MEASURING BANKING ACCESS IN 54 ECONOMIES 69961 the Poor Banking MEASURING BANKING ACCESS IN 54 ECONOMIES the Poor Banking © 2009 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved 1 2 3 4 5 12 11 10 09 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank Group. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. ISBN: 978-0-8213-7754-3 E-ISBN: 978-0-8213-7755-0 DOI: 10.1596/978-0-8213-7754-3 Cover by Studio Grafik. Library of Congress Cataloging-in-Publication Data Banking the poor. p. cm. Includes bibliographical references. ISBN 978-0-8213-7754-3 -- ISBN 978-0-8213-7755-0 (electronic) 1. Financial Services industry - developing countries. 2. Banks and Banking - Developing countries and 3. Microfinance - Developing countries. I. World Bank. HG1660.A4B36 2008 332.1'75086942096--dc22 2008037287 Preface How many people use formal banking services in poor countries? services by telephone or Internet appeals primarily to those who What services do they value most? How could access to banking already have accounts. services be expanded to include more people? This study explores • Basic banking, as practiced by commercial banks, increases access these questions and more through responses to surveys conducted in to banking services, but, in areas where commercial banks have 54 developing countries, mostly in Africa. not embraced basic banking, laws and regulations to promote it have done little to increase access. Banking the Poor presents new data collected from two sources: • Good credit information, effective protection of the legal rights central banks and leading commercial banks in each surveyed of creditors, and flexibility in collateral helps make more credit country. It explores associations between countries’ banking policies available to entrepreneurs. and practices and their levels of financial access, measured in terms • More information about loans, greater transparency in loan of the numbers of bank accounts per thousand adults. It builds on processes, and the existence of consumer protection laws tend to the previous work measuring financial access through information be associated with wider access to banking services. obtained from regulators, banks, and individual respondents to household surveys. Some of our results may be surprising. The absence of any association between access and tax-advantaged or matched savings Findings schemes is one example. The lack of an association may be explained by the fact that tax advantaged schemes are of little use to poorer The extent to which people are banked depends primarily on how clients, despite their wide prevalence, while matched savings wealthy they are. Even in the poorest countries, rich urban schemes, which could help attract poor clients, are rarely used. As customers get access to good banking. Although there are a range of regards better retail payments systems and expanded access, clients financial services used by the poorest, these are usually provided who are beginning to bank are not likely to explore sophisticated outside the formal banking system. Banks are used by those above issues such as the extent to which networks are interlinked. this threshold, especially salaried employees, who have the steady Eventually, however, such benefits may be reflected indirectly in income that makes being banked useful and that encourages banks reduced costs and greater efficiency in the provision of banking to seek them as clients. Expanding credit for enterprises leads to services, for example by reducing the distance to a usable ATM. the creation of a salaried class that wants to bank: this is the primary way to increase bank access. Banking via mobile technologies appears to be an additional service for existing clients. But a measurement issue makes it Our other main findings are as follows: difficult to give a definitive answer to this question. Some large • Complexity in opening accounts, applying for loans, or obtaining mobile banking services offered by telecommunications companies other financial services is associated with reduced access. (such as Vodafone’s M-Pesa service in Kenya) use a single clearing • Recurring costs, as well as multiple fees and charges, reduce access bank account for large numbers of clients. Users of such systems to banking. are not captured by the measure of access used in this study • Convenience features, such as after-hours services, may enhance (numbers of bank accounts). More generally, enabling clients to the usefulness of banking for those already banked, but they do use cell phone services for a wider spectrum of banking functions not bring new entrants across the threshold. Access to account could imply the need for linkages with retailers or “correspondents.��? PREFACE 3 What is next these bankers would not admit to discrimination. Yet a recent study—Doing Business: Women in Africa—suggests that Four areas relevant to banking the poor are not studied here. These discrimination is widespread. A mystery-shopper methodology will will be analyzed in future reports. be needed to study this topic. The first is how location affects the likelihood that an individual Finally, our study does not investigate the potential for micro- will use banking services. This study does not deal with rural financiers to grow into commercial banks that offer a wide range banking. In rural areas that have bank branches, bank usage seems of banking services. There are some examples of such growth—for as high as in urban areas. In areas where no bank branches or example, the Grameen Bank in Bangladesh and BancoSol in correspondent outlets operate, we would expect usage to be Bolivia. Their success presumably has brought greater numbers of substantially lower. people into the formal banking system. A second aspect is the impact of financial education. In Future work can also shed additional light on several questions that Afghanistan, the Central African Republic, Côte d’Ivoire, and were covered in this study. The basic measure of access used here, Gabon, bankers say that illiteracy, the lack of general education, accounts per thousand adults, can be refined. At present, there are and the lack of financial education are major barriers to opening variations across countries in regulators’ statistics for such accounts. a bank account. These assertions remain to be tested: bankers may In some countries, these may include accounts still open but little be putting the blame in the wrong place. Among developing used, while in others inactive accounts may be regularly or countries, programs of evaluation of financial literacy are being occasionally netted out. In some countries, a single client may have launched on a large scale in countries such as India and Indonesia. multiple accounts; in other countries, and in some banks, these multiple accounts are considered as subaccounts and not reported. Studies in rich countries suggest that households with limited education and financial literacy are less likely to seek out and use Likewise, one might go beyond the top five commercial banks to banking services. In the United States, participation in financial collect data. This is particularly so because, in some cases, banks education programs is a compulsory adjunct to participation in with relatively small asset bases may make efforts to cater to larger savings schemes and has been shown to be correlated with successful numbers of clients. Also, by focusing on the top five commercial outcomes from such savings programs. Other researchers have banks we may have overlooked banking institutions with large shown that school-based financial education programs in the client bases, such as savings banks, which are not listed in all United States increase asset accumulation in later life. Work-based countries as commercial banks. While bank clients make up the programs also suggest increased participation in voluntary savings largest part of those using financial services in most countries, schemes, especially for low-paid workers. incorporating other formal financial institutions would yield a more comprehensive picture of the population that enjoys access to Third, our analysis does not investigate bank usage by specific modern financial services. segments of the population—women, for example. When asked whether women require signatures of a male family member to open an account, only five banks in four countries (Namibia, Rwanda, Swaziland and Uganda) said yes. And when asked if a man’s signature was necessary for domestic money transfers, only three (in Lesotho, Sudan, and Uganda) said yes. However, even 4 BANKING THE POOR Contents Preface 3 1. Overview 7 2. Banking the poor 15 3. Starting to bank 23 4. Adding customer services 31 5. Building branchless banking 39 6. Promoting access to banking 47 7. Extending credit to entrepreneurs 53 8. Increasing disclosure 61 9. Data sources and methodology 65 References 71 Glossary of variables, indexes, and other terms 77 Indexes of financial access: mean values by region and country 80 Acknowledgments 104 CONTENTS 5 1. Overview In Chad, one of the world’s poorest countries, there are open an account in Swaziland is 3.6, compared with 2.0 barely nine bank accounts per thousand adults. In in Cape Verde. Rwanda, Liberia, and Madagascar, where per capita incomes are higher, the figure is about 35 accounts per The access to banking services in developing countries is thousand adults. South Africa is one of the richest even lower than suggested by the figures on density of countries in Africa, with a per capita income of more bank accounts, because some people have multiple bank than $5,300, about 10 times that of Chad. There, one accounts. Regulators count checking, savings, and time- finds about 550 bank accounts for every thousand adults. deposit accounts separately—yet each banked person Mauritius, another relatively rich country in Africa, has may have all three. Also, the numbers of accounts do not the subcontinent’s highest density of accounts—2,010 separate business accounts from individual accounts. per thousand adults (figure 1.1).1 Taking this into consideration, the number of banked adults in a country like Chad may be as low as three in Among countries with similar levels of per capita income, 1,000. one finds different levels of access to banking services. Mali and Bangladesh, for example, have a similar per Want to bank? Get a formal job capita income of around $425, but the number of accounts per thousand adults is five times as high in Among the several factors that affect the likelihood that Bangladesh as in Mali. Cape Verde and Swaziland also an individual will have a bank account, the most have similar levels of income ($2,200–$2,350 per capita). powerful is income (figure 1.2).2 The poorest people in But Cape Verde has almost twice as many bank accounts many countries do not use the banking system because per thousand adults as Swaziland. Cost and convenience they do not have enough money to make it worthwhile. may explain a good part of the difference. Monthly fees And banks may not find it worthwhile to provide them for holding an account in Mali are three times as high as with services. FinMark Trust of Africa notes that “the in Bangladesh, and the number of documents needed to greatest perceived barrier to access is the absence of FIGURE 1.1 The likelihood of being banked varies widely by country Number of bank accounts per thousand adults 2500 2000 1500 Central African Republic El Salvador South Africa Congo, Dem. Rep. of Philippines Honduras Swaziland Bangladesh Indonesia Mexico 1000 India Mozambique Burkina Faso The Gambia Côte d'Ivoire Nicaragua Sierra Leone Namibia Cape Verde Lesotho Pakistan Madagascar Afghanistan Guatemala Cameroon Singapore Tanzania Botswana Mauritius Angola Cambodia Nigeria Sri Lanka Ghana Uganda Lao PDR Vietnam Malaysia Thailand Gabon Ethiopia Senegal Kenya Zambia Malawi Sudan Nepal 500 Rwanda Burundi Benin Liberia Mali Niger Chad 0 Source: Getting Finance database. OVERVIEW 7 FIGURE 1.2 More bank accounts in rich countries and Duflo 2006). It is also true that few of the poor hold such jobs. In urban Indonesia, for example, only 38 percent Malaysia of the poor, defined as those whose daily per capita spending Singapore Accounts per thousand adults (log) Sri Lanka is less than $1 a day, are salaried.4 By contrast, 77 percent of India Bangladesh the lower-middle class, those with daily per capita spending Gabon of between $2 and $4, have a salaried job. For those in the Vietnam upper-middle class (with daily per capita spending of Cambodia between $6 and $10), that proportion jumps to nearly 90 Central African Republic percent. In poor countries, most people are employed in the Chad informal sector and lack a steady income. In Mozambique, Coef.=0.62, t-stat.=8.51 GDP per capita (log) for instance, more than 80 percent of the population works Source: Getting Finance database. in the informal sector. sufficient income��? (FinScope/Bankable Frontiers 2007).3 A People who have a formal job are more likely to need and use recent World Bank study, Finance for All?, reaches a similar formal banking services. In India, for example, 91 percent of conclusion (Demirguc-Kunt, Beck, and Honohan 2008). the households in which the main earner holds a salaried job are banked, in contrast to 38 percent of the households in Of the reasons cited for being unbanked in a survey done which the chief earner is a daily wage laborer. Those with in South Africa, (FinScope/Bankable Frontiers 2007) lack formal jobs that pay a regular salary not only have more of income (or lack of a job) was cited by 78 percent of income but also expect stable income in the future (figure respondents, compared with only 13 percent who cited 1.3). They are likely to find banking services useful, just as documentation needs, fees, or distance. Another 9 percent banks are likely to find them to be desirable customers. of respondents said that they were unbanked by choice, did not need an account, or did not trust banks. To create more formal jobs, businesses need to grow. But credit constraints often limit that growth. Easing access to In most developing countries, the middle class consists credit, by contrast, can boost revenue and employment largely of people who hold formal, salaried jobs (Banerjee growth. In a sample of Eastern European firms surveyed FIGURE 1.3 Salaried workers are more confident about future income Degree of confidence in stability of household income, by type of employment Regular salary 75.2 11 4.6 9.2 Self-employed 64.9 15.5 6.9 12.7 (nonagriculture) Self-employed 32.1 25.7 20.5 21.7 (agriculture) Daily wage labor 33.3 25.6 20.1 21 0% 20% 40% 60% 80% 100% Confident Less confident Least confident Can't say Note: The survey, conducted in 2006–07, focused on the saving and insurance habits of Indian households using a nationally representative sample of 63,000 households. Source: Shukla 2007. 8 BANKING THE POOR in 2002 and 2005, the firms that gained access to bank banking to more people. Making it easier to open an credit for investment and working capital registered 9 account is a good example. In many countries opening percent higher growth in employment, compared with an individual bank account is neither simple nor cheap. firms that had no change in access to bank credit.5 More stringent know-your-customer requirements that Revenue growth was higher by 36 percent for the firms have taken effect in recent years have contributed to this that enjoyed bank credit. Job creation was even bigger for situation, suggesting trade-offs between the need for small firms (those with fewer than 20 employees). Small regulatory precautions and the desirability of expanding firms that gained access to bank credit for new popular access to banking services. investments registered 20 percent more jobs. The typical bank in Côte d’Ivoire or Nicaragua, for No formal jobs, no stable income. No income, little reason example, demands five documents to establish an to use a bank. Some 90 percent of the unbanked in Mexico applicant’s identity, income, and residence. In Gabon, have below-median income. In Colombia, 53 percent of banks require a government-issued identity document to the unbanked earn less than half the minimum wage. In open an account. In Lesotho, identification cards do not Brazil, only 15 percent of persons in the bottom quintile of exist, which makes it difficult for poor people lacking other income have a bank account, compared with 64 percent in forms of identification to enter the banking system. Proof the top quintile. Moving from the bottom to the second of address is a standard requirement to open a bank account quintile increases the probability of having a bank account in many countries. Yet, many poor people live in informal by 11 percentage points. Even in the United States, 79 dwellings with no postal address. Some banks in Lesotho percent of the unbanked have below median income. In that target lower income groups reported a sharp drop in the United Kingdom, seven in ten of the unbanked are new account openings following the implementation of unemployed.6 Increasing income, through creating formal new regulations that called for three documents, including jobs, is the main way to increase bank access. proof of income, to open an account. In addition to relatively low incomes, the unbanked have Some banks charge fees to open an account. In Burundi, for another common characteristic: less education. Does more example, these fees are as much as 3 percent of annual per education increase the likelihood of using banking services? capita income. Many banks also impose other types of fees, In Brazil, education beyond the primary level doubles the such as charges for checkbooks and ATM cards. Accounts probability of having a bank account. In Mexico, banked are not always opened on the spot. It can take up to 3 days households are better educated than the unbanked in Gabon, Mexico and South Africa, for example, to open (Djankov and others 2008). However, education correlates an account. Across countries, the greater the bureaucratic very strongly with income.7 Controlling for income, there hurdles faced in opening an account, the lower the number is no evidence that education alone increases an individual’s of bank accounts (figure 1.4). inclination to use a bank. Businesses are also hurt by complexity in banking. To increase access to Entrepreneurs want to be able to access credit quickly and banking, reduce bureaucracy easily without complicated application procedures. Where loan application procedures are difficult, bank credit to the The surest way to increase popular use of formal banking private sector tends to be lower (figure 1.5a). The more services is to raise incomes. Short of that ambitious goal, complex the procedure, the less likely it is that entrepreneurs some small, practical steps can bring the benefits of will access bank credit for working capital (figure 1.5b). OVERVIEW 9 FIGURE 1.4 More documents, fewer accounts Some costs hinder Accounts per thousand adults and number of documents access to banking needed to open an account Malaysia Ongoing maintenance fees also reduce the number of Accounts per thousand adults Mauritius bank accounts. The monthly maintenance fee for a bank Sri Lanka Cape Verde account costs the typical formal worker in Rwanda 3 Thailand India The Gambia percent of her monthly wage, whereas a Singaporean Indonesia Nicaragua worker pays only 0.05 percent of her monthly wage for Angola Gabon Zambia the same service. More than half of all banks report that Namibia they charge a monthly account maintenance fee. Banks in Coef.=-153.5, t-stat.=-3.3 Number of documents required Africa are most likely to charge such fees; those in South Source: Getting Finance database. Asia, least likely. Moreover, the fees levied in Africa tend to be higher than elsewhere, even in much wealthier East Small firms are particularly hurt. Such firms often lack Asia. The average monthly fee for account maintenance the expertise to produce detailed business plans and is close to $4 in Africa, whereas it is only about 50 cents financial statements. In Botswana, it is common practice in South Asia. for new and small enterprises to use professional consultants to produce the business plans required as part Banks levy many other charges, such as fees for a of the credit application. The fee for the consultant’s checkbook or an ATM card. More than 20 percent of services depends on the loan amount. For small loans, it banks report other types of account-opening charges. may be a flat fee of 10,000 pulas, but for large loans Banks in the poorest countries of our sample—among running into several million pulas, the fee may be 1 them Burundi, Côte d’Ivoire, the Democratic Republic percent of the requested loan amount. of the Congo, and Mali—are more likely to impose a variety of such charges than those in richer countries. Countries where banks follow this practice tend to have lower rates of bank access, even after controlling for B FIGURE 1.5 Access to bank credit and complexity of loan applications a. Private credit and complexity of business-loan procedures b. Prevalence of working-capital loans among small firms and complexity of business-loan procedures Malaysia Thailand firms with working-capital loans Ratio of private credit to GDP Percentage of small and micro Vietnam Burundi Burkina Faso South Africa Namibia Ethiopia India Rwanda Thailand Niger Ethiopia Mauritius Kenya Pakistan Vietnam Nicaragua Cameroon Niger Angola Botswana Zambia Angola Botswana Coef.=-25, t-stat.=-1.72 Coef.=-31, t-stat.=-2.11 Complexity of business-loan procedures Complexity of business-loan procedures Note: In panel (a), a regression of private credit to GDP on the index of complexity of business loan procedures, controlling for income and legal rights, shows that a 0.1 increase in the complexity index is associated with a 2.5 percentage point drop in the ratio of private credit to GDP (t-stat: –1.72). In panel (b), a regression of the proportion of firms using bank credit for working capital on the index of complexity of business loan procedures (controlling for income and legal rights) shows that for small firms (those with fewer than 50 employees), a 0.1 increase in the index is associated with a 3.1 percent drop in small firms accessing credit (t-stat: –2.11). Source: Getting Finance database. 10 BANKING THE POOR income. Potential bank clients who may shrug off modest In addition, regulators in 12 countries, almost a quarter account-opening fees are likely to dislike paying a variety of the sample, report some form of regulation requiring of fragmented charges for other items, as this may make basic or simplified accounts for the poor. The scope of banking appear more complicated. A single, transparent such regulations varies widely, however. In some fee is best. countries, the law gives poor people the right to a basic bank account, but no policy exists to inform them of that Other costs that reduce bank usage are charges for right or encourage them to exercise it. The gap between remittances. The higher the fee charged for receiving a law and practice may explain why having regulations for foreign draft, the lower the number of accounts per basic banking are not linked to more accounts. Where thousand adults (figure 1.6). The same applies to the fee banks actually offer basic banking services, there is a charged for sending a foreign draft. positive association between principle and practice.8 Among factors associated with greater numbers of bank What doesn’t work? accounts, the availability of “basic banking��? services stands out. Basic banking is the provision of a package of It might be thought that the presence of a well-developed free or low-cost services to clients, usually with some retail payment system—with many different payment restrictions on the menu of services and on the amounts instruments, extensive interoperability of retail payment held in the accounts. For example basic accounts may be networks (such as ATM networks), and a greater choice entitled to free transactions up to some limit, provided of channels for retail payments—would be associated they are carried out using an ATM card rather than with more bank accounts, but we have not found that to through tellers. Alternatively, fees and charges for some be the case. Similar results are obtained for alternative transactions may be waived. payments channels and access (checks, payment cards on bank premises, payment cards at ATMs, direct credits on A majority of banks in all developing regions offer one bank premises, direct credits over the phone, and direct or more accounts with characteristics of a basic account. debits) across three different channels (person to person, person to business, and person to government). This is understandable. A prospective new client of modest FIGURE 1.6 Higher fees, fewer accounts means is unlikely to weigh sophisticated aspects of the Density of bank accounts and the banking infrastructure when making the decision of cost of receiving a banker’s draft of $250 whether or not to use a bank.9 Malaysia Accounts per thousand adults Mauritius Sri Lanka Sophistication aside, even the ordinary convenience Ethiopia India Thailand features of a standard bank account appear not to attract Tanzania Niger Ghana new bank customers, and probably for similar reasons. Burundi Such features include after-hours access to withdrawal Namibia Zambia Benin services, overdraft provisions, overdraft notification, and Coef.=-269.56, t-stat.=-2.13 exemption from the need to present a passbook when Cost of receiving a banker’s draft making withdrawals and deposits. In fact, an index of Note: A movement from the 25th to 75th percentile on the index cost of receiving such convenience features is negatively associated with a bank draft is associated with 269 fewer accounts per thousand adults, controlling for GDP per capita. The index is constructed by normalizing the cost ($) so that the lowest-cost country is assigned a 0 and the highest-cost the number of accounts per capita. This does not mean country a 1, and the rest are scaled in between. Source: Getting Finance database. that the availability of such features reduces access. OVERVIEW 11 Rather, banks appear more likely to provide such services generous free services to their rich clients. More marginal in countries where banking is the preserve of the elite clients may receive very limited free services, or none at all. few—perhaps to compete for the best clients. Poor countries with low rates of popular access to banking A surprising finding is that potential inconveniences such as services, such as Senegal, Kenya, and Chad, have some minimum balance requirements for opening or maintaining of the highest scores on the index of convenience features. an account do not appear to have a negative impact on the number of accounts.10 In 37 countries, all banks require a Once this is understood, it should not come as a surprise minimum deposit to open an account. The amount required that unlimited supplies of free services, such as balance can be as much as three times the per capita income of inquiries and withdrawals, do not increase the ranks of poorer countries (3.3 times per capita GDP in Burundi; 2.3 the banked. Such services are widespread. For example, times in Madagascar). Although the ratio of the minimum around 90 percent of banks in the sample offer unlimited deposit to per capita GDP tends to decline in richer free balance inquiries. Ethiopia, Mexico, and countries (to less than 1 percent in Malaysia, Thailand, and Mozambique are the only countries in which three or South Africa), the requirement for a minimum deposit is more of the sampled banks do not do so. Three quarters highly prevalent (figure 1.7). But neither the existence of of banks also offer unlimited free withdrawals on their such a requirement, nor its amount, is significantly standard accounts. But we found wide variation in ATM associated with access. When examining results separately and checking account services. In 10 countries, across all for savings accounts, as opposed to checking accounts, there regions and income levels, none of the banks offers a free is still no association. One explanation may be that opening ATM card, while in four, all banks do so. All banks in balances are less of a deterrent to access than are repeated Burkina Faso, Burundi, the Central African Republic, costs, such as the penalties for falling below a minimum Gabon, Madagascar, Nepal, Uganda, and Senegal provide balance. free checking account services, while in 16 of the 54 sampled countries none of the banks does so. Also, very Some special savings few banks (only 4 percent of all banks surveyed) offer a schemes hold promise free credit card. One reason for the apparent lack of association between free bank services and the density of Our analysis looks at regulations pertaining to matched accounts may be that banks offer unlimited or very savings schemes, tax-advantaged savings schemes, and so- FIGURE 1.7 Minimum deposit requirements are common—but more burdensome—in poor countries a. Standard account b. Interest-bearing savings account Fraction of banks with minimum deposit requirement Fraction of banks with minimum deposit requirement Median minimum deposit (% GDP per capita) Median minimum deposit (% GDP per capita) 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 Very Low Low Medium High Very Low Low Medium High Country income quartile (GDP per capita) Country income quartile (GDP per capita) Source: Getting Finance database. 12 BANKING THE POOR called doorstep collection schemes. At the level of sample. In Bangladesh, Cape Verde, the Central African commercial banks, we investigate an additional savings Republic, Gabon, India, Senegal, and Thailand, all product: periodic (or commitment) savings. sample banks offer commitment savings products; in another 10 countries, at least three out of five banks offer We find no correlation between account density and the such products. Such savings schemes are well adapted to availability of special savings vehicles offered by the needs of poor clients. governments or banks. A likely explanation is that of these various plans, those that are most likely to appeal to New technology has poor people, and thus to draw them into the banking yet to make a difference system, are little used. Only one country in our sample— Singapore—has a regulatory scheme for matched savings. What about technology? Has it helped, as many thought By contrast, 22 countries in our sample offer tax it would, to expand people’s access to banking services? incentives for savings. Matched savings plans require The answer is, not yet. Banking services can be provided governments to make matching contributions to private by ATMs, via the Internet, or via cell phones. All of these savings, in some agreed proportion. Such plans have been technologies have the potential to make banking services demonstrated to be effective vehicles for increasing less costly by reducing unit and marginal costs and savings by poor people in developed countries, among eliminating expenses associated with branches. There is them the United States and the United Kingdom. little evidence that this potential has been realized as yet. (Sherraden 2006). However, they require subsidies. Tax- advantaged savings schemes also represent a cost to the We found that banks are much more likely to offer services government, in terms of tax revenues forgone, but many over the Internet than over the phone (figure 1.8a). We governments prefer to forgo revenue through tax waivers also found only a few differences in the levels of service rather than spend collected revenues on subsidies. Tax- provided by cell phones compared with landline phones. advantaged schemes are useful for persons who do not Finally, we found that the range of remote assistance expect to need to access their savings for long periods remains narrow in most places. Balance inquiry is the (until their retirement, for example), and such people most frequently offered remote aid. In East Asia and tend to be better off than those who would be targeted by Latin America, account holders can check their balance matched-savings plans. Because they are better off, they over the Internet at 83 and 88 percent of all banks and by are also more likely already to have a bank account. cell phone at 60 percent and 35 percent of banks. All banks in Indonesia, Thailand, El Salvador, and Honduras Doorstep collection schemes, also known as commitment offer their customers balance inquiries via the Internet, schemes, have been used successfully by microfinance but only around half offer the same help via cell phone institutions. They are now being introduced by banks in (figure 1.8b). Next in popularity are “ministatements��? some countries to collect funds from small savers. In (information on the last few transactions) and checking Ghana, India, Indonesia, and the Philippines, regulators the status of payments. Remote services for money assert that doorstep collection schemes are in operation transfers and bill payments are rare. for low-income clients. Two countries—Indonesia and the Philippines—have adopted guidelines to manage the We constructed a mobile technology index that works as risks of doorstep collection schemes operated by nonbank follows: the more banks in a country that offer banking agents. Of the varieties of these schemes, those that services through cell phones, the higher the index. We provide for periodic savings are the most common in our found no correlation between account density and the OVERVIEW 13 FIGURE 1.8 Remote banking services have not kept pace with spread of cell phones a. Internet users as percentage of all telecommunications users, b. Share of banks offering various services by phone and Internet and cell phone subscribers as percentage of all telephone subscribers 100 80 75 60 (%) (%) 50 40 25 20 N/A 0 0 2004 2005 2004 2005 2004 2005 2004 2005 Balance Mini Payment Money Bill Selling cell inquiry statement status transfer payment phone Africa South Asia East Asia Latin America minutes Internet Cell phone Phone Cell phone Internet Source: Getting Finance database; World Development Indicators, World Bank, 2008a. index. The likely explanation is that mobile banking is a 1 The numbers of bank accounts are based on information provided by national regulatory authorities and central banks. convenience feature designed to retain existing customers 2 For the sample as a whole, there is a significant correlation between income per capita and accounts per 1,000 adults (R2=0.47; t=4.26). rather than attract new ones—only people who are 3 FinMark Trust has undertaken household surveys of access to finance in several African countries, using an approach referred to as Fin- already banked use it. For example, in Botswana, First Scope. FinScope finds that in Uganda, Kenya, and Namibia, 18, 19, and 51 people are banked per 1,000 adults, compared to 16, 16, and National Bank launched its mobile banking service in 32 in this study. For two countries, the estimates here are somewhat higher: 160 adults per 1,000 in Tanzania, and 550 in South Africa, November 2006. By May 2008, the service had 23,000 compared with 90 and 500 in FinScope. 4 subscribers, but it is available only to existing FNB account Household surveys typically lump daily and casual workers with salaried employees in one category—wage workers. But casual work- holders. ers have little job security, work fluctuating hours, and move often from job to job. They are usually paid by the hour or at the end of the day. Salaried employees are in secure jobs. These jobs pay on a weekly or a monthly basis. It appears that owning a cell phone in itself is not enough 5 The data are available at www.enterprisesurveys.org. 6 See surveys in Solo, Caskey, and Ruiz Duran (2006), Solo and Manroth to induce the unbanked to start banking. While mobile (2006), and Kempson and Whyley (2000). 7 The literacy rate does not affect the likelihood of being banked, con- banking via cell phones does not yet offer full banking trolling for income. 8 Some countries such as South Africa, where basic banking is popular solutions, it may grow in the future, especially for money and widespread, do not have a regulation for basic banking, as this has been adopted as a result of a voluntary agreement of banks. transfers. 9 However, there is a significant positive association between network interoperability and private credit. While such features of the retail payments infrastructure may facilitate overall financial flows, they are unlikely to entice the marginal consumer to open a bank account. 10 This is in contrast to earlier findings of others. See, for example, Beck, Demirguc-Kunt, and Martinez Peria (2007a). 