D I R E C T I O N S I N D E V E L O P M E N T D I R E C T I O N S I N D E V E L O P M E N T 34359 Migrant Labor Remittances in South Asia SAMUEL MUNZELE MAIMBO RICHARD H. ADAMS, JR. REENA AGGARWAL NIKOS PASSAS Migrant Labor Remittances in South Asia Migrant Labor Remittances in South Asia Samuel Munzele Maimbo Richard H. Adams, Jr. Reena Aggarwal Nikos Passas THE WORLD BANK Washington, D.C. © 2005 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 08 07 06 05 The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the Executive Directors of the International Bank for Reconstruction and Development / 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 judgment 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-10: 0-8213-6183-X ISBN-13: 978-0-8213-6183-2 eISBN-10: 0-8213-6184-8 DOI: 10.1596/978-0-8213-6183-2 Library of Congress Cataloging-in-Publication Data Migrant labor remittances in South Asia / Samuel Munzele Maimbo ... [et al.]. p. cm ­ (Directions in development) Includes bibliographical references and index. ISBN 0-8213-6183-X 1. Emigrant remittances--South Asia. 2. Alien labor, South Asian--Arab countries. 3. Banks and banking--South Asia. 4. Infrastructure (Economics)--South Asia-- Finance. 5. South Asia--Economic conditions. I. Maimbo, Samuel Munzele. II. World Bank. III. Directions in development (Washington, D.C.) HG 5720.3.A3M54 2005 332.1'78--dc22 2005043744 Table of Contents Acknowledgments vii Abbreviations and Acronyms viii Executive Summary ix 1. International Remittances: Impact, Policy, and Costs 1 The Impact of International Remittances on Poverty and Economic Development 2 The Efficiency and Effectiveness of the Public Infrastructure for Remittances 5 The High Transaction Costs Associated with Remittances through Formal Financial Institutions 5 The Level of Transparency and Accountability in the Informal Remittance Industry 6 Notes 6 2. The Development Impact of Remittances 7 Analyzing the Impact of International Remittances 7 A New Dataset on International Remittances and Poverty 8 Official International Remittances and Poverty: Econometric Results 10 Estimating Unofficial Remittances: Simulation Functions 12 Official and Unofficial Remittances and Poverty 14 Remittances, Investment, and Economic Development 16 Conclusion 18 Notes 18 3. Public Infrastructure for Remittances 21 Bangladesh 21 India 23 Pakistan 25 v vi TABLE OF CONTENTS Sri Lanka 28 Conclusion 30 Notes 30 4. Formal Financial Sector Infrastructure for Remittances 31 State Commercial Banks 31 Private Foreign and Local Commercial Banks 33 Money Service Businesses 36 Post Office Networks 37 Conclusion 39 Notes 41 5. Informal Financial Sector Infrastructure for Remittances 42 Types of Informal Remittance Systems 42 Historical Background of Informal Remittance Systems 46 Migrant Incentives for Using Informal Remittance Systems 47 Risks and Potential Abuse of Informal Remittance Systems 51 Conclusion 54 Notes 54 6. Conclusions and Recommendations 56 Enhancing the Development Impact of Remittances 56 Public Infrastructure for Remittances 58 Formal Financial Infrastructure for Remittances 59 Informal Financial Infrastructure for Remittances 61 Conclusion 62 Notes 62 Appendix 1: Summary of Dataset on Poverty, Inequality, International Migration, and Remittances 63 Bibliography 72 Index 77 Acknowledgments This study was prepared by a World Bank team led by Samuel Munzele Maimbo under the overall guidance of the South Asia Regional Initiatives Progam Team, notably Alastair McKechnie and Ejaz Syed Ghani, and the South Asia Finance and Private Sector management of Joseph Del Mar Pernia and Simon C. Bell. The peer reviewers, from whom the contributors benefited greatly, were Leora Klapper, Dilip K. Ratha, and John Wilson. Kazi Iqbal and Karina Karaan provided research assistance, and excellent administrative support was provided by Maria Marjorie Espiritu. The study draws on numerous interviews with public and private sector officials from selected South Asian countries. We are thankful to them all for their candid opinions, comments, and suggestions. As authors of an exploratory study of this nature, whose primary objective is to highlight the key policy discussion issues about remittances rather than an exhaustive paper on the subject as it affects every country in the region, we take full responsibility for failing to account for all the fascinating remittance innovations that are rapidly emerging in the region. vii Abbreviations and Acronyms ACH automated clearing house AML anti-money laundering ATM automated teller machines BMET Bureau of Manpower, Employment, and Training BOESL Bangladesh Overseas Employment and Services, Ltd CTF counter-terrorist financing CIRC Corporate and Industrial Rehabilitation Corporation FDI foreign direct investment GDP gross domestic product IMF International Monetary Fund IT informational technology LSMS Living Standards Measurement Survey MSBs money service businesses NFTS Nationwide Funds Transfer System NRFC nonresident foreign currency NRH nonreceiving household NRIs nonresident Indians NRPs nonresident Pakistanis ODA overseas development assistance OPF Overseas Pakistanis Foundation POSB Post Office Savings Bank RBI Reserve Bank of India RRH remittance-receiving household SAR South Asia Region SBI State Bank of India SBP State Bank of Pakistan SLBFE Sri Lanka Bureau of Foreign Employment UAE United Arab Emirates VSAT very small aperture terminals viii Executive Summary Bangladesh, India, Pakistan, and Sri Lanka have all experienced a sharp increase in remittances during the past decade. At the end of 2003, Bangladesh, India, Pakistan, and Sri Lanka were all among the top 20 receivers of remittances, with estimated receipts of US$3.2 billion, US$17.4 billion, US$4.0 billion, and US$1.3 billion, respectively (table 1; a billion is 1,000 million). This paper provides a strategic overview of key issues relating to the remittance industry in the South Asia region. ThepaperbuildsonrecentWorldBankresearchonremittancesthatpromi- nently features the SouthAsia region.1 Rather than duplicate that work, this study focuses only on the region's distinguishing characteristics, namely: · Alarge migrant population of semiskilled and unskilled workers largely concentrated in the Persian Gulf countries, particularly Saudi Arabia and the United Arab Emirates, contributing to rising remittance flows · The presence of dedicated public institutions and government financial incentives aimed at facilitating and providing support for temporary migration and remittance inflows Table 1 Workers' Remittances (US$ billions) 1999 2000 2001 2002 2003 Total 14.0 16.0 15.9 21.4 25.9 Bangladesh 1.8 2.0 2.1 2.8 3.2 India 11.1 11.7 11.1 13.7 17.4 Pakistan 0.1 1.1 1.5 3.6 4.0 Sri Lanka 1.0 1.2 1.2 1.3 1.3 Total, as a % of GDP 2.6 2.7 2.6 3.4 3.5 Bangladesh 3.9 4.2 4.5 6.0 6.1 India 2.5 2.6 2.3 2.7 2.9 Pakistan 1.7 1.8 2.5 6.0 5.8 Sri Lanka 6.8 7.1 7.5 7.9 7.1 Sources: Authors' calculations based on IMF, Balance of Payments Statistics; World Bank, World Development Indicators. ix x EXECUTIVE SUMMARY · The existence of large state bank branch networks with immense poten- tial for a more effective and efficient remittance financial market · The widespread use of trade-related informal remittance channels by both legal and illegal migrants The Development Impact of Remittances That this increase in the quantity of remittances is taking place at a time of declining overseas development assistance (ODA) flows to the region is of additional importance. Interest in the development impact of remittances has increased because they continue to be the largest financial flow to the region after foreign direct investment. The increase in remittance volumes has renewed academic and public policy interest in their potential to reduce poverty and economic vulnera- bility, improve family welfare, and stimulate local economic development in the face of much lower, sometimes temperamental, foreign direct invest- ment flows (figure 1). Despite this increasing interest, the specific development impact of remit- tances is still unclear and requires further research and debate. The following questions are particularly important: · What proportion of remittance monies is spent on consumption versus investment? Does the proportion vary with the income and educational level of the remittance-receiving family? Figure 1 Worker Remittances and Other Inflows India Pakistan 20 4.5 18 4.0 Remittances Remittances 16 3.5 14 3.0 12 2.5 billions 10 billions 2.0 8 US$, US$, 1.5 6 FDI 4 1.0 FDI 2 0.5 0 0 1988 1991 1994 1997 2000 2003 1988 1991 1994 1997 2000 2003 Source: Authors' calculations based on World Bank data. EXECUTIVE SUMMARY xi · What kinds of productive investment activities are remittances spent on? How does this vary by setting (urban or rural), and by district, region, or country? Addressing the questions listed above is methodologically challenging because of three factors. The most important of these is that remittance income is fungible; that is, remittances can be spent on anything. This makes it very difficult to associate remittances with any specific changes in house- hold patterns of consumption or investment. Second, remittances have mul- tiple effects on the local economy. For example, an increase in the volume of remittances in a community may lead to higher spending on housing, which in turn helps to generate more income and employment opportuni- ties for unskilled construction workers. Third, a robust theoretical and ana- lytical framework for determining the development impact of remittances is largely absent. Because remittance research is often based on data col- lected at just one point in time, it is difficult to measure how remittances change patterns of investment over time.2 This study analyzes the impact of international remittances on poverty using a growth-poverty model. This model, which has been used by a host of poverty researchers,3 assumes that economic growth--as measured by increases in mean per capita income--will reduce poverty. The results are interesting. The analysis finds that, when the estimated values for unofficial remittances are added to official remittance figures, total remittances (official and unofficial) reduce the level of poverty in South Asia. On average, the point estimates for the poverty headcount measure sug- gest that a 10 percent increase in total remittances (official and unofficial) will lead to a 0.9 percent decline in the level of poverty in South Asia. This means that for a "representative" country where exactly one-half of the popula- tion lives below the poverty line, a 10 percent increase in total remittances (official and unofficial) will bring the proportion living in poverty down to about 0.48 percent. The Public Infrastructure for Remittances All the countries in the region have taken active steps to support the migra- tion of their nationals. The public infrastructure to support the search for employment abroad, the migration of successful applicants, and their sub- sequent stay is, in principle, in place. In all four countries, the public infra- structure includes emigration legislation, government ministries and departments, and a plethora of incentives for nonresident nationals. · Legislation: Legislation such as the Emigration Act of 1983 (India) and the Emigration Ordinance of 1979 (Pakistan) form the legal basis for today's migration practices and influence the ease and convenience of xii EXECUTIVE SUMMARY dealing with the resulting government ministries and departments. These provide the basis for the support governments provide to premigrants, migrants, and returnees. · Government departments: The Bureau of Manpower Employment and Training (Bangladesh), the Bureau of Emigration and Overseas Employment (Pakistan), and the Bureau of Foreign Employment (Sri Lanka) provide a range of pre- and postmigration services and facil- ities. For first-time migrants in particular, premigration training, visa application processing, and other services are highly prized. However, these bureaus have a mixed record in terms of their effectiveness and efficiency. The lack of resources limits their outreach and the types of services that they offer. As government institutions, perceptions of bureau- cracies persist even where corrective actions have been undertaken. · Incentives: The most notable government actions directed toward migrants have been the plethora of incentives that have been announced in successive budgets that specifically target migrants. These incen- tives include special access to the merit-based quota system assigned in all public professional colleges and universities; generous duty-free import limits for items of personal convenience; preferential allocation of investment opportunities such as initial public offerings to be sub- scribed in foreign currency; eligibility for special lotteries such as those for prime plots in public housing schemes at attractive prices; and tax exemptions. Consistently improving the effectiveness of the public infrastructure will substantially aid the development impact of remittances. Formal Financial Institutions There are several types of formal financial institutions that dominate the official remittances market in the region. State Banks The South Asian remittances market is unique for the presence of an exten- sive branch network of state commercial banks. These banks have long dominated the official remittances business through monopolistic national foreign-currency legislation and large bank branch networks. Today, for example, India has over 32,000 rural commercial bank branches; Bangladesh has four nationalized banks with at least 3,346 branches in total; and Sri Lanka's largest state bank alone has 326 branches, 81 counter services, and 188 pawning centers. Physically, the infrastructure for an active far- reaching remittance network is already present. The challenge lies in making it more effective and efficient for that purpose. EXECUTIVE SUMMARY xiii Foreign Banks Foreign banks have also noted the huge remittance potential in the region. Previously reluctant participants, they are slowly investing in this business, albeit largely for the higher-income migrants--doctors, accountants, lawyers, and other professionals. Foreign banks have hitherto largely been inhibited by their limited branch networks, which are primarily centered in major cities. They have been careful about heightened anti-money laundering and counter-terrorist financing standards, and concerned about the changing remittance business model. Profits from remittance transactions have been declining. The removal of foreign exchange controls, the competition from money transfer compa- nies, and the pressure to make even speedier deliveries have reduced exchange gains, pushed down upfront commissions, and minimized the potential interest gains from cash floats. Local Banks It is largely the local banks that are taking the lead in the remittance service market. Investing heavily in various remittance application technologies-- credit cards, debit cards, Internet banking, and telebanking--the local banks are surely changing the nature of the remittance market. The increase in the number of automated teller machines (ATMs) has been impressive. In Sri Lanka for example, the total number of ATMs oper- ated by commercial banks increased to 705 at end of 2003 from 622 at end of 2002. In Pakistan one ATM network has over 392 ATMs in 26 cities; the other has 130. Yet the true potential for the ATMs to act as a remittance conduit or plat- form is yet to be realized. The majority of ATM transactions are only cus- tomers withdrawing cash from their own accounts. Only a limited number of transactions are customers using ATMs other than those installed by their own banks. Until this happens more frequently, the possibility of using the networks for remittance purposes will remain limited. Money Service Businesses The strongest remittance competition in the market comes from the emer- gence of licensed money service businesses (MSBs)--nonbank financial institutions that accept cash, checks, or other money instruments, or "stored value," in one location and pay the equivalent amount in another. Leading MSBs such as Western Union and Money Gram have extensive agent locations in the region that support their operations. However, Internet- based businesses such as e-exchange and Remit2India are having a posi- tive impact on the way business is managed. xiv EXECUTIVE SUMMARY Post Offices In the wake of declining traditional postal business, post offices in the region are exploring the possibility of participating in the remittance busi- ness more actively. Long used to delivering money orders to rural families, post offices are now investing in electronic money orders, partnering with money service businesses, and investing in Internet-based technology. The potential benefits for postal involvement are substantial. With 154,149 post offices and 554 sorting offices, India Post has the most extensive postal retail network in the world. Sri Lanka Post has 625 main post offices, 3,423 sub­post offices, and 632 agency post offices. However, as long as the postal systems are plagued with operational losses, the absence of management information systems, limited access to funds for investing in technology, and negative publicity--such as stories of insufficient cash on hand and delays in effecting payment--client demand and usage will remain low. Reliability, credibility, and efficiency are essen- tial ingredients