72331 REDUCING VULNERABILTY AND PROMOTING THE SELF-EMPLOYMENT OF ROMA IN EASTERN EUROPE THROUGH FINANCIAL INCLUSION 4 September 2012 The World Bank 1 World Bank ECA Vice President : Phillippe H. Le Houerou World Bank ECCU5 Country Director : Peter Harrold World Bank HD Sector Director : Ana Revenga and Mamta Murthi World Bank HDE Sector Manager : Roberta Gatti and Jesko Hentschel World Bank Task Team Leader : Joost de Laat This report has been prepared for the European Commission; however, it reflects the views only of the authors, and the Commission cannot be held responsible for any use which may be made of the information contained therein. This report draws on the results of the UNDP/World Bank/EC regional Roma 2011 survey. However, the report was prepared by the World Bank and does not necessarily represent the views of UNDP. 2 ACKNOWLEDGEMENTS This report is a product of the World Bank’s analytical and advisory services. It is based on a survey of Roma households and non-Roma living nearby in Bulgaria, Romania, Hungary, Slovakia, and the Czech Republic. The survey questionnaire was designed jointly by a team from the World Bank and UNDP. The financing to implement the survey and write the report came from the EC DG Regional Policy. The Nordic Trust Fund at the World Bank provided co-financing for a health module in these five countries. UNDP, which implemented the survey through the IPSOS polling agency, financed implementation of the same survey in Western Balkan countries and Moldova. The survey effort was closely coordinated with the FRA Roma pilot survey 2011, implemented simultaneously in 11 EU Member States.1 The report was prepared by Joost de Laat (Senior Economist ECSHD, and lead author 2), Rosen Asenov (Consultant), Stefan Hut (Consultant), Barbara Kits (Consultant), and Federico Torracchi (Consultant). The team worked under the direction of Peter Harrold (Country Director, ECCU5), Ana Revenga (Sector Director ECSHD), Roberta Gatti (Human Development Economics Sector Manager), Mamta Murthi (prior Acting Sector Director, ECSHD), and Jesko Hentschel (prior Sector Manager). The team received administrative support from Anne Anglio (Senior Program Assistant, ECSHD). The team would like to express its gratitude to the peer reviewers for their detailed feedback: Sandor Karacsony (Consultant), Zahia Msuya Lolila-Ramin, Senior Financial Sector Specialist (AFTFW), and György Molnár, Senior Research Fellow, Institute of Economics Hungarian Academy of Sciences (IE HAS). The team would also like to express its gratitude to comments and suggestions received, especially from the European Commission DG Regional Policy and DG Employment (including members of the EC Roma Pilots Steering Committee headed by Veronica Gaffey (Director)); the UNDP Bratislava Regional Center; and, Aurora Ferrari, Manager, World Bank MSME Finance Service Line, Katarina Mathernova, World Bank Senior Advisor, and Mohamed Ihsan Ajwad, Senior Economist, ECSHD. 1 Summary findings on key indicators from the two surveys have been published by the Fundamental Rights Agency and UNDP in The situation of Roma in 11 EU Member States; Survey Results at a Glance (2012). 2 For comments or questions: jdelaat@worldbank.org 3 Acknowledgements 3 List of Acronyms 7 Executive Summary 8 1 Background 13 1.1 Introduction 13 1.2 Data and Methodology 14 1.3 Welfare Conditions Among Roma 15 1.4 Labor Market Overview 17 2 Financial Exclusion among Roma 21 2.1 What is Financial Exclusion? 21 2.2 Physical and Electronic Access to Financial Services 22 2.3 Usage of Financial Services 24 2.4 Savings and Vulnerability 26 3 Roma Self-Employment and Entrepreneurship 30 3.1 Self-Employment Rates 30 3.2 Types of Businesses 31 3.3 Interest in Becoming Self-Employed and Starting Business 33 3.4 Barriers to Business Start-Ups 36 4 Accessing Micro-Credit Enterprise Financing 40 4.1 The Emergence of Microfinance 40 4.2 Usage of Credit among Roma 41 4.3 Demand Side Barriers to Accessing Microcredit 45 4.4 Supply Side Barriers to Accessing Microcredit 53 4.5 EU and Other International Initiatives for Microcredit and Entrepreneurship 57 5 A Comprehensive, Incremental Approach to Financial Inclusion of Roma 60 5.1 Introduction 60 5.2 Innovative Ways to Reach the Poor with Financial Services 62 5.3 Promote (Targeted) Savings 64 5.4 Support Financial Literacy and Debt Management 65 5.5 Support Access to Mainstream Microcredit 66 5.6 Use Monitoring and Evaluation to Promote Financial Inclusion 68 Annex A: EU Initiatives and Policies for SMEs 69 Policy: The European Charter for Small Enterprises and The Small Business Act (SBA) 69 Data Collection: SME Performance Review 71 Annex B: Estimation results 73 Bibliography 78 4 LIST OF FIGURES Figure 1: Poverty among Roma and Non-Roma Neighbors ____________________________________________ 16 Figure 2: Employment Rates for Men and Women ___________________________________________________ 18 Figure 3: Self-reported ethnic discrimination over the last 5 years among Roma adults (aged 15+) ____________ 20 Figure 4: % Households Living Less than 3KM Away from a Bank______________________________________ 23 Figure 5: Usage of Current Accounts and Credit Cards (%) ___________________________________________ 24 Figure 6. % Households with Savings among Roma and Non-Roma Neighbors ____________________________ 27 Figure 7. % Self-employed and Employed individuals among the working age population ____________________ 30 Figure 8. Share of Self-Employment among the Employed, Roma and General Population ___________________ 31 Figure 9. Interest in Self-Employment and Starting Business among working Age Individuals _________________ 34 Figure 10. Employment Status among Roma Interested in Starting a Business _____________________________ 34 Figure 11: Ease of Doing Business Rankings (1-183): EU27___________________________________________ 37 Figure 12: Use of Main Loan by Roma Households __________________________________________________ 43 Figure 14. Share of Roma and Non-Roma Neighboring Entrepreneurs with Bank Accounts __________________ 45 Figure 15. Share of Working Age Roma with Existing Businesses (Intensive Margin) and Aspiring to Start a Business (Extensive Margin) ___________________________________________________________________________ 46 Figure 16: Secondary Education Completion among (Potential) Credit Groups ____________________________ 48 Figure 17: Years Spent in Paid Work among (Potential) Credit Groups __________________________________ 48 Figure 18. Arrears among (aspiring) Roma Entrepreneurs ____________________________________________ 51 Figure 19. Household Economic Safety among (Aspiring) Roma Entrepreneurs (%) ________________________ 52 LIST OF TABLES Table 1: Distribution of Households in Sampled Roma Communities ......................................................................... 15 Table 2: Basic Needs and Economic Safety (% of Households) .................................................................................. 16 Table 3: Segregation characteristics - % of Roma Households................................................................................... 17 Table 4: % of Roma Among New Labor Market Entrants ........................................................................................... 17 Table 5. Labor Market Statistics for Roma: Men ......................................................................................................... 18 Table 6. Labor Market Statistics for Roma: Women .................................................................................................... 19 Table 7. Informal Employment: Written Contract and Social Contributions (% Roma) ............................................. 19 Table 8: Highest Individual Education Level Completed (%) ..................................................................................... 20 Table 9: Retail Network Physical Outreach................................................................................................................. 22 Table 10. % Households Living Less than 3 KM away from a Bank .......................................................................... 23 Table 11. Possession of Phones, Computers, and Internet Connection (%) ................................................................ 24 Table 12. Usage of Saving Accounts and Payment Services (%) ................................................................................. 25 Table 13. Correlates of Households’ Usage of Banking Services ................................................................................ 26 Table 14. Usage of Banking Services: Roma Households Using Saving Accounts and Payment Services.................. 28 Table 15. Comparison of Roma and Non-Roma Households’ Savings ........................................................................ 28 Table 16. Arrears on Household Mortgages, Loans, and Utility Bills (%) .................................................................. 28 Table 17: Comparison of Arrears between Roma and Non-Roma neighbors .............................................................. 29 Table 18. Types of Businesses among Self-Employed Roma ........................................................................................ 31 Table 19. Business Size for Self-Employed Roma ........................................................................................................ 32 Table 20. Business Structure of Non-Financial Enterprises. ....................................................................................... 32 Table 21. Type of Industry for Self-Employed Roma ................................................................................................... 33 Table 22: Average Working Time Experience (%)...................................................................................................... 35 Table 23. Ease of Doing Business: Indicators relevant for SMEs ............................................................................... 37 Table 24. Main Problems in Starting A Business Among Roma and Non-Roma Neighbors (%)................................. 38 5 Table 25. Main Problems Operating a Business Among Roma and Non-Roma Neighbors (%).................................. 39 Table 26. Borrowing Among Roma Households (%) ................................................................................................... 41 Table 27. Average Loans and Installments .................................................................................................................. 42 Table 28. Sources of Credit among Roma Households ................................................................................................ 42 Table 29. Sources of Funding for Businesses by Roma and Non-Roma neighboring Entrepreneurs .......................... 44 Table 30. Intensive and Extensive Margin: Number of Observations .......................................................................... 46 Table 31. Demographic Statistics for Roma (intensive and extensive margin) and General Population (comparison groups) ......................................................................................................................................................................... 47 Table 32. Professional and Computer Skills among Roma .......................................................................................... 49 Table 33. Use of Banking Services among (Aspiring) Roma Entrepreneurs (%)......................................................... 49 Table 34. Dwelling Type and Home Ownership among (Aspiring) Roma Entrepreneurs (%) .................................... 50 Table 35. Possessions among (Aspiring) Roma Entrepreneurs and General Population Comparators(%)................ 50 Table 36. Savings among Roma – Intensive and Extensive Margins ........................................................................... 50 Table 37: Temporary Employment among (Aspiring) Roma Entrepreneurs and General Population ........................ 51 Table 38: Possession of Personal Documents among Roma (%). ............................................................................... 52 Table 39: Summary of Barriers for (Aspiring) Roma Borrowers ................................................................................ 53 LIST OF BOXES Box 1: The Contribution of SMEs to National Employment and Value Added in Eastern Europe .............................. 32 Box 2: Microenterprise support through the New Széchenyi Plan 2011-13 ................................................................ 56 Box 3: Financial inclusion – in a broader sense – is increasingly being promoted at EU level ................................. 60 Box 4: The Graduation Approach: USING Financial Inclusion to Create Pathways to Graduate out of Poverty...... 61 Box 5: Bank on San Francisco, United States.............................................................................................................. 62 Box 6: ‘Correspondent Banking’ in Brazil .................................................................................................................. 63 Box 7: Mobile Financial Services: M-pesa Kenya ....................................................................................................... 63 Box 8: Earmarking Savings: United States and Ghana ............................................................................................... 64 Box-9: ‘Targeting Savings’ Project among Roma in Slovakia..................................................................................... 65 Box 10: Financial Literacy Training among Roma in Slovakia................................................................................... 66 Box 11: Experiences with Microcredit among Starting Roma Entrepreneurs: The KIUT Program in Hungary ........ 66 6 LIST OF ACRONYMS ATM Automated Teller Machine BDS Business Development Services BPA Basic Payment Account CEE Central and Eastern Europe CGAP Consultative Group to Assist the Poor CIDA Canadian International Development Agency CMZRB Czech and Moravian Guarantee and Development Bank CRS Catholic Relief Services CSR Corporate Social Responsibility EA Enumeration Area EBRD European Bank for Reconstruction and Development EC European Commission EGFE Expert Group of Financial Education EIB European Investment Bank EIF European Investment Fund EMN European Microfinance Network EPMF European Progress Microfinance Facility ESF European Social Fund EU SILC European Union Statistics on Income and Living Conditions FINCA Foundation for International Community Assistance FSP Financial Service Provider GBER General Block Exemption Regulation on State Aids GDP Gross Domestic Product HDI Human Development Index IMF International Monetary Fund JASMINE Joint Action to Support micro-finance Institutions in Europe JEREMIE Joint European Resources for Micro to Medium Enterprises JOBS Job Opportunities through Business Support LEA Local Enterprise Agency (Hungary) MCS National Microcredit Scheme (Hungary) MFI MicroFinance Institution MVA Hungarian Foundation for Enterprise Promotion NGO Non-Governmental Organization ODC Other Depository Corporation OECD Organization for Economic Co-operation and Development OMRO Opportunity Microfinance ROmania OPEI Operational Programme Enterprise and Innovations (Czech Republic) POS Point Of Sale PPS Purchasing Power Standard PPTS Probability Proportional To Size PSU Primary Sampling Unit ROSCA Rotating Saving and Credit Association SBA Small Business Act (for Europe) SDC Swiss Agency for Development and Cooperation SME Small- and Medium Enterprise SPE Statute for a Private European Company SZRB Slovak Guarantee and Development Bank UNDP United Nations Development Program USAID United States Agency for International Development VAT Value Added Tax 7 EXECUTIVE SUMMARY REDUCING VULNERABILTY AND PROMOTING (SELF-) EMPLOYMENT THROUGH FINANCIAL INCLUSION Roma households continue to face extremely high levels of vulnerability in relation to very basic needs. New United Nations Development Programme (UNDP)/World Bank/European Commission (EC) regional Roma survey (2011) data for Eastern Europe show that in each of the five countries surveyed, a large majority of Roma live in poverty and face difficulties in paying for basic goods and services. Moreover, more than one-third of the surveyed Roma households report that at least one person in the household went to bed hungry in the previous month in four out of the five countries—reaching 56 percent of Roma households surveyed in Romania. Policy efforts to promote Roma inclusion have focused on the employment, education, housing, and health sectors, as well as anti-discrimination programs. This multisectoral approach reflects the fact that the vulnerability of Roma households is rooted in multiple and interlinked deprivations. For example, only 20–43 percent of working-age Roma men in Eastern Europe are employed, and even fewer working- age Roma women—between 9 and 26 percent—have formal or informal jobs. This is closely linked to education, as less than one-third of Roma have finished upper secondary education, but discrimination, poor health, and poor housing conditions are important factors also. More generally, deprivation in one area is linked to deprivations in each of the other areas. One policy area for improving Roma welfare, the financial inclusion of Roma, has received relatively little policy attention. Poor and vulnerable households, Roma and non-Roma, need a broader range of financial services as much as, if not more than, nonpoor households. Savings, for example, allow the poor not only to respond to income shocks and smooth consumption, but also to invest in education, pay for health expenses, or set aside a down payment for a (micro) loan. Poor households also need careful financial planning to make ends meet with limited resources and to save for incremental investments in their well-being. Access to microcredit may enable poor Roma households to make home improvements, obtain a legal land title, or start a business and become self-employed. The role of financial inclusion to combat poverty is increasingly being recognized globally, and a growing number of governments are seeking comprehensive policy-level solutions to improve access to financial services. The 2010 Seoul G20 summit recognized financial inclusion as one of the main pillars of the global development agenda, and more than 60 countries have initiated financial inclusion reforms in recent years. A comprehensive strategy promotes not only the adoption of financial products and services, but also the ability of customers to take full advantage of them (World Bank, 2012a). Financial inclusion is also increasingly being promoted at the European Union (EU) level. The EC has followed an approach on financial inclusion geared towards basic bank account access and “financial education,� that is, improving EU citizens’ understanding of relatively complex financial products. There are also a number of EC-supported instruments described in the report, such as JASMINE, JEREMIE, PROGRESS, and CIP, promoting microcredit. With regard to the financial inclusion of Roma, there has been special EU attention to exploring the use of microcredit as a tool for promoting self-employment. The EC Communication on an EU Framework for National Roma Integration Strategies (EC 2011a), issued in April 2011 and endorsed by the European Council on June 24, 2011, highlights access to microcredit as a way to promote self- 8 employment. Furthermore, the European Parliament, in partnership with the EC DG Regional Policy, financed an EC Roma microcredit project in 2010 called the Kiút Program. This two-year pilot project in Hungary was an initiative of the Polgar Foundation in collaboration with Raiffeisen Bank, and had as its primary goal the systematic assessment of the viability of using microcredit to raise self-employment rates among Roma. In parallel, the EC DG Regional Policy requested the World Bank, in collaboration with UNDP, to assess Roma financial inclusion in Eastern Europe and, more specifically, the potential of microcredit in promoting Roma self-employment, by using a new regional survey of Roma households. This report therefore investigates two complementary questions: first, whether Roma in Eastern Europe lack access to a wide range of basic financial services; and second, whether improving access to microcredit can be a means to substantially raise self-employment. The analysis takes advantage of a new comprehensive UNDP/World Bank/EC regional Roma survey (2011), which provides detailed information on the living and employment conditions—including access to microcredit and financial services more broadly—of Roma in five countries: the Czech Republic, Hungary, Slovakia, Bulgaria, and Romania. The survey represents 78–90 percent of the entire Roma population in the countries analyzed. For comparison with the general populations, the report takes advantage of the EU Statistics on Income and Living Conditions (EU SILC) survey, among other data sources. ROMA LACK ACCESS TO A BROAD RANGE OF FINANCIAL SERVICES, ESPECIALLY VULNERABLE HOUSEHOLDS Roma lack access to a broad range of financial services, including basic current (or checking) accounts and savings accounts. The EU average for usage of current accounts is approximately 87 percent. In Eastern Europe, the rates are generally lower, but particularly so among Roma households. Standing out with a relatively high usage of current accounts are the Roma in the Czech Republic, Hungary, and Slovakia, where 47, 35, and 29 percent, respectively, of Roma households have such accounts, although this compares unfavorably with the general population, among whom more than 75 percent of households have a current account. In Bulgaria and Romania, very few Roma households have a current account, as the numbers are only 4 and 6 percent, respectively. In all five countries, less than 5 percent of Roma households have savings accounts. Those that do have a savings account are almost 26 percentage points more likely to be able to cover unexpected expenses compared with Roma households without savings accounts from the same community, after controlling for background characteristics such as household size, education, income, age structure, and home ownership. Roma are also much less likely to use formal financial instruments than their non-Roma neighbors living in the vicinity. For example, Roma households remain 17 percentage points less likely to save at all—regardless of where—compared to non-Roma neighboring households, even when controlling for background characteristics. Moreover, compared to their non-Roma neighbors, Roma households are also much more likely to have arrears, again after controlling for background characteristics. THERE IS CONSIDERABLE INTEREST IN SELF-EMPLOYMENT AMONG ROMA, BUT MICROCREDIT WILL NOT SUBSTANTIALLY RAISE SELF-EMPLOYMENT RATES UNLESS MORE BASIC FINANCIAL INCLUSION GAPS ARE ALSO ADDRESSED Increasing the extremely low employment rates among Roma is an economic necessity for the countries in Eastern Europe and a priority for the EC. As many as 7–20 percent of new labor market entrants in Bulgaria, the Czech Republic, Hungary, Romania, and Slovakia are Roma, yet their 9 employment rates continue to lag far behind those of the majority populations, and behind the Europe 2020 headline target of 75 percent of the population aged 20–64 employed. The low employment rates do not reflect preferences, as the vast majority of Roma express a desire for stable jobs, similar to the responses of non-Roma neighbors. Among working Roma, the share of people who are self-employed is low in the Czech Republic, Hungary, and Slovakia, while noticeably higher in Bulgaria and Romania. Among employed male Roma, the proportion of self-employed individuals, including many unskilled workers with occasional jobs, ranges from 5 percent in Hungary to 45 percent in Romania, compared to 14–26 percent among the general male populations. Similarly, between 1 and 34 percent of Roma working-age women are self- employed in the five countries, whereas it is 8–12 percent among the general female population. Among working Roma women, those in Bulgaria and Romania also have the highest shares of entrepreneurs. A large share of Roma not already self-employed report an interest in becoming so. Across the five countries, 24–40 percent of all working-age Roma men and between 14 and 35 percent of Roma women of working age report that they are interested or may be interested in becoming self-employed. Roma who report an interest in self-employment are found among the employed, the unemployed, and those not participating in the labor force. Among women, a relatively large share of this group is not in the labor force, whereas among men, most of those interested in self-employment do participate in the labor force, employed and unemployed. Indicators on the overall business environment and responses by surveyed entrepreneurs point to barriers—especially the lack of access to finance—to expanding self-employment that are both country specific and specific to Roma. The World Bank’s Doing Business survey results show that starting a business in the five countries is most difficult in Slovakia and especially the Czech Republic (World Bank, 2011). These findings may go some way in explaining why Roma in these two countries have lower self-employment rates than in the other countries. Yet, lack of start-up finance is highlighted as a particular challenge by entrepreneurs in each of the countries, along lack of customers and licensing. However, microcredit is unlikely to substantially raise employment rates among Roma unless a comprehensive approach to financial inclusion is undertaken that goes beyond strictly providing microcredit. At the moment, the potential group of Roma that can take advantage of microcredit is small, as most Roma interested in becoming self-employed face significant obstacles that limit access to credit. Not many suppliers of microfinance target start-up businesses in general, and even fewer are reaching (starting) Roma entrepreneurs, as a result of the substantial difficulties specific to Roma, including lack of savings and indebtedness, very low levels of education, even when compared to the segment of the general population that is being refused credit, and little employment experience, especially lacking experience starting and operating a business in the formal economy. The micro-lending challenges facing the recent Hungarian Kiút microcredit pilot program and the ways in which Kiút has overcome these challenges – through a comprehensive approach that includes considerable support with tax and business registration as well as business training - provides an important example in this regard. A COMPREHENSIVE APPROACH TOWARD THE FINANCIAL INCLUSION OF ROMA, FOCUSING ON ACCESS TO BANK ACCOUNTS, (TARGETED) SAVINGS, FINANCIAL LITERACY, AND BUSINESS SKILLS TRAINING To reduce vulnerability and promote self-employment of Roma through financial inclusion, European countries can take advantage of the experiences of both rich and poor countries in tackling this issue. A comprehensive, incremental approach to financial inclusion is suggested that 10 addresses basic barriers, such as the limited outreach of financial service providers in Roma communities and the (related) lack of access to savings technology, but also addressing low financial literacy and business skills. Given the extremely high poverty levels, low rates of education, and general degree of marginalization and segregation of Roma, some of the successful lower-income country efforts to achieve financial inclusion are therefore very relevant, while countries can also consider the experiences of rich countries seeking to achieve the “last mile� of financial inclusion, that is, the final leg of providing access to financial services to all citizens. Several examples are listed below, which the report highlights. INSTITTUTIONAL APPROACHES TO REACHING THE POOR WITH BASIC FINANCIAL SERVICES First, different countries are using a variety of institutional approaches to reaching financially excluded groups with basic financial services. For example, in a high income country like the United States, the Bank on San Francisco initiative seeks to bring unbanked individuals into the financial mainstream through a collaborative effort between the government, private sector financial institutions, and civil society. In middle income country like Brazil, where reaching poor clients in rural areas is often prohibitively expensive for financial service providers, banks started using local agents to manage their operations at the village level. And in a lower income country like Kenya, the highly advanced M-PESA mobile phone service allows customers to save money, transfer funds to other mobile phone users, and make payments for services such as utility bills. Countries can also promote financial inclusion by taking advantage of government social protection payment systems. Many households—Roma and non-Roma, poor and