WPS3834 Access to Financial Services in Colombia The "Unbanked" in Bogotá Tova Maria Solo and Astrid Manroth* World Bank Policy Research Working Paper 3834, February 2006 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. . *Tova Maria Solo is a Senior Urban Specialist in the Urban and Water Group of the Latin America and the Caribbean Region and. Astrid Manroth was a Young Professional in the Finance Group of the Latin America and the Caribbean region of the World Bank at the time this report was written. Questions and comments on this report should be addressed to Tsolo@worldbank.org ACKNOWLEDGEMENTS The authors of this report are Tova Maria Solo and Astrid Manroth. However the work is largely based on background work carried out by a team of consultants in Colombia: under the direction of Roberto Steiner and Alberto Carrasquilla from CEDE, Carlos Medina managed the principal survey and Beatriz Ariza the survey analysis, with support from the Ministry of Finance, in particular the Technical ViceMinistry and Director of Banking, Mauricio Rocilla and the Superintendency of Banks, under the direction of Patricia Correa. The authors also benefited from the inputs of World Bank advisors Marianne Fay, Anjali Kumar,, and Fernando Montes-Negret, and fellow collaborators and reviewers: Rodrigo Chavez, Mariluz Cortes, Michael J. Goldberg, Juan Carlos Mendoza, Marialisa Motta, John Daniel Pollner, Susana M. Sanchez, John Henry Stein, Augusto de la Torre and Anna Wellenstein. The authors acknowledge, in addition, the invaluable comments and support from Alberto Aspe (Harris Bank), Yolanda Brown (Pacific Bank), James Carr (Fannie Mae Foundation), John Caskey, (Swarthmore College), Amanda Fernandez (Annie Casey Foundation), Michele Kahane (Ford Foundation), Luis Nino (Banco Azteca) Michael Stegman (Center for Community Capitalism at the University of North Carolina), Carlos Ximenes (Unibanco-Brazil). The authors are grateful to Patricia Cardenas, Juan Carlos Pinzon and Carolina Baron of the Association of Colombian Banks, Asobancaria, for arranging invaluable meetings with banking sector officials and for providing their own insights. We would also like to thank Federico Ochoa, Roberto Vergara Ortiz of BanColombia, Patric Tissot and Eulalia Arboleda of Caja Social and others for their insightful and invaluable comments from the perspective of a banker. Editorial support was provided by Cristina Velazco-Weiss ABBREVIATIONS AND ACRONYMS ACIM Association of Media Investigation ATM Automated Teller Machine CADES Special District Attention Centers (Centros de Atención del Distrito Especial) CAV Savings and Loans Associations (Cajas de Ahorro y Vivienda) CCO Check Cashing Outlet CD Certificate of Deposit CDAT Saving Term Deposit (Certificado de Depósito de Ahorro a Término) CDT Term Deposit (Certificado de Depósito a Término) CEDE Center for Economic Development Studies (Centro de Estudios sobre Desarrollo Económico) COP Colombian Pesos CRA Community Reinvestment Act DANE National Administrative Statistics Department (Departamento Administrativo Nacional de Estadística) ECH Household Survey (Encuesta Continua de Hogares) EGM General Media Survey (Encuesta General de Medios) FDIC Federal Deposit Insurance Corporation FFIEC Federal Financial Institutions Examination Council FRB Federal Reserve Board GDP Gross Domestic Product IOEs Instituciones Oficiales Especiales IVA Value Added Tax (Impuesto sobre el Valor Añadido) LAC Latin America and the Caribbean NCUA National Credit Union Administration NPL Non-Performing Loans OCC Office of the Controller of the Currency OECD Organization For Economic Cooperation And Development OTS Office of Thrift Supervision US United States UVR Real Value Unit (Unidad de Valor Real) 3 INDEX CURRENCY AND EQUIVALENTS ........................... Error! Bookmark not defined. ACKNOWLEDGEMENTS.............................................................................................II ABBREVIATIONS AND ACRONYMS........................................................................ 3 EXECUTIVE SUMMARY............................................................................................ 10 CHAPTER 1: FINANCIAL ACCESS IN COLOMBIA................................................ 17 1.1 Introduction............................................................................................................. 17 1.2 Overview of the Financial Sector in Colombia....................................................... 19 1.3 Supply of Main Financial Services......................................................................... 22 1.4 Basic Access Indicators .......................................................................................... 27 CHAPTER 2: CONSTRAINTS ON THE SUPPLY OF FINANCIAL SERVICES..................................................................................................................... 30 2.1 Regulatory Considerations...................................................................................... 30 2.2 Cost Barriers ........................................................................................................... 35 2.3 Physical Barriers..................................................................................................... 41 2.4 Cultural Barriers...................................................................................................... 43 CHAPTER 3: USERS OF FINANCIAL SERVICES AND PROFILE OF THE UNBANKED................................................................................................................. 45 3.1 Profile of the Unbanked.......................................................................................... 47 3.2 Payment Services for the Unbanked....................................................................... 50 3.3 Savings Services for the Unbanked ........................................................................ 52 3.4 Credit Services for the Unbanked........................................................................... 54 3.5 The Costs of Being Unbanked................................................................................ 59 3.6 Reasons for Being Unbanked.................................................................................. 59 3.7 Are the Unbanked Bankable? ................................................................................. 60 CHAPTER 4: CONCLUSIONS AND RECOMMENDATIONS................................. 62 4.1 Regulatory Measures .............................................................................................. 63 4.2 Supply Side Measures............................................................................................. 64 4.3 Demand Side Measures........................................................................................... 65 4.4 Areas for Further Work........................................................................................... 66 BIBLIOGRAPHY.......................................................................................................... 68 4 FIGURES Figure 1.1 Financial depth and banking sector profitability 1998-2003....................... 19 Figure 1.2 Key financial system indicators................................................................... 20 Figure 1.3 Public sector bonds and loans as percentage of financial sector assets....... 21 Figure 1. 4 Breakdown of deposits and loans by type of product 1998-2003 .............. 22 Figure 1. 5 Deposits and loans to GDP, 1998-2003 ..................................................... 22 Figure 1. 6 Market shares of current and savings accounts, CDTs and CDATs .......... 23 Figure 1. 7 Average amounts of current and savings accounts, per type of institution ....................................................................................................................... 24 Figure 1. 8 Breakdown of loans in the Colombian banking system............................. 27 Figure 1. 9 Basic access indicators ............................................................................... 29 Figure 2. 1 Bank requirements for opening accounts ................................................... 31 Figure 2. 2 Number of cashed checks per person in Colombia .................................... 33 Figure 2. 3 Interest rates on checking, savings and CD services.................................. 37 Figure 2. 4 Distribution of bank branches by income strata......................................... 42 Figure 2. 5 Distribution of bank branches by income strata relative to population distribution................................................................................................... 42 Figure 3. 1 Use of Financial Services In Bogotá, Mexico City, urban Brazil and OECD countries............................................................................................................. 46 Figure 3. 2 Use of Banking Services as a Percentage of Population............................ 46 Figure 3. 3 Use of banking services in relation to income............................................ 47 Figure 3. 4 Employment profile of the unbanked......................................................... 48 Figure 3. 5 Education levels of the banked and unbanked ........................................... 48 Figure 3. 6 Home-Ownership among the banked and unbanked.................................. 49 Figure 3. 7 Payment of salaries to the banked and unbanked....................................... 50 Figure 3. 8 Location of cashing pay checks.................................................................. 51 Figure 3. 9 Banked versus unbanked users of informal savings services.................... 54 Figure 3. 10 Requests for informal lending services .................................................... 55 Figure 3. 11 Source of informal lending....................................................................... 55 Figure 3. 12 Loan request from formal institutions...................................................... 56 Figure 3. 13 Loan approval rates of banked and unbanked individuals ....................... 57 Figure 3. 14 Reasons for not requesting a loan from the formal sector........................ 58 Figure 3. 15 Reasons for not requesting a loan from the informal sector..................... 59 Figure 3. 16 Main reason for not having a bank account.............................................. 60 TABLES Table 1. 1 Average number of accounts per type of institution and deposit level........ 25 Table 1. 2 Average size of deposit amounts in active savings accounts....................... 25 Table 2. 1 Information requirements for opening a bank account in Colombia........... 31 Table 2. 2 Costs of savings services by income quintiles............................................. 36 Table 2. 3 Costs of checking account services by income quintiles............................. 38 Table 2. 4 Costs of domestic credit card services by income quintiles ........................ 38 Table 2. 5 Products and costs of service by selected financial institution.................... 40 5 Table 3. 1 Channels of receiving remittances............................................................... 52 Table 3. 2 Banked versus unbanked users of informal savings services ...................... 53 Table 3. 3 Banked versus unbanked users of informal credit services......................... 54 Table 3. 4 Reasons for requesting a loan...................................................................... 56 Table 3. 5 Loan approval rates by different credit providers........................................ 57 BOXES Box 1: The example of Banco Caja Social................................................................... 32 Box 2: Examples of Regulatory Measures to Enhance Access .................................... 34 Box 3: Micro Housing Finance in Colombia................................................................ 35 Box 4: BanColombia's Microfinance Program ............................................................ 39 6 FOREWORD The theme of bancarizacion is of vital importance for Colombia. Why do I believe this? Several years ago I had the opportunity to read declarations made by citizens of Bogotá in a collection of writings belonging to the Mayor's office on the experiences of selected citizens in "building community". The Mayor's office collected several such stories and one in particular which touched me greatly described a resident who had come to Bogotá with his family, fleeing from the violence of the 1950's and began the process of establishing relations and building assets in his new home. The story is dramatic and demonstrates how one person, through his own efforts to build a home by "self-help" ended up by paying twice the actual cost of his house, in terms of present value. Furthermore, he paid for his house in a much shorter period of time than the formal financial system would have allowed. That is he paid for the house in full in six years, starting from the moment he began to build, to the moment in which he finished construction. He paid in half the time a formal sector credit would have taken. Finally, throughout the construction work he did take out a series of short term loans and borrowed from suppliers to pay for materials ­ always at interest rates which were easily twice what qualifies as usury levels. These are characteristics of a life outside the formal financial system which say a good deal about the welfare benefits which could result from bringing the "unbanked" into a formal financial system. If the payment periods for such loans could be prolonged, if the economies of scale for home construction could be applied, and if the unbanked were able to eran a market interest rate on their savings, while paying market interest rates on the loans they take, the welfare gains would be enormous. If we consider that for Bogotá alone, approximately 50% of the inhabited areas is the result of self-help construction processes, squatter settlements etc., we can easily conclude that, for lack of debating the issue 40 or 50 years ago, the costs in resources and welfare have been onerous, and practically tossed away. For this reason, I propose urgent consideration of the theme of bancarizacion, discussion and conclusions. For me we are completing a first step of which I hope there are more. It seems that we must focus on a discussion of the Colombian case, in the light of the 7 vast experience from other countries which have been presented and discussed here. I believe that there are at least five themes which much be clarified in the next step. My invitation is to continue to our discussions, taking this as the first step, but moving, obviously, toward a convergence on concrete actions. The first is the need for indicators of potential demand. Where is the potential demand and what are the welfare gains it implies? What are the costs? What should we spend? The first fundamental study we must undertake is to look carefully at demand for bancarizacion and to establish indicators. On the supply side, I feel that we need an inventory of products on the market, indicating which can meet the potential demand, and also an inventory of the product problems, which prevent the products available from meeting the needs of the demand side. In third place we must do an inventory of existing regulation. It is the responsibility of the government to maintain a regulatory framework adjusted to reality. The fourth theme, which comes up repeatedly, and which appears very important is the adequacy of the technological platform. It is likely that the marginal cost of increasing technological services low although the initial costs could be high, so that the initial investment merits a discussion, particularly since we are now at the point of entering into new supervisory systems. The next theme is, simply stated ­ geography and the physical relation between banks and their clients, if we hope to realize our ambitions of building a broader client base and use of our financial systems. Finally there are the multiple themes of culture, on the part of the client, or potential client, as on the part of the financial entity, in the sense of education. Again, if we are going to achieve our goals, we must educate ourselves, side by side, and learn to trust one another; institutions and clients; banks and unbanked. Linking these themes is the need for follow-up, follow-through and monitoring. These themes should have measurable, concrete results and should lead to proposals for specific policy measures for discussion in the respective agencies. Along with monitoring and follow-up I must mention the need to dialogue and to share information here in Colombia. On that subject I digress to mention that when one asks why interest rates vary so widely from, for example a relatively large commercial loan, to a relatively small loan to a small business, or between medium and small firms, the answer 8 immediately refers to the risks which are not necessarily so different, but which are perceived as very different. And this happens as a result of imperfect and inadequate information. For this reason, credit histories are the catalyst for easing the loan process and should not be seen as agents of personal intrigue to destroy personal honor and reputation. Rather, they are the financial sector's basic tools for reducing perceived risks. Without information it is impossible to make decisions while with information the lending process can save time, bureaucracy and costs. Information, thus, is the final key in the process of bancarización. At this moment the theme of information is being considered in Congress for the third time and it seems to me that as in Congress as in the public at large we must make its importance understood ­ and bring attention to its relevance to the theme of bancarizacion. I am very pleased that the seminar on bancarizacion has taken place, very pleased with the conclusions, and, above all, with the conviction that this is a theme whose analysis is beginning today, but that on all sides, will require continued discussion and, even more, concrete bases, concrete information, and technical work to help us to understand the best path to follow, so that the next generation will not have to suffer the same losses and costs that the lack of bancarizacion has represented for this generation. Dr. Alberto Carrasquilla Ministro de Hacienda y Crédito Público 9 INTRODUCTION "Access to financial services" needs a broader definition. The theme of "access to financial services" is one of the most easily confused in the financial world. For many people access to banking services is limited exclusively to the possibility of any economic interest or person obtaining a loan under conditions which reflect their level of risk. In fact, "access to financial services" is a far broader concept, whose definition in English is the access to all financial services, such as holding a savings or checking account, a debit or credit card, or being able to pay for public services in a branch bank or at an ATM machine, or the possibility of having a range of investment and savings options for short and long term. From this point onwards, we will use this expression "access to financial services" instead of the Spanish term "access to financial services". A broad-based banking sector contributes to economic development In the Colombian Bank Association we firmly believe that access to financial services is a determinant in a country's development. In other words, the access to financial services means progress. For this reason we have supported studies and conferences on the subject and we welcome the publication of this paper. A range of academic studies have demonstrated the close relation between access to financial services and economic activity. As Ross Levine mentions in one of his articles1 "No shortage of evidence exists to suggest that the level of financial development is a good indicator of economic growth, capital accumulation and of the technological change of the future economy." There is no shortage of facts to support this argument. For example, an effective financial sector reduces the cost of information sharing between economic agents. Without credit institutions, every economic agent would have to approach potential lenders one by one to find just