14 BANKING THE POOR 2. Banking the poor Jyothi works as a wandering deposit collector in the slums of Few poor people the southeastern town of Vijayawada, India. Her clients are choose to save at banks slum dwellers, mostly women. Their savings are irregular, small, and often geared toward accumulating a lump sum for Seventy six percent of all bank accounts in the sample their family short-term needs: children’s textbooks, for countries are opened for savings. In half of the sample example. The largest expenditure to save for is a daughter’s countries, more than 80 percent of all deposit accounts are marriage.11 savings accounts (figure 2.1). In a fifth, more than 90 percent of all accounts are savings accounts. In only three countries— Jyothi’s clients pay to save. She gives each a simple card, Madagascar, Cape Verde, and Rwanda—do savings accounts divided into 220 boxes, and each client agrees to save a certain make up less than half of all deposit accounts. amount per box. At the end of 220 days her client takes back her savings. However, she does not get it all back, since Jyothi’s Time deposits account for up to 7 percent of total accounts. fees are equivalent to 20 out of the 220 payments, or 9 In just four countries—Bangladesh, India, Thailand, and percent. By adding back the interest forgone by the client Malaysia—they account for between 10 and 15 percent of (which Jyothi is able to earn by investing the money), Jyothi’s the total number of accounts. But in terms of value, time annual percentage rate of interest is about 30 percent. In other deposits account for more than a quarter of total accounts words, the client is “earning��? interest at minus 30 percent per (figure 2.2). In seven countries, they account for more than year.12 half the total value of all deposit accounts. Since poor people are unable to put aside money for long periods, it is the rich Yet Jyothi’s clients are happy with this arrangement. Jyothi is who make extensive use of this type of banking. flexible about the days when no contributions are possible and also helps arrange pickups and dropoffs at convenient Poor people prefer to save outside the home to avoid the times and locations. temptation to spend and diversion of funds by other family members for less urgent needs. Many women report, for FIGURE 2.1 In terms of numbers, most bank accounts are for saving Savings and time deposit accounts as shares of all bank accounts, by country 100 % Savings Accounts % Time Deposit Accounts 80 60 40 20 0 Ghana India Kenya Malawi Mexico Uganda Rwanda Liberia Malaysia The Gambia Sri Lanka Pakistan Namibia Nigeria Indonesia Nicaragua Guatemala Thailand Swaziland Honduras Bangladesh Cape Verde South Africa Tanzania Mauritius Afghanistan Madagascar Sierra Leone El Salvador Philippines Source: Getting Finance database. BANKING THE POOR 15 FIGURE 2.2 Measured by value, time deposits and checking accounts represent a large share of all accounts Value of time deposits, savings accounts, and checking accounts as shares of total account value in sample countries 100 80 Time Deposit Accounts Savings Accounts 60 (%) Checking Accounts 40 20 0 Tanzania Thailand Pakistan Bangladesh Cape Verde Uganda India Malawi Rwanda Namibia Swaziland Nigeria Afghanistan Guatemala Sri Lanka South Africa Malaysia Kenya Liberia Indonesia Ghana Nicaragua Honduras Mauritius The Gambia Sierra Leone Philippines El Salvador Mexico Madagascar Source: Getting Finance database. example, coming home to find that their husbands have In South Africa, a large range of stokvels meet different spent their hard-earned savings on alcohol (Rutherford needs, ranging from Christmas funds (saving for a 2000). Income earners also face great pressure to provide generous December food shopping spree) to burial funds. financial support to their extended families. Members contribute a fixed amount of money to a common pool, and money is drawn either in rotation or But only 14 percent of households living on less than $2 when a particular need or occasion arises. dollars a day have a formal savings account (Banerjee and Duflo 2006). For the most part, their incomes are too low Many microfinance institutions use the same group- and too unstable to make an account worthwhile. based commitment. They loan to members only, thereby Sometimes complicated bank procedures are to blame. preserving group liability and cohesion. Often, the reason is psychological: the poor feel unwelcome as clients. Or the bank is too far: the cost of getting to the Formal banks usually do not meet the needs of the poor, bank can be expensive for those with small incomes. who are not able to put money away for long and cannot Informal savings may be preferred, even if they are less afford to travel far. The time and expense of transport secure.13 explains why doorstep collection has been popular with poor savers in some countries. The poor often save for different needs through different means. They save most in the form of real assets such as Few poor people use banks to livestock, gold, jewelry, or, sometimes, construction channel remittances materials. Next in popularity are rotating savings groups or savings clubs, which are present in a variety of forms in most Oliveira Guzman, a 27-year-old Mexican emigrant in the countries. The rotulas of Brazil, the tandas of Mexico, the United States, sends an average of $300–$500 of his tontines and stokvels of South Africa, the chit funds of India, monthly salary of $1,800 to his parents in the town of and the hui of Taiwan, China, share many features in Chahuayito. Oliveira’s remittance is typical of the common. amounts sent by Mexican workers. His mother, Señora Guzman, explains that Oliveira’s contribution helps cover 16 BANKING THE POOR basic expenses. With spiraling prices of tortillas and other Formal money transfer operators are not cheaper than staples, Señora Guzman finds it harder to save (Malkin banks. Their minimum fees are around $15, amounting 2007; Roig-Franzia 2007; Hernandez-Coss 2007). The to 10 percent of the average transaction value (Sander typical Mexican family of four consumes about one kilo— and Maimbo 2003). Banks’ minimum fees range between 2.2 pounds—of tortillas each day. In some areas of Mexico, $5 to $50, or around 8 percent to 20 percent of the the price has risen from 63 cents a kilo in 2006 to more than remittance value, depending on sum and destination (El- $2 now. Señora Guzman and her husband spend more than Qorchi, Maimbo, and Wilson 2003). In contrast, $30 each month on tortillas alone. Important forms of informal remittance channels in Bangladesh have been saving for Mexican rural families, such as buying a cow or a estimated to cost around a third of the formal channels horse, cost between $800 and $1,400, which seems to be (Siddiqui and Abrar 2003). One estimate for sending a more than they can afford. remittance of R250 from South Africa, puts the costs at R150 through banks, R120 through money transfer Officially recorded remittances from overseas workers operators and from R25-R50 through informal channels amounted to $240 billion in 2007 (World Bank (Genesis Analytics 2003). Remittance Database 2007). Unrecorded flows are estimated to range from $80 billion to $185 billion Informal channels can take a variety of forms: ethnic (Freund and Spatafora 2008). The overwhelming share stores, travel agencies, money changers, the hawala of remittances passes through money transfer operators dealers of the Middle East (and equivalent systems (figure 2.3). In major recipient countries, banks have a elsewhere in the world), courier services, and hand share, too—mostly in partnership with money transfer delivery. In Mexico, designated members of a community operators. Examples in India are the ICICI and State are often responsible for remitting funds—usually a Bank of India; in Turkey, Esbank, Disbank, Pamukbank, person of integrity, like a teacher or a school principal, Isbank, and others; in the Philippines, National Bank and who is well known to members of the community and Equitable Bank; and, in Ghana, Commercial Bank, its emigrants abroad. In the town of Muna, in Yucatan, which partners with Fast International Money Transfers. Señor Pacheco, a primary school teacher, is responsible This trend is also popular in Gabon, where banks partner for receiving funds and disbursing remittances to with money transfer operators by acquiring a license and recipients in his town (Hernandez-Coss 2007). He is one operating under the Western Union brand. of the three options available in this town, together with FIGURE 2.3 Money transfer operators and other remittance channels Share of various channels for workers’ remittances to selected countries Dominican Republic 84 5 10 1 Cuba 83 15 2 Haiti 83 6 4 3 4 Paraguay 68 15 17 Ecuador 67 17 9 7 Mexico 47 33 9 10 1 Peru 34 24 15 27 Brazil 29 61 10 Bolivia 29 32 22 17 0 20 40 60 80 100 % Share Money transfer operators Bank & credit union Courier Travelers/relatives Other (mail, credit card) Source: Inter-American Development Bank and Bendixen Associates 2003-2006. BANKING THE POOR 17 a branch of the Banco Nacional de Mexico that opened How are remittances used? According to one estimate in in December 2003 and Caja Popular Crecencia A. Cruz, Ghana, 70 percent of the total is spent on immediate which has operated since 2004. However, Señor Pacheco household needs (consumption); less than 30 percent is establishes a link between the sending and recipient invested in assets such as land, cattle, or construction communities and offers easy access. (Schoorl and others 2000). Another estimate, in Mali, suggests that 80 to 90 percent is spent on consumption, If banks are no more expensive than money transfer with almost no investment in business (Martin and operators, why are they used so much less for remittance others 2002). In the Philippines, 68 percent is used to flows (see figure 2.3)? The slower pace of transfers repay debt or meet current needs, 13 percent on through banks and the difficulties faced by recipients in consumer goods, and 1 percent on education. Only 5 the retrieval of funds are the main reasons. If the family percent is used for business capital, only 3 percent on member in the home country has to travel to a distant land and buildings and 1 percent on personal savings.15 bank branch to retrieve the remittance, a money transfer Findings for Latin America are similar (figure 2.4). operator usually will be preferred to a bank. New technologies that sidestep the need for remittance Postal banks help payments recipients to have accounts further erode the banks’ edge. but offer limited services An example are the stored value cards offered by telephone companies in the Philippines and elsewhere. The branch networks of postal banks usually have greater reach than those of commercial banks. For instance, India Remittance recipients in Latin America are more likely has about 155,000 post offices, more than three times the to use banks than those in other regions, probably 45,000 estimated bank branches. In rural areas, the postal because households in the region tend to accumulate branch network has the potential to serve as a platform for their savings in banks (Orozco 2004). There is also some financial services. In many countries, post offices have a evidence that remittances are associated with the history of offering savings products and some limited availability of private bank credit.14 payment services, but the market penetration of postal savings accounts varies. For example, in Namibia, 20 percent of the adult population has an account at the Namibian Post Office Savings Bank. On the other hand, only 5 percent of FIGURE 2.4 Uses of remittances—consumption first, education next Share of remittance receipts used for various purposes in selected countries Cuba 97 12 Ecuador 61 2 8 8 4 17 Peru 60 21 8 6 1 4 Dominican Republic 60 17 5 5 4 9 Mexico 57 13 5 14 2 9 Paraguay 53 12 15 19 1 Brazil 46 13 10 9 7 15 Bolivia 45 21 17 12 4 1 0 20 40 60 80 100 % Total receipts Daily expenditures Education Business Savings Housing Other Source: Inter-American Development Bank and Bendixen Associates 2003-2006. 18 BANKING THE POOR households in India save in post office accounts, even though Cash transfers have been used, for example, in the the postal network reaches remote corners and has a long Kolomo district of Zambia. The aim of the program was history of offering savings products. Even where popular, to allow poor households to buy a second meal per day; postal savings accounts tend to be small deposits. In $10 could buy a 50-kilogram bag of maize. The Chilean Namibia, postal savings comprise 45 percent of all savings by Solidario program, which targets the nation’s poorest number of accounts, but only 13 percent by value. In Sri families, provides income support together with other Lanka, a country with a population of almost 20 million, forms of social support. Other national schemes that there are 4.8 million savings accounts, of which only combine income support with unemployment support 900,000 are active. And these are small deposits with an (linked to an obligation to seek work) are Argentina’s Jefes average balance of $75–100 (World Bank 2006). de Hogares program, introduced after the crisis at the beginning of this decade, and Bulgaria’s guaranteed In many countries, the postal branch network is a major minimum income program. In some poor countries, cash player in money orders and other cash payments. Sri transfers have been combined with public work Lanka’s post office, for example, processes more than 17 programs, such as the Rural Maintenance Program in million payments a year—money orders, pension Bangladesh, where poor women earn a stipend in disbursements, utility bill payments. But postal networks exchange for maintaining rural roads (Ahmed 2005). The rarely handle cashless payments. Nor do they disburse National Rural Employment Guarantee Program in India credit. Finally, they tend to account for a small share of is similar in scope (Grosh and others forthcoming; Ramji the flow of international remittances, estimated at less 2007). than 1 percent in Africa and Latin America. The main reason is that postal services are not full members of the Transfer programs to support health and education are interbank payment and clearing systems. widespread. One of the largest examples, Brazil’s Bolsa Familia program, covers 8.7 million families and 35 million Country experiences with banking services offered people, or close to a fifth of the country’s population. The through postal networks vary greatly. In general, postal program pays conditional stipends to poor families to keep branch networks typically lack the skills in marketing, their children in school. Mexico has a similar program, banking, and technology needed to offer anything more Oportunidades, that covers 5 million families. In than a very basic savings product and some payment Bangladesh, the Primary Education Stipend Program also services. They are further hampered by lack of a full provides cash transfers to keep children in school. banking license. For every success story, such as in Namibia, there are many failures. Banks can ensure that payments are made directly to beneficiaries. They also can provide wide geographical Government transfers to the poor coverage, using branches and other distribution channels, could make more use of banks such as ATMs, mobile units, and point-of-sale terminals at agents’ places of business. Moreover, banks have Banks could serve poor people by channeling government established systems for handling and accounting for cash. transfers. Rich and poor countries alike have programs For example, in Bangladesh’s Rural Maintenance of government support for poor families. These usually Program, uncollected funds are sent back to a central take the form of cash transfers, conditional upon looking account (Grosh and others forthcoming). for a job, keeping children in school, feeding children nutritional supplements, or another social target. BANKING THE POOR 19 The U.S. experience with electronic transfer accounts Credit services to small began in 1996 with legislation that required all federal borrowers are often through payments to be made electronically. Individual recipients special programs of federal payments were required to have an account at a bank. Because the account had to be provided at Most banks will not lend to the poor because they have “reasonable cost,��? the Electronic Transfer Account was few assets to serve as collateral. Some of the sampled created, with a maximum monthly fee of $3 and a banks, such as those in Gabon and Mozambique, said minimum number of free withdrawals. The U.S. Treasury that they had liquidity but lacked entrepreneurs to whom reimbursed banks that are ETA providers with a one-time they could offer credit. Banks find it expensive to screen payment of $12.60 per new account. Follow-up studies and monitor small borrowers, who often must resort to indicated that most beneficiaries already used electronic a combination of formal and informal markets to meet transfers. The biggest obstacle remained for those who their needs. In many countries, they turn to microfinance did not already have bank accounts: about 11 million institutions. One estimate of small enterprises in India people. Only 36,000 ETAs were opened in the first few puts the proportion of informal borrowing by small years, amounting to less than 1 percent of unbanked enterprises at two-thirds of the total credit used by this beneficiaries (Government Accountability Office 2002). segment (Jain 1999).16 A similar basic account was used in India for transfers When banks do lend to small borrowers it is often under the National Rural Employment Guarantee scheme through special programs that incorporate microcredit (Ramji 2007). Most of the new “no frills��? accounts were programs as a way of reducing risk. Take the example of opened for benefits transfers. In South Africa, Standard ICICI in India (Nair and von Pischke 2007). ICICI set Bank established a partnership with the Eastern Cape up a large-scale program of partnership with so-called Department of Social Development to disburse social self-help groups that practice group lending welfare grants to its 70,000 recipients. Arrangements allow (Chakraborty and Duflo 2006). By merging with a recipients to choose to receive payments automatically and smaller bank that already had a network of 1,200 self- free of charge into a newly opened or existing “E Plan��? help groups, ICICI was able to expand to 12,000 such account. Cash withdrawals from the account can be made partnerships within three years. at any ATM. The province agreed to pay the bank R13.50 ($1.65) per recipient per month to cover the account ICICI then began to experiment with partnerships with maintenance fee and two free ATM withdrawals. The microfinance institutions. Microfinanciers were willing arrangement was much less costly than the R31.50 to take on the risk of clients’ financial performance. They ($3.90) the province was spending to administer each would form groups, disburse and collect cash, and keep grant (Rose 2003). records. ICICI would lend to the microfinancier on the basis of its balance sheet and portfolio performance. Cost savings were also achieved in Bangladesh. The government found that the cost of the bank transaction Other commercial banks with large-scale microfinance required to deliver cash benefits under its Income programs have used the model of partnership with Generation for Vulnerable Groups program was nominal. microfinance groups or service companies. Banco ABN The Primary Education Support Program pays bank AMRO in Brazil has partnered with ACCION, a well- transaction fees equivalent to just 2.5 percent of the known microfinance group. Their operating subsidiary, benefits paid out by the government. Real Microcredito, in which ABN AMRO has a 97 20 BANKING THE POOR percent stake and ACCION has invested 3 percent, provides microfinance know-how. Banco de Pinchina, the largest bank in Ecuador, used a similar model to start Credife in 1999, and Sogebank, the largest bank in Haiti, began Sogesol, both in partnership with ACCION. The Unit Desas of the Banka Rakayat Indonesia and CrediAmigo of Banco Nordeste of Brazil are other examples in which a microfinance operation has become a profit center for the bank (Nair and von Pischke 2007). Grameen Bank of Bangladesh and BancoSol of Bolivia provide evidence that microfinanciers can grow into commercial banks. Another model that has been used with some success is that of Wells Fargo, whose business of lending to small and micro businesses relies heavily on credit-scoring models. The model can operate only in an environment rich in credit information, to compensate for the risks associated with less-known borrowers. 11 Rutherford (2000, 2002) provides a series of fascinating descriptions of savings services for the poor. The high discounting of future needs by the poor and consequent reluctance to start saving has been discussed by, among others, Hogarth, Hazembuller, and Wilson (2004), who de- scribe how hard it is for poor people in the United States to save. Banerjee and Duflo (2006) discuss spending patterns of the poor, pointing out that funds spent on festivals, for example, might other- wise have been saved. 12 Hirschland (2005) describes the importance of distance and doorstep collection. Ashraf, Karlan, and Yin (2006a) analyze the success of a doorstep savings arrangement in the Philippines. 13 Bertrand, Chugh, and Mullainathan (2004) describe the “rational be- havior theory��? of the poor to explain their preference for conven- ience and avoiding the formalities of formal systems. 14 Aggarwal, Demirguc-Kunt, and Martinez Peria (2006), in a 99-country study from 1975 to 2003, find an association between remittance flows and financial development. 15 http://www.ercof.org/papers/migrationimpact.html. Similar findings are reported by Gammeltoft (2002), Taylor (1999), and Acosta, Fajnzylber, and Lopez (2008). 16 Cole and Park (1983) and Biggs (1991) report similar findings for Republic of Korea and Thailand. BANKING THE POOR 21 3. Starting to bank Mempe, a young entrepreneur in Ghana, recently started Tanzania—impose the highest charges for basic payment earning enough to set aside some money on a regular services such as debit cards, bank drafts, and remittances. basis. She soon thought of opening a savings account that she could use to make and receive payments. Over time, Opening an account she will gain access to credit through a credit card. The average number of documents required to open an These are among the benefits of being banked. But often account varies from close to five in Côte d’Ivoire and opening an account is neither simple nor cheap. A typical Nicaragua to just one in Honduras, Indonesia, Singapore bank in Côte d’Ivoire or Nicaragua will ask for five and Thailand (figure 3.1). The overall average is three. documents to establish the applicant’s identity, income, Most commonly required document is a government and residence. Some banks levy fees to open an account; identity card, but many banks demand multiple identity in Burundi, the fee can be as high as 3 percent of annual documents as well as proof of income, proof of per capita income. Many banks also impose other types employment, and references from people who already of fees when an account is opened. For example, new hold bank accounts. customers may face separate fees for checkbooks and debit cards. Accounts are not always opened on the spot. Banks in richer countries require fewer documents: It can take up to three days to open an account in Mexico typically a driver’s license or a national identity card. But and South Africa. Across countries, the greater the in some very poor countries, such as the Democratic obstacles to opening an account, the lower the number of Republic of the Congo, Liberia, Mozambique, and the banked. Rwanda, only two documents are generally required. Banks in Africa are the most demanding in their Banks in countries with more competitive banking documentation requirements; East Asian banks are the sectors seem to be less bureaucratic. Thailand’s top three least demanding. banks account for less than 45 percent of the banking system’s total assets, making for a relatively competitive Producing multiple documents can be a challenge for environment. Also, Thai banks require, on average, only even wealthy and resourceful individuals. Many poor one document to open an account. By contrast, in people find it a daunting task that involves time- Nicaragua, where the top three banks hold 95 percent of consuming trips to government offices. Bribes are all banking assets, banks demand, on average, five sometimes exacted. Government officials may ask for documents to open an account. various types of documents, which applicants may not have ready at hand. In Angola, for example, many people Potential customers care about the recurring costs of displaced by the civil war do not have an identity card. being banked. These include account-maintenance fees Prospective bank clients wishing to obtain a replacement and charges for commonly used banking services. African have to travel to an earlier place of residence or even to banks—especially those in Burkina Faso, the Democratic their place of birth to obtain a birth certificate before Republic of the Congo, Liberia, Malawi, Senegal, and applying for a new ID. STARTING TO BANK 23 FIGURE 3.1 Country differences in the documents required to open a bank account Average number of documents required to open a bank account, by country 5 4 3 2 1 0 Lao PDR Tanzania Cameroon Pakistan Cape Verde Afghanistan Malawi Benin Bangladesh Nepal Burundi Gabon Chad Mali Sudan Senegal Thailand Angola Swaziland Zambia Malaysia Liberia Kenya Vietnam Botswana Sri Lanka Nigeria India Uganda South Africa The Gambia Nicaragua Ghana Cambodia Ethiopia Guatemala Namibia Rwanda Indonesia Congo, Dem. Rep. of Côte d'Ivoire Singapore Burkina Faso Lesotho Sierra Leone Mexico Mauritius Mozambique Honduras Philippines Central African Republic El Salvador Niger Madagascar Source: Getting Finance database. Producing other types of documents can be difficult for Passports and national identity cards are the most trusted certain groups. The self-employed and those working in identity documents, with almost all of the sampled transient jobs cannot provide proof of income or banks regarding them as acceptable forms of ID. Only employment. Migrant workers do not have proof of domicile. about half of banks will take a driver’s license or military The more onerous the documentation requirement, the less ID card. Less than a quarter take voter ID cards, which attractive it is for a person on the banking threshold to open may be because many countries do not issue separate an account. The numbers tell the story: the more documents voter ID cards. About half of banks accept other ID required, the fewer the bank accounts (figure 3.2). For every formats. For instance, banks in Cambodia, Ethiopia, additional document required, one finds 153 fewer bank Indonesia, Malaysia, Philippines, and Swaziland take accounts per 1,000 adults.17 student IDs. Banks in many countries also accept employer-issued IDs. FIGURE 3.2 Burdensome documentary requirements, fewer accounts Banks decide on how many documents they need to Density of bank accounts and number of open an account. In part, the decision is dictated by the documents required to open a bank account regulatory authorities. But even within a single country, Malaysia one finds variations in the number of documents Accounts per thousand adults Mauritius required. For example, in Madagascar and Mali, some Sri Lanka Thailand banks ask for six documents; others just one. In Angola, India The Gambia Kenya, Malawi, and Sierra Leone, the difference can be Benin Bangladesh Indonesia as great as four documents. Some banks in Mozambique Zambia Namibia seek up to five documents to open an account— Coef.=-153.5, t-stat.=-3.3 national identification card, passport, proof of address, Number of documents required proof of employment, and a utility bill—while others Note: Each additional document implies 153 fewer accounts per thousand, significant at 1 percent. Controls for GDP per capita, population, need only a government-issued ID. In richer countries, and population density. Source: Getting Finance database. all banks ask for the same type and number of 24 BANKING THE POOR TABLE 3.1 documents. This is to be expected—richer countries Where do banks accept account applications? have widely acceptable standard IDs. Location Percentage of Banks At any branch 95 Stringent know-your-customer standards are becoming At correspondent outlets 27 the norm as a result of the worldwide campaign against Over the Internet 15 At select branches only 8 money laundering and terrorist financing. The new Over the telephone 4 measures have increased documentation requirements Other (work, home, etc.) 14 and reduced the discretion available to banks. Lesotho implemented new rules in January 2007. The country’s Source: Getting Finance database. banks must now ask for three documents, including proof of income, when an account is opened. Some banks that target lower-income segments have noticed a sharp Commentators often cite the difficulty of getting to the drop in new account openings after new regulations went bank as a reason why people remain unbanked. But into effect. using an index of accessibility—an index that rises with the number of locations at which to apply for an Where can a bank account account—we found no significant association between be opened? the average index value for a country and the number of bank accounts. The share of banks offering nonbranch Almost all banks allow clients to open an account at any and mobile options to apply for an account does not branch (table 3.1). More than a quarter of the banks vary much by country income (figure 3.3a). In fact, accept applications at correspondent outlets such as African banks are more likely than others to offer kiosks at retail outlets, post offices, and so on. Only 17 electronic options to open an account (figure 3.3b), yet percent allow clients to apply remotely either over the the density of account holders in Africa remains low. Internet or the phone—the Internet much more so than Making it physically easier for people to apply for bank the phone. About 14 percent of banks offer other ways accounts by itself does not affect bank usage. to apply for an account. In particular, many banks in Africa use mobile sales agents to approach potential customers at home or work. FIGURE 3.3 Nonbranch and mobile options for opening an account do not vary by income a. Share of banks that have nonbranch options for opening b. Share of banks that allow customers to submit an an account, by country income quartile (GDP per capita) account application by Internet or telephone, by region 0.50 Central America South Asia 0.25 East Asia Africa 0.00 Q1 Very Low Q2 Low Q3 Medium Q4 High 0.00 0.05 0.10 0.15 0.20 Any nonbranch option Internet/telephone options Source: Getting Finance database. STARTING TO BANK 25 TABLE 3.2 Account-opening fees—top 10 countries banks do not have an account-opening fee but instead Country rank Account Other opening charge separately for checkbooks or ATM cards. One opening fees charges bank passes through to customers the tax on services to 1 Côte d’Ivoire Chad which it is subject. 2 Burundi Sudan 3 Congo, Dem. Rep. of Namibia 4 Mali Nigeria Banks in the poor countries in our sample—such as 5 Madagascar Pakistan Burundi, Côte d’Ivoire, the Democratic Republic of the 6 Vietnam Zambia Congo, and Mali—are more likely to charge account- 7 Mexico Burundi 8 Cameroon Rwanda opening and other types of fees than those in the richer 9 Sudan Tanzania countries (figure 3.4). The pattern is especially striking 10 Cambodia Cambodia for the latter category of charges. Across regions, banks in Source: Getting Finance database. Africa are more likely to assess application fees and other charges than are banks in other regions. This may be because income from such fees is a big part of banks’ The cost of opening an account earnings, all the more necessary where there are few sound lending opportunities. Charging fees to open an account or to obtain a checkbook or ATM card discourage people from using Countries in which banks tack on charges when opening banks. Fortunately, account-opening fees are rare—only an account tend to have lower rates of financial access, 9 percent of the sample banks have them. They are even after adjusting for income. Potential bank clients charged by one or more banks in Burundi, Cambodia, may shrug off account-opening fees, but they dislike Côte d’Ivoire, the Democratic Republic of the Congo, paying separate charges for items such as checkbooks and Laos, Mali, Madagascar, Sierra Leone, and Sudan (table ATM cards. They may prefer a transparent single fee 3.2). The fees in some cases can be substantial. For instead of a complicated fee structure. example, banks in Burundi charge fees equivalent to 2–3 percent of per capita GDP to open an account. Many more banks, more than 20 percent of the sample, report other types of account-opening charges. Most of these FIGURE 3.4 Account-opening charges are relatively higher in poorer countries a. Share of banks that assess application fees and other charges b. Share of banks that assess account-opening charges, to open an account, by country income quartile (GDP per capita) by region 0.5 South Asia 0.4 0.3 East Asia 0.2 Central America 0.1 Africa 0.0 Q1 Very Low Q2 Low Q3 Medium Q4 High 0.0 0.1 0.2 0.3 0.4 Application fee Other charges Source: Getting Finance database. 26 BANKING THE POOR The time required TABLE 3.3 How long must customers wait to open an account? to open an account Longest wait Shortest wait Country Days Country Days Eighty-six percent of banks open a new account within Mexico 3.0 Cameroon 0.5 Pakistan 2.8 Chad 0.5 minutes of the application or on the same day. But banks South Africa 2.8 Malaysia 0.5 in Mexico, Pakistan, and South Africa can take up to Gabon 2.5 Thailand 0.5 three days to open an account. Countries in which banks Botswana 2.3 Ethiopia 0.6 take longer to open an account tend to have a lower India 1.8 Angola 0.67 Nigeria 1.8 Burkina Faso 0.7 density of bank accounts, even after adjusting for national The Gambia 1.7 Sri Lanka 0.7 income. Banks in Central America take the longest, on Tanzania 1.7 Benin 0.75 average, to open an account—about one and a half days. Malawi 1.6 Honduras 0.8 East Asian banks are much faster, taking only half as long, Congo, 1.5 Niger 0.8 Dem. Rep. of on average (figure 3.5). On average, customers wait six times longer to open an account in Mexico than in Note: In the numbers reported here, responses such as “on the spot,��? “immediately,��? or “10 minutes��? are treated as 0.5 days. Thailand (table 3.3) Source: Getting Finance database. The cost of maintaining Banks in Africa are the most likely to charge such fees, an account while those in South Asia are least likely. The fees are higher in Africa, as well. For instance, the average The monthly fee to maintain a bank account in Liberia monthly charge for account maintenance is close to $4 in is equivalent to half of the typical factory worker’s Africa, but just over $0.50 in South Asia, the other low- monthly income. A Singaporean worker pays far less— income region (figure 3.6). Strikingly, these fees are 0.05 percent. More than half the banks in our sample higher in Africa than in much wealthier East Asia. For a report charging a monthly account-maintenance fee. In bank customer in Africa, such high fees imply a much five countries—Afghanistan, India, Namibia, Nepal, and higher burden relative to income. As noted, the fees are Vietnam—none of the sampled banks charges as much as 51 percent of average monthly income in such a fee. Liberia, 45 percent in Ethiopia, 28 percent in Niger, and more than 23 percent in Malawi. FIGURE 3.5 Opening an account takes longer The cost of making payments in some regions than others Number of days to open an account One of the main reasons people open bank accounts is to make and receive payments. Banked households have East Asia access to a variety of payment instruments—among them checks, bank drafts, direct debits, and debit and credit Africa cards. But the use of such services can come with a price South Asia tag, sometimes a hefty one. For example, the annual fee Central America for a credit card represents almost 132 percent of the 0.0 0.5 1.0 1.5 2.0 average monthly income in Mali. And that fee comes on top of other regular charges, such as account- Source: Getting Finance database. maintenance fees. When one begins to add up the costs STARTING TO BANK 27 FIGURE 3.6 Account-maintenance fees are most common in Africa a. Share of banks that charge an account-maintenance fee, b. Average monthly account-maintenance fee, by region (US$) by region South Asia South Asia East Asia East Asia Central America Central America Africa Africa 0.0 0.2 0.4 0.6 0.8 0 1 2 3 4 Source: Getting Finance database. of payment services, they can take a big bite out of the followed by East Asia ($7.52), Central America ($4.79), budget of the average worker in a developing country. and South Asia ($3). Here, as elsewhere, fees vary. For example, it costs $42 to issue a banker’s draft in the Central African Republic, The same pattern holds for fees on cross-border compared with just $0.57 in the Philippines. transactions. It costs about four times as much to send a $250 bank draft from Africa as it does from South Asia, We considered five major payment-related banking and six times as much to receive the same amount. Central services: (i) the fee charged for a checkbook, (ii) the annual America comes in second for cross-border sending and fee for a debit card, (iii) debit card fees per transaction at receiving fees, followed by East Asia. Residents of the last merchants, (iv) the monthly fee for account maintenance, two regions pay about the same to receive remittances. and (v) the fee for issuing a banker’s draft. Two banking But, on average, it costs slightly more to send a remittance services have a fixed cost—an annual or monthly fee— from East Asia than from Central America. whereas the others are assessed by transaction. The fixed costs are the ones that impose the greatest burden on Checkbooks are relatively expensive in East Asia because customers. Across countries, the mean fee for annual the countries of the region are moving toward a cashless account maintenance is as high as $36. FIGURE 3.7 Cash cards and debit cards also cost more in Africa Africa is the region where payment services are most Annual fees for cash cards and debit cards, by region (US$) expensive. The average cost per checkbook issued in Africa is $4.11, compared with $3.87 in East Asia and South Asia $2.15 in South Asia. The price of bank drafts, annual fees for cash cards (figure 3.7), and transaction-based East Asia fees for cash cards are higher in African banks than Central America elsewhere. For example, the average fee assessed per card transaction at merchants is $0.18 in Africa, $0.08 in Africa 0 2 4 6 8 Central America, $0.03 in South Asia, and $0.02 in U.S. dollars per year East Asia. Africa is also the most expensive region for obtaining a banker’s draft ($9.56 per draft, on average), Source: Getting Finance database. 