nonpoor—receive government social protection payments for pensions, family benefits, social assistance, education support, unemployment insurance, and so on. In many countries, these payments are not necessarily made to people’s bank accounts but can be made in cash, for example, through the post office. This is the case with regard to social assistance payments in Slovakia, for instance. Governments can play a significant role in incentivizing the financial inclusion of Roma by transferring various benefits to accounts that beneficiaries, Roma and non-Roma, open in commercial banks. This is also attractive for governments, since it frequently reduces the administrative costs of transfer programs. PROMOTING (TARGETED) SAVINGS A key component of the incremental approach to financial inclusion is savings facilitation and linking savings activities to human development outcomes. For example, governments could support households in opening up targeted education or housing savings accounts, in which households are encouraged to save with an explicit purpose. International evidence suggests that targeted savings are successful in raising overall savings. Several locally implemented initiatives can be adopted and scaled up, such as ETP Slovakia’s “Individual Development Account,� which provides incentives to save for housing improvements. This report also describes other international models, such as the “Kindergarten- to-College (K2C)� savings initiative used in the United States, and the labeling of savings accounts to ensure that savings are “earmarked� for housing improvements, school fees, and so forth. An example of a comprehensive approach is the Consultative Group to Assist the Poor (CGAP)- Ford Foundation Graduation Program. This is a global effort to understand how safety nets, livelihoods, and microfinance can be sequenced to create pathways for the poorest to graduate out of extreme poverty. It provides an example of how poor households facing very high economic insecurity are being targeted in low- and middle-income countries through an incremental approach that first addresses 11 the most obstructive barriers that poor households face in accessing financial services, including through financial literacy training and a focus on savings. Its focus on savings, for example, before business skills training and asset transfer is as relevant for the vast majority of Roma lacking a systematic means of savings as it is for the beneficiaries of the Graduation pilots. PROMOTING FINANCIAL LITERACY AND BUSINESS SKILLS To successfully raise savings and increase household investments, basic financial literacy and debt management training must accompany a focus on savings facilitation. Excellent examples from around the world are plentiful, but countries can also build on locally implemented initiatives that can be scaled up. There are many financial literacy and business skill training modules, including for school children. In Slovakia, the nongovernmental organization (NGO) most active in the area of small-scale financial inclusion activities with disadvantaged Roma is ETP. Together with the Autonomia Foundation, ETP has been working with standardized financial training modules that have proven to be successful entry points into microsavings and microcredit programs for clients living in marginalized communities. For example, an important entry point to increasing financial literacy among Roma are schools where teenagers can be targeted. MAINSTREAMING MICROCREDIT ACCESS FOR ROMA ENTREPRENEURS Not-for-profit initiatives can promote greater access to microcredit for Roma entrepreneurs by identifying promising candidates and helping them meet the eligibility requirements of mainstream microcredit providers. While some small-scale projects have successfully reached poor Roma families, the number of loans disbursed is often small making the cost per loan relatively large. The institutional set-up frequently leaves out depository institutions, thus limiting the scope of the savings promotion. A more promising avenue may be the establishment of partnerships between pro-social, not-for-profit initiatives on the one hand, and mainstream credit providers—ideally depository ones—that provide microcredit to the self-employed on the other. The Kiut microcredit experience shows that not-for-profit support can focus on (a) identifying promising (aspiring) Roma entrepreneurs, and (b) helping them meet the conditions of mainstream credit providers, including through business plan, financial literacy, and savings support (for down payment), possibly in combination with asset transfer (for collateral or as a start-up grant). Countries can leverage EU structural funds to support these pro-social initiatives as well as specific programs such as JASMINE and PROGRESS. PROMOTING RESULTS THROUGH THE SOUND MONITORING OF INDICATORS AND RIGOROUS EVALUATION OF FINANCIAL INCLUSION PROGRAMS The consistent monitoring of financial inclusion indicators and rigorous evaluation of programs promoting financial inclusion are critical to achieving results. This is also recognized by the April 2011 EU Framework for National Roma Integration Strategies, which calls upon EU member states to include strong monitoring and evaluation (M&E) components. Fortunately, there are well-established monitoring surveys and considerable global experience with rigorous impact evaluations of financial inclusion initiatives, ranging from financial literacy support and (targeted) savings to the impact of microcredit and start-up grant support. This report highlights several useful global resources on M&E of financial inclusion, including the World Bank’s financial inclusion program, the financial inclusion (FINDEX) database, and the Global Financial Inclusion initiative at Innovations for Poverty Action. 12 1 BACKGROUND This chapter highlights the growing global and European financial inclusion agenda and provides a brief assessment of vulnerability and labor market outcomes for Roma in Eastern Europe. Financial inclusion is considered by many to be among the main instruments to improve welfare, especially among the poor, which need financial services as much as, if not more, than non-poor, to reduce vulnerability and improve self-employment. Following a brief description of the main data sources used in the analysis, the chapter highlights the very high levels of vulnerability and exclusion among Roma in Eastern Europe and the poor labor market outcomes by any measure, including extremely low employment rates and high informality. Against this background, the following chapters assess financial exclusion among Roma, and self-employment levels and (credit) barriers to starting businesses. The final chapter highlights international experiences promoting comprehensive, incremental approaches to financial inclusion. 1.1 INTRODUCTION Roma households continue to face extremely high levels of vulnerability in relation to very basic needs. New United Nations Development Programme (UNDP)/World Bank/European Commission (EC) regional Roma survey (2011) data for Eastern Europe show that in each of the five countries surveyed, a large majority of Roma live in poverty and face difficulties in paying for basic goods and services. Moreover, more than one-third of the surveyed Roma households report that at least one person in the household went to bed hungry in the previous month in four out of the five countries—reaching 56 percent of Roma households surveyed in Romania. Policy efforts to promote Roma inclusion have focused on the employment, education, housing, and health sectors, as well as anti-discrimination programs. This multisectoral approach reflects the fact that the vulnerability of Roma households is rooted in multiple and interlinked deprivations. For example, only 20–43 percent of working-age Roma men in Eastern Europe are employed, and even fewer working- age Roma women—between 9 and 26 percent—have formal or informal jobs. This is closely linked to education, as less than one-third of Roma have finished upper secondary education, but discrimination, poor health, and poor housing conditions are important factors also. More generally, deprivation in one area is linked to deprivations in each of the other areas. One policy area for improving Roma welfare, the financial inclusion of Roma, has received relatively little policy attention. Poor and vulnerable households, Roma and non-Roma, need a broader range of financial services as much as, if not more than, nonpoor households (Ehrbeck, 2011). Savings, for example, allow the poor not only to respond to income shocks and smooth consumption, but also to invest in education, pay for health expenses, or set aside a down payment for a (micro) loan. Poor households also need careful financial planning to make ends meet with limited resources and to save for incremental investments in their well-being. Access to microcredit may enable poor Roma households to make home improvements, obtain a legal land title, or start a business and become self-employed. In similar vein, financial exclusion is harmful not only for households and individual businesses, but can also limit economic growth and compromise stability (v. e.g.: World Bank, 2008: 23 a.o.; Carbo et al., 2005: 100; Peachey and Roe, 2004), and it may increase the inequality of income distributions of an economy (Karlan and Morduch, 2009: 2). The role of financial inclusion to combat poverty is increasingly being recognized globally, and a growing number of governments are seeking comprehensive policy-level solutions to improve access to financial services. The 2010 Seoul G20 summit recognized financial inclusion as one of the main 13 pillars of the global development agenda, and more than 60 countries have initiated financial inclusion reforms in recent years. A comprehensive strategy promotes not only the adoption of financial products and services, but also the ability of customers to take full advantage of them (World Bank, 2012a). Financial inclusion is also increasingly being promoted at the European Union (EU) level. The EC has followed an approach on financial inclusion geared towards basic bank account access and “financial education,� that is, improving EU citizens’ understanding of relatively complex financial products. There are also a number of EC-supported instruments described in the report, such as JASMINE, JEREMIE, PROGRESS, and CIP, promoting microcredit. With regard to the financial inclusion of Roma, there has been special EU attention to exploring the use of microcredit as a tool for promoting self-employment. The EC Communication on an EU Framework for National Roma Integration Strategies (EC 2011a), issued in April 2011 and endorsed by the European Council on June 24, 2011, highlights access to microcredit as a way to promote self- employment. Furthermore, the European Parliament, in partnership with the EC DG Regional Policy, financed an EC Roma microcredit project in 2010 called the Kiút Program. This two-year pilot project in Hungary was an initiative of the Polgar Foundation in collaboration with Raiffeisen Bank, and had as its primary goal the systematic assessment of the viability of using microcredit to raise self-employment rates among Roma. In parallel, the EC DG Regional Policy requested the World Bank, in collaboration with UNDP, to assess Roma financial inclusion in Eastern Europe and, more specifically, the potential of microcredit in promoting Roma self-employment, by using a new regional survey of Roma households. This report therefore investigates two complementary questions: first, whether Roma in Eastern Europe lack access to a wide range of basic financial services; and second, whether improving access to microcredit can be a means to substantially raise self-employment. The analysis takes advantage of a new comprehensive UNDP/World Bank/EC regional Roma survey (2011), which provides detailed information on the living and employment conditions—including access to microcredit and financial services more broadly—of Roma in five countries: the Czech Republic, Hungary, Slovakia, Bulgaria, and Romania. For comparison with the general populations, the report takes advantage of the EU Statistics on Income and Living Conditions (EU SILC) survey, among other data sources. 1.2 DATA AND METHODOLOGY The findings in this report rely on a new household survey with detailed information on financial inclusion: the UNDP/World Bank/EC regional Roma survey (2011). This survey (henceforth: the regional Roma survey (2011)) was designed by the World Bank and UNDP in partnership, financed by the European Commission DG Regional Policy, and implemented by UNDP through the IPSOS polling agency in May-July 2011. Interviews were held with a random sample of Roma in Bulgaria, Romania, Hungary, Slovakia, and the Czech Republic. In each of the countries, approximately 750 Roma households and approximately 350 non-Roma households living in the same neighborhoods or vicinity were interviewed (Table 1). The ‘non-Roma neighbors’ sample is important because it allows for comparisons in financial access and usage outcomes between Roma and non-Roma facing similar local environments. The regional Roma survey represents 78% - 90% of the entire Roma population in the countries analyzed. The sample was purposefully not representative of all Roma in these countries, but rather focused on those communities where the share of the Roma population equals or is higher than the national share of the Roma population. This sampling frame approach covers 88% of the Roma population in Bulgaria, 90% in the Czech Republic, 78% in Hungary, 89% in Romania, and 83% in Slovakia. Once 14 the primary sampling units (PSUs) were identified, a random sample of enumeration areas was drawn, and households were randomly sampled within these. In each of the countries rural households are in the majority with the exception of the Czech Republic where almost all sampled households are found in urban locations. For comparison with the general populations, the report mainly takes advantage of the 2008 EU Statistics on Income and Living Conditions (EU SILC) survey. The EU-SILC does not distinguish between Roma and non-Roma and provides household survey information that is representative of the general population in each of the countries. 3 In general, the comparisons between Roma and non-Roma neighbors in the 2011 regional Roma survey will display smaller gaps than comparisons between Roma and the general populations in the EU SILC. TABLE 1: DISTRIBUTION OF HOUSEHOLDS IN SAMPLED ROMA COMMUNITIES Czech Bulgaria Hungary Romania Slovakia Total Republic Roma 763 756 753 757 756 3,785 Non-Roma neighbors 366 350 354 350 350 1,770 Source: UNDP/World Bank/EC regional Roma survey (2011). 1.3 WELFARE CONDITIONS AMONG ROMA Despite overall gains in the standard of living in these countries over recent years, Roma households still face extremely high levels of insecurity in relation to very basic needs. The figure below shows that more than two third of all Roma represented by the survey data live under the poverty line, much higher even than non-Roma nearby. More than half of Roma households in each country report facing difficulties in paying for basic goods and services. Overall, self-reported economic insecurity affects 54% of all Roma households in the Czech Republic and reaches 70% in Hungary. This lack of economic safety reflects low levels of disposable income. Roma have a per capita disposable income less than half of the average disposable income in each country – and as little as 26% in Romania. As shown in the table below, over 35% of the surveyed households report that at least one person in the household went to bed hungry in the previous month in four out of five countries– reaching 56% in Romania. Even in the Czech Republic – the country with the lowest proportion – this figure exceeds 20%. 3 Since the Roma are a minority in each of the countries analyzed, their inclusion in the EU SILC data slightly biases comparisons between ‘the general population’ and ‘Roma’ specifically, meaning that the actual differences between non-Roma in the general population and the Roma represented by the Roma Regional Survey are likely to be slightly larger than reported by the figures in this study. 15 FIGURE 1: POVERTY AMONG ROMA AND NON-ROMA NEIGHBORS 100% 87% 90% 81% 80% 74% 71% 71% 70% 60% 50% 45% 40% 40% 33% 30% 26% 22% 20% 10% 0% Bulgaria Czech Hungary Romania Slovakia Republic Non-Roma neighbors Roma Source: UNDP/World Bank/EC regional Roma survey (2011). Figure represents: Share of people living in the households where per capita income is below the defined poverty line in the total number of people in the interviewed households (60% of the median equivalised disposable income= poverty). TABLE 2: BASIC NEEDS AND ECONOMIC SAFETY (% OF HOUSEHOLDS) Czech Bulgaria Hungary Romania Slovakia Republic Roma Living Conditions: Go to Bed Hungry (%) A 44% 24% 35% 56% 40% Unsafe Economic Situation (%) 62% 54% 71% 67% 56% Social Assistance (%) 17% 61% 60% 27% 56% Monthly Disposable Income per Capita: Roma (€) B 62 190 102 35 125 General Population (€) C 148 398 285 134 291 Roma Relative to General 42% 48% 36% 26% 43% Population (%) Sources: UNDP/World Bank/EC regional Roma survey (2011) for Roma, EU SILC (Eurostat, 2008) for general population GDP per capita and Finance Statistics; Eurostat, 2011 for nominal exchange rates. A Figures represent the share of Roma- / Non-Roma individuals living in households where in the last month, one or more household members went to bed hungry because there was not enough money for food. B Monthly income calculated in current prices using the July 2011 nominal exchange rate; C Country average of household disposable income. Regarding ethnic segregation, a mixed picture emerges with considerably segregation but not exclusively so. Most sampled Roma households in Hungary, Romania, and Slovakia live in rural communities, while in Bulgaria approximately half do, and in the Czech Republic virtually all surveyed Roma live in urban neighborhoods. Apart from the Czech Republic, at least two thirds live more than 3 km away from the nearest city or city center. On the other hand, in two of the five countries – the Czech Republic and Romania – is a slight majority of the sampled Roma households located in predominantly Roma communities. In the other three countries, a substantial minority of households is located in such ‘segregated’ neighborhoods, but most Roma live in mixed neighborhoods in these countries. In addition, 16 households where Romani is spoken as the main language in the home are far from universal. Although a majority of Roma in Bulgaria and Slovakia speak Romani at home, and almost half in the Czech Republic and Romania do, this holds for only 7% of all households in Hungary. TABLE 3: SEGREGATION CHARACTERISTICS - % OF ROMA HOUSEHOLDS Czech Bulgaria Hungary Romania Slovakia Republic Household (N=3,785) Rural Community 48 2 66 61 72 Predominantly Roma community or neighborhood 32 60 40 54 42 Romani Spoken at Home 64 44 7 41 67 Nearest City or City Center More than 3 km 71 40 73 77 68 Source: UNDP/World Bank/EC regional Roma survey (2011) 1.4 LABOR MARKET OVERVIEW Increasing the extremely low employment rates among Roma is an economic necessity for the countries in Eastern Europe and a priority for the EC. As many as 7–20 percent of new labor market entrants in Bulgaria, the Czech Republic, Hungary, Romania, and Slovakia are Roma (see table). Majority populations throughout much of Central and Eastern Europe are declining while the Roma population continues to be young and growing. For example, in Roma communities in Bulgaria, the Czech Republic, Hungary, Romania, and Slovakia, those aged 0-14 years old – the next generation of working-age people – make up 27%-to-38% of the Roma populations. In contrast, among the general populations in these countries, those aged 0-14 years old only make up 13%-to-15% of the population. In other words, in each of these countries, new labor market entrants are 2 to 3 times as likely to come from Roma communities as their total population share would suggest. Yet their employment rates continue to lag far behind those of the majority populations, and behind the Europe 2020 headline target of 75 percent of the population aged 20–64 employed. TABLE 4: % OF ROMA AMONG NEW LABOR MARKET ENTRANTS % Roma all % Roma ages Roma Ratio Roma % Roma % Roma population according new Roma-to- population (all all ages new labor (all ages) to non- labor general ages) according market according to census market population according to to census entrants non-census estimates entrants 0-14 years census (1) (1) (1) (2) (2) (2) Bulgaria 2.0 370,000 4.7% 9% 700,000 9.0% 18% Czech Republic 2.8 70,000 0.7% 2% 250,000 2.4% 7% Hungary 2.4 189,984 1.9% 5% 550,000 5.3% 13% Romania 2.4 535,140 2.5% 6% 1,800,000 8.3% 20% Slovakia 2.5 128,242 2.3% 6% 320,000 5.9% 15% Notes: A concern with census based estimates that Roma substantially underreport their ethnicity. An alternative set of expert estimates is commonly reported (e.g. European Commission (2011) Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, An EU Framework for National Roma Integration Strategies up to 2020, and United Nations Development Programme (2002), ‘Avoiding the Dependency Trap’. 17 There are very few Roma who have (formal or informal) jobs, and exceptionally many Roma who actively look for jobs but cannot find them. The figures and tables below compare labor market outcomes between Roma and the general population. Among the working-age population, participation in the labor market is somewhat lower among Roma than among the general population. However, differences in employment 4- and unemployment rates are much larger. Among Roma, employment rates vary from as little as 20% for men and 9% women in Slovakia to slightly more than 40% for Roma men in Bulgaria, the Czech Republic, and Romania, and 26% for Roma women in Bulgaria. In comparison, employment rates among men from the general population vary between 60% in Hungary to 74% in the Czech Republic, and among women from 51% in Hungary to 56% in Bulgaria and the Czech Republic. For the EU-27 as a whole, the figures are respectively 70% for men and 58% for women. Unemployment rates among Roma vary from 28% in Hungary to 64% in Slovakia for men, and 44% in Romania and 87% in Slovakia for women. FIGURE 2: EMPLOYMENT RATES FOR MEN AND WOMEN a. Men b. Women 80 74 70 80 63 66 65 70 60 70 56 56 58 60 60 51 52 52 50 42 43 42 50 40 34 40 26 30 20 30 19 19 20 20 13 9 10 10 0 0 General Population Roma General Population Roma Source: UNDP/ World Bank/EC regional Roma survey (2011); 2010 Labor Force Survey (Eurostat, 2011) for the general population. Sample restricted to working-age population (15-64). TABLE 5. LABOR MARKET STATISTICS FOR ROMA: MEN Czech Bulgaria Hungary Romania Slovakia EU- Republic 27 Roma All Roma All Roma All Roma All Roma All Employment rate (%) 42 63 43 74 42 60 34 66 20 65 70 Participation rate (%) 64 71 63 79 57 68 61 72 57 76 78 Unemployment rate (%) 35 8 33 7 28 8 44 8 64 13 10 Source: Rome employment levels: UNDP/ World Bank/EC regional Roma survey (2011); general population: 2010 Labor Force Survey (Eurostat, 2011). Working-age population (15-64). 4 The employment rates follow the official ILO definition and thus include informal employment. 18 TABLE 6. LABOR MARKET STATISTICS FOR ROMA: WOMEN Czech EU- Bulgaria Hungary Romania Slovakia Republic 27 Roma All Roma All Roma All Roma All Roma All Employment rate (%) 26 56 19 56 13 51 19 52 9 52 58 Participation rate (%) 50 62 37 62 34 57 33 56 47 61 65 Unemployment rate (%) 47 5 48 5 61 6 44 6 87 8 8 Source: Rome employment levels: UNDP/ World Bank/EC regional Roma survey (2011); general population: 2010 Labor Force Survey (Eurostat, 2011). Working-age population (15-64). These low employment rates do not reflect preferences: the vast majority of Roma express a desire for stable jobs, similar to the responses by non-Roma neighbors. Consistent with the finding that Roma with jobs report greater levels of happiness and life satisfaction, between 71 and 92 percent of Roma adults 16 years and older report preferring “Secure employment but low paid� instead of “Having a higher income but insecure and irregular�. These responses are very similar to the responses by non-Roma neighbors. Similarly, between 61 and 90 percent of Roma across the five countries prefer “Having secure employment but having to be at work 8 hours a day, 5 days a week, and not having the freedom to manage your time� compared with “Having irregular employment but being free to manage your time�, again a pattern almost equal to that among non-Roma neighbors. For those Roma with a job, informal employment is common, particularly in Romania and Bulgaria. As measured by the proportion of employees working without a formal contract, the table below illustrates that over 40% of employed Roma in Romania, and almost 40% in Bulgaria work informally. This proportion falls to less than a quarter in the Czech Republic, and to even slightly lower rates in Hungary and Slovakia. When measuring informality based on the absence of social contribution payments, similar results are found in most countries. In Romania, this definition yields a much higher rate of informal employment. TABLE 7. INFORMAL EMPLOYMENT: WRITTEN CONTRACT AND SOCIAL CONTRIBUTIONS (% ROMA) Czech Bulgaria Hungary Romania Slovakia Republic Written contract (%) A 62 78 83 58 85 Pension and/or health care contributions (%) B 60 80 87 40 78 Source: UNDP/World Bank/EC regional Roma survey (2011). A Sample restricted to working-age employees in private and public enterprises (i.e. excluding self-employed); B Sample including all employed working-age respondents. Extremely low education levels among Roma men and even lower among Roma women are a central cause for these low employment rates. Although some improvements in levels of education have been made in the past few years (World Bank, 2010), education levels among surveyed Roma aged 25 to 64 remain extremely low. The table below shows that a majority of Roma has had some education, but only a minority has finished upper secondary education. Among men it ranges from 12% in Romania to 29% in the Czech Republic, while among Roma women the upper secondary completion rates are even lower; between 6% in Romania and 21% in the Czech Republic. 19 TABLE 8: HIGHEST INDIVIDUAL EDUCATION LEVEL COMPLETED (%) Czech Bulgaria Hungary Romania Slovakia Republic M F M F M F M F M F Below lower secondary (ISCED1) 37 49 14 16 21 32 54 67 16 19 Lower secondary (ISCED 2) 48 41 57 62 61 57 34 27 63 66 Upper secondary (ISCED 3) 15 10 29 21 18 10 12 6 20 14 Tertiary (ISCED 4+) 0 1 0 0 0 0 0 0 0 0 Source: UNDP/World Bank/EC regional Roma survey (2011). Sample restricted to individuals aged 25-64. In all five countries, a substantial share of Roma adults report that they have experienced discrimination because of their ethnicity. Although there is considerable variation between countries, self-reported ethnic discrimination is very high across the board. Discrimination is reported to occur in various realms: ranging from education to healthcare, housing and the labor market. Data on discrimination among entrepreneurs specifically are not available; neither are there data on ethnic discrimination in the financial sector, e.g. among Roma applying for loans. FIGURE 3: SELF-REPORTED ETHNIC DISCRIMINATION OVER THE LAST 5 YEARS AMONG ROMA ADULTS (AGED 15+) A. Overall Ethnic Discrimination B. Ethnic Discrimination among Job Seekers 70 61 100 78 Percent (among Job 60 54 73 43 80 50 55 Percent 35 60 45 40 Seekers) 26 30 30 40 20 10 20 0 0 Source: UNDP/World Bank/EC regional Roma survey (2011). Sample restricted to individuals aged older than 14. Only one randomly selected adult per household was sampled for this question. 20 2 FINANCIAL EXCLUSION AMONG ROMA This chapter investigates access to a broad range of financial services for Roma. The background chapter underscored the poor living conditions and high levels of economic insecurity among Roma surveyed in Eastern Europe. Poor households – regardless of ethnicity - need a broad range of financial services just as much as, if not more than, non-poor households. For example, poor people need access to savings products to protect themselves against inherent income vulnerabilities, or need savings to build up a down payment for a (micro-) loan. 5 Similarly, financial literacy can help poor households better plan for the limited resources they have. For these reasons, this chapter first goes beyond micro-credit and explores access to a broader range of basic financial services. 