the one who is ready to let him borrow the amount he needs according to the terms he can afford. Such an exercise would be virtually impossible to carry out nowadays, without taking into account the exaggerated costs it would represent. Financial intermediation, therefore, consists in gathering, and processing information which is disseminated in the market and putting it in the hands of actors interested in finding resources2. Thus, the financial sector serves to reduce the risk of liquidity and to promote optimal decision-making in the economy. In general, savers are risk-averse and 1Levine, Ross "Financial Development and Economic Growth: Views and Agenda" Journal of Economic Literature Volume XXXV June 1997 2Campos, Alexander. "Naturaleza y funciones del sistema financiero" Asobancaria, 2000 10 prefer short-term investments, while important projects require support in the form of resources over the long-term. In this case, the financial sector serves to obtain those resources needed to cover the investor while satisfying the requirements of the saver. In the same direction, the financial sector facilitates the recycling of savings into the economy, especially towards profitable projects with a pay-back for society, as with new technologies, which typically require major investments. Access to financial services can support sector growth as well At the same time, the access to financial services benefits the general population and tangibly improves community welfare. It strengthens the "savings culture", facilitates asset-accumulation, increasing capacity to deal with possible economic shocks and periods of financial difficulty. In addition, financial services can bring about important improvements in day to day life-styles. Consider, for example, how payments of public services (water, electricity, telephone services) can be made easier by branch payments, checks or internet (rather than personal cash payments in the company office downtown ­tr.). Also consider the possibilities for making transactions safer, through PIN numbers or personal identification, or even "biometry" (identification by personal physical characteristics). In the same way, access to banking services reduces transaction costs and increases efficiency for the economy at large. As stated by World Bank on a paper about access to financial services in Mexico, for the `unbanked" "day to day transactions ­ making and receiving payments- cost more, and saving becomes much more difficult, particularly in inflationary circumstances." The "unbanked" face higher costs associated with unregulated lenders and borrowers, those known better as `predatory lenders' or "loan sharks" whose interest rates and service charges far surpass those set by law. Access to financial services has advanced in Colombia ... but it still has a long way to go Recently, access to financial services has become a topic of study in many countries, both developed and developing. For example, the United States passed the Community Reinvestment Act, in the 1970's, with the objective of increasing the number of low-income and minority groups who benefit from the financial system, by offering incentives to financial institutions. In Germany, the law establishes a universal financial coverage program, through the municipal savings banks. 11 We have also seen special policies and practices developed by the market, in its efforts to broaden the business spectrum to include different population groups. South Africa's Standard Bank offers a good example with its "E-Plan", which puts automatic teller machines at the service of clients who can't read or write. Standard Bank's policy has succeeded to the point of increasing its client base by 600.000 in its first 10 months of operation, and by 1.400.000 one year later. In Colombia, sector efforts of the past few years are now producing results. The number of savings accounts has increased by 18% between 1999 and 2003, and the number of debit and credit cards grew by 21% and 24% respectively during the same period; and internet operations skyrocketed from 10,000 to 11,514,000 between 1998 and 2003. At this moment, the Colombian banking sector has 3,789 branches all over the country, plus 5,583 automatic teller machines and 42,224 payment points, which last year registered transactions for $10 billion Colombian pesos, an increase of 23% over the previous year. Equal efforts have been made in technological development. Compared to the 1970's, when a large percent of banking transactions were personal and manual and when access depended on the physical presence of clients in branches, the 1990's boasted new techniques which have increased both the range of services provided and their diffusion throughout the nation, to areas without a physical bank presence. Despite these advances, the level of access to financial services continues to be very low in our country. According to one of the few studies on the subject,3 in Bogotá only 40% of the population has any kind of access to financial services. If a similar study were applied to the rest of the country this percentage would fall considerably. The high percentage of "unbanked" or the low rate of access to financial services has several explanations. Among others: the lack of financial education in Colombia, the high costs for the banking sector of reaching low-income population; the loss of confidence on the financial system which stemmed from the juridical instability following the crises at the end of the 1990's; and finally, the difficult conditions of security and civil unrest which affect some of our cities. All these have held back the possibilities for increasing coverage in certain regions of the country. 3Ariza, Beatriz y Medina, Carlos A. "The Demand of Services from the formal financial sector in Bogotá", Julio 2003. 12 The financial sector is ready to meet the challenges We all face great challenges to improve access to financial services: the financial institutions; the government; the regulators and the users of the financial system. For the financial sector the challenges are: first recovering public trust as financial intermediaries; second, continuing to improve the quality and efficiency of day-to-day service delivery. Recent surveys suggest that there has been tangible progress in this area. Sector studies report that the cost of managing $Col 100 have dropped from $6.2 five years ago to $5.8 today. Third, maintaining the process of technological innovation in service delivery and meeting the consumers' needs. Finally, building a stable institutional framework, since it is clear that a healthy business environment is essential to financial sector depth. And here the government must play a fundamental role. Factors such as the guaranteed creditor rights, a stable regulation framework, interest rates based on market realities, an efficient judicial system, market transparency based on public disclosure of financial information and clear accounting regulations are crucial4 for this development. How does Colombia fare in each of these points? a. According to an IDB study of indicators of credit risk, Colombia has one of the worst ratings of protection to creditors, along with Peru and Mexico5; this necessarily means higher transaction costs and generates inefficiencies in financial intermediation. b. Regarding a stable regulatory framework, the most impeding issue is the current tax on financial transactions. Initially it was meant as a "short term measure", with a tax of 0.2% on financial transactions; today, we find ourselves with a permanent tax of 0.4%. We have repeatedly remarked the negative effects of this tax on the overall economy: it has greatly reduced the average returns on savings, making them less attractive; it has increased preference for cash transactions and has generated an increase in the "unbanked" population. The tax on financial transactions has effectively set back the Colombian economy by fifty years; preference for cash has grown and intermediation has shrunk. 4De Nicoló, Gianni, Geadah, Sami y Rozhkov, Dmitriy. "Cómo cerrar la "gran brecha" 5BID (2001), "Mercado financiero para el crecimiento empresarial" en Competitividad, El motor de desarrollo del crecimiento, Washington DC 13 c. Colombia is characterized by strict regulations on interest rates. In contrast, China, a country which has set and met impressive challenges in the field of economic development and growth, offers an interesting case for analysis. At present the Chinese government is considering a full liberalization of interest rates; this with the sole objective of improving the access of its great rural population, and of small and medium businesses, a measure which guarantees the creation of nearly 10 million jobs per year until 2010. d. The protection of financial information is of paramount importance, given that it creates a basic input for granting loans. Both the financial sector and the government must defend the use of such information, assuring its protection and correct use. Addressing these issues would assure a better development of our financial sector. However, what is important here is that efforts do not rest on financial institutions alone. Commitments must come forth from the regulatory agencies, the government, the Central Bank, the congress and international agencies, like the World Bank, which can lend support through its vast knowledge and international experience. We hope to take together the first step in building a medium-term plan for deepening the financial sector in our country. In our opinion, the best way would be through institutional development which allows private and public initiatives to work together to broaden financial markets. There is no place for arbitrary regulation which could lead to high risks for the sector. Hopefully the lessons acquired through academic discussions and studies, will help us to design a plan for the future. Not only because access to financial services is a natural outcome financial deepening, but also because it is indeed, a condition for development and economic growth. 14 EXECUTIVE SUMMARY 1. The link between financial services and development is by now broadly acknowledged in economic literature. Over the past years, the World Bank has explored the theme of access to financial services in various countries in Latin America, including Colombia. The present report provides an initial overview of the state of financial access in Colombia, drawing on the limited available information to date. 2. Access to financial services has been declining in Colombia, as illustrated by decreasing levels of financial depth and intermediation and decreasing availability of basic financial services in relation to population size. In addition, a household survey in Bogotá finds that 61% of the adult population lacks access to any form of formal financial services, such as checking or savings accounts, payment services or loans. 3. An overall decline in financial intermediation, especially following the banking and fiscal crisis in the second half of the 1990s offers a partial explanation for the low level of financial access in Colombia. The level of intermediation was at 85% in 2003, down from 116% in 1998. the number of current and savings accounts per inhabitant has also been decreasing in the last four years. In light of lower risk investment opportunities in form of government bonds, Colombia's main financial intermediaries, private banks, have limited incentives to increase their level of intermediation and provide services to low income clients. With the recovery of the financial system, the challenge today is to increase intermediation overall, and especially, intermediation to the poor. 4. A brief supply side analysis shows that Colombia's banks provide costly services mainly catered towards high income clients. Existing fees and costs of checking, savings and loan services average 5-10% of a monthly minimum wage, making them hard to afford for low income clients. In addition, high minimum balance requirements create strong barriers to access even for basic financial services such as savings accounts. The situation is exacerbated by multiple documentation requirements resulting from strict anti-money laundering regulation which reduces the eligibility of low income clients for formal financial services. 5. Financial exclusion is associated with lower and more uncertain incomes, lower education and closer links to the informal sector. A household survey conducted in Bogotá for this study reveals that 70% of the unbanked earn less than one minimum wage per month, are three times more likely to be unemployed than the banked, and have lower education levels. The unbanked save and borrow largely in the informal sector. At the same time, however, high home ownership rates show that the unbanked have the capacity to build assets, demonstrating that they have `bankable' characteristics. 15 6. Location plays a role in financial exclusion. Banks are over represented in terms of bank branch presence in higher income areas, while the majority of the unbanked lives more than 10 blocks away from a bank branch. 7. The high share of financial exclusion creates significant welfare losses for the economy as whole, as informal savings cannot be transformed into formal savings and investments. In light of Colombia's declining savings rate, improved access to financial services could therefore be an important factor contributing towards the economic development of the country. 8. Policies to improve the penetration of financial services among Colombia's low income population include regulatory incentives, supply side measures and demand side initiatives. In terms of regulation, many countries adopt an obligation for banks to open basic accounts for low income clients with a limited number of low-cost services, which could reduce cost barriers. Account opening requirements for low income people could be reduced to one simple form of identification. In addition, low financial intermediation in general would benefit from an exemption of the existing 0.4% tax on financial transactions. 9. Supply side measures could include `smart subsidies' to bridge the higher transaction costs associated with the low income segment. This requires a more detailed study of banks' cost structure. In addition, banks should leverage low-cost technology such as ATMs, and focus on the development of simple product solutions for low income clients. 10. Demand side measures need to focus on identified issues of income, education and culture. As lack of resources is a key factor in financial exclusion, growth and income generating policies are highly relevant for financial access. In addition, financial education is necessary to raise the level of awareness and understanding of formal financial services among the unbanked. Lastly, the Government should consider innovative measures to increase demand for low-cost electronic banking services, such as electronic transfers of wages and pensions. 11. Follow-up work should focus on a better understanding of banks' cost structure, on the potential role of credit cooperatives and other non-bank financial intermediaries in enhancing access, on a better understanding of demand side factors, and on a quantification of the benefits of financial access in order to secure political support. Finally, building better financial infrastructure could also be investigated. For example, the strengthening of credit information systems, better registries for collateral for secured credit, strengthened channels for deposits and remittances, are all areas for exploration. 16 CHAPTER 1: FINANCIAL ACCESS IN COLOMBIA 1.1 INTRODUCTION 1. The present report is part of a series of recent World Bank studies which explore the theme of access to financial services for low-income populations. To date, regional efforts include work on financial access in urban Brazil (Brazil: Access to Financial Services, 2003), in Mexico (Broadening Access to Financial Services Among the Urban Population: Mexico City's Unbanked, 2003) and in Nicaragua (The State of Microfinance in Nicaragua, 2002). World Bank studies and operations on financial access have also been conducted in India (Access to Finance in India, 2004) and Romania (Romania Rural Finance Program). 2. The World Bank has worked with the Colombian government over the past years to address the challenge of low penetration of banking services for the poor. To this extent, a comprehensive study has been conducted on access to finance in rural Colombia (Colombia Rural Finance, Access Issues, Challenges and Opportunities, 2003). In addition, a conference jointly organized by Asobancaria, the Colombian Banking Assocation and the World Bank on `Bancarizacion ­ Una Herramienta para el Desarollo' in Cartagena de Indias in April 2004 raised awareness of the importance of access to financial services with Colombia's private banks, regulators and politicians. Finally, the World Bank's recently approved Second Programmatic Financial Sector Adjustment Loan to the Republic of Colombia includes a component on improvements in the provision of housing finance for the poor. 3. In this context, the present report focuses on access to financial services in Bogotá and urban Colombia, drawing on existing publications and internal reports, a demand side survey and an initial overview of supply side issues. This report also highlights a number of issues requiring further research. Due to the parallel efforts of the World Bank, rural finance and housing finance in Colombia are not included in the scope of this report. 4. The term unbanked has traditionally been applied to individuals without deposit accounts in a regulated financial institution. In this report, it refers more broadly to individuals without access to formal financial services. Financial services are understood as products and services provided by formal financial institutions, including savings and checking accounts, debit and credit cards, other payment services, consumer loans, microcredit and remittances. 5. The importance of access to financial services in the context of macroeconomic growth and financial sector stability is well documented in economic literature. Development theory has proved the existence of 17 a link between financial services and economic development6. By transforming informal savings into savings in the formal financial sector, access to financial services helps increase aggregate savings and investments, thus stimulating economic growth. 6. The costs of being unbanked are equally well documented7. Financial exclusion is an obstacle to micro-economic development, as it significantly increases transaction and financial costs leading to substantial welfare and efficiency losses8. Furthermore, lack of access to financial savings instruments leads to sub-optimal investment choices, as individuals are forced to resort to alternative forms of asset accumulation or income generation9. By encouraging asset accumulation, access to financial services provides households with a financial safety net10. In addition, access to banking services enhances the probability of credit access by, for example, establishing a credit record. 7. Access to financial services can be good business, aligning private sector interests with the social aspect of access to basic financial services as a public good. The financial self-sufficiency of many microfinance institutions or the experience of community banks in the US demonstrate that working with low income clients can be profitable. In Colombia, recent initiatives such as the launch of BanColombia's microfinance program reflects the growing interest of commercial banks in doing business with the poor (see Box 1). 8. This report explores the issue of financial access in Colombia in four steps. Chapter 1 provides an overview of Colombia's financial sector and some basic access indicators. Chapter 2 focuses on three areas of constraints on the supply of financial services in Colombia. Chapter 3 presents key results from a demand side survey and Chapter 4 concludes with initial recommendations on potential policy measures and areas for further work. 6Rostow, 1960, 1962. 7Dunham, 2001; Caskey, 1997; and Stegman 1999; Carr and Scheutz, 2001; Skricki and Fernández, 2001; Rhine et al., 2000; Green and Lichter 1996. 