to a remittance business. Informal Financial Institutions Informal remittance systems--courier services, in-kind remittances, and hawala systems--have a long history in the region. Originally developed at a time when conventional banking instruments were either absent or weak, informal channels offered a speedy, low-cost, convenient, accessible, and (when necessary) anonymous option. · Courier services: These are the simplest and oldest way of moving value in the region. Physical transfer of cash by migrants either on their own behalf or on behalf of others is said to be high. This is not unique to the region, as such transfers are still observed in other regions. · In-kind remittances: Some migrants choose to invest in durable goods such as refrigerators, stoves, televisions, VCRs, and other valuable commodities such as gold or precious stones, which they send to their families for sale or personally carry home at the end of their contracts. · Hawala: Also referred to as hundi or chit, hawala systems owe their pop- ularity to the long history of trade in the region. Hundi, a form of bill of exchange or promissory note, was used in the region before the advent of modern banking. The interface between traders and hawala dealers also owes much to the region's traditionally complex frame- work of regulating import and export transactions. Until very recently burdensome legislation regarding international payments, rebates, refunds, and other financial payments encouraged the use of informal alternatives. EXECUTIVE SUMMARY xv Overall, migrants use informal systems for a variety of reasons, partic- ularly their reliability, lower cost, speed, convenience and accessibility, and for some the anonymity that informality can provide users and recipients alike. Stongly based in trust and rooted in age-old cultural traditions, infor- mal systems deliver cash much faster than conventional banking systems at a much more cost-effective rate. Importantly, the convenience of door- to-door collection and delivery services is invaluable to migrants working long shifts and recipient families in communities with absent or weak formal financial services. Unfortunately, the informal system's success--speedy transactions with minimal or no documentation--has also been its undoing. The anonymity associated with such transactions has long raised concern with law enforce- ment communities. This has led some countries to issue an outright ban on these systems; others have proceeded to or are considering issuing new regulations in an effort to improve the transparency of the sector. Conclusion and Recommendations Remittances from migrant workers are increasingly important and are a stable source of external finance for countries in the South Asia region. This preliminary analysis of migrant remittances to the region has been based on a review of widely dispersed data, documentation, and interviews. Its purpose is to stimulate and inform discussion on the role remittances play in the region and to help stakeholders design, develop, and implement appropriate policy. By exploring the actual and potential impact of remit- tances in the region, and the capacity of the public and private infrastruc- ture to facilitate that impact, this study identifies opportunities for greater contributions. With an extensive state and private bank network already in the region, a key component of the requisite building blocks for an effective remittance industry is already in place. However, much more needs to be done to maxi- mize the network's full potential, including: · Shared payments systems platforms: Greater investment in open archi- tecture payments system information technology is required. Recipients should increasingly be able to receive remittances from any state or private bank branch or ATM machine, whether or not they hold an account with that institution. The potential for nonbank financial insti- tutions, such as post office networks and money transfer businesses, to connect to banks' branches and technological advantages promises great potential for expedient, cost-effective remittance networks and services. xvi EXECUTIVE SUMMARY · Public-private partnerships: Strategic partnerships between the public and private sectors across infrastructure, products, and services are neces- sary for countries in the region that seek the types of gains attained elsewhere. · Cross-selling: It is essential that, as the profitability of the conventional formal financial sector remittance business model continues to decline, formal financial institutions invest in cross-selling complementary finan- cial services and products. Encouraging recipients to open and maintain bank accounts that might lead to short- and long-term auto and hous- ing loans, for example, should be part of a business strategy for banks. In the long run, strengthening the formal financial sector for remittance purposes will facilitate the move away from informal remittance systems, which, though beneficial to the direct users, may not have as effective an impact on the macro level as formal remittance transfers. High transaction costs, long delays in transferring remittances, foreign cur- rency controls, and overly bureaucratic policies and procedures for simple money transfers have no place in a vibrant and still-growing remittance industry. Notes 1. See El-Qorchi, Maimbo, and Wilson (2003); Ratha (2003, 2004); Maimbo (2003). 2. For example, using data collected in a series of 14 interviews with 469 households in rural Pakistan over a five-year period (1986­7 to 1990­1), Adams (1998) found that the availability of remittance income helps to increase invest- ment in rural assets by raising migrant households' marginal propensity to invest. Contrary to the common notion that remittances are primarily used for consump- tion, the results suggested important statistical effects on the accumulation of rural assets. This is because households receiving remittances tend to treat them as transitory income, and the marginal propensity to invest transitory income is higher than it is for total labor income (excluding remittances and rental income). 3. See, for example Ravallion (1997) and Ravallion and Chen (1997). 1 International Remittances: Impact, Policy, and Costs As volumes of international remittances have swelled, policy makers have become increasingly concerned with their impact on poverty and economic development. Over the past decade, the literature on the macro- and micro- economic implications of migrant remittances has grown with the finan- cial flows. Only recently, however, has that research taken decidedly regional and global perspectives. In a much-quoted study of remittances, Ratha (2003) underscored the importance of remittance flows for developing countries and delineated their regional patterns. In a follow-up study, Ratha (2004) observed that workers' remittances continued to rise in 2003 to an estimated US$93 bil- lion, up from US$88.1 billion in 2002. Latin America and the Caribbean region continued to lead in the volume of remittance receipts--receiving US$30 billion, nearly a third of remittance flows to all developing coun- tries. South Asia, East Asia, and the Pacific each received some US$18 bil- lion, whereas Sub-Saharan Africa received just US$4 billion. Bangladesh, India, Pakistan, and Sri Lanka all are among the top 20 receivers of remittances, with estimated receipts of US$3.2 billion, US$17.4 billion, US$4.0 billion, and US$1.3 billion, respectively (table 1.1). These four coun- tries made the South Asia region the second largest regional recipient of remittances in the world after Latin America and the Caribbean. The figures above ensured that remittances remained the second-largest financial flow to developing countries in 2003 after foreign direct invest- ment, more than double the size of net official development assistance. Moreover, remittance volumes increased, despite expectations to the con- trary based on weak labor markets and the tightening of border controls in the industrial countries after the terrorist attacks of September 11, 2001. Ratha (2004) attributes the continued increase in remittances to three factors. First, more remittances are being diverted from alternative channels to formal channels as a result of efforts to curb money laundering. Second, the increased focus on remittances has brought better reporting of data in many developing countries. And third, the fear of being deported or inves- tigated may have prompted some migrant workers to remit their entire savings to their home countries. 1 2 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Table 1.1 Workers' Remittances Received by Selected Countries in South Asia, 1995­2003 (US$ billions) 1995 1996 1997 1998 1999 2000 2001 2002 2003 Bangladesh 1.2 1.3 1.5 1.6 1.8 2.0 2.1 2.8 3.2 India 6.2 8.8 10.3 9.5 11.1 11.7 11.1 13.7 17.4 Pakistan 1.7 1.3 1.7 1.2 0.1 1.1 1.5 3.6 4.0 Sri Lanka 0.8 0.8 0.9 1.0 1.0 1.2 1.2 1.3 1.3 Total 10.0 12.3 14.6 13.3 14.0 16.0 15.9 21.4 25.9 Source: World Bank 2004b: 197. The economic policy implications of these trends are significant. Policy makers are rightly debating the following key issues: · The impact on poverty and economic development of migrant remit- tances, especially the proportion of remittances directed toward pro- ductive investment and savings in developing countries · The efficiency and effectiveness of the public infrastructure for remittances · The high transaction costs associated with remittances through formal financial institutions · The level of transparency and accountability in the informal remittance industry, particularly in light of the growing security concerns relating to money laundering and terrorist financing This paper addresses each of the issues listed above with specific ref- erences to the South Asian countries of Bangladesh, India, Pakistan, and Sri Lanka. This study is not an exhaustive examination of these countries to emphasize, illustrate, and argue their policy relevance to the region. At best, the study is a platform for further, more detailed investigative research. The Impact of International Remittances on Poverty and Economic Development Chapter 2 is based on the argument that, in part, the impact of remittances on economic development has come to the fore in the development debate with the steady decline in the volume of overseas development assistance (ODA) toSouthAsia,andtothedevelopingworldasawhole(figures1.1and1.2).The downward trend in ODA has increased interest in the effect of remittances oneconomicdevelopment.Butmoreimportant,thepaperisalsopremisedon INTERNATIONAL REMITTANCES: IMPACT, POLICY, AND COSTS 3 Figure 1.1 Official Development Assistance to Bangladesh, India, Pakistan, and Sri Lanka (aggregated, 1990­2001) 8,000 7,000 6,000 5,000 millions 4,000 US$, 3,000 2,000 1,000 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Source: World Bank, Global Development Finance 2003. Figure 1.2 Official Development Assistance and Official Aid to Bangladesh, India, Pakistan, and Sri Lanka (by country, 1990­2001) 3,000 Bangladesh India Pakistan Sri Lanka 2,500 2,000 millions 1,500 US$,1,000 500 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Period Source: World Bank, Global Development Finance 2003. 4 MIGRANT LABOR REMITTANCES IN SOUTH ASIA suggestions in the recent literature that remittances raise the incomes of migrant households and increase the recipient country's foreign exchange reserves.At the household level, the literature suggests that remittance income helps meet a variety of family needs, such as increased consumption--of food, housing, and durable items--and increased investment--in business, microenterprise, education, and financial enterprises.1 Through these means, remittances support a wide variety of devel- opment purposes: improving family welfare, reducing economic vul- nerability, and boosting the local economy, while increasing the use of formal banking services by the poor. If remittances are invested in productive activities such as local business and the education of young children, they can contribute to output growth and generate positive mul- tiplier effects for the economy as a whole. Yet the developmental impact of remittances is still unclear, and requires more work on such policy questions as: · What proportion of remittance monies is spent on consumption versus investment? Does the proportion vary with the income and educational level of the remittance-receiving family? · What kinds of productive investment activities are remittances spent on? How does this vary by setting (urban or rural) and by district, region, or country? · What is the impact of the channels through which remittances arrive (formal or informal) on spending patterns? A reasonable hypothesis is that hawala funds may be used for less productive investments, facili- tate tax evasion, and negatively affect governance and exchange reserves. Using countries in the region, particularly Pakistan, chapter 2 highlights the potential development impact of remittances. The discussion is brief because methodological hurdles make it difficult to test the impact of remit- tances on migrant households.2 The most important of the methodological problems is that remittance income is fungible; that is, remittances can be spent on anything. This makes it very difficult to associate remittances with any specific changes in household patterns of consumption or investment. Second, remittances have multiple effects on the local economy. For exam- ple, an increase in the volume of remittances in a community may lead to higher spending on housing, which in turn would help generate more income and employment opportunities for unskilled construction work- ers. Third, a robust theoretical and analytical framework for determining the development impact of remittances is largely absent. Because data used in remittance research are often based on information collected at just one point in time, it is difficult to measure how patterns of investment change over time.3 INTERNATIONAL REMITTANCES: IMPACT, POLICY, AND COSTS 5 The Efficiency and Effectiveness of the Public Infrastructure for Remittances Chapter 3 takes stock of the public infrastructure for remittances in the four countries and comments on the opportunities for improving the effective- ness and efficiency of the existing institutions. Governments in the region are increasingly taking active note of the development impact of remittances and are taking various forms of initiatives--with varying degrees of success-- to facilitate migration of their nationals, the welfare of their citizens abroad, and the remittance of earnings to their home countries. Chapter 3 highlights the similarities among the four countries with respect to their basic public infrastructure for remittances. It commends the efforts made to facilitate easier migration for the average migrant, particularly low-income migrants. However, it concludes by contending that more could be done to improve the effectiveness and efficiency of the existing public infrastructure. The High Transaction Costs Associated with Remittances through Formal Financial Institutions Chapter 4 argues that while government should be commended for allo- cating public infrastructure for the migration and remittance industry, in the long term, it is the private formal financial institutions that should take the lead in improving the quality of the remittance infrastructure available to migrants. At present, inefficient financial systems and high-cost trans- actions in some parts of the region remain a significant impediment to the more effective use of remittances in many labor-exporting countries. Chapter 4 also discusses the transaction costs associated with the formal financial sector with special regard to the unique presence in the region of dense formal state commercial bank branch networks. These networks have tremendous potential for facilitating remittances, especially to the rural poor. While extensive state bank branch networks provide a relatively dense infrastructure for remittance services, questions remain regarding the capac- ity of these networks to provide affordable payments systems attractive to migrant workers. Hitherto, weak access to formal finance has resulted in a heavy reliance among poorer rural households on informal finance--mostly moneylen- ders and shopkeepers--for credit and deposit services, including remit- tance services. Making the formal financial institutions more attractive for remittance agencies remains a fundamental challenge for governments. Despite recent innovations by private financial institutions, more needs to be done. This issue of high transaction costs and difficult-to-access formal institutions is not unique to the region. The World Bank estimates that the average cost of transferring remittances remains about 13 percent, and some- times exceeds 20 percent, of the amount remitted. In addition to these high 6 MIGRANT LABOR REMITTANCES IN SOUTH ASIA transaction costs, recipients often face exchange losses and slow check clear- ance because of inefficient banking systems. Reducing the transaction cost of remittances to less than 10 percent would imply an annual saving of US$3.5 billion to overseas workers.4 The Level of Transparency and Accountability in the Informal Remittance Industry In Bangladesh, India, Pakistan, and Sri Lanka, informal remittance systems-- used for both legal and illegal remittance purposes--are commonplace. They are particularly attractive for their speed, low cost, convenience, acces- sibility, and potential anonymity, especially in rural communities when conventional banking instruments are either absent or weak. While acknowl- edging the potential benefits to migrants for using such systems, govern- ments in the region are concerned about the risks of financial abuse and criminal activity. With a long and rich history of commercial trade, there is widespread mingling of trade and remittance businesses among different countries. Unfortunately, some of the businesses are illegal and have raised legitimate concerns among law enforcement agencies. Understanding how these systems work, as detailed in chapter 5, and the incentives for migrants to use them, is an essential prerequisite to an effective formal remittance development strategy. The study's conclusions and recommendations are presented in chap- ter 6. Overall, this study complements and extends the existing literature by reviewing remittances in a development context in four countries of South Asia. The objective of the study is to deepen understanding of the poverty impact of remittances, the developmental roles of remittances, the public and private infrastructure for remitting migrant earnings, and cur- rent concerns about transparency and accountability of remittance flows. Notes 1. See, for example, Adams (1991,1998). 