28 BANKING THE POOR economy. They have started to introduce banking policies circumstances, many commercial banks charge their in which checks are made more expensive in order to customers in dollars rather than in the local currency. encourage a switch to electronic modes of payment. Such fees are onerous for the average consumer, however, as there is no adjustment for purchasing power. Several factors may account for Africa’s high costs. The infrastructure needed to support effective electronic Countries such as Sierra Leone, Sudan, and the payment systems is weak. Electricity is expensive; power Democratic Republic of the Congo still operate in a cash- outages are frequent, and Internet connections unreliable. based economy. Weak infrastructure—unreliable Most African banks lack economies of scale in their ATM electricity, poor roads, insufficient technological networks, thus raising costs. Africa is also the region where resources, and a shortage of skilled professionals—imply asset concentration among banks is highest. The resulting that most transactions are done by cash or check. Debit lack of competition can result in higher fees for consumers. and credit cards are rarely used, thus accounting for the Finally, the credit-to-deposit ratio is low in Africa, and high fees. lending is not always a big source of revenue. African banks appear to derive a larger-than-normal share of their revenue In contrast, Singapore, the richest country of our sample, from fees on depositors’ transactions. has 2,058 accounts per 1,000 adults and, as expected, is among the countries where banking fees, expressed in No single country can boast the highest or lowest fees in income per capita, are lowest. The other nine countries every cost category. For example, the fee for a checkbook is with the lowest bank fees are El Salvador, Mauritius, 45 percent of average monthly income in Liberia, second- Afghanistan, The Gambia, India, Madagascar, Niger, highest in our sample, but Liberia is not among the countries Philippines, and Vietnam. In this group, the average that charges high fees for a debit card (tables 3.4 and 3.5). number for accounts per 1,000 adults is 629, with six However, bank fees are consistently high in some countries, having over 100 accounts per 1,000 adults. El Salvador’s and occasionally higher, when added up, than the average banking industry is well-diversified. Of its twelve large monthly income. For example, the cost of a new checkbook commercial banks, five are privately owned, two are state- and a debit card may be as high as 133 percent of monthly income in the Democratic Republic of the Congo, which TABLE 3.4 Checkbooks and debit cards are unaffordable in also has the lowest access in our sample. some countries Fees for checkbooks and debit cards, as percentage of average monthly income, by country In most of the high-cost countries the banking sector is Highest fees, % Highest fees, % checkbook issue debit card dominated by foreign banks that tend to charge higher Congo, Dem. Rep. of 75 Burkina Faso 70 fees. For example, in Tanzania, apart from a state Liberia 45 Congo, Dem. Rep. of 58 controlled bank, the six largest by asset size are foreign Burundi 39 Madagascar 47 subsidiaries (Bank of Tanzania 2005). The same applies Malawi 33 Senegal 47 to Liberia, whose five major banks have a majority of Central African Republic 30 Ghana 36 Portuguese and South African capital or are foreign- Tanzania 26 Mozambique 35 owned. Moreover, both in Liberia and in the Democratic Afghanistan 25 Mali 27 Republic of the Congo, the banking industry is highly Sierra Leone 23 Chad 26 dollarized, and most of the banks’ clients are international Ethiopia 21 Nepal 20 Mozambique 21 Ethiopia 19 and top-tier local companies, the public sector, and wealthy individuals (IMF 2007). Under such Source: Getting Finance database. STARTING TO BANK 29 TABLE 3.5 Highest and lowest fees for five banking services Number of times country is ranked in the top 10 for highest or lowest fees in five distinct categories, and bank accounts per thousand adults Country Highest fees Accounts per 1,000 Country Lowest fees Accounts per 1,000 Tanzania 4 159 El Salvador 4 693 Congo, Dem. Rep. of 4 2 Mauritius 4 2,011 Burkina Faso 3 121 Afghanistan 3 19 Ethiopia 3 66 The Gambia 3 165 Liberia 3 36 India 3 656 Malawi 3 108 Madagascar 3 31 Senegal 3 79 Niger 3 15 Philippines 3 566 Singapore 3 2,058 Vietnam 3 82 Note: Values for the Democratic Republic of the Congo, Ethiopia, and Vietnam are imputed. See chapter 9. Source: Getting Finance database. owned, and five are foreign-owned. This diversification More people are banked in countries where banks charge fosters competition and lowers fees. Moreover, currency lower account-maintenance fees. The costs of using dollarization has mitigated exchange-rate risks. certain types of payment services, such as bank drafts remittances, are also associated with account density. The In Mauritius, domestic bank assets represent more than higher the fee charged for receiving a foreign banker’s 100 percent of GDP, and savings exceed 40 percent of draft of $250, the lower the number of bank accounts GDP. Systems for payment, securities trading, and (figure 3.8). The same applies to the fee charged for settlements are efficient, which explains the low fees of sending a foreign draft.18 the country’s commercial banks. Other countries in the top 10—such as India, the Philippines, and Vietnam— 17 Beck, Demirguc-Kunt, and Martinez Peria (2007a) report similar find- ings from a survey of 193 banks across 58 countries—more documen- have experienced high growth in the last decade and also tation at account opening is associated with fewer accounts per capita and lower demographic branch penetration. have relatively well-developed banks. 18 These findings are consistent with Beck, Demirguc-Kunt, and Martinez Peria (2006b). They report a significantly negative association be- tween accounts per capita and annual fees to checking and savings accounts. They also find a negative association between the cost of FIGURE 3.8 making international payments and accounts per capita, although the Fees for common services are higher where effect is not significant. the density of accounts is lower Cost of receiving a banker’s draft for $250 and accounts per thousand adults Malaysia Accounts per thousand adults Mauritius Sri Lanka India Thailand Ghana Niger Zambia Namibia Benin Coef.=-269.557, t-stat.=-2.13 Cost of receiving a banker’s draft Note: A movement from the 25th to 75th percentile on the index of costs of receiving a bank draft is associated with 269 fewer accounts per thousand adults, controlling for GDP per capita. Source: Getting Finance database. 30 BANKING THE POOR 4. Adding customer services The quality and convenience features of a bank checks as a convenience feature despite a widespread account—after-hours withdrawals, overdraft provisions, problem of bounced checks. But newer entrants to the and overdraft notification—add to its ease of use. But market are offering more innovative products; their do they encourage more people to open accounts? We standard product is a transactions account without a constructed an index combining the conveniences just checkbook. mentioned, plus exemption from the need to present a passbook when making withdrawals and deposits. Only 4 percent of the banks surveyed offer a free credit card. Banks in poorer countries have high scores on the Nicaragua is the only country where as many as half of all index—Senegal, Kenya, and Chad have the highest banks surveyed offer a credit card free of charge. scores (figure 4.1). The index of free usage is an equally weighted sum of the Free services five features discussed above. Again, low-income countries such as Côte d’Ivoire and Rwanda have high Banks offer some free services. For example, about 90 mean values on this index (figure 4.1); those values show percent of banks in the sample permit unlimited balance no significant association with the density of accounts in inquiries at no cost. Ethiopia, Mexico, and Mozambique the country.19 are the only countries in which three or more of the five sampled banks do not offer unlimited balance inquiries. We also examined potential inconveniences to see if they Three-quarters of banks also allow free withdrawals on affect the number of accounts per thousand adults. Most their standard accounts. In 28 out of 54 surveyed common among the inconveniences are banks’ requirements countries, all commercial banks do so. But in Liberia, for minimum opening deposits or minimum balances. Namibia, and South Africa none of the banks offer free Eighty-eight percent of banks reported minimum-deposit withdrawals, and in India, Mexico, and Swaziland less requirements to open their standard account. Rwanda and than a third do so. Gabon are the only two countries in which none of the banks requires a minimum deposit to open a standard There is wide variation in the availability of ATM services account. In 37 countries, all banks require a minimum and checking services. In 10 countries (Afghanistan, deposit to do so. The amount required can be as much as Burundi, Cameroon, Cape Verde, Central African three times per capita income in poorer countries (3.3 times Republic, Chad, Gabon, the Gambia, Liberia, and in Burundi, 2.3 times in Madagascar). While the proportion Thailand) none of the banks offers a free ATM card. In of the minimum deposit to income per capita tends to Côte d’Ivoire, Honduras, Nigeria, and Rwanda, all the decline for richer countries, to less than 1 percent of per banks offer free ATM cards. All banks in Burkina Faso, capita income in Malaysia, Thailand, and South Africa, the Burundi, the Central African Republic, Gabon, requirement of a minimum deposit remains highly Madagascar, Nepal, Senegal, and Uganda offer free prevalent. The relationship of minimum deposit and income checking services, but in 16 of 54 sampled countries none is not linear; thus in Mexico and Singapore, minimum of the banks does so. It is interesting that in countries requirements are as high as 2 percent and 6 percent of such as Botswana, long-established traditional banks offer income per capita, respectively. ADDING CUSTOMER SERVICES 31 FIGURE 4.1 Convenience features and free usage features offered with bank accounts, by country a. Performance of sampled banks on index of convenience features 1.0 0.8 0.6 0.4 0.2 0.0 Tanzania Cameroon Cape Verde Gabon Nepal Afghanistan Burundi Benin Pakistan Malawi Bangladesh Mali Senegal Sudan Chad Thailand Cambodia Swaziland Vietnam Rwanda Uganda Indonesia The Gambia Nigeria Sri Lanka Angola Zambia Namibia Kenya Botswana India Nicaragua Liberia Ethiopia South Africa Ghana Malaysia Singapore Mexico Lesotho Mozambique Sierra Leone Côte d'Ivoire Honduras Burkina Faso Mauritius Lao PDR Philippines El Salvador Niger Madagascar Congo, Dem. Rep. of Central African Republic b. Performance of sampled banks on index of free-usage features 1.0 0.8 0.6 0.4 0.2 0.0 Tanzania Cameroon Pakistan Burundi Senegal Mali Malawi Cape Verde South Africa Cambodia Bangladesh Gabon Ghana The Gambia Afghanistan Swaziland Sri Lanka Benin Nepal Liberia Thailand Malaysia Botswana Zambia Sudan Ethiopia Namibia Nicaragua Chad Rwanda Vietnam Indonesia Uganda Angola India Kenya Nigeria Singapore Côte d'Ivoire Mexico Burkina Faso Mozambique Sierra Leone Honduras Lao PDR Mauritius Philippines Lesotho Central African Republic Congo, Dem. Rep. of Madagascar El Salvador Niger Note: The index of convenience features is the equally weighted sum of after-hours access to withdrawal services, overdraft provision, overdraft notification, and waiver of the need to present a passbook when making withdrawals and deposits. Free usage index is the equally weighted sum of free balance inquiries, free withdrawals, a free ATM card, free checking, and a free credit card. Source: Getting Finance database. However we find no significant association between the Requirements for minimum ongoing balances needed to density of accounts and the presence of a minimum maintain an interest-bearing savings account are more opening-deposit requirement. Separating savings accounts varied. Such requirements exist in 43 percent of the from checking accounts, we find minimum-balance sampled countries. Mozambique, Singapore, and Sudan requirements to be similarly widespread. All commercial are the only countries in which no bank has them. In banks in 43 of 54 sample countries have an opening Angola, El Salvador, and Mauritius, a couple of the banks balance requirement for savings accounts. Sudan is the only in each country require a minimum balance to maintain country in which none of the banks imposes a minimum a savings account. In 23 of our 54 sample countries, all opening-balance requirement. While, as expected, there is of the sampled banks impose such a minimum on a negative association between account density and the interest-bearing savings accounts. In Africa, all of the minimum balance required to open a standard savings banks in 16 of 35 countries have an ongoing balance account, the association is insignificant. requirement for savings accounts. In the South Asia 32 BANKING THE POOR region, banks in India, Pakistan, and Sri Lanka, together and taxes, and receipt of government transfers. Fourth, we with Indonesia, Laos and Malaysia in East Asia and examined the time it takes to complete a range of Nicaragua in Latin America, all require a minimum domestic payment transactions. balance to sustain an account without incurring a penalty. Ongoing balance requirements are less frequent than Payment instruments available opening balances in countries of all income categories with standard accounts (figure 4.2). However, the levels of ongoing balances, as a percentage of income per capita, vary less than do those Banks in all regions are more likely to offer checking and of opening balances. Thus, while the levels of ongoing debit facilities than other payment instruments (figure balances in poorer countries are not as high as opening 4.3a). Virtually all banks in all regions offer checking balances, the gap between the two requirements is less facilities. Debit cards and direct debit facilities are more than it is in richer countries. common than cash cards in all regions, confirming a universal trend toward the substitution of more Facilities for making and sophisticated services for less sophisticated ones. Some receiving payments interregional variations emerge: fewer African banks offer debit cards (58 percent) compared with other regions (78 It is often said that more people would use bank services percent and 82 percent in South Asia and Central if it were easier to make payments through the banking America; 88 percent in East Asia). None of the banks system. To determine if that is true, we examined four offers debit cards in Chad, Liberia, or Niger, whereas all aspects of retail payment services in relation to national do so in Indonesia, Malaysia, Thailand, and Vietnam. levels of popular access to banking. First, we looked at East Asia leads in terms of direct debit services (offered by the range of payment instruments offered by banks with 97 percent of banks, compared with 79 percent in Africa their standard account—checks, cash cards, debit cards, and only 54 percent in Latin America), debit cards (88 direct debit facilities (interbank and intrabank), and percent, compared with 82 percent in Central America credit cards. Second, we examined the quality of ATM and 58 percent in Africa), and cash cards (78 percent). networks and their degree of interoperability. Third, we East Asia also has the lowest ratio of check use (84 explored the payment options offered by banks for retail percent, compared with 100 percent in South Asia), payments, person-to-person payments, payment of bills which probably reflects a move away from checks in favor FIGURE 4.2 Minimum-balance requirements for savings accounts can be very high in poor countries a. Share of banks with balance requirements for savings b. Balance requirements for savings accounts as share of accounts, by country income quartile (GDP per capita) per capita income, by country income quartile (GDP per capita) 25 100 20 80 15 60 (%) (%) 10 40 5 20 0 0 Very Low Low Medium High Very Low Low Medium High % Banks with opening balance requirement Opening balance as % GDP per capita % Banks with ongoing balance requirement Ongoing balance as % GDP per capita Source: Getting Finance database. ADDING CUSTOMER SERVICES 33 FIGURE 4.3 Some payment instruments are more common than others, regardless of region a. Share of banks offering selected payment instruments with b. Share of African banks offering selected payment instruments standard accounts, by world region with standard accounts, by subregion 100 100 80 80 60 60 40 40 20 20 0 0 Checks Cash cards Debit cards Direct debit Credit cards Checks Cash cards Debit cards Direct debit Credit cards Africa South Asia East Asia Central America West Africa East Africa South Africa Central Africa Source: Getting Finance database. of electronic payments. Only half of the sampled banks Central Africa lags similarly in the use of debit cards. in Cambodia, Malaysia, and Thailand offer checking Only 30 percent of sampled banks offer such cards on services with a standard account. Latin American and their standard account, compared with at least 55 percent East Asian banks lead in terms of credit card services (35 elsewhere, and 76 percent in South Africa (figure 4.3b). percent and 34 percent, compared with 16 percent and The region’s banks are far behind in credit cards as well 14 percent in South Asia and Africa). For example, four (3 percent, compared with 19 percent in the next-lowest of five banks in Honduras offer a credit card with their region of West Africa and 24 percent in Southern Africa). standard account, as do three of five banks in Malaysia No banks offer debit cards in the Central African and Vietnam. Republic, Chad, or Niger; only one bank does so in Burundi and the Democratic Republic of the Congo. Even within Africa, variation by instrument is high compared with variation by region, although some We found no significant association between the range regional patterns are discernible. Southern African banks of standard payment instruments offered and the number use a lower proportion of checks (76 percent) than do of accounts per thousand adults or the ratio of private their regional neighbors (more than 90 percent in each of credit to GDP. One reason may be that newer electronic the other regions); they also use a high proportion of models of payments substitute for more traditional direct debit (88 percent). Only two of five banks offer modes as countries grow wealthier. checks in Swaziland, and three of five in Botswana and Lesotho, versus all in the West African group—Benin, Network payment capabilities Cameroon, Cape Verde, Côte d’Ivoire, Gabon, The Gambia, Ghana, Liberia, Senegal, and Sierra Leone. East We next examined the relationship between the Africa’s banks engage equally heavily in direct debit, and capabilities of the payment network and bank accounts West Africa is not far behind, at 81 percent. Only Central per thousand adults. The first part of this question Africa appears to lag (53 percent).20 involves the extent to which networks are interoperable. Can ATM cards be used only at the issuing bank, or are they interoperable across networks of different banks? Can ATM cards also be used to make payments at point- 34 BANKING THE POOR of-service terminals, such as those available in stores? Better network capabilities are positively correlated with Most of the bankers we sampled reported that ATM the density of accounts, but the association is not networks in all regions had fairly high levels of significant. By contrast, there is a significant positive interoperability. While Latin America and East Asia lead association between network interoperability and the in terms of linked networks (88 percent and 90 percent overall depth of the financial system, as measured by the respectively), South Asia and Africa also achieve ratio of private credit to GDP. This suggests that while substantial interoperability (62 percent and 54 percent, better network capabilities are associated with an respectively). All banks in Honduras, Mexico, and aggregate increase in flows through the financial system, Nicaragua allow other banks’ ATM cards to be used at individual retail customers do not take such capabilities their networks and at merchants’ point-of-service into account in their decisions about whether to open terminals. Guatemala’s lower level of network linking is and maintain a bank account. the exception in Central America. Furthermore, since 2005, all commercial banks in Pakistan have been Payment channels directed to join one of two ATM networks, M-Net or I- Link, and to offer their customers the possibility of Payment channel options are more numerous in East Asia electronic banking. After the change, transactions rose by and Central America than in South Asia and Africa. They 47 percent from 2005 to 2007. In Kenya, many debit are most highly developed for person-to-person cards can operate through KenSwitch, Pesapoint, or payments; somewhat less so for payments to or from VISA—interoperability of networks is high. Regionally, businesses. The options for payments to and from the the proportion of ATM cards that can also be used to government are the least developed (figure 4.4). make payments at point-of-service terminals is not much lower than the rates of interoperability, at 82 percent and For example, 66 percent of East Asian banks (including 87 percent in Latin America and East Asia, and 62 all banks in Indonesia, Malaysia, the Philippines, percent and 53 percent in South Asia and Africa. Thailand, and Vietnam) and 47 percent of Latin American banks permit payments between persons over FIGURE 4.4 Some payments are more common than others, with some regional variations Percentage of sampled banks offering various payment options for individual payments to other individuals, to businesses, and to government, by region 100 80 60 40 20 0 Direct debits Payments cards at ATMs Direct credit / premises Direct credit/ premises Payment cards/ premises Payment cards/ premises Direct credit- phone / Internet Direct debits Payment cards/ premises Direct credit/ premises Payments cards/ at ATMs Checks Payments cards at ATMs Checks Direct debits Direct credit- phone / Internet Checks Direct credit – phone / Internet Person to person Individual to business Individual to government East Asia Central America Africa South Asia Source: Getting Finance database. ADDING CUSTOMER SERVICES 35 ATMs at their premises, as opposed to 27 percent for that rate is a good deal higher than the 12 percent Africa and 14 percent for South Asia. Fifty-three percent recorded for East Africa, 10 percent for West Africa, and 3 of East Asian banks and 29 percent of Latin American percent for Central Africa. Only one bank in the entire banks are capable of transactions with the government via Central African region, in Burundi, offers this service. ATMs at their premises. Only 11 percent of African banks and 3 percent of South Asian banks enable such We examined the association between retail payment transactions with the government (figure 4.4). Indeed, no systems and the availability of alternative payment banks in India, Nepal, or Sri Lanka offer such channels using an index of network capabilities consisting transactions. This illustrates the limitations that present of equally weighted responses to the six payment options payment networks have in facilitating transfer payments across the three different channels (figures 4.5a and to citizens. 4.5b).21 As with network capabilities, the availability of alternative payment options is positively correlated with In Africa, patterns again vary more by service type than account density—but not significantly so. However, the by region. Southern Africa tends to use more association between network capabilities (and the sophisticated payment options than do the other regions; availability of payment channels) with the ratio of private East Africa comes in second. West Africa and Central credit to GDP is highly significant, even after controlling Africa typically have lower levels of service provision. for income (figure 4.5a). This finding suggests that while Payment channels between government and citizens are the features of the national infrastructure for retail less-developed than are person-to-person and individual- payments may facilitate overall financial flows, they do to-business channels. For example, 58 percent of not entice the marginal consumer to open a bank account. Southern African banks allow person-to-person payments via ATMs (especially in Angola, South Africa, and Processing domestic payments Zambia), compared with 40 percent in East Africa, 25 percent in West Africa, and 23 percent in Central Africa. There can be considerable variation between countries in Only 21 percent of Southern African banks offer the the time required to process payment transactions (table same facility for government payments or transfers—and 4.1). Sending a domestic payment to a rural area takes FIGURE 4.5 Private credit, network interoperability, and availability of payment channels a. Network capabilities and financial depth b. Retail instruments and financial depth Malaysia Malaysia Vietnam Singapore Vietnam Thailand Maritius Namibia Thailand Namibia India Burundi Singapore India Nicaragua Philippines Guatemala Indonesia Nigeria Swaziland Cameroon Swaziland Liberia Cameroon Gabon Botswana Gabon Coef.=21.43, t-stat.=2.11 Coef.=52.49, t-stat.=3.24 Index of network capabilities Index of retail payment instruments Note: A cross-country regression of the ratio of private credit to GDP on an index of network capabilities (panel a) shows that a 10 percent increase in the index value is associated with a 2.1 percentage point increase in the credit ratio (t-stat: 2.11). A similar regression on an index of the range of available retail payment instruments (panel b) shows that a 10 percent increase in the index value is associated with a 5.2 percentage point increase in the credit ratio (t-stat: 3.24). Source: Getting Finance database. 36 BANKING THE POOR barely half a day in the Philippines, but it can take eight A more efficient system of retail payments does not days or more in the Central African Republic. Sending a appear to be associated with access to banking services, as payment to an urban area takes two days in India but half measured in terms of the number of accounts per a day in the Philippines. Cross-border bank drafts take a thousand adults. However, better network month in the Philippines but evidently less than four days interoperability and better retail payment services are in Sudan positively and significantly associated with financial depth, which may, over time, make banking more useful We found the expected negative association between the for more people. index of domestic payment transactions and the density 19 If low-end users are sensitive to costs, how do we explain the lack of of accounts. However, the relationship was not sensitivity to free features? More analysis is needed. 20 Southern Africa includes Angola, Botswana, Lesotho, Namibia, South significant. The cross-border payments index likewise has Africa, Swaziland, Zambia. East Africa in our sample comprises Ethiopia, Kenya, Madagascar, Mauritius, Malawi, Mozambique, no significant association with the density of accounts. Sudan, Uganda and Tanzania. West African countries are Benin, Cameroon, Cape Verde, Côte d’Ivoire, the Gambia, Ghana, Liberia, This could reflect the low cost of processing time to the Nigeria, Senegal and Sierra Leone. The Central region is defined to in- relatively poor marginal client, whose immediate clude Burkina, Burundi, Chad, the Democratic Republic of the Congo, Mali, Niger and Rwanda. preoccupation is with the cash cost of the service. Such a 21 The index is defined in the glossary. See “Index of network quality and interoperability.��? client may place a relatively low marginal value on the additional time taken for the transaction. Also, longer payment time reflects weaknesses in the financial system that may make the system unattractive to the marginal client for various other reasons, such as cumbersome account-opening procedures. There is no significant association between the time indices we constructed and the ratio of private credit to GDP. TABLE 4.1 Processing domestic and cross-border payments Mean, minimum, and maximum time (in days) required to process various transactions in sample countries Number of Country Country Time required for payment to… countries Mean Minimum Maximum (minimum) (maximum) Domestic transactions Reach the second-largest city in the country 52 1.1 0.4 2.3 Thailand India Reach a rural area 50 2.1 0.6 8.5 Philippines Central African Republic Clear a check from another bank 52 2.6 0.8 6.3 Chad South Africa Clear a direct credit 49 1.6 0.6 3.3 Malaysia Ethiopia Clear a direct debit 50 1.7 0.5 3.4 Niger Sri Lanka Cross-border transactions Send a cross-border bank draft 52 11.6 3.3 30.8 Sudan Philippines Send a cross-border wire transfer 52 2.3 0.8 5.5 Indonesia Nicaragua Source: Getting Finance database. ADDING CUSTOMER SERVICES 37 5. Building branchless banking Reducing the distance between clients and services can municipalities that had no banking facilities was reduced increase the number of bank accounts. Distance imposes from 1,659 to zero, and 6.5 million new accounts were costs of transportation and the opportunity cost of time opened. Uganda built a bank-agency partnership system forgone. Some of the ways to bridge that distance and based on point-of-service terminals at merchants and gas thereby increase the density of bank services are to add stations (Firpo 2005). In South Africa, Standard Bank’s E branches and nonbank service points, or to make services plan provided staffed ATM machines; the bank claims available over telephone and Internet. that, as a result of this innovation, its client base grew by 12 million between 1994 and 2004 (Cracknell 2004; The problem of distance is the most acute in rural areas, Porteous and Hazelhurst 2006). where populations are sparse and communications infrastructure is generally less extensive than in urban Studies point out that lack of physical access to a bank areas. For instance, the Ghanzi district in Botswana has branch does not translate into fewer accounts in urban a population of 33,000 in an area of 118,000 square areas. Regional differences in the availability of banking kilometers. In Mozambique, 100 of 128 districts have no services, especially in Mexico and Brazil, are shown to banking institution, ATM, or point-of-sale banking reflect income differences.23 In South Africa, most rich facility. Those 100 districts cover almost 78.5 percent of people and almost half the urban population agree that the national territory. there is a bank nearby. In other words, banks tend to open branches in wealthy and middle-income One study of the effect of mandated rural branch banking neighborhoods where they can reach a larger and more in India suggests that the resulting rise in deposits and profitable clientele. Furthermore, studies of the United credit has had a positive effect on rural poverty (Burgess States show that more than 90 percent of nonbank service and Pande 2004). Increases in deposits made possible by providers are located no more than a mile from a bank or shrinking distances have also been documented in Mexico credit union (Fellowes and Mabanta 2008). (Aportela 1999). A study in the Philippines shows that doorstep collection schemes produced net growth in the Banking over the telephone and the Internet has grown number of households saving in banks (Ashraf, Karlan, rapidly and offers great potential for further growth. Cell and Yin 2006a). A common thread in all these studies is phones, becoming ubiquitous even in poor countries, that small savers are prepared to trade positive financial account for a large share of that potential. However, returns for the convenience of shorter trips to deposit- banks are still more likely to offer common services such taking facilities (Rutherford 1998).22 as balance inquiries, bill payment, and statements over the Internet than via cell phones. The scale of cell-phone Recent attention has focused on the advantages of based banking services has remained small, with some nonbank partnerships in bringing people and banks exceptions, such as M-Pesa in Kenya and Smart in the closer together. Brazil introduced 32,000 new bank Philippines. The services offered tend to be simple, service outlets in five years through partnerships with the usually balance inquiries and money transfers. postal franchise, lottery shops, drug stores, and other agents (Kumar and others 2006). The number of BUILDING BRANCHLESS BANKING 39 The degree of availability of electronic banking does not option (telephone and/or Internet) for opening an appear to be associated with more accounts per capita. account. About 42 percent of banks in the richest quartile One reason could be that such services are primarily offer nonbranch options, as compared with 36 percent intended for and used by those who already have bank in the poorest quartile. The percentages for the Internet accounts. Electronic banking faces infrastructural and telephone options are 20 and 16, respectively. constraints, especially unreliable power supply, which Country scores on the index of nonbranch accessibility hampers coverage and increases the cost to users of are shown in figure 5.1. The index of nonbranch charging their phones. There is also a challenge of secure accessibility is set to 1 if the bank offers one or more of cash handling when using the Internet or a cell phone to the following options to apply for an account: over the transfer money. telephone, via Internet, at a correspondent outlet, and at other locations, such as home or the workplace. Opening an account— Singapore and South Africa have high values for branch vs. branchless banking nonbranch accessibility, but so do Senegal and Chad. Conversely, Mexico and Malaysia have relatively low As noted earlier, almost all banks (95 percent) accept scores. Overall, there is no cross-country evidence that account applications at any branch—Mali and Chad have nonbranch options for opening accounts are associated the most restrictions. Just 18 banks (of 235) reported that with a higher number of bank accounts per capita. they accept applications only at select branches; they include banks in Mozambique and Ghana in Africa, While banks in higher-income countries may offer Cambodia and Vietnam in East Asia, and Honduras in convenient electronic access for transactions, most appear Central America. A quarter of banks (27 percent) report to require physical presence at a branch to open an that correspondent outlets are at other locations where account, as do their counterparts in low-income people can apply to open an account. That subsample countries. The relation between nonbank options for includes three or more of the sampled banks in Burkina opening an account and the number of bank accounts Faso, Chad, and South Africa. However in 10 of the 35 per thousand adults is positive but insignificant. African countries covered, all banks agreed that correspondent outlets could not be used for account- Electronic banking opening purposes. Sixteen percent of banks accept applications over the Internet, and just 4 percent accept ATMs, the Internet, cell phones, and other new applications by telephone. Singapore is the only country technologies make banking available without need for in which most banks accept applications by phone; yet tellers or branches. They aid financial transactions in two most or all banks accept applications over the Internet in ways. First, they make services accessible when bank Côte d’Ivoire, Mauritius, Namibia, and Sri Lanka. branches are closed or far away. They can also be used to make long-distance domestic and foreign payments. We considered separately, and by country income, the Second, automated services are cheap, because electronic options for opening accounts at branch and at nonbranch messaging is costless. locations. There is little discernible difference across countries (grouped by income quartiles) with respect to The cost reduction potential of automated services is the share of banks that offer a nonbranch option enormous. Conducting a banking transaction through a (correspondent banks, Internet, phone, or other teller at a bank costs more than a dollar. The same locations, such as home or workplace) or an electronic transaction costs less than $0.30 at an ATM and less than 40 BANKING THE POOR FIGURE 5.1 Opening an account remotely is easier in some countries than in others Country performance on index of nonbranch options for opening an account 1.0 0.8 0.6 Central African Republic The Gambia Congo, Dem. Rep. of Afghanistan Indonesia 0.4 Malaysia Lao PDR Liberia Côte d'Ivoire Burkina Faso Sierra Leone South Africa Mozambique Bangladesh Cape Verde Madagascar Cameroon Singapore Nicaragua Swaziland Cambodia Guatemala El Salvador Botswana Philippines Honduras Mauritius Sri Lanka Tanzania Vietnam Namibia Burundi Rwanda Senegal Uganda Thailand Pakistan 0.2 Ethiopia Malawi Zambia Angola Nigeria Lesotho Mexico Ghana Gabon Kenya Sudan Nepal Benin Niger Chad India Mali 0.0 Source: Getting Finance database. $0.10 via Internet or mobile phone. The low unit cost of Internet users still make up just a small fraction of all Internet banking and mobile phone banking in particular telecommunication users, whereas cell phones now permits banks and other operators to make profits on dominate fixed lines (figure 5.2a). But banks are much very small transactions. In a case study of the WIZZIT more likely to offer services over the Internet than over system in South Africa, a cell-phone based banking the phone. We considered five common needs—balance facility aimed at the estimated 16 million unbanked or inquiries, ministatements, status-of-payment checks, underbanked South Africans, annualized costs were money transfers, and bill payments—all of which were estimated at around $70 per year, compared with typical more widely available by Internet than by phone (figure costs of $103 at one of South Africa’s four big banks.24 5.2b). In India, all banks, public and private, allow The cost of using a basic Mzansi account, which has no payments over the Internet, but only public sector banks minimum balance, no monthly maintenance fee, and a provide (limited) services by phone. There are few limited number of free monthly deposits and withdrawals, differences in the levels of service provided through cell is in between, at $94 a month (Cracknell 2004). phones and landline phones, which suggests that commercial banks are not, at present, providing services in a manner designed to facilitate access to banking for FIGURE 5.2 Remote banking services have not kept pace with spread of cell phones a. Internet users as percentage of all telecommunications users, b. Share of banks offering various services by phone and Internet and cell phone subscribers as percentage of all telephone subscribers 100 80 75 60 (%) (%) 50 40 25 20 0 0 2004 2005 2004 2005 2004 2005 2004 2005 Balance Mini Payment Money Bill Selling cell Africa South Asia East Asia Latin America inquiry statement status transfer payment phone minutes Internet Cell phone Phone Cell phone Internet Source: Getting Finance database; World Development Indicators, World Bank, 2008a. BUILDING BRANCHLESS BANKING 41 cell phone users. This situation may be ascribable to the for balance inquiries. Ratios are lower still for money lower costs and greater security of Internet transactions. transfers and bill payments. Only 9 percent of banks in Africa offer bill payments by cell phone, and 12 percent The most frequently offered remote service, in all regions, offer money transfers, compared with 43 percent and 37 is balance inquiry. In East Asia and Latin America, this percent, respectively, in East Asia. service is offered over the Internet by 83 and 88 percent of all banks and via cell phone by 60 percent and 35 After simple transactions such as checking one’s balance percent of all banks (figure 5.3). All banks in Indonesia, and recent activity, payments and other money transfer Thailand, El Salvador, and Honduras allow their services are the most common banking transactions customers to make balance inquiries over the Internet, available by cell phone (Ivatury and Pickens 2006; CGAP but only about half offer the same service via cell phones. 