8A recent study comparing financial exclusion in the US and Mexico estimates the cost of being unbanked for a medium income household in the US at about 2% of annual income for check cashing services and higher for savings and borrowing services. (Caskey and Ruiz, 2004) 9The difficulties of saving without access to banking services are widely documented. (Stegman 1999, Carr and Scheutz 2001, Skricki and Fernández 2001, Rhine et.al 2000, Green and Lichter 1996). 10Stegman, 1999; Carr and Scheutz, 2001; Skricki and Fernández, 2001; Rhine et al., 2000; Green and Lichter, 1996. 18 1.2 OVERVIEW OF THE FINANCIAL SECTOR IN COLOMBIA 9. Colombia's financial sector11 underwent a significant re-shaping as a result of a banking crisis in 1998. Following a period of strong expansion in the first half of the 1990s, excessive loan growth combined with an economic slowdown in the second half of the 1990s caused a rapid deterioration in loan portfolio quality and coverage ratios, triggering systemic liquidity problems. In order to avoid a broader systemic crisis and a deposit run, several emergency measures were introduced between 1999 and 2001. As a result of liquidation and merger activity, the number of total financial institutions decreased from 148 to 64. In particular, most of the Cajas de Ahorro y Vivienda (CAVs) were absorbed by commercial banks, and the banking crisis effectively marginalized the financial cooperative sector in Colombia. 10. Following the crisis, banking sector performance has been recovering, mainly driven by a partial resumption in economic growth, improved earnings and portfolio quality, stable net interest margins and significant government recapitalizations. Financial system assets steadily increased from 57% of GDP in 1998 to 70% of GDP in 2003, while deposits grew even faster than total assets over the same period. For the sector as a whole, return on average equity rose from ­32.6% in 1999 to 16.4% in 2003, capital adequacy recovered from 10.9% in 1999 to 14.4% in 2003, and non-performing loans decreased from 16.4% in 1999 to 10% of total loans in 2003. Figure 1.1 Financial depth and banking sector profitability 1998-2003 20.0% 16.4% 80 Chile 9.6% 70 10.0% 0.1% 1.8% 1.1% 60 -2.5% -3.5% -2.3% 1.1% ) 0.0% (% PD 50 Colombia G 40 -10.0% totiderC Mexico 30 Peru -20.0% 20 -21.0% -20.7% -30.0% 10 Venezuela -32.6% 0 -40.0% 0 1 2 3 4 5 6 7 8 9 0 1 2 3 1998 1999 2000 2001 2002 2003 199 199 199 199 199 199 199 199 199 199 200 200 200 200 Colombia Chile Peru Mexico Venezuela Return on Average Equity Return on Average Assets Source: World Bank calculations based on SIMA and Superintendencia Bancaria data 11The term financial sector refers to all financial intermediaries supervised by the Superintendencia Bancaria, including credit institutions (private and public banks including mortgage banks), financial cooperatives, corporate finance companies and leasing companies. 19 Figure 1.2 Key financial system indicators 80 Assets/GDP 10 Loans/GDP, Deposits/GDP, Loans/Deposits 50 120 70 9 45 8 100 60 PD 40 % 7 % PD 35 50 6 80 sti /Gst % 30 h 40 5 25 60 sseal depos/ growt % 30 4 s/Gtisoped,PD 20 taoT oansl 3 40 15 s 20 2 s/Gnaol 10 20 osrG 10 1 PDG/tessA ssorG 5 0 - 0 - Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Gross Loans/GDP Total Deposits/GDP Total assets / GDP Asset/GDP growth Gross loans/deposits Intermediation by type of institution Share of total assets in Colombian banking system 180% 100 13 160% 18 16 15 14 14 0 0 0 1 1 1 140% 80 16 16 % 17 18 18 st 18 sio 120% 12 12 60 11 11 12 12 epd 100% s/naol % 80% 40 ssorG 60% 52 54 54 56 58 59 20 40% 20% 0 1998 1999 2000 2001 2002 2003 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Private local banks Public banks Foreign banks Financial coops Private Domestic Banks Public Banks Others Private Foreign Banks Fin. cooperatives Others (*) Private local banks include mortgage banks Gross Loans Total Deposits 100 8 1 14 14 13 12 11 1 100 1 8 1 0 1 11 10 10 9 0 0 80 18 17 1 1 0 17 16 26 18 17 15 80 15 18 18 17 12 8 8 9 7 7 1 60 12 11 15 11 11 11 % 60 % 11 40 60 59 60 62 63 65 40 60 60 62 64 65 20 47 20 0 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 0 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Private Domestic Banks Public Banks Private Foreign Banks Fin. coops Private Domestic Banks Public Banks Others Private Foreign Banks Fin. coops Others Source: World Bank calculations based on Superintendencia Bancaria data 11. At the same time, financial intermediation in Colombia has been steadily declining since 1998. Total financial sector loans as a percentage of total assets decreased from 65% in 1999 to 51% in 2003 and from 116% in 1998 to 85% in 2003 as a percentage of total deposits (see Figure 1.2). As a result, financial depth in Colombia stagnated around crisis levels with a ratio of credit to GDP of about 34%. This is significantly below the level of Chile, while at par with Mexico and above the levels of Peru and Venezuela (see Figure 1.1). The decline in intermediation by Colombian banks can be attributed to their risk- adverse behavior following the banking crisis and a resulting shift in 20 business activity and banks' revenue composition. 12. Declining intermediation is partly due to crowding out by the public sector. Between 1998-2003, the exposure of Colombian banks to the public sector through bonds and loans increased by 166%, compared with a mere 6% growth of loans to the private sector, and a total asset growth of 17%. As of December 2003, banks invested 20% of their assets in public bonds and provided an additional 5% of total assets in loans to the public sector (see Figure 1.3). Public bonds account for over 70% of banks' total securities portfolio. With a share of 30% of banks' operating income12 at the end of 2003, interest income from government securities is almost as important a revenue source as interest income from loans (40% of banks' operating income)13. Given attractive interest rates14 and the zero risk weighting of government bonds for solvency requirements, Colombian banks currently lack incentives to increase lending to the private sector. Figure 1.3 Public sector bonds and loans as percentage of financial sector assets Breakdown of total system assets, COP trillion System assets by type, %to total assets 120 100% 90% 100 31% 32% 30% 29% 28% 27% 80% 80 27 70% 25 s 5% 7% 60% 7% 15% 19% 20% 22% noliilrt 25 26 24 24 sseta 6% 60 21 50% 7% 6% 6% 4% 4 18 5 6 16 5 12 P 5 4 40% OC 40 5 5 altotfo % 30% 58% 54% 48% 46% 46% 47% 20 46 43 39 39 41 46 20% 10% - 0% 1998 1999 2000 2001 2002 2003 1998 1999 2000 2001 2002 2003 Loans to private sector Loans to public sector Loans to private sector Loans to public sector Public sector bonds Other securities & assets Public sector bonds Other securities & assets Source: World Bank calculations based on Superintendencia Bancaria data 13. Private banks are the most important financial intermediary in Colombia. Private domestic banks account for 59% of financial system assets and 65% of gross loans and deposits. Following private domestic banks, foreign banks' activity accounts for 16% of financial sector assets and a 17% and 15% of gross loans and deposits, respectively. Private domestic and foreign banks have an intermediation ratio of 85% and 91% respectively. 14. Public banks are also important financial actors. The share of public banks in the financial system has been relatively stable over the past 12Defined as net interest income plus other net non-interest income, before operating expenses 13Non-interest income from fees and commission, trading and leasing activities accounts for the remaining 30% of operating income. 14For example, the 10 year government bond due April 2012 is yielding 13.54% and the two year government bond due September 2005 is yielding 8.42% (Banco de la República). 21 years around 8-12% of total assets, deposits and loans (see Figure 1.2). While collecting 11% of deposits, public banks provide 8% of total loans of the financial sector, or 24% of total lending if public second-tier institutions are included. Public banks intermediation ratio is of 57%. 1.3 SUPPLY OF MAIN FINANCIAL SERVICES 15. The following provides a brief overview of the supply structure of basic financial services in Colombia based on the absolute development of financial services and the relative importance of the different intermediaries in providing these services. Figure 1. 4 Breakdown of deposits and loans by type of product 1998-2003 Total deposits by type of product 4 Gross loans by type of product 70 60 4 3 60 50 3 3 2 22 21 50 22 10 no 40 24 16 n 16 13 12 illirt 23 23 40 10 8 P 30 23 llioirt 10 7 7 8 CO 24 30 21 20 15 17 POC 14 16 20 31 33 10 28 28 28 28 11 12 13 10 6 7 9 0 0 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Current Account Savings Account CDT (*) Others Commercial loans Consumer loans Mortgage loans (*) CDTs: Certificados de Depósito a Término Source: World Bank calculations based on Superintendencia Bancaria data Figure 1. 5 Deposits and loans to GDP, 1998-2003 Loans to GDP, Deposits to GDP 50 39 4441 45 40 38 3735 33 3436 38 34 30 % 20 10 0 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Gross Loans/GDP Total Deposits/GDP Source: World Bank calculations based on Superintendencia Bancaria data 16. Deposits have continuously increased, despite the 1998 crisis. As Figures 1.4 and 1.5 show, the level of deposits, both in absolute numbers and as a percentage of GDP, has grown steadily in the last six years. This is a significant development, given the 1998-99 banking crisis, 22 which could have led to a loss of public confidence. However, given that no deposits were lost, trust in the banking system was not eroded. 17. Increased preference for high liquidity instruments. Looking into the evolution of deposits by type, we see a significant growth in current accounts versus savings instruments. Indeed, current accounts represented 21% of all deposits in 2003, and just 13% in 1998. On the other hand, term deposits plus savings accounts decreased to 72% of deposits in 2003, from 83% in 1998. Figure 1. 6 Market shares of current and savings accounts, CDTs15 and CDATs16 by type of institution Current Accounts Savings Accounts 100 3 100 14 14 15 15 22 17 23 21 19 19 25 80 80 33 11 11 10 11 11 15 19 17 18 17 60 21 60 10 % % 10 40 40 66 72 74 75 74 74 58 61 61 64 51 47 20 20 0 0 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Private Domestic Banks Public Banks Private Domestic Banks Public Banks Private Foreign Banks Financial cooperatives Foreign Banks Financial cooperatives Others Others Term Deposits (CDT) Saving Term Deposits (CDAT) 1 1 100 100 4 4 2 16 22 20 21 29 23 23 24 23 31 80 80 31 0 33 0 0 0 0 17 3 6 0 17 19 18 15 16 9 2 6 60 16 60 6 5 5 5 4 8 17 % 7 % 21 5 40 40 75 72 62 48 54 53 54 56 56 53 20 20 37 42 0 0 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Private Domestic Banks Public Banks Private Domestic Banks Public Banks Private Foreign Banks Financial cooperatives Private Foreign Banks Financial cooperatives Others Others Source: Superintendencia Bancaria 18. Domestic private banks collect the vast majority of deposits, accounting for between 47% to 74% of deposits held in current accounts, savings accounts, term deposits and saving term deposits (see Figure 1.6). Except for savings term deposits, foreign banks have an 15 Certificados de Depósito a Término. Term certificate of 30 days or above. Can be traded in the secondary market. 16Certificados de Depósito de Ahorro a Término. Term between 1 and 29 days. It cannot be traded in the secondary market. 23 additional 15-20% market share in each of these services. Public banks collect a stable 11% share in savings accounts and a declining 15% share of deposits in current accounts. They provide a significantly lower share of term deposits. The significant market shares of financial cooperatives in saving term deposits (21%) and other credit institutions (such as commercial finance companies and leasing companies) in term deposits (23%) is the result of their regulatory restrictions on the type of deposits they can accept. 19. Private domestic banks cater more towards small savers compared to public banks. Private domestic banks in Colombia collect almost 90% of savings deposits below COP 5 million and have the smallest estimated average savings amount (see Figure 1.7), as they hold by far the largest number of savings accounts in this category. Private domestic banks therefore appear to cater more towards small savers than public banks, which can likely be attributed to the specific business focus of some banks (such as Banco Agrario, Caja Social, Colmena or Megabanco) on low income clients. 20. Checking accounts by public banks may be more accessible for lower income clients. Public banks show a significant market share in current accounts with a deposit size below COP 5 million, with 27% (see Figure 1.7). Public banks also have by far the largest number of accounts in the small deposit segment below COP 5 million (see Table 1.1). This may indicate that the checking account products of public banks serve smaller savers and lower income clients better than those of private banks. Figure 1. 7 Average amounts of current and savings accounts, per type of institution Market share of current accounts by deposit level Market share of savings accounts by deposit level 100 100 3 3 5 7 13 tsnu 15 16 17 18 20 11 9 18 27 80 tsnu 10 10 co 80 10 accot 21 20 18 17 16 12 ac enrr 60 20 sgn 60 cufo e savifo ` ar 40 e 86 88 85 ar 40 82 77 sh 64 64 65 65 65 sh 69 ketra 53 ket M 20 ar 20 M % % 0 0 <5 >5 & <10 >10 & <20 >20 & <50 >50 & < >100 <5 >5 & <10 >10 & <20 >20 & <50 >50 & < >100 100 100 Million COP Million COP Private Domestic Banks Private Foreign Banks Public Banks Private Domestic Banks Private Foreign Banks Public Banks Source: Own calculations based on Superintendencia Bancaria data 24 Table 1. 1 Average number of accounts per type of institution and deposit level COP 1-5 COP 5-10 COP 10-20 COP 20-50 COP 50-100 Over COP100 million million million million million million CURRENT ACCOUNTS Private Domestic Banks 59,191 2,629 1,851 1,459 576 612 Private Foreign Banks 43,385 1,677 1,079 782 277 279 Public Banks 172,102 3,495 2,552 2,178 907 1,071 Total current accounts 65,967 2,423 1,684 1,323 518 556 SAVINGS ACCOUNTS Private Domestic Banks 1,029,685 15,647 8,365 4,305 948 711 Private Foreign Banks 247,421 2,963 1,878 1,022 239 234 Public Banks 172,102 3,495 2,552 2,178 907 1,071 Total savings accounts 698,198 10,453 5,750 3,066 724 601 Source: Own calculations based on Superintendencia Bancaria data 21. An analysis of average savings amounts shows that the average savings amount with credit institutions (banks) is about COP306,700 (US$11917) in the smallest range. This suggests that small savers have a savings capacity of about one minimum salary18. Higher average savings amounts for other financial intermediaries are explained by their regulatory restrictions on taking savings deposits. These numbers also show that average savings levels by lower-end clients do not significantly exceed minimum account opening or balance requirements by banks (see Chapter 2). Table 1. 2 Average size of deposit amounts in active savings accounts COP million Less than $5 $5 -$10 $10 -$20 $20 -$50 $50-$100 Over $100 Credit institutions 306,639 6,576,824 12,986,545 27,912,260 55,450,203 1,041,420,302 Financial 701,846 8,214,000 15,426,750 20,255,000 68,123,000 3,977,280,081 corporations Commercial 768,236 6,939,667 14,337,250 32,820,500 83,057,000 824,328,000 finance companies Financial 429,409 6,775,080 13,558,793 27,872,800 68,375,462 242,573,867 cooperatives Total active accts 306,791 6,577,066 12,987,120 27,912,376 55,461,244 1,042,339,394 Source: Own calculations based on Superintendencia Bancaria data Note: Inactive accounts are those which have not been used in the past six months 22. Following the banking crisis, the relative supply of loan services has changed. The total amount of gross loans in the period 1998-2003 has remained flat at around COP53 trillion. However, commercial loans increased by 18% over this period and now account for over 60% of available credit. In contrast, mortgage loans suffered a reduction of 38% between 1998-2003. The decline in mortgage lending is a direct result of the crisis, as asset problems in the mortgage sector and the closing of 17 Exchange rate of 1US$ = COP2,585 (as of October 27, 2004). 18 Minimum monthly salary in Colombia for year 2004 is COP358,000, or US$138. 25 the majority of the CAVs adversely affected supply. Consumer lending has shown a steady increase since 2000, with a particularly strong 25% growth between 2002 and 2003. 23. Private banks are the main providers of commercial and consumer lending, accounting for 77% of commercial lending and 86% of consumer lending as of December 2003 (see Figure 1.7). In line with their strategic business focus, foreign banks provide a higher share of commercial lending (27% of total private bank lending) than consumer lending (19% of total private bank lending) in Colombia. However, domestic banks are gaining market share at the expense of foreign banks, pointing to increased competition in the commercial loan segment. Public banks accounted for 9% of commercial lending in 2003, but are a small player in consumer loans with a constant 3% market share. With a 3% market share, financial cooperatives are as relevant as public banks in the provision of consumer loans, while they do not provide commercial loans. 24. Mortgage loans are almost exclusively provided by private domestic banks which account for over 90% of mortgage lending as they mostly absorbed the remaining mortgage banks. With the market share of public banks declining to 5%, foreign banks are aggressively expanding into this business segment, increasing their market share by 50% in 2002- 2003. 25. In contrast, public banks are the main provider of microcredit in Colombia. Public banks account for almost 50% of microcredit services, pointing to limited supply from the private sector. However, it should be noted that microcredit activities only started to be reported separately from commercial lending in 2002. Thus, there is a risk that private banks might not have adequately reclassified their portfolios yet. The market share of private domestic banks increased by 8% in 2003 and domestic banks now account for 42% of microcredit services in Colombia. Although cooperatives currently only provide a small share in relation to the total market (0.44% in 2003), they expanded their micro- loan portfolio by more than 60% in the last year, demonstrating the most rapid growth rate among any supplier. 26 Figure 1. 8 Breakdown of loans in the Colombian banking system Gross Commercial Loans Gross Consumer Loans 100 100 12 9 7 8 18 15 13 14 12 3 3 0 22 22 19 3 0 4 4 5 80 0 15 17 17 0 0 0 80 27 17 21 21 21 19 4 3 3 25 26 25 5 60 60 13 8 9 7 13 % 9 7 8 % 40 40 69 69 69 62 47 53 56 56 45 49 20 43 48 20 0 0 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Private Domestic Banks Public Banks Private Domestic Banks Public Banks Private Foreign Banks Financial Cooperatives Private Foreign Banks Financial Cooperatives Others Others Gross Mortgage Loans Microcredit 1 2 3 8 100 2 21 2 9 2 100 10 10 8 7 7 5 2 1 80 80 49 49 60 60 % 88 88 89 91 91 92 % 40 40 20 20 39 42 0 0 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Dec 02 Dec 03 Private Domestic Banks Public Banks Private Domestic Banks Public Banks Private Foreign Banks Financial Cooperatives Private Foreign Banks Financial Cooperatives Others Others Source: Superintendencia Bancaria 1.4 BASIC ACCESS INDICATORS 26. On the basis of simple access indicators, access to financial services in Colombia has been declining. Figure 1.9 shows that both the number of current and savings accounts per 100 inhabitants decreased over the past years. Equally, the number of cashed checks decreased from 3 checks per inhabitant in 1998, to 1.6 in 2003. This reflects, among other factors, the impact of financial sector taxation (discussed in Chapter 2), although it could also be a consequence of the increased use of electronic payments. Declining intermediation and available credit to the economy also demonstrate the negative trend in access to financial services in Colombia. 