2. See Adams (1998). 3. For example, using data collected in 14 interviews with 469 households in rural Pakistan over a five-year period (1986­7 to 1990­1), Adams (1998) found that the availability of remittance income helps to increase investment in rural assets by raising migrant households' marginal propensity to invest. Contrary to the common notion that remittances are primarily used for consumption, the results suggest important statistical effects on the accumulation of rural assets. This is because households receiving remittances tend to treat them as transitory income, and the marginal propensity to invest transitory income is higher than it is for total labor income (excluding remittances and rental income). 4. See Ratha (2003). 2 The Development Impact of Remittances International migration is one of the most important factors affecting eco- nomic relations between developed and developing countries in the twenty- first century. It is currently estimated that about 175 million people--roughly 3 percent of the world population--live and work outside the country of their birth.1 The international remittances sent home by these migrant work- ers have a profound impact on the living standards of people in the devel- oping world. Analyzing the Impact of International Remittances The purpose of this chapter is to analyze the impact of international remit- tances (official and unofficial) on poverty and economic development in South Asia. Despite the ever-increasing size of international remittances in South Asia, little attention has been paid to the effect of these transfers on poverty and development in the region. At least three factors are respon- sible for this lack. The first is an absence of poverty data: until recently, it has been difficult to estimate accurate poverty headcounts in the five coun- tries of South Asia. The second factor relates to international remittances: as of early 2005, the data on international remittances did not include the large (and unknown) sum of remittance monies that are transmitted through informal, unofficial channels. This makes it very difficult to pinpoint the total impact of remittances--official and unofficial--on poverty in South Asia. The final factor concerns measuring the effect of remittances on invest- ment and development. Although household budget surveys represent the best source of information on how migrant households spend or invest their remittance earnings, few of these surveys ask any questions about how international remittances are used by their recipients. This chapter seeks to overcome these and other problems by proceed- ing in the following manner. First, to set the South Asian experience in a larger international context, it builds a new dataset composed of 71 devel- oping countries.2 This dataset includes all those low- and middle-income developing countries for which reasonable information on poverty and international remittances could be assembled. Second, the chapter uses this 7 8 MIGRANT LABOR REMITTANCES IN SOUTH ASIA dataset to estimate the impact of official remittances on the level and depth of poverty in South Asia. Third, the chapter uses new techniques to simu- late an estimate of the size of unofficial remittance flows to South Asian countries. The analysis then takes these estimates of unofficial remittances, combines them with the official remittance figures, and re-estimates the impact of remittances (official and unofficial) on poverty in South Asia. Finally, the chapter uses data from a five-year panel household survey in rural Pakistan to analyze how remittances are spent or used. This section sheds light on the question of how households use international remittances for economic development at the local level. A New Dataset on International Remittances and Poverty To set the South Asia experience into a larger context, this section develops a new dataset that includes information on international remittances and poverty for as many developing countries and time periods as possible. Initially the goal was to include all 157 countries that are classified as either low-income or middle-income countries by the World Bank in the World Development Report, 2000/01.3 However, it proved impossible to find remit- tances and poverty data for many of these developing countries. Since these data problems have constrained past work on this topic, it is useful to spell out the nature of these difficulties.4 Because of their importance to labor-exporting countries, remittance flows tend to be the best-measured aspect of the migration experience. For instance, the International Monetary Fund (IMF) keeps annual records of the amount of international remittances received by each labor-exporting country.5 However, the IMF reports only data on official international remit- tance flows--that is, remittance monies that are transmitted through official banking channels. Since a large (and unknown) proportion of remittance monies is transmitted through private, unofficial channels, the level of remit- tances recorded by the IMF underestimates by an unknown percent the actual flow of remittance monies returning to labor-exporting countries. With respect to poverty, many developing countries--especially coun- tries with smaller populations--have not conducted the types of nationally representative household budget surveys that are needed to estimate poverty. For example, of the 157 countries classified as low or middle income by the World Bank, 76 countries (48 percent) have not published the results of any household budget survey. Within these data limitations, the dataset in this chapter includes infor- mation from 71 low-income and middle-income developing countries. Appendix 1 gives the countries, geographical regions, poverty, and remit- tances indicators included in this new dataset. The dataset includes 184 observations from the 71 developing countries; an observation is any point THE DEVELOPMENT IMPACT OF REMITTANCES 9 in time for which complete data on poverty and remittances exist. The dataset includes 20 observations (from five countries) in South Asia. Appendix 1 reports three different poverty measures. The first, the head- count index, set at $1 per person per day, measures the percentage of the population living beneath that poverty line at the time of the survey.6 However, the headcount index ignores the "depth of poverty," that is, the amounts by which the average income and expenditures of the poor fall short of the poverty line. The table therefore reports the poverty gap index, which measures in percentage terms how far the average income and expen- ditures of the poor fall short of the poverty line. The third poverty measure-- the squared poverty gap index--indicates the severity of poverty. The squared poverty gap index has useful analytical properties because it is sensitive to changes in distribution among the poor.7 Table 2.1 summarizes the poverty and remittances information in the dataset by comparing the observations for South Asia with those for the rest of the developing world. On average, poverty is higher in South Asia. For example, both the mean poverty headcount and poverty gap measures for South Asia are higher than those for other countries in the developing world, and the differences between South Asia and the rest of the world for these two poverty measures are statistically significant. However, income inequality (measured by the Gini coefficient) is lower, on average, in South Asia than in the rest of the developing world, and this difference is signif- icant. With respect to migration and remittances, international migration Table 2.1 Summary of Descriptive Statistics on Poverty, Inequality, International Migration, and Remittances in South Asia Compared with the Rest of the Developing World South Asia Rest of developing world T-statistic Statistic (N 20) (N 164) (two-tailed) Poverty headcount ($1/person/day) 34.67 16.05 4.37** Poverty gap (percent) 9.06 6.16 2.24* Squared poverty gap (percent) 2.53 3.25 1.08 Gini coefficient 0.324 0.423 8.65** International migration as share of country population 0.49 1.59 3.42** Official remittances as share of country GDP 2.59 1.41 2.76** Source: Authors' calculations. Note: N 184 observations. See Appendix 1 for countries and survey dates. *Significant at the 0.05 level. **Significant at the 0.01 level. 10 MIGRANT LABOR REMITTANCES IN SOUTH ASIA as a share of country population is lower in South Asia, but official inter- national remittances as a share of country GDP are higher than in other developing countries. The reason for this paradox is that while South Asian countries send large numbers of international migrants to the Persian Gulf countries, no data are available on the number of migrants working there.8 Thus, while the migration information in the dataset tends to underesti- mate the extent of international migration from South Asia, the remittances information reflects the impact of remittances more accurately on this region of the world. Mean official remittances as a share of country GDP are higher in South Asia (2.59 percent) than in any other region of the world save one: the Middle East and North Africa (9.60 percent). Official International Remittances and Poverty: Econometric Results In this section the cross-country data are used to analyze how international remittances that are sent back through official channels affect poverty in South Asia. Using the basic growth-poverty model suggested by Ravallion (1997) and Ravallion and Chen (1997), we want to estimate variations in poverty as a function of three variables: mean per capita income, income distribution, and level of official international remittances.9 This growth-poverty model, which has been used by a host of poverty researchers,10 assumes that economic growth--as measured by increases in mean per capita income--will reduce poverty. The relationship between poverty and the income variable is therefore expected to be negative and significant. The model also assumes that the level of income inequality affects poverty reduction. Since past work has shown that a given rate of economic growth reduces poverty more in low-inequality countries than in high-inequality countries,11 the income inequality variable is expected to be positive and significant. The innovation in this study is to introduce into the model a variable measuring the level of international remittances. All other things being equal, it is hypothesized that countries receiving more international remittances will have less poverty.12 This growth-poverty model is often measured in first differences in order to deal with possible correlation problems between the variables, since the dependent and independent variables are drawn from the same single source of data (household budget surveys). In this chapter, however, the model is estimated as a level equation since the dependent and inde- pendent variables come from different sources of data: the dependent vari- able is drawn from household budget surveys and the independent variables (for official international remittances) from various other sources. Regression results for the growth-poverty model are presented in table 2.2. As expected, the coefficients for the per capita income and Gini coefficient variables are negative and positive, respectively, and statistically significant Table 2.2 Elasticity of Poverty, Estimated Using Official International Remittances Dependent variable Poverty headcount (US$1.08/person/day) Dependent variable Poverty gap Variable (1) (2) (3) (4) (5) (6) (7) (8) Per capita survey 2.263 2.090 2.007 2.002 2.584 2.434 2.438 2.340 mean income ( 20.60)** ( 18.61)** ( 13.47)** ( 13.40)** ( 20.54)** ( 18.36)** ( 12.29)** ( 12.25)** Gini coefficient 3.832 4.217 4.134 4.131 4.792 5.126 5.539 5.534 of country (14.32)** (15.59)** (11.75)** (11.72)** (15.55)** (16.05)** (12.28)** (12.28)** Per capita official 0.188 0.181 0.207 0.196 remittances ( 4.73)** ( 4.46)** ( 4.07)** ( 3.78)** South Asia dummy 0.856 0.800 1.120 0.745 0.836 1.353 (1 if South Asia country) (4.15)** (3.76)** (2.61)* (3.06)** (3.07)** (2.47)** 11 (South Asia dummy) 0.153 0.247 (per capita official ( 0.86) ( 1.09) remittances) Constant 21.768 20.759 20.577 20.522 23.557 22.679 22.932 22.843 (24.63)** (23.75)** (17.43)** (17.33)** (23.13)** (21.98)** (15.15)** (15.05)** N 153 153 95 95 153 153 95 95 Adjusted R2 0.764 0.787 0.791 0.791 0.770 0.782 0.761 0.761 F-Statistic 247.39 188.49 90.45 72.30 255.72 183.11 75.83 61.02 Source: Authors' calculations. Note: Estimates obtained using ordinary least squares. All variables expressed in logs. T-ratios shown in parentheses. See appendix 1 for countries and survey dates. *Significant at the 0.05 level. **Significant at the 0.01 level. 12 MIGRANT LABOR REMITTANCES IN SOUTH ASIA in all cases. The magnitude of the poverty elasticities with respect to income and income inequality (Gini coefficient) is consistent with other recent analy- ses of poverty reduction (Adams 2003a; Ravallion 1997). When a shift dummy for South Asia is introduced into the model (columns 2­3 and 7­8), the dummy coefficients are positive and highly significant. This means that, controlling for level of income and income inequality, poverty is signifi- cantly higher in South Asia than in other regions of the world. This result parallels the findings in table 2.1. In table 2.2 the remittances variable--per capita official remittances-- has a negative and significant impact on poverty in developing countries (columns 4 and 8). In other words, with all other variables held constant, a higher level of official international remittances can be statistically expected to reduce both the poverty headcount and the poverty gap in the devel- oping world. However, the results for the interactive dummy variable for South Asia (South Asia dummy times official remittances) are negative but not statistically significant. These results suggest that although official international remittances reduce poverty in the developing world, in South Asia official remittances have no statistical impact on the level and depth of poverty. This is an unexpected outcome, one that needs to be further analyzed. Estimating Unofficial Remittances: Simulation Functions It is possible that official international remittances have no statistical impact on poverty in South Asia because these remittance figures count only a por- tion of total remittance flows. In other words, if unofficial remittances could be estimated and added to official remittances, then the impact of total remittances (official and unofficial) on poverty might be different. The hypothesis here is that unofficial remittances are more likely to be earned by the poor, because the poor are more likely to avoid banks and official channels (Adams 2003b). It is therefore possible that when remittance fig- ures include unofficial remittances, they will tend to statistically reduce poverty. This hypothesis seems particularly plausible in South Asia, where various observers have noted that a large proportion of remittances flows through unofficial channels.13 One way to estimate the size of the remittance monies that is sent back through unofficial channels is to use the simulation model suggested by El-Qorchi, Maimbo, and Wilson (2003). This model states that the share of unofficial remittances in total (official and unofficial) remittances can be analyzed as a function of the black market premium prevailing in different developing countries.14 Results for the model are obtained by assigning dif- ferent values for the minimum (MIN) and maximum (MAX) share of unof- ficial remittances in total remittances, according to country characteristics THE DEVELOPMENT IMPACT OF REMITTANCES 13 and exchange rate histories. For example, El-Qorchi, Maimbo, and Wilson (2003) note that the hawala system of unofficial remittances is deeply entrenched in Pakistan, and so the choice value for the minimum and max- imum shares of unofficial remittances in total remittances would be high. As El-Qorchi, Maimbo, and Wilson (2003, Appendix II) emphasize, this method of selecting values for minimum and maximum shares of unofficial remit- tances in total remittances is "judgmental," based on current knowledge of the various factors in individual countries. Table 2.3 summarizes the results of the model simulation by listing the values for official remittances (RP), unofficial remittances (RI) and total remittances (official and unofficial) (R) for the five South Asian countries.15 For the five South Asian countries as a whole, when the values for unofficial remittances are combined with those for official remittances, total remittances Table 2.3 Comparison of Official and Unofficial Remittances for South Asia US$ millions Unofficial Total remittances remittances (official and Official (estimated from estimated unofficial Country Survey year remittances simulation model) remittances) Bangladesh 1984 527 631 1,158 Bangladesh 1986 497 805 1,302 Bangladesh 1989 771 902 1,673 Bangladesh 1992 848 1,010 1,858 Bangladesh 1996 1,217 1,110 2,327 India 1983 2,311 444 2,755 India 1986 2,105 1,233 3,338 India 1988 2,402 1,101 3,503 India 1990 1,875 1,425 3,300 India 1995 7,685 4,123 11,808 India 1997 10,688 4,044 14,732 Nepal 1985 39 0 39 Nepal 1995 101 75 176 Pakistan 1988 2,013 442 2,455 Pakistan 1991 1,848 1,262 3,110 Pakistan 1993 1,562 1,642 3,204 Pakistan 1997 1,409 4,013 5,422 Sri Lanka 1985 292 25 317 Sri Lanka 1990 401 601 1,002 Sri Lanka 1995 790 646 1,436 Total 39,381 25,534 64,915 Source: Authors' calculations. 