2008), although purchasing air time (or cell phone Forty-nine percent of banks in South Asia and Africa also minutes) is increasingly common. For example, more offer balance inquiries by cell phone. Internet services for than half of the transactions on the mobile banking balance inquiries are offered by 60 percent of African service recently launched by Botswana’s First National banks (including all banks in Burkina Faso, the Bank are purchases of air time. The use of cell phones to Democratic Republic of the Congo, the Gambia, and pay bills lags in many environments because bill Senegal) and 49 percent of South Asian banks, aggregation services have not yet been developed. particularly in India and Sri Lanka. The promise of cell phones Next in popularity are ministatements (information on the last few transactions). Twenty-six percent of banks in The intense interest in the use of mobile phones to Africa and 32 percent in South Asia offer this service by deliver banking services in developing countries is cell phone, compared with 58 percent and 46 percent explained by the wide and rapidly growing use of mobile that offer it over the Internet. Only one bank in Burkina phones in these countries. Internet-based banking, Faso offers ministatements via cell phone, whereas four though well established in developed countries and banks offer statements via the Internet. Cell phone growing in the developing world, requires access to a availability for ministatements is considerably lower than computer, ideally a personal computer, a luxury item to FIGURE 5.3 Balance inquiries are the most common service – Bill payments are the least common Percentage of sampled banks offering selected services by telephone and Internet, by region 100 75 (%) 50 25 0 Phone Cell Internet Phone Cell Internet Phone Cell Internet Phone Cell Internet Africa South Asia East Asia Latin America Balance inquiries Money transfers Bill payments Source: Getting Finance database, 2007. 42 BANKING THE POOR all but a minority in most developing countries. By Philippines) to transmit payment instructions, with the contrast, cell phones are relatively inexpensive, and their bank serving as the cash-processing point. recent growth has been colossal. More than 800 million Telecommunications companies dominate other models mobile phones were sold in developing countries from in which a bank’s services may be retained mainly for the 2002 to 2005 (Ivatury and Pickens 2006). purpose of cash processing (M-Pesa in Kenya, WIZZIT in South Africa). In some cases the telecommunications The marked growth in both Internet and mobile phone company may also take on cash-processing functions users is not matched by increases in the ranks of those (Globe in the Philippines). employing landline phones to reach their bank. In Africa, East Asia, and South Asia, landline phone use rose from Is cell phone banking associated with more bank 11 percent to 13 percent between 2004 and 2005. accounts? We constructed a mobile technology index, During the same period, the number of Internet users defined as the equally weighted responses from sampled increased by 24 percent in East Asia and as much as 71 banks regarding the availability of five transactions percent in Africa. Annual Internet growth rates of 41 (balance inquiries, ministatements, checking payment percent and 53 percent were observed in Latin America status, transferring money, and bill paying bills), using and South Asia, respectively. Slower growth in East Asia cell phones. reflects high levels at the beginning of the period, compared with the low base in Africa. Cell phone use Although the density of accounts is positively associated during the same period grew by 66 percent in Africa and with the index, the association is not significant (figure 98 percent in South Asia. 5.4a). Conversely, there is a positive and significant association between the index and the ratio of private Most important in terms of the potential to extend credit to GDP (figure 5.4b). One explanation is that financial services, the absolute number of cell phone mostly people who are already banked utilize mobile subscribers now greatly outstrips that of Internet users banking. Mobile banking services are convenience and landline subscribers (see figure 5.2). In 2005, there features designed to retain existing customers rather than were 17 landline telephone subscribers per thousand attract new ones. In Botswana and Lesotho, for example, persons in Africa and 39 in South Asia, far fewer than the mobile banking services are tied to bank account 123 cell phone subscribers per thousand in Africa and 79 ownership. in South Asia. Four times as many people use cell phones than the Internet in Africa, 2.5 times as many in Latin Cell phone banking is penetrating new markets—among America, and 1.6 times as many in South Asia. them China, India, Indonesia, Mongolia, North Africa, and Pakistan, and tentatively Ecuador, Mexico, and Emerging cell phone banking models encompass a range Nicaragua. For instance, Barclays India launched a mobile of relationships between banks, telecommunications banking service for its urban customers using a technology providers, and nonbank institutions. Bank-led models are with mass-market potential and trials to reach more rural linked to clients’ bank accounts; users can send funds via customers (Kamel and Hassan 2003; Donner 2003, 2006). any telecommunications provider. Those funds can be dispensed or received at banks and other bank-authorized Botswana provides another example of the successful use outlets. Banks may also enter into joint ventures with of cell phones in the banking sector. Roughly the size of specific telecommunications providers (such as MTN France but with less than 3 percent of the inhabitants, Mobile Money in South Africa, or Smart in the Botswana is among the most sparsely settled countries in BUILDING BRANCHLESS BANKING 43 FIGURE 5.4 Cell-phone banking; good for financial depth, no relation to account density a. Mobile technology and density of bank accounts b. Mobile technology and the ratio of private credit to GDP Malaysia Malaysia Accounts per thousand adults Private credit/GDP Vietnam Cape Verde Sri Lanka South Africa India Singapore Guatemala Honduras Honduras Philippines Indonesia Pakistan Kenya Indonesia Nicaragua Kenya South Africa Guatemala Botswana Angola Mexico Angola Coef.=245, t-stat.=1.01 Coef.=25.36, t-stat.=1.75 Mobile technology index Mobile technology index Note: A cross-country regression of the number accounts per 1,000 adults on an index of mobile banking services (panel a) shows that the association is positive but insignificant (t-stat: 1.01). A cross-country regression of the ratio of private credit to GDP on an index of mobile banking services shows that the association is positive and significant (t-stat: 1.75). Source: Getting Finance database. the world. To bridge its great distances, the First National The future growth of mobile banking faces Bank has made a serious commitment to cell phone infrastructural, technological, and legal challenges. As banking. The bank acquired 23,000 subscribers within described by The Economist (February 7, 2008), “To go months of releasing its new service, most of whom high tech, you need to have gone medium tech first.��?25 already had an account with First National. To date, most The lack of reliable power supply is a major barrier. of the cell phone transactions services used have been for Motorola now provides free solar-powered charging the purchase of air time. kiosks to female entrepreneurs in Uganda, who use them to sell air time. Also being tested are wind- and solar- But most mobile payment models still operate at limited powered base stations in Namibia, aimed at reducing the scale. Kenya and South Africa have relatively major service cost of extending coverage to remote areas. In the absence providers among African countries. Yet WIZZIT in South of reliable power, users face problems charging their Africa, one of the better known examples, has only 50,000 phones. Many have to charge their phones at shops that clients (Ivatury and Pickens 2006; Donner 2007). Larger have diesel-fueled generators. In Bangladesh, Grameen scale examples are the operators of the Philippines, Bank is now equipping its “cell phone ladies��? (poor estimated to have 5.5 million clients (CGAP 2008). women who received a loan to buy a cell phone that they lend to villagers for a fee) with kits that contain long- Most users of cell phone banking are urban dwellers who lasting batteries—but the kits cost $150 each. are relatively well off. M-Pesa users in Kenya and Afghanistan appear to be less wealthy than their peers Issues in cash handling also pose a challenge to the wider - elsewhere but may well belong to the already banked. In use of cell phones for banking. In principle, a migrant a recent paper, CGAP (2008) estimates that of about a worker could send money home by texting his or her million mobile banking customers in South Africa, fewer family to collect the remittance from a local point of than 100,000 fall below South Africa’s poverty line. service in their village—for example, the grocer. The Globe Telecom’s GXI Inc., which offers the G-Cash migrant instructs the grocer to pay the remitted amount mobile wallet service in the Philippines, estimates that (possibly less a commission) using text messaging. If the nearly all of its 500,000 active users live in urban areas grocer and the emigrant have accounts in a common bank and thus are likely to be wealthier than rural clients. or are at least linked to a common clearinghouse, the 44 BANKING THE POOR grocer’s account is credited while the worker’s is debited. But this is not a likely scenario in most poor countries. Alternatively, as in Uganda, the worker may transfer air time up to the value of the remittance if the grocer and the worker are with the same cell phone provider. In practice, there remain the issues of security and the liquidity of the grocer. For instance, the population of San Juan in the Mixtec region of Mexico depends on remittances from younger family members in the United States. The nearest big town is Tlaxiaco, 35 kilometers away. Recipients must travel to Tlaxiaco to pick up their remittances, but the trip takes two to three hours and is a dangerous one since travelers are often robbed. Branchless banking has great potential for improving banking services—and possibly for attracting new clients. However, owing to the multiple impediments discussed in this chapter, banks in developing countries still have a way to go before mobile technology can become an integral part of banking services. 22 In a cross-country sample of 91 countries, Beck, Demirguc-Kunt, and Martinez Peria (2007b) suggest that wider branch and ATM penetra- tion is associated with a more even distribution of deposit and loan services among clients. 23 Djankov and others (2008) and Solo, Caskey, and Ruiz Duran (2006) for urban Mexico; Solo and Manroth (2006) for Colombia; and Kumar (2005) for Brazil show that income differentials largely explain bank- ing differences. 24 http://www.wizzit.co.za/. 25 Citing the World Bank’s Global Economic Prospects (2008b). BUILDING BRANCHLESS BANKING 45 6. Promoting access to banking In December 2005, India introduced a new type of bank cost services, with some restrictions on the menu of account designed for the poor. The Zero Balance account services and on the amounts held in accounts. For required no minimum balance. Application forms were example, basic accounts may enjoy free transactions up to simple, and few identity documents were needed. a specified limit, provided they are carried out using Moreover, account-maintenance charges were low, a small ATM cards and not through tellers. Fees and charges for overdraft facility was included, and the limits of free some transactions may be waived—subject to limits on transactions were plainly disclosed. Banks were urged to the number and volume of transactions. give wide publicity to the new no-frills accounts. Governments are also exploring new ways to promote India, like many developing countries, was seeking new savings. Many rich countries and some developing ways to increase access to banking. Until recently, countries are experimenting with matching schemes and extending credit to small clients was thought to be too tax-advantaged schemes. Savings methods that have costly and too risky for commercial banks. To support worked for microfinance—doorstep collection and small clients, India’s national government, and others, periodic-contribution programs (also known as created publicly owned development finance institutions. commitment programs)—are being offered by some In some cases, these institutions supplemented their banks in developing countries. credit activities by providing subsidies that allowed commercial banks to offer credit at a lower rate of interest Basic banking than would otherwise have been possible. Basic banking has long been offered in some countries— However, none of these initiatives significantly increased among them Canada, France, Sweden, and the United the numbers of bank accounts. Special programs were States. In the United States, a concept known as lifeline used largely by the rich or politically connected (Adams, banking was introduced in several states during the 1980s Graham, and Von Pischke 1984; Yaron, Benjamin, and and 1990s. For example, New York State mandates basic Piprek 1997). One study found lending to be correlated accounts in which initial deposits need not exceed $25, with election years.26 Development-finance institutions and the minimum balance is no more than $0.10. At least had high failure rates, and commercial banks handling eight free withdrawals are permitted, as well as unlimited large government-directed credit programs fared only free deposits. In the United Kingdom, basic bank slightly better (Caprio and others 2004). Cross-country accounts were launched at bank branches and post offices studies found that government ownership of banks and in 2003. They allow account holders to withdraw cash directed-credit programs was associated with slower through bank branches, ATMs, and post office counters. subsequent financial development, lower productivity, By November 2005, an estimated 1.52 million accounts and retarded economic growth.27 had been opened (British Bankers Association 2005). This experience prompted a search for other solutions to Other rich countries have less detailed provisions. In increase bank accounts. One measure introduced in Sweden, banks are not allowed to refuse applicants recent years is basic banking—a package of free or low- wishing to open a savings or deposit account. French PROMOTING ACCESS TO BANKING 47 citizens have a right to open a bank account, in the sense Employees of firms that have a payroll-deposit that if a person is refused an account by three banks, the arrangement with a bank need not maintain a minimum government will select a bank and require it to open an balance in their personal accounts with the bank. account. In Canada, all citizens have the right to open a However, when the employee leaves the firm, the bank personal bank account and to cash government checks may charge commissions on the account. free of charge at any bank. Even when not mandated by regulation, banks in some There has been no rigorous evaluation of the success of countries offer accounts with basic characteristics: no these schemes. A study of the effectiveness of lifeline opening fee, no maintenance fees, and no minimum- banking in the United States, based on a 1995 survey of balance requirements. In 16 countries of our sample, at consumer finances, points out that many commercial least half of the banks offer such accounts. This suggests banks already offered services as cheaply as lifeline that bankers also consider basic accounts to be lucrative, schemes (Doyle, Lopez, and Saidenberg 1998). That as they draw in new customers. finding has not stopped others from trying similar schemes. A growing group of developing countries, South Africa offers a good example of voluntary including 12 countries in our sample, is adopting the commitment to basic banking. The country’s Financial concept of basic banking.28 For example, Pakistan Charter of 2003 called for open access to banking; the introduced a basic-account regulation in November 2005 following year, the government encouraged banks to offer that includes a minimum initial deposit of Rs1000 so-called Mzansi accounts. Now offered by four major (around $17) for a transactions account. There is no South African banks, the accounts have no minimum maintenance fee and no minimum balance. Two free balance, no monthly maintenance fee, and a limited deposits and two free check withdrawals are allowed each number of free monthly deposits and withdrawals. In month. An unlimited number of withdrawals may be addition, ministatements are available by cell phone. made at no charge from the banks’ own ATMs and an annual statement of account is issued. If the account runs While basic banking has grown in popularity, a zero balance for six months, it will be closed. information about its impact is still limited. One assessment was done in India’s Gulbarga district, where More recent is Mexico’s basic banking regulation of July 400,000 new no-frills accounts were opened between 2007. It provides a list of minimum services that banks August 2006 and June 2007 during a government- must offer free of charge to anyone who meets the supported drive (Ramji 2007). The study found that requirements, and accounts must be below a certain size. most respondents (75 percent) opened accounts for the For such accounts there is no minimum opening amount, purpose of receiving government funds under the though banks may set their own requirements for an National Rural Employment Guarantee Program. Very ongoing minimum balance. The maximum balance is few persons opened accounts to save (4 percent) or to 165 times the daily minimum wage, above which the make transactions. Overall, the study found no large- bank may charge commissions for client transactions. scale impact on bank access. The minimum services include: opening and maintaining an account, providing a debit card, accepting free The evidence from South Africa is more positive. The deposits, and allowing free withdrawals and account voluntary code led to the opening of a million new accounts inquiries from the bank’s ATMs. Mexican regulations also in the first year alone, amounting to an additional 8.5 provide for employer-sponsored basic accounts. percent of total accounts, representing 4 percent of the 48 BANKING THE POOR population. More than 91 percent of the new account four account characteristics: no opening fee, no monthly holders were new to the banking system. Recent estimates of fee, no minimum balance, and free transaction bundles. the use of the Mzansi account suggest 3.5 to 4 million users, Such features are fairly common. For example, the Central of whom 60 percent are new to the banking system (Teschler African Republic is the only country in which banks do and Schneider 2008). But graduation from the Mzansi not provide an account without an opening fee. In just account to regular banking is difficult. Banks complain that four countries of the sample (Indonesia, Laos, Nepal, and these accounts are unprofitable, with relatively low use and Uganda) the majority of banks fail to offer any of the basic- high rates of dormancy and abandonment. account features. Our index of basic account availability is positively associated with accounts per thousand adults In 2003, regulators in the areas covered by the BCEAO (figure 6.1). Clearly, the practice of offering basic banking (Banque Centrale des Etats de l'Afrique de l'Ouest) matters more than regulations mandating such a practice. adopted the basic banking concept developed in France. All citizens now have a right to a bank account; if they are Encouraging savings denied by three commercial banks, the Central Bank can order one of the banks to provide services. But many poor Providing a way for poor people to save is an important people, unaware of complaint procedures, will probably aim of basic banking, and saving services are important to find it difficult to exercise that right. poor clients. Most of the accounts opened in our sample countries are savings accounts (76 percent, on average). Our examination of the links between basic banking and The share of savings accounts in total accounts increases access focused on regulations, on the one hand, and the as countries move away from poverty (figure 6.2). activities of commercial banks, on the other. Twelve countries, almost a quarter of our sample, reported At financial institutions, savings deposits provide funds having regulatory provisions for basic accounts. Among for credit operations and at the national level, increased them are some countries with sophisticated banking savings provide funds for investment. Governments offer systems—such as India, Mexico, Pakistan, and South incentives to expand savings because higher private Africa—as well as countries in Central and West Africa, savings imply less dependence on public savings for such as Burkina Faso, Chad, Mali, and Niger. health and education as well as for social security. Across the sample, we found no association between FIGURE 6.1 Where commercial banks offer regulations mandating a right to basic banking and accounts basic banking, more people use banks per thousand adults. One reason for the weak association Accounts per thousand adults and between basic banking regulations and bank accounts could country scores on index of basic banking be that while regulations in the BCEAO countries provide Malaysia Accounts per thousand adults for the right to a bank account, the procedures for exercising Mauritius Sri Lanka that right are not widely understood. A second reason is that, Cape Verde Thailand Burundi The Gambia India in all countries, the introduction of basic banking is relatively Nepal Madagascar recent—dating from 2003 or later. More time may have to Cameroon Zambia Philippines Sudan pass before the impact of basic banking is felt. Gabon Namibia Coef.=482.8, t-stat.=2.2 Next, we constructed an index of basic account provision Basic banking index at commercial banks. The index compiles the following Source: Getting Finance database. PROMOTING ACCESS TO BANKING 49 FIGURE 6.2 Savings accounts make up a greater share of total accounts in wealthier countries, and in Asia a. Share of savings accounts in total accounts, b. Share of savings accounts in total accounts, by region by country income quartile (GDP per capita) 100 100 75 75 (%) (%) 50 50 25 25 0 0 Q1 Very Low Q2 Low Q3 Medium Q4 High Africa East Asia Central South Asia America Source: Getting Finance database. We examined three savings vehicles offered by governments: helping poor people save (Sherraden 2006), they imply matched savings, tax-advantaged savings, and doorstep cash subsidies. Tax-advantaged savings schemes also collection. We also investigated periodic or commitment represent a cost to the government, in terms of tax savings, a product offered by commercial banks. revenues forgone. But many governments prefer to forgo tax revenue than to spend cash that they have collected. Only Singapore has a regulatory scheme for a matched savings product.29 In contrast, 22 countries in our sample The regressive implications of preferences for tax- offer tax incentives for savings (figure 6.3). Low-income advantaged schemes have been widely discussed. In the countries are less likely to offer tax incentives for savings; United States, for example, the distributive effects of such incentives are more prevalent in East Asia matched savings schemes, such as Individual Development (Indonesia, Philippines, Singapore) than elsewhere. Accounts, are compared with tax-advantaged retirement Matched savings plans require governments to make savings schemes.30 In developing countries, tax-advantaged contributions to private savings in some agreed savings schemes are most common in East Asia and are used proportion. Although they have been shown effective in less in Africa and South Asia (figure 6.3b). Around two- FIGURE 6.3 Tax-advantaged savings schemes are more common in wealthier countries, and in East Asia a. Countries offering tax incentives b. Percentage of countries offering for savings, by country income quartile (GDP per capita) tax incentives for savings, by region 1.0 100 Average countries' score 0.8 75 0.6 (%) 50 0.4 25 0.2 0.0 0 Q1 Very Low Q2 Low Q3 Medium Q4 High Africa East Asia Central South Asia America Note: Country scores on tax-advantaged savings measure the following five subcategories: the mandated saving period, ceilings on contributions, limits on access, contribution exemptions, withdrawal exemptions, and earned-interest exemptions. Source: Getting Finance database. 50 BANKING THE POOR FIGURE 6.4 Some common characteristics successfully by microfinance institutions to collect funds of tax-based savings schemes from small savers. Their convenience is so great, especially Frequency distribution (%) of six characteristics in remote areas, that small savers are prepared to put up of tax-based savings schemes in 22 countries with a negative rate of return.31 Banks in some countries are 100 attempting to adopt the practice to serve small depositors. 75 In Ghana, India, Indonesia, and the Philippines, banks have doorstep-collection schemes for low-income clients. Two (%) 50 countries—Indonesia and the Philippines—have adopted 25 guidelines to manage the risks of doorstep-collection 0 schemes practiced by nonbank agents. Mandated period Ceilings on contributions Limits on access Contribution exemptions Withdrawal exemptions Earned interest exemptions In seven countries—Bangladesh, Cape Verde, the Central Note: The 22 countries sampled are: Benin, Burkina Faso, Cameroon, Cape Verde, African Republic, Gabon, India, Senegal, and Central African Republic, Chad, Côte d’Ivoire, Gabon, El Salvador, India, Indonesia, Malaysia, Mali, Mexico, Niger, Philippines, Senegal, Singapore, South Africa, Thailand—all sampled banks offer commitment savings Sri Lanka, Thailand, and Vietnam. Source: Getting Finance database. products; in another ten, at least three out of five banks offer such products. However, there are exceptions. For fifths of South Asian and Central American countries offer instance, no banks in Cambodia, Laos, and Pakistan, and tax-advantaged savings schemes. Many poorer countries, none in seven African countries, offer commitment including Benin, Burkina Faso, Côte d’Ivoire, Indonesia, savings schemes. the Philippines, and Sri Lanka, offer such schemes. After tax incentives for savings, periodic savings schemes Tax-advantaged schemes typically have mandated are the most widely available savings device in our sample investment periods and restrictions on access to saved countries (figure 6.5). These allow savers to prevent funds. This makes them unsuitable for lower income family members from siphoning off savings. They also persons who face greater income fluctuation and need to have been shown to produce superior saving outcomes draw on savings to maintain consumption levels when among poor people than other types of savings plans income drops. The schemes take many forms, with tax exemptions applied during the contribution, growth, or FIGURE 6.5 withdrawal phases. Tax exemptions on interest earned are Tax-advantaged schemes are the most common and periodic savings schemes come next the most common benefit (figure 6.4). Thailand provides Percentage of countries offering tax-advantaged tax exemptions on interest earned on special fixed and periodic savings schemes, by region deposits of up to 20,000 baht, or $650; Malaysia’s tax 100 exemptions apply to special fixed deposits of up to % Countries offering 100,000 ringgit, around $25,000. 75 the scheme 50 Doorstep collection and 25 commitment savings 0 Africa East Asia Central South Asia America We examined two additional types of periodic savings Doorstep Tax-advantaged Commitment products: doorstep collection and commitment savings Source: Getting Finance database. products. Doorstep collection schemes have been used PROMOTING ACCESS TO BANKING 51 FIGURE 6.6 The prevalence of periodic savings schemes rises with country income Percentage of sampled banks in 30 countries offering periodic savings schemes 100 75 (%) 50 25 0 Malawi Liberia Sierra Leone Afghanistan Rwanda Madagascar Uganda Tanzania The Gambia Bangladesh Ghana Kenya Nigeria Pakistan India Nicaragua Cape Verde Honduras Sri Lanka Philippines Indonesia Swaziland El Salvador Guatemala Namibia Thailand Mauritius South Africa Malaysia Mexico GDP per capita lower higher Note: Data on periodic savings schemes were available from only 30 countries of the sample. Source: Getting Finance database. (Ashraf, Karlan, and Yin 2006a, 2006b). Periodic savings schemes are present in all regions (figure 6.6), and their prevalence increases with a country’s income. 26 See Khwaja and Mian (2005) on government banks in Pakistan. 27 For example, La Porta and others (2002), in a 92-country study. 28 Benin, Burkina Faso, Côte d'Ivoire, India, Malaysia, Mali, Mexico, Mozambique, Niger, Pakistan, Senegal, and Vietnam. Brazil, not included in our sample, introduced such legislation in 2003. 29 An example is the Post-Secondary Education Account (PSEA), to support investment in children’s further education. Starting in 2008, all the Singaporean citizens between the age of 17 and 20 have PSEA accounts opened for them and the government periodi- cally adds funds to the PSEA accounts of low-income families. 30 For example, Sherraden and Schreiner (2007). 31 Hirschland (2005) discusses the value of doorstep schemes. 52 BANKING THE POOR 7. Extending credit to entrepreneurs Business is booming for Farida, a young entrepreneur in additional capital. In Gabon, some banks are trying to Bangladesh. She needs to purchase more equipment and lower lending risks with innovative solutions such as hire two more employees to meet growing demand. Since group lending to small and medium enterprises, a she has no credit history she will need to apply for a concept similar to Grameen Bank’s schemes for women startup loan at the local branch of her commercial bank. entrepreneurs. If her application is approved, she will be able to expand her client base and compete with better-established Complex applications screen out potential customers businesses in Dhaka. who may be deterred by unnecessary and onerous requirements. We found that high complexity is Farida’s business is the kind of small firm that can grow associated with lower rates of use of the banking system and generate jobs. Many middle-class people in poor and lower levels of private credit. The efficiency of countries are entrepreneurs; most run small-scale application processing is important to firms, which may businesses with few assets. Family members supply labor; have short-term cash-flow needs or may need to exploit paid staff are few. opportunities with limited time horizons. But banks sometimes take a long time to evaluate applications—up Obtaining startup finance is not always simple or easy, to three weeks in Thailand, for example. When however, even for successful entrepreneurs. Business loan application processes take longer, firms report being more applications can be onerous, often requiring detailed financially constrained. financial projections and business plans. Proof of employment, letters of reference, or other documents Ratios of collateral to loan values are high in Africa and may be required in some countries, such as in Liberia and South Asia, ranging up to 150 percent in the Central Zambia. New entrepreneurs and small firms typically lack African Republic and 137 percent in Pakistan. However, the expertise to prepare detailed business plans. In higher values of collateral relative to loan size are not Botswana and Lesotho, professional consultants are hired associated with lower access to bank credit among firms. to prepare the business plans that are required as part of But when entrepreneurs can use business assets such as credit applications. The fees charged by these consultants equipment, accounts receivable, and inventory to secure can be as much as 1 percent of the requested loan loans they are less constrained. amount. In Sudan, the fee to apply for a startup loan can be up to 5 percent of the value of the loan. The lending environment is an important factor in increasing access to credit for entrepreneurs. Strong Loan terms typically are short and inflexible, making it creditors’ rights, a functioning credit information system, challenging to borrow for larger, longer-term and a competitive banking sector mean more finance for investments. In Rwanda and Vietnam, for example, the entrepreneurs. Unfortunately, these conditions are not longest term available for a startup is one year. Collateral met in many of the countries we sampled. requirements can present a further obstacle. When banks accept only houses or buildings as collateral, Still, there are some success stories. A bank in entrepreneurs are unable to leverage their assets to obtain Mozambique has begun to offer credit for salaried EXTENDING CREDIT TO ENTREPRENEURS 53 individuals who have worked for the same company for Microcredit helps, but can it at least a year. The minimum amount available is $500. provide jobs? People have up to five years to repay the loan, at 25 percent interest. An additional 1.5 percent is charged The enterprising poor are hobbled by a lack of credit. with each delayed payment. People use this product not Banks will not lend to them because they have few assets only to buy consumer goods, such as bicycles and to post as collateral and because banks find it expensive washing machines, but also to start income generating to screen and monitor small-scale borrowers. In response activities. The bank has about 12,000 clients who use this to the dearth of credit, microcredit programs have taken mass credit. off in developing countries over the past decade. Grameen Bank in Bangladesh and Bank Rayat in Poor people as entrepreneurs Indonesia are leading examples. As of December 2005, 3,133 such institutions were reported to be lending to Many poor people in low-income countries are more than 110 million people (Daley-Harris 2006). entrepreneurs. In Indonesia, Nicaragua, and Pakistan, Microcredit has made a difference to the lives of many about half of the urban poor run their own business, but beneficiaries, especially women. A study from Bangladesh they are usually not specialized. In fact, poor households finds that annual household consumption increases by tend to have multiple occupations. Consider the typical 18 percent of the amount borrowed by women (Pitt and case of a poor slum inhabitant from Guntur, a town in Khandker 1998). India. In the morning, she sells dosas and pancakes made of rice, dal, or beans on the roadside. In the afternoons, Microcredit helps the poor smooth consumption and she makes and sells saris, traditional attire for Indian build assets. However, it plays only a modest role in women. Almost half of the poor urban households in creating new jobs. Often the loans are used for household Côte d’Ivoire and Indonesia derive their income from consumption and not invested in household businesses. more than one source. Likewise, the rural poor do more In Indonesia, for instance, low-income households use than just work on their own land; they often work as microcredit for consumption purposes about 30 percent nonagricultural labor. For instance, in Pakistan, 51 of the time. Even households with enterprises use loans percent of very poor rural households earn some income for consumption purposes such as school fees, medical from supplying labor to nonagricultural firms, and 35 needs, and social and holiday spending (Johnston and percent run a nonagricultural business. Murdoch 2007). So the poor are enterprising, but their businesses operate Also, the beneficiaries of microcredit programs are not on a small scale. Most use family members as labor and the poorest of the poor. In northeast Thailand, paid staff are rare. Only one in ten businesses run by the microfinance schemes run by nongovernmental urban poor in Côte d’Ivoire employ a paid staff member. organizations that explicitly target the very poor tend to These businesses also lack tangible assets. In Pakistan, operate in the wealthier villages. Even within a given only 4 percent of such businesses had a motor vehicle, village, the schemes lend more to wealthier households and none had any machinery. The overall picture is one (Coleman 2006). Similarly in Indonesia, Bank Rayat of penniless entrepreneurs—small in scale, unskilled, and borrowers have incomes that are 40 percent higher than with little capital. those of nonborrowing households in the target group (Hulme and Mosley 1996). 