27. At the same time, growing use of electronic banking services shows increased acceptance of alternative banking technologies by the banked (see Figure 1.9). The penetration of debit and credit cards in Colombia has been increasing over the past few years. At the end of 2003, one in five inhabitants (or one out of two banked persons) had access to a debit card, and 74% of active savings accounts had an 27 associated debit card19. On the other hand, only 56 out of 1,000 inhabitants had access to a credit card. The use of debit cards with both checking and savings account explains the relatively higher penetration compared to credit cards. In addition, the number of ATM and Internet transactions has grown as well to an average of 9 ATM transactions and about 3 Internet transactions per inhabitant in 2003. Colombia's low internet penetration (4.6% in 2002) explains the relatively lower share of Internet transactions. In light of declining account penetration, increased penetration of electronic banking services points to growing acceptance of alternative banking technologies among the banked population. 28. Access to bank branches has decreased, while access to ATMs is improving. Using population per branch and ATM as a basic indicator for geographic access, trends over the past few years show that the average number of individuals served by one bank branch increased by almost 25% from almost 9,000 people in 1999, to almost 12,000 people in 2003. An increasing number of clients per bank branch points to a decreasing geographical bank branch network. In contrast, the number of people covered by one ATM has been slightly decreasing, indicating improving ATM coverage by Colombia's banks. While the trend away from physical branch outlets to cheaper technology solutions is in line with the experience in Western banking systems, it implies increased barriers to access for the unbanked which are generally less educated and less equipped to deal with electronic banking services (see Chapter 3). 19Marulanda 2004. 28 Figure 1. 9 Basic access indicators No. of accounts per 100 inhabitants No. of cards per 100 inhabitants 35 30.5 30.6 35 30.5 28.0 30 30 25.5 25 25 18.6 18.7 19.0 18.7 21.0 20 20 %15 %15 10 10 4.7 4.7 4.4 5.6 4.6 4.8 4.6 4.8 5.1 5 4.4 5 0 0 1999 2000 2001 2002 2003 1999 2000 2001 2002 2003 Current Account Saving Accounts Credit Card Debit Card No. of transactions per capita Population per branch, ATM and point of 10 sale 8.9 9 8.4 14,000 ati 7.9 8 7.7 7.6 12,000 10,863 10,982 11,295 11,767 cap 7 10,000 8,718 erp 6 7,663 7,973 8,071 7,986 sn 8,000 7,440 5 000)'( io 4 6,000 3.13.1 s.oN sact 3 2.6 4,000 anrt 2.1 2.2 2 1.8 1.3 2,000 925 882 836 812 811 oN 1 0.0 0.3 0.8 - 0 1999 2000 2001 2002 2003 1999 2000 2001 2002 2003 ATM Internet POS Branch ATM POS Source: Asobancaria 29. In conclusion, the structure of Colombia's financial system still reflects the consequences of the banking crisis of the late 1990s, as financial depth is relatively low compared to other countries in the region and financial intermediation remains below pre-crisis levels. Accordingly, access to basic financial services has been declining, as reflected in a declining number of current and savings accounts over the past years, and in a decreasing share of credit available to the economy. 30. As private banks provide the majority of financial services, the structure of the financial system offers an explanation for the low level of financial access in Colombia. In light of profitable alternative investment opportunities such as government bonds, the main actors in Colombia's financial system currently have limited incentives to provide services to low income clients, in particular access to credit services. This is exacerbated through the absence of regulation on financial access, as discussed in Chapter 2. 29 CHAPTER 2: CONSTRAINTS ON THE SUPPLY OF FINANCIAL SERVICES 1. This chapter examines supply side aspects of access to financial services in Colombia from the perspective of constraints on financial institutions in Colombia that prevent low income clients from gaining access to formal banking services. Three groups of supply side issues have been identified: (i) regulatory issues, (ii) cost constraints and (iii) physical difficulties. The following will look at each one of these in more detail. 2.1 REGULATORY CONSIDERATIONS 2. In Colombia, existing regulation has a mixed impact on the use of financial services. On the one hand, norms such as strict account opening requirements and taxation on financial services might discourage the use of formal financial services. On the other, regulation on income tax returns and sales tax refunds on credit and debit card transactions could encourage the use of formal banking services. 3. Existing regulation constraints entry to formal financial services for low income people. Anti-money laundering regulation introduced in 200320 created a number of complex account opening requirements. In order to open an account with a Colombian bank, individuals have to provide multiple documents, including proof of identity, employer information and monthly income and expense statements, such as proof of salary and rental payments (see Table 2.1). These information requirements apply to all clients regardless of their socio-economic characteristics. By implicitly assuming that clients are active in the formal economy, this regulation creates significant barriers to entry for low income clients. As the demand survey shows (see Chapter 3.1), a significant share of the unbanked is not active in the formal sector, and hence, cannot fulfill the formal eligibility criteria, such as proof of income or employment. Consequently, low income people are severely disadvantaged under existing regulation. In addition, multiple documentation requirements are needed to open a savings account in Colombia21. Banks require proof of employment, personal references or, in some cases, even proof of tax income for opening an account that entails no credit risk (see Figure 2.1). Finally, many banks also require a minimum deposit or a minimum salary. Banco Caja Social is one of the few financial institutions that do not require minimum opening amounts (see Box 2 and Table 2.5). 20Superintendencia Bancaria, Prevención y Control del Lavado de Activos, Circular Externa 025 de 2003, Annex 1. 21The only exception is savings accounts opened by an employer in the name of their employees for salary payments which do not require additional documentation. (Prevención y Control del Lavado de Activos, Circular Externa 025 de 2003). 30 Table 2. 1 Information requirements for opening a bank account in Colombia Basic information requirements: · Complete name and last names · Identification number: citizenship number, identity card, or resident card · Date and place of birth · Mailing address and home telephone number · Job or occupation · Description of the occupation: Employed or autonomous, and position held · Monthly income and expenses · Other income, or income originated in activities other than main one · Total assets and liabilities · In case of a trust, identification of the good and type of resource that is being handed in · Disclosure authorization for control and reporting to risk management centers · Signature and fingerprint of the potential client In case the potential client's activity involves transactions in foreign currency: · Type of foreign currency transaction that the person usually undertakes · Foreign currency current accounts that the person may have: account number, bank, city, country and currency · If the potential customer does not undertake foreign currency transactions, a declaration including this statement is required Documents that must be annexed to the application form: · Photocopy of the identification document (citizenship number, identity card, or resident card) · Proof of income (professional fees, tax retention certificate, or other document) · Tax filing declaration corresponding to last available fiscal year · Last available audited financial accounts, when applicable Source: Superintendencia Bancaria, Circular Externa 025 de 2003 4. Similarly, documentation requirements for credit applications require a high degree of formality. The minimum information that needs to be provided by the applicant includes proof of identity, financial statements and the last rent payments if applicable, plus bank references and a positive reply from the credit bureau22. These requirements are clearly geared towards people who already are active in the formal financial sector. Requirements such as bank references and credit bureau information are difficult to fulfill by the unbanked and those mainly active in the informal sector. Figure 2. 1 Bank requirements for opening accounts 25 20 knaBfore s 15 mb 10 Nu 5 0 off oorP tnemyolp laicr alr sreh Ot em mmeoC secnreefre la onrs Pe secnreefre rotnara fofooPr Gu atelloc emocnix Ta Checking Saving Source: Superintendencia Bancaria, June 2002 22Marulanda (2004) 31 Box 1: The example of Banco Caja Social Banco Caja Social has been a pioneer in providing savings and lending services to the urban low income segment in Colombia. Banco Caja Social serves over one million clients with saving products, and 200,000 clients with active loans totaling more than $414 million. More than 70 percent of these loans have a maximum credit limit of $2,070, and are mostly oriented toward the personal banking and entrepreneurial sectors, with an average monthly income of $274 per family. Key to Banco Caja Social's success has been its ongoing efforts in offering a portfolio of products and services that is specifically tailored for the low-income customer, paying particular attention to minimum amount requirements when opening accounts as well as product simplicity. Banco Caja Social opens personal current accounts and savings accounts with no minimum balance requirements, and with simpler requirements than most Colombian banks. An example of Banco Caja Social's product focus on the low income segment is its savings account "Reach your Home", ("Alcance su Casa") targeted in particular to individuals and families with income below 4x the minimum salary, who are saving in order to reach the minimum level that gives access to governmental housing subsidies. 5. Taxation of financial transactions increases the cost of access. To address the fiscal impact of the banking crisis, a general tax of 0.2% on current and savings account withdrawals was introduced. This tax was subsequently made permanent and increased to 0.3% on all current and savings account transactions and has recently been raised further to 0.4%23. The tax has to be paid by all users and clients of supervised financial institutions at the time of the transaction24. A few specific exemptions include savings dedicated to housing finance below a certain limit25, transfers between checking and savings accounts of the same owner in the same institutions, loan disbursements, and some transactions relative to payroll accounts. 6. As a result, the use of formal financial services has decreased over the past few years. This is illustrated by a decline in the use of check cashing services (Figure 2.1) and a simultaneous increase in the amount of cash in the economy26. In addition, focus groups confirm that the transaction tax discourages low income people from using formal financial services. Consequently, the poor prefer cash based transactions, they save under their mattress and disregard other credit and transaction services with formal financial institutions27. Combined with complex account opening requirements, the taxation of financial transactions creates strong disincentives for low income clients to access formal financial services. 23Law 863 of 2003. 24This includes all types of withdrawals including check, debit cards, ATM transactions etc. 25Decreto 449 of 2003. 26This may also make tracing money laundering transactions more difficult. 27Marulanda, 2004. 32 Figure 2. 2 Number of cashed checks per person in Colombia 4 ati 3.0 3 cap erp 3 2.4 2.0 2 1.8 ecks 1.6 ch 2 de sh 1 cafo.oN1 0 1999 2000 2001 2002 2003 No. of cashed checks Source: Superintendencia Bancaria 7. Proposed tax refunds will only be available to people with bank accounts. In an economy with a low savings rate such as Colombia, tax returns contribute to the savings capacity of low income people. This is supported by evidence from income tax returns in the United States. Recently introduced legislation in Colombia intends to return 2% of the 10% sales tax charged on purchases with debit and credit cards as a tax rebate28. The purpose of this law is to foster payments with cards, which makes evading VAT more difficult. The amount will be refunded through financial institutions by crediting the bank account of the purchaser. The tax refund is therefore only available to buyers who have a bank account. While the measure encourages the use of lower cost non-cash transactions among the banked, it precludes people without a bank account from access to an additional source of savings. 8. Under the current regulation, no obligation exists for banks to open accounts for low income clients. In view of their higher risk and cost of serving such clients, banks are often reluctant to provide such services. Recognizing the importance of access to basic financial services, other countries have introduced measures to make basic account services available to all clients with limited costs. These vary from government guidelines on basic account services to disclosure- based models such as the Community Reinvestment Act in the US. 28Ley 863 de 2003, Articulo 850-1. 33 Box 2: Examples of Regulatory Measures to Enhance Access Examples of experience in developed countries show different models how access to basic financial services can be achieved. The different models are based on various levels of government intervention. Example Germany ­ Raising awareness: A public debate about the right of everyone to have a checking account resulted in a voluntary undertaking in 1995 by all associations of Germany's private credit institutions to open checking accounts for everyone on a deposit basis. These accounts provide the client with access to cash withdrawal and payments, transfer and deposit services, but not automatically to overdraft facilities. Income levels and credit history are not accepted reasons for a bank to reject opening an account. The only reason why a bank can reject opening an account is client engagement in money laundering or other forms of fraudulent behavior. This declaration preempted national regulation and government intervention in the private banking sector. Consumer protection agencies monitor the compliance of private banks with this undertaking. Separately, public credit institutions, such as savings banks, are obliged by federal state law to open a checking account for everyone under similar conditions. Example Canada ­ Government guidelines: The Government of Canada set out guidelines in 2001 to ensure access to basic financial services for all Canadian citizens. These obliges banks to provide low cost accounts with the following features: (i) no charge for deposits, (ii) the use of a debit card, (iii) check-writing privileges, (iv) free updated account statement or passbook record, (iv) 8 to 15 debt transactions per month, (v) a maximum monthly fee of $4. The Financial Consumer Agency of Canada monitors if financial institutions respect their public commitments according to these guidelines. Example US ­ Government regulation: Through the community reinvestment act (CRA) of 1977, the US government obliged banks to reinvest deposits in the communities where they are received from, and having institutions serve the areas in which they have a physical presence. CRA established a point system which rates banks according to their provision of lending and savings services to low income neighborhoods. These ratings were subsequently published and taken into consideration in the approval of decisions on bank mergers and acquisitions and other regulatory decisions. Bank compliance with the CRA is subject to regular reporting and monitoring by the Federal Financial Institutions Examination Council. 9. With the exception of housing finance, no regulatory incentives exist to enhance the provision of credit to low income clients. Following the significant decline in available housing finance after the banking crisis, the Government put in place a framework to increase micro housing lending (see Box 3). As the household survey shows, demand for personal loans is high, in particular among the lower income groups (see Chapter 3). Measures to provide incentives for the "downscaling" of banks toward the microcredit market could be considered, based on the experience of other countries29. 29However, as the case of Brazil has shown, the provision of microcredit at subsidized interest rates, or based on earmarked resources, is not a sound basis for expansion. 34 Box 3: Micro Housing Finance in Colombia During the banking crisis, banks suffered considerable losses from residential mortgages with non-performing loan ratios exceeding 20%. As a result, banks remain reluctant to lend to the sector. The Government has taken a number of measures to promote lending to the sector, these include: (i) a rediscounting program with a state owned second-tier financial intermediary (FINDETER) for banks, credit cooperatives and finance companies to offer mortgage and real estate micro-credits to lower income communities; (ii) a voluntary agreement between Government and the banking sector to allocate a small portion of bank funding to the micro- housing finance sector (alternatively, banks can purchase bonds issued by FINDETER whose proceeds are used jointly with FINDETER funds for FINDETER's micro-credit and social housing programs); and (iii) an insurance program established by the Government's National Guarantee Fund to cover financial institutions against losses from housing micro-credit and social mortgage loans. Under the FINDETER social mortgage program, mortgage loans at subsidized interest rates are available for households which earn less than four minimum legal monthly salaries. Social mortgage loans are only available for the construction of new housing. 2.2 COST BARRIERS 10. The demand survey has identified lack of resources and high costs of bank accounts as key reasons why people do not use formal financial services. The following pages take a look at the costs of different banking services in the formal sector based on the limited material available to date. Sources include a survey by the Superintendencia Bancaria of financial services of 26 banks as of June 2002 and subsequent update, as well as interviews with five financial institutions in Bogotá, including four banks (Bancolombia, Colmena, Megabanco and Caja Social) and one financial corporation (Finamerica). A more comprehensive supply side analysis is recommended to validate the initial findings presented here. 11. Access to financial services for low income clients increases transaction costs. Serving low income clients increases transaction costs, as smaller transaction amounts reduce margins above fixed technological infrastructure costs30. On this basis, a comprehensive analysis of cost barriers to financial access should include an analysis from a client affordability perspective and an analysis of banks' cost structures. In the absence of detailed information on the latter, the following mainly focuses on the aspect of affordability. 12. Even basic savings accounts are hard to afford for low income clients. As demand studies in other countries have shown, low income clients are primarily interested in savings accounts as most important financial services. In the absence of any credit risk, they should also be the most affordable financial service for any type of client. In Bogotá, banks which explicitly focus on lower income clients, such as Banco Caja Social or Colmena, require a minimum deposit of COP50,000 to 30Interview with Patrick Tissot, Banco Caja Social, February 2004 35 open a savings account, corresponding to 14% of one minimum monthly wage. Additional costs for ancillary services such as a registration book and monthly administrative fees range from an additional 1.3% to 8.1% of a minimum salary, bringing the potential requirements for opening a savings account up to 26.5% of one minimum monthly salary. These costs rapidly decline as a percentage of income for higher income earners, falling to a significantly lower 4.1% of an average monthly income in the fifth income quintile. Excluding the minimum balance requirements, the monthly cost of a savings account still amounts to 12.6% of a minimum income. This represents high costs for the majority of the unbanked, who earn less than one minimum wage, effectively making even the most basic financial service hard to afford. Table 2. 