of credit reports to be made mandatory. There with a French multilateral to help some of the smaller MFIs was also wide support for the private sector to take the lead replace inadequate loan management systems with a core in establishing a comprehensive credit bureau that would banking solution that allowed them to improve service to cater to the entire lending community, including retailers, 96 Credit Reporting Knowledge Guide their borrowers and to interact with the credit bureau in a reporting council of between 5 and 11 members chaired manner compliant with legal requirements. by the NBC. This council was set up to prevent excessive fees and reflects the importance of MFIs in the Cambodian A key issue that took considerable time to resolve was the market. pricing of credit reporting products for lenders, particularly for the smaller MFI lenders. Given the small size of MFI Lenders were given nine months’ notice to adapt their loans, the MFI sector wanted the banking sector to pay a systems and to provide the required data or face possible greater proportion of the costs of using the bureaus, despite penalties. As the Cambodian market consists of 68 banks the fact that MFIs, would, by virtue of their lending volumes, and MFIs (with loan volumes ranging from 150 to over be the greatest users and beneficiaries of the bureaus. IFC 400,000), with systems and sophistication to match, special supported several rounds of discussions and supplied efforts had to be made by the bureau to meet these unique global pricing benchmarks to enable stakeholders to come challenges. to agreement on a trial pricing regime, whereby loans below $500 were charged a modest fee, and those above The Prakas also covers other usual items such as permissible that amount, were priced slightly higher than the average purposes; data sources, collection, distribution, quality, and price charged by startup bureaus in other jurisdictions. retention periods; security; consumer rights; access and At the end of the trial period, pricing and usage data will challenges to data; dispute resolution, and offenses and be reviewed so that an appropriate pricing model can be penalties. adopted that will provide a financially sustainable model for the bureau and a reasonable return to the shareholders With IFC assistance, stakeholders selected Veda Advantage on their investment. from New Zealand as its strategic partner, and then set up the private Credit Bureau of Cambodia. The company was IFC also facilitated site visits for NBC representatives to licensed by NBC, and the bureau was launched on March Egypt, Peru, and Singapore, so that the NBC could gain 19, 2012. The initial upload of 1,405,722 loans from 67 insight into how the central banks in other countries have of the 68 banks and MFIs was a great achievement, and successfully created environments in which they work a testament of the efforts of all lenders to clean up their closely with the private sector to support the establishment data. Some of the big banks in Cambodia decided to invest of credit reporting systems. in the bureau, resulting a shareholding of 51 percent by local stakeholders and 49 percent by Veda Advantage. After a wide consultation process, a Prakas (set of Veda Advantage proved to be a strong fit as a technical and regulations) designed to meet the needs of the Cambodian strategic partner, given its experience developing bureaus in market was formally approved by NBC. The Prakas other markets and its commitment to work proactively with included some unique requirements, such as the right stakeholders to address the various challenges in developing of the NBC to nominate a director to the board of the the bureau. Veda, in conjunction with the Credit Bureau of private credit bureau, and a provision for an independent Singapore has provided dual language, in-depth training for director to represent shareholder interests. The Prakas local stakeholders, which was seen as critical to the overall required that all regulated lenders provide mandatory success of the bureau. data on all existing loans to the bureau (notwithstanding the loan size) and all new applications for loans, as well as Although it is too soon to know the impact of this credit any renewals or restructuring of existing credit facilities be reporting effort that took about seven years, some initial subject to a credit inquiry. This requirement was designed bureau statistics provide interesting analysis of the market. to enhance the quality of lending, increase access to finance The initial upload of 1.4 million records to the bureau and financial services to those without real collateral, and showed a total portfolio across banks and other regulated prevent overindebtedness. According to the Prakas, pricing lenders (including MFIs) of over $3.6 billion and a borrower policies must to be signed off by an independent