2.1 WHAT IS FINANCIAL EXCLUSION? Financial inclusion refers to adequate access to financial services for all households and businesses. Financial inclusion can be defined as the ability of all businesses and individuals to access formal financial services that meet their needs (v. Carbo et al., 2005: xv; Sinclair, 2001). Financial exclusion, therefore, is the inability of businesses and individuals to access formal financial services due to “price or non-price barriers� (World Bank, 2008: 27). From the point of view of households and businesses, these barriers are related to physical access to financial intermediaries, eligibility, affordability, and availability and awareness of appropriate products and services (World Bank, 2006a; Carboni et al. 2010). Financial exclusion partly exists because small-scale financial services are – in relative terms – more costly to provide than larger-scale services. All types of financial services have some ‘fixed cost’ incurred by the provider. Fixed costs arise, for instance, because even a deposit or savings account with no deposits on it at all needs to be administered, or because processing a loan applicant, disbursing the loan, and administering repayment all require some resources regardless of the loan size, making it relatively cheaper for financial service providers (FSPs) to supply large loans or bank accounts for large sums of money (World Bank, 2008: 32). Another factor that gives rise to financial exclusion is information asymmetry. Information asymmetries are perhaps the most important friction in credit markets. Before the transaction, financial intermediaries may find it difficult to assess – in the case of loans – the credit-worthiness of potential borrowers. To lessen such effects, FSPs need to monitor borrowers, requiring the use of resources. However, in anticipation of the costs this implies, they may decide not to provide their services to groups which are deemed to either be exceptionally risky, or exceptionally hard and costly to monitor. Generally, existing formal businesses are easier for FSPs to serve, because they possess some sort of formal documentation – providing information on, for example, their banking history and their credit-worthiness, as compared to start-ups and informal businesses. Financial exclusion can also be deeply rooted in non-economic challenges. Self-exclusion, for instance, can be caused by minorities not entering credit markets in anticipation of discriminatory behavior by financial intermediaries. Thus, the specific causes of financial exclusion, their relative importance, and hence, the nature of financial exclusion itself, often remain unknown (v. Pytkowska et al., 2008; World Bank, 2008). 5 For a discussion on microcredit vis-à-vis financial inclusion broadly, e.g. see an August 19, 2011 interview with Mr. Tilman Ehrbeck, the CEO of CGAP: http://www.cgap.org/gm/document- 1.9.53154/Tilman_Ehrbeck_Transcript.pdf. 21 2.2 PHYSICAL AND ELECTRONIC ACCESS TO FINANCIAL SERVICES This section explores the extent of financial access and financial usage by Roma and the general population in the five countries. The two most concrete macro-level measures of financial inclusion are the physical outreach of FSPs and the usage of financial services by the population, i.e. by both households and businesses. Each year, the IMF surveys central banks or other regulatory institutions to gather a wide range of data on both depository and non-depository institutions to evaluate these two dimensions in the Financial Access Survey (IMF, 2010). 6 The World Bank Consultative Group to Assist the Poor (CGAP) performs a similar survey; the Financial Access Indicators (CGAP, 2010). The physical outreach of depositary institutions’ retail networks in the five countries lags behind that of the rest of Europe. Table 9 shows that there are fewer depository institution branches and ATMs in the Czech Republic, Hungary, Romania and Slovakia per adult compared with the EU average. 7 Moreover, they are distributed twice less densely. Bulgaria is an exception: its commercial branch network is denser than the European average. Moreover, outreach is only an average measure that can hide important rural-urban divides. For example, in Hungary and Slovakia, there are on average 16 and 25 commercial bank branches per 1,000 km2, but only 10 and 17 branches per 1,000 km2 if the three largest cities are excluded. The retail networks are not becoming more widespread. Data on depository institutions for Hungary show that between 2003 and 2010 the outreach of the branch retail network has not changed apart from some small year-to-year variations. Romania shows a similar pattern between 2008 and 2010. Data for Bulgaria and the Czech Republic for commercial banks show only a minor increase. In contrast, the branch retail network in Europe is on average becoming less widespread. This could reflect a shift towards more ATMs and POS terminals for withdrawal-, deposit-, and payment services, and a shift towards online banking. The data available suggests indeed an increase in the outreach of ATMs - from 86.8 to 93.5 ATMs per 1,000 km2 between 2007 and 2010. Large gains in ATMs density are observed in all five countries, although the absolute density remains low compared to Western Europe. TABLE 9: RETAIL NETWORK PHYSICAL OUTREACH Czech Bulgaria Hungary Romania Slovakia EU C Republic n.a. n.a. 38 34 28 61 D Branches per 100,000 adults (91) (23) (17) (33) (27) (36) E n.a. n.a. 34 27 26 62 F Branches per 1,000 km2 (55) (26) (16) (26) (25) (35) G B ATMs per 100,000 adults 81 42 57 56 50 93 H 2 B ATMs per 1,000 km 49 49 54 44 48 94 POS terminals per 100,000 adults 790 869 831 540 713 n.a. POS terminals per 1,000 km2 480 1,007 792 428 676 n.a. Source: Financial Access Survey (IMF, 2010) for branches and ATMs; Financial Access Indicators (CGAP and the World Bank, 2010) for POS terminals. Data includes all depository corporations. Financial corporation headquarters are excluded from the calculation. Data in brackets refer to commercial banks only. B 2009 data. C Authors’ calculations. Land area and adult population data from the World Bank Development Indicators 2010; D-I Data not available for all EU countries. 6 Since this data is at the aggregate level, important within-country differences are not taken into account. 7 The data presented here concerns all depository institutions other than the central bank (ODCs), defined as “all resident financial corporations and quasi-corporations […] mainly engaged in financial intermediation and that issue liabilities included in the national definition of broad money� (Financial Access Survey, Explanatory Notes, 2010, p.1). 22 Many fewer Roma households live within 3 kilometers from the nearest bank branch. The table and figure below show that the proportion of Roma households living less than 3 km away from the nearest bank branch is as high as 69% in the Czech Republic and as low as 35% in Slovakia. The Czech Republic and Hungary are the only two countries where most Roma households live less than 3 km away from the nearest bank. In all five countries, the average household (EU SILC 2008) reports to be located closer to a bank branch than the average Roma household in the regional Roma survey (2011). Moreover, a sharp divide exists between Roma households in urban neighborhoods and in rural communities. Slovakia – with 81% of Roma households in urban neighborhoods but only 18% of those in rural communities living less than 3 km away – provides the most extreme example. This implies that physical access to banks may be a substantial problem for Roma households, especially those in the country-side. FIGURE 4: % HOUSEHOLDS LIVING LESS THAN 3KM AWAY FROM A BANK 100 91 92 85 79 80 69 64 60 60 47 44 35 40 20 0 Bulgaria Czech Hungary Romania Slovakia Republic General population Roma households TABLE 10. % HOUSEHOLDS LIVING LESS THAN 3 KM AWAY FROM A BANK Czech Bulgaria Hungary Romania Slovakia Republic General population 91 79 92 85 64 Roma households 47 69 60 44 35 Roma households in urban 73 70 79 73 81 neighborhoods Roma households in rural 19 50 50 26 18 communities Source: UNDP/World Bank/EC regional Roma survey (2011) for Roma household data, and EU SILC (Eurostat, 2008) for general population data. For Roma, the data refers to the proportion of households living less than 3 km away from the nearest bank branch. For the general population, the data refers to the proportion of households living close to a bank branch, though the term ‘close’ is left unspecified in the questionnaire administered to households. Thus, survey design differences do not allow a strict comparison between the data. Although a majority of Roma households possess either a landline or a mobile phone, the home usage of computers and internet is much more limited. With the increasing digitalization of banking services, possession of phones, computers and Internet increasingly allows households to access formal financial services. In all five countries, a majority of Roma households possess a mobile phone, a landline, or both. By contrast, only a quarter or less of Roma households possess a computer, and a similar share of Roma adults knows how to use a Word processing program. The number of households with an Internet connection also lags behind the number of households with a computer at home. Hence, internet services would most likely have only a marginal effect in the accessibility of financial services to Roma, unless households are accessing Internet at public or private providers outside the home. The table below also 23 shows that ownership of phones and computers is much lower among Roma households than the general population. TABLE 11. POSSESSION OF PHONES, COMPUTERS, AND INTERNET CONNECTION (%) Czech Bulgaria Hungary Romania Slovakia Republic Roma All Roma All Roma All Roma All Roma All Mobile phone/landline 62 92 76 97 70 91 65 80 65 95 Computer 16 34 24 56 27 50 16 32 18 51 Household Head Able to use Word Processing 7 n.a. 24 n.a. 20 n.a. 10 n.a. 19 n.a. Program Adults Able to Use Word 12 n.a. 24 n.a. 28 n.a. 15 n.a. 25 n.a. Processing Program Internet connection 13 n.a. 19 n.a 19 n.a 9 n.a 13 n.a. Source: UNDP/World Bank/EC regional Roma survey (2011) for Roma household data, and EU SILC (Eurostat, 2008) for general population data. 2.3 USAGE OF FINANCIAL SERVICES When compared to the general population, Roma households make much less use of current (checking) accounts and credit cards. The EU average for usage of current accounts is 87%, meaning that almost 9 in every 10 households have a current account in the EU. The figures below show that in Eastern Europe, the rates are generally lower, but particularly so among Roma households. Hungary and the Czech Republic stand out with a relatively high usage of current accounts among Roma households, reaching almost 50% of households in the first case. At the opposite end, fewer than one in 20 Roma households in Bulgaria have a current account. The overall EU average for usage of credit cards stands at 45% of all households. However, even among the general population in the five countries, credit- or store cards are not very widespread, reaching 20% of households in Slovakia. Usage of credit cards among Roma households is even lower. Less than 4% of Roma households in Bulgaria, Hungary, and Romania have a credit card. FIGURE 5: USAGE OF CURRENT ACCOUNTS AND CREDIT CARDS (%) A. Current Accounts B. Credit Cards and Store Cards 90 79 100 77 90 80 73 70 80 60 70 47 Percent Percent 60 50 35 50 40 29 40 30 23 30 16 18 20 20 20 12 10 11 4 6 8 8 10 10 2 3 3 0 0 Bulgaria Czech Hungary Romania Slovakia Bulgaria Czech Hungary Romania Slovakia Republic Republic Roma General Population Roma General Population Source: UNDP/World Bank/EC regional Roma survey (2011) for Roma household data, and EU SILC (Eurostat, 2008) for general population data. 24 Usage of savings accounts and payment-related services is also extremely low among Roma households. Although no data is available for comparison with the general population of these countries, the table below shows that savings accounts and term deposits are almost absent among Roma households. In all countries but Slovakia, less than 3% and 1% of Roma households in communities use savings accounts and term deposits, respectively. These figures are only marginally higher for Slovakia – 5% and 2%, respectively. Money transfer services are also used little in all five countries. By contrast, usage is higher for debit cards – particularly in Bulgaria and Hungary where over a third of Roma households use them. TABLE 12. USAGE OF SAVING ACCOUNTS AND PAYMENT SERVICES (%) Czech Bulgaria Hungary Romania Slovakia Republic Saving Account 2 3 2 2 5 Term Deposits 1 0 1 1 2 Money Transfer Services 5 7 4 2 2 Debit/Payment Card 36 19 36 8 10 Source: UNDP/World Bank/EC regional Roma survey (2011). Roma are much less likely to use formal financial instruments than their non-Roma neighbors, even after controlling for background characteristics such as income. The table below shows that Roma households are 17 percentage points less likely than their non-Roma neighboring households in similar communities to have a current account, 10 percentage points less likely to have a savings accounts, and 19 percentage points less likely to have a debit card. 8 This model accounts not only for background characteristics like income, household size, and whether or not the household receives any income from employment, but also for the language spoken at home. Although smaller than the correlation with Roma ethnicity itself, households speaking Romani at home are significantly less likely to use banking services such as current accounts (7 percentage points), savings accounts (2 percentage points), and debit cards (4 percentage points). Credit card usage is also lower among Roma households, with the latter being 9 percentage points less likely to use credit cards than non-Roma neighbors. In sum, geographic- and basic household-level background characteristics do not - or at least not fully - account for the lower usage of banking services among Roma. 8 The econometric specification of these models is of the form: usage=constant +b1*Roma households+b2*Romani spoken at home+b3*enumeration area+b4*control variables where each household represents an observation. The coefficients b1 and b2 are interpreted as the change in the probability of the household using the specific financial services associated with being a Roma household and speaking Romani at home, respectively. Annex B provides the full model. 25 TABLE 13. CORRELATES OF HOUSEHOLDS’ USAGE OF BANKING SERVICES Current Savings Debit Card Credit Card Account Account (N=5,202) (N=5,191) (N=5,296) (N=5,186) -0.17*** -0.10*** -0.19*** -0.09*** Roma Household (0.02) (0.01) (0.02) (0.01) -0.07*** -0.02* -0.04** -0.03** Romani Spoken at Home (0.02) (0.01) (0.019) (0.01) Enumeration Area Fixed Effects Yes Yes Yes Yes Household Controls Yes Yes Yes Yes Adjusted-R2 0.45 0.21 0.38 0.25 F-statistic 41.11*** 13.99*** 48.71*** 11.70*** Source: UNDP/World Bank/EC regional Roma survey (2011). Cluster-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1. Household controls included in the model are the household head’s gender, age, and education level, home ownership, household size, household income quintile (measured by country and ethnic group), and whether the household receives at least some income from employment. Income quintiles were based on Adult Equivalent Income levels. Usage of financial services among Roma households is strongly correlated with the household’s head education, whether Romani is spoken at home, dwelling ownership, and (employment) income. The regression estimates in Annex Table 4 show that households whose ‘head’ has completed at least secondary education are 10 percentage points more likely to have a current account. Those with income from employment are almost 10 percentage points more likely to use a current account. On the other hand, being close to a bank only slightly increases the likelihood of using a current account, and although Roma households are often remotely located, ‘not living in a city’ does not decrease the likelihood of using financial services. For using savings accounts, debit cards and credit cards, the results are similar; determinants that have the biggest impact are income levels and whether or not the household has earnings from employment. 2.4 SAVINGS AND VULNERABILITY For poor households being able to save in a secure way is perhaps the most crucial form of financial inclusion since this is an important means of increasing economic security. Savings allow people to smooth consumption and respond to income shocks, but also to borrow money, increasing their potential for self-employment. For poor households, financial exclusion often implies a complete lack of secure means to save, or forces these households to make use of expensive alternatives to regular banking services. Nonetheless, even under these far from ideal circumstances the poor still can and do save. “The evidence generated thus far on the impact of access to formal savings is limited but very promising," (Karlan, 2011). The ability of a Roma household to face unexpected expenses is also associated with the household’s use of banking services, and especially the use of savings accounts. The regression estimates in Annex Table 5 show that there is a strong correlation between usage of formal financial services and the household’s ability to cope with unexpected expenses; Roma households with a savings account are over 25 percentage points more likely to be able to cover unexpected expenses than Roma households without savings accounts. These effects are robust to different model specifications - comparing Roma households living in the same community or including households’ background characteristics does not change the coefficients substantially. The usage of a current account is only significantly correlated to the likelihood 26 of being able to face unexpected expenses when households in the same neighborhood are compared (Model 2). Once background characteristics are also taken into account, the correlation disappears (Model 3). The same holds for the usage of credit cards. In short, these models (unsurprisingly) suggest that financial services are being used to mitigate the impact of unexpected expenses (i.e. smooth consumption over time) – whether this requires borrowing against future income, or drawing on savings accumulated in the past. For all households, including poor households, savings are an important means of increasing economic security, but few households have savings. Savings allow people to smooth consumption and respond to income shocks, but also to borrow money, increasing their potential for self-employment. For poor households, financial exclusion often implies a complete lack of secure means to save, or forces these households to make use of expensive alternatives to regular banking services. As shown above, however, Roma households have savings accounts only in 5% or less of all cases. In fact, very few Roma households indicate that they have savings. Roma households have no savings at all in about 9 out of 10 cases in the Czech Republic and Slovakia, and in over 19 out of 20 cases in the other three countries. In comparison to non-Roma neighboring households, these levels of savings are extremely low. Although non-Roma neighbors do not often save either, rates among Roma are still only about a quarter of the rates for non-Roma living in the vicinity. FIGURE 6. % HOUSEHOLDS WITH SAVINGS AMONG ROMA AND NON-ROMA NEIGHBORS 70 60 60 50 43 Percent 40 30 20 20 13 15 14 9 10 4 3 3 0 Bulgaria Czech Hungary Romania Slovakia Republic Roma Non-Roma Source: UNDP/World Bank/EC regional Roma survey (2011) Among Roma households that do save, the amounts of savings are very low, and most do not save in formal bank accounts. The table below shows that among the few Roma households which are able to save, the average reported number of months during which the household could live on its savings as the sole source of income never exceeds 5, i.e. is always less than half a year. This suggests high vulnerability in dealing with unexpected shocks such as the breadwinner’s job loss. For non-Roma households in the vicinity, these figures (not shown) are higher in all countries: about twice as high even in the Czech Republic, Hungary and Slovakia. Moreover, the share of Roma households with savings that has a formal bank account never exceeds 40%. Again, among non-Roma neighbors, saving ‘formally’ is much more common, with the share of ‘saving households’ that possesses a bank account being as much as 5 times as high among non-Roma households as compared to Roma in the Czech Republic. 9 9 Savings comparisons with the general population are not possible because this information is not available in the EU SILC 2008. 27 TABLE 14. USAGE OF BANKING SERVICES: ROMA HOUSEHOLDS USING SAVING ACCOUNTS AND PAYMENT SERVICES Czech Bulgaria Hungary Romania Slovakia Republic Household has Savings (%) 4 13 3 3 9 Number of Months the Household Could Rely on Savings A 4 3 2 5 3 Household with Savings has a Savings Account (%) B 29 9 18 40 33 Source: UNDP/World Bank/EC regional Roma survey (2011). A Conditional on household having some form of savings. B Percentages of Roma households with savings (including e.g. cash, bank deposits and highly valued commodity items like gold), which also has a formal savings account. When controlling for background characteristics, Roma households remain 17 percentage points less likely to save at all as compared to non-Roma neighboring households. This implies that issues not controlled for such as financial literacy or discrimination in the banking sector may play a role, limiting Roma households’ ability to accumulate savings. Whether Romani is spoken as the main language at home is not correlated with the likelihood of having savings. TABLE 15. COMPARISON OF ROMA AND NON-ROMA HOUSEHOLDS’ SAVINGS Savings (N=5,525) Roma Household -0.17*** (0.02) Romani Spoken at Home -0.01 (0.02) Enumeration Area Fixed Yes Effects Household Controls Yes Adjusted-R2 0.30 F-statistic 27.44*** Source: UNDP/World Bank/EC regional Roma survey (2011). Cluster-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1. Household controls included in the model are the household head’s gender, age, and education level (at least secondary education or less), home ownership, household size, and household income quintile (measured by country and ethnic group). Finally, not only do Roma save less, they are also more likely to have arrears on utility bills than the general population. This is shown in the table below; between 15% of Roma households in Slovakia and 46% of Roma households in Hungary have such arrears, as compared to slightly lower levels in Bulgaria and Romania, and much lower levels in the remaining three countries among the general population. TABLE 16. ARREARS ON HOUSEHOLD MORTGAGES, LOANS, AND UTILITY BILLS (%) Czech EU- Bulgaria Hungary Romania Slovakia Republic 27 Roma All Roma All Roma All Roma All Roma All Utility Bills 38 32 34 2 46 0 29 23 15 3 1 Source: UNDP/World Bank/EC regional Roma survey (2011) for Roma household data, and EU SILC (Eurostat, 2008) for general population data. Bills for any housing related utilities (e.g. electricity and water supply, heating or phone bills. 28 And, compared to their non-Roma neighbors, Roma households are also much more likely to have arrears after controlling for background characteristics. The regression estimates in the table below show that a Roma household is 21 percentage points more likely to have arrears than its non-Roma neighbors, even when comparing households within the same neighborhoods (by accounting for enumeration area fixed effects) and controlling for household characteristics. TABLE 17: COMPARISON OF ARREARS BETWEEN ROMA AND NON-ROMA NEIGHBORS Utility, Tax, Education, and Utility Payment Health Payment Arrears Arrears (N=5,553) (N=5,553) Roma Household 0.21*** 0.19*** (0.02) (0.02) Romani Spoken at Home 0.00 0.00 (0.02) (0.02) Enumeration Area Fixed Yes Yes Effects Household Controls Yes Yes Adjusted-R2 0.29 0.27 F-statistic 41.02*** 38.71*** Source: UNDP/World Bank/EC regional Roma survey (2011). Cluster-robust standard errors in parentheses. OLS Estimates. *** p<0.01, ** p<0.05, * p<0.1;Household controls included in the model are the household head’s gender, age, and education level (at least secondary education or less), home ownership, household size, and household income quintile (measured by country and ethnic group). In short, this chapter demonstrates that Roma lack access to a wide range of financial services, including basic current accounts and savings accounts. As highlighted in the introduction and underscored by the finding above that less vulnerable Roma households take advantage of financial services more so than vulnerable Roma households, this finding is concerning since poor households need a broader range of financial services just as much as, if not more than, non-poor households. As such, improving access to a wider range of financial services, savings mechanisms especially, but also financial literacy and business skills training, should be promoted. In fact, as the next chapter two chapters will demonstrate, such a comprehensive financial inclusion strategy is also necessary to more narrowly expand access to microcredit. The final chapter discusses examples of ways in which countries are promoting this broader financial inclusion agenda. 29 3 ROMA SELF-EMPLOYMENT AND ENTREPRENEURSHIP Before focusing more specifically on access to microcredit (Chapter 4), this Chapter 3 will first review self-employment patterns among Roma more generally, including the types of businesses Roma entrepreneurs operate, the interest among aspiring Roma entrepreneurs to start a (formal) business and become self-employed, as well as business barriers – credit and non-credit – reported by the World Bank’s Doing Business survey for each of the countries as well as the experiences reported by surveyed entrepreneurs themselves. 3.1 SELF-EMPLOYMENT RATES Self-employment ranges from 3-19 percent among Roma working age men and 0-7 percent among Roma working-age women. Individuals are classified as self-employed according to a broad definition that includes non-registered businesses. In particular, people are considered self-employed if they are members of producers’ cooperatives, individuals working in family businesses, as well as those reported to be working on their own account or as employers. Respondents with occasional jobs as scrap collectors or as unskilled workers lacking a formal contract are included under this definition. As shown in the figure below, self-employment among Roma is most common in Romania, where 19% of the Roma working-age men and 7% of Roma working-age women are self-employed. Many of these, however, are likely to operate unregistered businesses as only few – less than 4% in each of the countries - report operating a business. Note that in all countries, employed Roma are more likely to work as an employee in a private or public corporation than as a self-employed person. FIGURE 7. % SELF-EMPLOYED AND EMPLOYED INDIVIDUALS AMONG THE WORKING AGE POPULATION A. Men B: Women 100 100 80 80 60 60 10 4 19 40 2 40 4 3 1 7 20 39 20 1 0 32 32 23 22 17 18 12 12 9 0 0 Bulgaria Czech Hungary Romania Slovakia Bulgaria Czech Hungary Romania Slovakia Republic Republic Employed by employer Self-employed Employed by employer Self-employed Source: UNDP/World Bank/EC regional Roma survey (2011). Self-employment figures are according to the broad definition explained at the start of this section. The share of self-employed among working Roma is considerably lower than among the general working population in the Czech Republic, Hungary, and Slovakia, while noticeably higher in Bulgaria and Romania. The figures below illustrate that, among those male Roma who are employed, the proportion of self-employed individuals – according to the broad definition outlined above – ranges from 5% in Hungary to 45% in Romania, compared with 14 to 26% for the populations as a whole. Similar to the figures for men, women in Bulgaria and Romania also have the highest shares of Roma 30 entrepreneurs. For women, the range among Roma in all countries is 1-34%, whereas it is 8-12% among the general population. FIGURE 8. SHARE OF SELF-EMPLOYMENT AMONG THE EMPLOYED, ROMA AND GENERAL POPULATION A. Men B: Women 50 45 50 Percentage of self-employed Percentage of self-employed among employed women 40 40 34 among employed men 30 24 26 30 21 21 19 20 14 15 20 13 12 9 10 9 11 8 8 10 10 5 10 6 4 1 0 0 Roma General Population Roma General Population Source: UNDP/World Bank/EC regional Roma survey (2011) for Roma, and Labor Force Survey (Eurostat, 2011) for the general population. 