2 Costs of savings services by income quintiles % half % min. % Quintile % Quintile % Quintile % Quintile % Quintile min. wage wage 1 2 3 4 5 Average monthly 179,000 358,000 379,241 727,087 846,080 917,584 2,321,841 income (in COP) Savings account Min. balance 50,000 27.9% 14.0% 13.2% 6.9% 5.9% 5.4% 2.2% Registration book 29,082 16.2% 8.1% 7.7% 4.0% 3.4% 3.2% 1.3% Monthly admin. Fee 4,808 2.7% 1.3% 1.3% 0.7% 0.6% 0.5% 0.2% Debit card 5,563 1.6% 1.5% 0.8% 0.7% 0.6% 0.2% processing 3.1% Other monthly fees 5,500 3.1% 1.5% 1.5% 0.8% 0.7% 0.6% 0.2% TOTAL 94,953 53.0% 26.5% 25.0% 13.1% 11.2% 10.3% 4.1% Excl. Min. balance 44,953 25.1% 12.6% 11.9% 6.2% 5.3% 4.9% 1.9% Source: Superintendencia Bancaria; Encuesta de Calidad de Vida 2003 13. Real interest rates for small savings amounts are frequently negative. The average interest rate paid by banks on savings accounts ranges from 2-7%, depending on the bank and the type of service provided (see Figure 2.3). The majority of banks do not pay interest below a certain minimum balance. For example, BanColombia, the largest bank in the country, starts paying interest on savings accounts on a minimum balance of COP300,000. Factoring in inflation of 7% in 2003, the real interest rate for small deposits is therefore negative. This provides limited incentives for savers to convert informal savings into savings with the formal banking sector. Focus groups confirmed that low income people generally do not trust formal financial institutions as they feel that they do not receive any remuneration on savings accounts, and fees are too high31. 31Marulanda, 2004. 36 Figure 2. 3 Interest rates on checking, savings and CD services 25 20 sknaBforebm 15 10 Nu 5 0 0% 2%- 5%- 7%- %7 NA > 0% 2% 5% Checking Savings CD >30 days CD <30 days Source: Superintendencia Bancaria, June 2002 14. Cost barriers for checking accounts are high. A bank such as Colmena requires a minimum balance of COP300,000 to open a checking account, corresponding to almost one minimum monthly salary. Additional account fees can add up to about 5% of one minimum salary. While fees are similar to those of savings accounts, the minimum balance requirements provide a prohibitively high barrier to entry for low income clients, in particular for the 50% of the unbanked who earn less than half a minimum wage, and for the additional 20% of the unbanked which earn less than one minimum wage (see Chapter 3). However, checking accounts have a relatively high cost even for higher income quintiles. The use of additional services such as ATM transactions costs low income earners an additional 0.4% to0.8% of their monthly income per transaction, while comparable fees for income earners in the highest income quintile correspond to a significantly lower cost of 0.1% of monthly income per transaction. However, some banks such as Conavi, Granahorrar, Megabanco, Santander or Banco Superior offer ATM withdrawals from their own network free of charge32. 32Marulanda, 2004. 37 Table 2. 3 Costs of checking account services by income quintiles Avg. % half % min. % Quintile % Quintile % Quintile % Quintile % Quintile Cost min.wage wage 1 2 3 4 5 Average monthly 358,000 379,241 727,087 846,080 917,584 2,321,841 income (in COP) Checking account Min. balance 300,000 167.6% 83.8% 79.1% 41.3% 35.5% 32.7% 12.9% Check value 2,057 1.1% 0.6% 0.5% 0.3% 0.2% 0.2% 0.1% Monthly admin. Fee 3,200 1.8% 0.9% 0.8% 0.4% 0.4% 0.3% 0.1% Monthly handling fee 7,050 3.9% 2.0% 1.9% 1.0% 0.8% 0.8% 0.3% Monthly fee other 5,250 2.9% 1.5% 1.4% 0.7% 0.6% 0.6% 0.2% TOTAL 317,557 177.4% 88.7% 83.7% 43.7% 37.5% 34.6% 13.7% Excl. min. balance 17,557 9.8% 4.9% 4.6% 2.4% 2.1% 1.9% 0.8% ATM transactions Cost of statement 1,594 0.4% 0.4% 0.2% 0.2% 0.2% 0.1% (own bank) 0.9% Cost of statement 2,888 0.8% 0.8% 0.4% 0.3% 0.3% 0.1% (other bank) 1.6% Cost of withdrawal 1,532 0.4% 0.4% 0.2% 0.2% 0.2% 0.1% (own bank) 0.9% Cost of withdrawal 2,918 0.8% 0.8% 0.4% 0.3% 0.3% 0.1% (other bank) 1.6% Source: Superintendencia Bancaria; Encuesta de Calidad de Vida 2003 15. Interestingly, credit cards are available in Colombia even for people without bank accounts, as outstanding balances can be settled by cash or by check. However, in the practice, only high income banked clients use credit cards. Processing and monthly fees for credit cards are relatively high, corresponding to 4-5% of a minimum monthly salary. In addition, eligibility criteria de facto preclude low income clients from access to credit cards. For example, banks such as Bancolombia require a minimum salary of at least 2.5 times the regulatory minimum salary for credit card issuance. Table 2. 4 Costs of domestic credit card services by income quintiles % half % min. % Quintile % Quintile % Quintile % Quintile % Quintile min. wage wage 1 2 3 4 5 Average monthly 179,000 358,000 379,241 727,087 846,080 917,584 2,321,841 income (in COP) Domestic credit cards Processing costs 5,250 2.9% 1.5% 1.4% 0.7% 0.6% 0.6% 0.2% Monthly fee 11,925 6.7% 3.3% 3.1% 1.6% 1.4% 1.3% 0.5% Other monthly fees 1,137 0.6% 0.3% 0.3% 0.2% 0.1% 0.1% 0.0% TOTAL 18,312 10.2% 5.1% 4.8% 2.5% 2.2% 2.0% 0.8% Source: Superintendencia Bancaria; Encuesta Calidad de Vida 2003 16. More bank services are available for higher income clients. Less than 10 out of the 26 banks surveyed by the Superintendencia Bancaria provide savings services for the lowest population quintile, by income level. In addition, none of the banks offer credit card services to this segment. This supply side finding confirms the demand side finding that income is a key factor in determining access to financial services (see Chapter 3). 38 17. While a growing number of banks provide microcredit services, interest rates are high. All five financial institutions interviewed for the purpose of this chapter provide microcredit services, with BanColombia having most recently started its microlending program (see Box 4). Low average loan amounts starting as low as COP1 million (approximately US$360) point to these institutions targeting small microentrepreneurs and their employees. However, with the exception of Banco Caja Social, the interviewed banks all lend at the regulatory lending limit for microcredit of UVR + 11%, currently equivalent to 29%. Even at these high rates, microcredit is frequently the only source of financing available to micro borrowers apart from informal lenders, or "loan sharks", with even higher interest rates. Box 4: BanColombia's Microfinance Program As part of a collaboration agreement between the national government and the financial sector, Colombia's largest bank, BanColombia launched a microfinance program at the end of 2002. The bank committed COP45 billion to investments in loans available to small low income entrepreneurs in Colombia. Companies eligible under BanColombia's program are those that have 10 or less employees, total assets of less than 501 times the monthly minimum wage, and a maximum indebtedness level of 25 times the monthly minimum wage. In August 2003, less than a year after the program was launched, BanColombia had disbursed over COP73 billion, outperforming its projections by 160%. By the end of 2003, BanColombia's micro-loan portfolio stood at COP 83 billion, an increase of over 20% in one year, illustrating significant demand for the provision of microfinance services from private banks. 18. The basic cost analysis shows that financial services in Colombia are hard to afford for low income clients, in particular for those earning less than one minimum wage per month. As this corresponds to 70% of Bogotá's unbanked, existing cost barriers result in the exclusion of a significant share of the population from access to financial services in Colombia. In addition, the higher relative cost of financial services for the poor results in a regressive income distribution: as the poor have to pay a significantly higher share of their income for the use of financial services, their remaining income decreases. In contrast, higher income people have to spend relatively less on banking fees, increasing their income accumulation for alternative uses. 19. Cost barriers need to be evaluated in light of banks' operational costs. In order to cover the costs of bank branches and technology such as ATMs, banks charge a minimum fee. Some analysts estimate transaction costs per transaction in a bank branch at between COP2,500 and COP3,000, while costs of an electronic transaction are estimated at a substantially lower COP50033. As Table 2.2 suggests, however, banks seem to be able to recover these costs, though further analysis of the banking sector cost structure would be necessary to confirm this. Banks with an unclear view of their own costs may have been failing to efficiently price their products, leading them to actually forsake revenue. 33Marulanda, 2004. 39 Table 2. 5 Products and costs of service by selected financial institution CURRENT ACCOUNT SAVINGS ACCOUNT Info requirements at opening Min. Min. Info requirements at opening Min. Min deposit Min balance Interest opening balance opening for interest balance deposit accrual Bancafe Salary of at least 3x min. wage. Not required Not required 1) Completed application form, 2) photocopy of Not available Not Not available Not available Other requirements not available national ID card available Ban- Work certificate; 2) passport or ID card; 3) last 3 Not available Not available Info requirements not available (See note A) 50% of Not required COP300,000 $300,001-$1 mill. 1.50% Colombia monthly income or account statements monthly min. $1,000,001-$5 mill. 2.25% wage $5,000,001-$15 mill. 3.50% $15,000,001-$ 4 5mill. 3.70% $45,000,001-$300 mill. 4.25% over $ 300 mill. 2.75% Ganadero- Not available COP 1 million Not required Info requirements not available (See note B) COP100,000 Not required Not available Not available BBVA Davivienda (See note C) COP200,000 Not required (See Note D) COP25,000 COP 10,000 COP 25,000 2.50% annual rate over Individuals: 1) completed application form, 2) photocopy Individuals: 1) ID card of citizenship card. If foreign (except for daily balance of national identity card. citizen, residence card or NIT. 2) If owner of a payroll and commercial store: certificate of registration in retirement Entities: 1) completed application form, 2) photocopy of Chamber of Commerce. accounts) NIT and ID card of representing person, 3) certificate of foundation and management, 4) photocopy of tax filing Entities: 1) Certificate of foundation and documentation and/or financial accounts, 5) if newly management, 2) NIT, 3) articles of incorporation, created company, initial constitution balance 4) Chamber of Commerce registration and proof of legal representation, 5) last tax filing documentation, 6) list of shareholders. Occidente 1) Completed application form, 2) two photocopies of Not available Not available 1) Completed application form, 2) two Not available Not available Not available Not available the national ID card, including fingerprint, 3) printouts of photocopies of the national ID card, including all bank account movements during last 3 months, 4) fingerprint. certificate from employer, no older than 30 days, stating position, years of experience and salary; 5) newest available tax filing documents, or income certificate and retentions; 6) if the individual is not eligible for tax filing, letter expressing this circumstance. Santander COP 500,000 Not required 1) Completed application form, 2) presentation of Less than 18 COP20,000 COP300,000 $300,000 - $1 mill. 1.00% Colombia Employees: 1) completed application form, 2) photocopy of $1 mill. - $5 mill. 1.98% the national identity card, 3) photocopy of the income and the national ID card. years old: $5 mill. - $10 mill. 3.20% retentions certificate, 4) Certificate from employer no older COP100,000 $10 mill. - $50 mill. 3.44% than 30 days, that includes position, years of experience $50 mill. - $3000 mill. 3.92% and salary, or last three payrolls. Older than 18: Over $3000 mill. 3.44% COP300,000 Self employed: 1) completed application form, 2) photocopy of national ID card, 3) photocopy of last available tax filing documents or letter stating status as non-tax payer, 4) report of bank activity in last 3 months. Banco Employees: 1) Completed appliction form, 2) photocopy of Not required Not required ID card Not required Not required Not required 3-5% for sight deposits Caja national ID card; 3) photocopy of income and fiscal 5-8% for time deposits Social retentions certificate; 4) Letter from employer stating position, salary and years of experience, or proof of retirement; 5) newest available tax filing documents, or letter stating status as non tax payer. Self employed with non-registered business: 1) application form, 2) photocopy of owner's ID card. Self employed with registered business: 1) Registration certificate in the Chamber of Commerce, no older than 60 days, 2) financial accounts, 3) tax filing document, or letter stating non tax payer status, 4) proof of bank activity of last 3 months, if available. Megabanco 1) Completed application form, 2) photocopy of ID card, COP500,000 Not required (See Note E) COP50,000 COP10,000 COP150,000 Not available 3) instruction letter and signed blank IOU, 4) certificate 1) Completed application form, 2) presentation of from workplace, 4) income and fiscal retentions the national ID card. certificate, or tax filing documentation. Colmena Same as Banco Caja Social Min. salary of COP300,000- Not available Not required Not required Not required $0 - $500,000 0,10% COP 1 million 500,000 Over $500,001 0.20% Notes: A) Bancolombia savings account conditions are based on "Bancolombia Ahorro" account; B) Ganadero-BBVA savings account conditions are based on "El Libreton" account; C) Davivienda current account conditions are based on the "Cuenta Corriente sin Sobregiro" account; D) Davivienda savings account conditions are based on the "Cuenta FijoDiario" account; E) Megabanco savings account conditions are based on the "Cuenta Ahorro Tradicional" account. Source: Interviews with bank managers, banks' websites 40 2.3 PHYSICAL BARRIERS 20. Another factor in access to financial services is the physical accessibility of formal financial sector services for low income clients. The limited economic literature available to date points to location as one determinant of access, while its relative importance compared to income and education is inconclusive34. Nevertheless, location is one of the barriers to enter formal financial sector services for low income households. The lack of bank branches in the neighborhood can prevent the unbanked from gaining access to formal financial sector services at all, while remote branch locations increase the costs of access for the poor. The following therefore takes a look at the geographical location of bank branches in Colombia and Bogotá drawing on the limited work conducted to date on this subject. 21. On a national basis, banks have more branches in higher populated areas. An analysis by Medina and Steiner35 based on data from the Superintendencia Bancaria shows that Colombia's banks focus their branch network on municipalities with more than 20,000 inhabitants. Seventeen out of the 27 analyzed banks by Medina and Steiner do not have a branch office in a municipality with less than 20,000 inhabitants and most other banks have less than 10% of their branch offices in municipalities with less than 20,000 inhabitants. An analysis by the Superintendencia finds that over 70% of total loans are provided to and about 70% of total deposits are collected in municipalities with more than 2 million inhabitants. 22. Bank branches are located in higher income areas. Medina and Steiner's analysis of bank branch distribution by income strata36 in Bogotá shows that all 27 of the surveyed banks focus on income strata 3 and above (see Figure 2.4). The average income strata for all bank branches is 3.6, showing an orientation of Colombia's banks towards higher income clients. In particular, the analysis found that foreign banks cater towards high income clients, as about 50% of their branches are located in areas of income strata 4 and 5. As Figure 2.5 shows, these strata are over-served in terms of bank branch presence relative to their share of population. In contrast, income strata 1 and 2 account for 25% of the population but are only served by 10% of bank branches. This demonstrates that the low income population in Bogotá is underserved in terms of bank branches, highlighting one reason for their frequent exclusion from financial services. 23. Anecdotal evidence from interviews with banks supports these findings. 34World Bank, 2004. 35Medina and Steiner, 2002. 36The income strata used in this analysis are the following: strata 1 (individuals earning less than half a minimum wage), strata 2 (half to one minimum wage), strata 3 (one to 1.5x the minimum wage), strata 4 (1.5x to 2x), strata 5 (2x to 4x) and strata 6 (more than 4x). 41 In interviews with the World Bank, major commercial banks confirmed their business focus on clients of income strata 3 and above. While Colombia's largest bank BanColombia has recently initiated a microcredit program which specifically caters towards low income entrepreneurs, this will remain a small proportion of the bank's overall business. Banks that traditionally serve lower end clients from income strata 1-3, such as Banco Caja Social and Megabanco, are the minority among Colombian banks. In fact, the five banks which have been identified as serving the lower income market37 only account for 18% of financial system assets at the end of 2003. Figure 2. 4 Distribution of bank branches by income strata 4.50-5.00 2 )(* s nk 4.00-4.50 ba 6 of taarts 3.50-4.00 14 gvA 3.00-3.50 5 - 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 # of entities in each income strata (*) Avg. income strata of banks is an average of the number of branches that each banking institution has, w eighted by the income level strata of the neighborhood w here each brank is located Source: Medina and Steiner (2002), based on Superintendencia Bancaria data Figure 2. 5 Distribution of bank branches by income strata relative to population distribution 50% 48.4% 45% 47.5% 40% 35% % noital 30% 25% 22.1% 19.7% opuP 20% 17.5% 15% 11.7% 10.6% 8.9% 10% 6.7% 3.4% 5% 2.8% 0.9% 0% 1 2 3 4 5 6 Population distribution per income strata Distribution of bank branches per neighborhood income strata Distribution of population per income strata Source: Medina and Steiner (2002), based on Superintendencia Bancaria data 24. Credit cooperatives have a higher presence in lower income areas than banks. A separate analysis by Medina and Núñez38 about bank representation in Bogotá compares branch presence of banks with financial credit cooperatives. The analysis shows that bank branches are concentrated in higher income areas. Medina's analysis statistically 37Includes Banco Popular, Banco Agrario, Banco Caja Social, Banco Colmena and Megabanco. 38Medina and Núñez, 2001. 42 confirms the hypothesis that bank branches are underrepresented in low income neighborhoods in Bogotá, while the hypothesis is not valid for financial credit cooperatives. When taking financial credit cooperatives into account, Medina's analysis finds that lower income areas in Bogotá are significantly better served in terms of branch presence. 25. However, within an urban area, bank branch presence is driven by other factors in addition to current income levels. Medina's analysis finds that the number of bank branches in Bogotá is positively correlated with the level of permanent household income and welfare. In addition, positive correlations are found between the number of bank branches and the security of the neighborhood, the health of its infrastructure and the presence of religious and cultural institutions. On the other hand, the presence of bank branches is found to be negatively correlated with population density, an area's unemployment rate and the percentage of an area occupied by industrial activity. The particular importance of security in access to finance is highlighted by analysis which shows that the cost of handling cash is a key factor in bank branch profitability39. 