14 MIGRANT LABOR REMITTANCES IN SOUTH ASIA (official and unofficial) increase by 64.8 percent: from US$39.4 billion to US$64.9 billion. In other words, in the latter situation unofficial remittances represent about 39 percent (US$25.5 billion/US$64.9 billion) of the total value of remittances. Official and Unofficial Remittances and Poverty It is possible to take the estimated values for unofficial remittances, com- bine them with the official remittance figures, and then re-estimate the poverty equations used in the section on the new dataset. This is done in table 2.4. The most important results in table 2.4 are contained in columns (4) and (8), which show the findings for the interactive dummy variable for South Asia. For the poverty headcount measure, this interactive variable (South Asia dummy times official and unofficial remittances) is nega- tive and statistically significant. For the poverty gap measure, this inter- active variable is also negative and just misses the 5 percent level of significance. In other words, when the estimated values for unofficial remittances are added to official remittance figures, total remittances (offi- cial and unofficial) reduce the level of poverty in South Asia.16 On aver- age, the point estimates for the poverty headcount measure (column 4) suggest that a 10 percent increase in total remittances (official and un- official) will lead to a 0.9 percent decline in the level of poverty. This means that for a "representative" country, where exactly one-half of the population lives below the poverty line, a 10 percent increase in total remittances (official and unofficial) will bring the proportion of the popu- lation living in poverty down to about 0.48 percent. The question now arises: Why do total remittances (official and unof- ficial) have a statistical impact on reducing poverty in South Asia, while official remittances do not? The answer to this policy question has already been broached above. Unofficial remittances are more likely to be earned by the poor, because these are the people who are more likely to shun banks and to avoid official financial channels. Thus, when remittance fig- ures include only those funds that enter official banking channels, the impact of these transfers on the poor is underestimated. However, when remittance figures are revised to include estimates of unofficial remit- tances, the impact of these resource flows on the poor becomes more evident. These factors are all the more critical in those regions of the world--such as South Asia--where a large share of remittance monies has tended to flow through unofficial and informal channels. In South Asia (and perhaps the Middle East) it is very important to estimate the size of unofficial remittance flows to arrive at a proper, more correct under- standing of the impact of international remittances (official and unofficial) on poverty. Table 2.4 Elasticity of Poverty, Estimated Using Official and Estimated Unofficial International Remittances (unofficial remittances estimated from simulation model) Dependent variable Poverty headcount ($1.08/person/day) Dependent variable Poverty gap Variable (1) (2) (3) (4) (5) (6) (7) (8) Per capita survey mean 2.263 2.090 1.829 1.832 2.584 2.434 2.200 2.206 ( 20.60)** ( 18.61)** ( 14.84)** ( 14.81)** ( 20.42)** ( 18.36)** ( 14.29)** ( 14.33)** Gini coefficient of country 3.832 4.217 4.085 4.080 4.792 5.126 4.942 4.931 (14.32)** (15.59)** (13.98)** (13.92)** (15.55)** (16.05)** (13.55)** (13.52)** Per capita official and 0.162 0.156 0.145 0.133 unofficial remittances ( 3.85)** ( 3.62)** ( 2.76)** ( 2.47)* 15 South Asia dummy 0.856 0.945 1.202 0.745 0.781 1.348 (1 if South Asia country) (4.15)** (4.64)** (2.46)* (3.06)** (3.07)** (2.22)* (South Asia dummy) 0.099 0.220 (per capita official and ( 1.98)* ( 1.53) unofficial remittances) Constant 21.768 20.759 19.262 19.263 23.557 22.679 21.304 21.306 (24.63)** (23.75)** (21.05)** (21.00)** (23.13)** (21.98)** (18.66)** (18.66)** N 153 153 134 134 153 153 134 134 Adjusted R2 0.764 0.787 0.792 0.792 0.770 0.782 0.764 0.765 F-Statistic 274.39 188.49 128.21 102.10 255.77 183.11 109.15 87.57 Source: Authors' calculations. Note: Estimates obtained using ordinary least squares. All variables expressed in logs. T-ratios shown in parentheses. See appendix 1 for countries and survey dates. See text for details on estimating unofficial remittances using simulation model. *Significant at the 0.05 level. **Significant at the 0.01 level. 16 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Remittances, Investment, and Economic Development From the standpoint of economic development, the central question about international remittances is quite straightforward: How are these monies spent or used? Do migrant workers channel international remittances into productive investments at home, or do they use such monies merely to underwrite the consumption of newly desired consumer goods? Stated more baldly, do international remittances help provide the investment needed to facilitate economic development, or do they merely foster new pat- terns of dependence on status-oriented consumer goods for migrant work- ers and their families? Unfortunately, these questions are difficult to answer because of three related methodological problems. The first is the design of household budget surveys, which represent the best source for collecting information on how migrant households spend or use remittances. However, as noted above, few of these household surveys ask any questions at all about how inter- national remittances are used. For example, an analysis of all the World Bank Living Standard Measurement Surveys (LSMS) done on South Asian countries failed to uncover even one survey question concerning how remit- tance monies were spent.17 The second problem is fungibility: because inter- national remittances are like any other form of cash income, it is difficult to associate this income source with any particular changes in household expenditure behavior. Thus, in designing household surveys, it is not enough to just ask a household how it spent its remittance income, because that income could have freed other resources for expenditure on saving or invest- ment. The most effective approach is to design household panel surveys to collect information on the expenditure patterns of international migrant households over time. The final problem relates to the multiple-round effects of international remittances on development. For example, an inflow of remittances into a local economy may lead to a surge in expenditures in housing, which may, in turn, create new income and employment oppor- tunities for the poor and unskilled. Unfortunately, few studies have tried to evaluate the second- and third-order effects of international remittances on wages and employment.18 One of the few empirical studies that has examined how international remittances are spent or used in South Asia is by Adams (1998). Adams used a five-year panel dataset from rural Pakistan to examine how inter- national remittances were spent or invested in different assets over time.19 Although this study was based on a limited number of rural households (500), and was not representative of either rural Pakistan or Pakistan as a whole, its findings are suggestive. Adams found that international remittances (official and unofficial) account for only a small share--between 4.8 and 11.7 percent--of total THE DEVELOPMENT IMPACT OF REMITTANCES 17 household income in rural Pakistan. When households are ranked into quin- tile groups based on total per capita income, these international remittances go mainly to the rich. According to the data, households in the top income quintile receive 13.8 percent of their household income from international remittances, while households in the poorest quintile receive only 1.0 per- cent of their household income from international remittances. One of the main reasons for this phenomenon seems to be the high entry costs to inter- national migration. In rural Pakistan most international migrants go to work in the Persian Gulf, and the average costs of international migration--about US$1,300--are too high for lower-income households. In his study Adams used an asset accumulation model to estimate the impact of international remittances on the accumulation of four different types of assets. These are all physical assets: irrigated land owned; rain-fed land owned; livestock assets (value of animals); and nonfarm assets (value of vehicles, bikes, stores, or buildings). Efforts to estimate the impact of international remittances on the accumulation of human capital assets, such as change in male and female education, failed to produce meaningful results.20 The analysis found that international remittances had a positive and sig- nificant effect on the accumulation of two assets in rural Pakistan: irrigated and rain-fed land. These results, which parallel those of other studies regard- ing the propensity of international migrants to invest in land,21 are impor- tant for two reasons. First, they show that despite the high cost of land in rural Pakistan--an estimated US$3,420 per acre for irrigated land in Punjab province--remittances can and do lead to rural asset accumulation. Second, these findings show that rather than wasting their remittance earnings on increased consumption (more food, more lavish social ceremonies), inter- national migrants prefer to invest their remittance money. In rural Pakistan international migrants do invest and they actually exhibit a higher propen- sity to invest than do their nonmigrant counterparts. However, it should be emphasized that this study found that interna- tional remittances do not have a significant effect on the accumulation of nonfarm assets (value of stores, vehicles, or buildings). This finding is dis- appointing because some studies have suggested that international remit- tances may provide the capital for increased investment in new business and mercantile activities. However, in rural Pakistan international migrants do not invest in business activities because apparently the rates of return on this investment are lower than those for land. Moreover, since many of the international migrants from rural Pakistan are government bureaucrats or farmers, they probably lack the necessary mercantile skills to start a new business when they return. In their investments migrants seem to avoid investing in areas that they do not know (such as business) in favor of com- mitting their earnings to what they know best (namely, land). 18 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Conclusion This chapter has shown that international migration and remittances are important to poverty and economic development in South Asia. International remittances (official and unofficial) reduce poverty in South Asia, and international remittances do lead to asset accumulation--mainly land accumulation--in rural areas. Obviously, more research work needs to be done to clarify the impact of international remittances on poverty and economic development in South Asia. Such work should be focused on three areas. First, with respect to data, the IMF should make greater efforts to count the amount of remit- tance monies that are currently transmitted through informal, unofficial channels. Since it is likely that poor people remit more through infor- mal channels, a full and complete accounting of the impact of international remittances (official and unofficial) on poverty in South Asia needs more accurate data on the large level of unofficial remittances returning to that region. Second, with respect to research methods, more household panel sur- veys need to be implemented in South Asian countries to collect informa- tion on the expenditure patterns of international migrant households. These surveys need to be conducted on nationally representative samples of urban and rural households, and should include modules designed to measure the second- and third-round effects of international remittances on wages and employment. A full and complete accounting of the impact of interna- tional remittances on local-level economic development requires more detailed and complete household-level data. Third and finally, more research and policy work needs to be undertaken to encourage policy makers in South Asia to improve the investment cli- mate for returning migrant workers. If, for example, policy makers would like to see a larger share of international remittance monies invested in busi- ness, then they need to undertake the types of infrastructure improvements-- better roads, improved telephone systems and electrical grids--that would encourage expanded commerce and trade. Providing small business loans to new firms would also do much to improve the business climate for return- ing migrants. The point to emphasize here is that international migrants do invest their remittance earnings, and they tend to invest these earnings in areas where financial rates of return are the greatest. Notes 1. See United Nations (2002). 2. These 71 countries are Algeria, Bangladesh, Belarus, Bolivia, Botswana, Brazil, Bulgaria, Burkina Faso, the Central African Republic, Chile, Colombia, THE DEVELOPMENT IMPACT OF REMITTANCES 19 Costa Rica, Côte d'Ivore, the Czech Republic, the Dominican Republic, Ecuador, the Arab Republic of Egypt, El Salvador, Estonia, Ethiopia, The Gambia, Ghana, Guatemala, Honduras, Hungary, India, Indonesia, the Islamic Republic of Iran, Jamaica, Jordan, Kazakhstan, Kenya, the Kyrgyz Republic, Latvia, Lesotho, Lithuania, Madagascar, Mauritania, Mexico, Moldova, Morocco, Mozambique, Namibia, Nepal, Nicaragua, Nigeria, Pakistan, Panama, Paraguay, Peru, the Philippines, Poland, Romania, the Russian Federation, Senegal, Sierra Leone, South Africa, Sri Lanka, Thailand, Trinidad and Tobago, Tunisia, Turkey, Turkmenistan, Uganda, Ukraine, Uruguay, Uzbekistan, Venezuela, the Republic of Yemen, Zambia, and Zimbabwe. 3. The full list of these 157 countries appears in World Bank (2001: 334). 4. More details on data problems, especially those concerned with estimating international migration variables, can be found in Adams and Page (2003). 5. The IMF records annual flows of official international remittances in its annual publication, Balance of Payments Statistics Yearbook. 6. To ensure compatibility across countries, all of the poverty lines in appendix 1 are international poverty lines, set at estimates of $1.08 per person per day in 1993 purchasing power parity (PPP) exchange rates. The PPP exchange rates are used so that $1.08 is worth roughly the same in all countries. PPP values are calculated by pricing a representative bundle of goods in each country and comparing the local cost of that bundle with the U.S. dollar cost of the same bundle. In calculat- ing PPP values, the comparison of local costs with U.S. costs is done using conver- sion estimates produced by the World Bank. 7. Although a transfer of expenditures from a poor person to a poorer person will not change the headcount index or the poverty gap index, it will decrease the squared poverty gap index. 8. Although the Persian Gulf countries (Kuwait, Oman, Saudi Arabia, and the United Arab Emirates) appear to keep detailed records on the number of interna- tional migrants working within their borders, they do not publish such data. 9. The growth-poverty model can be written as log Pit i 1log it 2 log (git) 3log (xit) it where P is the measure of poverty in country i at time t, 1 is the "growth elastic- ity of poverty" with respect to mean per capita income given by , 2 is the elastic- ity of poverty with respect to income distribution given by g. 3is the elasticity of poverty with respect to variable x (such as official international remittances), and is an error term that includes errors in the poverty measure. 10. In addition to Ravallion (1997) and Ravallion and Chen (1997), see Collier and Dollar (2001). 11. See, for example, Birdsall and Londoño (1997) and Ravallion (1997). 