54 BANKING THE POOR In low-income countries, the middle-class consists largely and near villages. Between 1980 and 1999, rural factory of people who hold steady salaried jobs. This is what employment multiplied tenfold. Those factories employed distinguishes them from the poor. In urban Indonesia, mostly unskilled labor. The growth in employment was for example, only about 38 percent of the very poor, twice as effective in boosting rural income as were gains in whose daily per capita spending is less than $1 a day, hold agricultural productivity (Foster and Rosenzweig 2004). a salaried job.32 In contrast, 77 percent of the lower- middle class, those with daily per capita spending of Families with one or more salaried earners are much more between $2 and $4, hold a salaried job. For those in the likely to use banking services. In India, 91 percent of the upper-middle class—whose daily per capita spending is households in which the chief earner holds a salaried job are between $6 and $10—that proportion exceeds 87 banked, in contrast to 38 percent of those in which the chief percent. earner is a daily wage laborer (figure 7.1). More than 58 percent of daily-wage households hold their cash at home, Needless to say, the middle class has its entrepreneurs. compared with 20 percent of the salaried. Conversely, 68 However, the average middle-class entrepreneur runs a percent of the salaried households save in bank deposits, small business with few, if any, paid employees—as few as whereas only 25 percent of daily-wage families do so. 0.2 paid staff per business in Côte d'Ivoire. The businesses have few assets and do not require much in Those with formal jobs that pay a regular salary not only way of skills. For instance, in Hyderabad, India, lower- have more income; they also have longer planning horizons middle-class households run general stores, phone and are more confident about the stability of future income, booths, tailor shops, and fruit-and-vegetable stands. making a bank account seem worthwhile. Their enterprises are as undercapitalized as those run by the poor. Yet they have better access to bank credit. For Credit to firms boosts jobs example, banks in urban Indonesia account for just 23 percent of the loans taken out by very poor households Growing businesses generate jobs. But credit constraints but 74 percent of the loans contracted by upper-middle- hamper firm growth, as a study from India finds (Foster and class households. Of course, much of credit may be tied Rosenzweig 2004). The researchers used the following to purchases of consumer durables and not be available reasoning: if firms become eligible for directed credit at for investment. Yet the middle class generates enough income to be able to save. They buy assets such as FIGURE 7.1 The salaried are more likely to bank televisions, radios, and larger houses. Their businesses, Percentage of households holding accounts meanwhile, tend to remain undercapitalized—savings are in financial institutions, by source of income not invested in growing the business. Regular salary 90.8 For poor people who aspire to increase their income and Self-employed 84 (nonagriculture) enter the middle class, entrepreneurship is not the most Self-employed 67.7 (agriculture) common route. The easiest way to middle-class status is a Daily wage labor 38.4 secure, well-paid job. The jobs provided by a new factory, for example, can stimulate investments in children’s health 0 25 50 75 100 and education, and those children in turn can move up the % Households income ladder. In India, an investment policy favoring rural Source: Shukla 2007. development led to the rapid growth of factories located in EXTENDING CREDIT TO ENTREPRENEURS 55 subsidized, below-market rates of interest, then both capital (Banerjee and Munshi 2004). Tirupur has been constrained and unconstrained firms would seek to use it. traditionally dominated by Gounders, a Tamil-speaking But the two types of firms would use it differently. If a firm people of southern India. In recent years, people from other is credit-constrained, then it will use the directed credit to parts of India have set up shops in Tirupur. New Gounder expand production. If a firm is unconstrained, then it will firms are set up with almost three times as much fixed replace existing market loans with the directed credit. It will capital as comparable outside firms. Gounder firms also stay not expand production. When additional firms become much more capital-intensive than outsiders’ firms. eligible for directed credit, they expand production, and revenue and employment rise. The balance of this chapter presents a new analysis of requirements for business startup loans across countries. If firms are operating in a credit-constrained environment, We examine the association between those requirements easing access to credit can greatly affect revenue and and firms’ access to bank credit. employment growth. A sample of East European firms was surveyed in 2002 and again in 2005. We find that some of Business-loan applications: the firms gained access to bank credit for investment and how simple? working capital during that period. These firms registered 9 percent higher growth in employment between 2002 and Entrepreneurs, particularly small entrepreneurs, want to 2005 compared with the firms that had no change in access access credit quickly, without lengthy application to bank credit. Moreover, revenue growth was higher by procedures. But do they succeed in doing so in our sample 36 percent. The effect on employment was even greater for countries? To answer that question, we constructed an small firms with less than 20 employees. Small firms that index that captures four dimensions of complexity: gained access to bank credit for new investments registered whether the borrower is required to have an account with 20 percent higher employment growth. the bank, whether a letter of reference is required, whether the borrower needs to provide proof of employment, and Another example of large variation in access to credit comes whether there is a fee to process the application. The index from the knitted-garment industry in Tirupur, South India. ranges from 0 to 1, and the average score in our sample of Here, community ties mediate large differences in access to countries is 0.74 (figure 7.2). FIGURE 7.2 Applying for a business loan is much simpler in some countries than in others Country performance on index of business-loan complexity 1.0 0.8 Central African Republic 0.6 Congo, Dem. Rep. of 0.4 Mozambique Burkina Faso Sierra Leone South Africa Afghanistan Madagascar The Gambia Bangladesh Cape Verde El Salvador Philippines Cameroon Nicaragua Cambodia Swaziland Singapore Botswana Indonesia Honduras Mauritius Sri Lanka Tanzania Thailand Malaysia Vietnam Pakistan Namibia Ethiopia Lao PDR Rwanda Burundi Lesotho Senegal 0.2 Malawi Zambia Nigeria Angola Liberia Gabon Ghana Kenya Sudan Nepal Benin Niger Chad India Mali 0 Index of business loan complexity complex simple Source: Getting Finance database. 56 BANKING THE POOR FIGURE 7.3 Complexity of loan applications limits access especially for small firms a. Prevalence of working-capital loans, and complexity b. Prevalence of working-capital loans and complexity of business-loan procedures, all firms of business-loan procedures, small and micro firms Burkina Faso with loansfor working capital Burkina Faso % of firms with bank loans % of small and micro firms for working capital Thailand Ethiopia Vietnam Thailand Ethiopia KenyaEl Salvador Vietnam Burundi Niger Zambia Senegal Niger Botswana Ghana Botswana Angola Angola Coef.=29, t-stat.=-2 Coef.=31, t-stat:-2.11 Complexity of business-loan procedures Complexity of business-loan procedures Note: In panel a, a cross-country regression of the proportion of firms using bank credit for working capital on the index of complexity of business-loan procedures, controlling for income and the legal rights environment, shows that a 0.1 increase in the complexity index is associated with a 2.9 percent point drop in firms accessing credit for working capital (t-stat: –2). A similar regression for small firms (less than 50 employees) shows, in panel b, that a 0.1 increase in the index is associated with a 3.1 percent drop in small firms accessing credit (t-stat: –2.11). Source: Getting Finance database. We found that bank access varied inversely with the index, Another dimension of the loan application process is the even after controlling for a country’s income and legal efficiency with which applications are processed. On environment (figure 7.3). The result also holds for the average, it takes just over a week (7.5 days) to process a amount of private credit available. For instance, Ethiopia business startup loan in the sample countries (figure 7.4), and Sierra Leone have similar GDP per capita, but with Botswana and South Africa having the fastest Ethiopia’s loan procedures are much less complicated than processing times (1.5 and 2 days respectively), compared those in Sierra Leone (which has the maximum possible with a high of 17.5 days in Liberia and 21 days in index score of 1). Correspondingly, Ethiopia has a much Thailand. Viewed regionally, loan applications take the higher ratio of private credit to GDP (27 percent) than does longest to process in Central America, where the average Sierra Leone (just over 4 percent). The association between is almost 15 days, while African banks are the quickest, complexity and bank access also holds for small enterprises, processing applications in an average of just over 6 days. with a correlation that is stronger and more robust. FIGURE 7.4 Countries differ in the efficiency with which their banks process loans Days required to process a business startup loan, by country 25 Business loan processing time - 20 Central African Republic Congo, Dem. Rep. of 15 The Gambia Mozambique Burkina Faso Cape Verde Lao PDR Vietnam Ethiopia Philippines Singapore Swaziland Senegal Sierra Leone Mauritius Malawi Afghanistan 10 Kenya South Africa Ghana Indonesia Zambia Benin Malaysia Uganda India Madagascar Bangladesh Botswana Sudan Guatemala El Salvador Lesotho Chad Cameroon Nicaragua Cambodia Honduras Sri Lanka Tanzania Thailand Pakistan Namibia Rwanda Burundi 5 Angola Nigeria Liberia Gabon Nepal Niger Mali 0 Source: Getting Finance database. EXTENDING CREDIT TO ENTREPRENEURS 57 There are positive and significant associations between at 150 percent of the value of the loan. South Africa has loan-processing time and various dimensions of access to the lowest, at less than 15 percent. investment finance.33 The associations hold for small and micro firms as well as large and medium-sized enterprises. The value of the collateral required to secure a loan is not Longer processing times are also associated with more associated with lower bank access for firms, but the type firms reporting access to finance as a major constraint. of collateral that is acceptable to a bank does appear to One explanation is that banks in low-income countries affect access. We constructed an index of acceptable may have fewer automated systems to evaluate business collateral for a business startup loan that captures the applications, particularly loans for startup businesses. range of assets that a bank might consider as valid Applications have to be reviewed more carefully by loan collateral to secure a business startup loan. The assets officers and may require more face-to-face screening or include immovable property (such as land, a house, or a other labor-intensive evaluation processes. building), movable property (such as inventory and equipment), and liquid assets (cash). Higher scores How much does it cost to apply for startup loans? The fees indicate a wider range of acceptable collateral (figure 7.6). vary considerably across countries, with banks in Uganda charging as much as 3 percent of the loan value, while Where banks accept as collateral only titled houses, land, banks in Vietnam charge only 0.01 percent of loan value or other immovable property, many businesses are excluded (figure 7.5). In general, fees are considerably higher in from credit markets. However, these same businesses may Africa and Central America than in the two Asian regions. have an array of productive assets that could be harnessed to serve as collateral. When businesses can use productive Terms and conditions business assets, such as inventory, accounts receivable, or for startup loans equipment, as means to secure additional capital, financing constraints are lower (figure 7.7a). This pattern also holds The terms available for business startup loans vary for small and micro firms (figure 7.7b). considerably across countries. They are shortest in the Democratic Republic of the Congo, Rwanda, and Regulations that Vietnam—at 1.5, 1, and 1 years, respectively. At the can increase access other extreme, in Honduras, banks offer 10-year startup loans. (Banks in Central America offer longer terms, on Banks’ loan procedures, terms, and conditions matter for average.) The length of a bank loan does not appear in business borrowers, but the broader institutional measures of access, but it does affect the overall depth of environment for lending matters for creditors. The the credit market. Loan terms are significantly and soundness of that environment therefore affects access to positively associated with the ratio of private credit to finance. The degree to which creditors can credibly GDP, but not with measures of bank access for firms. enforce contracts and the availability of information on current and prospective borrowers both play important Collateral requirements are relatively high in our sample, roles in reducing credit risks for banks. Better legal with countries in South Asia and Africa having higher protections enable lenders to offer entrepreneurs money requirements (on average, about 90 percent of loan value) on better terms, and creditors’ rights are associated with than those in Central America and East Asia (on average, higher ratios of private credit to GDP. Stronger creditors’ 75 percent of loan value). Banks in the Central African rights are also associated with longer loan terms and lower Republic have the highest average collateral requirement, interest rates.34 We found that the strength of legal 58 BANKING THE POOR FIGURE 7.5 The cost of applying for a loan to start a business Application fees for business startup loan as a percentage of loan value, by country 2.5 2.0 Congo, Dem. Rep. of Mozambique 1.5 South Africa Madagascar Cambodia Mauritius Afghanistan Lesotho Nepal Burkina Faso Senegal 1.0 Lao PDR Benin Angola Ethiopia Honduras Bangladesh India Sierra Leone The Gambia Cape Verde Pakistan Sri Lanka Philippines El Salvador Cameroon Nicaragua Swaziland Botswana Indonesia Chad Singapore Namibia Tanzania Thailand Rwanda Burundi Malaysia Vietnam 0.5 Malawi Nigeria Liberia Gabon Ghana Sudan Niger Mali 0.0 Application fees for business start-up loans higher fees lower fees Source: Getting Finance database. protections for creditors is associated with higher ratios of borrowers (Jappelli and Pagano 2002b; Padilla and private credit to GDP, higher shares of bank borrowing in Pagano 1997; Jappelli and Pagano 1993). An abundance firms’ financing structure, and higher percentages of firms of empirical evidence supports the links between credit with access to loans from commercial banks. information and financial access. The presence of credit registries is associated with a higher ratio of private credit Systems that facilitate the sharing of credit information to GDP. In countries with no registries, firms perceive are equally important. Well-functioning credit higher financing constraints; while in countries where information systems improve banks’ knowledge of private credit registries are present, firms perceive lower applicants’ characteristics and permit more accurate financing constraints and have a higher share of bank predictions of the probability of repayment. Credit borrowing in their financing structure (Djankov, information also increases repayment incentives for McLeish, and Shleifer 2007). FIGURE 7.6 Banks’ flexibility concerning collateral for business startup loans Average scores of sample banks on collateral flexibility index, by country 1 Collateral acceptability index 0.8 0.6 Congo, Dem. Rep. of 0.4 Mozambique Burkina Faso Sierra Leone South Africa The Gambia Bangladesh El Salvador Philippines Cameroon Cambodia Swaziland Singapore Botswana Mauritius Sri Lanka Tanzania Malaysia Thailand Vietnam Pakistan Ethiopia Lao PDR Burundi Senegal Lesotho 0.2 Malawi Zambia Angola Nigeria Sudan Kenya Nepal Niger Chad India Mali 0 Collateral flexibility index less flexible more flexible Note: The collateral flexibility index is an equally weighted index of the sum of responses concerning the acceptability of various types of assets as collateral for a business startup loan. The asset types are: cash and liquid assets, land, house with title, house without title, equipment, livestock, inventory, accounts receivable, and other. Higher scores indicate a wider range of acceptable collateral. Source: Getting Finance database. EXTENDING CREDIT TO ENTREPRENEURS 59 FIGURE 7.7 Use of credit increases with the range of assets accepted as collateral a. Percentage of working capital obtained from bank b. Percentage of working capital obtained from bank loans and collateral flexibility, all firms loans and collateral flexibility, micro and small firms (fewer than 50 employees) % of working capital from bank loans Thailand Thailand % of working capital from Mauritius Mauritius Kenya Ethiopia El Salvador Vietnam Ethiopia bank loans Kenya El Salvador Senegal South Africa Burundi Niger Malawi Philippines South Africa Zambia The Gambia Philippines Tanzania Swaziland Botswana Angola Swaziland Angola Botswana Coef.=33.4, t-stat.=1.78 Coef.=28.17, t-stat.=1.9 Collateral flexibility index Collateral flexibility index Note: A cross-country regression of the percentage of firms with working capital funded by bank loans on the index of acceptable collateral (panel a) shows that a 10 percent increase in the index is associated with a 3.3 percent increase in firms using bank loans to fund working capital (t-stat: 1.78). A cross-country regression of the percentage of small firms with working capital funded by bank loans on the index of acceptable collateral (panel b) shows that a 10 percent increase in the index is associated with a 2.8 percent increase in small firms using bank loans to fund working capital (t-stat: 1.9). Source: Getting Finance database. Our results corroborate the earlier findings. The use of In the sample of 54 economies studied here, competition credit registries (public or private) is associated across among banks is significantly and positively associated countries with more private credit to GDP and a larger with domestic private credit, but not with the number of percentage of firms with access to commercial bank loans. bank accounts per 1,000 adults. In other words, it The presence of credit registries is associated with lower increases private credit but not bank access. financing constraints for small and micro firms as well. 32 Household surveys typically lump daily and casual workers with (This makes sense, because small and micro firms are the salaried employees in one category—wage workers. But casual workers have little or no job security, work fluctuating hours, and often move most opaque.) from job to job. They are usually paid by the hour or at the end of the day. Salaried employees will be in relatively secure, stable jobs that pay on a weekly or monthly basis. Frequency of payment is a useful proxy for job security. Finally, the level of competition in the banking sector, 33 We measure access to investment finance in two ways: first, the average share of investment financing obtained through commercial measured by the top three banks’ share of banking sector bank loans, and, second, the average share of firms with access to assets, also affects financial access for enterprises, investment finance from commercial banks. 34 See Djankov, McLeish, and Shleifer (2007); Safavian and Sharma particularly in low-income countries and countries with (2007); Qian and Strahan (2005); Jappelli and Pagano (2002a); La Porta and others (1997). underdeveloped financial and institutional infrastructure.35 35 In high-income countries, or those with well-developed institutions, the empirical evidence is more mixed. See Beck, Demirguc-Kunt, Competition in these countries has been associated with and Maksimovic (2004); Bonaccorsi di Patti and Dell’Ariccia (2003); Black and Strahan (2002); Berger and others (1998); Peterson and more credit for the private sector, and lower levels of Rajan (1995). collateral requirements. This is because banks’ scope for - extracting rent through collateralization decreases with competitive pressure and leads to a higher share of domestic credit being made available to enterprises (Beck, Demirguc-Kunt, and Maksimovic 2004; Hainz 2003). 60 BANKING THE POOR 8. Increasing disclosure Better disclosure of the procedures and fees involved in measured in terms of the legal rights accorded to creditors opening and maintaining bank accounts is associated with and borrowers. more people being banked. Although many countries recognize the importance of better disclosure, there have Guidelines and support been few analyses of its beneficial impact. Stipulations on for credit applications disclosure are often embedded in financial regulations pertaining to access to banking. In the area of basic banking, First, we examine existing regulations pertaining to the type U.S. Lifeline Banking laws call for the obligatory disclosure and number of documents required to obtain credit. We of charges assessed after the customer exceeds the limited measure the extent of this regulation using an index of credit- number of free transactions. Pakistan’s basic banking application guidelines, which includes guidelines for regulation has similar provisions. In India, basic banking prospective borrowers on proof of identity, tax identification, regulations require banks to publicize the plans they offer. credit history, and the need for personal guarantors. We found a strong positive association between our constructed In the area of credit, a truth-in-lending law has existed in the variable and the overall percentage of firms with bank loans United States since 1968. The U.S. Community Reinvest- in a given country. Countries such as Bangladesh, Cameroon, ment Act of 1977 protects against discriminatory lending Niger, and Rwanda have highly regulated credit application practices, and at least 25 U.S. states also have regulations requirements and procedures. In these countries, the per- against predatory lending. Similarly, Australia adopted a centage of firms with bank loans is comparatively high. But Uniform Consumer Credit Code in 1993 and the European in Ghana, Malawi, and Tanzania, regulations on credit app- Community issued its first directive for consumer protection lications are absent, and few small businesses have bank loans. in credit markets in 1986. There is also a strong positive association between our Our analysis reveals that greater disclosure of loan processes variable and the percentage of firms that use loans to is associated with a higher density of bank accounts. These finance working capital (figure 8.1). In Benin, Cape associations are significant after controlling for differences Verde, and Uganda, many small businesses use bank in income per capita and the overall credit environment, loans to finance working capital, while in countries in FIGURE 8.1 Small business loans are more prevalent in countries with credit application guidelines a. Correlation between index of government-regulated b. Correlation between index of government-regulated credit credit guidelines and percentage of firms using bank loans guidelines and percentage of firms using working capital loans Thailand Mauritius Bangladesh Honduras % of firms % of firms Cameroon South Africa Rwanda Nicaragua El Salvador Malawi Namibia Kenya Benin Burundi Malawi Ghana Mali Ghana Botswana Swaziland Uganda The Gambia Guatemala Mexico Lesotho Cape Verde Tanzania The Gambia Uganda Pakistan Angola Mexico Angola Coef.=21.30, t-stat.=2.83 Coef.=16.76, t-stat.=3.2 Index of credit-application guidelines Index of credit-application guidelines Source: Getting Finance database. INCREASING DISCLOSURE 61 which credit applications are unregulated or only loosely whether banks are required to advise applicants of credit regulated (El Salvador, Ghana, and Lesotho, for decisions is strongly positive (figure 8.2a). For instance, example), a smaller proportion of firms use bank loans Benin, Botswana, and Senegal have regulations on to finance working capital. advising applicants of credit denials and of any unfavorable changes in loan terms. The percentage of Changes in credit terms small businesses with bank loans in these countries is and denial of credit higher than in Angola, Ghana, Malawi, and Kenya, where such regulations are absent. The implication is that We focused next on the degree to which information on entrepreneurs are more likely to initiate a loan application the credit application process and on the outcome of if they are reasonably sure that they will be notified of applying for credit is associated with access to banking. the outcome. Our index of disclosure of credit decisions measures the extent to which central banks have adopted guidelines Does disclosure help? that require banks to advise clients of unfavorable changes in credit terms and of denials of credit. The final piece of our analysis assesses the relationship between overall transparency in banking procedures and In 60 percent of the countries in our sample, regulators bank access. It is based on an index with 12 components require banks to advise the account holder of a change in that measure the extent of national regulations that terms that may be considered unfavorable. However, 14 pertain to public disclosure of banking practices.36 Most of 32 African counties surveyed do not have such countries have adopted some disclosure requirements to requirements. Nor do Cambodia or the Lao People’s protect consumers, some more than others. For example, Democratic Republic in East Asia or Afghanistan and 89 percent of the countries we sampled have regulations Bangladesh in South Asia. Guidelines for advising on disclosing interest rates and other fees. Thirty-six applicants when credit is denied are even less common. percent of countries require that banks employ an ombudsman to deal with customer complaints. The three We found a weak positive association between bank countries that lack any sort of disclosure regulation are access and better information on loan processes (figure Lesotho, Malawi, and Rwanda. The countries that have 8.2b). By contrast, the association between access and the most extensive disclosure requirements are India, Sri FIGURE 8.2 Small business loans are more prevalent in countries with better disclosure of credit decision outcomes a. Correlation between percentage of firms using bank loans and b. Correlation between density of bank accounts and whether whether banks are required to advise applicants of credit decisions banks are required to advise applicants of credit decisions Thailand % Firms using bank loans to invest Malaysia Accounts per thousand adults Mauritius Sri Lanka Cape Verde Thailand India Mauritius Vietnam Malawi Singapore Guatemala Kenya Tanzania Malawi Senegal Zambia Ghana Cambodia Pakistan Senegal Nicaragua Tanzania Pakistan Guatemala Cameroon Angola Mexico Laos Sudan Coef.=5.87, t-stat.=2.34 Coef.=139.2, t-stat.=1.04 Index of disclosure of credit decisions Index of disclosure of credit decisions Source: Getting Finance database. 62 BANKING THE POOR FIGURE 8.3 Lanka, and Vietnam. Among the 13 countries in the Firms in countries with better disclosure on banking lowest income quartile, the highest degree of transparency terms are less likely to see access to finance as a problem is found in Madagascar and Uganda. Correlation of firm perceptions of finance as an obstacle to doing business and country performance on index of national banking disclosure regulations India and Sri Lanka each have positive responses for 10 of Benin access to finance as an obstacle Ghana Cameroon 12 items in the transparency index. Also, both India and % Firms that view Kenya Sri Lanka have high levels of access relative to their income Malawi Burundi Uganda per capita (figure 8.3). In Africa, Uganda has the highest Lesotho Pakistan Swaziland Honduras score on the index of transparency and consumer India protection, with seven positive responses; Madagascar has Mexico Vietnam Cambodia six, and Cape Verde has five. Conversely, in Malawi and Coef.=-12.75, t-stat.=-2.11 Rwanda, which both have a transparency index of zero, Index of national banking disclosure regulations many firms view access to finance as a constraint. Note: A cross-country regression of the percentage of firms that view access to finance Entrepreneurs’ perceptions of finance as a constraint as an obstacle on our index of transparency and consumer protection reveals that a 10 percent increase in the index is associated with a 1.3 percent point drop in the decrease significantly as countries’ scores on the index rise. proportion of firms that view finance as an obstacle (t-stat: –2.11.) Source: Getting Finance database. 36 The components ask whether or not (i) whether the central bank pro- vides public information on fees and charges on commercial banking transactions; (ii) whether banks disclose interest rates and charges to customers; (iii) whether banks disclose the interest rates applied to overdrawn accounts; (iv) whether banks provide printed information on charges applied to savings accounts and credit arrangements; (v) - whether banks read or explain terms to illiterate customers; (vi) whether banks provide information in local or regional languages in addition to the official national language; (vii) whether banks have internal procedures for customer complaints; (viii) whether banks are subject to regulations for maintaining records of complaints for a specified period; (ix) whether a consumer protection law applicable to bank customers is in force; (x) whether an ombudsman is in place to deal with customer complaints; (xi) whether the ombudsman’s deci- sions are binding on the parties; and (xii) whether rules or guidelines are in place to support advisory services for certain categories of clients (such as women entrepreneurs and first-time borrowers). INCREASING DISCLOSURE 63 9. Data sources and methodology This study is based on data from 54 countries. The core of Constructing a measure the sample are poor countries in Africa (35 countries) and of access as an outcome variable Asia (12), along with 4 Central American countries. Malaysia, Mexico and Singapore were added as comparators. The analysis in this study is based on the association between access to banking services, as measured by the We designed two questionnaires for collecting data. One number of bank accounts per thousand adults in each was designed to be administered to the top five country, and several other factors. Those factors are commercial banks in each country; the other to central transactions offered at banks, or required by banks, and banks and financial regulatory agencies. The goal was to regulations adopted by country authorities that may capture information from commercial banks on the affect banking access. Our principal outcome variable, transactional aspects of bank access and to examine access to banking services, is based on information on the obstacles to banking from the regulatory perspective. The total numbers of bank accounts per country provided by regulators’ questionnaire also collects basic statistical regulators or central banks. Such data were available for information on access to banking in each country. 45 countries. Data on another nine countries were added by extrapolation from the figures on numbers of accounts We collected information from commercial banks in two provided by the commercial banks sampled in those ways. The first was through local consultants who knew countries, using information from Bankscope on the the countries’ financial systems and had contacts in the shares of the sampled banks in the assets of the national banking system. The second was through direct financial system. Data on the numbers of bank accounts communication with institutions via electronic mail, fax per thousand adults are computed from numbers of total and telephone, undertaken by World Bank staff. accounts and World Bank population data, notably the Information from central banks was collected only by proportion of the population aged 15 and older. World Bank staff. Of 282 surveys sent to commercial banks, 235 were returned, an overall response rate of 82 We recognize that our data on the numbers of accounts percent (table 9.1). In the case of central banks and may exceed the total numbers of the banked, as some regulators, responses were obtained for 52 of the 54 people may have multiple accounts (though in some countries solicited (a 96 percent overall response rate). countries net numbers are clearly presented). It is difficult to separate business and individual accounts, and to make Each questionnaire contained questions pertaining to adjustments for accounts that have fallen into disuse. three aspects of financial services: opening and using a Some countries systematically close inactive accounts bank account, savings services, and credit services. The after a defined period of inactivity; others allow inactive commercial bank questionnaire also included a fourth accounts to remain open indefinitely. We hope that ways section on payments and remittance services. to overcome these shortcomings can be found in time; for the moment, it is the best measure of access that we Questions about payment services were not included in were able to devise. the regulators’ questionnaire, as information on the regulation of retail payment services was being collected Our access measures correlate strongly with several other in a parallel exercise (World Bank 2008c). access measures used in the literature. FinScope has DATA SOURCES AND METHODOLOGY 65 produced estimates, based on household surveys, of the clients per institution, for 46 of the 54 countries of our proportion of households with access to banking in some sample.38 By adding these to the numbers of bank clients, African countries. The correlation between our measure as reported by central banks, we were able to estimate the and the FinScope measure across five countries (Kenya, proportion of people who use banks.39 Banks serve more Namibia, South Africa, Tanzania, and Uganda) is 0.7. than 90 percent of all those who consume financial services Moreover, Finance for All? (in table A.1) provides a in 25 of the 46 countries, and more than 80 percent in 34 composite measure of access to financial services—the of 46 common countries. Banks therefore capture the vast percentage of adult population that has an account with a majority of people who use formal financial services. The financial intermediary.37 For the 51 overlapping countries, only countries in which bank coverage falls below 50 the correlation between the Finance for All? measure and percent are Afghanistan, Cambodia, and Nicaragua. the measure used here is 0.82. Nonbank microfinance institutions in those countries appear to serve a relatively large number of people.40 To what extent does our measure of financial access, a measure based on the banking system alone, capture the full Our analysis of terms and conditions in business startup extent to which persons are served by financial institutions? loans employs outcome variables constructed from World To estimate the relative importance of banks, as compared Bank enterprise surveys. These surveys collect data from with microfinance institutions and other nonbank or firms on their operations, and on their perceptions of the nongovernmental providers of financial services, we business environment. The respondents are managing assembled central-bank data on numbers of account holders directors, accountants, human resource managers, and at banks in each of our countries. We then compared those other staff. Among the data collected are measures of the data with the numbers of clients of nonbank institutions in fraction of working capital and new investment financed each of these countries, using data from the Microfinance by bank credit, and firm perceptions of the degree of Information Exchange (MIX) database (figure 9.1). difficulty in accessing finance. Between 2002 and 2007, the World Bank conducted enterprise surveys in 38 of The MIX database contains cross-country data on the countries sampled here. Sample sizes ranged from 98 microfinance providers, including data on numbers of to 1,480 firms, with an average being 445. FIGURE 9.1 Banks serve the majority of financial service clients substantially exceeding microfinance Percentage of all consumers of financial services who are served by banks 100 80 60 40 20 0 Cambodia Afghanistan Nicaragua Bangladesh Congo, Dem. Rep. of Ethiopia Benin Senegal Malawi Chad Mali Madagascar Sierra Leone Cameroon Niger Nepal Burkina Faso Kenya Mexico Sri Lanka Philippines Uganda Ghana Honduras Tanzania Pakistan El Salvador Rwanda Vietnam Côte d’Ivoire Guatemala Liberia Nigeria Zambia Swaziland India Mozambique South Africa Burundi Indonesia Angola The Gambia Sudan Namibia Thailand Gabon Source: Microfinance Information Exchange database (http://www.mixmarket.org/) and Getting Finance database. 66 BANKING THE POOR Explanatory variables—data economic development, as represented by income per from commercial banks and capita. We use the literacy rate as a proxy for financial from regulators sophistication and ability to deal with formal institutions. Population density and urbanization are used as proxies The survey of commercial banks targeted the top five for remoteness. The Gini coefficient, which measures banks in each country, based on banking assets. Lower- income inequality and may also affect overall access, ranking banks were substituted if the largest ones would have been used as a control if information on this declined to participate. Asset shares were computed from variable had not been unavailable for too many countries the latest available Bankscope data. The commercial bank in our sample. sample consists of 235 banks in 54 countries (table 9.1). The World Bank’s annual Doing Business report publishes The largest banks in terms of assets may not necessarily a legal rights index and a creditors’ information index for be those with the most accounts. In particular, some 181 countries. These are used as controls when analyzing institutions, such as specialized savings banks or postal the effects of the terms and conditions of business startup banks, may have a large number of accounts but a loans on firm-level access to bank credit. relatively small share of national assets because many of their accounts are very small.41 Nevertheless, we had two Methodology reasons for defining our sample using asset shares. First, bank-level data on the number of accounts are not We tested our hypotheses using cross-country regressions available on a cross-country basis, whereas data on asset of bank access on measures of the cost and complexity of shares are available from Bankscope. Second, our study using services offered in each of four transaction focuses on access to a standard bank account, which categories: opening an account, making payments, allows the full spectrum of commercial banking saving, and borrowing. Data were constructed at the services—checking, saving, payments, and credit. Many country level. To date, we have not done bank-level of the excluded institutions, such as postal banks, analysis or examined intracountry variation among specialize only in savings. As a check, however, we banks. Bank-level results are averaged by country using a compared asset shares of the participating banks to their simple average to construct one data point per country. shares of accounts (using numbers of accounts reported by the banks as a proportion of the total number of Examining the correlation accounts in the nation where those data were reported by between access and alternative the central bank). The correlation is positive and explanatory variables robust—a regression of account share on asset share yields a coefficient of 0.81 (and a significant t-statistic of 7.2). Given the large numbers of explanatory variables and the limited degrees of freedom in terms of countries covered, Control variables the analysis examines the effect of each explanatory variable on the outcome variable separately, through Several identifiable factors affect the level of banking regressions that control for income per capita, literacy, access across countries. In all of our cross-country population density, and urbanization. These depict the regressions, we use four country-level socioeconomic and cross-country association between the outcome of demographic control variables that plausibly affect the interest—usually accounts per thousand adults but propensity to bank. Most obvious is a country’s level of sometimes the proportion of firms with access to bank DATA SOURCES AND METHODOLOGY 67 finance—and the explanatory variable. The slope 37 The measure used in Finance for All? draws on household surveys where available, surveys of bank regulators, data from the World coefficient and t-statistic accompany each figure in which Bank Savings Association, and interpolations based on the regression fit between share of households with accounts and aggregate indica- regression results are presented. tors of deposit accounts and branch penetration. See Beck, Demirguc- Kunt, and Maksimovic (2007a, 2007b), Honohan and Beck (2007), Peachy and Roe (2006), and World Bank (2008c). 