26. Focus groups also revealed that remote bank locations are a key factor in financial exclusion for people in income strata 1, while people in income strata 2 and 3 did not mention location as a key reason for lack of financial access40. For Bogotá, the demand survey confirms that 62% of the unbanked are over 10 blocks away from the next bank branch (see Chapter 3). Greater distance to bank branches increases transaction costs for low income clients and the unbanked. 27. Alternative access mechanisms are limited. ATMs provide a cost- effective alternative of physical access to formal financial services, such as payment services, check deposits or savings deposits. While the number of ATMs in Colombia has remained broadly stable over the past few years, the number of ATM transactions increased, pointing to increasing acceptance of ATMs among clients. No further data is available regarding the location of ATMs and their proximity to low income neighborhoods. 2.4 CULTURAL BARRIERS 28. In addition to regulatory, cost and location barriers, other subtle barriers exist between the providers of formal financial services and the unbanked. Banks are not used to dealing with a different client profile of low income clients. Given their smaller transaction size, potential irregular income, limited asset base and collateral, and on average lower education (see Chapter 3) different business models are required for this 39Arango, 1999. Banks with comprehensive national coverage in Colombia, such as Banco Agrario, frequently have to resort to transporting cash by helicopter due to the security situation. 40Marulanda, 2004. 43 group of clients. In addition, different means of attire and communication are necessary in order to familiarize this client base with banking services41. Focus groups have shown that low income clients frequently do not know or are confused about the features and the small print of the banking products they use, resulting in higher costs for them than they expected42. Finally, the frequent practice of having a separate line for unbanked clients increases the feeling of discrimination and distrust towards banks from the part of the unbanked. 41A simple comparison of website design of high-end banks in Colombia with that of banks focused on low income people has shown that the latter work with very different means of communication, including a much simpler design, bright colors and simple, slogan-like phrases. 42Marulanda, 2004. 44 CHAPTER 3: USERS OF FINANCIAL SERVICES AND PROFILE OF THE UNBANKED 1. This chapter presents the demand side perspective of financial exclusion in Bogotá. To this end, the chapter draws on data from several sources. The introductory part is drawn from the results of a nationwide survey by the Colombian Association of Media Investigation (ACIM). The presentation of the socio-economic characteristics of the unbanked is the result of a joint household survey by the Colombian government's department of statistical studies, DANE, the World Bank and Universidad de los Andes. 2. ACIM carried out a household consumer survey (the General Media Survey or EGM, acronym in Spanish) through the National Consulting Center in 2000. A section of the survey centers on the possession of financial services. The survey was applied to 12,218 households in 16 of Colombia's most important cities: Bogotá, Barranquilla, Cartagena, Medellín, Pereira, Manizales, Cali, Popayán, Ibagué, Bucaramanga, Neiva, Cúcuta, Villaviciencio, Santa Marta, Armenia and Tunja. This sample is representative for a population of 11,850,560 inhabitants and was extrapolated to the total population. The survey tabulated use of: savings and checking accounts, credit and debit cards, certificates of deposit (CDT), mortgages, personal credit, banking services by telephone, savings account with special awards, stocks and bonds. 3. Financial services, whether provided by banks, by cooperatives or by micro-finance institutions, reach a very limited percentage of Colombia's urban population. The ACIM survey, which covered 16 Colombian cities, suggests that the percentage of unbanked increases outside of Bogotá. Figure 3.2 shows the percentages of households using different types of financial services. Considering the unbanked to be households without access to savings accounts, the ACIM survey suggests that the unbanked in urban Colombia reach close to 70%. A similar survey by the same company, in Mexico, reported findings of unbanked households at 85%. 4. The DANE survey, applied to a representative sample of 1,600 Bogotá households in 2002, was sponsored by the World Bank, but designed and implemented by the Center for Economic and Development Studies in the University of Los Andes (CEDE) in conjunction with the National Administrative Statistics Department (DANE)'s program of Household Surveys (ECH). 5. In this study the term `banked' refers to households and to individuals who maintain deposits (savings, checking or CDT account) in some formal sector financial institutions, including commercial and social interest banks, regulated cooperatives and credit unions (these last 45 including the "cajas de compensación" associated with certain employee organizations) . The `unbanked', on the other hand, have no deposits or savings in any such products mentioned above. 6. Less than 40% of Bogotá's adult population use formal sector financial services. This percentage is higher than in Mexico, but lower than the proportion prevailing in Brazil (Figure 3.1). Figure 3. 1 Use of Financial Services In Bogotá, Mexico City, urban Brazil and OECD countries 100 50,000 91 90 45,000 onita 80 37,610 40,000 70 35,000 $)SU( popul de 60 30,000 2003 nkab 50 ni 43 25,000 39 ati of 40 20,000 cap geatnecreP 30 24 15,000 erp 20 10,000 6,230 GNI 10 1,810 2,710 5,000 - - Mexico D.F Bogotá Brazil USA Banked population GNI per capita Source: Mexico, Bogotá and Brazil data, own studies,- US Federal Reserve Bank, 2002 Figure 3. 2 Use of Banking Services as a Percentage of Population Funds/ Stocks/ Bonds 0.7 CDT 2.5 Mortgage 2.9 t Savings Account w ith special aw ards 3.0 uc odrp Personal Credit 3.8 of Use of telephone for banking services 3.8 pe Ty Current Account 7.2 Credit Card 8.0 Debit Card 17.9 31.7 Savings Account 0 5 10 15 20 25 30 35 40 % of population that use each product Source: Colombian Association of Media Investigation ACIM, General Media Survey EGM for the year 2000 46 3.1 PROFILE OF THE UNBANKED 7. Being unbanked is clearly associated with lower incomes. The majority of the unbanked are poor, but the unbanked are not just the poor. As Figure 3.3 shows, the percentage of unbanked persons increases as incomes fall. Slightly more than half (51%) of bank account holders belong to the top three deciles of income distribution, and about a third (34%) belong to the group earning less than the median income. Of the unbanked, over half (53%) earn less than half the minimum wage and 70% earn less than one minimum wage. On the other hand, of those earning over 4 minimum wages, 84.22% are banked and 15.78% unbanked. At the other end of the income scale, it is notable that no less than a quarter of the families in the lowest 40% of income distribution (those earning less than half a minimum wage) are banked. This data shows that even very poor households have the capacity to be banked. Figure 3. 3 Use of banking services in relation to income 100% 90% 81% 80% 74% 75% 73% 76% 70% 70% 64% 59% 60% 55% 56% % 50% 45% 41% 44% 40% 36% 30% 30% 26% 25% 27% 24% 19% 20% 10% 0% 1 2 3 4 5 6 7 8 9 10 Population distribution by income level Banked Unbanked Source: DANE household study of Bogotá, 2003 8. The unbanked have three times greater unemployment than the banked (10% are unbanked and unemployed versus 3.6% banked and unemployed). Figure 3.4 shows that the banked population prevails among the government employees43, formal sector employees, and employers, most of whom also benefit from social security and health insurance programs. On the other hand, the unbanked tend to work in areas of lower job security and beyond the social safety networks. The unbanked are more likely to be among the self-employed (72% of the total, which includes informal sector workers), domestic workers (81.5%) and unpaid workers in family businesses (78%). The financial 43Perhaps due to recent policy decisions encouraging payment by direct deposit to bank accounts, 83% of government employees are banked 47 exclusion of informal, or irregular workers is likely also be a function of banks' preference for clients with steady employment, reflected in requirements for references and proof of work (see Chapter 2). At the same time, almost 50% of salaried workers are unbanked, suggesting that informality is not a primary reason for financial exclusion. Figure 3. 4 Employment profile of the unbanked Banked and unbanked by type of activity Banked and unbanked by type of job 100% 50% Banked Banked 50% 90% 48% Unbanked 78% Unbanked 80% 76% 73% 40% 70% 36% 60% 60% 48%52% 30% % 50% 40% 24% 40% % 30% 22% 24% 27% 20% 14% 20% 9% 10% 10% 7% 0% 2% 4% 3% 3% ngi s 1% 0% es ed 0% ousH horc udenttS oypl hertO ed 0% oypl cti rek nemU mE deiral nirekr rot ecs deyolp es rekr eryopl ortces herto sa wo eatvi me-fl dom wo wor em yli deiral nirekr sa cibl pr se mfa yaptuoht wo wi pu Source: DANE household study of Bogotá, 2003 9. The unbanked have lower educational attainment levels than those with bank accounts. Figure 3.5 shows a clear relation of education to the use of formal banking products. As educational levels increase, the percentage of unbanked falls and the percentage of banked persons rises. Regression analysis shows a high correlation between educational levels and likelihood of being banked or unbanked. At the same time, it is noteworthy that approximately 15% and 25% of the group with less than elementary and elementary school only respectively do have bank accounts in formal sector institutions. Figure 3. 5 Education levels of the banked and unbanked 100% 90% 85% 79% 80% 75% 70% 61% 60% 50% 50% % 50% 39% 40% 30% 25% 21% 20% 15% 10% 0% Incomplete Complete Complete Incomplete Complete Primary Primary - Secondary Superior Superior Education and Incomplete Education Education Education and less Secondary More Banked Unbanked Source: DANE household study of Bogotá, 2003 48 10. Notwithstanding their lower income, education levels, and location, home ownership among the unbanked is high. 61% of the unbanked in Bogotá own their home, showing a similar level of home ownership than the banked (66%, see figure 3.6). These levels are similar to home ownership rates in the United States (65%). The minimal difference in home-ownership rates for banked and unbanked suggests that access to bank loans is not a limiting factor in accessing homes. The high rate of ownership overall suggests that both unbanked and banked Colombians view their homes as key to their asset-building strategies. The high rate of home ownership among the unbanked indicates their capacity to accumulate assets, and, as a result, to potentially benefit from access to financial services. Figure 3. 6 Home-Ownership among the banked and unbanked Banked Unbanked 6% 4% 13% 9% 53% 52% 30% 33% Owned by Household, totally paid Owned by Household, totally paid Rented by Household Rented by Household Owned by Household, being paid Owned by Household, being paid Used by household without paying Used by household without paying Source: DANE household study of Bogotá, 2003 11. The unbanked are further from access to banks than the banked. The DANE survey does reveal that over 60% of the unbanked either don't know where a bank branch is in relation to their homes, or report the closest banking institution to be over ten blocks away. In the case of the banked population in Bogotá, 50% of respondents report a distance to the next bank of over 10 blocks or an unknown distance. In particular, 22% of the unbanked live more than 20 blocks from the next bank compared only 13.5% of the banked in Bogotá. 12. In conclusion, the majority of the unbanked share the characteristics of the urban underclass with lower income, lower education, closer links to the informal sector (and consequent economic and social insecurity), with a greater tendency to live outside higher income census districts where banks, commerce, religious and health centers are located. The unbanked show no particular difference of age or gender distribution from either the banked population or from the population at large nor do they differ from the overall population in terms of home-ownership. At the same time, the data reviewed above 49 suggests that the unbanked are far from a homogeneous group. While they tend to be poorer and less educated as a group than the banked, the unbanked of Bogotá also include individuals from high income groups. 3.2 PAYMENT SERVICES FOR THE UNBANKED 13. The majority of the unbanked receives their salary in cash, while the banked are more likely to receive paychecks or have their salaries deposited directly into savings accounts (see Figure 3.7). Handling cash payments makes the unbanked more susceptible to loss and robbery than the banked. Unlike in Brazil, where the majority of respondents prefers to be paid in cash, over 90% of the banked individuals in Colombia who receive payments by check and direct deposits responded to the question how they would like to be paid that they prefer the existing system. Figure 3. 7 Payment of salaries to the banked and unbanked 0.3% Other 0.1% Deposit in checking 0.5% account 4.4% 5.0% Check 9.4% Deposit in savings 5.0% account 52.4% 89.2% Cash 33.8% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Population by income level Banked Unbanked Source: DANE household study of Bogotá , 2003 14. Cashing checks is costly for the unbanked. One third of the people paid by check are unbanked, corresponding to 5% of the unbanked. The unbanked cash their checks in banks in 83% of the cases. The remainder change the checks with family members of friends, check cashing outlets (CCO's), commercial stores, or other unidentified methods. In addition to the fees charged by banks or check cashing outlets (between 5% and 20% of the check's face value), the time spent in accessing check cashing locations represents an additional cost for the unbanked. As Colombian banks do not honor checks drawn on other banks except for deposit account holders, checks have to be cashed at the bank on which they are drawn. According to the survey, 60% of the unbanked live over ten blocks away from a formal bank. Access to CCO's appears slightly easier for the unbanked as 32% of the unbanked live closer than 10 blocks to a CCO. 50 Figure 3. 8 Location of cashing pay checks 100% 90% 83% 80% 70% 60% 53% 47% % 50% 40% 30% 20% 13% 10% 5% 1% 0% Cash it in a bank Other Deposit in account Banked Unbanked Source DANE household study of Bogotá, 2003 15. The unbanked are also disadvantaged when making payments. Colombian banks process payments for public services (water, sanitation, electrical energy, garbage collection, telephones etc.) for both deposit holders and non-deposit holders. Banked individuals can make payments using traditional means such as check or cash, or instruments such as ATM, debit, credit cards, internet and telephone payments. Public utility payments can also be made with debit, credit and ATM cards using data phones at a wide variety of places such as supermarkets, video stores and malls. However, the unbanked who need to pay in cash are obliged to travel to Administrative Centers of Services44. Commercial banks only accept payments in cash for public utilities at specified hours and days. In addition, only one line is available for the unbanked to make payments in banks and processing late or overdue payments bears additional costs. 16. Both the banked and unbanked receive income from remittances. Approximately 6% of Bogotá's population receives remittances from other cities (3.2%) and from abroad (2.4%). Of these, 58% of recipients are banked, while 42% are unbanked. When receiving domestic remittances, 60% of the banked rely on safer and cheaper wire transfers. In contrast, the majority of the unbanked receive domestic remittances in hand carried cash. Interestingly, both the banked and unbanked receive remittances from abroad mainly through remittance agencies. However, the banked individuals use direct wire transfers and checks in another 40% of the cases, while the unbanked rely on cash for another 43% of the cases. 44 Special District Attention Centers (CADES, or Centros de Atención del Distrito Especial), are administrative offices located in different neighborhoods of Bogotá and its periphery, where citizens pay taxes and basic public services. 51 Table 3. 1 Channels of receiving remittances Domestic remittances Remittances from abroad Channels Banked % Unbanked % Banked % Unbanked % Cash 20,175 28% 32,740 56% 7,067 11% 17,716 43% Remittance Agency 7,337 10% 5,798 10% 28,232 45% 19,199 47% Deposit in Account 25,186 35% 9,244 16% 2,909 5% 0 0% Transfer in Account 18,054 25% 10,821 19% 15,747 25.% 0 0% Other 1,271 2% 0 0% 4,237 7% 3,954 10% Total 72,023 100% 58,603 100% 62,707 100% 40,869 100% Source: DANE household study of Bogotá , 2003 17. The cost of sending and receiving remittances is high. Receiving hand carried cash remittances is not secure and takes time for the unbanked. In order to have access to more secure alternatives such as deposit and account transfer, the unbanked frequently have to pay an obligation or cost for using someone else's account45. Remittance agencies regularly charge up to 25% of the received amount in fees. International remittance agencies charge substantial fees both at the sending and receiving end of remittances. For example, to send US$100 from New York to Bogotá, Western Union charges 12% to the sender46 and another 7% to the receiver for changing dollars to Colombian pesos. In addition, the remittance sender is responsible for advising the receiver, either by phone, mail or personal message and transmit the order number, since the remittance company does not offer this service. The cost of an international phone call thus needs to be added to the explicit charges. In contrast, fees charged by the formal banking system can be substantially lower, illustrating the high cost of being unbanked. Fees for transferring money from a US bank to a Colombian bank can be as low as $9 (fixed fee) payable by the sender with no additional charges to the recipient. Given the importance of remittances to the Colombian economy47, reducing their costs by providing the unbanked with access to the formal banking system, can thus yield significant economic benefits. 3.3 SAVINGS SERVICES FOR THE UNBANKED 18. The savings capacity of the unbanked in Bogotá is low. In the survey, only 9% of Bogotá's unbanked or 5.5% of the total population reported having any form of financial savings. This percentage reflects 45Further work is necessary to understand better the exact costs in time and money for the use of third persons' accounts. 46Many banks charges a flat US$60- $100 for sending any amount up to $1,000 ­ a costly rate for small amounts. Further study should advise how other institutions scale remittance charges and whether they are reduced or increased as the amount sent increases. 47Remittances represent about 90% of foreign direct investment in Colombia and are the second largest source of foreign currency entering the country after petroleum exports (Colombian Foreign Relations Ministry). 52 the generally low savings capacity in Colombia compared to other countries48. A simultaneous survey of banked and unbanked people in Mexico City revealed that 45% of the total population held financial savings and 47% of the unbanked were savers. The considerable difference in savings capacity between Bogotá and Mexico City can be attributed to the generally low savings capacity in Colombia. However, the numbers need to be treated with caution due to certain limitations of the Bogotá survey49. 