credit composition of 97 percent women. Borrowers holding 97 Credit Reporting Knowledge Guide more than one loan represented about 13 percent of total security or if the lender wants to lodge a security to do that of unique borrowers identified in the system, signaling on their behalf in an agent capacity. The bureau also intends potentially increasing levels of overindebtedness. to secure access to public data sources, such as company and business records, so that these can be incorporated into Currently, the Credit Bureau of Cambodia caters only to their reports. lender’s consumer borrowers but will move in the near term to deliver reporting for SMEs. It is also intended that The Credit Bureau of Cambodia was a successful launch the bureau will shortly develop an application processing in a reasonably quick time, illustrating the importance solution for major users and will create a link to the of securing buy in from stakeholders and of learning the collateral registry so that lenders may either check to see lessons from home and abroad. whether moveable collateral offered is subject to another 98 References Barron, J.M., and Michael Staten. 2003. “The Value of Beck, Thorsten, Asli Demirgüç-Kunt, and Ross Levine. Comprehensive Credit Reports: Lessons from the US 2007. “Finance, Inequality and the Poor.” Journal of Experience.” http://www.privacyalliance.org/resources/ Economic Growth 12(1): 27–49. staten.pdf Beck, Thorsten, Asli Demirgüç-Kunt, and Vojislav Basel Committee on Banking Supervision. 2011. Basel III: Maksimovic. 2005. “Financial and Legal Constraints to A Global Regulatory Framework for More Resilient Banks Firm Growth: Does Firm Size Matter?” Journal of Finance and Banking Systems. Bank for International Settlements. 60(1): 137–77. http://www.bis.org/publ/bcbs189.pdf Beck, Thorsten, Ross Levine, and Norman Loayza. 2000. _____. 2010. Basel III: International Framework for “Finance and the Sources of Growth.” Journal of Financial Liquidity Risk Measurement, Standards and Monitoring. Economics 58(1–2): 261–300. Bank for International Settlements. http://www.bis.org/ publ/bcbs188.htm CFPB (Consumer Financial Protection Bureau). 2012. “Consumer Financial Protection Bureau to Supervise _____. 2009a. Guidelines for Computing Capital for Credit Reporting.” http://www.consumerfinance.gov/ Incremental Risk in the Trading Book - Final Version. pressreleases/consumer-financial-protection-bureau-to- Bank for International Settlements. http://www.bis.org/ superivse-credit-reporting/ publ/bcbs159.htm Demirguc-Kunt, Asli, and Leora Klapper. 2012. _____. 2009b. Revisions to the Basel II Market Risk “Measuring Financial Inclusion: The Global Findex.” Framework - Final Version. Bank for International World Bank Policy Research Working Paper 6025, World Settlements. http://www.bis.org/publ/bcbs158.htm Bank, Washington DC. _____. 2009c. Enhancements to the Basel II Framework. Doing Business Indicators (database). World Bank, Bank for International Settlements. http://www.bis.org/ Washington DC. http://www.doingbusiness.org publ/bcbs157.htm Economist Intelligence Unit. 2012. Global Microscope _____. 2006. Basel II: International Convergence on the Microfinance Business Environment 2012. of Capital Measurement and Capital Standards: A The Economist. Revised Framework – Comprehensive Version. Bank for International Settlements. http://www.bis.org/publ/ Global Partnership for Financial Inclusion and bcbs128.htm International Finance Corporation. 2011. “SME Finance Policy Guide.” http://www.gpfi.org/sites/default/files/ documents/SME%20Finance%20Policy%20Guide.pdf 99 Credit Reporting Knowledge Guide Girault, Matías Gutiérrez, and Jane Hwang. 2010. Mix Market. 2012a. Asia Regional Snapshot, 2011. “Public Credit Registries as a Tool for Bank Regulation http://www.themix.org/publications/mix-microfinance- and Supervision.” World Bank Policy Research Working world/2012/03/2011-asia-regional-snapshot Paper 5489, World Bank, Washington DC. Mix Market. 2012b. Countries and Regions, 2012. http://www.mixmarket.org Hilgert, Marianne A., Jeanne M. Hogarth, and Sondra G. Beverly. 2003.”Household Financial Management: OECD (Organisation for Economic Co-operation The Connection between Knowledge and Behavior.” and Development). 2005. “Improving Financial Federal Reserve Bulletin, July, 309–322. http://www. Literacy: Analysis of Issues and Policies.” federalreserve.gov/pubs/bulletin/2003/0703lead.pdf http://www.oecd.org/document/28/0,3343, en_2649_15251491_35802524_1_1_1_1,00.htm IFC (International Finance Corporation). 2010. “The SME Banking Knowledge Guide.” _____. 2004. “Principles of Corporate Governance.” http://www.ifc.org/ifcext/gfm.nsf/AttachmentsByTitle/ http://www.oecd.org/corporate/corporateaffairs/ SMEBankingGuide2010-E.pdf/$FILE/ corporategovernanceprinciples/31557724.pdf SMEBankingGuide2010-E.pdf _____. 