38 The MIX database did not contain information on the following eight The results reported are cross-country correlations, countries included in our sample: Botswana, Cape Verde, Central African Republic, Lao PDR, Lesotho, Malaysia, Mauritius, and Singa- controlling for income and four country-level pore. In the few cases where MIX data included commercial banks that provide microfinance services, these were not included in the to- socioeconomic and demographic variables. They do not tals of nonbank providers so as to avoid double counting. 39 While the MIX database may not be comprehensive and the com- prove or imply causation. For that, panel data estimates pleteness of its coverage may vary over countries, it is the best cur- rently available global database of microfinance service providers and are needed. As this is the first year of collecting data on gives a reasonable first estimate of the extent of coverage of clients. 40 bank access, such estimates will be available only after For instance, in Afghanistan, nonbank microfinance institutions and commercial banks appear to serve a similar number of clients— additional years of work. 41 297,808 and 267,435 respectively. A FinScope survey provides a list of the top five banks by number of accounts in seven African countries. The overlap with our sample is as follows: Botswana (4), Kenya (3), Namibia (4), South Africa (4), Tanzania (3), Uganda (3), and Zambia (4). The banks missing from our sample but appearing on the FinScope list are almost all savings institutions such as the Botswana Savings Bank and Uganda’s Post Bank. TABLE 9.1 Characteristics of data sample Commercial banks, asset shares in the financial system and in the sample Outcome variable Asset share % coverage % asset share of top 5 banks of total of banks Y = number Number of (% total financial reporting Firm access of accounts sampled financial system data on to bank per 1,000 Country banks assets)a assetsb accountsc Regulators credit adults Afghanistan 5 100 100 100 X — 19 Angola 6 86 89 60 X X 202 Bangladesh 7 54 59 59 X — 373 Benin 3 100 73 73 X X 60 Botswana 5 94 93 56 No X 923 Burkina Faso 5 78 78 64 X X 121 Burundi 5 100 100 100 X X 25* Cambodia 5 76 74 65 X X 32 Cameroon 5 81 81 66 X X 77 Cape Verde 2 100 36 25 X X 1,139 Central African Republic 1 100 30 30 X — 21* Chad 5 100 100 73 X — 9* Côte d'Ivoire 1 90 18 0 X — 71 Congo, Dem. Rep. of 3 74 33 26 No X 2* El Salvador 4 90 60 60 X X 693 Ethiopia 5 94 94 87 No X 66* Gabon 4 100 18 0 X X 183 The Gambia 4 100 100 93 X X 165 Ghana 5 72 72 46 X X 204 Guatemala 3 61 32 12 X X 910 Honduras 5 65 65 42 X X 625 India 5 48 47 15 X X 657 Indonesia 5 58 32 32 X X 464 Kenya 6 59 61 46 X X 169 Lao PDR 5 100 100 48 X — 108* continued 68 BANKING THE POOR TABLE 9.1 continued Characteristics of data sample Commercial banks, asset shares in the financial system and in the sample Outcome variable Asset share % coverage % asset share of top 5 banks of total of banks Y = number Number of (% total financial reporting Firm access of accounts sampled financial system data on to bank per 1,000 Country banks assets)a assetsb accountsc Regulators credit adults Lesotho 5 100 100 66 X — 307 Liberia 5 100 100 100 X X 36 Madagascar 4 96 53 53 X — 31 Malawi 5 99 92 91 X X 108 Malaysia 5 56 56 56 X X 2,177 Mali 4 97 70 70 X — 76 Mauritius 3 80 59 59 X X 2,011 Mexico 3 81 49 32 X X 631 Mozambique 5 96 93 33 X — 103 Namibia 4 96 65 62 X X 323 Nepal 5 69 63 42 X — 178 Nicaragua 2 96 43 18 X X 232 Niger 5 100 100 100 X X 16 Nigeria 5 67 58 44 X — 185 Pakistan 5 60 60 18 X — 268 Philippines 5 60 57 57 X X 566 Rwanda 2 92 32 32 X X 35 Senegal 4 89 55 20 X X 79 Sierra Leone 5 100 100 100 X — 65 Singapore 2 99 61 43 X — 2,058 South Africa 3 96 69 48 X X 552 Sri Lanka 5 84 84 84 X X 1,578 Sudan 6 58 25 14 X — 144* Swaziland 5 94 94 76 X X 483 Tanzania 6 75 66 51 X X 159 Thailand 5 65 62 62 X X 1,352 Uganda 3 59 37 37 X X 141 Vietnam 5 80 77 53 X X 83* Zambia 5 82 82 82 X X 115 Note: Number of countries covered: 54; number of commercial banks covered: 235. X = survey received. — = data not available. * Imputed values. Imputed by extrapolation from figures on numbers of accounts provided by the commercial banks sampled in these countries, using information from Bankscope on the shares of those banks in the national financial system. a. BankScope Data, 2006–07. b. BankScope Data, 2006–07. Based on survey coverage of this study. c. Survey data collected for this study. DATA SOURCES AND METHODOLOGY 69 References Acosta, Pablo, Pablo Fajnzylber, and Humberto Lopez. ———. 2007. “What Is Middle Class about the 2008. Remittances and Development: Lessons from Middle Class Around the World?��? MIT Latin America. World Bank Publications. Department of Economics Working Paper 07-29. Washington, D.C. Cambridge, Mass. Adams, Dale W., Douglas H. Graham, and J. D. Von Banerjee, Abhijit, and Kaivan Munshi. 2004. “How Pischke. 1984. Undermining Rural Development Efficiently is Capital Allocated? Evidence from the with Cheap Credit. Boulder, Colorado: Westview Knitted Garment Industry in Tirupur.��? Review of Press. Economic Studies 71: 19–42. Aggarwal, Reena, Asli Demirguc-Kunt, and Martinez Basu, Priya, and Pradeep Srivastav. 2006. “Access to Peria. 2006. “Do Workers’ Remittances Promote Rural Finance in India.��? World Bank Policy Financial Development.��? World Bank Policy Research Working Paper 3646. Washington, D.C. Research Working Paper 3957. Washington, D.C. Bayer, Patrick J., Douglas B. Bernheim, and John Karl Ahmed, Shaikh S. 2005. “Delivery Mechanisms of Scholz. 1996. “The Effects of Financial Education Cash Transfer Programs to the Poor in Bangladesh.��? in the Workplace: Evidence from a Survey of World Bank Social Protection Discussion Paper Employers.��? NBER Working Paper W5655, Series. Washington, D.C. National Bureau of Economic Research, Cambridge, Apgar, William, and Mark Duda. 2003. “The Twenty- Mass. Fifth Anniversary of the Community Reinvestment Beck, Thorsten, Asli Demirguc-Kunt, and Vojislav Act: Past Accomplishments and Future Regulatory Maksimovic. 2004. “Bank Competition and Access Challenges.��? Economic Policy Review 9 (2): 169–192. to Finance: International Evidence.��? Journal of Aportela, Fernando. 1999. “Effects of Financial Access Money, Credit and Banking 36 (3): 627–648. on Savings by Low-Income People.��? Banco de Beck, Thorsten, Asli Demirguc-Kunt, and Maria México Research Department. Chapter 1 of Soledad Martinez Peria. 2007a. “Banking Services doctoral dissertation, MIT Department of for Everyone? Barriers to Bank Access and Use Economics, Cambridge, Mass. around the World.��? Policy Research Working Paper Ashraf, Nava, Dean Karlan, and Wesley Yin. 2006a. 4079. World Bank, Washington, D.C. “Household Decision Making and Savings Impacts: ———. 2007b. “Reaching Out: Access to and Use of Further Evidence from a Commitment Savings Banking Services across Countries.��? Journal of Product in the Philippines.��? Discussion Paper 939. Financial Economics 85(1): 234–66. Yale University, New Haven, Conn. Berger, Allen, Anthony Saunders, Joseph Scalise, and ———. 2006b. “Tying Odysseus to the Mast: Gregory F. Udell. 1998. “The Effects of Bank Evidence from a Commitment Savings Product in Mergers and Acquisitions on Small Business the Philippines.��? Quarterly Journal of Economics 121 Lending.��? Journal of Financial Economics (50): (2): 635–672. 187–229. Bank of Tanzania: Banking Supervision Annual Bernheim, B. Douglas, and Daniel M. Garrett. 2003. Reports, 2005. http://www.bot- “The Effects of Financial Education in the tz.org/Publications/EconomicAndOperationsAnnua Workplace: Evidence from a Survey of Households.��? lReports/June_2005.pdf Journal of Public Economics (87): 1487-1519. Banerjee, Abhijit, and Esther Duflo. 2004. “Do Firms Bernheim, B. Douglas, Daniel M. Garrett, and Dean Want to Borrow More? Testing Credit Constraints M. Maki. 2001. “Education and Saving: The Long- Using a Directed Lending Program.��? BREAD Term Effects of High School Financial Curriculum Working Paper No. 005. Mandates,��? Journal of Public Economics, Elsevier 80 ———. 2006. “The Economic Lives of the Poor.��? (3): 435–465. Journal of Economic Perspectives 21(1): 141–167. REFERENCES 71 Bertrand, Marianne, Dolly Chugh, and Sendhil Cracknell, David. 2004. “Electronic Banking for the Mullainathan. 2004. “A Behavioral Economics View Poor: Panacea, Potential and Pitfalls.��? Small of Poverty.��? American Economic Association 94 (2): Enterprise Development 15 (4): 8–24. 419–423. Daley-Harris, Sam. 2006. “State of the Microcredit Biggs, Tyler. 1991. “Heterogeneous Firms and Efficient Summit Campaign Report 2006.��? Microcredit Financial Intermediation in Taiwan.��? In Markets in Summit Campaign, Washington, D.C. Developing Countries: Parallel, Fragmented, and Demirguc-Kunt, Asli, Thorsten Beck, and Patrick Black, ed. Michael Roemer and Christine Jones Honohan. 2008. Finance for All? Policies and Pitfalls (167–197). San Francisco: ICS Press. in Expanding Access. Washington, D.C.: World Black, Sandra, and Philip Strahan. 2002. Bank. “Entrepreneurship and Bank Credit Availability.��? Djankov, Simeon, Caralee McLeish, and Andrei Journal of Finance 57 (6): 2807–2833. Shleifer. 2007. “Private Credit in 129 Countries.��? Bonaccorsi di Patti, Emilia, and G. Dell’Ariccia. 2003. Journal of Financial Economics 84 (2): 299–329. “Bank Competition and Firm Creation.��? Journal of Djankov, Simeon, Pedro Miranda, Enrique Seira, and Money, Credit, and Banking 36 (2): 225–251. Siddharth Sharma. 2008. “Who Are the Braunstein and Welch. 2002. “Financial Literacy: An Unbanked?��? World Bank Enterprise Analysis Unit, Overview of Practice, Research and Policy.��? Federal Washington, D.C. Reserve Bulletin (85): 445-457. Donner, Jonathan. 2003. “Microentrepreneurs and British Bankers Association. 2005. “Basic Bank Mobiles: An Exploration of the Uses of Mobile Account Growth Continues.��? November. Phones by Small Business Owners in Rwanda.��? http://www.bba.org.uk/bba/jsp/polopoly.jsp?d=446 Earth Institute, Columbia University, New York. Burgess, Robin, and Rohini Pande. 2004. “Do Rural ———. 2006. “The Use of Mobile Phones by Banks Matter? Evidence from the Indian Social Microentrepreneurs in Kigali, Rwanda: Changes to Banking Experiment.��? CEPR Discussion Paper Social and Business Networks.��? Information 4211. Centre for Economic Policy Research, Technologies and International Development 3 (2): London. 3–19. Caprio, Gerard, Jonathan Fiechter, Michael ———. 2007. “M-Banking and M-Payments Services Pomerleano, and Robert E. Litan, eds. 2004. The in the Developing World: New Channel, Same Ties? Future of State-Owned Financial Institutions Policy Presentation at the conference Home/Community- and Practice. Washington, D.C.: Brookings Oriented ICT for the Next Billion, IIT Madras, Institution Press. Chennai, India. CGAP. 2008. “Regulating Transformational Branchless Doyle, Joseph J., Jose A. Lopez, and Marc R. Banking: Mobile Banking and Other Technologies Seidenberg. 1998. “How Effective Is Lifeline to Increase Access to Finance��? Focus Note 43. Banking?��? Current Issues in Economics and Finance World Bank, Washington, D.C. (Federal Reserve Bank, New York) 4 (6). Chakraborty, Suvalaxmi, and Annie Duflo. 2006. Duflo, Esther, William Gale, Peter Orszag, and “Transformation to a Full-Service Financial Emmanuel Saez. 2007. “Savings Incentives for Low- Institution.��? Paper Presented to the Global and Moderate Income families in the United States: Conference on Access to Finance. World Bank, Why Is the Saver’s Credit Not More Effective?��? Washington, D.C. Journal of the European Economic Association 5 (2–3): Cole, David C., and Yung Chul Park. 1983. Financial 647–661. Development in Korea 1945–1978. Cambridge, Economist. 2008. “Technology and Development.��? Mass.: Harvard University Press. February 7. Available at: Coleman, Brett E. 2006. “Microfinance in Northeast http://www.economist.com/opinion/displaystory.cf Thailand: Who Benefits and How much?��? World m?story_id=10650775 Development 34 (9): 1612–1638. 72 BANKING THE POOR El-Qorchi, Mohammed, Samuel Maimbo, and John Government Accountability Office. 2002. “Use by Wilson. 2003. “Informal Funds Transfer Systems: Federal Payment Recipients Has Increased but An Analysis of the Hawala System.��? IMF Obstacles to Greater Participation Remain.��? Report Occasional Paper 222, International Monetary to the Subcommittee on Oversight and Fund, Washington, D.C. Investigations, Committee on Financial Services, Fellowes, Matt, and Mia Mabanta. 2008. “Banking on House of Representatives. Paper 913, Washington, Wealth: America’s New Retail Banking D.C. Infrastructure and Its Wealth-Building Potential.��? Grosh, Margaret, Carlo del Ninno, Emil Tesliuc, and Brookings Institution Research Brief. Washington, Azedine Oeughi, with Annamaria Milazzo and D.C. Christine Weigand. Forthcoming. “For Protection FinScope. 2003. “A Study of the Financial Access and and Promotion: the Design and Implementation of Behavior of the South African Population—Towards Effective Safety Nets.��? World Bank, Washington, Defining a Segmentation Model that Will Classify a D.C. Person Into Eight FSM Tiers.��? FinMark Trust, GTZ. 2008. Social Cash Transfer Scheme: Johannesburg, South Africa. Dependency. Available at: ———. 2003. “Namibia.��? Available at: http://www.socialcashtransfers- http://www.finscope.co.za/namibia.html zambia.org/pageID_2680193.html. Accessed on ———. 2005. “Uganda.��? Available at: March 10, 2008. http://www.finscope.co.za/uganda.html Hainz, Christa. 2003. “Bank Competition and Credit ———. 2006. “Kenya.��? Available at: Markets in Transition Economies.��? Journal of http://www.finmark.org.za/Documents/FPres_FSAf Comparative Economics (31): 223–245. rica.pdf Hanson, James. 2003. “Banking in Developing FinScope/Bankable Frontiers 2007. Countries in the 1990s.��? World Bank Policy http://www.bankablefrontier.com/weblog/index.php. Research Working Paper 3168. Washington, D.C. Firpo, Janine. 2005. “Banking the Unbanked: Hernandez-Coss, Raul. 2005. “A Proposed Framework Technology’s Role in Delivering Accessible Financial to Analyze Informal Funds Transfer Systems.��? In Services to the Poor.��? SEMBA Consulting and U.K. Remittances: Development Impact and Future Department for International Development, Prospects, ed. Samuel Maimbo and Dilip Ratha. London. Washington, D.C.: World Bank. Foster, Andrew D., and Mark R. Rosenzweig. 2004. ———. 2005. “The U.S-Mexico Remittance Corridor: “Agricultural Productivity Growth, Rural Economic Lessons on Shifting from Formal to Informal Diversity and Economic Reforms: India, Transfer Systems.��? World Bank Working Paper 47. 1970–2000.��? Economic Development and Cultural Washington, D.C. Change 52 (3): 509–42. Hilgert, Marianne A., and Jeanne M. Hogarth. 2003. Freund, Caroline, and Nikola Spatafora. 2008. “Household Financial Management: The “Remittances: Transaction Costs, Determinants and Connection between Knowledge and Behavior.��? Informal Flows.��? Journal of Development Economics. Federal Reserve Bulletin (July): 309–22. 86(2): 356-366. Hirschland, Madeline. 2005. Savings Services for the Gammeltoft, Peter. 2002. “Remittances and Other Poor. Sterling, Va.: Kumarian Press. Financial Flows to Developing Countries.��? Working Hogarth, Jeanne, Amberly Hazembuller, and Michael Paper 02.11. Centre for Development Research, D. Wilson. 2004. “How Much Can the Poor Save?��? Copenhagen. Federal Reserve Board, Washington, D.C. Genesis Analytics. 2003. “African Families, African September. Money: Bridging the Money Transfer Divide.��? Honohan, Patrick, and Thorsten Beck. 2007. Making Report prepared for FinMark Trust, Johannesburg, Finance Work for Africa. World Bank. Washington, South Africa. D.C. REFERENCES 73 Hulme, David, and Paul Mosley, eds. 1996. Finance La Porta, Rafael, Florencio Lopez-de-Silanes, and Against Poverty. London: Routledge. Andrei Shleifer. 2002. “Government Ownership of IMF (International Monetary Fund). 2007. Banks.��? Journal of Finance 57 (1): 265–301. “Democratic Republic of Congo: Selected Issues La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei and Statistical Appendix.��? IMF Country Report Shleifer, and Robert Vishny. 1997. “Legal 07/329. Washington, D.C. September. Determinants of External Finance.��? Journal of Ivatury, Gautam, and Mark Pickens. 2006. “Mobile Finance 52 (3): 1131–1150. Phone Banking and Low-Income Customers: Malkin, Elizabeth. 2007. “Thousands in Mexico City Evidence from South Africa.��? CGAP, Washington, Protest Rising Food Prices.��? New York Times, D.C. February 1. Jain, Sanjay. 1999. “Symbiosis versus Crowding Out: Martin, Philip, Susan Philip, and Patrick Weil. 2002. The Interaction of Formal and Informal Credit “Best Practice Options: Mali.��? Cooperative Efforts Markets in Developing Countries.��? Journal of to Manage Emigration, University of California, Development Economics 59 (2): 419–44. Davis. Jappelli, Tullio, and Marco Pagano. 1993. “Information Nair, Ajai, and J. D. von Pischke. 2007. “Commercial Sharing in Credit Markets.��? The Journal of Finance Banks and Financial Access.��? In Building Inclusive 48 (5): 1693–1718. Financial Systems: A Framework for Financial Access ———. 2002a. “Courts and Banks: Effects of Judicial Financial Services, and Financial Markets, ed. Enforcement on Credit Markets.��? Journal of Money, Michael S. Barr, Anjali Kumar, and Robert E. Litan. Credit and Banking 37 (2): 223–44. Washington, D.C.: Brookings Institution. ———. 2002b. “Information Sharing, Lending and Orozco, Manuel. 2004. The Remittance Marketplace: Defaults: Cross-Country Evidence.��? Journal of Prices, Policies, and Financial Institutions. Banking and Finance 26: 2017–2045. Washington, D.C. Pew Hispanic Center, June. Johnston, Don, and Jonathan Murdoch. 2007. Padilla, Jorge, and Marco Pagano. 1997. “Endogenous “Microcredit versus Microsaving: Evidence from Communication Among Lenders and Indonesia.��? Paper prepared for conference on Access Entrepreneurial Incentives.��? Review of Financial to Finance, March 15–16, 2007. Washington, D.C. Studies 10 (1): 205–236. Kamel, Sherif, and Ahmed Hassan. 2003. “Assessing Peachy, Stephen, and Alan Roe. 2006. “Access to the Introduction of Electronic Banking in Egypt Finance: Measuring the Contribution of Savings Using the Technology Acceptance Model.��? In Banks.��? World Savings Bank Institute, Brussels, Annals of Cases on Information Technology, vol. 5, ed. Belgium. Mehdi Kosrow-Pour. Hershey, Penn.: Idea Group. Peterson, Mitchell A., and Raghuram Rajan. 1995. Kempson, Elaine, and Claire Whyley. 2000. “The Effect of Credit Market Competition on “Understanding Small Savers: Saving on a Low- to Lending Relationships.��? Quarterly Journal of Middle-Income.��? Pearl Assurance, Peterborough, Economics 110 (2): 407–443. England. Pitt, Mark M., and Shahidur R. Khandker. 1998. “The Khwaja, Asim, and Atif Mian. 2005. “Do Lenders Impact of Group-Based Credit Programs on Poor Favor Politically Connected Firms? Rent Provision Households in Bangladesh: Does the Gender of in an Emerging Financial Market.��? Quarterly Participants Matter?��? Journal of Political Economy Journal of Economics 120 (4): 1371–411. 106 (5): 958–96. Kumar, Anjali. 2005. Access to Financial Services in Porteous, David. 2005. “The Access Frontier as an Brazil. Washington, D.C.: World Bank. Approach and Tool in Making Markets Work for Kumar, Anjali, Ajai Nair, Adam Parsons, and Eduardo the Poor.��? Unpublished paper prepared for the Urdapilleta. 2006. “Expanding Bank Outreach Department for International Development, through Retail Partnerships: Correspondent London. Banking in Brazil.��? World Bank Working Paper 85. Washington, D.C. 74 BANKING THE POOR ———. 2006. “The Enabling Environment for Mobile Sherraden, Michael. 2006. “Schemes to Boost Small Banking in Africa.��? Department for International Savings: Lessons and Directions.��? Paper presented at Development, London. Access to Finance: Building Inclusive Financial ______. 2008. “Mobile Phone Banking: Is M-banking Systems, May 31, Washington, D.C. Advancing Access to Basic Banking Services in Sherraden, Michael, and Mark Schreiner. 2007. “Can South Africa?��? FinMark Trust and Bankable Frontier the Poor Save? Saving and Asset Accumulation in Associates, Johannesburg. Individual Development Accounts.��? Piscataway, Porteous, David, and Ethel Hazelhurst. 2006. Banking N.J.: Transaction Publishers. on Change: Democratic Finance in South Africa, Shukla, Rajesh. 2007. How India Earns, Spends, and 1994–2004 and Beyond. FinMark Trust, Saves: Results from the Max New York Life–NCAER Johannesburg. India Financial Protection Survey. New Delhi: Qian, Jun, and Philip E. Strahan. 2005. “How Law and National Council of Applied Economic Research. Institutions Shape Financial Contracts: The Case of http://www.ncaer.org/featuredproject1.html. Bank Loans.��? NBER Working Paper 11052, Siddiqui, Tasneem, and C. R. Abrar. 2003. “Migrant National Bureau of Economic Research, Cambridge, Worker Remittances and Microfinance in Mass. Bangladesh.��? Working Paper 38. International Ramji, Minakshi. 2007. “100 Percent Financial Labour Organization, Geneva. Inclusion: Improving Access and Usage. A Study Solo, Tova Maria, and Astrid Manroth. 2006. “Access Report on Gulbarga District Initiative.��? to Financial Services in Colombia: The Unbanked http://ifmr.ac.in/cmf/eomf6-financialinclusion.html in Bogota.��? World Bank Policy Research Working Roig-Franzia, Manuel. 2007. “A Culinary and Cultural Paper 3834. Washington, D.C. Staple in Crisis.��? Washington Post, January 27. Solo, Tova Maria, John P. Caskey, and Clemente Ruiz Rose, R. 2003. Welfare Boost for Low-End Banking. Duran. 2006. “The Urban Unbanked in Mexico Business Day (Johannesburg), April 23. and the United States.��? World Bank Policy Research Rutherford, Stuart. 1998. “The Savings of the Poor.��? Working Paper 3835. Washington, D.C. Journal of International Development 10, (1). South African Reserve Bank. 2001. “Labour Markets ———. 2000. The Poor and Their Money. Oxford and Social Frontiers: Financial Development and University Press. New York. the Unbanked.��? Pretoria. ______. 2002. “Money Talks: Conversations with Poor Sweden’s Banking Business Act of 1987. Available at: Households in Bangladesh about Managing http://www.bankforeningen.se/upload/bankr%C3% Money.��? Working Paper 45. Institute for B6relselagen.pdf Development Policy and Management, University Taylor, Edwards. 1999. “The New Economics of Labor of Manchester. Migration and the Role of Remittances.��? Safavian, Mehnaz, and Siddharth Sharma. 2007. International Migration 37 (1): 63–86. “When Do Creditor Rights Work?��? Journal of Tescher, Jennifer, and Rachel Schneider. 2008. “Lessons Comparative Economics 35 (3): 484–508. from South Africa: Innovations and Opportunities Sander, Cerstin, and Samuel Maimbo. 2003. “Migrant in Underbanked Services.��? Center for Financial Labor Remittances in Africa.��? Africa Region Services Innovation, Shorebank Corporation, Working Paper 64. World Bank, Washington, D.C. Chicago. Schoorl, J. J., L. Heering, I. Esveldt, G. Groenewold, MIF (Multilateral Investment Fund of the Inter- R. F. van der Erf, A. M. Bosch, H. de Valk, and B. J. American Development Bank). 2003–06. de Bruijn. 2000. “Push and Pull Factors of “Remittance Recipients in Bolivia, Brazil, Cuba, International Migration: A Comparative Report.��? Dominican Republic, Ecuador, Haiti, Mexico, Research report 3/2000/E/no.14. Eurostat. Paraguay, Peru.��? Produced by Bendixen Associates for the Multilateral Investment Fund of the Inter- American Development Bank. REFERENCES 75 World Bank. 2006. “The Role of Postal Networks in Expanding Access to Financial Services.��? Global Information and Communication Technologies Department, Washington, D.C. ———. 2008a. World Development Indicators. Washington, D.C. ———. 2008b. Global Economic Prospects. Washington, D.C. ———. 2008c. Payment Systems Worldwide: A Snapshot—Outcomes of the Global Payment Systems Survey 2008. Washington, D.C. World Bank Remittance Database. 2007. www.worldbank.org/prospects/migrationandremitta nces. Yaron, Jacob, McDonald P. Benjamin, and Gerda L. Piprek. 1997. “Rural Finance: Issues, Design, and Best Practices.��? Environmentally and Socially Sustainable Development Studies and Monographs Series 14. World Bank, Washington, D.C. Zia, Bilal, and Shawn Cole. 2008. “Bounded Rationality, Financial Literacy, or Consumer Sovereignty: Field Experiments Exploring Why People Do Not Use Financial Services?��? World Bank. Washington, D.C. 76 BANKING THE POOR Glossary of variables, indexes, and other terms Variable/index Definition Number of accounts Accounts per thousand adults The total number of deposit accounts (checking, savings, and time deposits) per thousand adults aged 15 years and older, based on data from World Development Indicators. Data for 45 countries were obtained from surveys of central banks and regulators conducted by the Getting Finance project. For nine countries, marked with an asterisk, this value was imputed on the basis of data on numbers of accounts reported to the Getting Finance project by surveyed commercial banks. Those data were scaled up to represent total numbers of accounts in the country based on the share of the surveyed banks in the total assets of the banking system, using data from BankScope. Opening an account Index of accessibility The index of accessibility measures the ease of opening an account in terms of the range and number of locations where an account application can be submitted. The index is the weighted sum (weights in parentheses) of survey responses on different modes of applying: at any branch (.33), phone (.167), Internet (.167), correspondent banks (.167), and other (.167). At least one nonbranch option to apply for This value is set to 1 if the bank offers any one or more of the following options to apply for an account (0–1) an account: over the telephone, via Internet, at a correspondent outlet, and at other locations, such as home or the workplace. At least one electronic This value is set to 1 if the bank offers one or both of the following options to apply for an option (0–1) account: over the telephone, via Internet. Convenience features Index of convenience features (0–1) The index of convenience is an equally weighted mean of responses to the following survey questions: passbook not required, after-hours access available, overdraft provision, and notification of overdraft. Index of free features (0–1) The index of free features is an equally weighted mean of responses to the following survey questions: unlimited free withdrawals, unlimited free balance inquiries, free ATM card, free checking, and free credit card. Domestic and cross-border payments Index of time required for domestic payments The domestic time index comprises five variables: time for a remittance to reach its (0–1) destination in the country’s second-largest city and in a remote rural area, time to clear a check from another bank, time to clear a direct credit sent to a recipient at another bank, and time to clear a direct debit if initiated by another bank. The five variables are normalized and given equal weight (0.2). Index of time required for cross-border The cross-border time index comprises two variables: time for a check to be sent to the payments (0–1) country's most common overseas destination and collected by the recipient, and time for a wire transfer sent to the country's most common overseas destination to be collected by the recipient. These two variables are normalized and given equal weight (0.5). GLOSSARY OF VARIABLES, INDEXES, AND OTHER TERMS 77 Variable/index Definition Retail payments, availability and quality Index of range of payment services with stan- The range of payments services index consists of five variables that reflect the availability, with dard bank account (0–1) a standard account at a commercial bank, of checks, debit cards, interbank services, intra- bank services, and credit cards, with all variables equally weighted (0.2). Index of mobile banking technology (0–1) The mobile banking technology index reflects the availability by cell phone of five transac- tions: checking one’s balance, obtaining a statement of recent activity, checking the status of a payment, transferring money, and paying bills. Each transaction has the same weight. Index of network quality and interoperability The index of network quality and interoperability comprises two variables that assess the in- (0–1) teroperability of payments networks: whether debit cards can be used at ATMs belonging to other banks that share the bank’s network; and whether debit cards can be used at mer- chants through point-of-service devices. The two variables have different weights: 0.4 for the former and 0.6 for the later. Index of retail payment channels (0–1) The index of retail payment channels comprises six retail payment transactions and assesses their availability in each of three different channels: person to person, individual to business, and individual to government. Each has equal weight. The six transactions are: check pay- ment; payment cards used on bank premises; payment cards used at ATMs; direct credits initi- ated on bank premises; direct credit via telephone, Internet, or mobile banking technology; and direct debits initiated by the beneficiary. Credit Index of complexity of business-loan applica- This index is an equally weighted sum of survey responses on four aspects of the process of tion applying for a business loan: whether the borrower is required to have an account at the bank, whether a letter of reference is required, whether the borrower has to furnish proof of employment, and whether an application fee is charged. Index of collateral flexibility The index of collateral flexibility is an equally weighted sum of survey responses on the ac- ceptability of various types of assets as collateral for a business startup loan. The asset types are: cash and liquid assets, land, house with title, house without title, equipment, livestock, accounts receivable, and other. Basic banking Government, offers basic banking* Based on regulators’ answers to the question of whether national regulations require that a basic or simplified account must be made available to some or all segments of the popula- tion. The questionnaire explains that such accounts may be designed for low-income persons and may have the following characteristics: no opening fee, no monthly fee, a basic package of transactions free of charge (such as a limited number of free withdrawals, balance in- quiries, and payments), and restrictions on check writing and card-only transactions. Government, basic accounts exempted from Value set to 1 if regulator responds positively to the following question: “Are there provisions ID requirements* for any exemptions from ID requirements for low-income or other clients?��? Basic banking (0–1)* This variable measures whether commercial banks actually offer accounts that have the fea- tures of basic accounts, based on the equally weighted sum of survey responses on the fol- lowing four account characteristics: no opening fee, no monthly fee, no minimum balance, and bundle of free transactions. 78 BANKING THE POOR Variable/index Definition Savings schemes Doorstep collection Value set to 1 if regulator responds positively to the following question: “Are there any doorstep deposit collection schemes for low-income clients in your country?��? Government-matched savings Value set to 1 if regulator responds positively to the following question: “Are there any guide- lines on government-supported matched savings schemes?��? Tax incentives for savings Value set to 1 if regulator responds positively to the following question: “Does your country provide tax incentives for specific saving accounts?��? Periodic deposit The questionnaire sent to commercial banks provides the following description of periodic de- posit schemes: “Some banks offer savings schemes in which persons have to agree to con- tribute a fixed amount at regular intervals of time—e.g., on a monthly basis. These periodic savings schemes offer clients a disciplined way to save, especially for small savers.��? The ques- tionnaire then asks: “Does your bank offer any periodic savings schemes?��? The value of this variable is set to 1 if the response to the preceding question is positive. Credit-related transparency and consumer protections Advising applicants of credit decisions (0–1) The index of disclosure of credit decisions reflects the presence of national guidelines requir- ing banks to advise clients of an unfavorable change of credit terms and to notify clients when credit is denied. Documents required for credit applications The index of documents required for credit applications ranges from 0 to 1, based on the (0–1) number of additional documents required, expressed as a proportion of the maximum num- ber of documents. Countries are averaged within each income and regional category, using equal weights. Guidelines related to credit documents (0–1) The index of guidelines on credit documents consists of four subindexes that measure the presence or absence of guidelines on identity papers, tax identification, recorded credit his- tory, and the need for personal guarantors. Transparency and consumer protection (0–1) The variable is composed of 12 subindexes that require a yes or no answer: The central bank provides public information on fees and charges levied on transactions at commercial banks as well as disclosure of interest rates and charges to the customer. GLOSSARY OF VARIABLES, INDEXES, AND OTHER TERMS 79 Indexes of financial access: mean values by region and country Latin All countries South Asia East Asia America Africa Afghanistan Angola Bangladesh Benin Botswana Burkina Faso Burundi Number of accounts No. of accounts per thousand adults1 463.87 512.27 854.91 618.12 240.21 19.05 202.00 372.95 59.80 923.00 120.92 25* No. of accounts per thousand inhabitants1 317.23 363.50 825.30 389.00 158.70 10.67 108.52 243.47 33.51 .. 65.31 13.90 80 BANKING THE POOR Opening an account Number of documents required 2.82 2.75 2.00 2.52 3.05 2.60 2.33 4.43 3.67 4.00 2.60 2.60 Third-party reference required (average yes/no) 0.38 0.40 0.10 0.72 0.39 0.60 0.17 1.00 0.33 0.80 0.20 0.00 Index of accessibility (0–1)1 0.42 0.41 0.41 0.35 0.44 0.37 0.47 0.40 0.44 0.50 0.53 0.46 At least one nonbranch option to apply for an account (0–1)1 0.40 0.37 0.33 0.18 0.45 0.20 0.50 0.43 0.67 0.80 0.80 0.40 At least one electronic option (0–1)1 0.17 0.10 0.16 0.04 0.20 0.00 0.17 0.00 0.00 0.20 0.20 0.20 Application fee? (average yes/no) 0.10 0.00 0.10 0.11 0.11 0.00 0.00 0.00 0.00 0.00 0.00 0.60 Other costs? (average yes/no) 0.18 0.16 0.10 0.09 0.22 0.20 0.17 0.14 0.00 0.00 0.00 0.60 Days to open an account 1.29 1.41 0.81 1.44 1.36 1.00 0.67 1.29 0.75 2.30 0.70 0.90 Convenience features Index of convenience features (0–1)1 0.52 0.48 0.39 0.66 0.55 0.31 0.58 0.29 0.50 0.70 0.70 0.45 Index of free usage features (0–1)1 0.52 0.54 0.53 0.48 0.52 0.40 0.67 0.56 0.60 0.64 0.64 0.60 Minimum balance required when opening a standard account? (average yes/no) 0.90 0.97 0.89 1.00 0.87 1.00 0.83 1.00 1.00 0.80 1.00 1.00 Mean minimum deposit for a standard account (fraction GDP per capita) 0.30 0.09 0.05 0.03 0.43 0.21 0.14 0.04 0.34 0.00 0.08 3.42 Minimum balance required when opening a savings account? (average yes/no) 0.93 0.93 0.94 1.00 0.92 1.00 0.50 1.00 1.00 1.00 1.00 0.80 Minimum ongoing balance needed for a savings account? (average yes/no) 0.71 0.85 0.64 0.68 0.71 0.50 0.33 0.80 1.00 0.60 1.00 0.40 Fees and costs for standard domestic transactions (% monthly income) Cost per checkbook issued 10.67 5.95 3.72 2.00 14.42 24.63 1.08 0.87 0.00 0.49 0.00 39.46 Annual fee for cash card/debit card 11.51 6.36 3.88 1.69 15.97 0.00 0.00 10.74 2.09 0.00 70.10 .. Cash card/debit card fees per transaction at merchants 0.16 0.05 0.02 0.08 0.22 0.00 0.00 0.04 0.00 0.03 0.67 .. Fee for issuing a banker's draft 21.61 10.32 4.34 2.92 30.44 44.07 4.68 0.75 13.79 1.61 10.05 85.05 Monthly fee for account maintenance 8.30 0.91 0.51 1.03 12.04 0.00 0.10 3.34 7.17 1.08 13.50 12.76 Fees and costs for standard domestic transactions (US$) Cost per checkbook issued 3.67 2.15 3.88 2.19 4.11 5.87 2.67 0.30 0.00 2.32 0.00 3.47 Annual fee for cash card/debit card 5.19 2.35 3.29 3.56 6.47 0.00 0.00 3.74 1.00 0.00 28.01 .. Cash card/debit card fees per transaction at merchants 0.13 0.03 0.02 0.08 0.18 0.00 0.00 0.01 0.00 0.17 0.27 .. Fee for issuing a banker's draft 8.01 2.99 7.52 4.79 9.56 10.50 10.67 0.26 6.63 7.63 4.02 7.48 Monthly fee for account maintenance 3.05 0.55 0.96 3.75 4.23 0.00 0.22 1.16 3.45 5.14 5.39 1.12 Fees and costs for remittances (% monthly income) Fee to send payment via bankers' draft, foreign currency 69.75 15.13 14.56 10.27 98.26 22.81 13.08 22.18 47.21 2.24 161.42 136.08 Fee to send payment via wire transfer 67.79 30.77 18.23 18.06 93.00 53.19 14.28 32.26 48.05 2.03 127.38 117.18 Fee to use money transfer operator to send money 66.67 28.57 12.06 4.37 84.37 91.23 .. 5.74 32.21 .. 85.42 382.10 Fee to receive a check issued abroad 59.91 7.75 12.50 4.76 86.13 0.00 9.48 2.50 42.83 3.74 39.57 391.23 Fee to receive payment via wire transfer 26.54 16.08 3.54 5.70 36.08 62.77 10.92 4.08 32.03 2.10 48.36 30.24 Fee to receive remittance through money transfer operator 19.17 4.80 1.17 0.00 29.37 0.00 .. 4.08 0.00 .. 0.00 191.06 Fees and costs for remittances (US$) Fee to send a $250 payment via banker's draft in foreign currency 24.19 6.12 17.26 0.90 31.82 5.43 29.33 7.72 22.69 10.61 64.50 11.97 Fee to send $250 payment via wire transfer 27.49 12.60 20.55 23.78 32.23 12.67 31.89 11.22 23.09 9.62 50.90 10.31 Fee to use money transfer operator to send $250 18.58 7.86 13.25 4.08 24.36 21.74 .. 2.00 15.48 .. 34.13 33.61 Fee to receive $250 check issued abroad 18.85 3.63 9.29 4.06 25.60 0.00 21.27 0.87 20.58 17.74 15.81 34.41 Fee to receive $250 payment via wire transfer 9.93 5.56 4.89 5.95 12.40 14.96 24.41 1.42 15.39 9.95 19.32 2.66 Fee to receive $250 remittance through money transfer operator 5.44 0.91 1.76 0.04 7.96 0.00 .. 1.42 0.00 .. 