19. Informal savings instruments are used by both banked and unbanked individuals. Informal savings are used by 13% of the population in Bogotá, compared to 39% of the population in Mexico City. In Bogotá and in Mexico City, banked individuals outnumber the unbanked in their use of informal savings services. As Table 3.2 shows, 19% of banked individuals in Bogotá (7.5% of the total sample) use informal savings, twice as much as the unbanked, with 9% usage of informal savings services (corresponding to 5.4% of the total sample). 20. The majority of the unbanked saves at home. Home savings is used three times more frequently by unbanked persons than by the banked population, while other forms of informal savings are as likely to be used by persons with bank accounts as without. There was no notable difference between unbanked and banked regarding the use of "tandas" (savings clubs with friends), savings clubs at stores and other organizations and loans to friends and families. 21. The cost of informal savings is high. Informal savings systems do not offer interest, so that savings lose value with inflation. In the case of savings clubs and "lay-away plans" at stores, informal savings can be shown to lose a margin in excess of the inflationary loss, since savings are frequently traded in for certain items sold above list price. In addition, informal savings systems do not benefit from any institutional deposit insurance, having to rely on personal trust mechanisms instead. Table 3. 2 Banked versus unbanked users of informal savings services Banked % Unbanked % Total % Uses informal savings 318,211 19.0% 237,424 9.1% 555,635 12.9% Does not use informal savings 1,359,091 81.0% 2,379,201 90.9% 3,738,292 87.1% Total 1,677,302 100% 2,616,625 100% 4,293,927 100% Source: DANE household study of Bogotá, 2003 48The savings rate in Colombia has been constantly declining since 1990 from 24% of GDP to 14% of GDP in 2003. 49 The DANE did not gather sufficiently robust information on the savings behavior of the interviewed people. A second application of the survey with revised questions is therefore recommended. In contrast, the México City survey included several in-depth questions on savings behavior. 53 Figure 3. 9 Banked versus unbanked users of informal savings services 100% 90% 80% 42.7% 70% 63.6% 60% %50% 40% 30% 57.3% 20% 36.4% 10% 0% Used informal savings during the last Did not use informal savings during the 12 months last 12 months Banked Unbanked Source: DANE household study of Bogotá, 2003 3.4 CREDIT SERVICES FOR THE UNBANKED 22. Informal credit is used by both the banked and unbanked. 14% of the survey population uses informal credit services. In line with the finding on informal savings services, demand for informal credit is higher from banked than from unbanked individuals. 58% of informal credit users are banked, while 42% are unbanked. This finding could be partly explained by the fact that the among the unbanked only 9% applied for an informal loan during the past three years, while 20.2% of the banked did, increasing the likelihood of the banked population over the unbanked to obtain informal credit. Table 3. 3 Banked versus unbanked users of informal credit services Banked % Unbanked % Total % Uses informal credit 339,061 7.9% 244,300 5.7% 583,361 13.6% Does not use informal 1,338,240 31.2% 2,370,769 55.2% 3,709,009 86.4% credit Total 1,667,301 39.1% 2,615,069 60.9% 4,292,370 100.0% Source: DANE household study of Bogotá, 2003 54 Figure 3. 10 Requests for informal lending services 100% 90.7% 86.4% 90% 79.8% 80% 70% 60% 50% 40% 30% 20.2% 20% 13.6% 9.3% 10% 0% Banked Unbanked Total Population Requested an informal loan during the past 3 years Did not request an informal loan during the past 3 years Source: DANE household study of Bogotá, 2003 23. The unbanked have less access to credit sources with a higher degree of formality. As Figure 3.11 shows, the unbanked mostly use friends or family and money lenders as sources of credit. In contrast, the banked are more frequent users of credit services provided by commercial stores, employment funds or informal savings funds. This suggests that access for the unbanked to these sources of lending is limited. Figure 3. 11 Source of informal lending 100% 88.8% 90.5% 90% 76.8% 80% 74.7% 70% 60% 51.3% 54.0% 48.7% 50% 46.0% 40% 30% 25.3% 23.2% 20% 11.2% 9.5% 10% 0% Friends or Money Commercial Employment Informal Other family lenders store fund savings fund Banked Unbanked Source: DANE household study of Bogotá , 2003 24. While reasons to request a loan vary by income level, personal expenses and housing finance are key demand drivers. In particular, up to the 40th income percentile, demand to borrow for personal expenses rates higher than any other reason. In the "middle class" (41st to 80th percentile) housing finance is the most important reason for borrowing. At all income levels, demand for personal expenses plus health and education is greater than demand for either housing or business expenses. This range of demand could also help explain why 55 even banked persons request loans from informal lenders, as personal loans may be harder to obtain from formal lending institutions than business loans or subsidized housing loans. However, as discussed in Chapter 1, consumer lending in Colombia has grown significantly over the past two years. Table 3. 4 Reasons for requesting a loan Income percentile 0-20% 21-40% 41-60% 61-80% 81-00% Personal Expenses 34.5% 32.2% 23.4% 17.5% 21.9% Education or Health 8.3% 7.7% 17.4% 20.2% 17.9% Vehicle 7.4% 11.1% 2.1% 6.7% 16% Housing, including equipment and furniture 25.6% 31.4% 30.3% 28.5% 13.5% Business, investment, equipment and repairs 14.0% 9.2% 12.3% 10.6% 20.3% Payment of debts 8.0% 6.5% 10.6% 13.2% 9.1% Other 2.3% 2% 4% 3.3% 1.4% Totals 100.0% 100.0% 100.0% 100.0% 100.0% Source: DANE survey, 2003 25. More people seek loans in the informal than in the formal sector. The Bogotá survey reveals that the banked population is more likely to seek loans than the unbanked - by a factor of six to one (see Figure 3.12). Among the population surveyed 13.6% of the total population successfully sought loans with the informal sector in the past thirty six months while another 8.6% successfully sought loans from the formal sector. This proportion is higher than that reported in other studies. Only 10% of those surveyed in Mexico City reported requesting a loan in the past 36 months from either formal or informal sources. Figure 3. 12 Loan request from formal institutions 100% 97% 91% 90% 82% 80% 70% 60% 50% 40% 30% 18% 20% 9% 10% 3% 0% Banked Unbanked Total population Requested a formal loan in the past 3 years Did not request a formal loan in the past 3 years Source: DANE household study of Bogotá, 2003 26. Higher loan approval rates help explain the relatively higher importance of credit obtained outside the banking system. Table 3.5 shows the approval rate, according to the would-be borrowers surveyed, 56 of different lenders. Workers' Compensation Funds50 associated with large scale employers show the highest approval rates, while their funding is restricted in purpose and amount per client. The second best option is borrowing from a friend or relative. While the formal institutions (banks, finance corporations and cooperatives) offer lower active interest rates, they are also much more selective, turning down three out of four loan requests. As figure 3.13 shows, the banked population is far more likely to get a bank loan approved than the unbanked population. Higher loan approval rates in the informal sector therefore drive the unbanked towards borrowing at higher interest rates, reflecting a significant cost of being unbanked. Table 3. 5 Loan approval rates by different credit providers Commercial Finance Friend or Workers' Informal Bank Corporation Cooperative Relative Prestamista51 Store Compensation Fund Saving Approval Rate 21% 23% 25% 95% 87% 87% 100% 80% Annual Int. Rate 15% 16% 13% N/A 150% 150% 15% N/A Source: DANE household study of Bogotá, 2003 Figure 3. 13 Loan approval rates of banked and unbanked individuals 100% 90% 77.4% 80% 72.1% 70% 60% 51.0% 50% 43.8% 40% 30% 25.8% 21.3% 20% 10% 5.1% 1.4% 2.1% 0% Banked Unbanked Total Approved Not approved In course Source: DANE household study of Bogotá, 2003 27. Credit is not used because it is not perceived as a necessary economic strategy or desirable financial service. Lack of demand is given as the main reason for not requesting a loan, independent of whether people are banked or unbanked or whether loans come from informal or formal sources. When asked about their reasons for not seeking loans, the majority listed: i) not needing loans; ii) not liking to be in debt; iii) preferring to use own resources; or iv) not enough income. All these answers, accounting for over 75% of the population, suggest a general aversion towards going into debt. In other words, most people do not see borrowing as a strategy for survival or upward 50We consider this as an informal credit mechanism because it falls outside the banking system. 51An informal lender, loan-shark, pawn-broker. 57 mobility. In the case of formal sector loans, a small percentage list paper work as a reason for not seeking loans, in particular among the banked. Interestingly, the percentage of persons who give the high cost of credit as reason for not taking loans is higher in the formal sector than in the informal sector ­ even though interest rates tend to be higher for informal credit. 28. In contrast, it is worth noting that the majority of banked and unbanked consider savings and deposits as highly desirable financial services, as evidenced by other studies such as surveys in Mexico and Brazil. Despite these findings, policy makers tend to focus on credit for the unbanked as a primary financial service, and tend to discount the importance of savings and payment services in this context. As the DANE questionnaire did not include questions about savings habits and costs, a second, more detailed survey could yield important information on these points. Figure 3. 14 Reasons for not requesting a loan from the formal sector Other 1.7% 1.8% Lack of collateral/guarantees 2.6% 2.7% High cost of credit 3.3% 7.2% Too much paper w ork 5.9% 11.2% No w ork/job 11.9% 4.8% Prefers ow n resources 15.1% 10.6% Don't like to get into debt 17.6% 19.1% Not enough income 19.0% 8.2% Didn't need 22.9% 34.4% 0% 5% 10% 15% 20% 25% 30% 35% 40% Banked Unbanked Source: DANE household study of Bogotá, 2003 58 Figure 3. 15 Reasons for not requesting a loan from the informal sector Other 1.5% 1.5% High cost of credit 2.0% 5.4% Lack of collatera/guarantees 2.6% 2.7% No w ork/job 12.3% 4.3% Not enough income 16.0% 7.2% Prefers ow n resources 16.4% 15.2% Don't like to get into debt 22.4% 22.2% Didn't need 26.6% 41.5% 0% 10% 20% 30% 40% 50% Banked Unbanked Source: DANE household study of Bogotá, 2003 3.5 THE COSTS OF BEING UNBANKED 29. The survey does not itemize costs of informal and formal financial services. Further analysis is encouraged in this regard. Conceptually, the costs paid for financial exclusion by the unbanked include: · The cost of informal savings in terms of insecurity, values lost to inflation and to interest charges from lay away plans, or of outright losses in lottery programs. · The costs of cashing checks at CCO's · The costs of remittances · The costs of lost IVA returns (i.e. 2% of all purchases which are returned through bank accounts) · The costs of paying bills ­ time lost traveling to utility companies to make payments in person · The costs of informal loans, averaging 10% per month or 150% annually, versus 15- 27% annual interest rates for bank loans and 36% for credit cards. 3.6 REASONS FOR BEING UNBANKED 30. In answer to the question "what is your main reason for not having a bank account?" the majority of the unbanked pointed to their lack of resources. The percentages of persons stating they lacked resources 59 either to pay for or to justify having bank accounts declined as income levels increased. While 75% among the lowest 20% of income distribution pointed to the cost of banking, or to their lack of resources, this is only an issue for 50% of the wealthiest 20%. The higher income groups point, instead, to not needing an account (24%) and to high management costs (13%). While banking services are considered to be costly by persons at all income levels, lower income groups in particular believe that their incomes are too low to justify using banking services. This confirms the finding of the supply side cost analysis (see Chapter 2) that the costs of banking services in Colombia are too high for low income people to afford, in particular those who earn less than one minimum wage per month. Figure 3. 16 Main reason for not having a bank account 12% Q5 13% level 24% 51% 7% ome Q4 11% 17% nci 65% by 6% seli Q3 9% 15% nt 71% qui 6% onita Q2 7% 14% 73% ul opP 8% Q1 6% 10% 76% 0% 10% 20% 30% 40% 50% 60% 70% 80% % Lack of Resources Does not need an account High Costs Other Source: DANE household study of Bogotá, 2003 3.7 ARE THE UNBANKED BANKABLE? 31. Given their different socio-economic characteristics, the question has been raised whether the unbanked have "bankable characteristics". Home ownership rates by the unbanked above 60%, with minimal differences to the level of the banked, illustrate that the unbanked have asset building capacity. Access to formal financial services would therefore provide the unbanked with the opportunity to build financial assets through savings and increase their asset base through access to credit services. 32. The DANE survey shows a high correlation between educational levels and likelihood of being banked or unbanked. At the same time, it is noteworthy that approximately 15% and 25% of the group with less than elementary and elementary school only respectively do have bank accounts in formal sector institutions. This suggests that while 60 educational attainment may be an impediment for accessing formal banking services, it can be addressed. Financial education can therefore play an important role in bridging the gap between formal financial institutions and the unbanked. 33. While the unbanked are generally associated with lower income, interviews with formal bankers confirm that they are not necessarily perceived as higher risk clients (see Chapter 2). Therefore the question is not `whether the unbanked are bankable' but `how to bank the unbanked'. To this end, innovative products and low cost approaches are necessary. 61 CHAPTER 4: CONCLUSIONS AND RECOMMENDATIONS 34. This paper attempted an initial overview of the state of access to financial services for low income clients in Colombia. Taking a look at the financial system as a whole, the recent history of Colombia's financial sector suggests a decline in financial access. The impact of the banking crisis has resulted in decreasing financial intermediation in Colombia, as the main financial intermediaries, private banks, have turned towards less risky sources of income, in particular government bonds. Currently, Colombia's main financial intermediaries have limited incentives to enhance access to their services for the poor, however, further growth of their balance sheets will require reassessing the importance of the retail market, including lower income clients. Low levels of access are illustrated by decreasing numbers of checking and savings accounts per inhabitant, as well as a declining financial depth in terms of credit to GDP. 35. The report identifies the low penetration of basic financial services among low income people in Colombia as both a supply and demand side issue. On the supply side, lack of access for the poor is the result of high costs of most existing financial services, of limited incentives to provide services to low end clients, of the relatively higher costs of dealing with small transaction amounts and higher risks of this client segment, of a lack of sufficient information, and of the lack of adequate products for a clientele with different socio-economic characteristics. 36. On the demand side, lack of access to basic financial services for the poor and unbanked stems from the relatively high price of financial services, and from a lack of information ­most unbanked are less familiar with formal banking services­. In addition, cultural barriers prevent a constructive dialogue between the unbanked and formal financial intermediaries. 37. As a result, the existing gaps between supply and demand lead to the exclusion of a significant share of the population from formal financial services. This study found that less than 40% of Bogotá's adult population has access to formal financial services. Financial exclusion creates significant welfare losses not only for the unbanked as a result of higher transaction costs, but also for the economy as a whole, as their informal savings cannot be transformed into formal savings and investments. 38. The regulatory framework plays a role in financial access. Apart from housing finance, Colombia's current regulatory framework does not explicitly address the provision of financial services for low income populations. In addition, the existing anti-money laundering regulation and financial transaction tax create high barriers to entry into the formal banking system. 62 39. The following provides an overview of potential measures to improve the access to formal financial sector services for the low income population in Colombia. It draws on the areas for correction which became apparent from the supply and demand side analysis and potential regulatory steps, including experience from other countries. 4.1 REGULATORY MEASURES 40. As the experience of other countries shows (Chapter 2, Box 2), regulation to enhance financial access can take various forms. Models such as the US Community Reinvestment Act track the extent which banks to provide specific services to low income communities. In this case, service provision is enhanced through detailed information and disclosure requirements, and affects approval on other matters, such as bank branches or mergers and acquisitions. Other measures include the provision of basic services, as in Germany, where banks provide a voluntary undertaking to open accounts to all clients under a few basic conditions. This is monitored by consumer protection associations with the right of appeal of clients who have been rejected by a bank. A middle-of-the-ground model, such as Canada, provides government guidelines or directives for the provision of basic account services to the poor, with enforcement through monitoring by a government agency. 41. In light of the low penetration of basic financial services, there are a number of initiatives that could help improve access to financial services in Colombia, from the enforcement of minimum service requirements that are already in the regulation, to the introduction of a basic account available to low income clients, or to encouraging that certain types of payments (such as pension payments and payroll) are done through bank accounts. Such basic accounts have been recently adopted elsewhere in Latin America, for example, in Brazil. 42. Given the extent of financial exclusion in Colombia as revealed in the Bogotá survey and banks' risk-averse behavior following the crisis, it is unlikely that private banks will provide basic account services at their own initiative, and thus, Regulation could be catalyst. The Superintendencia Bancaria could support this by, for example, monitoring the share of basic accounts as a percentage of total accounts in defined lower income categories. 43. All of the initiatives will require the concerted efforts of the Colombian Bankers' Association, as well as a continued dialogue within the banking sector, so that they can effectively lead to the acquisition of new profitable clients and the expansion of the banking sector. 44. Existing anti-money laundering regulation clearly is an impediment for the unbanked to access formal financial services. In particular, multiple account opening requirements could be reduced to one simple 63 identification requirement for clients with low incomes for basic accounts. 