3.2 TYPES OF BUSINESSES The vast majority of self-employed Roma have one-(wo)man’s businesses, but there is also a substantial proportion of Roma businesses with additional employees. In Hungary, family businesses account for 50% of all self-employment. In the other countries, family businesses account for a much smaller, though still substantial, share. These figures imply that the average size of Roma businesses is very small, which possibly points to barriers for business expansion among Roma entrepreneurs. At the same time, there are also Roma businesses with additional employees, especially in the form of family businesses. TABLE 18. TYPES OF BUSINESSES AMONG SELF-EMPLOYED ROMA Czech Bulgaria Hungary Romania Slovakia Republic Among Self-Employed: Self-employed on own account 85 78 45 92 93 Employer 1 0 0 0 0 Member of producers’ 4 2 5 2 3 cooperative Work in family business 10 20 50 5 3 Source: UNDP/World Bank/EC regional Roma survey (2011). In Bulgaria and Romania, (close to) 90% of self-employed Roma work in businesses with less than 5 workers. In the other countries businesses run by Roma have a bigger average size. In Hungary, less than 50% of self-employees run businesses with less than 6 workers. In fact, Hungary, Slovakia and the Czech Republic show a more widespread distribution, with approximately a quarter or more of all Roma businesses employing more than 10 workers. However, recall that self-employment rates in these three countries are extremely low. In order to provide a more general frame of reference, Box 1 outlines to what 31 extent SMEs – regardless of the ethnicity of the owner - contribute to employment and value added in the economies of the five countries analyzed. TABLE 19. BUSINESS SIZE FOR SELF-EMPLOYED ROMA Czech Bulgaria Hungary Romania Slovakia Republic 1-5 90 62 46 87 36 6-10 6 15 31 5 18 11-20 3 13 15 1 0 21-50 1 3 0 2 36 51-250 0 3 8 4 0 >250 0 5 0 1 9 Source: UNDP/World Bank/EC regional Roma survey (2011). BOX 1: THE CONTRIBUTION OF SMES TO NATIONAL EMPLOYMENT AND VALUE ADDED IN EASTERN EUROPE The contribution of SMEs to employment varies considerably across the five countries. Data on non- financial enterprises (and not engaged in agriculture, forestry and fishing, in public administration or in other largely non-market services like health and education) show that in the EU-27, 29% of the labor force employed in non-financial enterprises work in micro establishments and produces 21% of value added. Bulgaria, the Czech Republic and Romania are in line with the EU in terms of employment, but contribution to value added is smaller: 14% in Bulgaria, 19% in the Czech Republic and 12% in Romania. In Hungary, more than a third of the labor force in non-financial enterprises work in micro-businesses, contributing to 18% of value added. Lastly, the figures for Slovakia are 15% and 13% respectively, mirroring the smaller importance of SMEs in the national economy. TABLE 20. BUSINESS STRUCTURE OF NON-FINANCIAL ENTERPRISES. Czech Bulgaria Hungary Romania Slovakia EU-27 Republic Contribution to Employment (%) Micro (1-9) 28 29 35 30 15 30 Small (10-49) 23 19 19 21 18 21 Medium (50-249) 23 20 16 14 23 17 SMEs (1-249) 74 68 71 67 55 67 Large (>249) 26 32 29 33 45 33 Contribution to Value Added (%) 14 19 18 12 13 21 Micro (1-9) 19 16 16 14 15 19 Small (10-49) 21 20 18 18 18 18 Medium (50-249) 54 55 52 42 47 58 SMEs (1-249) 46 45 48 58 54 42 Large (>249) 14 19 18 12 13 21 Source: SBA Fact Sheets: Bulgaria, Czech Republic, Hungary, Romania, Slovakia, (European Commission Enterprise and Industry, 2009); Excluding corporations engaged in agriculture, forestry and fishing, in public administration or in other largely non-market services like health and education. 32 Self-employed Roma mostly work in agriculture, construction, and trade. There is considerable cross- country variation in terms of the sectors. In Bulgaria, Hungary and Romania, a substantial share of these businesses operated by Roma can be found in the agricultural sector. By contrast, in the Czech Republic most businesses are in construction and transportation. In Hungary, construction plays an important role as well, just like it does in Slovakia. In general, other sectors such as commercial businesses and education play a trivial role. TABLE 21. TYPE OF INDUSTRY FOR SELF-EMPLOYED ROMA Czech Bulgaria Hungary Romania Slovakia Republic Agriculture and forestry 25 4 57 44 4 Trade 24 4 10 13 8 Construction 19 46 24 16 64 Leisure services 6 7 0 2 0 Public utilities 6 0 0 4 12 Other commercial services (hairdressing, tailoring, etc.) 6 7 0 6 4 Industry or mining 5 7 0 8 4 Other (specify) 5 0 0 4 0 Culture and arts 3 2 0 0 0 Transportation 0 22 0 1 0 Education and science 0 0 5 0 0 Non-governmental sector 0 2 5 0 0 Finances 0 0 0 0 4 Source: UNDP/World Bank/EC regional Roma survey (2011). Figures in bold represent sectors hosting >20% of self-employment in each country. 3.3 INTEREST IN BECOMING SELF-EMPLOYED AND STARTING BUSINESS A large share of Roma in the labor force report being interested in becoming self-employed and starting a business. In total, 24 to 40 percent of all working age Roma men and between 14 and 35 percent of Roma women of working age report that they are interested or may be interested in becoming self-employed and starting a business. This excludes individuals already in self-employment (broadly defined, a majority of whom operate informal enterprises). Hence, there are substantial groups of Roma who could potentially be assisted in realizing their business start-up aims. The next section therefore examines why the barriers facing those Roma wishing to start their own businesses. 33 FIGURE 9. INTEREST IN SELF-EMPLOYMENT AND STARTING BUSINESS AMONG WORKING AGE INDIVIDUALS A. Men B. Women 70 70 19 Percentage of working age Percentage of working age men women interested in self- interested in self-employment 60 60 50 10 50 40 40 4 7 employment 30 20 4 12 3 30 2 19 1 13 0 20 20 1 18 12 28 14 10 10 20 10 21 12 12 11 16 11 9 0 8 0 3 7 Yes Maybe Already self-employed Yes Maybe Already self-employed Source: UNDP/World Bank/EC regional Roma survey (2011). The share of missing values only exceeds the 10% threshold in Slovakia: 12% for men and 14% for women. Roma that report being interested in self-employment and starting a business are found among the employed, the unemployed, and those not participating in the labor force. Among women, a relatively large share of this group is inactive, whereas among men, most of those interested in self-employment do participate in the labor force. Roma men interested in self-employment are inactive in 19-30% of all cases. Roma women interested in self-employment are inactive much more often: in 36-52% of all cases. Furthermore, among men, the group of employed individuals interested in self-employment is generally larger than the group of unemployed individuals with this interest. Among women, these two groups are similar in size. Those who are currently not employed may be particularly hindered in making the jump to self-employment. The next chapter will examine this in more detail. FIGURE 10. EMPLOYMENT STATUS AMONG ROMA INTERESTED IN STARTING A BUSINESS A. Men B. Women 100 100 Percentage of women aged 20-54 Percentage of men aged 20-54 19 interested in self-employment 24 20 30 29 36 interested in self-employment 80 80 44 52 49 44 60 30 28 60 43 19 43 36 24 40 40 24 39 53 35 20 46 37 50 20 28 28 32 27 14 17 0 0 Employed Unemployed Employed Unemployed Out of the laborforce Out of the laborforce Source: UNDP/World Bank/EC regional Roma survey (2011). Interest in self-employment includes all subjects who answered ‘yes’ or ‘maybe’ when asked whether they were interested in becoming self-employed and starting a business. 34 Most Roma interested in becoming self-employed and starting a business, both current employees and unemployed, do have some work experience. The table below shows the proportion of respondents’ ‘working live’ (i.e. since the age of 15) that an individual has worked, for three categories: (1) those currently self-employed; (2) employees interested in becoming self-employed; and, (3) unemployed interested in becoming self-employed. Roma male employees interested in becoming self-employed have worked more than half of their working life in all countries, and the same holds for those who are already self-employed. Those who are currently unemployed have worked less in most countries, but the differences with employees are relatively small. Data on women shows that on average, employed Roma interested in becoming self-employed have worked around 40% of their adult working life, while unemployed women interested in self-employment have worked between 17% (Romania) and 30% (Hungary) of their lives since age 15. TABLE 22: AVERAGE WORKING TIME EXPERIENCE (%) Bulgaria Czech Republic Hungary Romania Slovakia M F M F M F M F M F Self-employed 56 55 65 55 59 51 51 39 37 37 Employees interested in self-employed 51 44 56 44 49 44 53 40 48 44 Unemployed interested in self-employed 29 24 33 25 41 30 18 17 26 19 Source: UNDP/World Bank/EC regional Roma survey (2011). Sample restricted to individuals aged 20-54 years old; Interest in self-employment includes both answers ‘yes’ and ‘maybe’; (1) Self-employed, (2) Employees interested in becoming self-employed, (3) Unemployed interested in becoming self-employed; Figures calculated by dividing the total years of work by the number of years in which the respondent could have worked since turning 15. More specifically, Roma men, and especially Roma in Bulgaria, are most interested in becoming self-employed and starting a business. Annex Table 1 shows the outcomes of an estimation model 10 in which interest in self-employment among those currently employed as employees or unemployed is being predicted based on a number of background characteristics. When all five countries are combined in one model, the probability that unemployed and currently employed Roma men are interested in self- employment is 8 percentage points higher than for Roma women. These gender differences are largest in the Czech Republic and Slovakia. The odds of being interested in becoming self-employed do not increase linearly with schooling - rather, these odds decrease in some of the countries analyzed. 11 In Bulgaria, rural residents are more often interested in self-employment than urban residents, whereas in Romania, the opposite holds. Overall, Bulgaria is the country with the highest level of interest in self-employment followed by Romania. The unemployed are more often interested in self-employment and starting a business than those who are employed, be it formally or informally. In the pooled model, being unemployed increases by 12 percentage points the odds of being interested in self-employment as compared to those who are formally employed, with the strongest effect in Hungary. Unemployment has no effect in the Czech Republic and Slovakia, however. Compared to those who are employed in the formal sector, those with informal employment are more likely to be interested in self-employment in Hungary, but in not in the other countries. Measures of whether someone possesses ‘skills relevant to employment’, such as being able to use a word processing program and having completed an adult learning course, are not correlated 10 A Probit model was used. 11 In Bulgaria, one year of extra schooling leads to a 6 percentage point decrease in the likelihood that someone is interested in self-employment. In Slovakia, this effect is twice as strong. The number of years of work experience only makes a difference in Romania, but the effect is much smaller than for years of schooling: having one extra year of work experience leads to a one percentage point increase in one’s chance to be interested in self-employment. 35 with interest in most countries. Having done an adult learning course or apprenticeship does correlate with interest in the Czech Republic and Hungary. The distance from the household’s dwelling to the nearest bank branch is only weakly correlated in Bulgaria, whereas being located further away from a municipal office does decrease one’s chance of being interested in self-employment in most countries, albeit with only a few percentage points. The ethnicity of the majority of people living in the settlement again makes no difference in most cases: only in Slovakia does living in a predominantly Roma settlement increase the chance of being interested in self-employment. 3.4 BARRIERS TO BUSINESS START-UPS BUSINESS CLIMATE The financial crisis that started in 2008 and the subsequent fall in global trade had a somewhat larger economic impact on the five countries analyzed than on the rest of the EU. Between 2008 and 2010 (the year before the regional Roma survey), real incomes declined by more than 5% in Bulgaria, and almost 7% in Hungary and Romania, compared to a 4.6% average decline in the EU-27. The Czech Republic and Slovakia experienced a fall in income more in line with the EU average: 4.7% and 5.0%, respectively. In addition, the global credit crunch diminished foreign capital inflows and tightened credit conditions. SMEs suffered more from this credit crunch than large corporations. In Bulgaria, SMEs were the most affected by the crisis. Similarly, in the Czech Republic, lower demand for goods and services worsened the performance of most SMEs. The Czech government made a priority of assisting these businesses, for example by offering guarantees and state loans (OECD, 2009). In Hungary, the number of insolvencies and bankruptcies among SMEs increased (OECD, 2009). The Hungarian government put in place measures to strengthen the capital base of businesses and increase long-run investment (OECD, 2009). An example in this regard is the New Hungary Micro Credit Program, enabling entrepreneurs to obtain micro credits, portfolio guarantees, and venture capital. The five countries examined rank lower than most other EU27 countries on the overall ease of doing business, but are not at the bottom. Each year the World Bank compiles a list of nine indicators to measure the ease of doing business in a country. The figure below shows that Slovakia had the most favorable business climate among these five countries in 2011, with a ranking of 48 out of 183 countries. Starting a new business is also relatively easy in Hungary, whereas Bulgaria, the Czech Republic and Romania have the most unfavorable business climates of the five countries. However, the ranking of overall ease of doing business is not necessarily reflected in actual self-employment rates among both Roma and non-Roma: for example, Slovakia is not the country with the highest rate of self-employment, whereas it does have the most favorable ‘doing business’ ranking. 36 FIGURE 11: EASE OF DOING BUSINESS RANKINGS (1-183): EU27 120 100 100 87 80 72 59 62 64 60 48 50 51 37 40 44 40 27 28 29 30 31 32 14 19 21 24 20 5 7 10 11 0 Source: Doing Business Indicators (World Bank, 2011). The overall ease of doing business rankings indicate the relative ease of doing business in one country as compared to other countries. The country with ranking 1 has the most favorable business climate, while the country with ranking 183 has the most unfavorable business climate in the world. When looking at more specific indicators that are relevant for SMEs, starting a business is particularly burdensome in the Czech Republic, whereas obtaining credit is the most difficult in the Czech Republic and Hungary. For instance, the paid-in minimum capital to be deposited in a bank or with a notary prior to the business registration is 31% of average GDP per capita in the Czech Republic, 15% in the OECD countries, and 0 in Bulgaria. Taxation is most favorable in Bulgaria, with an average corporate tax rate of 29%. The other four countries all have average corporate tax rates higher than the OECD average of 43%. Similarly, the number of tax payments and the time necessary to prepare, file and pay taxes are also higher. In the Czech Republic it takes on average 557 hours per year to fulfill all tax obligations, more than twice the OECD average. Finally, Bulgaria, Romania, and Slovakia perform well worldwide in terms of borrowers’ legal rights protection and depth of credit history information.12 TABLE 23. EASE OF DOING BUSINESS: INDICATORS RELEVANT FOR SMES Bulgaria Czech Hungary Romania Slovakia Republic Overall Ease of Doing Business 59 64 51 72 48 Starting a business 49 138 39 63 76 Getting credit A 8 48 48 8 24 Paying taxes 69 119 117 154 130 Source: Doing Business Indicators (World Bank, 2011). This indicator only assesses the legal rights of borrowers and availability of information when obtaining credit. It does not provide any information on access to or availability of credit. 12 To evaluate the strength of legal rights, 8 aspects related to legal rights in collateral law and 2 aspects in bankruptcy law are evaluated. Thus, the index (out of 10) indicates the number of features present in the legal system of the country under consideration. Similarly, the depth of credit history index is measured by evaluating 6 features related to the rules and practices affecting the coverage, scope and accessibility of credit information available through either a public credit registry or a private credit bureau. A one-unit difference in the index implies the presence of an additional feature. For more information, the reader should consult http://www.doingbusiness.org/methodology/getting-credit#legalRights. 37 START-UP AND OPERATING CHALLENGES BY SURVEYED ENTREPRENEURS The large majority of existing self-employed Roma and nearby non-Roma entrepreneurs in all countries report experiencing difficulties when starting and operating a business, especially lack of finance. The regional Roma survey asked those operating a formal business detailed questions about challenges faced when starting and while operating a business. Challenges were pre-coded and entrepreneurs were asked to identify up to three main challenges they faced in starting and operating a business, including the option ‘no major difficulties’. There were few such observations in Hungary (15) and Slovakia (11), even after including the neighboring non-Roma entrepreneurs (11 in Hungary and 8 in Slovakia). The table below shows the responses by Roma and non-Roma entrepreneurs jointly in light of the small sub-samples, but the responses were very similar within each country between the two groups.13 The particular problems are most likely also barriers for those interested in starting up a business. First, the table below shows that when starting a business, a large majority report difficulties. Consistent with the Doing Business findings, the Czech Republic stands out as the country with the smallest proportion of interviewed entrepreneurs – 13 percent – indicating no major difficulties in starting a business. Note that in each of the five countries, lack of finance was the most commonly reported barrier to starting their business, mentioned by 38 (Czech Republic) to 55 (Slovakia) percent of entrepreneurs. Several non-credit barriers were also frequently mentioned consistent with the Doing Business indicators. First, lack of customers was mentioned by 22 (Bulgaria) to 38 (Czech Republic) percent of entrepreneurs, taxation by 18 (Slovakia) to 30 (Czech Republic) percent, government licensing and regulations by 16 (Bulgaria) to 36 (Czech Republic) percent, and business skills by 9 (Slovakia) to 25 (Czech Republic) percent of the entrepreneurs. Other challenges – listed below – were mentioned less consistently across the countries. TABLE 24. MAIN PROBLEMS IN STARTING A BUSINESS AMONG ROMA AND NON-ROMA NEIGHBORS (%) Czech Bulgaria Hungary Romania Slovakia Republic (N=55) (N=15) (N=36) (N=11) (N=56) Lack of finance 40 38 47 44 55 Taxation 27 30 27 28 18 Lack of job specific skills 24 14 7 14 0 Lack of customers 22 38 27 28 36 Lack of business skills 18 25 13 19 9 Gov. licensing and regulations 16 36 27 33 27 Problems obtaining power or water 9 5 0 3 0 Lack of raw material 7 2 7 8 9 Lack of adequate labor 7 13 0 3 0 Household/family duties 5 4 6 6 9 Unable to work (disability, age) 2 2 0 3 0 No major difficulties 25 13 33 28 27 Source: UNDP/World Bank/EC regional Roma survey (2011). The figures are the shares of households with at least one self-employed household member, indicating for each issue whether this is one of (up to) three main problems they faced in starting a business. Figures in bold highlight issues that were mentioned in more than 20% of all cases in each country; The number of Roma entrepreneurs whose answers are included are, respectively: 31 (BG), 20 (CZ), 4 (HU), 22 (RO) and 3 (SK). 13 One exception: in the Czech Republic, lack of finance was mentioned by 15% of the Roma entrepreneurs (20 observations) and 50% of the non-Roma entrepreneurs (36 observations). 38 In the operating phase, lack of finance, lack of customers, and taxation stand out as the greatest challenges. As the table below indicates, the responses are more concentrated in the operating phase. Lack of finance is most commonly mentioned as a key challenge in Bulgaria (49 percent) and Hungary (53 percent), while lack of customers is a challenge in each of the countries, and especially Romania (63 percent). Taxation is also a challenge, particularly in Hungary (60 percent). TABLE 25. MAIN PROBLEMS OPERATING A BUSINESS AMONG ROMA AND NON-ROMA NEIGHBORS (%) Czech Bulgaria Hungary Romania Slovakia Republic (N=53) (N=15) (N=35) (N=13) (N=52) Lack of finance 49 33 53 23 23 Lack of customers 38 42 53 63 46 Gov. licensing and regulations 28 35 0 9 8 Taxation 30 38 60 37 31 Lack of business skills 19 2 0 3 8 Lack of job specific skills 11 2 0 0 0 Problems obtaining power or water 6 10 0 6 0 Lack of raw material 8 0 7 20 15 Lack of adequate labor 8 15 0 17 0 Unable to work (disability, age) 2 6 0 3 0 Household/family duties 0 8 7 9 8 No major difficulties 15 21 20 20 31 Source: UNDP/World Bank/EC regional Roma survey (2011). Figures represent the shares of households with at least one self-employed household member indicating each issue as one of the three main problems in operating a business. Figures in bold highlight issues, which were mentioned in more than 20% of all cases in each country; figures in red highlight the most frequently mentioned problem in each country. In short, the country macro indicators and responses by surveyed entrepreneurs point to barriers that are both country specific and specific to Roma and their non-Roma neighbors. The Doing Business showed that starting a business is most difficult in Slovakia and especially the Czech Republic. These findings may go some way in explaining why Roma in these two countries have lower self- employment rates than in the other countries. Lack of start-up finance is highlighted as a particular challenge among surveyed entrepreneurs in each of the countries, even though Bulgaria, Romania, and Slovakia have relatively favorable ‘getting credit’ rankings in the Doing Business survey. The next chapter will explore in more detail the specific credit barriers faced by Roma. Finally, in addition to lacking credit, the entrepreneurs also identified several non-credit barriers, especially taxation and government regulation, and lack of customers once they are operating their businesses. 39 4 ACCESSING MICRO-CREDIT ENTERPRISE FINANCING This chapter will focus specifically on barriers to accessing credit, for both aspiring Roma entrepreneurs and Roma businesses. Following a brief description of the development of the microfinance sectors in Eastern Europe, the chapter first examines current usage of credit among Roma. It then analyzes barriers to accessing credit for Roma. These may take various forms. For example, credit providers may find it risky to lend to individuals with little education and training. Or, a lack of collateral can prevent borrowers from being able to secure their loan. The chapter also reviews the microfinance sectors for each of the countries, as well as EU and other international initiatives available to these countries promoting microfinance and entrepreneurship. 4.1 THE EMERGENCE OF MICROFINANCE For a long time, global microfinance typically involved small loans disbursed using unconventional lending methods. Reflecting this, the terms ‘microcredit’ and ‘microfinance’ were initially used almost interchangeably. Over time, microfinance providers have started offering other financial services, including saving accounts, payment services, and insurance, as well as accompanying non-financial services such as Business Development Services (BDS), (McGuire and Conroy, 2000: 15). Microfinance activities in Eastern Europe emerged in the early 1990s as part of the countries’ transition from centrally planned systems to market economies. In this changing economic climate, top-down microfinance initiatives were established as a development tool to facilitate the re-emergence of entrepreneurship and the private sector, which had been largely interrupted in the centrally planned systems. Initially, non-profit NGOs and other non-depository microfinance institutions were providing microfinance services in the region. Only later did depository enterprises, such as commercial- and development banks, enter the market (Kraemer-Eis and Conforti, 2009: 21). Benefitting from international assistance, the microfinance sectors in Bulgaria, Hungary and Romania developed relatively fast. In Bulgaria, international donor organizations such as USAID and the EU-Phare program supported new microfinance programs. A breakthrough in the development of the Bulgarian microfinance sector occurred in 1997 when legal permission was granted to the EU-Phare program to provide loans as a credit cooperative. Following this decision, other providers chose the same model, increasing the scale and outreach of microfinance services in the country (Carboni et al., 2010: 104). In Hungary, the government cooperated with the European Commission and EU-Phare to set up the National Microcredit Scheme (MCS). By 1996, a network of Local Enterprise Agencies (LEAs) covered the whole of Hungary (ibid., 190-191). The EU did not originally fund microfinance activities in Romania, but other forms of international support did. In 1995, Opportunity International established its Romanian chapter - Opportunity Microfinance Romania (OMRO) (ibid.,308). Since then, several other organizations have entered the microfinance sector, the importance of which grew rapidly, as recognized by the Microfinance Companies Law adopted in 2005 (EMN, 2009: 75). Development of the microfinance sector has lagged behind in the Slovak Republic and the Czech Republic. One of the wealthier countries in Central and Eastern Europe (CEE) after the fall of communism, Slovakia experienced a slow emergence of microfinance activities (EMN, 2009: 257). Several initiatives were set up, but the scope of the Slovak microfinance sector remains limited. In the Czech Republic, there were historically no financial institutions offering microfinance services or, more generally, lending to microenterprises. Only in 2000 did support programs financed by the European Bank for Reconstruction and Development (EBRD), EIB, and the Czech government spur growth in SME 40 financing. Microfinance is not recognized as a separate section of finance provision, however (Carboni et al, 2010: 204, 216). 4.2 USAGE OF CREDIT AMONG ROMA A significant minority of Roma households borrows money, with half to most of the loans provided by formal financial providers. The table below shows that Hungary and the Czech Republic are the two countries with the highest proportion of Roma households currently borrowing from either formal or informal loan providers, respectively 31 percent and 27 percent. In contrast, only approximately 15 percent of households in Bulgaria and Slovakia are currently borrowing. Comparable data from EU SILC suggest that in Bulgaria, the Czech Republic, and Romania, Roma households borrow less frequently than the average household in the general population. Further analysis shows that while borrowing from formal institutions is predominant in all countries, only slightly more than half of borrowing Roma households in the Czech Republic, two thirds in Bulgaria and three quarters in Romania get their loans from such institutions. TABLE 26. BORROWING AMONG ROMA HOUSEHOLDS (%) Czech Bulgaria Hungary Romania Slovakia Republic Roma Households, Currently Borrowing 17 28 31 22 15 Currently Borrowing from Formal Financial Providers: Roma 12 20 31 18 8 General Population 21 18 21 17 24 Source: UNDP/World Bank/EC regional Roma survey (2011), and EU SILC (Eurostat, 2008). Formal providers include commercial banks and non-commercial, credit cooperatives, non-bank credit institutions, and non-depository MFIs. The average amount of credit disbursed to Roma households remains far below 25,000 Euro, which is the threshold commonly used to define ‘microloans’. 14 The table below shows that Roma households on average have borrowed between 1,672 Euro in Romania and 7,500 Euro in Hungary. Since these are currently disbursed loans, the data does not enable a more detailed temporal assessment. Average monthly loan installments constitute a substantial share of monthly household disposable income in all countries – particularly in Romania, where loan installments are 133% of household income. The burden is lowest – but still substantial – in Slovakia and the Czech Republic, standing at 13% and 15%, respectively. 14 The European Commission defines microcredit as “a loan of up to EUR 25,000 for business initiatives, from any institution whose purpose includes lending smaller amounts� (2010: 5), “to micro-entrepreneurs, employees who wish to become self-employed, to people working in the informal economy and to the unemployed and others living in poverty who are not considered bankable� (2007: 12). 