12. One possible problem with this model is that it assumes that all of the right- hand side variables--including international remittances--are exogenous to poverty. However, it is possible that international remittances may be endogenous to poverty. Reverse causality may be taking place: international remittances may 20 MIGRANT LABOR REMITTANCES IN SOUTH ASIA be reducing poverty, but poverty may also be affecting the level of international remittances being received. One means of accounting for this reverse causality is to pursue an instrumental variables approach. When this is done, results for the instrumented variable (international remittances) are more negative and of greater statistical significance than those for the noninstrumented variable. The extent of this reverse causality is small, however. For example, while the instrumented estimates for the poverty headcount measure suggest that a 10 percent increase in per capita official international remittances will lead to a 3.5 percent decline in the share of people living in poverty, the noninstrumented estimates in table 2.2 suggest that a similar increase in official international remittances will lead to a 1.8 percent decline in the share of poor people. 13. See, for example, El-Qorchi, Maimbo, and Wilson (2003). 14. The full model elaborated by El-Qorchi, Maimbo, and Wilson (2003) can be written as: RI/R 0 1B 2B2 3B3 , where RI is unofficial remittances, R is total remittances (official and unofficial), RP is official remittances (as recorded in the balance of payments accounts), R equals RI plus RP, B is the black market premium (in percent of the official exchange rate), MIN is the intercept ( ) or the minimum share of unofficial remittances in total remittances (official and unoffi- cial) and MAX is the maximum share of unofficial remittances in total remittances (official and unofficial). 15. Full results for all observations from the 71 countries in the dataset are available from the author. 16. Columns (4) and (8) of table 2.4 also show that total remittances (official and unofficial) significantly reduce the level and depth of poverty in the developing world as a whole. 17. None of the following LSMS household surveys done in the South Asia region contained any questions on how remittances were used or spent: India, 1997­8; Nepal, 1991; Pakistan, 1991; and Sri Lanka, 1999­2000. In all fairness, it should be noted that this problem is not confined to South Asia. The following LSMS surveys also failed to include any questions on the uses of remittances: Brazil, 1996­7; Morocco, 1991; and Vietnam, 1997­8. 18. For one notable exception to this statement, see Taylor and Adelman (1996). 19. Adams's study (1998) was based on a panel of 500 households in four rural Pakistani districts over the period 1986­7 to 1990­1. 20. There was very little change observed in male and female education over the course of the five-year panel. 21. See, for example, Adams (1991) and Russell (1986). 3 Public Infrastructure for Remittances All the countries in the region have taken active steps to support the migra- tion of nationals. The public infrastructure needed to support the search for employment abroad, the migration of successful applicants, and their subsequent stay is, in principle, in place. This chapter considers the basic public infrastructure and incentives that have been established in recent times in each of the four countries. Bangladesh Remittances to Bangladesh have been growing steadily over the last decade. Since its independence in 1971, more than 3 million Bangladeshis have left the country in search of employment. The central bank estimates their cumulative remittances during 1976­2003 at around US$22 billion.1 Recognizing their economic importance, the government for years has had legislation, policies, and an institutional structure in place to facilitate the migration of its citizens. Currently three government agencies manage the welfare of migrants: the Ministry of Expatriates' Welfare and Overseas Employment; its Bureau of Manpower, Employment, and Training (BMET); and Bangladesh Overseas Employment and Services, Ltd (BOESL).2 Administratively, the Ministry of Expatriates' Welfare and Overseas Employment is responsible for all public sector organizations and compa- nies dealing with overseas employment. It also registers recruitment agen- cies, manages the Labour Wing in Bangladeshi missions abroad, and administers the Wage Earners' Welfare Fund. Because it is directly respon- sible for migrants, the ministry handles complaints from expatriates, man- ages projects for expatriate participation in local economic and social welfare activities, facilitates expatriate investments in the country, promotes Bangladeshi culture among expatriates abroad, and maintains ties with associations of Bangladeshis abroad. BMET has seven specific functions: (1) to promote employment in Bangladesh and abroad; (2) to protect the interests of emigrants; (3) to ensure the welfare and remittances of Bangladeshis working abroad; (4) to provide 21 22 MIGRANT LABOR REMITTANCES IN SOUTH ASIA vocational guidance and employment counseling; (5) to promote self- employment by distributing tool kits and preparing investment schedules; (6) to maintain liaison with international agencies regarding development of training and employment; and (7) to compile and publish reports on employment and to carry out research and studies on action-oriented programs. BOESL projects Bangladesh as a reliable source of quality manpower through regular publicity and advertising in foreign media, and through employment-promotion campaigns in labor-importing countries. Its work is carried out either independently through its numerous agencies, branches, and offices in Bangladesh and abroad, or in collaboration with private sector agents. To this end it acts as a consultant and recruitment agency for Bangladeshis looking for foreign employment, particularly on large civil and mechanical construction projects. It arranges employment tests, med- ical tests, tickets, and other facilities for persons selected for foreign employ- ment, either independently or in collaboration with other organizations. Thereafter, it continues to look into the welfare of migrants while motivat- ing and mobilizing Bangladeshi citizens living abroad into productive investment in the country through joint ventures and other commercial enterprises. Together, and in coordination with other arms of government, the three agencies facilitate the marketing of potential migrants, the migration of suc- cessful candidates, and their sojourn abroad and eventual return home. · Premigration: Bangladesh actively markets its labor resources abroad through labor attachés posted to missions in foreign countries. These attachés coordinate visits of ambassadors and ministers with their coun- terparts, as well as employer groups, in host countries. Licensed agents in the country market their employer clients by contacting prospective foreign workers directly. Interested candidates are briefed at centers run by the BMET. Training is provided, particularly for semiskilled and unskilled workers who form the majority of Bangladeshi migrants (table 3.1). First-time travelers receive instructions on procedures for leaving the country. · Migrants: Migrants are said to have access to welfare officers in their governments' missions to submit complaints of nonpayment or under- payment of wages, lack of medical facilities, or poor food. In cases where no amicable resolution of the problem is available, migrants are given legal assistance in seeking redress. Sometimes, when the migration expe- rience fails, the government is called upon to assist in the repatriation of migrant workers from host countries, including the repatriation of the remains of workers who have died abroad. To assist with these and other expensive support systems, migrants are expected to contribute to a PUBLIC INFRASTRUCTURE FOR REMITTANCES 23 Table 3.1 Official Flow of Bangladeshi Workers by Their Skill Composition Year Professional Skilled Semiskilled Unskilled Total 1998 9,574 74,718 51,590 131,785 267,667 1999 8,045 98,449 44,947 116,741 268,182 2000 10,669 99,606 26,461 85,950 222,686 2001 5,940 42,742 30,702 10,581 188,965 2002 14,450 56,265 36,025 109,285 216,025 Source: Siddiqui and Abrar 2003. welfare fund. The fund helps with legal support in the host country, initial sustenance, repatriation, and legal support to family members in the country of origin. Throughout this process, particular attention is paid to the convenience of sending remittances to Bangladesh. Numerous initiatives address this specific concern--among them (1) the appointment of bank representatives in major labor-importing countries, (2) agreements with banks in host coun- tries to facilitate remittances, (3) special fiscal incentives to encourage and facilitate investments in privatized Bangladeshi industries, (4) special savings incentives in the form of a Wage Earners Bond, and (5) housing opportu- nities in government-land developments. India The Emigration Act of 1983 safeguards the interests and ensures the wel- fare of Indians employed overseas on a contractual basis. The Ministry of Labour administers the Act and supervises the emigration of Indian work- ers through the eight offices of the Protectors of Emigrants. The other min- istries involved in the emigration of Indians are the Passport Issuing Authority, India's diplomatic missions under the Ministry of External Affairs, and the Airport Immigration Authorities of the Bureau of Immigration in the Ministry of Home Affairs. · Premigration: Although some state governments have set up manpower corporations to promote labor exports from their states, the central gov- ernment assumes no role in promoting emigration. Centrally organized predeparture orientations or services are not yet available to migrant workers. Instead, the private sector plays a significant role in expand- ing labor migration. Recruiting agencies search out new areas and 24 MIGRANT LABOR REMITTANCES IN SOUTH ASIA avenues for deployment of Indian workers abroad. Under the Emigration Act, only agencies registered with the Ministry of Labour and holding a registration certificate can conduct the business of recruiting for over- seas employment.3 A foreign employer may recruit any Indian citizen for employment abroad either through a recruitment agent authorized under the act to perform such recruitment, or directly in accordance with a valid permit issued by the government. Indian workers are also deployed by Indian firms on project sites abroad. The recruitment agent is authorized to collect service charges from each worker, as fixed by the government. · Migrants: India's diplomatic posts in countries where there is a large con- centration of Indian expatriate workers have a dedicated labor attaché or welfare officer. Posts provide various forms of assistance to migrants, including assistance in settling disputes. Most of the complaints received from Indian migrants relate to nonpayment or delayed payment of wages, irregular working hours, modification of the labor contract to their dis- advantage, poor working and living conditions, stranding because of foreign employers' unwillingness to meet workers on arrival, and non- payment of compensation in the case of injuries resulting from accidents in the course of employment.4 Reports of physical abuse of domestic workers, especially housemaids, are received from Kuwait and other countries of the Middle East. Overall, most complaints are received from workers in the Gulf region and in Southeast Asia, notably Malaysia. Although India does not have specific national programs and services to assist in the reintegration of returning migrants, it does provide vari- ous forms of incentives for nonresident Indians (NRIs) to reinvest in the country. Many of India's migrants have established themselves in other countries and now hold dual citizenship. One of the most important incentives for NRIs to invest in India is the ability to move capital freely between their two homes without undue bureaucratic constraints. In 2004 alone, this group remitted substantially through formal banking channels (table 3.2). In this regard, the government has sought to be accommodating. On January 10, 2004, the Minister of Finance announced that "any citizen can walk into a bank, buy dollars up to US$25,000 a year and remit it abroad-- anywhere he likes, for whatever he wants."5 Despite the annual cap of US$25,000 on free outward remittances, this is the first time that a resi- dent need not specify the underlying purpose of the foreign exchange transaction. Previously, capital convertibility existed only for NRIs. Full convertibility is expected to take years and may depend on whether India can achieve lower fiscal deficits and manage inflation, among other con- ditions. The announcement of January 2004 was driven by the need to PUBLIC INFRASTRUCTURE FOR REMITTANCES 25 Table 3.2 NRI Deposits as at End March 2004 (US$ millions) Year FCNR(A) FCNR(B) NR(E)RA NR(NR)RD Total 1991 10,103 -- 3,618 -- 13,721 1992 9,792 -- 3,025 -- 12,817 1993 10,617 -- 2,740 621 13,978 1994 9,300 1,108 3,523 1,754 15,685 1995 7,051 3,063 4,556 2,486 17,156 1996 4,255 5,720 3,916 3,542 17,433 1997 2,306 7,496 4,983 5,604 20,389 1998 -- 8,467 5,637 6,262 20,367 1999 -- 7,835 6,045 6,618 20,498 2000 -- 8,172 6,758 6,754 21,684 2001 -- 9,076 7,147 6,849 23,072 2002 -- 9,673 8,449 7,052 25,174 2003 -- 10,199 14,923 3,407 28,529 2004 -- 10,961 20,559 1,746 33,266 Source: Reserve Bank of India 2004. Note: FCNR(A): Foreign Currency Nonresident (Accounts); FCNR(B): Foreign Currency Nonresident (Banks); NR(E)RA: Nonresident (External) Rupee Accounts; NR(NR)RD: Nonresident (Nonrepatriable) Rupee Deposits. --. Program not in place or discontinued. create a demand for the U.S. dollar, which had been sliding against the Indian rupee. In the face of the US$7 billion in foreign direct investment (FDI) inflow and even greater remittances, the Indian currency appreci- ated by more than 5 percent in 2003 despite regular intervention by the central bank. There have been other reforms as well. Indian companies can now invest up to 100 percent of their net worth overseas. Previously, a ceiling of US$100 million had been imposed. Indian companies are now allowed to undertake agricultural ventures abroad. Places are reserved in educa- tional institutions for the children of nonresident Indians in the Persian Gulf region and Southeast Asia. Those in the Gulf countries are eligible to pay lower resident fees at educational institutions. Pakistan Pakistan has a comprehensive system of emigration under the supervision of the Ministry of Labour, Manpower, and Overseas Pakistanis. Under the Emigration Ordinance of 1979, the ministry's Bureau of Emigration and Overseas Employment regulates labor emigration in the private sector, while the Overseas Employment Corporation deals with labor emigration 26 MIGRANT LABOR REMITTANCES IN SOUTH ASIA in the public sector. Emigration of Pakistanis for purposes other than foreign employment is controlled by the Ministry of Interior through its depart- ments--the Federal Investigations Agency, the Directorate General of Passports, and the Immigration and National Database and Registration Authority. · Premigration: The government organizes premigration briefings on the customs, local conditions, and relevant laws of destination countries. The briefings are held for all registered migrants in offices of the Protector of Emigrants in Karachi, Lahore, Rawalpindi, and Peshawar. The gov- ernment tries to improve the skills and attitudes of the workers in demand abroad to bring them in line with international norms and standards. Foreign recruiting, however, is primarily managed by overseas employ- ment promoters in the private sector. About 70 percent of foreign employ- ment is handled by licensed overseas employment promoters. The remaining 30 percent of workers find employment through their own efforts. · Migrants: Community welfare attachés are posted in Pakistan's diplo- matic missions in manpower-importing countries. Attachés maintain liaison with Pakistani workers and help them solve problems in coordination with authorities in the host country. A welfare fund estab- lished under the Emigration Ordinance of 1979 is managed by the Overseas Pakistanis Foundation (OPF) for the welfare of migrant workers and their dependents in Pakistan. OPF provides education, training, housing, and medical facilities for the families of overseas Pakistanis. Over the past few years, the government has announced several mea- sures to encourage Pakistanis abroad to remit their funds to Pakistan, espe- cially those in the Persian Gulf States (table 3.3). In June 2001 the minister of finance announced the following benefits: Table 3.3 Top Five Countries of Destination for Pakistani Contract Migrant Workers Country Number of migrants (Jan­Nov 2002) Saudi Arabia 94,852 United Arab Emirates 29,593 Kuwait 2,755 Bahrain 955 Qatar 410 Source: IOM 2003. PUBLIC INFRASTRUCTURE FOR REMITTANCES 27 Overseas Pakistanis remitting US$2,500 per annum through banking channels to Pakistan are entitled to: · Separate immigration and customs counters at all international airports at arrival and departure · Free renewal of passports on a priority basis · Duty-free import of items of personal convenience up to US$700 during any year Nonresident Pakistanis (NRPs) remitting a minimum of US$10,000 through banking channels are entitled to: · Special access to the merit-based quota system assigned in all public pro- fessional colleges and universities · Duty-free import of items of personal convenience of a value up to US$1,200 per year · An allocation