0.00 16.81 Latin All countries South Asia East Asia America Africa Afghanistan Angola Bangladesh Benin Botswana Burkina Faso Burundi Domestic and cross-border payments, time indexes Domestic payment time index (0–1)* 0.49 0.56 0.39 0.45 0.51 0.38 0.50 0.55 0.39 0.58 0.36 0.46 Cross-border payment time index (0–1)* 0.44 0.38 0.40 0.43 0.46 0.46 0.39 0.27 0.28 0.53 0.69 0.44 Retail payments, availability and quality Range of payment services with standard bank account index (0–1)1 0.64 0.68 0.75 0.60 0.61 0.52 0.77 0.49 0.60 0.72 0.64 0.32 Mobile banking technology index (0–1)1 0.26 0.36 0.33 0.22 0.23 0.20 0.10 0.17 0.13 0.20 0.28 0.08 Network quality and interoperability index (0–1)1 0.62 0.69 0.76 0.83 0.54 0.20 0.67 0.63 0.33 0.80 0.88 0.12 Retail payment channels index (0–1)1 0.47 0.43 0.65 0.50 0.44 0.29 0.48 0.31 0.28 0.63 0.37 0.40 Credit Business loan processing time (days) 7.42 8.02 7.64 13.41 6.53 3.33 10.75 7.67 5.00 1.50 6.00 7.40 Startup loan processing fee (%) 0.89 0.38 0.54 0.92 1.07 0.50 0.40 0.13 0.54 0.88 0.35 1.99 Collateral-to-loan-value ratio (%) 88.32 96.12 80.14 73.82 90.45 41.67 58.75 137.50 83.33 100.00 83.33 106.00 Maximum terms for startup loan (years) 4.55 4.43 3.31 6.25 4.60 1.83 5.00 3.67 7.00 2.00 5.00 4.50 Index of business loan application complexity1 0.66 0.68 0.73 0.77 0.75 0.75 0.75 0.81 0.83 0.50 0.56 0.69 Index of collateral flexibility* 0.48 0.66 0.80 0.52 0.71 0.00 0.61 0.56 0.00 0.56 0.56 0.61 Basic banking No opening fee (average yes/no) 0.77 0.81 0.80 0.94 0.74 1.00 0.83 0.86 1.00 0.60 1.00 0.60 No monthly fee (average yes/no) 0.61 0.83 0.73 0.76 0.53 1.00 0.83 1.00 0.33 0.60 0.40 0.00 No minimum balance (average yes/no) 0.52 0.42 0.51 0.54 0.53 0.80 0.67 0.14 0.67 0.80 0.60 0.40 Basic package of free transactions (average yes/no) 0.56 0.60 0.58 0.66 0.54 0.80 0.67 0.43 0.67 0.60 0.80 0.25 Basic banking (0–1)* 0.61 0.67 0.64 0.73 0.59 0.90 0.75 0.61 0.67 0.65 0.70 0.31 Government policy (commercial bank response) (average yes/no) 0.57 0.70 0.62 0.36 0.56 0.60 0.00 1.00 0.67 0.40 0.00 0.00 Government, offering basic banking1 0.23 0.33 0.25 0.20 0.22 0.00 0.00 0.00 1.00 0.00 1.00 0.00 Government, exempt basic accounts from ID requirements1 0.06 0.17 0.13 0.00 0.03 0.00 0.00 1.00 0.00 .. 0.00 0.00 Savings schemes Doorstep collection1 0.08 0.17 0.25 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Government-matched savings1 0.02 0.00 0.13 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Tax incentives for savings1 0.42 0.33 0.75 0.40 0.38 0.00 0.00 0.00 1.00 0.00 1.00 0.00 Periodic deposit1 0.40 0.53 0.48 0.46 0.36 0.40 0.33 1.00 0.33 0.00 0.50 0.60 Transparency and consumer protection Advising applicants index (0–1)1 0.46 0.58 0.56 0.40 0.42 0.00 0.00 0.00 1.00 0.50 1.00 0.50 Additional document requirements index (0–1)1 0.10 0.17 0.13 0.13 0.08 0.38 0.00 0.13 0.00 .. 0.00 0.00 Guidelines on credit documents index (0 –1)1 0.71 0.67 0.72 0.90 0.68 0.00 0.75 1.00 1.00 0.75 1.00 0.75 Transparency and consumer protection index (0–1)1 0.40 0.60 0.50 0.40 0.33 0.25 0.33 0.58 0.25 0.83 0.25 0.08 Firms using bank loans (% all firms) Firms with bank loans 31.06 50.80 .. 36.30 27.78 .. 4.14 50.80 .. 27.24 .. 35.26 Firms that use loans for investment 16.25 12.71 23.95 10.83 15.70 .. 2.13 11.64 20.81 11.32 .. 12.25 Firms that use loans for expenses 28.20 34.47 33.27 28.16 25.58 .. 2.32 43.07 29.44 15.88 .. 25.51 Firms that view access to finance as constraint 35.62 27.76 15.11 22.06 46.17 .. 55.26 42.55 73.20 41.43 .. 50.94 Small and micro firms using bank loans (% firms) Small firms with bank loans 27.23 37.00 28.00 27.00 27.00 .. 2.68 .. 27.01 15.36 107.87 23.23 Micro firms with bank loans 20.33 34.00 21.00 22.00 19.00 .. 2.44 .. 19.46 13.60 14.41 19.65 Small firms that use loans for working capital 10.76 13.63 13.18 12.41 9.82 .. 1.22 .. 10.63 4.66 7.22 7.74 Micro firms that use loans for working capital 8.97 10.48 9.55 11.19 8.35 .. 1.08 .. 7.11 4.30 5.63 6.05 Small firms that use loans for investment 13.65 10.41 16.87 16.80 12.51 .. 4.74 .. 10.56 12.66 10.88 14.48 Micro firms that use loans for investment 11.20 6.25 13.77 15.27 10.06 .. 3.45 .. 5.97 11.83 9.91 11.83 Small firms that view access to finance as constraint 75.25 34.00 53.00 74.00 81.00 .. 98.59 .. 94.85 83.92 96.40 94.81 Micro firms that view access to finance as constraint 59.67 18.00 34.00 56.00 67.00 .. 92.24 .. 84.54 71.35 .. 88.15 INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 81 Latin All countries South Asia East Asia America Africa Afghanistan Angola Bangladesh Benin Botswana Burkina Faso Burundi Demographic and socioeconomic variables Per capita GDP (US$) 2041.69 662.42 5486.02 3136.88 1334.40 263.37 2686.17 429.25 549.28 5876.38 456.70 102.97 Population (millions) 53.25 292.20 63.38 27.20 20.51 32.00 16.00 144.00 9.00 2.00 14.00 8.00 Percentage of the population aged 15 years and older 61.00 65.00 70.00 63.00 57.00 56.00 53.72 65.28 56.04 64.91 54.01 55.28 Population density (/sq km) 245.56 370.96 920.71 124.00 87.11 38 13.15 1108.9 78.59 3.1 49.66 305.03 Urban population (%) 39.14 24.08 47.71 58.07 37.05 23.28 53.96 25.54 40.5 58.16 18.7 10.32 82 BANKING THE POOR Literacy (15+) 64.17 53.40 85.79 79.46 58.89 28.00 67.41 47.49 34.66 81.19 22.69 59.30 Ratio of private credit to GDP 28.85 37.00 55.25 35.00 20.53 .. 8.00 36.00 17.00 20.00 17.00 26.00 Banking assets covered (%) 68.35 68.67 64.88 49.80 72.06 100.00 89.00 59.00 73.00 93.00 78.00 100.00 Legal rights index 4.12 4.00 4.75 3.60 4.06 0.00 3.00 7.00 4.00 4.00 .. 1.00 Credit information index 2.22 2.50 3.00 5.60 1.44 0.00 4.00 2.00 1.00 1.00 0.00 1.00 1Construction of index or variable explained in glossary of terms. * Figures for accounts per thousand adults are imputed. See chapter 9. Source: Getting Finance database. Cent. Congo, Latin African Dem. All countries South Asia East Asia America Africa Cambodia Cameroon Cape Verde Rep. Chad Côte d'Ivoire Rep. of Number of accounts No. of accounts per thousand adults1 463.87 512.27 854.91 618.12 240.21 32.08 78* 1138.98 21* 9* 71.41 2* No. of accounts per thousand inhabitants1 317.23 363.50 825.30 389.00 158.70 20.30 .. 694.92 .. .. 41.85 .. Opening an account Number of documents required 2.82 2.75 2.00 2.52 3.05 4.00 3.60 2.00 3.00 3.00 5.00 2.00 Third-party reference required (average yes/no) 0.38 0.40 0.10 0.72 0.39 0.60 0.00 0.00 .. 0.00 .. 1.00 Index of accessibility (0–1)1 0.42 0.41 0.41 0.35 0.44 0.47 0.43 0.58 0.33 0.46 0.50 0.33 At least one nonbranch option to apply for an account (0–1)1 0.40 0.37 0.33 0.18 0.45 0.40 0.60 1.00 0.00 0.60 1.00 0.00 At least one electronic option (0–1)1 0.17 0.10 0.16 0.04 0.20 0.20 0.00 0.50 0.00 0.00 1.00 0.00 Application fee? (average yes/no) 0.10 0.00 0.10 0.11 0.11 0.20 0.20 0.00 0.00 0.00 1.00 0.67 Other costs? (average yes/no) 0.18 0.16 0.10 0.09 0.22 0.40 0.20 0.00 0.00 1.00 0.00 0.00 Days to open an account 1.29 1.41 0.81 1.44 1.36 1.00 0.80 0.50 1.00 1.80 1.00 1.50 Convenience features Index of convenience features (0–1)1 0.52 0.48 0.39 0.66 0.55 0.38 0.60 0.50 0.75 0.65 0.25 0.38 Index of free usage features (0–1)1 0.52 0.54 0.53 0.48 0.52 0.56 0.55 0.40 0.60 0.52 0.80 0.60 Minimum balance required when opening a standard account? (average yes/no) 0.90 0.97 0.89 1.00 0.87 1.00 0.80 1.00 1.00 1.00 1.00 1.00 Mean minimum deposit for a standard account (fraction GDP per capita) 0.26 0.09 0.05 0.03 0.43 0.10 0.33 0.05 0.83 0.81 0.11 3.97 Minimum balance required when opening a savings account? (average yes/no) 0.93 0.93 0.94 1.00 0.92 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Minimum ongoing balance needed for a savings account? (average yes/no) 0.71 0.85 0.64 0.68 0.71 0.60 1.00 0.50 1.00 0.50 1.00 0.50 Fees and costs for standard domestic transactions (% monthly income) Cost per checkbook issued 9.26 5.95 3.72 2.00 14.42 16.39 0.00 3.09 30.10 10.88 .. 149.16 Annual fee for cash card/debit card 10.31 6.36 3.88 1.69 15.97 7.50 5.33 3.09 .. 26.28 .. 115.00 Cash card/debit card fees per transaction at merchants 0.16 0.05 0.02 0.08 0.22 0.00 0.52 .. .. .. .. 0.00 Fee for issuing a banker's draft 20.41 10.32 4.34 2.92 30.44 23.43 0.00 2.47 132.82 .. .. 115.00 Monthly fee for account maintenance 7.42 0.91 0.51 1.03 12.04 0.49 3.94 .. 9.30 5.55 13.33 69.00 Fees and costs for standard domestic transactions (US$) Cost per checkbook issued 3.48 2.15 3.88 2.19 4.11 6.92 0.00 5.97 9.56 6.23 .. 17.88 Annual fee for cash card/debit card 5.05 2.35 3.29 3.56 6.47 3.17 5.12 5.97 .. 15.06 .. 13.80 Cash card/debit card fees per transaction at merchants 0.13 0.03 0.02 0.08 0.18 0.00 0.50 .. .. .. .. 0.00 Fee for issuing a banker's draft 7.76 2.99 7.52 4.79 9.56 9.90 0.00 4.78 42.16 .. .. 13.80 Monthly fee for account maintenance 3.01 0.55 0.96 3.75 4.23 0.21 3.78 .. 2.95 3.18 11.04 5.64 Fees and costs for remittances (% monthly income) Fee to send payment via bankers' draft, foreign currency 63.59 15.13 14.56 10.27 98.26 35.41 43.46 9.28 189.74 83.21 .. 577.44 Fee to send payment via wire transfer 64.26 30.77 18.23 18.06 93.00 61.99 42.21 7.73 189.74 104.07 72.72 345.60 Fee to use money transfer operator to send money 65.05 28.57 12.06 4.37 84.37 30.69 17.92 17.32 .. 58.94 .. 120.00 Fee to receive a check issued abroad 54.06 7.75 12.50 4.76 86.13 73.80 37.30 6.19 24.19 90.73 35.15 576.00 Fee to receive payment via wire transfer 25.31 16.08 3.54 5.70 36.08 17.71 16.17 7.73 24.19 6.89 0.00 115.00 Fee to receive remittance through money transfer operator 19.17 4.80 1.17 0.00 29.37 0.00 0.00 17.32 .. 0.00 .. 0.00 Fees and costs for remittances (US$) Fee to send a $250 payment via banker's draft in foreign currency 23.14 6.12 17.26 0.90 31.82 14.96 41.76 17.91 60.23 47.68 .. 69.24 Fee to send $250 payment via wire transfer 26.75 12.60 20.55 23.78 32.23 26.19 40.56 14.93 60.23 59.63 60.23 41.54 Fee to use money transfer operator to send $250 18.39 7.86 13.25 4.08 24.36 12.96 17.22 33.44 .. 33.78 .. 14.40 Fee to receive $250 check issued abroad 17.94 3.63 9.29 4.06 25.60 31.17 35.84 11.94 7.68 51.99 29.11 69.24 Fee to receive $250 payment via wire transfer 9.54 5.56 4.89 5.95 12.40 7.48 15.54 14.93 7.68 3.95 0.00 13.84 Fee to receive $250 remittance through money transfer operator 5.44 0.91 1.76 0.04 7.96 0.00 0.00 33.44 .. 0.00 .. 0.00 INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 83 Cent. Congo, Latin African Dem. All countries South Asia East Asia America Africa Cambodia Cameroon Cape Verde Rep. Chad Côte d'Ivoire Rep. of Domestic and cross-border payments, time indexes Domestic payment time index (0–1)* 0.49 0.56 0.39 0.45 0.51 0.45 0.64 0.44 .. 0.27 0.62 0.36 Cross-border payment time index (0–1)* 0.44 0.38 0.40 0.43 0.46 0.41 0.43 0.45 0.26 0.51 0.46 0.30 Retail payments, availability and quality Range of payment services with standard bank account index (0–1)1 0.64 0.68 0.75 0.60 0.61 0.68 0.56 0.80 0.20 0.48 0.40 0.47 84 BANKING THE POOR Mobile banking technology index (0–1)1 0.26 0.36 0.33 0.22 0.23 0.20 0.24 0.30 0.00 0.16 0.40 0.20 Network quality and interoperability index (0–1)1 0.62 0.69 0.76 0.83 0.54 0.20 0.72 0.80 0.00 0.00 0.40 0.53 Retail payment channels index (0–1)1 0.47 0.43 0.65 0.50 0.44 0.39 0.27 0.69 0.17 0.26 0.17 0.37 Credit Business loan processing time (days) 7.42 8.02 7.64 13.41 6.53 10.33 12.67 6.00 .. 4.50 .. 3.50 Startup loan processing fee (%) 0.89 0.38 0.54 0.92 1.07 0.67 0.80 1.63 .. 0.30 .. 0.56 Collateral-to-loan-value ratio (%) 88.32 96.12 80.14 73.82 90.45 75.00 100.00 105.00 .. 100.00 .. 85.00 Maximum terms for startup loan (years) 4.55 4.43 3.31 6.25 4.60 3.83 4.50 3.00 .. 2.50 .. 1.50 Index of business loan application complexity1 0.66 0.68 0.73 0.77 0.75 0.83 0.50 0.88 0.00 0.88 0.00 1.00 Index of collateral flexibility* 0.48 0.66 0.80 0.52 0.71 1.00 0.81 0.00 0.00 0.63 0.00 0.72 Basic banking No opening fee (average yes/no) 0.77 0.81 0.80 0.94 0.74 0.80 1.00 0.50 0.00 1.00 1.00 1.00 No monthly fee (average yes/no) 0.61 0.83 0.73 0.76 0.53 0.80 0.40 0.50 0.00 0.60 0.00 1.00 No minimum balance (average yes/no) 0.52 0.42 0.51 0.54 0.53 0.80 0.40 0.50 1.00 1.00 0.00 1.00 Basic package of free transactions (average yes/no) 0.56 0.60 0.58 0.66 0.54 0.40 1.00 0.50 0.00 0.80 0.00 1.00 Basic banking (0–1)* 0.61 0.67 0.64 0.73 0.59 0.70 0.70 0.50 0.25 0.85 0.25 1.00 Government policy (commercial bank response) (average yes/no) 0.57 0.70 0.62 0.36 0.56 0.60 0.40 0.00 1.00 0.60 0.00 0.00 Government, offering basic banking1 0.23 0.33 0.25 0.20 0.22 0.00 0.00 0.00 0.00 0.00 1.00 .. Government, exempt basic accounts from ID requirements1 0.06 0.17 0.13 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 .. Savings schemes Doorstep collection1 0.08 0.17 0.25 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 .. Government-matched savings1 0.02 0.00 0.13 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 .. Tax incentives for savings1 0.42 0.33 0.75 0.40 0.38 0.00 1.00 1.00 1.00 1.00 1.00 .. Periodic deposit1 0.40 0.53 0.48 0.46 0.36 1.00 0.00 0.20 1.00 0.40 0.00 .. Transparency and consumer protection Advising applicants index (0–1)1 0.46 0.58 0.56 0.40 0.42 0.00 0.50 0.00 0.50 0.50 1.00 .. Additional document requirements index (0–1)1 0.10 0.17 0.13 0.13 0.08 0.13 0.13 0.13 0.13 0.13 0.00 .. Guidelines on credit documents index (0 –1)1 0.71 0.67 0.72 0.90 0.68 0.25 1.00 1.00 1.00 1.00 1.00 .. Transparency and consumer protection index (0–1)1 0.40 0.60 0.50 0.40 0.33 0.08 0.58 0.42 0.58 0.58 0.25 .. Firms using bank loans (% all firms) Firms with bank loans 31.06 50.80 .. 36.30 27.78 .. 42.11 46.94 .. .. .. .. Firms that use loans for investment 16.25 12.71 23.95 10.83 15.70 6.76 18.02 25.51 .. .. .. .. Firms that use loans for expenses 28.20 34.47 33.27 28.16 25.58 5.77 41.86 19.39 .. .. .. .. Firms that view access to finance as constraint 35.62 27.76 15.11 22.06 46.17 9.39 68.02 38.78 .. .. .. .. Small and micro firms using bank loans (% firms) Small firms with bank loans 27.23 37.00 28.00 27.00 27.00 4.08 32.00 19.35 .. .. .. 5.03 Micro firms with bank loans 20.33 34.00 21.00 22.00 19.00 3.49 22.58 18.92 .. .. .. 3.68 Small firms that use loans for working capital 10.76 13.63 13.18 12.41 9.82 1.41 10.55 11.26 .. .. .. 1.62 Micro firms that use loans for working capital 8.97 10.48 9.55 11.19 8.35 1.25 7.76 11.69 .. .. .. 1.42 Small firms that use loans for investment 13.65 10.41 16.87 16.80 12.51 1.79 5.37 18.32 .. .. .. 3.61 Micro firms that use loans for investment 11.20 6.25 13.77 15.27 10.06 1.79 4.75 15.14 .. .. .. 2.90 Small firms that view access to finance as constraint 75.25 34.00 53.00 74.00 81.00 99.75 80.81 96.94 .. .. .. 95.88 Micro firms that view access to finance as constraint 59.67 18.00 34.00 56.00 67.00 87.56 70.35 82.65 .. .. .. 88.82 Cent. Congo, Latin African Dem. All countries South Asia East Asia America Africa Cambodia Cameroon Cape Verde Rep. Chad Côte d'Ivoire Rep. of Demographic and socioeconomic variables Per capita GDP (US$) 2041.69 662.42 5486.02 3136.88 1334.40 501.22 1098.14 2207.17 362.87 654.95 946.51 143.97 Population (millions) 53.25 292.20 63.38 27.20 20.51 14.00 17.00 1.00 4.00 10.00 18.00 59.00 Percentage of the population aged 15 years and older 61.00 65.00 70.00 63.00 57.00 63.29 58.53 61.01 57.52 53.78 58.61 52.74 Population density (/sq km) 245.56 370.96 920.71 124.00 87.11 81.30 35.85 128.61 6.57 7.93 58.07 26.17 Urban population (%) 39.14 24.08 47.71 58.07 37.05 20.32 55.46 58.04 38.18 25.76 45.44 32.72 Literacy (15+) 64.17 53.40 85.79 79.46 58.89 70.47 67.90 81.22 48.57 25.65 48.73 67.17 Ratio of private credit to GDP 28.85 37.00 55.25 35.00 20.53 10.00 9.00 50.00 7.00 2.00 14.00 .. Banking assets covered (%) 68.35 68.67 64.88 49.80 72.06 74.00 66.00 .. 100.00 100.00 18.00 24.00 Legal rights index 4.12 4.00 4.75 3.60 4.06 0.00 3.00 5.00 3.00 3.00 .. .. Credit information index 2.22 2.50 3.00 5.60 1.44 0.00 2.00 3.00 2.00 1.00 .. .. 1 Construction of index or variable explained in glossary of terms. * Figures for accounts per thousand adults are imputed. See chapter 9. Source: Getting Finance database. INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 85 Latin Gambia, All countries South Asia East Asia America Africa El Salvador Ethiopia Gabon The Ghana Guatemala Honduras Number of accounts No. of accounts per thousand adults1 463.87 512.27 854.91 618.12 240.21 693.45 66* 183.24 164.81 204.12 909.72 624.80 No. of accounts per thousand inhabitants1 317.23 363.50 825.30 389.00 158.70 459.70 .. .. 97.24 125.33 519.65 378.51 Opening an account Number of documents required 2.82 2.75 2.00 2.52 3.05 2.25 2.00 2.75 4.00 2.80 2.00 1.00 Third-party reference required (average yes/no) 0.38 0.40 0.10 0.72 0.39 0.25 0.00 0.00 0.67 0.60 0.67 1.00 86 BANKING THE POOR Index of accessibility (0–1)1 0.42 0.41 0.41 0.35 0.44 0.25 0.33 0.33 0.39 0.37 0.33 0.40 At least one nonbranch option to apply for an account (0–1)1 0.40 0.37 0.33 0.18 0.45 0.00 0.00 0.00 0.25 0.40 0.00 0.40 At least one electronic option (0–1)1 0.17 0.10 0.16 0.04 0.20 0.00 0.00 0.00 0.25 0.20 0.00 0.20 Application fee? (average yes/no) 0.10 0.00 0.10 0.11 0.11 0.00 0.00 0.00 0.00 0.00 0.00 0.20 Other costs? (average yes/no) 0.18 0.16 0.10 0.09 0.22 0.25 0.00 0.00 0.33 0.40 0.00 0.20 Days to open an account 1.29 1.41 0.81 1.44 1.36 0.88 0.67 2.50 1.67 1.00 .. 0.88 Convenience features Index of convenience features (0–1)1 0.52 0.48 0.39 0.66 0.55 0.50 0.15 0.56 0.42 0.65 0.00 0.65 Index of free usage features (0–1)1 0.52 0.54 0.53 0.48 0.52 0.40 0.28 0.40 0.40 0.40 .. 0.56 Minimum balance required when opening a standard account? (average yes/no) 0.90 0.97 0.89 1.00 0.87 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Mean minimum deposit for a standard account (fraction GDP per capita) 0.26 0.09 0.05 0.03 0.43 0.00 0.02 0.11 0.33 0.06 0.03 0.02 Minimum balance required when opening a savings account? (average yes/no) 0.93 0.93 0.94 1.00 0.92 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Minimum ongoing balance needed for a savings account? (average yes/no) 0.71 0.85 0.64 0.68 0.71 0.25 0.40 1.00 1.00 0.60 .. 0.80 Fees and costs for standard domestic transactions (% monthly income) Cost per checkbook issued 9.26 5.95 3.72 2.00 14.42 0.16 21.14 1.69 6.37 9.11 0.92 4.97 Annual fee for cash card/debit card 10.31 6.36 3.88 1.69 15.97 0.84 18.87 2.53 0.00 36.43 3.46 0.00 Cash card/debit card fees per transaction at merchants 0.16 0.05 0.02 0.08 0.22 0.00 0.00 0.00 0.00 .. 0.00 0.40 Fee for issuing a banker's draft 20.41 10.32 4.34 2.92 30.44 0.05 5.28 1.35 48.49 18.21 2.02 3.47 Monthly fee for account maintenance 7.42 0.91 0.51 1.03 12.04 0.23 45.29 .. 0.20 9.47 0.58 1.61 Fees and costs for standard domestic transactions (US$) Cost per checkbook issued 3.48 2.15 3.88 2.19 4.11 0.34 3.19 10.04 1.75 4.35 2.10 5.24 Annual fee for cash card/debit card 5.05 2.35 3.29 3.56 6.47 1.83 2.85 15.06 0.00 17.39 7.87 0.00 Cash card/debit card fees per transaction at merchants 0.13 0.03 0.02 0.08 0.18 0.00 0.00 0.00 0.00 .. 0.00 0.42 Fee for issuing a banker's draft 7.76 2.99 7.52 4.79 9.56 0.11 0.80 8.03 13.30 8.70 4.59 3.65 Monthly fee for account maintenance 3.01 0.55 0.96 3.75 4.23 0.50 6.83 1.46 0.05 4.52 1.31 1.69 Fees and costs for remittances (% monthly income) Fee to send payment via bankers' draft, foreign currency 63.59 15.13 14.56 10.27 98.26 0.26 40.76 3.38 64.48 245.87 0.00 2.01 Fee to send payment via wire transfer 64.26 30.77 18.23 18.06 93.00 1.47 95.11 .. 70.46 132.04 12.29 39.98 Fee to use money transfer operator to send money 65.05 28.57 12.06 4.37 84.37 1.10 .. .. .. 56.91 0.06 10.45 Fee to receive a check issued abroad 54.06 7.75 12.50 4.76 86.13 0.68 18.12 2.99 16.25 145.70 0.00 4.77 Fee to receive payment via wire transfer 25.31 16.08 3.54 5.70 36.08 0.47 89.07 .. 11.44 22.77 2.31 11.85 Fee to receive remittance through money transfer operator 19.17 4.80 1.17 0.00 29.37 0.00 41.52 .. 0.00 0.00 0.06 .. Fees and costs for remittances (US$) Fee to send a $250 payment via banker's draft in foreign currency 23.14 6.12 17.26 0.90 31.82 0.57 6.15 20.08 17.68 99.67 0.00 2.12 Fee to send $250 payment via wire transfer 26.75 12.60 20.55 23.78 32.23 3.20 14.35 .. 19.32 63.04 27.95 42.12 Fee to use money transfer operator to send $250 18.39 7.86 13.25 4.08 24.36 2.40 .. .. .. 27.17 0.13 11.01 Fee to receive $250 check issued abroad 17.94 3.63 9.29 4.06 25.60 1.49 2.73 17.73 4.46 69.57 0.00 5.03 Fee to receive $250 payment via wire transfer 9.54 5.56 4.89 5.95 12.40 1.03 13.44 .. 3.14 10.87 5.25 12.49 Fee to receive $250 remittance through money transfer operator 5.44 0.91 1.76 0.04 7.96 0.00 6.26 .. 0.00 0.00 0.13 .. Latin Gambia, All countries South Asia East Asia America Africa El Salvador Ethiopia Gabon The Ghana Guatemala Honduras Domestic and cross-border payments, time indexes Domestic payment time index (0–1)* 0.49 0.56 0.39 0.45 0.51 .. 0.66 .. .. 0.64 .. 0.45 Cross-border payment time index (0–1)* 0.44 0.38 0.40 0.43 0.46 0.31 0.55 .. 0.54 0.44 .. 0.26 Retail payments, availability and quality Range of payment services with standard bank account index (0–1)1 0.64 0.68 0.75 0.60 0.61 0.65 0.40 0.65 0.40 0.56 0.13 0.84 Mobile banking technology index (0–1)1 0.26 0.36 0.33 0.22 0.23 0.30 0.12 0.00 0.40 0.16 0.00 0.60 Network quality and interoperability index (0–1)1 0.62 0.69 0.76 0.83 0.54 0.90 0.20 0.75 0.00 0.60 0.27 1.00 Retail payment channels index (0–1)1 0.47 0.43 0.65 0.50 0.44 0.47 0.33 0.60 0.38 0.38 0.06 0.64 Credit Business loan processing time (days) 7.42 8.02 7.64 13.41 6.53 12.50 7.00 1.00 7.33 5.00 15.00 11.13 Startup loan processing fee (%) 0.89 0.38 0.54 0.92 1.07 1.50 0.33 1.00 0.83 2.00 .. 0.25 Collateral-to-loan-value ratio (%) 88.32 96.12 80.14 73.82 90.45 75.00 116.67 .. 140.00 90.00 131.95 63.33 Maximum terms for startup loan (years) 4.55 4.43 3.31 6.25 4.60 6.00 6.00 .. 2.33 5.00 5.00 9.00 Index of business loan application complexity1 0.66 0.68 0.73 0.77 0.75 0.88 0.33 0.00 0.75 0.88 0.00 0.50 Index of collateral flexibility* 0.48 0.66 0.80 0.52 0.71 0.00 0.44 0.89 0.56 0.00 0.00 0.00 Basic banking No opening fee (average yes/no) 0.77 0.81 0.80 0.94 0.74 0.75 0.60 0.75 1.00 0.80 .. 1.00 No monthly fee (average yes/no) 0.61 0.83 0.73 0.76 0.53 0.75 0.60 0.75 0.67 1.00 .. 0.80 No minimum balance (average yes/no) 0.52 0.42 0.51 0.54 0.53 0.25 0.40 0.25 1.00 1.00 .. 0.40 Basic package of free transactions (average yes/no) 0.56 0.60 0.58 0.66 0.54 0.50 0.60 0.00 1.00 0.60 0.50 0.80 Basic banking (0–1)* 0.61 0.67 0.64 0.73 0.59 0.56 0.55 0.44 0.92 0.85 .. 0.75 Government policy (commercial bank response) (average yes/no) 0.57 0.70 0.62 0.36 0.56 0.25 0.20 0.00 1.00 1.00 .. 0.20 Government, offering basic banking1 0.23 0.33 0.25 0.20 0.22 0.00 .. 0.00 0.00 0.00 0.00 0.00 Government, exempt basic accounts from ID requirements1 0.06 0.17 0.13 0.00 0.03 0.00 .. 0.00 0.00 0.00 0.00 0.00 Savings schemes Doorstep collection1 0.08 0.17 0.25 0.00 0.03 0.00 .. 0.00 0.00 1.00 0.00 0.00 Government-matched savings1 0.02 0.00 0.13 0.00 0.00 0.00 .. 0.00 0.00 0.00 0.00 0.00 Tax incentives for savings1 0.42 0.33 0.75 0.40 0.38 1.00 .. 1.00 0.00 0.00 0.00 0.00 Periodic deposit1 0.40 0.53 0.48 0.46 0.36 0.75 .. 1.00 0.00 0.40 0.50 0.40 Transparency and consumer protection Advising applicants index (0–1)1 0.46 0.58 0.56 0.40 0.42 0.00 .. 0.50 0.50 0.00 1.00 0.00 Additional document requirements index (0–1)1 0.10 0.17 0.13 0.13 0.08 0.13 .. 0.13 0.00 0.00 0.13 0.25 Guidelines on credit documents index (0 –1)1 0.71 0.67 0.72 0.90 0.68 1.00 .. 1.00 0.25 0.00 1.00 0.75 Transparency and consumer protection index (0–1)1 0.40 0.60 0.50 0.40 0.33 0.33 .. 0.58 0.33 0.33 0.25 0.25 Firms using bank loans (% all firms) Firms with bank loans 31.06 50.80 .. 36.30 27.78 48.93 .. .. 16.55 22.18 33.56 46.87 Firms that use loans for investment 16.25 12.71 23.95 10.83 15.70 17.27 .. .. 7.59 16.02 12.81 8.51 Firms that use loans for expenses 28.20 34.47 33.27 28.16 25.58 39.04 .. .. 14.32 21.43 22.77 36.04 Firms that view access to finance as constraint 35.62 27.76 15.11 22.06 46.17 24.82 .. .. 40.32 66.24 20.50 26.71 Small and micro firms using bank loans (% firms) Small firms with bank loans 27.23 37.00 28.00 27.00 27.00 42.46 32.95 .. 13.29 12.26 26.61 32.89 Micro firms with bank loans 20.33 34.00 21.00 22.00 19.00 36.66 26.72 .. 7.14 10.00 24.69 25.34 Small firms that use loans for working capital 10.76 13.63 13.18 12.41 9.82 21.26 17.67 .. 4.51 3.56 10.12 16.38 Micro firms that use loans for working capital 8.97 10.48 9.55 11.19 8.35 19.84 14.16 .. 3.29 2.79 10.53 12.71 Small firms that use loans for investment 13.65 10.41 16.87 16.80 12.51 27.53 5.41 .. 8.00 3.11 16.32 16.10 Micro firms that use loans for investment 11.20 6.25 13.77 15.27 10.06 26.52 3.70 .. 4.75 2.49 16.32 12.13 Small firms that view access to finance as constraint 75.25 34.00 53.00 74.00 81.00 71.66 81.94 .. 92.53 94.97 71.51 69.91 Micro firms that view access to finance as constraint 59.67 18.00 34.00 56.00 67.00 51.19 65.91 .. 75.86 86.36 50.69 54.63 INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 87 Latin Gambia, All countries South Asia East Asia America Africa El Salvador Ethiopia Gabon The Ghana Guatemala Honduras Demographic and socioeconomic variables Per capita GDP (US$) 2041.69 662.42 5486.02 3136.88 1334.40 2618.81 183.19 6790.60 328.90 572.95 2735.13 1255.61 Population (millions) 53.25 292.20 63.38 27.20 20.51 7.00 73.00 1.00 2.00 23.00 13.00 7.00 Percentage of the population aged 15 years and older 61.00 65.00 70.00 63.00 57.00 66.29 55.85 64.61 59.00 61.40 57.12 60.58 Population density (/sq km) 245.56 370.96 920.71 124.00 87.11 337.44 72.71 5.46 155.27 99.03 118.99 65.73 Urban population (%) 39.14 24.08 47.71 58.07 37.05 60.10 16.28 84.08 54.74 48.54 47.66 46.96 88 BANKING THE POOR Literacy (15+) 64.17 53.40 85.79 79.46 58.89 80.64 35.90 84.02 43.00 57.90 69.10 80.01 Ratio of private credit to GDP 28.85 37.00 55.25 35.00 20.53 44.00 27.00 9.00 16.00 18.00 27.00 49.00 Banking assets covered (%) 68.35 68.67 64.88 49.80 72.06 60.00 94.00 100.00 .. 61.00 32.00 65.00 Legal rights index 4.12 4.00 4.75 3.60 4.06 3.00 .. 3.00 4.00 5.00 3.00 6.00 Credit information index 2.22 2.50 3.00 5.60 1.44 6.00 .. 2.00 0.00 0.00 5.00 6.00 1Construction of index or variable explained in glossary of terms. * Figures for accounts per thousand adults are imputed. See chapter 9. Source: Getting Finance database. Latin All countries South Asia East Asia America Africa India Indonesia Kenya Lao PDR Lesotho Liberia Madagascar Number of accounts No. of accounts per thousand adults1 463.87 512.27 854.91 618.12 240.21 656.78 464.47 169.39 108* 306.95 35.97 30.63 No. of accounts per thousand inhabitants1 317.23 363.50 825.30 389.00 158.70 443.09 334.31 97.18 .. .. 19.06 17.29 Opening an account Number of documents required 2.82 2.75 2.00 2.52 3.05 3.25 1.00 2.67 2.63 3.00 2.00 4.25 Third-party reference required (average yes/no) 0.38 0.40 0.10 0.72 0.39 0.60 0.00 0.33 0.20 0.40 0.40 0.25 Index of accessibility (0–1)1 0.42 0.41 0.41 0.35 0.44 0.43 0.37 0.53 0.30 0.33 0.37 0.33 At least one nonbranch option to apply for an account (0–1)1 0.40 0.37 0.33 0.18 0.45 0.40 0.20 0.67 0.20 0.00 0.20 0.00 At least one electronic option (0–1)1 0.17 0.10 0.16 0.04 0.20 0.00 0.00 0.50 0.00 0.00 0.20 0.00 Application fee? (average yes/no) 0.10 0.00 0.10 0.11 0.11 0.00 0.00 0.00 0.20 0.00 0.00 0.50 Other costs? (average yes/no) 0.18 0.16 0.10 0.09 0.22 0.00 0.20 0.17 0.20 0.00 0.00 0.00 Days to open an account 1.29 1.41 0.81 1.44 1.36 1.80 0.50 1.25 1.00 1.00 1.00 1.38 Convenience features Index of convenience features (0–1)1 0.52 0.48 0.39 0.66 0.55 0.75 0.38 0.79 0.30 0.50 0.10 0.56 Index of free usage features (0–1)1 0.52 0.54 0.53 0.48 0.52 0.56 0.48 0.57 0.48 0.36 0.20 0.70 Minimum balance required when opening a standard account? (average yes/no) 0.90 0.97 0.89 1.00 0.87 1.00 1.00 0.17 0.80 1.00 0.80 0.75 Mean minimum deposit for a standard account (fraction GDP per capita) 0.26 0.09 0.05 0.03 0.43 0.05 0.01 0.08 0.18 0.01 0.17 2.30 Minimum balance required when opening a savings account? (average yes/no) 0.93 0.93 0.94 1.00 0.92 1.00 1.00 0.67 1.00 1.00 0.80 0.75 Minimum ongoing balance needed for a savings account? (average yes/no) 0.71 0.85 0.64 0.68 0.71 1.00 1.00 0.60 1.00 0.60 0.60 0.50 Fees and costs for standard domestic transactions (% monthly income) Cost per checkbook issued 9.26 5.95 3.72 2.00 14.42 1.62 0.72 6.16 7.00 6.87 45.43 0.00 Annual fee for cash card/debit card 10.31 6.36 3.88 1.69 15.97 0.97 1.17 0.00 12.00 0.00 .. 46.66 Cash card/debit card fees per transaction at merchants 0.16 0.05 0.02 0.08 0.22 0.00 0.10 0.03 .. 0.43 .. 0.00 Fee for issuing a banker's draft 20.41 10.32 4.34 2.92 30.44 1.30 2.67 11.04 2.00 16.11 42.99 1.17 Monthly fee for account maintenance 7.42 0.91 0.51 1.03 12.04 0.00 1.22 11.23 1.00 5.91 51.52 3.36 Fees and costs for standard domestic transactions (US$) Cost per checkbook issued 3.48 2.15 3.88 2.19 4.11 1.13 1.00 3.21 3.59 4.59 6.89 0.00 Annual fee for cash card/debit card 5.05 2.35 3.29 3.56 6.47 0.68 1.62 0.00 6.16 0.00 .. 11.92 Cash card/debit card fees per transaction at merchants 0.13 0.03 0.02 0.08 0.18 0.00 0.14 0.01 .. 0.29 .. 0.00 Fee for issuing a banker's draft 7.76 2.99 7.52 4.79 9.56 0.90 3.70 5.76 1.03 10.76 6.52 0.30 Monthly fee for account maintenance 3.01 0.55 0.96 3.75 4.23 0.00 1.69 5.86 0.51 3.95 7.82 0.86 Fees and costs for remittances (% monthly income) Fee to send payment via bankers' draft, foreign currency 63.59 15.13 14.56 10.27 98.26 4.09 3.37 22.36 61.41 16.75 155.13 53.46 Fee to send payment via wire transfer 64.26 30.77 18.23 18.06 93.00 22.25 4.81 63.50 41.20 25.34 188.37 38.40 Fee to use money transfer operator to send money 65.05 28.57 12.06 4.37 84.37 0.00 4.11 338.20 44.84 .. 137.40 .. Fee to receive a check issued abroad 54.06 7.75 12.50 4.76 86.13 15.82 1.92 24.02 8.29 25.77 182.83 19.44 Fee to receive payment via wire transfer 25.31 16.08 3.54 5.70 36.08 7.31 1.52 19.05 6.17 10.31 166.21 16.03 Fee to receive remittance through money transfer operator 19.17 4.80 1.17 0.00 29.37 0.00 1.30 5.19 3.90 8.59 0.00 0.00 Fees and costs for remittances (US$) Fee to send a $250 payment via banker's draft in foreign currency 23.14 6.12 17.26 0.90 31.82 2.85 4.66 11.67 31.50 11.19 23.53 13.65 Fee to send $250 payment via wire transfer 26.75 12.60 20.55 23.78 32.23 15.48 6.65 33.14 21.13 16.93 28.57 9.81 Fee to use money transfer operator to send $250 18.39 7.86 13.25 4.08 24.36 0.00 5.68 17.10 23.00 .. 20.84 .. Fee to receive $250 check issued abroad 17.94 3.63 9.29 4.06 25.60 11.01 2.66 12.54 4.25 17.22 27.73 4.97 Fee to receive $250 payment via wire transfer 9.54 5.56 4.89 5.95 12.40 5.08 2.11 9.94 3.17 6.89 25.21 4.09 Fee to receive $250 remittance through money transfer operator 5.44 0.91 1.76 0.04 7.96 0.00 1.80 2.71 2.00 5.74 0.00 0.00 INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 89 Latin All countries South Asia East Asia America Africa India Indonesia Kenya Lao PDR Lesotho Liberia Madagascar Domestic and cross-border payments, time indexes Domestic payment time index (0–1)* 0.49 0.56 0.39 0.45 0.51 0.71 0.28 0.54 0.54 0.60 .. 0.65 Cross-border payment time index (0–1)* 0.44 0.38 0.40 0.43 0.46 0.40 0.16 0.36 0.24 0.71 0.78 0.48 Retail payments, availability and quality Range of payment services with standard bank account index (0–1)1 0.64 0.68 0.75 0.60 0.61 0.80 0.80 0.83 0.40 0.52 0.60 0.55 Mobile banking technology index (0–1)1 0.26 0.36 0.33 0.22 0.23 0.32 1.00 0.40 0.04 0.00 0.32 0.15 Network quality and interoperability index (0–1)1 0.62 0.69 0.76 0.83 0.54 1.00 1.00 0.93 0.20 0.48 0.00 0.40 90 BANKING THE POOR Retail payment channels index (0–1)1 0.47 0.43 0.65 0.50 0.44 0.60 0.61 0.62 0.30 0.33 0.24 0.31 Credit Business loan processing time (days) 7.42 8.02 7.64 13.41 6.53 4.75 3.00 6.00 7.00 2.25 17.50 10.67 Startup loan processing fee (%) 0.89 0.38 0.54 0.92 1.07 0.49 1.00 2.13 0.50 0.69 1.00 0.75 Collateral-to-loan-value ratio (%) 88.32 96.12 80.14 73.82 90.45 73.33 125.00 121.75 70.00 56.67 140.00 106.16 Maximum terms for startup loan (years) 4.55 4.43 3.31 6.25 4.60 5.00 3.00 4.00 3.00 4.33 1.75 5.00 Index of business loan application complexity1 0.66 0.68 0.73 0.77 0.75 0.42 1.00 0.81 0.65 1.00 0.92 0.75 Index of collateral flexibility* 0.48 0.66 0.80 0.52 0.71 0.33 0.00 0.70 0.67 0.89 0.00 0.00 Basic banking No opening fee (average yes/no) 0.77 0.81 0.80 0.94 0.74 1.00 0.40 1.00 0.20 0.60 0.40 0.33 No monthly fee (average yes/no) 0.61 0.83 0.73 0.76 0.53 1.00 0.20 1.00 0.20 0.40 1.00 0.00 No minimum balance (average yes/no) 0.52 0.42 0.51 0.54 0.53 1.00 0.20 0.33 0.00 0.20 0.60 0.50 Basic package of free transactions (average yes/no) 0.56 0.60 0.58 0.66 0.54 0.80 0.40 0.67 0.20 0.00 1.00 0.00 Basic banking (0–1)* 0.61 0.67 0.64 0.73 0.59 0.95 0.30 0.75 0.15 0.30 0.75 0.17 Government policy (commercial bank response) (average yes/no) 0.57 0.70 0.62 0.36 0.56 1.00 0.00 0.67 0.00 0.60 1.00 0.00 Government, offering basic banking1 0.23 0.33 0.25 0.20 0.22 1.00 0.00 0.00 0.00 0.00 0.00 0.00 Government, exempt basic accounts from ID requirements1 0.06 0.17 0.13 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Savings schemes Doorstep collection1 0.08 0.17 0.25 0.00 0.03 1.00 1.00 0.00 0.00 0.00 0.00 0.00 Government-matched savings1 0.02 0.00 0.13 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Tax incentives for savings1 0.42 0.33 0.75 0.40 0.38 1.00 1.00 0.00 0.00 0.00 0.00 0.00 Periodic deposit1 0.40 0.53 0.48 0.46 0.36 1.00 0.20 0.17 0.00 0.40 0.20 0.25 Transparency and consumer protection Advising applicants index (0–1)1 0.46 0.58 0.56 0.40 0.42 1.00 1.00 0.00 0.00 0.00 0.00 1.00 Additional document requirements index (0–1)1 0.10 0.17 0.13 0.13 0.08 0.13 0.13 0.00 0.13 0.00 0.00 1.00 Guidelines on credit documents index (0 –1)1 0.71 0.67 0.72 0.90 0.68 0.25 1.00 0.50 1.00 0.00 0.00 1.00 Transparency and consumer protection index (0–1)1 0.40 0.60 0.50 0.40 0.33 0.83 0.67 0.25 0.75 0.00 0.17 0.50 Firms using bank loans (% all firms) Firms with bank loans 31.06 50.80 .. 36.30 27.78 .. .. .. .. .. .. .. Firms that use loans for investment 16.25 12.71 23.95 10.83 15.70 19.41 13.88 25.70 13.82 6.67 .. 12.97 Firms that use loans for expenses 28.20 34.47 33.27 28.16 25.58 33.92 29.73 43.66 17.89 13.33 .. 19.80 Firms that view access to finance as constraint 35.62 27.76 15.11 22.06 46.17 15.83 17.53 44.07 19.11 39.44 .. 59.14 Small and micro firms using bank loans (% firms) Small firms with bank loans 27.23 37.00 28.00 27.00 27.00 .. 15.10 52.76 .. .. .. 12.78 Micro firms with bank loans 20.33 34.00 21.00 22.00 19.00 .. 5.41 48.61 .. .. .. 10.58 Small firms that use loans for working capital 10.76 13.63 13.18 12.41 9.82 .. 7.51 27.93 .. .. .. 6.28 Micro firms that use loans for working capital 8.97 10.48 9.55 11.19 8.35 .. 2.57 28.53 .. .. .. 4.47 Small firms that use loans for investment 13.65 10.41 16.87 16.80 12.51 .. 8.02 38.04 .. .. .. 10.68 Micro firms that use loans for investment 11.20 6.25 13.77 15.27 10.06 .. 3.21 38.72 .. .. .. 7.65 Small firms that view access to finance as constraint 75.25 34.00 53.00 74.00 81.00 .. 36.89 56.30 .. .. .. 71.68 Micro firms that view access to finance as constraint 59.67 18.00 34.00 56.00 67.00 .. 8.27 43.33 .. .. .. 53.41 Latin All countries South Asia East Asia America Africa India Indonesia Kenya Lao PDR Lesotho Liberia Madagascar Demographic and socioeconomic variables Per capita GDP (US$) 2041.69 662.42 5486.02 3136.88 1334.40 816.63 1634.22 602.96 590.46 825.05 186.68 288.10 Population (millions) 53.25 292.20 63.38 27.20 20.51 1110.00 223.00 35.00 6.00 2.00 3.00 19.00 Percentage of the population aged 15 years and older 61.00 65.00 70.00 63.00 57.00 67.46 71.98 57.37 61.05 59.86 52.99 56.44 Population density (/sq km) 245.56 370.96 920.71 124.00 87.11 373.27 123.12 61.75 24.95 58.94 35.09 32.82 Urban population (%) 39.14 24.08 47.71 58.07 37.05 28.98 49.22 21.00 21.00 18.98 58.78 27.08 Literacy (15+) 64.17 53.40 85.79 79.46 58.89 61.01 90.38 73.61 68.73 82.22 51.94 70.68 Ratio of private credit to GDP 28.85 37.00 55.25 35.00 20.53 46.00 25.00 28.00 6.00 9.00 8.00 10.00 Banking assets covered (%) 68.35 68.67 64.88 49.80 72.06 47.00 32.00 61.00 100.00 100.00 100.00 .. Legal rights index 4.12 4.00 4.75 3.60 4.06 6.00 5.00 8.00 2.00 5.00 4.00 1.00 Credit information index 2.22 2.50 3.00 5.60 1.44 4.00 3.00 4.00 0.00 0.00 0.00 0.00 1 Construction of index or variable explained in glossary of terms. * Figures for accounts per thousand adults are imputed. See chapter 9. Source: Getting Finance database. INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 91 Latin All countries South Asia East Asia America Africa Malawi Malaysia Mali Mauritius Mexico Mozambique Namibia Number of accounts No. of accounts per thousand adults1 463.87 512.27 854.91 618.12 240.21 107.72 2176.89 76.24 2011.11 630.69 102.68 322.69 No. of accounts per thousand inhabitants1 317.23 363.50 825.30 389.00 158.70 57.05 1503.11 39.93 1528.25 440.09 .. 199.07 Opening an account Number of documents required 2.82 2.75 2.00 2.52 3.05 2.75 1.20 3.25 2.00 2.33 2.00 3.00 Third-party reference required (average yes/no) 0.38 0.40 0.10 0.72 0.39 0.80 0.00 0.00 0.00 0.67 0.40 0.67 Index of accessibility (0–1)1 0.42 0.41 0.41 0.35 0.44 0.37 0.37 0.50 0.50 0.33 0.27 0.67 92 BANKING THE POOR At least one nonbranch option to apply for an account (0–1)1 0.40 0.37 0.33 0.18 0.45 0.60 0.20 0.50 0.67 0.00 0.00 0.50 At least one electronic option (0–1)1 0.17 0.10 0.16 0.04 0.20 0.20 0.20 0.00 0.67 0.00 0.00 0.50 Application fee? (average yes/no) 0.10 0.00 0.10 0.11 0.11 0.00 0.00 0.50 0.00 0.33 0.00 0.00 Other costs? (average yes/no) 0.18 0.16 0.10 0.09 0.22 0.25 0.00 0.25 0.00 0.00 0.00 0.67 Days to open an account 1.29 1.41 0.81 1.44 1.36 2.33 0.50 1.50 1.50 3.00 1.00 1.00 Convenience features Index of convenience features (0–1)1 0.52 0.48 0.39 0.66 0.55 0.75 0.25 0.67 0.50 0.75 0.55 0.67 Index of free usage features (0–1)1 0.52 0.54 0.53 0.48 0.52 0.32 0.48 0.65 0.53 0.47 0.28 0.33 Minimum balance required when opening a standard account? (average yes/no) 0.90 0.97 0.89 1.00 0.87 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Mean minimum deposit for a standard account (fraction GDP per capita) 0.26 0.09 0.05 0.03 0.43 0.12 0.01 0.22 0.02 0.02 0.12 0.01 Minimum balance required when opening a savings account? (average yes/no) 0.93 0.93 0.94 1.00 0.92 1.00 1.00 1.00 1.00 1.00 0.75 1.00 Minimum ongoing balance needed for a savings account? (average yes/no) 0.71 0.85 0.64 0.68 0.71 0.60 1.00 1.00 0.33 0.67 0.00 0.67 Fees and costs for standard domestic transactions (% monthly income) Cost per checkbook issued 9.26 5.95 3.72 2.00 14.42 33.20 0.28 1.69 0.54 0.00 20.68 1.54 Annual fee for cash card/debit card 10.31 6.36 3.88 1.69 15.97 0.78 1.08 26.92 0.00 0.79 35.34 0.00 Cash card/debit card fees per transaction at merchants 0.16 0.05 0.02 0.08 0.22 0.00 0.00 .. 0.00 0.00 0.00 0.17 Fee for issuing a banker's draft 20.41 10.32 4.34 2.92 30.44 70.26 0.17 0.00 0.77 1.37 13.09 4.91 Monthly fee for account maintenance 7.42 0.91 0.51 1.03 12.04 23.42 0.44 10.77 0.19 1.71 3.40 0.00 Fees and costs for standard domestic transactions (US$) Cost per checkbook issued 3.48 2.15 3.88 2.19 4.11 4.58 1.42 0.63 2.04 0.00 6.08 3.87 Annual fee for cash card/debit card 5.05 2.35 3.29 3.56 6.47 0.11 5.38 10.04 0.00 5.33 10.40 0.00 Cash card/debit card fees per transaction at merchants 0.13 0.03 0.02 0.08 0.18 0.00 0.00 .. 