41 TABLE 27. AVERAGE LOANS AND INSTALLMENTS Czech Bulgaria Hungary Romania Slovakia Republic 3,457 2,735 7,387 1,661 2,849 Amount of Credit (Euro) (650) (428) (1,580) (258) (773) 281 114 142 124 80 Monthly Loan Installment (Euro) (113) (14) (49) (50) (8) Monthly Loan Installment as Proportion of Monthly 119 15 37 133 13 Income (%) Source: UNDP/World Bank/EC regional Roma survey (2011). Response rate: 100% of those currently borrowing reported the amount they borrow. 90% of those currently borrowing reported their monthly loan installments; 79% of those currently borrowing reported both their monthly installments and their total income. Cluster-robust standard deviations in brackets. Among Roma households that secured a loan, the most important sources of credit are banks and credit cooperatives, followed by microfinance institutions (MFIs) and friends or relatives. Yet, the share of households using non-formal loans remains high. The table below shows that there is quite some cross-country variation in terms of credit sources. For instance, in the Czech Republic approximately a quarter of households borrow from commercial banks or credit cooperatives, and microfinance institutions, and only slightly fewer from friends and relatives, and informal lenders. In Hungary, commercial banks and credit cooperatives account for an overwhelming majority of credit. In contrast, in Bulgaria only half of Roma households borrow from these formal venues, and approach shopkeepers, friends and relatives, and informal lenders relatively more. In contrast to the existing literature (v. Ivanov and Tursaliev, 2006: 37; UNDP 2002: 50), the current data does not report a high reliance on informal lenders by Roma households – with the possible exception of the Czech Republic and Bulgaria. However, these results could be biased, given that Roma may fear to disclose any transactions with informal lenders, and especially with loan sharks. TABLE 28. SOURCES OF CREDIT AMONG ROMA HOUSEHOLDS Bulgaria Czech Hungary Romania Slovakia Republic Formal Provider 61 52 93 77 87 Depository Formal Institution (Commercial 58 24 88 71 46 Bank, Credit Cooperative or other bank) Non-Depository Formal Institution 3 33 5 6 43 Microfinance Institution (MFI) 2 31 1 6 36 Other Non-Depository Credit 1 1 3 0 7 Institution Informal Provider 39 48 7 23 13 Shopkeepers 19 9 2 6 9 Friends or Relatives 6 25 6 17 7 Other Informal Lender 7 33 0 3 8 Households which currently have a loan 100 100 100 100 100 Source: UNDP/World Bank/EC regional Roma survey (2011). Sample restricted to Roma households reporting that they are borrowing or have borrowed in the past. Since households can have borrowed from multiple sources, figures do not add up to 100. 42 When comparing Roma households to their non-Roma neighbors, no significant difference exists in the likelihood of borrowing money. The regression estimates in Annex Table 2 shows that if Romani is spoken at home, the household is 5 percentage points less likely to have ever borrowed and 4 percentage points less likely to be currently borrowing. On the other hand, the coefficient on Roma ethnicity is insignificant, suggesting that once controlling for the language spoken at home as well as other background characteristics, Roma and non-Roma households in the same or nearby communities have the same likelihood to have a loan. Among Roma households, those earning higher incomes and located closer to a bank are more likely to have loans. The regression estimates in Annex Table 3 show a linear probability model that describes the likelihood of having received a loan among Roma households. Speaking Romani at home decreases the probability by 8 percentage points. Furthermore, larger households are slightly more likely to have loans. The income of the household significantly increases the chance that the household has loans – suggesting that usage of credit is higher among richer households. Perhaps surprisingly, Roma living in rural settlements are more likely to borrow than Roma living in urban settlements. Lastly, households located at walking distance (<3km) from a bank are 3 percentage points more likely to borrow or to have borrowed in the past. Most Roma households with loans use their largest loan for non-productive purposes such as food and house improvements. As illustrated in the figure below, approximately one third of loans are used for house improvements in all countries but the Czech Republic. The second most important use of loans is to purchase food, and in some countries, current consumption products. On the other hand, only 10% or less of all loans granted to Roma households are used for income generating activities. This pattern may be explained by a number of reasons, such as the prevalence of hunger, the lack of stable income sources, or the potential difficulty of obtaining loans that are large enough to fund business investments. FIGURE 12: USE OF MAIN LOAN BY ROMA HOUSEHOLDS 45 39 38 40 36 34 35 30 26 27 25 Percent 25 18 19 19 20 17 1615 15 15 13 12 15 10 10 11 8 9 10 10 6 7 67 6 4 5 22 322 22 2 5 1 1 0 Bulgaria Czech Republic Hungary Romania Slovakia Source: UNDP/World Bank/EC regional Roma survey (2011). 43 When focusing specifically on Roma and non-Roma neighboring entrepreneurs, external credit is only a secondary source of business start-up funding, far behind after household savings. The table below reports the main sources of funding specifically for entrepreneurs. Again, the figures combine the Roma and non-Roma entrepreneur sub-samples because of the few observations. Household savings are by far the most frequently mentioned source to fund a business, mentioned by as few as 60 percent (Slovakia) to as many as 82 percent (Romania) of entrepreneurs. Not shown is that savings are slightly more commonly mentioned by non-Roma neighbors than Roma, underscoring the limited savings finding by Roma. The next most common source of financings comes from loans or gifts from family and relatives, mentioned by 10 percent of entrepreneurs in Slovakia to 43 percent in the Czech Republic. These are followed by bank loans, which were mentioned rarely in the case of Romania (3 percent highlights consumer credit, none highlight business credit) and somewhat more frequently in Bulgaria (18 and 16 percent, respectively, for consumer and business credit) and the Czech Republic (23 and 21 percent, respectively), with formal credit use lower among the Roma entrepreneurs (not shown) than the joint sample figures reported in the table. Lastly, fewer than 10 percent of entrepreneurs indicated not needing any financings. TABLE 29. SOURCES OF FUNDING FOR BUSINESSES BY ROMA AND NON-ROMA NEIGHBORING ENTREPRENEURS Czech Bulgaria Republic Hungary Romania Slovakia (n=50) (n=56) (n=15) (n=39) (n=10) Pesonal/Informal Funding Sources: Household savings 76 61 73 82 60 Sale of Assets 4 14 7 8 10 Inheritance 22 11 7 18 20 Loans or Gifts from Relatives/Family 18 43 20 23 10 Loans or Gifts from Friends 14 14 0 15 10 Private Money Lender 4 7 0 10 0 Institutional Funding Sources: Bank Loan – Consumer Credit 18 23 13 3 10 Bank Loan - Business Credit 16 21 7 0 20 Loan from Government Agency 0 0 0 0 0 Grant from Government Agency 0 0 7 3 0 Other Source 8 0 13 8 20 Did not need any money 4 14 13 10 0 Source: UNDP/World Bank/EC regional Roma survey (2011). Roma households in each country reporting type of funding as one of the three main sources of funding for their business. Up to three answers possible. Most Roma businesses do not even have a bank account. In the three countries for which data are available, most Roma businesses do not have a formal bank account, or, in the case of the Czech Republic, only a slight majority of businesses do. In Bulgaria and the Czech Republic, non-Roma neighboring entrepreneurs have bank accounts much more often than Roma entrepreneurs. 44 FIGURE 13. SHARE OF ROMA AND NON-ROMA NEIGHBORING ENTREPRENEURS WITH BANK ACCOUNTS 100 82 80 57 Percent 60 48 40 34 32 31 20 0 Bulgaria Czech Hungary Romania Slovakia Republic Roma Non-Roma Neighbors Source: UNDP/World Bank/EC regional Roma survey (2011). A more restrictive definition of entrepreneurship was used: only those households in which the interviewer confirmed that one or more household members had an own (formal) business provided information on whether the business had a formal bank account. Missing values for non- Roma exceed 10% for Bulgaria (22%) and Romania (14%). Data on Hungary and Slovakia were not reported because of limited number (<10) of responses. 4.3 DEMAND SIDE BARRIERS TO ACCESSING MICROCREDIT The analysis below distinguishes between credit provision at the intensive margin of the microcredit market – i.e. to entrepreneurs with existing businesses – and at the extensive margin –i.e. to those aspiring entrepreneurs interested in starting a new business as well as those currently operating informal businesses. In particular, for the purposes below, the extensive margin focuses on three groups of people aged 20-54 years old that can be considered aspiring entrepreneurs. First, those who already classified as self-employed, but working informally (i.e. without a contract) as unskilled workers in periodic jobs. Second, those people who answered ‘yes’ or ‘maybe’ to the question on being interested in "becoming self-employed and starting a business", but are currently working as an employee. And third, those people who also answered ‘yes’ or ‘maybe’ to the question on being interested in "becoming self- employed and starting a business", but who are currently unemployed. In short, whereas the intensive margin is synonymous with credit targeting established businesses, the extensive margin is synonymous with start-up business credit. The distinction between intensive and extensive margin is useful because providing credit to these two groups involves different challenges for credit providers. For example, it is easier to assess credit worthiness for existing businesses, since they can typically provide evidence of past credit behavior, tax payments, and/or future profitability prospects. The reverse also holds. For example, those at the extensive margin may be more likely to lack information on existing possibilities for accessing credit or the functioning of the banking system. Moreover, the extensive margin of aspiring entrepreneurs in particular represents a potential target group for microcredit initiatives which – if successful – can raise self-employment levels. As shown in the figure below, a substantial part of the aspiring entrepreneurs in the extensive margin (4 to 10 percent of the total working age population) is currently unemployed. If this group were to be successfully supported in starting up businesses, this would significantly decrease overall unemployment rates among Roma in each of the five countries. 45 FIGURE 14. SHARE OF WORKING AGE ROMA WITH EXISTING BUSINESSES (INTENSIVE MARGIN) AND ASPIRING TO START A BUSINESS (EXTENSIVE MARGIN) 40 35 30 25 Percent 7 20 10 15 12 10 12 4 7 7 5 7 9 4 5 4 0 2 1 1 Bulgaria Czech Hungary Romania Slovakia Republic Intensive Extensive: Employed Extensive: Unemployed Source: UNDP/World Bank/EC regional Roma survey (2011). Sample sizes for the existing entrepreneurs at the intensive margin are small, compromising to some extent inference on this group. Given the fact that in Slovakia, Hungary, and the Czech Republic, self- employment rates are relatively low, the subsample of potential Roma borrowers belonging to the intensive margin in these three countries is relatively small. For Slovakia, the sample size of this group is so small (11 observations) that data were left out altogether. TABLE 30. INTENSIVE AND EXTENSIVE MARGIN: NUMBER OF OBSERVATIONS Bulgaria Czech Hungary Romania Slovakia Republic Intensive 98 42 19 192 11 Extensive 450 209 237 394 230 Total 541 251 256 582 241 Source: UNDP/World Bank/EC regional Roma survey (2011). Roma sample only. To assess credit barriers, the two groups of potential Roma borrowers are compared with each other and with two ‘credit’ groups from the general population using EU SILC 2008 information: (1) ‘credit individuals’ - those currently self-employed in the general population, between the ages of 20- 54 inclusive, that are in households with credit or other non-housing loans; and, (2) ‘discouraged credit individuals’ – those currently employed (self-employed, employee or family worker) in the general population, between the ages 20-54 inclusive, who report that banks would refuse them credit if they applied. A comparison of background characteristics among these two Roma and two general population groups provides an indication of the For example, if those from the general population who expect to be refused loans have higher education levels and are less likely to have debt than aspiring Roma entrepreneurs, then the latter will unlikely be able to get loans either. DEMOGRAPHIC BACKGROUND CHARACTERISTICS Basic demographics such as age and sex do not point to obvious barriers. First, aspiring Roma entrepreneurs are on average in their mid-thirties, between 28-42% are female, and the proportion in rural areas varies from 3% in the Czech Republic, 44% in Romania, 52% in Bulgaria, 62% in Hungary to more than 70% Slovakia. Compared with this group, existing Roma entrepreneurs in each of the countries are 46 older, more often male, and more often reside in rural areas, with the exception of Bulgaria. Second, aspiring Roma entrepreneurs are – on average – younger than the two comparison groups from the general population, and are male slightly more often, especially in comparison with the ‘discouraged credit’ group. TABLE 31. DEMOGRAPHIC STATISTICS FOR ROMA (INTENSIVE AND EXTENSIVE MARGIN) AND GENERAL POPULATION (COMPARISON GROUPS) Bulgaria Czech Republic Hungary Romania Slovakia Roma Int Ext Int Ext Int Ext Int Ext Int Ext Age (av.) 40 35 39 33 43 35 36 33 . 33 Female (%) 34 39 29 26 32 32 30 35 . 34 Rural (%) 33 51 5 2 68 62 75 48 . 71 General population Cred Disc Cred Disc Cred Disc Cred Disc Cred Disc Age (av.) 42 39 40 39 41 39 38 38 39 39 Female (%) 34 45 31 54 40 46 39 43 27 53 Rural (%) 46 63 44 26 43 66 63 76 41 45 Source: UNDP/World Bank/EC regional Roma survey (2011), and EU SILC (Eurostat, 2008). Int = intensive margin: currently self-employed Roma with established businesses; Ext = extensive margin: currently not-self- employed Roma, in the labor force, between the ages of 20-54 inclusive, showing an interest in becoming self- employed; or self-employed but working informally as unskilled in periodic jobs without a contract; Cred = currently self-employed in the general population, between the ages of 20-54 inclusive, that are in households with credit or other non-housing loan; Disc = discouraged credit applicants, i.e. currently employed people (self- employed, employee or family worker) in the general population, between the ages 20-54 inclusive, who report that bank would refuse them credit if they applied. LOW LEVELS OF HUMAN CAPITAL AND LIMITED WORK EXPERIENCE Limited business- and financial skills may be an important barrier to accessing credit for Roma entrepreneurs. External financing is generally more readily available if the entrepreneur is able to construct a credible financing proposal, is willing to share control, has a credible business plan, and if the uncertainties and risks associated to implementation are limited (World Bank, 2008: 57). The skills needed for this, in turn, depend importantly on previous education and employment experience. Aside from familiarity with financial services, more basic literacy- and numerical skills – and increasingly computer skills – are required for a successful loan application. Education levels among both groups of potential Roma business credit borrowers are considerably lower than among the two comparison groups in the general population. The vast majority of potential Roma borrowers do not have secondary education, unlike the general population comparison groups. Even in the Czech Republic, where this rate is the highest, only just over half of the intensive-, and just under half of the extensive margin have completed secondary education. These shares only amount to one third in Slovakia, one fourth in Hungary, and less than one fifth in Bulgaria and Romania. In the general population comparison groups, on the other hand, secondary education enrollment is at minimum approximately 80% for individuals belonging to households with credit, and nearly 100% in Slovakia and the Czech Republic. The attainment rate is lower among individuals in households who expect to be refused credit, ranging from 50% in Bulgaria to over 85% in Slovakia. However, these rates are still considerably higher than among either existing- or aspiring Roma entrepreneurs. 47 FIGURE 15: SECONDARY EDUCATION COMPLETION AMONG (POTENTIAL) CREDIT GROUPS 100 98 94 98 87 86 90 80 80 69 70 62 62 57 Percent 60 50 50 44 40 26 23 27 30 14 17 15 20 10 10 0 Bulgaria Czech Hungary Romania Slovakia Republic Intensive Extensive With Credit Discouraged Source: UNDP/World Bank/EC regional Roma survey (2011) for Roma, and EU SILC (2008) for the general population. For a definition of the four groups, see notes above Table 31. Aspiring Roma entrepreneurs typically also have much less work experience as compared to existing Roma entrepreneurs, or to the comparison groups from the general population. The number of years spent in paid work among aspiring Roma entrepreneurs is on average below 10 years in each country, compared with between 11 and 19 years of work experience among existing entrepreneurs. This may partly be a result of the age difference, since existing entrepreneurs are slightly older on average than aspiring entrepreneurs. In the general population sample, in four out of five countries - Hungary being the exception - individuals belonging to households with credit have more work experience than those who expect to be refused credit. Yet, work experience of aspiring Roma entrepreneurs remains well below that of general population individuals in both of these groups, implying that a lack of work experience may be a challenge when these individuals at the extensive margin would apply for credit. FIGURE 16: YEARS SPENT IN PAID WORK AMONG (POTENTIAL) CREDIT GROUPS 20 18 19 18 18 17 17 16 16 16 15 14 14 14 14 14 13 11 Percent 12 10 9 9 9 8 8 7 6 4 2 0 Bulgaria Czech Hungary Romania Slovakia Republic Intensive Extensive With Credit Discouraged Source: UNDP/World Bank/EC regional Roma survey (2011) for Roma and EU SILC (2008) for the general population. 48 Levels of participation in adult training courses and computer literacy are low. Adult learning courses are followed most frequently in Hungary, but even there only one in five existing entrepreneurs and a quarter of aspiring ones have experience with such courses or apprenticeships. Moreover, in three out of five countries, fewer than one in three potential borrowers know how to use a computer word processing program. TABLE 32. PROFESSIONAL AND COMPUTER SKILLS AMONG ROMA Czech Bulgaria Hungary Romania Slovakia Republic Int Ext Int Ext Int Ext Int Ext Int Ext Adult learning courses or 19 7 13 14 21 27 12 11 . 14 professional apprenticeship Able to use computer word 28 17 58 41 21 35 15 28 . 28 processing program Source: UNDP/World Bank/EC regional Roma survey (2011). Relatively little experience with banking services likely points to limited financial literacy, especially in Romania. Usage of banking services among existing and aspiring entrepreneurs varies considerably between countries. In the Czech Republic, Hungary and Slovakia, both groups often possess current accounts, whereas such accounts are almost never held by either group in Bulgaria and Romania. A similar pattern appears for debit / payment cards, which are held by approximately half of all existing and aspiring entrepreneurs in Bulgaria, the Czech Republic and Hungary, but are accessed much less in Romania and Slovakia. Possession of savings accounts is very low among both the intensive and the extensive margin. The same holds for term deposits, money transfers, and credit. Credit or store cards are held relatively frequently by self-employed in the Czech Republic, but not in the other countries. TABLE 33. USE OF BANKING SERVICES AMONG (ASPIRING) ROMA ENTREPRENEURS (%) Bulgaria Czech Republic Hungary Romania Slovakia Int Ext Int Ext Int Ext Int Ext Int Ext Current bank 20 3 73 47 58 59 6 8 . 43 account Savings account 5 4 8 6 6 3 3 2 . 17 Debit/payment card 45 49 47 26 53 51 5 16 . 19 Credit/store cards 12 2 45 14 5 5 2 3 . 13 Term deposit 4 1 3 0 0 0 1 0 . 5 Money transfers 2 1 22 7 16 8 1 2 . 7 Credit 15 22 30 30 28 39 19 35 . 17 Source: UNDP/World Bank/EC regional Roma survey (2011). Data refers to the percentage of individuals belonging to the intensive / extensive margin and living in a household that makes use of different banking services. LACK OF COLLATERAL / (PROOF OF) CREDIT WORTHINESS Potential Roma business credit borrowers may be unable to meet collateral requirements set by microfinance providers. Collateral is often provided in the form of physical assets, cash savings or deposits, and/or a proof of receivables. However, such assets are far from abundant in most Roma households. Most Roma at the intensive and extensive margin own their dwelling, but housing conditions 49 are poor and proof of ownership often does not exist. Furthermore, literature on, for example, Romania and Slovakia suggests that although ownership may be clear for the Roma themselves, a majority of the households lack official property documents to prove ownership (ENAR, 2011a: 20; ENAR, 2011b: 1215). TABLE 34. DWELLING TYPE AND HOME OWNERSHIP AMONG (ASPIRING) ROMA ENTREPRENEURS (%) Czech Bulgaria Hungary Romania Slovakia Republic Family Owns the Dwelling Intensive 86 33 89 86 . Extensive 85 14 82 87 69 Source: UNDP/World Bank/EC regional Roma survey (2011). Potential Roma borrowers frequently do not own other physical assets that can be used as a loan guarantee such as a car. Among aspiring Roma entrepreneurs, car ownership rates never exceed one third. Rates are higher among existing entrepreneurs, although they remain low in Hungary and Romania. In all countries, car ownership is higher among households in the general population with credit as compared to both Roma groups, and among households that expect to be refused credit it exceeds rates for the extensive margin. For the ownership of computers, very similar results hold. TABLE 35. POSSESSIONS AMONG (ASPIRING) ROMA ENTREPRENEURS AND GENERAL POPULATION COMPARATORS(%) Bulgaria Czech Republic Hungary Romania Slovakia Roma Int Ext Int Ext Int Ext Int Ext Int Ext Car 60 28 78 33 42 23 20 14 . 22 Computer 47 21 46 34 37 38 18 25 . 17 General Population Cred Disc Cred Disc Cred Disc Cred Disc Cred Disc Car 77 40 96 38 93 39 53 23 88 42 Computer 64 23 88 41 88 38 53 29 85 47 Source: UNDP/World Bank/EC regional Roma survey (2011) for Roma; EU-SILC (Eurostat, 2008) for the general population. The level of savings is very low for both groups of potential Roma borrowers. Fewer than 10% of Roma households belonging to either group report having savings except for the Czech Republic. Among those with savings, amounts saved are generally very low, covering living expenses for at most 8, but often no more than 2 or 3 months. TABLE 36. SAVINGS AMONG ROMA – INTENSIVE AND EXTENSIVE MARGINS Czech Bulgaria Hungary Romania Slovakia Republic Int Ext Int Ext Int Ext Int Ext Int Ext Savings (%) 10 3 49 22 0 3 3 2 . 15 Months of savings (among 2 5 3 3 . 1 1 8 . 2 those with savings) Source: UNDP/World Bank/EC regional Roma survey (2011). 15 Further information can be found in ENAR Romania report: p. 20; Slovakia report: p. 12. 50 The ability to put up collateral – as well as overall credit worthiness – is further worsened by an already high indebtedness among a considerable part of the (aspiring) Roma entrepreneurs. In all countries, at least one quarter of all potential Roma borrowers live in households, which report to have outstanding payments on their utility bills. Between-country differences are not very pronounced, with the exception of Slovakia. Mortgage arrears do not seem to be substantial but this also reflects few households having mortgages. FIGURE 17. ARREARS AMONG (ASPIRING) ROMA ENTREPRENEURS 60 54 50 46 35 36 37 40 34 Percent 28 30 16 18 20 14 11 10 3 1 0 1 0 1 1 0 Bulgaria Czech Hungary Romania Slovakia Republic Utility Bills Intensive Utility Bills Extensive Mortgage Intensive Mortgage Extensive Source: UNDP/World Bank/EC regional Roma survey (2011). Data refers to percentages of individuals living in households of which the household head has indicated the household has arrears. Finally, the large share of temporary employment prevents a substantial proportion of Roma from using a proof of regular income obtained in a formal bank account. In all countries, over half of the aspiring Roma entrepreneurs have a temporary or seasonal job. Conversely, none of the individuals in the general population belonging to households with credit report to have a temporary job. This percentage is slightly higher for those who expect to be refused credit, but still markedly lower than either of the two Roma groups examined here. TABLE 37: TEMPORARY EMPLOYMENT AMONG (ASPIRING) ROMA ENTREPRENEURS AND GENERAL POPULATION Bulgaria Czech Republic Hungary Romania Slovakia Int Ext Int Ext Int Ext Int Ext Int Ext Roma 56 65 57 61 74 71 88 66 . 51 Cred Disc Cred Disc Cred Disc Cred Disc Cred Disc General 0 30 0 34 0 26 0 5 0 20 Population Source: UNDP/World Bank/EC regional Roma survey (2011), and EU SILC (2008). The precarious situation of (aspiring) Roma entrepreneurs is captured by the self-valuation of their economic status. With the exception of the Czech Republic, a majority of aspiring Roma entrepreneurs live in households, which, according to the household head, are economically unsafe. In Romania and Slovakia, the same holds for individuals in the intensive margin. If these perceptions are adequate, lenders may be inclined to doubt the credit worthiness of many aspiring entrepreneurs. These figures support the 51 abovementioned low income levels, high uncertainty, and unclear prospects for income generation among (aspiring) Roma entrepreneurs in general. FIGURE 18. HOUSEHOLD ECONOMIC SAFETY AMONG (ASPIRING) ROMA ENTREPRENEURS (%) A. Intensive (existing entrepreneurs) B. Extensive (aspiring entrepeneurs) 100 8 6 0 4 9 100 2 5 5 3 7 90 90 80 30 18 80 37 24 32 70 70 49 40 53 Percent Percent 60 74 60 50 83 50 40 73 40 71 30 66 30 62 65 46 53 20 40 20 10 26 10 11 0 0 Totally unsafe Relatively safe Totally safe Totally unsafe Relatively safe Totally safe Source: UNDP/World Bank/EC regional Roma survey (2011). Data refers to percentages of individuals living in households of which the household head has indicated to feel economically unsafe, relatively safe or totally safe. DOCUMENTATION REQUIREMENTS However, proof of identity does not seem to be a major barrier for potential Roma borrowers. Credit providers usually require loan applicants to submit formal documents as well, such as proof of identity, address, and/or employment, as well as the provision of a credit history. However, nearly all potential Roma borrowers are in the possession of a birth certificate and an ID card. Very few respondents, particularly in the extensive margin, reported to be in possession of a passport, but since banks tend to accept either ID cards or passports – and not necessarily both – as proof of identity (Pytkowska et al., 2008: 9), this does not seem to be a major obstacle. TABLE 38: POSSESSION OF PERSONAL DOCUMENTS AMONG ROMA (%). Bulgaria Czech Republic Hungary Romania Slovakia Int Ext Int Ext Int Ext Int Ext Int Ext Birth certificate 100 100 100 100 100 100 99 99 . 100 ID Card 100 100 100 100 100 99 97 99 . 98 Passport 30 16 45 35 21 8 17 20 . 22 Source: UNDP/World Bank/EC regional Roma survey (2011). OVERVIEW OF DEMAND-SIDE BARRIERS TO ACCESSING MICROCREDIT AMONG ROMA Overall, these results show that potential Roma borrowers belonging to both groups, but especially aspiring entrepreneurs, face substantial barriers applying for a loan, including low levels of education and professional experience, little experience with financial services, and limited availability of collateral. The table below presents an overview of whether the two groups of potential 52 Roma borrowers on average reported worse results than those from the general population who expect to be refused credit, or whether the results that could not be compared to the general population were favorable or unfavorable. All rows with ‘yes’ refer to issues that are likely to forms substantial challenges for Roma wishing to access business credit. In general, the abundance of ‘yes’ rows implies that aspiring as well as existing Roma entrepreneurs face a very challenging credit market if they would apply for loans. TABLE 39: SUMMARY OF BARRIERS FOR (ASPIRING) ROMA BORROWERS Intensive Extensive Margin Margin Indebtedness Yes Yes Human capital Yes Yes Work experience No Yes Collateral Yes Yes Proof of regular income Yes Yes Personal identification No No Source: UNDP/World Bank/EC regional Roma survey (2011). Table reports those barriers for which either group has ‘worse’ results than the ‘credit refusal group’ from the general population. 