of up to 25 percent for initial public offerings to be sub- scribed in a foreign currency · Participation in a lottery for choice plots in public housing schemes at attractive prices if paid in a foreign currency · Adiscount in the auctions of the Corporate and Industrial Rehabilitation Corporation (CIRC) if payment is made in a foreign currency · Expedited allocation of shares in privatizations In addition, bona fide remittances are not subject to tax. Overseas Pakistanis may remit money freely to their families in Pakistan for con- sumption or investment without any interference from tax authorities. The income of Pakistani seafarers working on foreign vessels remitted to Pakistan through normal banking channels also is exempt from income tax. Also, banks have been directed to review their arrangements for remit- tances to speed up the remittance process and ensure outreach to labor migrants by establishing prior contacts with the recipients. Both residents and nonresidents may now maintain foreign currency accounts without risk of freezing or seizure. In addition, the following measures are being adopted to encourage the participation of professionals in remittance and investment: · Exclusive investment products will be marketed to nonresident Pakistanis. · Banks will be encouraged to offer new products for nonresident Pakistanis, who now may hold and manage their own foreign currency deposits. · The market for private pension funds will be developed and promoted with a view to attracting investment from nonresidents. · A Web site will host information about charities for the benefit of non- residents who wish to support them. 28 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Sri Lanka Two departments are directly active in the migration and remittances busi- ness in Sri Lanka: the Ministry of Employment and Labour and its imple- menting arm, the Sri Lanka Bureau of Foreign Employment (SLBFE). The ministry is responsible for formulating policies and monitoring the overall administration of foreign employment as well as coordinating with relevant stateagencies.SLBFEimplementsawiderangeofworkers'welfareprograms, both locally and in host countries, that protect migrant workers during the entire migration process. SLBFE is responsible for monitoring the migration of individuals, the activities of employment agencies, and the promotion of foreignemploymentopportunities,especiallytothePersian Gulfcountries to which the majority of unskilled workers emigrate (table 3.4). Programs to ensure the welfare of families of migrant workers are being prepared. · Premigration: Sri Lanka has a comprehensive range of premigration facili- ties and services that include identifying foreign employment opportu- nities, predeparture training, and predeparture loan schemes. · Identification of foreign employment opportunities: SLBFE seeks out employ- ment opportunities for its nationals abroad. The majority of oppor- tunities, however, are identified and filled by the numerous foreign employment agencies registered and supervised by the SLBFE. Traditionally, the largest markets have been for unskilled laborers, par- ticularly housemaids, in the Middle East. However, new markets are opening up in Europe and Southeast Asia. Recent efforts have been made to increase the number of male migrants, increase opportunities for skilled workers, and diversify the geographical destinations of Sri Lankan migrants. Table 3.4 Official Flow of Sri Lankan Workers to the Middle East by Skill Composition Professional Middle Clerical Country level level related Skilled Unskilled Housemaid Total Saudi Arabia 319 2,810 1,853 13,774 16,349 40,728 75,830 Kuwait 51 378 408 6,518 4,725 26,178 38,258 United Arab Emirates 472 1,912 2,414 10,141 8,469 8,762 32,173 Qatar 284 878 920 9,138 9,198 3,281 23,699 Lebanon 5 14 10 55 304 12,787 13,175 Jordan 12 264 108 3,527 445 2,720 7,076 Oman 85 416 265 1,361 385 1,572 4,084 Bahrain 55 205 175 976 573 1,729 3,713 Source: Sri Lanka Bureau of Foreign Employment 2004: 5. PUBLIC INFRASTRUCTURE FOR REMITTANCES 29 · Training: In coordination with the Ministry of Vocational Training and other government bodies, SLBFE provides training to prospective migrants at 22 training centers and works with an additional eight private insti- tutions. Provided free of charge to all who desire foreign employment (as long as they are registered with the SLBFE or a private employment agency), the training includes immigration procedures, customs and etiquette in the host country, and basic financial management skills. The training for housemaids destined for the Middle East, who constitute as much as 80 percent of the migrant labor force, is particularly important given the number of complaints received from that region and the high social costs associated with female migrants: sexual abuse, nonpayment of wages, excessively long working hours, and general physical abuse. · Predeparture loan schemes: Through the state banks, the SLBFE also pro- vides predeparture loans to cover departure expenses, including travel costs. The loans are available at a flat interest rate of 8 percent. The differ- ence between that rate and the market interest rate is subsidized by the government. The loans can be used for the purchase of a plane ticket, for Sri Lankan Foreign Bureau registration fees, and for medical expenses and other necessities. To obtain the loan, applicants must open an account with a state bank and must be 18­50 years of age, hold a valid visa, a letter of employment from an employment agency in Sri Lanka or employer abroad, and a certificate from the SLBFE. Once these requirements are met, the applicant need only complete an application, obtain a draft for at least US$50 as an initial deposit, and provide a photocopy of the page of the passport that contains the applicant's photograph and personal data. · Predeparture insurance schemes: The SLBFE also provides free life insur- ance to eligible migrants and their families.6 A migrant becomes eligible on producing the following documents: passport, return ticket and pur- chase receipt, receipt of registration payment with the SLBFE, medical records if the migrant has been sick, documents from the employer or employment agency, a letter from the agency (if the migrant returns within three months, the migrant must then provide another letter from the agency), and a bank passbook. · Migrants: Once abroad, migrants are permitted to maintain nonresident foreign currency (NRFC) accounts through which they may send their remittances. Although all banks provide NRFC accounts, there is a clear market segmentation between the foreign banks and the state-run banks. The former concentrate on middle- and higher-income skilled profes- sionals, while the latter hold the accounts of the majority of lower-income unskilled laborers. The following categories of migrants are eligible to open NRFC accounts: (1) Sri Lankans employed abroad or within 90 days of return to Sri Lanka; (2) nonnationals of Sri Lankan origin living abroad; and (3) Sri Lankan nationals who go oversees to study or participate in conferences, subject to producing documentary evidence that they have brought into the country any funds received from overseas institutions.7 30 MIGRANT LABOR REMITTANCES IN SOUTH ASIA In its recently published National Employment Policy, the government made it clear that it will seek to "adopt a proactive approach to identifying global employment opportunities and uplift the image and skills of migrant human capital by providing them with appropriate training to enhance their competitiveness."8 To achieve this, the government plans to reorganize the SLBFE to make it more market oriented and effective in its promotional and training activities, particularly for domestic female workers. Conclusion Remittances to the South Asia region have been growing steadily over the past decade. The economic benefits of that growth are clear--remittance income enables families to supplement their local income. To the extent remittances are transferred through formal financial institutions, the gov- ernment's balance of payments benefits from the increased foreign exchange reserves. Governments have noticed and taken various forms of initiatives-- with varying degrees of success--to facilitate migration of their nationals, the welfare of their citizens abroad, and the remittance of earnings to their home countries. Government legislation, dedicated departments, labor training centers, labor attachés in embassies abroad, and a plethora of tar- geted incentives are part of a far-reaching public framework for remittances in the region. Consistently improving the effectiveness of this framework will substantially aid the development impact of migrant remittances in the years to come. Notes 1. See Azad (2005). 2. See the International Organization for Migration (IOM) (2003). 3. The certificate is granted after checking, among other things, the agent's financial soundness, trustworthiness, adequacy of premises, and experience in dealing with manpower export, and after depositing financial security. 4. A proposal is under consideration to establish a Central Manpower Export Promotion Council and an Indian Overseas Workers' Welfare Fund that would fund return tickets for workers stranded abroad, pay for repatriation of deceased workers, and grant assistance to workers who become partially or permanently disabled, among other purposes. 5. See The Economic Times (Channai), January 10, 2004. 6. In addition to this scheme, migrants have access to a welfare fund financed through a fee of US$25 levied on employers abroad. 7. The accounts may be held individually or jointly. Spouses may open a joint account in their names provided at least one is working abroad and the other is living with him or her. 8. See Sri Lanka Bureau of Foreign Employment (2004). 4 Formal Financial Sector Infrastructure for Remittances Most studies of workers' remittances have focused on the senders and receivers of remittances with little attention given to intermediaries, official or unofficial, that facilitate the process. This chapter discusses the formal financial infrastructure for remittances in South Asia available to migrant workers, with particular attention to the extensive state financial infrastructure already in place. The South Asia region is unique in its vast state banking and nonbanking financial insti- tutions. Although Latin America, Africa, and East Asia have largely pri- vatized their state banking infrastructure, the South Asia region continues to be dominated by large state banks with dense urban and rural branch networks. The formal financial system for remittances includes four principal actors: state commercial banks, private foreign commercial banks, local commer- cial banks, and money service businesses or companies. Remittancesthroughbanksaretypicallybywayofbankdrafts(10­15days) and telegraphic money transfers (48 hours). The former cost about US$1 from the Persian Gulf countries, while the latter cost slightly more, at US$4. Though the banks do not always have branches in destination areas such as the Persian Gulf countries, they are able to participate in the remittance busi- ness by partnering with remittance companies eager to take advantage of the significant branch networks the banks in the region have to offer. Slowly, other actors--such as post offices and microfinance institutions-- have begun exploring opportunities in the remittance industry. However, their participation is still in its infancy. State Commercial Banks In South Asia, the extensive branch network of state commercial banks is unique. These banks have long dominated the official remittance business, in part because of their large commercial branch networks, but also because they have enjoyed monopolistic control of foreign exchange transactions. In India, the 1970s and 1980s saw a rapid expansion of India's financial system into rural areas. Following Indira Gandhi's bank nationalization 31 32 MIGRANT LABOR REMITTANCES IN SOUTH ASIA drive, launched in 1969, commercial banks were required to open rural branches. Between 1973 and 1985, bank branches in rural areas grew at an average of 15.2 percent each year, about double the growth rate of branches in semi-urban (6.4 percent), urban (7.8 percent), and metropolitan (7.5 per- cent) areas. Rural branches grew from 1,833 in 1969 to 30,186 in 1985, an increase of 1,547 percent. Rural branches continued to expand in the 1980s to 35,000 by 1991, declining slightly to 32,400 in 2001. Today, India has over 32,000 rural branches of commercial banks and regional rural banks, and some 14,000 cooperative bank branches.1 The State Bank of India (SBI) alone, for example, has 9,000 branches in urban and rural India. Although it has no branches in the Middle East, it attracts remittances from that region through money exchange companies that receive local currency from remitters and convert the funds to a demand deposit that eventually is deposited in the SBI accounts in India. In the United States, SBI allows individuals to obtain demand drafts in Indian rupees using that day's exchange rate. SBI accepts cash for demand drafts valued up to US$1,000. For amounts greater than US$1,000, a cashier's check from the customer's U.S. bank is required. If a personal check is used, then time is required for clearance. Acharge of US$10 applies to drafts of Rs 50,000 or less, with no charge for greater amounts. If the recipient has an account with SBI in India, a wire transfer can be made. The amount is transferred in two business days for a fee of US$30 per transaction. SBI offers various remittances modes, including personal checks, cashier's checks, wire trans- fers, Automated Clearing House (ACH) debit authorizations, and account debits. Recipients may choose to receive the money in rupees or foreign currency, either in cash or as a direct deposit into their account. The banking system in Bangladesh is dominated by four nationalized commercial banks, which together operate at least 3,346 branches. Sonali Bank, the largest commercial bank in Bangladesh and fully state-owned, handles 40 percent of all remittances to Bangladesh. Sonali Bank has sub- sidiaries in the United States and the United Kingdom, a representative office in Saudi Arabia, and 23 agency offices in the Middle East. The U.S. subsidiary has six offices from New York to California. The U.K. subsidiary has five offices. The bank has arrangements with 22 foreign correspondent banks in the United States (including American Express and Banker's Trust) for facilitating remittances to Bangladesh. Sonali Bank guarantees to pay the remitted funds in 48 hours. A bank account can be opened without an initial deposit, and funds sent by workers can be used to purchase Wage Earners' Development Bonds. In Sri Lanka, People's Bank--the largest state bank--has 326 branches, 81 counter services, 188 pawning centers, and 97 ATMs,2 while Bank of Ceylon has 299 branches.3 Between them, the two state banks make a sub- stantial contribution to a Sri Lankan bank-branch density (per 10,000 people) of 0.69 in 2003.4 FORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 33 Private Foreign and Local Commercial Banks Increasingly, however, it is the private commercial banks that are eyeing the remittances market. Sensing a growing market for remittances to South Asia, several banks have introduced remittance services and started aggres- sive marketing programs. Foreign Commercial Banks Foreign banks in Bangladesh, India, Pakistan, and Sri Lanka are newcom- ers to the remittance business. Where they are involved, they have tended to focus on skilled migrants--doctors, accountants, lawyers, and other professionals--mainly account holders who have migrated. They are not known to offer competitively priced services to walk-in clients. A number of reasons are given for their reluctant participation in the sector--which is quite different from the experience in Latin America and Southeast Asia, where all banks are taking a serious interest in the remittances sector. · Limited branch networks: Most foreign banks are concentrated in the cap- ital cities. Their branch networks are designed for corporate clients and more affluent individual clients. They do not have the rural outreach of the state banks and local private banks. · Transaction costs: The high minimum account balances required by com- mercial banks are a deterrent for lower-income migrants. People are reluctant to give up access to sizeable funds in order to maintain a bank account.5 Further, the experience of banks that have held such accounts is that the recipient rarely keeps the money in the account, but rather withdraws it as it comes in. The transaction costs for maintaining an account with no float potential are high. · Remittance business model: The remittances business has three direct sources of income--profit on the exchange rate, commissions and fees paid by the remitter and sometimes by the recipient, and the interest earned on remitted funds before they are paid out to the recipient. On all three counts, the profits on remittances to the region have been falling. With the recent removal of exchange rate controls, the margins have declined dras- tically. Competition among transfer agents has pushed down commis- sions and fees to US$1 dollar or less. And pressure to make payments within 24 hours has minimized the potential for gains on cash float. Unless clients open bank accounts and banks are able to cross-sell other more profitable