1980. “Guidelines on the Protection of Privacy _____. 2012. “Secured Transactions Systems and and Transborder Flows of Personal Data.” Collateral Registries.” http://www1.ifc.org/wps/wcm/ http://www.oecd.org/internet/interneteconomy/ connect/c5be2a0049586021a20ab719583b6d16/ Powell, A., Nataliya Mylenko, Margaret Miller, and SecuredTransactionsSystems.pdf?MOD=AJPERES Giovanni Majnoni. 2004. “Improving Credit Information, Bank Regulation and Supervision: On the Role and Love, Inessa, and Nataliya Mylenko. 2003. “Credit Design of Public Credit Registries.” World Bank Policy Reporting and Financing Constraints.” World Bank Policy Research Working Paper 3443, World Bank, Washington, Research Working Paper 3142, World Bank, Washington DC. DC. Rajan, Raghuram, and Luigi Zingales. 1998. “Financial Lyman, Timothy, Tony Lythgoe, Margaret Miller, Xavier Development and Growth.” The American Economic Reille, and Shalini Sankaranarayan. 2011. “Credit Review 88(3): 559–86. Reporting at the Base of the Pyramid: Key Issues and Success Factors.” Access to Finance Forum, No. 1. Reille, Xavier. 2010. “The Rise, Fall and Recovery of the Washington, DC, CGAP and IFC. Microfinance Sector in Morocco.” Consultative Group http://www1.ifc.org/wps/wcm/connect/c88f08004a17 to Assist the Poor (CGAP). January 2010. http://www. 5970b19dfddd29332b51/CreditBureau-IFC-CGAP- cgap.org/sites/default/files/CGAP-Brief-The-Rise-Fall- Credit%2BReportingPyramid.pdf?MOD=AJPERES and-Recovery-of-the-Microfinance-Sector-in-Morocco- Jan-2010.pdf Meghana, Ayyagari, Asli Demirguc-Kunt, and Vojislav Maksimovic. 2011. “Small v. Young Firms across the Safavian, Mehnaz, Heywood Fleisig, and Jevgenijs World: Contribution to Employment, Job Creation and Steinbuks. 2006. “Unlocking Dead Capital: How Growth.” Policy Research Working Paper 5631. 2011. Reforming Collateral Laws Improves Access to Finance.” The World Bank, Washington DC. Private Sector Development Viewpoint, No. 307, World Bank, Washington DC. 100 Credit Reporting Knowledge Guide Goland, Tony, Robert Schiff, and Peer Stein. 2010. World Bank. 2004. “Credit Bureau Development in “Two Trillion and Counting: Assessing the Credit Gap South Asia, Finance and Private Sector Development, for Micro, Small, and Medium-Size Enterprises in the South Asia Region.” http://siteresources.worldbank. Developing World.” Mckinsey & Co and International org/INTSOUTHASIA/Resources/Credit_Bureau_ Finance Corporation. http://www.mspartners.org/ Development_in_South_Asia.pdf download/Twotrillion.pdf World Bank, DFID (Department for International Stiglitz, Joseph E., and Andrew Weiss. “Credit Rationing Development), OECD (Organisation for Economic in Markets with Imperfect Information.” The American Cooperation and Development), and CGAP (Consultative Economic Review 71 (3): 393-–410. Group to Assist the Poor. 2009. “The Case for Financial Literacy in Developing Countries: Promoting Access Turner, Michael A. 2010. “The Consequences of to Finance by Empowering Consumer.” OECD, Paris. Prohibiting Credit Inquiry Data in Chilean Credit Files: http://www.oecd.org/dataoecd/35/32/43245359.pdf PERC White Paper.” Policy and Economic Research Council (PERC). http://perc.net/files/Chile%20 World Bank, IFC. 2009. Financial Infrastructure: White%20Paper_web.pdf Building Access through Transparent and Stable Financial Systems. http://siteresources. World Bank. 2011a. General Principles for Credit worldbank.org/FINANCIALSECTOR/ Reporting. Washington DC: World Bank. http:// Resources/282044-1252596846652/ siteresources.worldbank.org/FINANCIALSECTOR/ FinancialInfrastructureReport.pdf Resources/Credit_Reporting_text.pdf World Development Indicators (database). World Bank, World Bank. 2011b. Doing Business 2012: Doing Washington DC. http://data.worldbank.org/indicator Business in a More Transparent World. Washington DC: World Bank. http://www.doingbusiness.org/reports/ global-reports/doing-business-2012 101 Annexes Annex 1: Regional Maps of Positive and Negative Credit Bureaus................................. 103 Annex 1.1: Europe and Central Asia Region ............................................................... 104 Annex 1.2: Middle East and North Africa Region ....................................................... 105 Annex 1.3: Africa Region .............................................................................................. 106 Annex 1.4: South Asia Region and East Asia and Pacific Region ............................... 107 Annex 1.5: Latin America and Caribbean Region ....................................................... 108 Annex 2: Overview of Credit Registries and Credit Bureaus around the World ......cover 3 103 Annex 1.1: Europe and Central Asia Region 104 Annex 1.2: Middle East and North Africa Region 105 Annex 1.3: Africa Region 106 Annex 1.4: South Asia Region and East Asia and Pacific Region 107 Annex 1.5: Latin America and Caribbean Region 108