4.4 SUPPLY SIDE BARRIERS TO ACCESSING MICROCREDIT INSTITUTIONAL TYPES The current status of the microfinance sector in the region reflects the historical trends mentioned in the introduction. According to the EMN, in Bulgaria and Romania, the innovative models of microfinance activities have “proven to be sustainable and replicable at larger scale� (2009: 16). The microcredit-sector has taken an increasingly important role in the lending industry, and nearly all MFIs provide non-financial business development services (Carboni et al, 2010; 319; EMN, 2009: 163). In Hungary, the microfinance industry is considered to be a “success� too, particularly because of the strong development of the National Microcredit Fund (Carboni et al, 2010; 10). The Slovak microfinance sector, however, has “struggled to succeed� (13) and “remains undeveloped� (EMN, 2009: 17), and in the Czech Republic there are no specific microfinance institutions at all (EMN, 2009: 17; Kraemer-Eis and Conforti, 2009: 33). The institutional nature of the main microcredit providers in the five countries differs widely. In Slovakia and the Czech Republic, a state-owned development bank dominates the SME and micro-lending markets, providing both direct loans and guarantees. No MFIs are actively engaged, and commercial banks are reluctant to provide micro-loans. In Hungary, commercial banks, a development bank, and several non-depository MFIs all provide micro-loans, with non-bank organizations still being the main actors in the market. By contrast, in Bulgaria and Romania, a thriving and diverse group of MFIs coexist with commercial banks. Romania has perhaps the most diversified microcredit sector. Commercial banks have not only down-scaled their product menu to a make products more accessible for small businesses, but have often also established subsidiaries that specifically target micro-entrepreneurs. Some of the larger MFIs have transformed from non- profit into for-profit organizations. In Slovakia, a state-owned development bank is the main provider of credit to micro-businesses, both directly and through guarantees. No non-bank institution is primarily engaged in micro-lending, 53 though several organizations offer small-scale microloan programs alongside other activities (Kraemer-Eis and Conforti, 2009: 41; EMN, 2009: 258), such as ETP Slovakia. The Slovak Guarantee and Development Bank (SZRB) has a statuary mandate to support and provide lending for those SMEs without access to commercial bank lending. In addition, it offers guarantees for microloans that partner commercial banks disburse. Several commercial banks provide microcredit services, but almost exclusively under the SZRB guarantee scheme. In 2003, the SZRB launched specific microloan programs targeting micro- entrepreneurs (Carboni et al., 2010: 217). A state-owned development bank also dominates the Czech Republic microcredit market. Like the SZRB, the Czech and Moravian Guarantee and Development Bank (CMZRB) has a mandate to sustain SME finance through direct loan provision and guarantees. However, unlike the SZRB in Slovakia, the CMZRB does not provide a specific microloan product, and average loan size is reported to be substantially above that of a microloan. It is expected that this might change with the JEREMIE EU- sponsored program, for which the CMZRB is expected to provide the microcredit services (EMN, 2009: 227). Aside from this, several commercial banks provide micro-loans under their mainstream lending products, although they are reluctant to develop specific microcredit activities (205). Another similarity with Slovakia concerns the absence of MFIs. Nonetheless, there are other non-depository financial institutions, such as leasing- and consumer loan companies, that in a few instances disburse micro-loans used to finance business activities. Their overall importance on the market, however, is minor. In Hungary, government-supported non-depository institutions are the main microcredit actors, though depository organizations are becoming increasingly important. An important actor is the government-sponsored Hungarian Foundation for Enterprise Promotion (MVA) (MVA, no date: 1). Responsible for the National Microcredit Fund, it operates through a network of 20 Local Enterprise Agencies (LEAs) that independently and directly disburse microloans (Szefku, 2007: 9). In addition, there are several other government-, EU-, and private-sector corporations which provide or guarantee microloans and SME lending more generally (Carboni et al., 2010: 193-194). Turning to depository institutions, although commercial banks have become an increasingly important provider of SME finance, the extent of their provision of micro-loans remains unclear. Lastly, the state-owned Hungarian Development Bank (MFB) developed the ‘Microcredit Plus’ program, in cooperation with the network of LEAs, in order to establish additional microcredit facilities. In this respect, guarantee facilities provided by some depositary institutions play an important role as collateral. In Bulgaria, commercial banks are the main providers of micro-loans, accompanied by a growing and diverse group of MFIs. All major commercial banks and most of the smaller ones have gradually recognized the growth potential of start-ups and micro-business (Bulgarian National Bank, 2011). They are now the largest providers of microloans, though they specifically target those businesses with the highest growth potential and not necessarily those groups in society most in need of financial support (Damianova et al, 2008: 75). Alongside these for-profit depository enterprises, the privately-owned development bank ProCredit Bank offers micro-loans and credit to SMEs more generally to combat poverty as well as socio-economic marginalization (ProCredit Bank, 2010: 14). In addition, there are several cooperatives and credit unions that sustain micro-lending programs for their members, and many MFIs that provide micro-loans to the financially excluded through supporting micro-entrepreneurs and start-ups. Some target particular social groups, like women or the young, or economic groups such as farmers in rural areas (EMN, 2009: 158). Romania has arguably the most diversified microcredit market. As in Bulgaria, commercial banks are the largest providers of microloans, but they have not merely downscaled to reach micro-entrepreneurs. In several instances, commercial banks outsourced the provision of micro-loans to newly established 54 subsidiaries acting as non-depository microfinance institutions (Ibid., 87). This contrasts starkly with the Czech commercial bank business model, where lending products are not tailored to micro-entrepreneurs. As in Bulgaria, ProCredit Bank is also present in the Romanian market, offering micro-loans and credit to SMEs. Lastly, there also exists a thriving group of socially-oriented MFIs that offer micro-loans to micro- entrepreneurs and start-ups. Interestingly, some of the larger and more experienced MFIs such as Opportunity Microcredit Romania and Credit Patria choose the path of commercialization while maintaining their socially-oriented focus. MICROCREDIT PRODUCTS WITHIN REACH OF (ASPRING) ROMA ENTREPRENEURS In all five countries, it seems that microcredit is to a large extent provided to established businesses, particularly by commercial banks, and generally beyond the reach of the average poor Roma micro- entrepreneur seeking start-up funds. Recall that the data from the regional Roma survey (2011) showed that there are few potential Roma borrowers with existing businesses and that there is a much larger group of aspiring entrepreneurs – although facing considerable barriers to accessing credit. While it is difficult to assess the extent to which credit is provided to existing businesses or start-ups, the available evidence suggest that most of the existing small credit products in the region are beyond the reach of potential new Roma entrepreneurs. In the Czech Republic and Slovakia, very little information is available, suggesting that the supply of credit for small start-ups is equally limited. Following a series of interviews in the Czech Republic, the EIF Market Study on Micro Lending concluded that “none of the interviewed institutions [were] financing pure start-ups (excluding CMZRB) and the informal sector� (EMN, 2009: 210). In addition, many of the institutions did not create specialized products for disadvantaged or excluded groups, such as the Roma (EMN, 2009: 213). The CMZRB set up a special lending facility for start-up enterprises in 2004, but it was discontinued. In Slovakia, the main microcredit provider (the SZRB), has as its objective to support the segment of SMEs that commercial banks do not target (EMN, 2009: 258). Entrepreneurs lacking the means to secure loans or representing a higher risk-group to commercial banks can obtain a guarantee up to 60% of the loan principal, or may apply for direct SZRB lending. However, data available indicates that this program targets businesses with relatively large loan requests rather than micro- businesses. An exception was the (small) Integra Foundation microcredit program that disbursed 161 micro-loans worth approximately 350,000 Euro to start-up entrepreneurs - particularly women - between 1999 and 2007, but this program was discontinued (Carboni et al., 2010: 217). In Hungary, Romania, and Bulgaria, loans by depository institutions will generally also be beyond the reach of most potential Roma entrepreneurs. In Hungary, available data show that of all commercial bank loans extended in 2010, 5% were investment loans while the remaining 95% were current account loans providing short-term operating capital (Hungarian Supervisory Authority, 2010). Micro-loans are not commonly disbursed to the vulnerable or financially excluded (EMN and Latham & Watkins, 2008: 35). For example, even the Microcredit Plus program provided relatively large loans in 2009, when it disbursed loans to 418 clients for an average of approximately 37,750 Euro (MFB, 2009: 28). In Romania and Bulgaria, commercial banks generally include more microcredit products but similarly focus on working capital and investment needs of existing businesses (EIF, 2009: 87). There are, however, some non-depository institutions in these three countries supplying microloans to start-ups and existing businesses. In Romania and Bulgaria especially, there are several microcredit initiatives and MFIs which target client groups such as the unemployed, economically challenged groups, 55 agricultural producers, women, ethnic minorities, the young, the disabled, and the rural population (EMN, 2009; Carboni et al, 2010; Nachala; Ustoi; Mikrofond; Eurom Consultancy and Studies, 2010). Also, the national management authorities have used EU funds to target small- and microenterprises. For example, The New Hungary Microcredit Program focused on the poorer regions in the country. Between January 2008 and September 2010, a total of 2,649 micro-loans averaging almost 21,000 Euro were disbursed, although the extent to which start-ups were targeted remains unclear (Vingelman, 2010). Similarly, on 14 January 2011, the Hungarian government launched a new EU funded action plan called the New Széchenyi Plan 2011-13 (Box 2). This action plan includes seven focus areas, among which enterprise development, and will provide SMEs with support in the form of small grants combined with micro loans. BOX 2: MICROENTERPRISE SUPPORT THROUGH THE NEW SZÉCHENYI PLAN 2011-13 In 2011, the Hungarian government launched a call for tenders to support small and medium enterprises with a combination of grants and loan support. The program, funded in part by the European Regional Development Fund (ERDF), aims to support between 3500 and 15000 SMEs between 2011 and 2013. The program conditions specify that projects much bring in at least 10% in own resources. The grant value ranges from a minimum of (approximately) Euro 3.5k to Euro 14k, and can be no more than 45% of the project value. The loan component can be up to (approximately) Euro 29k, but cannot exceed 60% of the project value. Even this program will likely be beyond the reach of potential poor Roma micro-entrepeneurs starting a new business. Because the minimum grant funding is approximately Euro 3.5k and because this can maximally constitute 45% of the total project value, the minimum project size is (approximately) Euro 8k and, by extension, anyone wishing to participate must provide at least Euro 800 – 10% of the project value – in own resources. The minimum loan size is then approximately Euro 3.7k. Information from the Roma Regional Survey on savings, and experience from the new Kiut microfinance program in Hungary – described below – show that even these relatively modest amounts will be a considerable hurdle for most potential Roma entrepreneurs seeking to start new businesses. In addition, given the limited access to information (e.g. through the internet) and computer use, the registration requirements may similarly be an obstacle to reaching the average potential Roma entrepreneur. Source: Hungarian Development Agency, 2011. The only credit programs within reach of the average aspiring Roma micro-entrepeneur are small scale, often pilot, initiatives. There have been a number of MFI’s and NGOs in Bulgaria, Romania, Hungary, and Slovakia providing microcredit support specifically targeting (new) Roma entrepreneurs16. These initiatives are often collaborations between MFIs and NGOs and reach out to potential Roma entrepreneurs by targeting communities with known Roma populations. The number of loans have tended to be few (up to a few hundred depending on the project), loan sizes small (under Euro 1500), and often relying on some kind of group solidarity concept in which members provide guarantees – formally or informally – as collateral. Such initiatives also frequently combine the small loans with a range of non- 16 Examples include the Job Opportunities through Business Support (JOBS), a joint initiative of the Bulgarian Ministry of Labor and Social Policy and UNDP; the Alliance for Inclusive Business Development of Roma Communities in Bulgaria, an initiative of the CRS, USAID, USTOI AD and Mikrofond EAD; ETP Slovakia; the Microcredit Pilot Program for Roma in Romania, implemented by the Economic Development Center (CDE), which (part of the Soros Open Network); and, the microcredit programme for disadvantaged groups in Hungary with a special focus on Roma managed by Autonomia Foundation. 56 financial services, such as business registration and training. However, little is known about the impact or the sustainability of these projects. One of the projects currently targeting Roma entrepreneurs is the Kiut microcredit project in Hungary. This pilot project is part of the EU Parliament funded “Roma Pilots� initiative, which runs from summer 2010 until summer 2012. Their experience underscores that a comprehensive approach to micro-lending for (aspiring) Roma entrepreneurs is necessary (described in the concluding chapter). 4.5 EU AND OTHER INTERNATIONAL INITIATIVES FOR MICROCREDIT AND ENTREPRENEURSHIP International organizations, including the EU, provide funding through programs aimed specifically at improving access to finance for SMEs and micro-enterprises throughout Europe. This section examines to what extent such programs are available in the five countries, and to what extent they are being used. Special attention is given to programs initiated by the EU. International donor organizations are an important source of funding for non-bank microfinance institutions in all countries, except in the Czech Republic. In Romania, particularly, major MFIs receive funding from donor organizations. Aside from Opportunity International (see above), the Swiss Agency for Development and Cooperation (SDC), USAID, and World Vision International, among others, are active donor organizations in the Romanian microfinance market (Carboni et al., 2010: 312-313). In Bulgaria, too, several large international donors currently support microfinance activities. USAID is an active foreign donor that has supported several major MFIs such as Nachala and Ustoi, as well as a Roma- specific microfinance project (Alliance for Inclusive Business Development of Roma Communities in Bulgaria) together with Catholic Relief Services (CRS) (Rowe et al, 2009: 235). The Soros Economic Development Fund has also funded several MFIs targeting the financially excluded, Roma in particular (Soros Economic Development Fund, 2006). In Slovakia, several non-bank institutions that offer microloan programs on the side have been supported by international donors such as CIDA, the Open Society Foundation, UNDP, and USAID (EMN, 2009: 292; Carboni et al., 2010: 222). Hungary is involved with international donor organizations particularly through the Hungarian Foundation for Enterprise Promotion (MVA). This organization mainly draws funds from EU Structural Funds. The European Commission and the European Investment Bank (EIB) have recognized that access to finance is often a major challenge for SMEs, and have dedicated financial resources to address this. “The Commission shares the assessment of the European Investment Bank Group (0...) that there is a need to further improve SMEs’ access to finance,� (EC, 2008: 11; v. EC, 2005; EC, 2010: xxi). Because of this, by 2013, the EU Cohesion Policy intends to provide some €27 billion explicitly dedicated to the support of SMEs. Around €10 billion will be contributed through financial engineering measures (see below), and some €3.1 billion through venture capital. The European Agricultural Fund for Rural Development also benefits SMEs as it promotes entrepreneurship and encourages the economic diversification of rural areas. (EC, 2008: 11) Among other things, the EIB Group has established a new “Microfund� with an initial capital of around €40 million to support non-bank MFIs. The four most consistent EU instruments to support SMEs are JEREMIE, JASMINE, PROGRESS and CIP. JEREMIE, JASMINE and PROGRESS are specific initiatives of the European Commission to support the development of microcredit in the EU and are part of the EU Structural Funds. The Competiveness and Innovation framework Program (CIP) addresses SME's needs for financing at various stages of development and complements financial schemes at the national level. Part of CIP's budget is used to finance microcredit providers in the EU. 57 Cohesion policy does not finance individual projects directly at the EU level, but instead funds multi-annual national or regional operational programs aligned with EU objectives and priorities and managed by national or regional authorities. EU Structural Funds are implemented within the framework of shared management and the legislation defines a clear division of responsibilities between Member States and the Commission. Where the managing authorities decide to use financial engineering instruments to support enterprises, they contribute resources from an operational program to a holding fund or directly to financial engineering instruments such as venture capital funds, loan- or guarantee funds ("financial intermediaries"). Where a structure with a holding fund is used, this could be a suitably qualified financial institution selected through a transparent procedure, or alternatively the managing authorities can award a contract directly to the EIF to carry out the holding fund tasks. The role of a holding fund is to organize operations on behalf of the managing authorities, including for instance the selection of financial intermediaries. JEREMIE - Joint European Resources for Micro to Medium Enterprises - promotes the use of financial engineering instruments to improve access to finance for SMEs via Structural Funds interventions by means of equity, loans, or guarantees. It is an initiative of the European Commission's Directorate General for Regional Policy (DG REGIO) developed together with the European Investment Fund 17 . In the framework of 2007-2013 Structural Funds operational programs, Member States and regions can use financial engineering instruments to finance SMEs by means of repayable structures: resources invested in or paid to SMEs can, once paid back, be reinvested for the benefit of SMEs. Special emphasis is given to supporting the growth and jobs agenda, by emphasizing technology transfer, innovation, start-ups, and microcredit. JEREMIE has been implemented in all countries. Evaluation studies have been carried out in the context of the JEREMIE program for all five countries. These evaluation studies were aimed at identifying ‘market gaps’ in the financial services market for SMEs and micro-enterprises. Subsequently, JEREMIE has been implemented in all countries. In some cases, implementation has been challenging, and subsequent usage of the allocated funds has been low. Overall, however, implementation is considered to have increased the available funding for SMEs and micro-enterprises. Start-ups are only explicitly targeted by the JEREMIE program in Bulgaria, Hungary, Slovakia and Czech Republic. JASMINE - Joint Action to Support Microfinance INstitutions in Europe - targets non-bank micro- credit providers in the EU and makes available technical support and limited funding. By October 2011, JASMINE technical assistance was granted to 14 non-bank microcredit providers in Bulgaria (4), Hungary (3) and Romania (7), focusing particularly on microcredit for start-up businesses. This pilot initiative was launched in late 2008 for an initial 3 years ending in 2011; the final year of implementation has been extended for two years ending in 2013. Although the bulk of the support is technical assistance, JASMINE also includes a small provision of funding (EUR 4m) used to provide seed funding to five non- bank micro-credit providers to complement other sources of EU funding, such as EIB's 18. JASMINE's technical assistance is targeted specifically at the development of non-bank microcredit providers and includes services for selected institutions such as delivering assessments and ratings, training their staff, helping improve internal processes and disseminating best market practices. Generally speaking, non-bank microcredit providers address those individuals or entrepreneurs that are socially and financially excluded (i.e. cannot access bank credit), including ethnic minorities, who want to become self-employed. 17 See for details: http://ec.europa.eu/regional_policy/archive/funds/2007/jjj/doc/pdf/jeremie_sme_access.pdf 18 Personal communication with DG Regional Policy 58 The European PROGRESS Microfinance Facility is part of the ESF, and aims to increase accessibility of microcredit for those aiming to start or develop their own business by providing funding instruments, as well as by reducing the risk borne by Microcredit providers through the provision of guarantees. Implemented in 2010 as a joint initiative of the European Commission and the EIB, it is managed by the EIF with an overall budget of EUR 200m. PROGRESS issues guarantees to microcredit providers, sharing their potential risk of loss, and provides funding to increase microcredit lending volumes. Funding is provided through long-term loans to intermediaries, which in turn lend to micro- entrepreneurs, particularly those facing difficulties in accessing credit from traditional banking sources. More specifically, individuals who have lost or are at risk of losing their job and have difficulties (re-) entering the labor market are targeted by Progress. As of July 2012, the EIF had signed with microfinance providers under PROGRESS in Bulgaria (Mikrofond AD, JOBS Micro Financing Institution EAD) and Romania (Patria Credit, FAER, Banca Transilvania) 19. These intermediaries target existing SME, as well as start-ups among individuals or households undergoing financial or geographical exclusion. In sum, the findings reveal substantial challenges to substantially raising Roma self-employment levels through microcredit but also some opportunities to expand access to microcredit. This chapter has shown that there are substantial barriers – on the supply side and especially on the demand side - to starting a business and accessing start-up credit among aspiring Roma entrepreneurs. At the same time, funding from the EU and other international organizations, and to a certain extent from governments and local NGOs, could be used to initiate more tailored microcredit initiatives for Roma start-ups. There are several pilots, among which the Kiut program, which provide important lessons. However, substantially raising self-employment rates will require a comprehensive, incremental approach that goes beyond microcredit. 19 http://www.eif.org/what_we_do/microfinance/progress/Progress_intermediaries.htm 59 5 A COMPREHENSIVE, INCREMENTAL APPROACH TO FINANCIAL INCLUSION OF ROMA This final chapter highlights international experiences promoting comprehensive, incremental approaches to financial inclusion, highlighting how innovative approaches from rich-, middle income-, and poor countries to reaching the poor with financial services as well as specific examples to promoting (targeted) savings, financial literacy, and microcredit. 5.1 INTRODUCTION In June 2012, the World Bank published a Financial Inclusion Reference Framework as a resource for preparing new financial inclusion strategies. It emphasizes integrating supply side financial services provision with support to customers to take full advantage (World Bank, 2012a). This Framework highlights that: “Financial inclusion is emerging as a priority for policymakers and regulators in financial sector development, with an increasing number of countries introducing comprehensive measures to improve access to and usage of tailored financial services, informed by a fast-growing body of experience and knowledge. More than 60 countries have initiated financial inclusion reforms in recent years.� And, according to the Framework “A comprehensive strategy promotes not only the adoption of financial products and services, but also the ability of customers to take full advantage of them. Informed and financially educated customers will benefit more from financial services— for example, through a better understanding of what type of financial product best suits their individual needs, and more accurately assessing risks associated with that product.