products, remittances are not an attractive business propo- sition for the larger private commercial banks. · Anti-money-laundering and counter-terrorist financing (AML/CTF) standards: Heightened international concerns over money laundering and terrorist 34 MIGRANT LABOR REMITTANCES IN SOUTH ASIA financing have increased the pressure on international banks to adhere strictly to international AML/CTF standards. High volumes of individ- ual money transfers, especially between nonaccount holders, make it difficult to comply with know-your-customer and suspicious-activity- reporting requirements. However, the limitations noted above have not limited all the foreign banks. Many have been aggressive in investing in the sector. Many are com- petitively offering a variety of products to migrants including life, accident, and baggage insurance, and predeparture loans at concessional rates to be repaid from remittance proceeds. The size of the loan is determined by the level of income offered by the prospective employer and the duration of employment. Some migrants use the loans to pay their employment agency fees, passage, and settling-in costs. Local Commercial Banks However, it is largely the local banks in all four countries that are taking the lead in investing heavily in remittance application technology. The com- bined use of noncash bank methods of payment such as credit cards, debit cards, Internet banking, and telebanking continues to rise. Technology is having a positive impact on the remittance market. As account holders in some of the more technologically advanced banks dis- cover the benefits of ATM machines, debit cards, and credit cards, the volume of account-to-account money transfers is increasing. The Internet is also being used as a method of transferring funds. Both developments give remitters ever more convenient access to banking services. For banks, the con- tribution of technology to the increase in account-to-account transfers is a welcome development. From a regulatory perspective, technology makes it easier to comply with know-your-customer requirements by automati- cally recording the identity of the remitter and the recipient. Financially, technology gives the bank the opportunity to cross-sell other financial products--such as credit cards, insurance products, and loans. Payment Card Infrastructure To facilitate the increased use of card-based small-value payment systems, there has been an equal increase in the number of automated teller machines (ATMs) installed in bank branches in the region. In Sri Lanka, for example, the total number of ATMs operated by commercial banks increased to 705 at the end of 2003 from 622 at the end of 2002, while there was a notable increase in e-banking and e-payment services (table 4.1). For a long time, the vast majority of the ATMs in Sri Lanka were used by customers only to withdraw cash from or deposit cash to their accounts FORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 35 Table 4.1 ATMs in Sri Lanka ATMS 2002 2003 1Q04 2Q04 Number of machines 636 721 760 783 Volume of transactions 25,124 33,491 9,938 8,862 Source: Central Bank of Sri Lanka Payments Bulletin 2004: 237. and to perform a range of nonfinancial transactions such as balance inquiries and requests for checkbooks. These transactions were largely made by cus- tomers using ATMs installed by their own bank. Only a limited number of ATM transactions were made by customers using ATMs other than those installed by their bank, and that required cross-institution clearing. Until banks increase the sharing of infrastructure across branches, ATM usage will remain limited. Fortunately, this is fast changing. As of June 2004, 92 percent (720) of ATMs in Sri Lanka were open-access ATMs, that could be used by the holder of an ATM card regardless of the issuing bank. The number of limited-access ATMs, which could be used by a holder of a card issued by the bank that owned the ATM, was 63. Out of 783 ATMs, domestic banks owned 535; the two state banks owned 196. At end 2003, the number of ATMs per 1 million inhabitants was 37. Ideally, this cross-sharing of ATM infrastructure should be voluntary and led by banks themselves as they see the business logic of sharing infra- structure and competing on service. However, in some instances more than moral suasion by the regulator has been used to encourage cross-bank shar- ing of infrastructure. For example, on August 5, 2002, the State Bank of Pakistan (SBP) asked all banks that were not part of the (then two) ATM switch networks man- aged by two private banks to join or come to agreement on their use. In a cir- cular the SBP argued that to improve the interconnecting of the ATM payment system, and to introduce enabling e-commerce technologies in the financial sectors, a single platform for the 250 ATMs then in operation was essential. Today, the link--called "1 link"--is the largest network of ATMs across Pakistan, with 392 ATMs in 26 cities, all of which are member banks,6 while the Muslim Commercial Bank network has 130 ATMs.7 In India, the central bank is taking an active interest in the interbranch connectivity of bank branches. With a view to providing for transfer of funds electronically across a large number of bank branches in the coun- try as a forerunner to the Nationwide Funds Transfer System (NEFT), the Special Electronic Funds Transfer System (SEFT) was introduced in April 2003. SEFT covers 2,312 branches of 29 banks situated in 127 cities across India.8 36 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Overall, new technology in the region will increase the speed of trans- actions through card-based remittance products used at ATM machines. ATMs represent a way for competitors to use technology to create wider distribution networks. However, multiplying ATM-networked branches in rural communities in the region will remain a tremendous challenge. In addition to the recip- ients' unfamiliarity with using them, banks face the challenge of physically distributing them and the cost of keeping them in operation. Poor road infrastructure, security challenges, and weak local economies unable to support them limit their expansion. Money Service Businesses Money transfer businesses play a leading role in providing remittance ser- vices to South Asian migrants. In the context of this study, money service businesses (MSBs) are financial service agents that accept cash, checks, other monetary instruments or forms of stored value in one location and pay a corresponding sum in cash or another form of value to a beneficiary in another location by means of a communication, message, or transfer, or through a clearing network to which the MSB belongs. Transactions per- formed by such services may involve one or more intermediary and final payment by a third party. Money services or value transfer services may be provided formally by persons (natural or legal) through the regulated financial system, for exam- ple through bank accounts, or informally through nonbank financial insti- tutions or other entities operating outside the regulated system. Primary among the formal MSBs are Western Union and MoneyGram, companies licensed and authorized to engage in foreign exchange activi- ties that do not involve acceptance of deposits (that is, money that can be withdrawn by check writing). Western Union allows customers to send money using credit and debit cards through the company's Web sites, through offices and agent locations, over the telephone, or through physi- cal mail, using a money order. With a 150-year history, Western Union is the global leader in money transfer services. It has more than 150,000 agent locations in 190 countries. Although estimates vary, it is commonly agreed that, after banks, Western Union handles the largest share of remittances from the United States to South Asia. The company came to India in 1993­4. Before obtaining its own license from the Reserve Bank of India in 1998, it operated through repre- sentatives. In 2001, Western Union handled US$85 million in remittances to India. Western Union has 13 principal agents and 14,500 subagent locations in India, of which 4,200 are a result of a partnership with the Indian post office, FORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 37 India Post. Principal agents can appoint subagents such as travel agents. India Post is an important partner for Western Union because of its geo- graphical outreach and experience in postal orders. From India Post's point of view, a new product such as money transfers helps it diversify and expand its business by collaborating with a reputable global firm. The venture with Western Union is part of India Post's Financial Services Group, which also distributes mutual funds and bonds, provides a money-changing facility, and issues prepaid credit cards for use by students and tourists. In addition to India Post, Western Union's partners include banks such as Andhra Bank and Punjab National Bank, travel agents such as Kuoni Travels, and financial firms such as Wall Street Finance Limited. These agents, whose participation with Western Union dates from the mid-1990s, are all authorized by the Reserve Bank of India to engage in the money transfer business, a small component of their business portfolio. Other MSBs are taking advantage of the Internet. These latest entrants in the remittance market typically make arrangements with money trans- fer companies or banks to carry out remittances. They generally remit smaller amounts than other players. C-Sam technology enables people to use cell phones and Personal Digital Assistants to transfer money, pay bills, payments, buy prepaid phone cards, and conduct other stored cash value transactions. Using C-Sam software, banks can offer their customers and mobile subscribers the ability to instantly remit money in real time using their mobile phones. Remit2India is another example of an Internet-based provider. The firm, part of TimesofMoney.com (owned by the Times of India), came into exis- tence in 1999, initially as a joint venture with Citigroup. No cash or credit cards are used in its business. The company provides solely a portal. TimesofMoney.com developed the technology behind the service, while UTI Bank performs the actual money transfer. Because a bank is involved in the transaction and the Times of India has instant name recognition, Remit2India inspires customers' confidence. The maximum one-time send- ing limit is $7,500. Remit2India's customers are not restricted by the $2,500 limit of the Reserve Bank of India because the company performs no cash transactions. Post Office Networks The post office, which offers access to an extended branch network of rural post offices in all four countries, is a potentially exciting opportunity for facilitating remittances. Many post offices are conveniently located and inte- grated into the social activities of the community. In many villages, post- masters enjoy a high level of trust. Importantly, they have an extensive rural reach. 38 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Box 4.1 India Post and Remittances The Indian post office has a long tradition of providing limited financial services to the general population through the Post Office Savings Bank (POSB), which was established in 1882. Presently, India Post is offering the following remittance-related businesses: · Money Orders: With a linkup to the post office's VSAT* connection, money transfers can now take place overnight in certain locations rather than the many days required under older systems that relied on hand-carried mail. Total money order traffic in 1999­2000 was 113,700,000 transactions with a value of around US$1 billion. · International Money Orders: An Inward International Money Order Service is also available with 27 countries whereby funds can be remitted to India. A two-way money order service is available with Nepal and Bhutan. During 1999­2000, a total of 49,363 foreign money orders brought foreign exchange into India. · Western Union: In late January 2001, the post office signed a memoran- dum of understanding with Western Union to support the transfer of funds into India using the post office network, permitting senders in 185 countries to remit money to India through electronic remittances of funds. Western Union, which has a five-year contract with India Post, initially began providing the service in 75 post offices and will cover 3,000­5,000 post offices within three years. The Department of Posts earns US$15 per transaction, which could become an important source of revenue. Post office services are provided by the 600,000 staff members working in the post office--supplemented by about 400,000 registered agents, who operate on a commission basis. With 154,149 post offices and 554 sorting offices, India Post has the most extensive postal retail network in the world. *Very small aperture terminals are small Earth stations that are used for the reliable transmission of data, video, or voice via satellite. India has a large post office network with 154,000 outlets that are required to focus on deposit mobilization and money transfers.9 Established in 1881, it holds the savings accounts of 130 million people. Sri Lanka has 4,600 post office branches (625 main post offices, 3,423 sub-post offices, and 632 agency post offices) that provide unequaled access to the country's rural regions (see box 4.1). FORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 39 Increasingly, post offices are exploring the possibility of increasing their remittance business. Electronic money orders and e-money orders are on the increase. India Post has commenced reforms aimed at transforming it into a front-end outlet for financial services by 2010, including remittances. The potential to innovate is high. However, this remittance channel remains largely spurned by clients who have experienced its inefficiency and poor service. Post offices are plagued by operational losses, absence of management information systems, limited access to funds for invest- ing in technology, and negative publicity. Stories of insufficient cash on hand and other delays are common in the region. Yet where they do oper- ate efficiently, postal money orders are a preferred option, especially for domestic remittances. Conclusion Overall, South Asia has an impressive density of rural formal financial insti- tutions. The combined state and private banking infrastructure, including the postal networks, provides a good basic infrastructure for remittances flows. However, the formal infrastructure's potential is not being maximized. The region remains largely cash dominated. The proportion of the rural poor in particular who have access to formal banking services, let alone formal payments systems, remains very low. A recent World Bank Access to Finance survey of 6,000 households in India, for example, found that 48 percent of the households surveyed reported receiving their income in the form of cash and 93 percent reported cash at home as the main mode of financing household expenses.10 In the case of payment services such as remittances, cash dominated: 82 percent of the households reported cash as the main mode of remittance, as opposed to 15 percent who reported checks or drafts and 1.7 percent who reported postal money orders as the mode of remittance. Several different factors account for the lower-than-possible penetration rate of remittance services across the region, including: · Rural outreach: Although progress has been made in the region on rural access to financial services, poorer households in rural areas still have little access to formal financial services. Rural banks primarily serve the needs of richer rural borrowers: some 66 percent of large farmers have a deposit account; 44 percent have access to credit (World Bank 2004b: 3). Thus, although the physical branch buildings have an extensive reach, this has not necessarily translated into financial service access for all. The lower-income earners--many of whom have migrant relatives--do not have access to financial services. 