45. In addition, an exemption from the financial transaction tax could be contemplated in order to reduce transaction costs for poor people. This could be achieved by exempting transactions on the proposed basic account from taxation. 46. Any form of regulation on financial access requires additional monitoring capacity on the part of the Superintendencia Bancaria or Asobancaria. However, costs of additional supervision could be reduced by tracking a few simple access indicators to be reported by banks on a quarterly basis. Such indicators could include the share of accounts for low income people out of total accounts, the average deposit amount on basic checking and savings accounts etc. Experience in the US has shown that disclosure of such indicators has been an effective tool in enforcing banks' compliance with the CRA. 4.2 SUPPLY SIDE MEASURES 47. This study identified cost barriers as a key reason for the gap between supply and demand of financial services to low income clients in Colombia. Costs are an issue in two aspects: (i) the fees charged by banks for existing financial services are prohibitively high and prevent low income clients from access to formal financial services; and (ii) the transaction costs for banks for processing smaller transactions are relatively high due to the fixed costs of financial infrastructure. 48. These cost barriers point to a lack of appropriate products for low income clients in Bogotá. A focused product design that takes into account the specific needs of low income clients can lead to cost structures that are affordable for the client and bring profits to the banks. Most of the low income individuals do not use credit services, while surveys in other countries show that savings accounts are the most demanded service. Banks should therefore focus on developing adequate savings products for low income clients. 49. One way of making bank services more affordable is to provide alternative products at a lower cost, for example through the basic account concept as proposed above. In case these services cannot be provided on a break-even basis, start-up and self-limiting support ("smart subsidies") could be explored. However, a thorough study of banks' transaction cost structure by product is a fundamental prerequisite in order to determine whether such measures are really needed. This concept could also be extended to incentives for banks to establish locations in low income neighborhoods. 50. The business potential of low income clients should be explored further and demonstrated to banks. Experience of successful microcredit 64 institutions shows that the poor have high loan repayment rates, indicating good payment morale and creditworthiness. In addition, many bankers do not see an inherent increase in risk from the smaller transaction sizes associated with savings and payment services for the poor, while pointing to the benefit of increased diversification. In addition, access to financial services itself provides low income clients with an opportunity of upward mobility, creating a potentially more affluent client base for banks over the medium term52. 51. Innovative solutions which could be explored by banks as an alternative to ATMs and full-fledged branches include mobile units which stop in specific locations at a specific time of the week; cooperation agreements with supermarkets to allow for banks' points of sale inside supermarkets; or access for banks to open branches or points of sales in public buildings, such as schools, municipalities etc. 52. Lottery savings models have also shown that low income people take interest in formal financial services if they are associated with some material incentive such as a lottery to win a consumer good or a share of the collective savings amount. In Colombia, lotteries have been successfully used in housing saving plans. It could be worth for banks to explore further uses of lotteries in other savings services. 53. As indicated in Chapter 3, financial cooperatives appear to serve low income areas in Bogotá better than commercial banks. This analysis should be explored further, including the potential of the credit cooperatives which are supervised by the Superintendencia de la Economia Solidaria. 4.3 DEMAND SIDE MEASURES 54. Demand side measures should be directed at addressing the identified issues of resources, education, location and culture. As in other countries, income levels and income security in Bogotá show a strong correlation with access to formal financial services. The survey shows that 70% of the unbanked earn less than one minimum wage per month and are three times as likely to be unemployed than the banked. In addition, lack of resources is the single most important reason given by the unbanked for not having a bank account. Consequently, growth and income generating policies are important to improve access to formal financial services. 55. As in similar studies in Brazil and Mexico, low levels of education are associated with high degrees of financial exclusion in Bogotá. About 52An example is Banco Popular which has been created as a subsidiary of Brazil's largest bank Banco do Brazil in response to the regulatory obligation to redirect 2% of sight deposits to microcredit services. Banco Popular screens its clients under this program also with a view to identify potential clients for Banco do Brazil over the medium term. 65 60% of the unbanked have an educational level of completed primary education or below, while the majority of the banked population has a completed secondary or university degree. Experience in the US has shown that low education is frequently associated with the lack of financial literacy. This points to the important role of financial education in increasing demand for formal financial services. 56. Focus groups in Bogotá have confirmed that lower income groups have limited knowledge about financial services and generally have preconceived notions that banks `rob or steal'. Consequently, financial education is an important ingredient of any financial access strategy for Colombia. To this end, various channels can be explored: innovative approaches in the US, for example, work with school children in order to reach their unbanked parents. Financial education can also be effective through NGOs or a comprehensive public awareness campaign. 57. The Government can also play a role in enhancing the use of formal financial services. For example, rather than paying government employees by check, the government could pay wages as well as pensions electronically into recipient accounts. This would significantly decrease transaction costs for employees, pensioners and banks, while encouraging recipients to use non-cash financial services. Paying employees electronically is successfully practiced in some US states. 4.4 AREAS FOR FURTHER WORK 58. In the scope of the present paper it was only possible to provide an initial overview of the state of access to financial services in Colombia, and highlight issues on both the supply and demand side. Further work is encouraged in order to explore a number of the presented topics in more detail. 59. On the supply side, a follow-up survey of Colombia's major financial intermediaries in Bogotá and other cities would be beneficial. Such a survey should address issues such as banks' transaction cost structure, account opening requirements, minimum balance requirements, fee structures for checking and savings accounts, payment services, consumer loans as well as microcredit, and interest rates on all of these products. These findings should be benchmarked against international bank practice taking countries' different income levels into account. On this basis it could be established whether banks in Colombia provide financial services at high costs relative to client income and international experience. The survey could also help identify the potential for start-up of self-limiting assistance to financial institutions for encouraging the provision of basic services. 60. The present paper limits itself to examining access issues among formal financial institutions supervised by the Superintendencia Bancaria. In 66 addition, the role of non-bank intermediaries such as microfinance institutions and credit cooperatives in the provision of basic financial services deserves a closer analysis. 61. Follow-up work can also be conducted on the state of microcredit in Colombia, as existing work and World Bank operations have mainly focused on housing microfinance. 62. Finally, the role of financial infrastructure can be further explored, including in this context the collection of information on low-income clients through credit information and registry systems. An investigation of this in the context of Colombia and other Latin American countries is already underway, financed by a FIRST initiative under the auspices of the World Bank. This could be expanded to include other aspects of financial infrastructure, such as secured credit transactions, and registries for collateral. Finally, small value payment systems could also be explored. Remittances are important for the Colombian economy. Their costs to the sender and recipients and alternative remittance models are an important topic for further analysis. 63. On the demand side, a second survey would provide the opportunity to obtain additional information on vital areas which were not covered by the first survey, including preferred banking services by the unbanked, additional questions on the use of savings, the role of public versus private banks etc. 64. Lastly, a study which would quantify the benefits of improved financial access for low-income people or, to the same extend, identify the costs of limited financial access could make a significant contribution to the political case for improving access to financial services for the unbanked in Colombia. 67 BIBLIOGRAPHY Bates, Timothy. 2000. "Financing the Development of Urban Minority Communities: Lessons of History." Economic Development Quarterly. 14 (3): 227-241. Baydas, Mayada M., Douglas H. Graham and Liza Valenzuela, (1997) "Commercial Banks in Microfinance: New Actors in the Microfinance World", Microfinance Best Practices Project. Development Alternatives Inc. Bethesda, MD. Barr, Michael 2003. "Banking the Poor", Working Paper, Brookings Institution, Washington D.C. Barr, Michael. 2001. "Access to Financial Services in the 21st Century: Five Opportunities for the Bush Administration and the 107th Congress", The Brookings Institution, Center on Urban and Metropolitan Policy and Harvard University Joint Center for Housing Studies, Washington, D.C. Bennett, Jr., Thomas E. 1987. "What do Bankers Want?", Inc. p. 149 Bernstein, Jared. 2003. "Savings Incentives for the Poor: Why the Scale doesn't Match the Promise." The American Prospect. Vol. 14 No. 5. May. Buckley, Graeme. 1997. "Microfinance in Africa: Is it either the problem or the solution?" World Development Vol.25, No. 7pp.1081-93, CGAP, Focus Note No. 15. 2000. "Raising the Curtain on the `Microfinancial Services Era" May Campion, Anita, and Victoria White. 2001. "NGO Transformation," DAI, September. Campos, S., and Wenner, M. 1998. "Lessons in Microfinance Downscaling: The case of Banco de la Empresa, S.A. Sustainable Development Department." IADB. Washington, D.C. Carr, James H. and Scheutz, Jenny. 2001."Financial Services in Distressed Communities: Framing the Issue, Finding Solutions", Fannie Mae Foundation, Washington, D.C. Carr, James H. and Lopa Kolluri. 2001. "Predatory Lending: An Overview", Fannie Mae Foundation, Washington, D.C. 68 Caskey, John P. Forthcoming. "Reaching Out to the Unbanked" (chapter in forthcoming book, Inclusion in Asset Building: Research and Policy, edited by Lisa Morris and Michael Sherraden Caskey, John P. 2001. "Bringing Unbanked Households into the Banking System" Capital Xchange Journal. The Brookings Institution, Center on Urban and Metropolitan Policy and Harvard University Joint Center for Housing Studies, Washington, D.C. Caskey, John P. 1994. Fringe Banking: Check-Cashing Outlets, Pawnshops, and the Poor, Russell Sage Foundation, New York: Caskey, John P. and Andrew Peterson. 1994. "Who Has a Bank Account and Who Doesn't: 1977 and 1989" Eastern Economic Journal, Winter. Caskey, John P. 1994. "Bank Representation in Low-Income and Minority Urban Communities" Urban Affairs Quarterly, June. _____________ 1997. Lower Income American, Higher Cost Financial Services (Madison, WI: Filene Research Institute) _____________ 1997. "Beyond Cash and Carry: Financial Savings, Financial Services, and Low-Income Households in Two Communities" a report for the Consumer Federation of America, December, _____________ 1989 "The IMF and concerted lending in Latin American debt restructurings: a formal analysis" Journal of International Money and Finance (1989) 8, 105-120, London, March. Carstens, Catherine Mansell.1995. Las Finanzas Populares en México, Editorial Milenio, S.A., México D.F. Christen, Robert Peck. 2000. Commercialization and Mission Drift ­ The Transformation of Microfinance in Latin America, unpublished paper for the World Bank Consultative Group to Assist the Poorest (CGAP) Christen, Robert and Deborah Drake. 2001. "Commercialization in Microfinance," DAI, October. Churchill, Craig F.1999. "Client ­Focused Lending: The Art of Individual Lending," Calmeadow. De Soto, Hernando, 2001. "Access to capital and financial services is the key to economic growth both in advanced economies and in the developing world. De Sousa-Shields, Marc and Enterprising Solutions. 2002. "Bringing Financial Services to the "Bankable" Poor in Mexico and Colombia Mexico: Supply Side Analysis" Study for the World Bank, Washington D.C. 69 Doyle, Joseph J. et al. 1998. "How Effective is Lifeline Banking in Assisting the `Unbanked' ?" Current Issues in Economics and Finance, Federal Reserve Bank of New York. Vol. 4, No. 6. Dhumale, Raúl, Amela Sapcanin, and William Tucker, "Comercial Banking and Microfinance in Egypt: National bank for Development, Case Study," UNDP, Regional Bureau for Arab States and World Bank, Middle East and North Africa Region. Durham, Constance R. 2001. "The Role of Banks and Nonbanks in Serving Low-and Moderate-Income Communities", presented at Federal Reserve System Conference, "Changing Financial Markets and Community Development", Washington D.C., April 5-6, Green, Mark and Leichter, Franz. 1996. Ranking Banking: The Consumer Bank Scorecard New York: Office of the Public Advocate for the City of New York Green, Mark and Leichter, Franz. 1998. Ranking Banking: The Consumer Bank Scorecard New York: OPffice of the PUblic Advocate for the City of New York Haag, Susan White. 2000. Community Reinvestment and Cities: A Literature Review of CRA's Impact and Future. The Brookings Institution, Center on Urban and Metropolitan Policy. Washington, D.C. Hannig, Alfred Dr. 1999. "Mobilizing Microsavings: The Millennium Challenge in Microfinance", presented at the Sixth Consultative Group Meeting of CGAP, Abidjan, 21-24 June Harvey, Tom, "The Bank Revolution. Positioning your Bank in the new financial services marketplace". Irwin Professional Publishing. Holyoke, Thomas T. 2001. "Community Organization and Community Reinvestment Act Lending in Washington, D.C.", presented at Federal Reserve System Conference, "Changing Financial Markets and Community Development", Washington D.C., April 5-6, Holden, Paul and Vassili Prokopenko. 2001. "Financial Development and Poverty Alleviation: Issues and Policy Implications for Developing and Transition Countries" IMF Working Paper, IMF, Washington D.C. Jenkins, Hatice. 2000. Commercial Bank Behaviour in Micro and Small Enterprise Finance. Development Discussion paper No. 741. Cambridge: Harvard Institute for International Development. Kennickell, Arther B. and Starr-McCleur, Martha. 1997. "Family Finances in the US: Recent Evidence from the Survey of Consumer Finances" Federal Reserve Bulletin, January 70 Kim, Anne. 2001."Taking the Poor Into Account: What Banks Can Do To Better Serve Low-Income Markets," Progressive Policy Institute Policy Report, August 2. Litan, Robert E., Nicholas P. Retsinas, et al. 2000. The Community Reinvestment Act after Financial Modernization: A Baseline Report. U.S. Department of the Treasury. Washington, D.C.: April. Marulanda, Beatriz. 2004. La Bancarizacion en Colombia, Informe Final. Bogotá, September 2004 Micro Banking Bulletin. 2001. No. 6, Statistical tables, April Minsky, Hyman P; Dimitri B. Papadimitriou et al. 1993. Community Development Banking: A Proposal to Establish a Nationwide System of Community Development Banks. The Jerome Levy Economics Institute of Bard College Policy Brief No. 3, Annandale-on-Hudson, NY. Navas, Sergio and Claudio Gonzalez-Vega. 2000. "Innovative Approaches to Rural Lending: Financiera Calpía in El Salvador," Ohio State University, June. Newsday. 1998. "Dollars and Cents: Trying to Lure More Banks to Needy Areas", June 11 New York State Banking Dept. 1996a. Check Casher Summary Figures, July 16 New York State Banking Dept. 1996b. Basic Banking Accounts/Lifeline Banking Accounts, August 7 New York State Banking Dept. 1998. "Proposed Part 8 of the General Regulations of the Banking Board, Baning Development Districts" Otero, Maria. 1986. "The Solidarity Group Concept: Its Characteristics and Significance for Urban Informational Sector Activities," Monograph Series No. 1, ACCION International, pp. 30-33. Poyo, Jeffrey and Robin Young. 1999. "Commercialization of Microfinance: The Cases of Banco Economico and Fondo Financiero FA$$IL, Bolivia," MBP, December. Radeki, Lawrence and Wenniger John and Orlow, Daniel. 1996. "Bank Branches in Supermarkets" Federal Reserve Bank of New York, Current Issues in Economics and Finance 2 no. 13 (December) Rhyne, Elizabeth. 2001."Microfinance Institutions in Competitive Conditions," Microenterprise Best Practices, Development Alternatives Inc, Bethesda, MD, USA. 71 Pristin, Terry. 1999. "Group Says Banks Don't Push Low-Cost Checking Accounts" New York Times, July 30, Rhine, Sherrie L.W., Toussaint-Comeau, Maude, Hogarth, Jeanne M., Greene, William H. 2001. "The Role of Alternative Financial Service Providers in Serving LMI Neighborhoods", March Rubin, Julia Sass. 2001. "Community Development Venture Capital: A Double- Bottom Line Approach to Poverty Alleviation" presented at Federal Reserve System Conference, "Changing Financial Markets and Community Development", Washington D.C., April 5-6, Schmidt, Reinhard and Claus-Peter Zeitinger, "Micro-Finance Banks: Building New Institutions Instead of Remodeling Existing Ones," Entwicklung und Zusammenarbeit, Vol. 41, No. 11. Squires, Gregory D. ed. 1992. From Redlining to Reinvestment: Community Responses to Urban Disinvestment. Temple University Press. Philadelphia. USA Stearns, Katherine. 1993. Leverage or loss? Guarantee Funds and Micro enterprises, ACCION monograph series no. 8. Cambridge, MA, Stegman, Michael A.1999. Savings for the Poor: Hidden Benefits of Electronic Banking, Washington, D.C.: Brookings Institution Press, Stegman, Michael A. 1998. "Electronic Benefit's Potential to Help the Poor"- Brookings Institute Policy Brief Series #32, March Theodore, Leslie, with Jacques Trigo Loubiere. 2001. "The Experience of Microfinance Institutions with Regulation and Supervision: Perspectives from Practitioners and a Supervisor," DAI, October Valenzuela, Lisa. 2001. "Getting the Recipe Right: The experiences and challenges of commercial bank downscalers," USAID, October. Williams, Marva. 2002. "Developing Opportunities for Affordable Lifeline Banking for Lower-Income Consumers", Testimony before the U.S. Senate Committee on Banking, Housing and Urban Affairs, May 2. Yaron, Jacob, Benjamin McDonald and Stephanie Charitonenko. 1998. "Promoting Efficient Rural Financial Intermediation," The World Bank Research Observer, Vol. 13, No. 2, August, p. 167. World Bank, Country Assistance Strategy for the United Mexican States, Washington, May, 1999 72 __________ Colombia Country Assistance Strategy Progress Report, Washington, November, 1999 __________ Colombia Rural Finance, Access Issues, Challenges and Opportunities, Washington, November, 2003 __________ Mexico, Rural Financial Markets, Report No 13599-ME, Washington, August, 1995 __________ Project Appraisal Document on a Rural Finance Project for Romania, Washington, March 2001 73