� Financial Inclusion is increasingly being promoted at EU level. The European Commission (EC) has followed an approach on financial inclusion geared towards basic bank account access, and focuses on ‘financial education’, i.e. citizens’ understanding of complex financial products. BOX 3: FINANCIAL INCLUSION – IN A BROADER SENSE – IS INCREASINGLY BEING PROMOTED AT EU LEVEL Financial Education: Since 2007, EU Guidelines have been in place to improve financial education in the Member States (EC, 2007). An Expert Group on Financial Education (EGFE) was established in 2008, and several initiatives were launched such as a Database for Financial Education and Dolceta, an online tool for consumer information and education that includes teaching materials and consumer learning modules on financial services. However, according to the EGFE, overall progress on the implementation and development of financial literacy initiatives has been limited between 2007 and 2011. Few Member States (the Czech Republic, the Netherlands, Portugal, Spain, the UK, and Sweden) have developed a country-wide strategy, and few have made financial education compulsory in school curricula. New goals and initiatives proposed by EGFE include: the design of national strategies by Member States, and to monitor progress in the various countries in this regard. Access to Basic Payment Accounts: In 2011, the European Commission issued a recommendation on access to a basic payment account (BPA) (EC, 2011b). This formally recognized the need for universal access to BPAs among EU consumers. The recommendation highlights the lack of national level guarantees for access to basic payment services, and the fact that providers of these services often do not ensure universal access either, at times even imposing restrictive eligibility criteria beyond what is required by law. The recommendation promotes a national guarantee of access to BPAs, as well as 60 fairness of treatment: access should be guaranteed irrespective of consumers’ financial circumstances, such as unemployment. Comprehensible information should be provided on the main features and conditions of BPAs. Charges for associated services should not prevent consumers from opening a BPA. Furthermore, providers should not offer overdraft facilities in conjunction with a BPA, and a payment order should not be executed where this would result in a negative balance. a Sources: EC, 2011b; EC, 2011c. To achieve financial inclusion of Roma, European countries can take advantage of the experiences of both rich and poor countries in tackling financial inclusion. Given the extremely high poverty levels, low levels of education, and general degree of marginalization and segregation of Roma, some of the successful lower income country efforts to achieving financial inclusion are very relevant. In particular, a holistic, incremental approach to financial inclusion that addresses basic barriers such as low financial literacy and the limited outreach of financial service providers in Roma communities, alongside rich country experiences that seek to achieve the ‘last mile’ of financial inclusion; i.e. achieving the final leg of providing access to financial services to all citizens. The ‘Graduation Approach’ provides an example of how poor households facing very high economic insecurity are being targeted in low and middle income countries through a comprehensive, incremental approach that addresses first the most obstructing barriers poor households face to accessing financial services, including through financial literacy training and a focus on savings: BOX 4: THE GRADUATION APPROACH: USING FINANCIAL INCLUSION TO CREATE PATHWAYS TO GRADUATE OUT OF POVERTY The Consultative Group to Assist the Poor (CGAP)-Ford Foundation Graduation Program is a global effort to understand how safety nets, livelihoods, and microfinance can be sequenced to create pathways for the poorest to graduate out of extreme poverty. The graduation model targets poor households facing extremely high economic insecurity. Ten pilots take place in Haiti, India in three places (with Bandhan, SKS, and Trickle Up), Pakistan, Honduras, Peru, Ethiopia, Yemen, and Ghana. The pilots involve diverse institutional forms, economic contexts, and cultures. The pilots are implemented through partnerships between financial service providers, nongovernmental organizations, and government units. Several of the pilots are measuring the program’s effects on people’s lives through rigorous randomized impact evaluations and qualitative research. The Graduation Approach addresses first the most obstructing barriers poor households face to accessing financial services. The initial focus on savings illustrates the latter’s crucial role in facilitating a successful transition to self-employment. The approach is built on five core elements: targeting, consumption support, savings, skills training and regular coaching, and an asset transfer. Program staff visit all selected households, enabling them to better understand the characteristics of poverty in their region. Consumption support addresses immediate consumption needs of the participating households. In addition, it is essential for the generation of trust towards program staff in an early stage of the program, and – through transparency and explicit emphasis on the duration of this support – allows households to plan ahead for when such support will no longer be available. Once the most immediate needs are addressed, households are encouraged and assisted to save, allowing the household to build assets and instilling financial discipline. Only in the final two steps, business skills training and the transfer of in-kind assets, does the purpose of self-employment come into play. The asset transfer is included to jump-start small businesses, increasing the chance of a successful transition to self- 61 employment. As such, the graduation approach provides a powerful illustration of a ‘step-wise’ approach toward financial inclusion and self-employment. Sources: GCAP Graduation Program Overview; Hashemi and De Montesquiou, 2011. CGAP: Haiti Chemen Lavi Miyo Program. The comprehensive, incremental graduation approach is very relevant to many Roma communities also facing very high levels of vulnerability as well as specifically for aspiring entrepeneurs. Its focus on savings, for example, before business skills training and asset transfer, is as relevant for the vast majority of Roma lacking a systematic means of savings as it is for the beneficiaries of the Graduation pilots. The evaluations being carried out of the graduation pilot programs may offer valuable lessons, for example on the methodology used and the impact of addressing ‘basic needs barriers’ and savings before focusing on microcredit. Moreover, elements from particular Graduation country pilots provide good examples of integrating financial inclusion with addressing other needs: in Haiti, for example, the program links financial inclusion with support for housing renovation and facilitates school attendance for participants’ children (CGAP: Haiti Chemen Lavi Miyo Program), both of which are pressing issues for many Roma. 5.2 INNOVATIVE WAYS TO REACH THE POOR WITH FINANCIAL SERVICES Different countries are using a variety of innovate means to reaching financially excluded groups with financial services such as savings and current accounts, incentives to save, and financial literacy. In promoting financial inclusion among Roma, European countries can benefit from these experiences. For example, in a high income country like the United States, the Bank on San Francisco initiative seeks to bring unbanked individuals into the financial mainstream through a collaborative effort between the government, private sector financial institutions, and civil society. In middle income country like Brazil, where reaching poor clients in rural areas is often prohibitively expensive for financial service providers, banks started using local agents to manage their operations at the village level. And in a lower income country like Kenya, the highly advanced M-PESA mobile phone service allows customers to save money, transfer funds to other mobile phone users, and make payments for services such as utility bills. BOX 5: BANK ON SAN FRANCISCO, UNITED STATES ‘Bank on San Francisco’ is a collaborative effort to bring 10,000 of the city’s estimated 50,000 un- banked individuals into the financial mainstream. The San Francisco Mayor’s Office, the Treasurer’s Office of the City and County of San Francisco, the Federal Reserve Bank of San Francisco, local non- profit EARN (Earned Assets Resource Network), and the city’s financial institutions have worked together to: - Increase the supply of starter account products that work for the low-income un-banked market by developing baseline product criteria that must be offered by all participating financial institutions; - Raise awareness amongst un-banked consumers about the benefits of account ownership and spur them to open accounts; - Make quality money management education more easily available to low-income San Franciscans; - Clamp down on the proliferation of check cashers and payday lenders; 62 - Raise city-wide awareness of the un-banked problem and potential solutions. Source: Stuhldreher, A.: Bank on San Francisco – An initiative to bring all residents into the financial mainstream (draft). BOX 6: ‘CORRESPONDENT BANKING’ IN BRAZIL Brazil’s ‘correspondent banking’ initiative exemplifies a creative solution to a widely occurring and highly obstructive barrier to accessing financial services. In Brazil, reaching poor clients in rural areas is often prohibitively expensive for financial service providers. In an initiative referred to as ‘correspondent banking’, banks in Brazil overcame this by using local agents to manage their operations at the village level. Agents may include retail vendors, lottery outlets, and post offices: these are trusted local establishments that can double as a substitute ‘bank branch’, processing everything from bill and pension payments to deposits, withdrawals, and money transfers. Banking agents’ lower set-up- and running costs promise to help banks and microfinance institutions reach more and poorer people living far from the nearest branch, with more financial products, at lower cost, than what could be accomplished through traditional banking channels alone. This approach is one example of an innovative way to bring poor people into the fold of formal financial services. (World Bank, 2006b) BOX 7: MOBILE FINANCIAL SERVICES: M-PESA KENYA Kenya’s mobile services have brought key financial services to millions of un-banked people in a matter of years. Since 2007, owners of mobile phones with a Safaricom cell phone company SIM card in Kenya can take advantage of a mobile phone-based payment and money transfer service called M-PESA – ‘mobile money’. With few people having access to regular bank accounts, M-PESA proved widely popular, and by August 2009, approximately 7.7 million M� PESA accounts had been registered, reaching 38 percent of the adult population. The service allows users to deposit money into an account stored on their cell phones, to send balances using SMS technology to other users, and to redeem deposits for regular money. The sender of so-called “e-float� is charged a flat fee of about 40 US cents, but the recipient only pays when s/he withdraws the funds (the cost of withdrawing Euro 100 is about Euro 1). There is no charge for depositing funds. Originally, transfers of e� float were expected to primarily reflect remittances between family members, but nowadays, e� float transfers are also used to pay directly for goods and services, including utility bills. In some cases, companies providing for example piped water or electricity even have pre-paid systems set-up where people pay for a certain quantity of water by sending e-float per SMS upon which water is released. Source: Jack, W., and T. Suri (2010). “The Economics of M-PESA� http://www.mit.edu/~tavneet/M-PESA.pdf; Safricom website: http://www.safaricom.co.ke/personal/m-pesa/how-to-register/faqs. Countries can also promote financial inclusion by taking advantage of government social protection payment systems. Many households – Roma and non-Roma, poor and non-poor – receive government social protection payments for pensions, family benefits, social assistance, education support, unemployment insurance etc. In many countries these payments are not necessarily made to people’s bank accounts but can be made in cash, for example through the post office. This is the case, for example, for social assistance payments in Slovakia. Governments can play a significant role in incentivizing financial 63 inclusion of Roma by transferring various benefits to accounts that beneficiaries – Roma and non-Roma - open in commercial banks. Providing social transfers into personal bank accounts may be an important incentive for private banks to agree with the Government on the provision of low cost “no frills� bank accounts for the poor. Also, a good system of E-payments can also provide significant savings to governments 20. For example, disbursing grants and funds electronically to beneficiaries with newly established bank accounts helped reduce the administrative costs of Brazil’s conditional cash transfer program Bolsa Familia from 14.7% to 2.6% of the disbursed grant value, and a recent McKinsey report suggests that the government of India could save $22.4bn annually by switching to electronic transfers21. There are many experiences from around the world that European countries can draw from to put in place a good - non-stigmatizing - system of E-payments to replace cash payments to poor households. See, for example: www.worldbank.org/paymentsystems. This website includes a report titled “General Guidelines for the Development of Government Payment Programs� (World Bank, 2012b), which presents a set of comprehensive guidelines that can assist governments and other stakeholders in designing and operating efficient government payment programs. 5.3 PROMOTE (TARGETED) SAVINGS A key component of the comprehensive incremental approach to financial inclusion is savings facilitation, and linking savings activities with human development outcomes. For example, governments can support households to open up targeted education- or housing savings accounts, in which households are encouraged to save with an explicit purpose. International evidence suggests that targeted savings are successful in raising overall savings. Several savings initiatives targeted vulnerable Roma can be adopted and scaled up, such as ETP Slovakia’s "Individual Development Account", which provides incentives to save for housing improvements. Similarly, there are international models such as the "Kindergarten-to-College (K2C)" savings initiative used in the USA or the labeling of savings accounts in Ghana. These are described below. BOX 8: EARMARKING SAVINGS: UNITED STATES AND GHANA The Kindergarten-to-College program in San Francisco, USA: In order to promote savings and economic mobility, children’s savings accounts have been established as long-term asset-building accounts that grow over time with additional deposits and earnings. San Francisco’s K2C initiative promises to open a college savings account for children entering kindergarten in San Francisco’s public schools, with an opening deposit of $50 from the City of San Francisco. Low-income families receive an additional $50 deposit from the city, and EARN, the program’s non-profit sponsor, has committed to provide an additional $100 to match the savings of the first group of students to participate. Source: www.earn.org Labeling savings accounts in Ghana raises overall savings levels: The focus on earmarking savings is not uncommon. In Ghana, the international organization Innovations for Poverty Action, which carries out 20 World Bank presentations at the Banking for Progress workshop in Bratislava on May 24, 2012. 21 Johnson, C., and P. Jaisinghani, We Must Do Better Than Cash. USAID Impact Blog, 2012. http://blog.usaid.gov/2012/02/we-must-do-better-than-cash/ 64 randomized counterfactual impact evaluations of innovative ideas, is working with Mumuadu Rural Bank (MRB) to study the response to, and impact of, a new account labelling savings product. It recruited 2100 study participants and found that customers with a labelled (i.e. ‘earmarked’ for housing improvements, school fees, etc.) savings account show a 31.2% increase in total deposits after nine months of account operations as compared to customers who were offered to open a savings account but were not offered the option to earmark it. Source: http://www.poverty-action.org/project/0071 A Slovak microsavings example focusing on Roma is ETP’s Individual Development Account. A central element of ETP’s Individual Development Account (IDA) project has been the provision of targeted savings toward home improvements, providing a way to link financial inclusion to improved housing conditions. BOX-9: ‘TARGETING SAVINGS’ PROJECT AMONG ROMA IN SLOVAKIA ETP’s Individual Development Account is a microsavings initiative which has been made available for 400 individuals in Eastern Slovakia. Since its launch in 2006, it teaches mainly Roma families how to resist loan sharks, better manage limited funds through the month, and – most importantly – think about the future prospects for the whole family. While the program is small, ETP has developed a very important knowledge base of financial inclusion in the marginalized communities through years of fieldwork, which can be capitalized on. ETP describes its project as follows: “The IDA project is focused on marginalized populations with inappropriate living and housing conditions. The savers are saving their money for a certain amount of time, usually 1 or 2 years, for a certain set goal. Some of them save for roof reconstruction or house expansion, others save for new windows or bathroom. The goal of this program is to teach the participants to think forward and plan their future as well as to teach them to save money, despite their modest income.� The terms and conditions of the Savings Program are clear but strict. The clients have to open a savings account and monthly deposit a certain amount of money, usually 10-50 Euros. During their saving period, the clients are neither allowed to make any cash withdrawals nor miss their monthly deposit. Those who meet the requirements of the program receive a bonus equal to the amount on their saving account. The clients can also participate in a Micro-loan Program after they have finished saving. Source: http://www.etp.sk/en/sporiaci-program-ida/#more-92 5.4 SUPPORT FINANCIAL LITERACY AND DEBT MANAGEMENT To successfully raise savings and increase household investments, basic financial literacy training and debt management training must accompany a focus on savings facilitation. Excellent examples from around the world are plentiful, but countries can also build on locally implemented initiatives that can be scaled up. There are many financial literacy training modules – even for school children. In Slovakia, the NGO most active in the area of small-scale financial inclusion activities with disadvantaged Roma is ETP. Together with Autonomia Foundation, it has been working with standardized financial training modules that have proven successful entry points into microsavings and microcredit programs for clients living in marginalized communities. An important entry point to increasing financial literacy among Roma are schools where teenagers can be targeted. 65 BOX 10: FINANCIAL LITERACY TRAINING AMONG ROMA IN SLOVAKIA The financial education course developed by ETP Slovakia aims to teach clients to use their finances wisely as well as to help them escape from the trap of debts and give them a new beginning. The financial education course consists of two modules. The first module („Don’t Be Afraid of Money“) is for entry-level program participants, children and youth. In this module, the client learns basic knowledge and information about financial management, household budgeting and not spending money beyond means. The second module consists of 6 volumes and is for the intermediate level (i.e. graduates of the first module). This program is conducted in groups in community centers. The participants gain knowledge, skills and attitudes that are necessary for good financial management of their earnings, expenses, savings and investments. The clients learn to plan their finances as well as saving for certain goals in the future. Both courses inform participants about the dangers associated with the excessive interest rates of loans provided by non-banking entities, and how this becomes a trap for clients by creating a vicious circle of growing debts. After the completion of both courses, the successful graduates become eligible to participate in microsavings and microcredit programs administered by ETP Slovakia. Source: www.etp.sk 5.5 SUPPORT ACCESS TO MAINSTREAM MICROCREDIT Microcredit can be an important tool for supporting (aspiring) Roma entrepreneurs. It is unlikely to substantially raise employment rates among Roma without a comprehensive financial inclusion approach in place addressing vulnerability and skills challenges. The 2011 regional Roma survey data show that the vast majority of Roma want to work, and many are interested in becoming self-employed. Among current entrepreneurs, lack of finance is considered a main obstacle to starting a business. Yet, the potential group of Roma that can take advantage of microcredit is small as most Roma interested in becoming self-employed face significant barriers that limit access to credit. These barriers include indebtedness, little employment- and professional experience, and very low levels of education: even when compared to that segment of the general population that is being refused credit. The recent social microcredit experience of the Hungarian Kiút program (box below) provides an important example. BOX 11: EXPERIENCES WITH MICROCREDIT AMONG STARTING ROMA ENTREPRENEURS: THE KIUT PROGRAM IN HUNGARY The Kiut Program is a two-year microcredit pilot program in Hungary implemented by the Polgar Foundation in collaboration with Raiffeisen Bank, with EU financing. The purpose of the project is to empower poor households, mainly Roma, to become self-employed through microcredit with a goal of reaching up to 300 loans by 2012. At the project start in mid-2010, its field officers sought out potential borrowers in poor communities, especially in Eastern Hungary, offering relatively small-sized, unsecured loans of up to EUR 3,500 for a period of 0.5-1.5 years. The loans are provided in a group setting, so that peer pressure – but not full joint liability – is used to secure the loans. Loans are generally provided sequentially 22 and saving is encouraged among members of the group. An ideal group consists of 5 members, but in practice contains a minimum of 3 and maximum of 6 members. The Kiut program supports selected entrepreneurs in registering their business and the first loan is disbursed. 22 I.e. credit is given to one borrower at a time under the condition that the previous borrower repaid her loan. 66 The program succeeded in reaching out to poor Roma families. Project summary statistics based on interviews with serious loan applicants 23 showed that a ‘typical’ applicant is Roma, male, in his late thirties, either married or living with a spouse, in a household of 5 members, and in a mixed neighborhood, most likely in a village. The applicant generally lives far (around half an hour or longer) from (a) a bank office where a loan could be taken and paid; (b) the local employment office, and (c) any secondary school. The applicant is typically literate but does not have advanced education – usually primary or some vocational. He has been unemployed for nearly two years although has a formal employment history spanning more than 10 years. However, he does not expect to be able to take a regular commercial loan of approximately 500 Euros. Furthermore, the typical applicant has low satisfaction with life in general, and especially with the financial situation of the household. The Kiut early experience underscores the challenges to lend to this group of borrowers. Staff drew several early lessons from the program in May 2011. First, despite the initial goal of concluding loan agreements with 100 customers by the end of the first year, fewer than 50 loans were disbursed, owing to the difficulty of finding good candidates. Even then, by early Winter 2011, many businesses experienced difficulty and there was an increase in the number of late- or non-payment of loan installments, and by May 2011 the program management decided to change the model of the program, most importantly requiring field agents to put greater emphasis on identifying potential entrepreneurial capabilities, even if this may mean in practice attracting fewer clients from the most disadvantaged groups. In addition, the program decided that counseling and training that had been conducted informally by the field workers and crisis managers so far, should be done in a more organized manner. As a result of these changes, Kiut has since successfully reached more than 100 borrowers thus far high with repayment rates. Source: Personal Communication and Kiut Monitoring Data (2011-2012) The report points to several key findings related to the use and potential of microcredit among Roma: (a) a significant minority of Roma do make use of credit, but not always from formal lending institutions such as banks and credit cooperatives, and rarely for productive purposes; (b) current levels of formal businesses are generally low among Roma, but there is substantial interest in self-employment and starting a business, both among those who are currently employed and among those who are currently unemployed; (c) an analysis of the demand side of the credit market shows that potential Roma start-up entrepreneurs, in particular, face various barriers to accessing (business) credit; (d) moreover, not many suppliers of microfinance target start-ups businesses in general, and few are reaching (potential) Roma entrepreneurs, while those that do reach Roma are usually small-scale and face various challenges; and, (e) there are several EU level programs supporting microfinance, also specifically targeting vulnerable populations such as the JASMINE program. Non-profit initiatives can promote greater access to microcredit for (aspiring) Roma entrepreneurs by identifying promising entrepreneurs and supporting these entrepreneurs meet the eligibility requirements of mainstream microcredit providers. While some small-scale projects set up by NGOs and MFIs have successfully reached some poor Roma families or MFIs, the number of loans disbursed is very small, the cost per loan large, and the institutional set-up frequently leaves out depository institutions thus limiting the scope to promote savings. Instead of parallel microcredit schemes, a more promising avenue may be the establishment of partnerships between pro-social non-for-profit initiatives on the one hand, and mainstream credit providers – ideally depository ones - that provide microcredit to self- employed. The non-for-profit support can focus on (a) identifying promising (aspiring) Roma entrepreneurs, (b) support these meet the conditions of mainstream credit providers – including business 23 Kiut field officers carry out a detailed interview with serious loan applicants. This extensive monitoring system was developed in collaboration with the World Bank and UNDP, and financed by DG Regional Policy. 67 plan-, financial literacy- and savings support (for down-payment), possibly in combination with asset transfer (for collateral or as start-up grant). Countries can leverage EU structural funds to support these pro-social initiatives. 5.6 USE MONITORING AND EVALUATION TO PROMOTE FINANCIAL INCLUSION Monitoring of financial inclusion indicators and rigorously evaluating programs promoting financial inclusion is critical to achieving results. This is also recognized by the April 2011 EU Framework for National Roma Integration Strategies 24, which calls upon EU member states to include strong monitoring and evaluation components. Fortunately, there are well-established monitoring surveys and there is considerable global experience with rigorous impact evaluations of financial inclusion initiatives, ranging from financial literacy support, to (targeted) savings, to the impact of microcredit and start-up grant support. There are several useful global resources on M&E of financial inclusion, including (a) the World Bank’s financial inclusion program – www.worldbank.org/financialinclusion; and, (b) the financial inclusion (FINDEX) database - http://datatopics.worldbank.org/financialinclusion. For example, FINDEX measures how people save, borrow, make payments, and manage risk. The Financial Capability Survey (CPFC) survey captures the way people manage their money, their ability to budget and to stick to the budget, to plan ahead, to make provisions for their retirement age as well as for expected and unexpected expenses and by identifying how people chose financial products and to which financial products and services people have access. Another resource is the (c) the Global Financial Inclusion initiative at Innovations for Poverty Action - http://www.poverty-action.org/financialinclusion - , which ‘manages two research funds seeking to test, evaluate and replicate interventions to improve products, delivery channels, and tools to help users make the most of their financial portfolios.’ Social policy experimentations – also called prospective randomized impact evaluations - provide rigorous evidence on financial inclusion project impacts. These impact evaluations involve randomly selected treatment and comparison groups. Randomization ensures the two groups have identical characteristics at the start of the program. This generally ensures that any differences that arise between the two groups – for example in education or labor market outcomes – can be attributed to the intervention and not to pre-existing differences that are often present if groups self-select into a project. For this reason, these randomized control trials are often considered the most robust methodology for conducting program impact evaluations. Promoted by the EU PROGRESS facility, and implemented around the globe by governments, civil society, and international organizations, these can identify the most cost-effective financial inclusion interventions and build public support around proven programs for Roma inclusion. The resources above provide examples of such impact evaluations. Other resources include: www.worldbank.org/dime and http://www.povertyactionlab.org. 24 COM(2011) 173 Final: An EU Framework for National Roma Integration Strategies up to 2020. http://ec.europa.eu/justice/policies/discrimination/docs/com_2011_173_en.pdf 68 ANNEX A: EU INITIATIVES AND POLICIES FOR SME S POLICY: THE EUROPEAN CHARTER FOR SMALL ENTERPRISES AND THE SMALL BUSINESS ACT (SBA) THE CHARTER The European Charter for Small Enterprises defines access to finance as one of its key priorities. The Charter – established in 2000 – is “a self-commitment from the Member States to improve the business environment for small enterprises� (EC, Enterprise and Industry: Charter). The charter was signed by the EU member states in 2000, and by the 13 at-the-time candidate member states in 2002 as the ‘Maribor Declaration’. “The Maribor Conference viewed inadequate access to finance as the most serious barrier to entrepreneurship� (Europolitics, 2002). The charter includes a commitment of the member states to: “Achieve a regulatory, fiscal and administrative framework conducive to entrepreneurial activity (0...); Improve access to finance throughout the entire life-cycle of an enterprise (0...); Listen to the voice of small business; Promote top-class small business support� (EC, Enterprise and Industry: Charter: 1). The Charter includes 10 lines of action, four of which are particularly relevant in the context of this report 25: “1. Education and training for entrepreneurship: (0...)Specific business-related modules should be made an essential ingredient of education schemes at secondary level and at colleges and universities. We will encourage and promote youngsters’ entrepreneurial endeavours, and develop appropriate training schemes for managers in small enterprises. 