0.00 0.00 0.00 0.43 Fee for issuing a banker's draft 7.76 2.99 7.52 4.79 9.56 9.69 0.85 0.00 2.91 9.19 3.85 12.34 Monthly fee for account maintenance 3.01 0.55 0.96 3.75 4.23 3.23 2.18 4.02 0.73 11.49 1.00 0.00 Fees and costs for remittances (% monthly income) Fee to send payment via bankers' draft, foreign currency 63.59 15.13 14.56 10.27 98.26 184.54 0.28 .. 1.44 .. 103.79 15.86 Fee to send payment via wire transfer 64.26 30.77 18.23 18.06 93.00 323.13 1.47 .. 2.21 3.69 119.76 15.86 Fee to use money transfer operator to send money 65.05 28.57 12.06 4.37 84.37 249.75 0.79 82.28 2.11 .. 2.62 .. Fee to receive a check issued abroad 54.06 7.75 12.50 4.76 86.13 117.98 1.58 80.76 0.96 0.56 29.45 15.86 Fee to receive payment via wire transfer 25.31 16.08 3.54 5.70 36.08 84.10 1.47 67.30 0.67 0.60 8.90 15.86 Fee to receive remittance through money transfer operator 19.17 4.80 1.17 0.00 29.37 74.26 0.85 82.28 0.77 0.00 52.35 .. Fees and costs for remittances (US$) Fee to send a $250 payment via banker's draft in foreign currency 23.14 6.12 17.26 0.90 31.82 25.45 1.42 .. 5.47 .. 30.54 39.89 Fee to send $250 payment via wire transfer 26.75 12.60 20.55 23.78 32.23 44.56 7.37 .. 8.39 24.82 35.23 39.89 Fee to use money transfer operator to send $250 18.39 7.86 13.25 4.08 24.36 34.44 3.97 30.68 8.01 .. 0.77 .. Fee to receive $250 check issued abroad 17.94 3.63 9.29 4.06 25.60 16.27 7.93 30.12 3.64 3.79 8.66 39.89 Fee to receive $250 payment via wire transfer 9.54 5.56 4.89 5.95 12.40 11.60 7.37 25.10 2.56 4.04 2.62 39.89 Fee to receive $250 remittance through money transfer operator 5.44 0.91 1.76 0.04 7.96 10.24 4.25 30.68 2.91 0.00 15.40 .. Latin All countries South Asia East Asia America Africa Malawi Malaysia Mali Mauritius Mexico Mozambique Namibia Domestic and cross-border payments, time indexes Domestic payment time index (0–1)* 0.49 0.56 0.39 0.45 0.51 0.58 0.32 0.50 0.45 0.44 0.41 0.54 Cross-border payment time index (0–1)* 0.44 0.38 0.40 0.43 0.46 0.60 0.45 0.41 0.76 0.29 0.28 0.32 Retail payments, availability and quality Range of payment services with standard bank account index (0–1)1 0.64 0.68 0.75 0.60 0.61 0.44 0.84 0.55 0.87 0.80 0.76 0.70 Mobile banking technology index (0–1)1 0.26 0.36 0.33 0.22 0.23 0.08 0.28 0.20 0.33 0.20 0.32 0.40 Network quality and interoperability index (0–1)1 0.62 0.69 0.76 0.83 0.54 0.48 1.00 0.35 1.00 1.00 0.80 0.75 Retail payment channels index (0–1)1 0.47 0.43 0.65 0.50 0.44 0.51 0.78 0.32 0.43 0.63 0.66 0.60 Credit Business loan processing time (days) 7.42 8.02 7.64 13.41 6.53 6.00 3.00 8.00 5.00 .. 6.75 9.67 Startup loan processing fee (%) 0.89 0.38 0.54 0.92 1.07 1.25 0.00 1.25 0.67 .. 0.77 0.11 Collateral-to-loan-value ratio (%) 88.32 96.12 80.14 73.82 90.45 123.33 60.00 110.00 100.00 .. 43.36 35.00 Maximum terms for startup loan (years) 4.55 4.43 3.31 6.25 4.60 3.33 .. 8.00 6.83 .. 5.50 11.00 Index of business loan application complexity1 0.66 0.68 0.73 0.77 0.75 0.92 0.75 0.50 0.58 0.00 1.00 0.75 Index of collateral flexibility* 0.48 0.66 0.80 0.52 0.71 0.44 0.67 0.67 0.83 0.00 0.44 0.00 Basic banking No opening fee (average yes/no) 0.77 0.81 0.80 0.94 0.74 0.60 1.00 0.75 1.00 1.00 0.80 0.50 No monthly fee (average yes/no) 0.61 0.83 0.73 0.76 0.53 0.60 1.00 0.75 1.00 1.00 0.40 0.33 No minimum balance (average yes/no) 0.52 0.42 0.51 0.54 0.53 0.40 0.80 0.25 0.67 1.00 0.60 0.00 Basic package of free transactions (average yes/no) 0.56 0.60 0.58 0.66 0.54 0.80 1.00 0.75 0.67 1.00 0.20 1.00 Basic banking (0–1)* 0.61 0.67 0.64 0.73 0.59 0.60 0.95 0.63 0.83 1.00 0.50 0.50 Government policy (commercial bank response) (average yes/no) 0.57 0.70 0.62 0.36 0.56 0.60 1.00 0.25 0.33 1.00 0.80 1.00 Government, offering basic banking1 0.23 0.33 0.25 0.20 0.22 0.00 1.00 1.00 0.00 1.00 1.00 0.00 Government, exempt basic accounts from ID requirements1 0.06 0.17 0.13 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Savings schemes Doorstep collection1 0.08 0.17 0.25 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Government-matched savings1 0.02 0.00 0.13 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Tax incentives for savings1 0.42 0.33 0.75 0.40 0.38 0.00 1.00 1.00 0.00 1.00 0.00 0.00 Periodic deposit1 0.40 0.53 0.48 0.46 0.36 0.00 0.60 0.50 0.00 0.67 0.20 0.67 Transparency and consumer protection Advising applicants index (0–1)1 0.46 0.58 0.56 0.40 0.42 0.00 1.00 1.00 0.50 0.50 0.00 0.00 Additional document requirements index (0–1)1 0.10 0.17 0.13 0.13 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.38 Guidelines on credit documents index (0 –1)1 0.71 0.67 0.72 0.90 0.68 0.00 0.50 1.00 0.75 0.75 1.00 0.50 Transparency and consumer protection index (0–1)1 0.40 0.60 0.50 0.40 0.33 0.00 0.58 0.25 0.58 0.67 0.08 0.50 Firms using bank loans (% all firms) Firms with bank loans 31.06 50.80 .. 36.30 27.78 29.56 .. .. .. 11.39 .. 24.00 Firms that use loans for investment 16.25 12.71 23.95 10.83 15.70 20.62 23.84 16.77 36.32 2.61 .. 8.07 Firms that use loans for expenses 28.20 34.47 33.27 28.16 25.58 33.75 51.77 21.29 53.30 5.93 .. 19.54 Firms that view access to finance as constraint 35.62 27.76 15.11 22.06 46.17 42.77 17.77 56.29 33.00 15.09 .. 18.43 Small and micro firms using bank loans (% firms) Small firms with bank loans 27.23 37.00 28.00 27.00 27.00 23.19 .. 20.97 52.27 6.97 .. 19.00 Micro firms with bank loans 20.33 34.00 21.00 22.00 19.00 9.52 .. 13.33 52.63 5.24 .. 18.07 Small firms that use loans for working capital 10.76 13.63 13.18 12.41 9.82 9.52 .. 7.28 28.94 2.11 .. 6.65 Micro firms that use loans for working capital 8.97 10.48 9.55 11.19 8.35 2.62 .. 4.22 32.89 1.74 .. 5.89 Small firms that use loans for investment 13.65 10.41 16.87 16.80 12.51 13.16 .. 10.74 29.86 5.41 .. 18.12 Micro firms that use loans for investment 11.20 6.25 13.77 15.27 10.06 8.00 .. 7.33 27.76 4.67 .. 15.37 Small firms that view access to finance as constraint 75.25 34.00 53.00 74.00 81.00 53.46 .. 90.07 55.67 74.93 .. 92.40 Micro firms that view access to finance as constraint 59.67 18.00 34.00 56.00 67.00 30.82 .. 72.85 34.98 56.82 .. 77.51 INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 93 Latin All countries South Asia East Asia America Africa Malawi Malaysia Mali Mauritius Mexico Mozambique Namibia Demographic and socioeconomic variables Per capita GDP (US$) 2041.69 662.42 5486.02 3136.88 1334.40 169.57 5778.80 426.21 5144.27 8052.09 377.68 3107.05 Population (millions) 53.25 292.20 63.38 27.20 20.51 13.00 26.00 14.00 1.00 104.00 20.00 2.00 Percentage of the population aged 15 years and older 61.00 65.00 70.00 63.00 57.00 52.96 69.05 52.37 75.99 69.78 55.69 61.69 Population density (/sq km) 245.56 370.96 920.71 124.00 87.11 139.91 78.43 11.40 617.46 54.60 25.69 2.49 Urban population (%) 39.14 24.08 47.71 58.07 37.05 17.66 68.20 31.06 42.48 76.26 35.28 35.68 Literacy (15+) 64.17 53.40 85.79 79.46 58.89 64.13 88.69 21.52 84.30 90.85 39.00 85.04 94 BANKING THE POOR Ratio of private credit to GDP 28.85 37.00 55.25 35.00 20.53 12.00 113.00 17.00 78.00 22.00 12.00 64.00 Banking assets covered (%) 68.35 68.67 64.88 49.80 72.06 64.00 56.00 70.00 59.00 49.00 93.00 65.00 Legal rights index 4.12 4.00 4.75 3.60 4.06 7.00 8.00 3.00 5.00 3.00 3.00 5.00 Credit information index 2.22 2.50 3.00 5.60 1.44 0.00 6.00 1.00 1.00 6.00 3.00 5.00 1Construction of index or variable explained in glossary of terms. * Figures for accounts per thousand adults are imputed. See chapter 9. Source: Getting Finance database. Latin All countries South Asia East Asia America Africa Nepal Nicaragua Niger Nigeria Pakistan Philippines Rwanda Number of accounts No. of accounts per thousand adults1 463.87 512.27 854.91 618.12 240.21 178.81 231.95 15.53 185.05 268.32 565.95 34.86 No. of accounts per thousand inhabitants1 317.23 363.50 825.30 389.00 158.70 110.40 145.63 8.08 103.45 170.73 363.26 19.84 Opening an account Number of documents required 2.82 2.75 2.00 2.52 3.05 2.25 5.00 3.40 2.60 2.00 2.20 2.00 Third-party reference required (average yes/no) 0.38 0.40 0.10 0.72 0.39 0.20 1.00 0.00 0.20 0.00 0.00 0.50 Index of accessibility (0–1)1 0.42 0.41 0.41 0.35 0.44 0.40 0.42 0.47 0.47 0.27 0.33 0.42 At least one nonbranch option to apply for an account (0–1)1 0.40 0.37 0.33 0.18 0.45 0.40 0.50 0.40 0.40 0.00 0.00 0.50 At least one electronic option (0–1)1 0.17 0.10 0.16 0.04 0.20 0.00 0.00 0.20 0.40 0.00 0.00 0.50 Application fee? (average yes/no) 0.10 0.00 0.10 0.11 0.11 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Other costs? (average yes/no) 0.18 0.16 0.10 0.09 0.22 0.00 0.00 0.00 0.60 0.60 0.00 0.50 Days to open an account 1.29 1.41 0.81 1.44 1.36 1.00 1.00 0.80 1.80 2.63 1.00 1.00 Convenience features Index of convenience features (0–1)1 0.52 0.48 0.39 0.66 0.55 0.70 0.75 0.40 0.50 0.50 0.35 0.63 Index of free usage features (0–1)1 0.52 0.54 0.53 0.48 0.52 0.65 0.50 0.64 0.60 0.60 0.60 0.70 Minimum balance required when opening a standard account? (average yes/no) 0.90 0.97 0.89 1.00 0.87 1.00 1.00 0.80 1.00 1.00 0.80 0.00 Mean minimum deposit for a standard account (fraction GDP per capita) 0.26 0.09 0.05 0.03 0.43 0.16 0.08 0.49 0.05 0.10 0.01 0.00 Minimum balance required when opening a savings account? (average yes/no) 0.93 0.93 0.94 1.00 0.92 1.00 1.00 0.80 1.00 0.80 1.00 1.00 Minimum ongoing balance needed for a savings account? (average yes/no) 0.71 0.85 0.64 0.68 0.71 0.80 1.00 1.00 0.60 1.00 0.60 0.50 Fees and costs for standard domestic transactions (% monthly income) Cost per checkbook issued 9.26 5.95 3.72 2.00 14.42 3.42 3.94 0.00 8.20 2.10 3.10 17.34 Annual fee for cash card/debit card 10.31 6.36 3.88 1.69 15.97 19.93 3.34 0.00 5.71 5.44 0.00 .. Cash card/debit card fees per transaction at merchants 0.16 0.05 0.02 0.08 0.22 0.00 0.00 0.00 0.00 0.20 0.00 .. Fee for issuing a banker's draft 20.41 10.32 4.34 2.92 30.44 12.13 7.68 190.44 2.16 1.73 0.47 .. Monthly fee for account maintenance 7.42 0.91 0.51 1.03 12.04 0.00 .. 28.25 0.25 0.41 0.34 6.45 Fees and costs for standard domestic transactions (US$) Cost per checkbook issued 3.48 2.15 3.88 2.19 4.11 0.84 3.28 0.00 5.38 1.40 3.72 3.92 Annual fee for cash card/debit card 5.05 2.35 3.29 3.56 6.47 4.92 2.78 0.00 3.74 3.61 0.00 .. Cash card/debit card fees per transaction at merchants 0.13 0.03 0.02 0.08 0.18 0.00 0.00 0.00 0.00 0.13 0.00 .. Fee for issuing a banker's draft 7.76 2.99 7.52 4.79 9.56 3.00 6.39 40.96 1.42 1.15 0.57 .. Monthly fee for account maintenance 3.01 0.55 0.96 3.75 4.23 0.00 .. 6.08 0.16 0.27 0.41 1.46 Fees and costs for remittances (% monthly income) Fee to send payment via bankers' draft, foreign currency 63.59 15.13 14.56 10.27 98.26 20.33 .. 297.98 57.90 15.58 6.97 335.77 Fee to send payment via wire transfer 64.26 30.77 18.23 18.06 93.00 36.78 25.05 243.19 71.33 23.80 11.35 248.12 Fee to use money transfer operator to send money 65.05 28.57 12.06 4.37 84.37 .. 3.34 150.21 .. 11.58 8.56 235.21 Fee to receive a check issued abroad 54.06 7.75 12.50 4.76 86.13 21.35 12.02 210.12 82.27 4.85 2.98 223.36 Fee to receive payment via wire transfer 25.31 16.08 3.54 5.70 36.08 28.47 8.35 22.87 17.12 2.47 3.38 223.36 Fee to receive remittance through money transfer operator 19.17 4.80 1.17 0.00 29.37 6.83 .. 0.00 0.00 2.47 2.10 223.36 Fees and costs for remittances (US$) Fee to send a $250 payment via banker's draft in foreign currency 23.14 6.12 17.26 0.90 31.82 5.02 .. 64.09 37.97 10.34 8.39 75.90 Fee to send $250 payment via wire transfer 26.75 12.60 20.55 23.78 32.23 9.09 20.83 52.30 46.78 15.79 13.66 56.08 Fee to use money transfer operator to send $250 18.39 7.86 13.25 4.08 24.36 .. 2.78 32.31 .. 7.68 10.30 53.17 Fee to receive $250 check issued abroad 17.94 3.63 9.29 4.06 25.60 5.27 10.00 45.19 53.95 3.22 3.58 50.49 Fee to receive $250 payment via wire transfer 9.54 5.56 4.89 5.95 12.40 7.03 6.94 4.92 11.23 1.64 4.07 50.49 Fee to receive $250 remittance through money transfer operator 5.44 0.91 1.76 0.04 7.96 1.69 .. 0.00 0.00 1.64 2.52 50.49 INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 95 Latin All countries South Asia East Asia America Africa Nepal Nicaragua Niger Nigeria Pakistan Philippines Rwanda Domestic and cross-border payments, time indexes Domestic payment time index (0–1)* 0.49 0.56 0.39 0.45 0.51 0.52 .. 0.29 0.49 0.62 0.39 0.66 Cross-border payment time index (0–1)* 0.44 0.38 0.40 0.43 0.46 0.27 0.84 0.40 0.38 0.52 0.67 0.38 Retail payments, availability and quality Range of payment services with standard bank account index (0–1)1 0.64 0.68 0.75 0.60 0.61 0.68 0.60 0.40 0.88 0.76 0.68 0.70 Mobile banking technology index (0–1)1 0.26 0.36 0.33 0.22 0.23 0.52 0.00 0.12 0.60 0.32 0.32 0.00 Network quality and interoperability index (0–1)1 0.62 0.69 0.76 0.83 0.54 0.72 1.00 0.08 1.00 1.00 1.00 1.00 96 BANKING THE POOR Retail payment channels index (0–1)1 0.47 0.43 0.65 0.50 0.44 0.38 0.72 0.18 0.70 0.48 0.83 0.39 Credit Business loan processing time (days) 7.42 8.02 7.64 13.41 6.53 7.50 15.00 7.67 8.67 11.67 5.00 10.00 Startup loan processing fee (%) 0.89 0.38 0.54 0.92 1.07 0.75 1.00 0.00 1.46 0.24 0.11 1.00 Collateral-to-loan-value ratio (%) 88.32 96.12 80.14 73.82 90.45 110.75 25.00 100.00 72.50 136.67 70.00 60.00 Maximum terms for startup loan (years) 4.55 4.43 3.31 6.25 4.60 4.33 5.00 4.00 3.00 5.33 4.00 .. Index of business loan application complexity1 0.66 0.68 0.73 0.77 0.75 0.67 1.00 0.33 0.92 0.67 0.88 0.75 Index of collateral flexibility* 0.48 0.66 0.80 0.52 0.71 0.67 0.00 0.89 0.83 0.83 0.94 0.00 Basic banking No opening fee (average yes/no) 0.77 0.81 0.80 0.94 0.74 0.20 1.00 0.80 0.60 1.00 1.00 1.00 No monthly fee (average yes/no) 0.61 0.83 0.73 0.76 0.53 0.20 0.50 0.40 0.40 1.00 1.00 0.50 No minimum balance (average yes/no) 0.52 0.42 0.51 0.54 0.53 0.00 0.50 0.40 0.40 0.60 0.80 0.50 Basic package of free transactions (average yes/no) 0.56 0.60 0.58 0.66 0.54 0.20 0.50 1.00 0.20 1.00 1.00 1.00 Basic banking (0–1)* 0.61 0.67 0.64 0.73 0.59 0.15 0.63 0.65 0.40 0.90 0.95 0.75 Government policy (commercial bank response) (average yes/no) 0.57 0.70 0.62 0.36 0.56 0.20 0.00 1.00 0.80 1.00 1.00 1.00 Government, offering basic banking1 0.23 0.33 0.25 0.20 0.22 0.00 0.00 1.00 0.00 1.00 0.00 0.00 Government, exempt basic accounts from ID requirements1 0.06 0.17 0.13 0.00 0.03 0.00 0.00 0.00 0.00 0.00 1.00 0.00 Savings schemes Doorstep collection1 0.08 0.17 0.25 0.00 0.03 0.00 0.00 0.00 0.00 0.00 1.00 0.00 Government-matched savings1 0.02 0.00 0.13 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Tax incentives for savings1 0.42 0.33 0.75 0.40 0.38 0.00 0.00 1.00 0.00 0.00 1.00 0.00 Periodic deposit1 0.40 0.53 0.48 0.46 0.36 0.20 0.00 0.20 0.20 0.00 0.40 0.50 Transparency and consumer protection Advising applicants index (0–1)1 0.46 0.58 0.56 0.40 0.42 1.00 0.50 1.00 0.00 0.50 0.50 0.00 Additional document requirements index (0–1)1 0.10 0.17 0.13 0.13 0.08 0.00 0.13 0.00 0.13 0.13 0.13 0.00 Guidelines on credit documents index (0 –1)1 0.71 0.67 0.72 0.90 0.68 1.00 1.00 1.00 0.50 0.75 1.00 1.00 Transparency and consumer protection index (0–1)1 0.40 0.60 0.50 0.40 0.33 0.42 0.50 0.25 0.58 0.67 0.42 0.00 Firms using bank loans (% all firms) Firms with bank loans 31.06 50.80 .. 36.30 27.78 .. 40.73 48.00 .. .. .. 37.58 Firms that use loans for investment 16.25 12.71 23.95 10.83 15.70 .. 12.97 14.40 .. 3.63 5.45 15.90 Firms that use loans for expenses 28.20 34.47 33.27 28.16 25.58 .. 37.04 26.40 .. 13.99 17.18 32.50 Firms that view access to finance as constraint 35.62 27.76 15.11 22.06 46.17 .. 23.16 55.65 .. 37.55 13.52 35.95 Small and micro firms using bank loans (% firms) Small firms with bank loans 27.23 37.00 28.00 27.00 27.00 .. 24.68 28.70 .. .. 15.62 27.51 Micro firms with bank loans 20.33 34.00 21.00 22.00 19.00 .. 20.46 27.37 .. .. 13.46 19.62 Small firms that use loans for working capital 10.76 13.63 13.18 12.41 9.82 .. 12.17 11.35 .. .. 6.25 12.55 Micro firms that use loans for working capital 8.97 10.48 9.55 11.19 8.35 .. 11.11 10.92 .. .. 5.01 9.92 Small firms that use loans for investment 13.65 10.41 16.87 16.80 12.51 .. 18.61 7.71 .. .. 7.93 16.18 Micro firms that use loans for investment 11.20 6.25 13.77 15.27 10.06 .. 16.72 4.58 .. .. 6.45 12.28 Small firms that view access to finance as constraint 75.25 34.00 53.00 74.00 81.00 .. 84.29 93.55 .. .. 51.60 89.62 Micro firms that view access to finance as constraint 59.67 18.00 34.00 56.00 67.00 .. 67.09 83.87 .. .. 34.74 78.30 Latin All countries South Asia East Asia America Africa Nepal Nicaragua Niger Nigeria Pakistan Philippines Rwanda Demographic and socioeconomic variables Per capita GDP (US$) 2041.69 662.42 5486.02 3136.88 1334.40 291.13 1022.77 245.83 792.41 810.05 1381.96 269.79 Population (millions) 53.25 292.20 63.38 27.20 20.51 28.00 5.00 14.00 145.00 159.00 85.00 9.00 Percentage of the population aged 15 years and older 61.00 65.00 70.00 63.00 57.00 61.54 62.79 52.02 55.90 63.63 64.19 56.92 Population density (/sq km) 245.56 370.96 920.71 124.00 87.11 193.41 43.24 11.38 158.90 206.26 283.70 374.72 Urban population (%) 39.14 24.08 47.71 58.07 37.05 16.28 59.38 17.00 49.00 35.32 63.44 20.22 Literacy (15+) 64.17 53.40 85.79 79.46 58.89 48.59 76.68 28.67 69.12 44.64 92.60 64.90 Ratio of private credit to GDP 28.85 37.00 55.25 35.00 20.53 .. 33.00 9.00 .. 29.00 30.00 .. Banking assets covered (%) 68.35 68.67 64.88 49.80 72.06 62.00 43.00 100.00 58.00 60.00 57.00 35.00 Legal rights index 4.12 4.00 4.75 3.60 4.06 4.00 3.00 3.00 7.00 4.00 3.00 1.00 Credit information index 2.22 2.50 3.00 5.60 1.44 2.00 5.00 1.00 0.00 4.00 3.00 2.00 1 Construction of index or variable explained in glossary of terms. * Figures for accounts per thousand adults are imputed. See chapter 9. Source: Getting Finance database. INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 97 Latin Sierra All countries South Asia East Asia America Africa Senegal Leone Singapore South Africa Sri Lanka Sudan Swaziland Number of accounts No. of accounts per thousand adults1 463.87 512.27 854.91 618.12 240.21 78.98 65.46 2057.76 551.63 1577.67 144* 482.96 No. of accounts per thousand inhabitants1 317.23 363.50 825.30 389.00 158.70 45.87 37.42 1671.00 375.42 1203.14 .. 293.59 Opening an account Number of documents required 2.82 2.75 2.00 2.52 3.05 3.75 3.20 1.00 3.50 2.00 3.67 3.60 Third-party reference required (average yes/no) 0.38 0.40 0.10 0.72 0.39 0.33 0.80 0.00 0.00 0.00 0.80 0.60 Index of accessibility (0–1)1 0.42 0.41 0.41 0.35 0.44 0.62 0.40 0.67 0.67 0.57 0.33 0.40 98 BANKING THE POOR At least one nonbranch option to apply for an account (0–1)1 0.40 0.37 0.33 0.18 0.45 0.75 0.40 1.00 1.00 0.80 0.33 0.40 At least one electronic option (0–1)1 0.17 0.10 0.16 0.04 0.20 0.50 0.40 0.50 0.00 0.60 0.17 0.00 Application fee? (average yes/no) 0.10 0.00 0.10 0.11 0.11 0.00 0.20 0.00 0.00 0.00 0.20 0.00 Other costs? (average yes/no) 0.18 0.16 0.10 0.09 0.22 0.00 0.40 0.00 0.00 0.00 0.67 0.40 Days to open an account 1.29 1.41 0.81 1.44 1.36 1.25 1.00 1.00 2.83 0.75 2.17 1.20 Convenience features Index of convenience features (0–1)1 0.52 0.48 0.39 0.66 0.55 0.81 0.60 0.63 0.58 0.35 0.38 0.45 Index of free usage features (0–1)1 0.52 0.54 0.53 0.48 0.52 0.67 0.52 0.60 0.33 0.48 0.56 0.44 Minimum balance required when opening a standard account? (average yes/no) 0.90 0.97 0.89 1.00 0.87 1.00 1.00 0.50 0.33 0.80 0.67 0.80 Mean minimum deposit for a standard account (fraction GDP per capita) 0.26 0.09 0.05 0.03 0.43 0.17 0.29 0.03 0.00 0.01 0.29 0.00 Minimum balance required when opening a savings account? (average yes/no) 0.93 0.93 0.94 1.00 0.92 1.00 1.00 0.50 1.00 0.80 0.00 1.00 Minimum ongoing balance needed for a savings account? (average yes/no) 0.71 0.85 0.64 0.68 0.71 1.00 1.00 0.00 1.00 1.00 0.00 1.00 Fees and costs for standard domestic transactions (% monthly income) Cost per checkbook issued 9.26 5.95 3.72 2.00 14.42 0.00 22.97 0.38 0.16 3.05 10.59 1.06 Annual fee for cash card/debit card 10.31 6.36 3.88 1.69 15.97 46.61 0.00 .. 1.81 1.06 0.00 3.62 Cash card/debit card fees per transaction at merchants 0.16 0.05 0.02 0.08 0.22 0.72 0.00 .. 0.10 .. 0.59 0.38 Fee for issuing a banker's draft 20.41 10.32 4.34 2.92 30.44 15.32 41.57 1.38 1.32 1.95 10.59 6.93 Monthly fee for account maintenance 7.42 0.91 0.51 1.03 12.04 13.74 7.08 0.05 0.50 1.69 17.64 1.06 Fees and costs for standard domestic transactions (US$) Cost per checkbook issued 3.48 2.15 3.88 2.19 4.11 0.00 4.88 9.80 0.72 3.34 8.96 2.01 Annual fee for cash card/debit card 5.05 2.35 3.29 3.56 6.47 30.55 0.00 .. 7.89 1.16 0.00 6.89 Cash card/debit card fees per transaction at merchants 0.13 0.03 0.02 0.08 0.18 0.47 0.00 .. 0.43 .. 0.50 0.72 Fee for issuing a banker's draft 7.76 2.99 7.52 4.79 9.56 10.04 8.83 35.95 5.74 2.14 8.96 13.20 Monthly fee for account maintenance 3.01 0.55 0.96 3.75 4.23 9.00 1.50 1.31 2.18 1.86 14.93 2.01 Fees and costs for remittances (% monthly income) Fee to send payment via bankers' draft, foreign currency 63.59 15.13 14.56 10.27 98.26 .. 48.61 2.11 4.94 4.91 42.35 6.56 Fee to send payment via wire transfer 64.26 30.77 18.23 18.06 93.00 45.95 57.56 2.11 4.71 10.33 58.81 8.67 Fee to use money transfer operator to send money 65.05 28.57 12.06 4.37 84.37 .. 35.32 .. 3.46 .. 88.22 9.04 Fee to receive a check issued abroad 54.06 7.75 12.50 4.76 86.13 .. 174.08 .. 3.46 1.27 .. 3.77 Fee to receive payment via wire transfer 25.31 16.08 3.54 5.70 36.08 19.71 29.59 0.25 3.23 2.96 .. 3.77 Fee to receive remittance through money transfer operator 19.17 4.80 1.17 0.00 29.37 .. 0.00 .. 3.46 0.64 .. .. Fees and costs for remittances (US$) Fee to send a $250 payment via banker's draft in foreign currency 23.14 6.12 17.26 0.90 31.82 .. 10.32 55.00 21.52 5.38 35.82 12.48 Fee to send $250 payment via wire transfer 26.75 12.60 20.55 23.78 32.23 30.12 12.22 55.00 20.52 11.33 49.75 16.50 Fee to use money transfer operator to send $250 18.39 7.86 13.25 4.08 24.36 .. 7.50 .. 15.06 .. 74.63 17.22 Fee to receive $250 check issued abroad 17.94 3.63 9.29 4.06 25.60 .. 36.96 .. 15.06 1.39 .. 7.17 Fee to receive $250 payment via wire transfer 9.54 5.56 4.89 5.95 12.40 12.92 6.28 6.54 14.06 3.25 .. 7.17 Fee to receive $250 remittance through money transfer operator 5.44 0.91 1.76 0.04 7.96 .. 0.00 .. 15.06 0.70 .. .. Latin Sierra All countries South Asia East Asia America Africa Senegal Leone Singapore South Africa Sri Lanka Sudan Swaziland Domestic and cross-border payments, time indexes Domestic payment time index (0–1)* 0.49 0.56 0.39 0.45 0.51 0.56 0.45 0.47 0.55 0.60 0.31 0.50 Cross-border payment time index (0–1)* 0.44 0.38 0.40 0.43 0.46 0.66 0.48 0.50 0.54 0.38 0.17 0.39 Retail payments, availability and quality Range of payment services with standard bank account index (0–1)1 0.64 0.68 0.75 0.60 0.61 0.70 0.40 1.00 0.93 0.84 0.60 0.48 Mobile banking technology index (0–1)1 0.26 0.36 0.33 0.22 0.23 0.55 0.36 0.30 0.73 0.64 0.20 0.04 Network quality and interoperability index (0–1)1 0.62 0.69 0.76 0.83 0.54 0.60 0.20 1.00 1.00 0.60 0.70 0.48 Retail payment channels index (0–1)1 0.47 0.43 0.65 0.50 0.44 0.43 0.36 0.78 0.95 0.51 0.48 0.57 Credit Business loan processing time (days) 7.42 8.02 7.64 13.41 6.53 6.00 3.25 5.00 2.00 13.20 3.33 4.75 Startup loan processing fee (%) 0.89 0.38 0.54 0.92 1.07 0.60 1.63 0.00 0.74 0.20 1.67 1.00 Collateral-to-loan-value ratio (%) 88.32 96.12 80.14 73.82 90.45 116.67 110.00 .. 10.00 76.78 121.00 25.00 Maximum terms for startup loan (years) 4.55 4.43 3.31 6.25 4.60 6.50 2.50 .. 5.00 6.40 1.70 5.75 Index of business loan application complexity1 0.66 0.68 0.73 0.77 0.75 0.42 1.00 0.50 0.58 0.80 0.75 0.75 Index of collateral flexibility* 0.48 0.66 0.80 0.52 0.71 0.67 0.70 0.78 0.72 0.61 0.69 0.83 Basic banking No opening fee (average yes/no) 0.77 0.81 0.80 0.94 0.74 0.75 0.60 1.00 1.00 0.80 1.00 0.60 No monthly fee (average yes/no) 0.61 0.83 0.73 0.76 0.53 0.25 0.80 1.00 1.00 0.80 1.00 0.40 No minimum balance (average yes/no) 0.52 0.42 0.51 0.54 0.53 1.00 0.20 0.50 0.67 0.00 1.00 0.20 Basic package of free transactions (average yes/no) 0.56 0.60 0.58 0.66 0.54 0.00 0.60 1.00 0.67 0.40 0.17 0.20 Basic banking (0–1)* 0.61 0.67 0.64 0.73 0.59 0.50 0.55 0.75 0.83 0.50 0.79 0.35 Government policy (commercial bank response) (average yes/no) 0.57 0.70 0.62 0.36 0.56 0.50 0.80 1.00 1.00 0.40 1.00 0.40 Government, offering basic banking1 0.23 0.33 0.25 0.20 0.22 1.00 0.00 0.00 0.00 0.00 0.00 0.00 Government, exempt basic accounts from ID requirements1 0.06 0.17 0.13 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Savings schemes Doorstep collection1 0.08 0.17 0.25 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Government-matched savings1 0.02 0.00 0.13 0.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 0.00 Tax incentives for savings1 0.42 0.33 0.75 0.40 0.38 1.00 0.00 1.00 1.00 1.00 0.00 0.00 Periodic deposit1 0.40 0.53 0.48 0.46 0.36 1.00 0.20 0.00 0.67 0.60 0.00 0.60 Transparency and consumer protection Advising applicants index (0–1)1 0.46 0.58 0.56 0.40 0.42 1.00 0.00 0.50 1.00 1.00 1.00 0.00 Additional document requirements index (0–1)1 0.10 0.17 0.13 0.13 0.08 0.00 0.00 0.50 0.00 0.25 0.00 0.00 Guidelines on credit documents index (0 –1)1 0.71 0.67 0.72 0.90 0.68 1.00 0.00 0.25 0.50 1.00 1.00 0.50 Transparency and consumer protection index (0–1)1 0.40 0.60 0.50 0.40 0.33 0.25 0.17 0.17 0.42 0.83 0.67 0.08 Firms using bank loans (% all firms) Firms with bank loans 31.06 50.80 .. 36.30 27.78 .. .. .. .. .. .. 21.92 Firms that use loans for investment 16.25 12.71 23.95 10.83 15.70 26.34 .. .. 24.21 16.15 .. 7.72 Firms that use loans for expenses 28.20 34.47 33.27 28.16 25.58 25.19 .. .. 43.45 46.90 .. 15.97 Firms that view access to finance as constraint 35.62 27.76 15.11 22.06 46.17 55.43 .. .. 12.60 15.11 .. 32.89 Small and micro firms using bank loans (% firms) Small firms with bank loans 27.23 37.00 28.00 27.00 27.00 20.77 .. .. 46.45 37.01 .. 18.22 Micro firms with bank loans 20.33 34.00 21.00 22.00 19.00 14.73 .. .. 37.31 34.00 .. 16.74 Small firms that use loans for working capital 10.76 13.63 13.18 12.41 9.82 10.47 .. .. 16.75 13.63 .. 6.73 Micro firms that use loans for working capital 8.97 10.48 9.55 11.19 8.35 6.51 .. .. 14.39 10.48 .. 6.47 Small firms that use loans for investment 13.65 10.41 16.87 16.80 12.51 14.66 .. .. 18.95 10.41 .. 13.36 Micro firms that use loans for investment 11.20 6.25 13.77 15.27 10.06 9.82 .. .. 11.94 6.25 .. 11.24 Small firms that view access to finance as constraint 75.25 34.00 53.00 74.00 81.00 77.52 .. .. 32.01 34.22 .. 85.67 Micro firms that view access to finance as constraint 59.67 18.00 34.00 56.00 67.00 62.02 .. .. 14.10 17.56 .. 75.24 INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 99 Latin Sierra All countries South Asia East Asia America Africa Senegal Leone Singapore South Africa Sri Lanka Sudan Swaziland Demographic and socioeconomic variables Per capita GDP (US$) 2041.69 662.42 5486.02 3136.88 1334.40 749.15 255.80 30091.90 5380.78 1364.10 1015.04 2351.36 Population (millions) 53.25 292.20 63.38 27.20 20.51 12.00 6.00 4.00 47.00 20.00 37.00 1.00 Percentage of the population aged 15 years and older 61.00 65.00 70.00 63.00 57.00 58.08 57.17 81.20 68.06 76.26 59.67 60.79 Population density (/sq km) 245.56 370.96 920.71 124.00 87.11 61.95 78.76 6376.21 39.02 305.92 15.57 65.47 Urban population (%) 39.14 24.08 47.71 58.07 37.05 41.86 41.44 100.00 59.78 15.10 41.68 24.38 Literacy (15+) 64.17 53.40 85.79 79.46 58.89 39.28 34.83 92.55 82.00 90.68 60.93 79.56 100 BANKING THE POOR Ratio of private credit to GDP 28.85 37.00 55.25 35.00 20.53 24.00 4.00 99.00 78.00 .. 13.00 21.00 Banking assets covered (%) 68.35 68.67 64.88 49.80 72.06 55.00 100.00 61.00 69.00 84.00 17.00 79.00 Legal rights index 4.12 4.00 4.75 3.60 4.06 3.00 5.00 9.00 5.00 3.00 4.00 5.00 Credit information index 2.22 2.50 3.00 5.60 1.44 1.00 0.00 4.00 6.00 3.00 0.00 5.00 1Construction of index or variable explained in glossary of terms. * Figures for accounts per thousand adults are imputed. See chapter 9. Source: Getting Finance database. Latin All countries South Asia East Asia America Africa Tanzania Thailand Uganda Vietnam Zambia Number of accounts No. of accounts per thousand adults1 463.87 512.27 854.91 618.12 240.21 159.07 1351.68 141.37 83* 114.60 No. of accounts per thousand inhabitants1 317.23 363.50 825.30 389.00 158.70 88.49 1062.35 71.74 .. 62.00 Opening an account Number of documents required 2.82 2.75 2.00 2.52 3.05 3.83 1.00 2.67 3.00 4.40 Third-party reference required (average yes/no) 0.38 0.40 0.10 0.72 0.39 1.00 0.00 0.67 0.00 1.00 Index of accessibility (0–1)1 0.42 0.41 0.41 0.35 0.44 0.47 0.33 0.39 0.46 0.43 At least one nonbranch option to apply for an account (0–1)1 0.40 0.37 0.33 0.18 0.45 0.67 0.00 0.33 0.60 0.40 At least one electronic option (0–1)1 0.17 0.10 0.16 0.04 0.20 0.17 0.00 0.00 0.40 0.00 Application fee? (average yes/no) 0.10 0.00 0.10 0.11 0.11 0.00 0.00 0.00 0.40 0.00 Other costs? (average yes/no) 0.18 0.16 0.10 0.09 0.22 0.50 0.00 0.00 0.00 0.60 Days to open an account 1.29 1.41 0.81 1.44 1.36 1.67 0.50 3.00 1.00 1.20 Convenience features Index of convenience features (0–1)1 0.52 0.48 0.39 0.66 0.55 0.54 0.33 0.67 0.50 0.60 Index of free usage features (0–1)1 0.52 0.54 0.53 0.48 0.52 0.50 0.40 0.70 0.60 0.40 Minimum balance required when opening a standard account? (average yes/no) 0.90 0.97 0.89 1.00 0.87 1.00 1.00 1.00 1.00 0.80 Mean minimum deposit for a standard account (fraction GDP per capita) 0.26 0.09 0.05 0.03 0.43 0.06 0.00 0.13 0.01 0.05 Minimum balance required when opening a savings account? (average yes/no) 0.93 0.93 0.94 1.00 0.92 1.00 1.00 1.00 1.00 1.00 Minimum ongoing balance needed for a savings account? (average yes/no) 0.71 0.85 0.64 0.68 0.71 1.00 0.40 1.00 0.50 1.00 Fees and costs for standard domestic transactions (% monthly income) Cost per checkbook issued 9.26 5.95 3.72 2.00 14.42 25.80 1.57 8.43 0.28 7.11 Annual fee for cash card/debit card 10.31 6.36 3.88 1.69 15.97 15.69 1.59 15.80 3.80 0.00 Cash card/debit card fees per transaction at merchants 0.16 0.05 0.02 0.08 0.22 0.89 0.00 0.84 0.00 0.49 Fee for issuing a banker's draft 20.41 10.32 4.34 2.92 30.44 58.22 0.23 31.61 .. 16.29 Monthly fee for account maintenance 7.42 0.91 0.51 1.03 12.04 19.42 0.50 7.37 0.00 7.52 Fees and costs for standard domestic transactions (US$) Cost per checkbook issued 3.48 2.15 3.88 2.19 4.11 6.91 4.38 2.30 0.17 5.45 Annual fee for cash card/debit card 5.05 2.35 3.29 3.56 6.47 4.20 4.44 4.31 2.28 0.00 Cash card/debit card fees per transaction at merchants 0.13 0.03 0.02 0.08 0.18 0.24 0.00 0.23 0.00 0.38 Fee for issuing a banker's draft 7.76 2.99 7.52 4.79 9.56 15.59 0.64 8.61 .. 12.48 Monthly fee for account maintenance 3.01 0.55 0.96 3.75 4.23 5.20 1.39 2.01 0.00 5.76 Fees and costs for remittances (% monthly income) Fee to send payment via bankers' draft, foreign currency 63.59 15.13 14.56 10.27 98.26 109.63 1.75 42.67 .. 27.04 Fee to send payment via wire transfer 64.26 30.77 18.23 18.06 93.00 109.63 4.97 47.94 .. 27.04 Fee to use money transfer operator to send money 65.05 28.57 12.06 4.37 84.37 111.98 8.45 26.34 .. 2.86 Fee to receive a check issued abroad 54.06 7.75 12.50 4.76 86.13 111.38 2.20 77.46 .. 18.34 Fee to receive payment via wire transfer 25.31 16.08 3.54 5.70 36.08 36.12 1.24 26.34 .. 2.92 Fee to receive remittance through money transfer operator 19.17 4.80 1.17 0.00 29.37 37.33 0.00 26.34 .. 0.04 Fees and costs for remittances (US$) Fee to send a $250 payment via banker's draft in foreign currency 23.14 6.12 17.26 0.90 31.82 29.36 4.88 11.63 .. 20.72 Fee to send $250 payment via wire transfer 26.75 12.60 20.55 23.78 32.23 29.36 13.87 13.06 .. 20.72 Fee to use money transfer operator to send $250 18.39 7.86 13.25 4.08 24.36 29.98 23.58 7.18 .. 2.19 Fee to receive $250 check issued abroad 17.94 3.63 9.29 4.06 25.60 29.82 6.13 21.11 .. 14.06 Fee to receive $250 payment via wire transfer 9.54 5.56 4.89 5.95 12.40 9.67 3.47 7.18 .. 2.24 Fee to receive $250 remittance through money transfer operator 5.44 0.91 1.76 0.04 7.96 9.99 0.00 7.18 .. 0.03 INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 101 Latin All countries South Asia East Asia America Africa Tanzania Thailand Uganda Vietnam Zambia Domestic and cross-border payments, time indexes Domestic payment time index (0–1)* 0.49 0.56 0.39 0.45 0.51 0.56 0.24 0.84 0.39 0.52 Cross-border payment time index (0–1)* 0.44 0.38 0.40 0.43 0.46 0.39 0.45 0.24 0.31 0.58 Retail payments, availability and quality Range of payment services with standard bank account index (0–1)1 0.64 0.68 0.75 0.60 0.61 0.77 0.68 0.87 0.92 0.76 Mobile banking technology index (0–1)1 0.26 0.36 0.33 0.22 0.23 0.17 0.24 0.07 0.28 0.28 Network quality and interoperability index (0–1)1 0.62 0.69 0.76 0.83 0.54 0.67 0.88 0.67 0.80 0.68 102 BANKING THE POOR Retail payment channels index (0–1)1 0.47 0.43 0.65 0.50 0.44 0.57 0.79 0.44 0.68 0.47 Credit Business loan processing time (days) 7.42 8.02 7.64 13.41 6.53 13.33 21.00 3.00 6.80 4.00 Startup loan processing fee (%) 0.89 0.38 0.54 0.92 1.07 1.75 2.00 3.00 0.01 2.25 Collateral-to-loan-value ratio (%) 88.32 96.12 80.14 73.82 90.45 135.00 50.00 60.00 111.00 80.00 Maximum terms for startup loan (years) 4.55 4.43 3.31 6.25 4.60 7.00 5.00 4.00 1.00 5.00 Index of business loan application complexity1 0.66 0.68 0.73 0.77 0.75 0.83 0.75 0.00 0.50 1.00 Index of collateral flexibility* 0.48 0.66 0.80 0.52 0.71 0.78 1.00 0.00 0.72 0.78 Basic banking No opening fee (average yes/no) 0.77 0.81 0.80 0.94 0.74 1.00 1.00 0.33 1.00 0.60 No monthly fee (average yes/no) 0.61 0.83 0.73 0.76 0.53 0.50 1.00 0.00 0.60 0.20 No minimum balance (average yes/no) 0.52 0.42 0.51 0.54 0.53 0.33 0.80 0.33 0.20 0.40 Basic package of free transactions (average yes/no) 0.56 0.60 0.58 0.66 0.54 0.67 0.60 0.33 0.00 0.40 Basic banking (0–1)* 0.61 0.67 0.64 0.73 0.59 0.63 0.85 0.25 0.45 0.40 Government policy (commercial bank response) (average yes/no) 0.57 0.70 0.62 0.36 0.56 0.83 0.60 0.67 0.75 1.00 Government, offering basic banking1 0.23 0.33 0.25 0.20 0.22 0.00 0.00 0.00 1.00 0.00 Government, exempt basic accounts from ID requirements1 0.06 0.17 0.13 0.00 0.03 0.00 0.00 1.00 0.00 0.00 Savings schemes Doorstep collection1 0.08 0.17 0.25 0.00 0.03 0.00 0.00 0.00 0.00 0.00 Government-matched savings1 0.02 0.00 0.13 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Tax incentives for savings1 0.42 0.33 0.75 0.40 0.38 0.00 1.00 0.00 1.00 0.00 Periodic deposit1 0.40 0.53 0.48 0.46 0.36 0.33 1.00 0.67 0.60 0.00 Transparency and consumer protection Advising applicants index (0–1)1 0.46 0.58 0.56 0.40 0.42 0.50 0.50 0.00 1.00 0.50 Additional document requirements index (0–1)1 0.10 0.17 0.13 0.13 0.08 0.25 0.00 0.00 0.00 0.13 Guidelines on credit documents index (0 –1)1 0.71 0.67 0.72 0.90 0.68 0.00 1.00 1.00 0.75 0.75 Transparency and consumer protection index (0–1)1 0.40 0.60 0.50 0.40 0.33 0.08 0.50 0.58 0.83 0.58 Firms using bank loans (% all firms) Firms with bank loans 31.06 50.80 .. 36.30 27.78 16.25 .. 17.22 .. .. Firms that use loans for investment 16.25 12.71 23.95 10.83 15.70 6.79 74.73 7.67 29.20 17.39 Firms that use loans for expenses 28.20 34.47 33.27 28.16 25.58 17.33 72.56 14.02 38.00 38.65 Firms that view access to finance as constraint 35.62 27.76 15.11 22.06 46.17 40.63 15.22 47.81 13.21 53.66 Small and micro firms using bank loans (% firms) Small firms with bank loans 27.23 37.00 28.00 27.00 27.00 14.53 61.38 10.58 41.71 32.93 Micro firms with bank loans 20.33 34.00 21.00 22.00 19.00 11.31 51.32 9.55 32.64 31.03 Small firms that use loans for working capital 10.76 13.63 13.18 12.41 9.82 5.08 36.11 3.61 14.61 11.88 Micro firms that use loans for working capital 8.97 10.48 9.55 11.19 8.35 4.14 28.05 3.24 10.87 9.34 Small firms that use loans for investment 13.65 10.41 16.87 16.80 12.51 3.58 50.64 9.93 15.95 10.71 Micro firms that use loans for investment 11.20 6.25 13.77 15.27 10.06 1.54 48.95 8.62 8.44 10.00 Small firms that view access to finance as constraint 75.25 34.00 53.00 74.00 81.00 84.01 30.44 91.30 44.73 54.15 Micro firms that view access to finance as constraint 59.67 18.00 34.00 56.00 67.00 72.32 10.85 76.20 29.45 38.54 Latin All countries South Asia East Asia America Africa Tanzania Thailand Uganda Vietnam Zambia Demographic and socioeconomic variables Per capita GDP (US$) 2041.69 662.42 5486.02 3136.88 1334.40 323.73 3185.81 312.04 723.83 919.74 Population (millions) 53.25 292.20 63.38 27.20 20.51 39.00 65.00 30.00 84.00 12.00 Percentage of the population aged 15 years and older 61.00 65.00 70.00 63.00 57.00 55.63 78.59 50.75 71.12 54.36 Population density (/sq km) 245.56 370.96 920.71 124.00 87.11 44.66 126.69 151.57 271.26 15.96 Urban population (%) 39.14 24.08 47.71 58.07 37.05 24.64 32.64 12.74 26.88 35.14 Literacy (15+) 64.17 53.40 85.79 79.46 58.89 69.43 92.65 66.81 90.28 68.00 Ratio of private credit to GDP 28.85 37.00 55.25 35.00 20.53 12.00 88.00 8.00 71.00 10.00 Banking assets covered (%) 68.35 68.67 64.88 49.80 72.06 66.00 62.00 37.00 77.00 82.00 Legal rights index 4.12 4.00 4.75 3.60 4.06 5.00 5.00 3.00 6.00 6.00 Credit information index 2.22 2.50 3.00 5.60 1.44 0.00 5.00 0.00 3.00 0.00 1 Construction of index or variable explained in glossary of terms. * Figures for accounts per thousand adults are imputed. See chapter 9. Source: Getting Finance database. INDEXES OF FINANCIAL ACCESS: MEAN VALUES BY REGION AND COUNTRY 103 Acknowledgments Banking the Poor was prepared by a team led by Anjali The report has benefited from comments received from Kumar under the direction of Simeon Djankov. The team reviewers at the World Bank and IFC, and from external comprised Hedia Arbi, Farah Dib, Maximilien Heimann, reviewers: Thorsten Beck, Asli Demirguc-Kunt, Michael Ying Lin, Joana Pascual, Vikram Pathania, Mehnaz Fuchs, Maria Soledad Martinez Peria, Ignacio Mas, Safavian, and Valentina Saltane. Nachikhet Mor (ICICI Foundation), David Porteous (Bankable Frontiers Associates), Roberto Rocha, Liliana The team thanks the staff of the World Bank and Rojas-Suarez (Center for Global Development), Consolate International Finance Corporation who helped us to Rusagara, Peer Stein, and Kiatchai Sophastienphong. conduct interviews and to collect data by making contact with local financial sector experts, commercial banks, and This report was supported by the U.K. Department for central bank officials. Special thanks are due to Juan International Development (DFID). We extend our Costain, Olivier Fremond, Djibrilla Issa, Guillemette thanks to Catherine Martin, Douglas Pearce, Karen Ellis, Jaffrin, Astrid Manroth, Samuel Munzele Maimbo, and and Nicholas Godfrey for their various contributions to Ravi Ruparel, all in the Africa region; Noritaka the project. Akamatsu, Margarete Biallas, Hanh Nam Nguyen, and Thomas Rose in East Asia; Niraj Varma, Varsha Marathe, and Sabin Raj Shreshta in South Asia; and Khalid Siraj and Ismail Dalla, who also helped with the identification of local experts in different countries. Consultants in several of the countries covered in this report made invaluable contributions to this effort, as did commercial bank representatives and central bank officials, who contributed their time, perspectives and assistance. 104 BANKING THE POOR ISBN 978-0-8213-7754-3