2. Cheaper and faster start-up: The costs of companies' start-up should evolve towards the most competitive in the world. Countries with the longest delays and most burdensome procedures for approving new companies should be encouraged to catch up with the fastest. Online access for registration should be increased. (0...) 7. Taxation and financial matters: (0...) In order to improve the access of small enterprises to financial services, we will: · Identify and remove barriers to the creation of a pan-European capital market and to the implementation of the Financial Services Action Plan and the Risk Capital Action Plan; · Improve the relationship between the banking system and small enterprises by creating appropriate access conditions to credit and to venture capital; · Improve the access to the structural funds and welcome initiatives by the European Investment Bank to increase funding available to start-ups and high-technology enterprises, including equity instruments. 9. Successful e-business models and top-class small business support: “We will co-ordinate Member States and EU activity to create information and business support systems, networks and services which are easy to access and understand, and relevant to the needs of business; ensure EU-wide access to 25 The remaining lines of action are: 3. Better legislation and regulation, 4. Availability of skills, 5. Improving online access, 6. More out of the Single Market – including “European and national competition rules�, which “should be vigorously applied to make sure that small businesses have every chance to enter new markets and compete on fair terms� (ECSE, 2000: 3) –, 8. Strengthen the technological capacity of small enterprises, and 10. Develop stronger, more effective representation of small enterprises’ interests at Union and national level (ECSE, 2000). 69 guidance and support from mentors and business angels, including through websites, and exploit the European Observatory on SMEs.� (ECSE, 2000: 1-4, italics removed from original) THE SMALL BUSINESS ACT The Small Business Act (SBA) puts into place a comprehensive SME policy framework for the EU and its Member States. The SBA is centered on the belief that SMEs thrive in an environment that recognizes the value of entrepreneurs. The SBA builds on the European Charter for Small Enterprises and the Modern SME policy. “The general climate in society should lead individuals to consider the option of starting their own business as attractive, and acknowledge that SMEs contribute substantially to employment growth and economic prosperity. (0...) [Hence,] entrepreneurship and the associated willingness to take risks should be applauded by political leaders and the media, and supported by administrations,� (EC, 2008: 3). The Small Business Act for Europe applies to all independent companies which have fewer than 250 employees: 99% of all European businesses. It was adopted in 2008, and is embedded in the EU Growth and Jobs Strategy. The SBA aims to: - Improve the overall approach to entrepreneurship, - Anchor the 'Think Small First' principle in policy making (i.e. prioritize policies for SMEs) - Promote SMEs' growth by helping them tackle problems which currently hamper their development. To realize these aims, the SBA includes ten guiding principles, including the following 26: 1. Create an environment in which entrepreneurs and family businesses can thrive and entrepreneurship is rewarded. In this light, the SBA points to the importance of the following: “Recognition of the special role of SMEs and in particular family-based enterprises, their typically local base, socially responsible attitudes and capacity to combine tradition with innovation, underpins the importance of simplifying the transfer of businesses and the skills associated with them.� The SBA also recognizes that “There is a continuing gender gap in terms of entrepreneurship, which translates into fewer women entrepreneurs. This adds to an unexploited potential for entrepreneurship among immigrants.� (EC, 2008: 5) “[T]he Member States are invited to: Stimulate innovative and entrepreneurial mindsets among young people by introducing entrepreneurship as a key competence in school curricula, particularly in general secondary education, and ensure that it is correctly reflected in teaching material (0...); provide mentoring and support for female entrepreneurs; provide mentoring and support for immigrants who wish to become entrepreneurs,� (EC, 2008: 6). 3. Design rules according to the “Think Small First� principle. In this regard, the Member States are invited to: “use specific measures for small and micro-enterprises, such as derogations, transition periods and exemptions, in particular from information or reporting requirements, and other tailor-made approaches, wherever appropriate (0...) and adopt the Commission proposal which would permit Member States to increase the threshold for VAT registration to €100 000.� (EC, 2008: 8) 26 The remaining guiding principles are: 2. Ensure that honest entrepreneurs who have faced bankruptcy quickly get a second chance; 5. Adapt public policy tools to SME needs: facilitate SMEs’ participation in public procurement and better use State Aid possibilities for SMEs; 7. Help SMEs to benefit more from the opportunities offered by the Single Market; 9. Enable SMEs to turn environmental challenges into opportunities; 10. Encourage and support SMEs to benefit from the growth of markets (EC, 2008). 70 4. Make public administrations responsive to SMEs’ needs. “Member States are invited to: reduce the level of fees requested by the Member States’ administrations for registering a business (0...); reduce the time required to set up a business to less than one week (0...) accelerate the start of SMEs’ commercial operations by reducing and simplifying business licences and permits (0...); establish a contact point to which stakeholders can communicate rules or procedures which are considered to be disproportionate and/or unnecessarily hinder SME activities; ensure full and timely implementation of the Services Directive, including the setting up of points of single contact, through which businesses can obtain all relevant information and complete all necessary procedures and formalities by electronic means.� (EC, 2008: 9). 6. Facilitate SMEs’ access to finance, in particular to risk capital, microcredit and mezzanine finance and develop a legal and business environment supportive to timely payment in commercial transactions. “Risk aversion often makes investors and banks shy away from financing firms in their start-up and early expansion stages. (0...) Furthermore, many entrepreneurs need guidance and education on the advantages and disadvantages of different forms of finance and on how to best present their investment projects to potential financiers. In addition, SMEs often have a weak equity position (0...). The Member States are invited to: (0...) ensure that the taxation of corporate profits encourages investment; [and] make full use of funding available in cohesion policy programs and the European Agricultural Fund for Rural Development, in support of SMEs.� (EC, 2008: 11-12) 8. Promote the upgrading of skills in SMEs and all forms of innovation. “More than 60% of companies responding to the consultation consider that schools do not provide the competences needed by entrepreneurs and their staff. (0...) SMEs suffer in particular from the lack of skilled labour in the field of new technologies. In this context, the European Reference Framework on key competences for lifelong learning, which aims at offering the necessary basic skills to all young people, needs to be implemented.� The SBA also includes the following legislative proposals: – General Block Exemption Regulation on State Aids (GBER): to exempt from prior notification categories of State Aid related to training, employment, R&D for SMEs. The new Regulation will simplify and harmonise existing rules for SMEs and increase investment aid intensities for SMEs. – Regulation providing for a Statute for a Private European Company (SPE): to provide one uniform statute for all member states. Amending proposals will also be made to ensure that this new company form can benefit from the existing corporate tax directives. – Directive on reduced VAT rates: to offer Member States the option of applying reduced value added tax (VAT) rates principally for locally supplied services, which are mainly provided by SMEs. Overall, the SBA emphasizes online communication, both between regulatory authorities and SMEs, and among SMEs and their customers. In addition, the SBA includes a commitment of the Commission to provide an EU-wide platform for best practice exchange (EC, 2008). DATA COLLECTION: SME PERFORMANCE REVIEW The SME Performance Review is the main EU Instrument for data collection on the performance of SMEs in Europe. The Observatory of European SMEs – established in December 1992 – had as its main aim to improve the monitoring of the economic performance of SMEs in Europe, providing information to policy-makers, researchers, economists and SMEs themselves. The last edition was published in 2007 71 (EC, SME Observatory, 2007). Since 2008, the Observatory has been replaced by the SME Performance Review. The SME Performance Review represents one of the main tools employed by the European Commission to monitor implementation of the Small Business Act (SBA), as well as a comprehensive source of information on SME performance in Europe. The SME Performance Review comprises: - An Annual Report - SBA fact sheets per country - Additional studies providing background information on particularly relevant issues for European SMEs THE SBA FACTSHEETS The Small Business Act (SBA) fact sheets are statistical information tools designed as an input for assessing the business and policy environment for SMEs in Member States (0...) and Partner Countries. They should be an interesting information source for policy makers in the Commission and in the individual countries, as well as for stakeholders and the public at large. The purpose of the fact sheets is to provide a snapshot of how the SME environment looks like in a given country (0...)� (Kwaak, 2010: 5). - Do not constitute a comprehensive assessment of member states’ policies, but only one input for it; - Should be regarded as supplement to, and not a substitute for, any other relevant in-depth studies of SME policies either by the European Commission or member-states-based publications - Data collected for the factsheets are harmonized by Eurostat after official submission by the member states. Definitions used by member states may differ. - The fact sheets do not provide a real-time reflection of the current situation. Most of the data in the fact sheet comes with a two-to-three-year delay. In most cases, the impact of this is limited, as many indicators are rather slowly changing structural indicators without huge and sudden oscillations. Nonetheless, there may be years and periods where such sudden changes occur. - The SBA fact sheets do leave out some policy areas with an either direct or indirect impact on SMEs performances, such as macro-economic or infrastructure indicators. 72 ANNEX B: ESTIMATION RESULTS ANNEX TABLE 1. INTERESTED IN BECOMING SELF-EMPLOYED CZECH POOLED BULGARIA HUNGARY ROMANIA SLOVAKIA REPUBLIC (N=3,646) (N=787) (N=784) (N=590) (N=742) (N=743) -0.08*** -0.08*** -0.12*** -0.08*** -0.03 -0.09*** Female (0.01) (0.03) (0.03) (0.03) (0.04) (0.03) 0.00 0.01 -0.02 0.03* 0.00 0.01 Age (0.01) (0.02) (0.01) (0.01) (0.02) (0.02) -0.00 -0.00 .00 -0.00** -0.00 -0.00 Age squared (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) -0.03*** -0.06*** -0.10 -0.02 .01 -0.12** Schooling years (0.01) (0.02) (0.07) (0.03) (0.02) (0.05) 0.00*** 0.00*** 0.01 0.00 0.00 0.01*** Schooling years squared (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) 0.00 -0.01 0.00 0.00 0.01* 0.01 Working years (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) Able to use a computer 0.05** 0.09 0.08* 0.00 0.07 -0.00 word processing program (0.03) (0.07) (0.05) (0.05) (0.07) (0.05) Completed adult learning 0.09*** 0.09 0.19** 0.11** 0.01 -0.02 course / apprenticeship (0.03) (0.08) (0.08) (0.05) (0.08) (0.06) 0.08** 0.04 0.03 0.14** 0.10 0.15 Informal employment (0.03) (0.07) (0.06) (0.07) (0.06) (0.13) 0.12*** 0.16*** 0.05 0.18*** 0.15** 0.02 Unemployed (0.02) (0.06) (0.05) (0.04) (0.06) (0.05) -0.03 0.14* -0.05 -0.03 -0.27*** -0.02 Rural area (0.03) (0.08) (0.16) (0.06) (0.08) (0.06) 0.00 -0.02* 0.01 0.00 -0.01 0.00 Distance from bank office (0.00) (0.01) (0.01) (0.00) (0.01) (0.00) Distance from municipal -0.01*** -0.03*** -0.03** -0.00 -0.01* -0.00 office (0.00) (0.01) (0.01) (0.01) (0.01) (0.01) Distance from employment -0.00 0.02 0.01 -0.01 0.02** -0.01* office (0.00) (0.01) (0.01) (0.00) (0.01) (0.00) -0.25*** Czech Republic (0.03) -0.19*** Hungary (0.03) 0.07* Romania (0.04) -0.25*** Slovakia (0.03) Predominantly Roma 0.06** 0.02 -0.02 0.07 0.00 0.19*** ethnicity neighborhood (0.02) (0.06) (0.04) (0.05) (0.06) (0.04) Source: UNDP/World Bank/EC regional Roma survey (2011). Cluster-robust standard errors in parentheses. Probit model. *** p<0.01, ** p<0.05, * p<0. a Cluster-robust standard errors. Clusters are households; b The two categories of the outcome variable are ‘interested (yes or maybe)’, and ‘not interested’ in becoming self-employed. c Sample restricted to employees and unemployed aged 20-54 years old, as well as those defined as ‘self-employed’ according to the broad definition explained on page …, but who are also a) not formally registered, b) unskilled, and c) in temporary employment; A unit change in a continuous variable (or a discrete change in a dummy variable) leads to a coefficient*100 percentage change in the probability of being interested in becoming self-employed. 73 ANNEX TABLE 2: LIKELIHOOD OF BORROWING AMONG ROMA AND NON-ROMA NEIGHBORING HOUSEHOLDS CURRENT AND PAST CURRENT BORROWING BORROWING (N=5,383) (N=5,383) Roma household -0.02 -0.01 (0.02) (0.02) Romani Spoken at Home -0.05** -0.04** (0.02) (0.02) Enumeration area fixed effects Yes Yes Household controls Yes Yes Adjusted-R2 0.17 0.17 F 17.79*** 13.99*** Source: UNDP/World Bank/EC regional Roma survey (2011). Cluster-robust standard errors in parentheses. OLS Estimates. *** p<0.01, ** p<0.05, * p<0.1 A Household controls included in the model are the household head’s gender, age, and education level (at least secondary education or less), home ownership, household size, and household income quintile (measured by country and ethnic group). 74 ANNEX TABLE 3. DIFFERENCES IN BORROWING BEHAVIOR AMONG ROMA HOUSEHOLDS. CURRENT AND PAST CURRENT BORROWING BORROWING (N=3,663) (N=3,663) Romani Spoken as Main Language at Home -0.08*** -0.06*** (0.02) (0.02) Female Household Head 0.01 0.00 (0.02) (0.02) Household Head has at least some secondary 0.02 0.03* education (0.02) (0.02) Household Head’s Age 0.02*** 0.01*** (0.00) (0.00) Household Head’s Age Squared -0.00*** -0.00*** (0.00) (0.00) Number of Household Members 0.02*** 0.02*** (0.00) (0.00) Dwelling Ownership 0.01 0.00 (0.02) (0.02) INCOME: Second Quintile 0.01 0.02 (0.03) (0.02) Third Quintile 0.05* 0.06** (0.03) (0.02) Fourth Quintile 0.12*** 0.12*** (0.03) (0.02) Fifth Quintile 0.13*** 0.11*** (0.03) (0.03) Missing Information on Household Income 0.03 -0.03 (0.02) (0.02) Bank Close to Dwelling (<3km) 0.03* 0.01 (0.02) (0.02) Household in Rural Settlement 0.04** 0.03 (0.02) (0.02) Constant -0.19*** -0.17** (0.07) (0.07) REGIONAL FIXED EFFECTS YES YES ADJUSTED-R2 0.07 0.06 F-STATISTICR2 11.23*** 10.89*** Source: UNDP/World Bank/EC regional Roma survey (2011). Cluster-robust standard errors in parentheses. OLS Estimates. *** p<0.01, ** p<0.05, * p<0.1 75 ANNEX TABLE 4. DIFFERENCES IN USAGE OF FINANCIAL SERVICES AMONG ROMA HOUSEHOLDS CURRENT SAVINGS DEBIT CARD CREDIT CARD ACCOUNT ACCOUNT (N=3,546) (N=3,540) (N=3,606) (N=3,546) Household Head has at Least Some Secondary 0.10*** 0.01 0.09*** 0.04*** Education (0.02) (0.01) (0.02) (0.01) -0.05*** 0.00 -0.06*** -0.02 Romani Spoken as Main Language at Home (0.02) (0.01) (0.02) (0.01) -0.00 -0.00 0.02*** 0.00 Number of Household Members (0.00) (0.00) (0.00) (0.00) 0.06*** 0.01 0.02 0.00 Dwelling Ownership (0.02) (-0.01) (0.02) (0.01) Household Income: B, C 0.01 -0.00 0.06** 0.01 Second Quintile (0.02) (0.00) (0.02) (0.01) 0.01 0.00 0.08*** 0.01 Third Quintile (0.02) (0.01) (0.03) (0.01) 0.09*** 0.02** 0.15*** 0.02 Fourth Quintile (0.03) (0.01) (0.03) (0.02) 0.17*** 0.06*** 0.25*** 0.05*** Fifth Quintile (0.03) (0.01) (0.03) (0.02) 0.10*** 0.02*** 0.10*** 0.03*** Household has Some Earnings from Employment (0.02) (0.01) (0.02) (0.01) 0.04* 0.01 0.01 0.02 Bank Close to Dwelling (<3km) (0.02) (0.01) (0.03) (0.02) -0.00 -0.01 Household in Rural Settlement (0.02) (0.01) Controls for basic Household Head characteristics B, C Yes Yes Yes Yes Regional Fixed Effects Yes Yes Yes Yes Adjusted-R2 0.23 0.04 0.38 0.18 F-statistic 14.60*** 3.53*** 15.42*** 2.88*** A Source: UNDP/World Bank/EC regional Roma survey (2011). Cluster-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1; B Basic characteristics controlled for in these models are the gender, age and squared age of the household head. None of these were significant. C Income quintiles were based on Adult Equivalent Income levels. In addition to the five quintiles, a sixth category was included for households in which information on income was missing. 76 ANNEX TABLE 5. ABILITY OF ROMA HOUSEHOLDS TO FACE UNEXPECTED EXPENSES MODEL 1 MODEL 2 MODEL 3 (N=3,466) (N=3,466) (N=3,464) 0.27*** 0.28*** 0.26*** Savings Account (0.05) (0.06) (0.06) 0.04 0.05** 0.02 Current Account (0.03) (0.02) (0.02) 0.05** 0.04** 0.03 Debit Card (0.02) (0.02) (0.02) 0.07** 0.08** 0.07* Credit Card (0.03) (0.04) (0.04) Regional Fixed Effects Yes No No Enumeration Area Fixed Effects No Yes Yes Household Controls A,B No No Yes 2 Adjusted-R 0.06 0.19 0.22 F-statistic 18.94*** 13.06*** 7.273*** Source: UNDP/World Bank/EC regional Roma survey (2011); Cluster-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1; B Data on the household’s ability to face unexpected expenses was based on the question: ‘Can you afford unexpected required expenses (in the amount of the household's monthly income) and pay these through the household’s own resources?’ Household controls included in the model are the household head’s gender, age, and education level (at least secondary education vs. less than secondary), home ownership, household size, household income quintile (measured by country and ethnic group), and whether Romani is spoken as the main language at home. B Income quintiles were based on Adult Equivalent Income levels. 77 BIBLIOGRAPHY Bulgarian National Bank. 2011. “Distribution of Banks by Group as of 31 January 2011.� Bulgarian National Bank, Sofia, available at http://www.bnb.bg/bnbweb/groups/public/documents/bnb_download/bs_201101_a0_en.pdf. Carbo, S., E. P. M. Gardener, and P. Molyneux. 2005. Financial Exclusion. New York: Palgrave MacMillan. Carboni, B. J., M. L. Calderon, S. R. Garrido, K. Dayson, and J. Kickul. 2010. A Handbook of Microcredit in Europe: Social Inclusion through Microenterprise Development. Cheltenham, UK: Edward Elgar. CGAP (Consultative Group to Assist the Poor). “Graduation Program Overview.� GCAP, http://www.cgap.org/p/site/c/template.rc/1.26.1467/. ———. “Haiti Chemen Lavi Miyo Program.� CGAP, http://graduation.cgap.org/pilots/chemen-lavi- miyo/. CGAP and the World Bank. 2010. “Financial Access Indicators, 2010: The State of Financial Inclusion through the Crisis.� CGAP and World Bank, Washington DC. Damianova, A. et al. 2008. “Annual Report on the Condition and Development of SMEs in Bulgaria.� Center for Economic Development Study, Bulgarian Ministry of Economy, Energy and Tourism, Sofia. EC (European Commission). 2000. “European Charter for Small Enterprises.� European Commission, Brussels, available at http://ec.europa.eu/enterprise/policies/sme/files/charter/docs/charter_en.pdf. ———. 2005. “Cohesion Policy in Support of Growth and Jobs, Community Strategic Guidelines 2007- 2013.� European Commission, Brussels. ———. 2007. “Financial Education.� Communication from the European Commission COM(2007)808, European Commission, Brussels. ———. 2008. “Think Small First: A ‘Small Business Act’ for Europe.� European Commission, Brussels. ———. 2009. Enterprise and Industry, SBA Factseets. http://ec.europa.eu/enterprise/policies/sme/facts- figures-analysis/performance-review/index_en.htm#maplinks. ———. 2010. “Gaining Scale in Microcredit: Can Banks Make it Happen?� European Commission, Brussels. ———. 2011a. “An EU Framework for National Roma Integration Strategies up to 2020.� Communication from the European Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, Brussels, April 5. ———. 2011b. “Commission Recommendation on Access to a Basic Payment Account.� European Commission, Brussels, available at http://ec.europa.eu/internal_market/finservices- retail/docs/inclusion/c_2011_4977_en.pdf. 78 ———. 2011c. “Review of the Initiatives of the European Commission in the Area of Financial Education.� Staff Working Document of the Internal Market and Services DG (March), European Commission, Brussels, available at http://ec.europa.eu/internal_market/finservices- retail/docs/capability/evaluation_financial_education_en.pdf. ———. n.d. “Small and Medium Enterprises (SMEs) Definition.� European Commission, http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/sme-definition/index_en.htm. ———. Enterprise and Industry. “Charter.� European Commission, http://ec.europa.eu/enterprise/policies/sme/best-practices/charter/. ———. SME Observatory. 2007. http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/sme- observatory/index_en.htm. Ehrbeck, T. 2011. “Full Financial Inclusion in Our Lifetime.� Consultative Group to Assist the Poor, Washington, DC. EIF. 2009. “Evaluation Studies on SME Access to Finance in EU Member States/Regions Carried out by the EIF in the Context of the JEREMIE.� European Investment Fund, Luxembourg, http://ec.europa.eu/regional_policy/funds/2007/jjj/doc/pdf/jeremie_sme_access.pdf. EMN (European Microfinance Network). 2009. “EIF Market Studies on Micro Lending in the European Union – Cross Country Study on Central and Eastern Europe: Bulgaria, Czech Republic, Romania, Slovakia.� European Microfinance Network, Brussels. EMN and Latham & Watkins. 2008. “Legal Survey on Microcredit in Europe.� http://www.european-microfinance.org/data/file/emn_lw_jurisdiction_summary_final_.doc. ENAR (European Network Against Racism). 2011a. “Racism and Discrimination in Romania.� European Network Against Racism Shadow Report 2009-2010. The International Law Research and Human Rights Monitoring Centre, European Network Against Racism, Brussels. ———. 2011b. “Racism and Discrimination in Slovakia.� European Network Against Racism Shadow Report 2009-2010. The International Law Research and Human Rights Monitoring Centre, European Network Against Racism, Brussels. Eurom Consultancy and Studies. 2010. “Benchmarking Microfinance in Romania: 2008-2009.� Eurom Consultancy and Studies, Bucharest. Europolitics. 2002. “Enlargement: Maribor Declaration on Small Firms Published.� May 13, http://www.europolitics.info/enlargement-maribor-declaration-on-small-firms-published- artr191079-5.html. Hashemi S., and A. De Montesquiou. 2011. “Reaching the Poorest: Lessons from the Graduation Model.� Focus Note 69, Consultative Group to Assist the Poor, Washington, DC. Hungary Development Agency. 2011. http://ujszechenyiterv.gov.hu/. Hungarian Supervisory Authority. 2010. “The HSFA’s Third Risk Report in 2010.� Hungarian Financial Supervisory Authority, Budapest, available at http://www.pszaf.hu/data/cms2288632/HFSA_riskreport_Nov2010.pdf . 79 IMF (International Monetary Fund). 2010. “Financial Access Survey.� International Monetary Fund, Washington, DC, http://fas.imf.org/ . ———. 2010. “Financial Access Survey, Explanatory Notes.� International Monetary Fund, Washington, DC, http://fas.imf.org/misc/Explanatory_Notes.pdf. Ivanov, A., and S. Tursaliev. 2006. “Microlending to the Roma in Central and Southeastern Europe: Mixed Results, New Approaches.� Comparative Economic Studies 48: 36–49. Jack, W., and T. Suri (2010). “The Economics of M-PESA� http://www.mit.edu/~tavneet/M-PESA.pdf. Karlan, D. 2011. “Why Savings?� Karlan, D., and J. Morduch. 2009. “Access to Finance.� In Handbook of Development Economics, vol. 5, ed. D. Rodrik and M. Rosenzweig. Maryland Heights, MO: Elsevier B. V. Kwaak, T. 2010. “SME Performance Review 2009: Objective and Methodology of the SBA Fact Sheets.� Panteia/EIM, Zoetermeer, Netherlands. Kraemer-Eis, H., and A. Conforti. 2009. “Microfinance in Europe: A Market Overview.� Working Paper 2009/001, European Investment Fund, Luxembourg. McGuire, P., and J. Conroy. 2000. “The Microfinance Phenomenon.� Foundation for Development Cooperation, Brisbane. MFB Hungarian Development Bank. 2009. “Annual Report 2009.� MFB Hungarian Development Bank. Budapest. MVA Network Hungary. n.d. “Mission of the Hungarian Foundation for Enterprise Promotion.� MVA Network Hungary. Budapest. OECD (Organisation for Economic Co-operation and Development). 2009: “The Impact of the Global Crisis on SME and Entrepreneurship Financing and Policy Responses.� OECD Center for Enterpreneurship, SMEs and Local Development, OECD, Paris. Peachey, S., and A. Roe. 2004. “Access to Finance: A Study for the World Savings Banks Institute.� Oxford Policy Management. Poverty Action: http://www.poverty-action.org/microsavings/mpii Procredit Bank. 2010. “Annual Report 2010.� Procredit Bank, Sofia. Pytkowska, J., A. Houyet, and A. Mach. 2008. “Inclusive Finance – the State of Microfinance Sector in ECA.� Microfinance Centre for Central and Eastern Europe and New Independent States, Warsaw. Rowe, W., M. Gavrilova, P. Velev, and T. Shaw. 2009. “Cutting the Economic Costs of Prejudice: Microfinance Services in Roma Communities in Bulgaria.� Enterprise Development and Microfinance 20 (3): 235–46. Sinclair, S. 2001. “Financial Exclusion: an Introductory Survey.� Center for Research into Socially Inclusive Services, Heriot-Watt University, Edinburgh. 80 Soros Economic Development Fund. 2006. “Creating Social and Economic Change. Annual Report 2006.� Soros Economic Development Fund, New York. Szefku, T. 2007. “The Microcrediting Sector in Hungary.� Fejer Enterprise Agency (FEA), Budapest. United Nations Development Programme (UNDP). 2002. “The Roma in Central and Eastern Europe: Avoiding the Dependency Trap.� Regional Human Development Report, UNDP, Bratislava. Vingelman, J. 2010. “Views from a National Holding Fund on the JEREMIE Implementation – the Case of Hungary: Challenges, Achievements, Prospectives.� European Investment Bank, Luxembourg. World Bank. 2004. Roma in an Expanding Europe: Breaking the Poverty Cycle. Ringold, D., M. Orenstein, and E. Wilkens. World Bank, Washington, DC. ———. 2006a. “Banking Services for Everyone? Barriers to Bank Access and Use around the World.� Beck, T, A. Demirguc-Kunt, and M. S. M. Peria, Policy Research Working Paper 4079, World Bank, Washington, DC. ———. 2006b. “Expanding Bank Outreach through Retail Partnerships Correspondent Banking in Brazil.� Kumar, A., A. Nair, A. Parsons, and E. Urdapilleta. World Bank Working Paper 85, World Bank, Washington, DC. ———. 2008. “Finance for All? Policies and Pitfalls in Expanding Access.� Demirgüç-Kunt, A., T. Beck, and P. Honohan. World Bank, Washington, DC. ———. 2010 “Roma Inclusion: An Economic Opportunity for Bulgaria, Czech Republic, Romania and Serbia. Policy Note.� Laat, J. de, C. Bodewig, N. Arnhold, T. Linden, D. Dulitzky, S. Kosko, and F. Torracchi. Washington, DC: World Bank, Washington, DC. ———. 2010. World Bank Development Indicators. http://data.worldbank.org/indicator. ———. 2011. Doing Business Indicators (database). World Bank, Washington, DC, http://doingbusiness.org/. ———. 2012a. “Financial Inclusion Strategies Reference Framework.� D. Pearce and C. Ruiz Ortega World Bank, Washington, DC. ———. 2012b. “General Guidelines for the Development of Government Payment Programs.� World Bank, Washington, DC. 81