40 MIGRANT LABOR REMITTANCES IN SOUTH ASIA In India, for example, the transaction costs of dealing with formal banks are high. In part, high transaction costs stem from distance from the recipients' home to the nearest financial institution. The mean dis- tance to the nearest financial institution for a typical rural household ranges from 2 kilometers (post office branches) to 5 kilometers (com- mercial banks or cooperative banks); the median time taken to travel to the nearest commercial bank, cooperative, or rural regional bank is 30 min- utes (available post offices are nearer).11 · Quality of financial services in rural areas: The limited access to financial services extends to the quality and choice of products and services pro- vided by rural bank branches, which appear to be generally lower than similar products and services in urban areas. On average, a rural bank branch serves almost three times the number of people served by a non- rural branch. · Regional disparities: Large regional differences exist in the distribution of financial services. These differences exist both in terms of the volume of transactions and in branch density, with clients in economically weaker regions having a disproportionately lower level of financial services. For example, in India the spread of branches appears to be closely associ- ated with regional shares in population: the eastern and central regions have larger shares in population and therefore, despite their low share in income, occupy the second and third positions in terms of share in branches. However, the presence of branches alone does not ensure access to finance; as expected, income is a determinant to financial access. Regional differences in the volume of financial services are largely explained by income differences. India's lesser-developed and low-income eastern, central, and northeastern regions account for 54 percent of the population and 40.5 percent of total branches, but only 20 percent of out- standing credit and 29 percent of deposits.12 · Payments system technology: Banks in the region are investing heavily in payments system technology. The use of credit cards, debit cards, and the installation of ATMs is on the increase. Hitherto, however, most banks have been investing in proprietary IT platforms for their payments sys- tems. There is very limited outsourcing or sharing on domestic branch net- works. Improved payment systems platforms could substantially reduce transaction costs by expediting check clearance, reducing exchange losses, and improving disclosure, especially in rural areas. Consequently, the formal financial sector still incurs higher transaction costs than the informal sector. The vast bank branch network is still largely less than optimal about handling remittances for the average migrant in the Persian Gulf region. Little wonder that the average cost of transferring remittances through banks using a bank draft is US$1.34, or 3.97 percent more expensive than other nonbank options, such as exchange companies, FORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 41 and US$2.72, or 6.8 percent more expensive than informal remittance chan- nels. If a remitter were to use a telex transfer, using a bank would be US$2.50, or 9.3 percent more expensive than exchange companies and 12.5­27.22 percent more expensive than informal remittance channels such as hawala (discussed in the next chapter).13 Notes 1. See World Bank (2004b). 2. See People's Bank (2005). 3. See Bank of Ceylon (2005). 4. See Central Bank of Sri Lanka, Payments Bulletin (2004: 237). 5. To encourage people to maintain a bank account, one bank in Sri Lanka has been operating a lottery scheme for account holders. To participate, account holders need to maintain a minimum balance of US$100. The prizes include cars, motorcycles, and gold coins. 6. See ABN AMRO (2005). 7. See MCB (2005). 8. See Reserve Bank of India 2004b Annual Report: 191. 9. See World Bank (2004b). 10. Ibid. 14. 11. Ibid. 8. 12. Ibid. 5. 13. It is important to note that this cost comparison is made only for transfer- ring US$100. If the amount of the transfer increases, the cost will not increase for banks and exchange companies. 5 Informal Financial Sector Infrastructure for Remittances In Bangladesh, India, Pakistan, and Sri Lanka, informal remittance systems are commonplace. Originally developed to facilitate debt settlement between distant regions at a time when conventional banking instruments were either absent or weak, their speed, low cost, convenience, accessibility, and anonymity led to their use for both legal and illegal remittance purposes. This chapter provides an overview of the informal infrastructure for remittances. The next section provides a broad description of the common types of informal remittance systems, with a special emphasis on the hawala system, which is the most widespread in the region. The following section summarizes the historical context within which informal remittance sys- tems have developed in the region. It highlights the role of trade practices, foreign exchange controls, weak financial systems, and regional conflicts in the sustained use of informal systems. The factors contributing to the sustained use of the hawala system are presented in the next section, fol- lowed by the potential risks and abuse of such systems, and finally by the chapter's conclusions. Types of Informal Remittance Systems Avariety of simple as well as quite sophisticated informal methods are used by South Asian expatriates. The most salient ones are couriers, in-kind remit- tances, and hawala. Of these three, couriers are the most inefficient and risky, as they require the physical transportation of cash across borders. The risk of the cash being lost or stolen is considerable, so hawala is by far the most trusted and efficient method. Each of these systems is briefly discussed below. Courier Services Courier services are the simplest and oldest way of moving value. In South Asia, as in many Persian Gulf countries where immigrants work, cash is commonly used even for large transactions. People take their cash with them or seek the assistance of trusted friends or relatives in moving it. There 42 INFORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 43 is evidence that physical transfer of cash by friends and relatives into the region is high. This is not unique to the region. Although there can be no precise estimates of the total volume carried, the use of courier services is booming in South Asia and elsewhere, including the Americas, Asia, Europe, and the Middle East. Transfers conducted in this way are observed in Egypt, Pakistan, the Philippines, and Sudan, for example, where some estimate that they may amount to double or triple the official remittance figure.1 According to one estimate, as much as 27 percent of all remittances to Pakistan may go by hand.2 In-Kind Remittances In-kind remittances take place through the provision of goods or services in one country, while the payment is made in another country. On a small scale, someone can offer to look after a sick parent or relative overseas while receiving payment in a country with strong currency. The tourism business, for instance, lends itself readily to this sort of practice. A travel agent may send groups to India and collect full payment from them in Europe. In India, an associate will cover all local expenses of the group and receive payment in an account maintained overseas. Some migrants choose to invest in durable goods such as refrigerators, stoves, televisions, VCRs, and so on, or in valuable commodities such as gold or precious stones, which they either send to their family for sale or haul back to the home countries upon the end of their contract. Other migrants pave the way for their return by building a house. A local com- pany may provide the construction work for the migrant and receive payment in an overseas account. The South Asian real estate market has provided many recent opportu- nities for such transactions. And where the migrants wish to buy or sell property at home without disclosing the true price to the tax authorities, balances between the real and declared price are settled through middle- men and payments outside the home country. Hawala There are many terms used to describe informal remittance systems such as hawala, including "alternative remittance systems," "underground bank- ing," "ethnic banking," and "informal value transfer systems." In other parts of the world, terms used include fei-ch'ien (China), hundi (Pakistan, Bangladesh), hawala (India and the Middle East), padala (the Philippines), hui kuan (Hong Kong), phei kwan (Thailand), kyeyo money (Uganda), and mali a mbeleko (Zambia). The operational mechanisms of the various systems are fundamentally the same.At its simplest, the typical hawala transaction involves the remitter, 44 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Figure 5.1 The Basic Informal Remittance Transaction fei-ch'ien (China), hundi (Pakistan, Bangladesh), hawala (Middle East), padala (Philippines), hui kuan (Hong Kong), phei kwan (Thailand), kyeyo money (Uganda), mali a mbeleko (Zambia) Worker Cash Telephone/ Middleman fax Shopkeeper Cash Reference Families Source: Authors. the recipient, and two intermediaries (hawaladars). For a basic transaction to takeplacethereneedstobearemittingparty,tworemittanceserviceproviders and a recipient (see figure 5.1). When the remitting party, for example a Sri Lankan migrant worker in Italy, wants to send money to Colombo, he makes payment in euros or another convertible currency to a remittance agent or middlemaninRome.Theserviceprovidercontactsapartnerserviceprovider counterparty (who might be a shopkeeper in Colombo), who arranges pay- ment in local (or other) currency to the remitter's family or other beneficiary on the production of a pre-agreed reference. As a form of identification for the transaction, the agent in the remitting country provides the remitter a code or reference that must be passed on to his designated beneficiary for presentation to the agent in the recipient country. Once the funds have been paid to the recipient, the agent in the remit- ting country is indebted to the agent in the recipient country. The princi- pals to the initial transaction do not play any role in subsequent clearing and balancing of this position. The agents can settle their positions in var- ious ways, including simple transactions going in the opposite direction, cash deliveries, and settlement by checks into the relevant accounts. Their INFORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 45 positions can also be transferred to other intermediaries. These other enti- ties can assume and consolidate the initial positions and settle at wholesale or multilateral levels, also by various means.3 Payment and delivery are made in the currency of the corresponding countries, unless a request is made for delivery in U.S. dollars, euros, or another currency. Payment instructions to hawaladars are sent by fax, e-mail, or telephone at the end of each day for delivery in about 24 hours. As a form of identification, a code such as "8 cents" may be sent along with the pay- ment instructions (in Bangladesh and Sri Lanka) or the serial number of a rupee note in the hands of the intended recipient (in India). In Pakistan, a passport or ID number is increasingly used. This is the remittance end of a hawala deal, which relates to the dealings between retail clients (remitters) and the intermediaries. More complex is the settlement end, which relates to the ways in which hawaladars interact with each other and with third parties--a process most associated with the illicit aspects of the business. In order to be able to make quick and regular payments upon request, each hawaladar needs to maintain a cash pool proportionate to anticipated demand. Because the cash pools in various locations are never symmetrical, reverse payments cannot balance accounts among operators. It is instructive to look into the sources of cash pools in labor-importing (A) and labor- exporting (B) countries. Hawala operations thrive not only in regions with less developed or unregulated formal banking infrastructures, but in all countries with significant South Asian communities, including North America and Europe. The following are the main sources for cash pool A (labor-importing countries): · Remittances of expatriates to their families · Payments for imports · Payments for services received overseas · Investment funds in excess of permitted limits · Traders who over-invoice exports or engage in bogus exports · Income not declared in the country of residence (tax evasion money) · Proceeds of misconduct or criminal enterprises of all types · Contributions to militant groups and ethnic causes Without further systematic study, it is not possible to determine the rel- ative importance of these categories. The same applies to cash pool B. Anecdotal evidence, however, suggests that the largest number of customers would fall under the categories of remitters to the home country and traders. All but the first three listed sources require secret transfers from labor- importing countries. Although hawala is certainly not the only means to make such transfers, it is a popular and efficient way. 46 MIGRANT LABOR REMITTANCES IN SOUTH ASIA The sources for the South Asian hawala pools are slightly different. Cash pool B (labor-exporting countries) is fed by: · Funds to cover students' tuition overseas · Funds to cover medical expenses overseas · Tourists' money beyond amounts allowed under currency controls · Capital flight · Payment for imports · Tax evasion · Misinvoicing of trade · Bribes of politicians and government officials · Money from criminal enterprises The first few categories are affected by conditions specific to South Asian and other labor-exporting countries, such as currency and capital controls. Citizens and residents are allowed to take out of the country a limited amount per year, generating a demand for a discreet channel for currency exports. As restrictions are eased, the demand for that channel declines sharply. Many of the issues that used to cause capital flight have also been addressed through recent economic and policy reforms. Historical Background of Informal Remittance Systems Informal remittance systems have a long history in South Asia. In the 1950s and 1960s, the main method of payment in the Indian subcontinent was through hundi, chiti, or hawala, all of which involve a draft drawn on a trad- ing associate.4 Import credit from the "money bazaars" generally took the form of loans against promisory notes in the form of hawalas or hundis. These were simple drafts drawn on correspondent traders in India, Iran, and Pakistan by traders and foreign exchange dealers from neighboring coun- tries, including Afghanistan. Trade Hundi, a form of bill of exchange or promissory note, was in use in India before the advent of modern banking. In India, no evidence has been found of the existence of paper money in early times, nor is there any reference to negotiable instruments in Hindu or Islamic texts. However, bills of exchange have been popular from very remote times. Hundi developed in India with a strong body of rules, usage, and customs, which the legislature and courts upheld. In the beginning, hundis were probably issued by brokers for the purpose of debt collection. They took diverse forms--as bills of exchange or other promissory notes--while being subject to local usage. These indige- nous instruments acquired such a high level of credibility that to dishonor a hundi was a rare event. They were freely circulated among Indian bankers INFORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 47 to finance internal trade and were gradually integrated into the activities of the emerging modern banking system. In an effort to reconcile and assim- ilate native and European law, the legislature did not abolish the numer- ous local customs and usages relating to hundis. Because the Indian commercial community was accustomed to their use, hundis were given the weight of formal bills of exchange in some Indian courts. Today, the speed and efficiency of hawala continues to attract traders in South Asia faced with foreign exchange controls or administrative inconveniences. Foreign Exchange Controls The interface between traders and hawala also owes much to the tradition- ally complex framework regulating the import-export business associated with the region. Until very recently, when countries liberalized their cur- rent accounts, the requirements regarding international payments, rebates, refunds, and other financial payments were perceived as too burdensome, and informal channels flourished. Further, tax evasion has also been noto- riously widespread. Multiple studies address the problem of "black money" and "unaccounted money." A good part of undeclared income is sent out of the country or otherwise laundered. Finally, in some parts of South Asia individuals and businesses pay "speed money" to get things done or to avoid taxes. Because companies do not wish to enter these payments in their official books, they purchase cash at a discount from hawaladars. An important implication of this interaction is that even when clean money is sent from labor-importing countries, dirty money may be flowing in the opposite direction. Conflict Areas of conflict are also more likely to be linked to informal methods of money transfers. The northern parts of Sri Lanka, Kashmir, and the north- western provinces of Pakistan have limited formal financial services. Migrants from those regions resort to the hawala system to make remit- tances. After the partition of India and Pakistan in 1947, virtually no pay- ments connected with trade between India and Pakistan were transacted through banks. Because it was long illegal to export currency from India and Pakistan, a considerable differential appeared between official and hawala exchange rates, which increased the popularity of the informal system. Migrant Incentives for Using Informal Remittance Systems Migrants use the hawala system for a variety of reasons. The most important are trust, cost, speed, convenience and accessibility, and for some, the anonymity that hawala can provide users and recipients alike. 48 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Trust Highly trust-based, informal remittance systems have a long history of being a reliable way of transferring funds. Few complaints from retail cus- tomers have been reported over decades of hawala operations in South Asia. Disputes and errors are settled among hawaladars, who assume the risk of a counterpart's default or law-enforcement actions. Trust and the cultural importance of one's reputation and good name perform an effective self- regulatory function. The element of trust is a defining characteristic of most informal remittance systems. Cost Although the costs of remittances through the formal financial sector are falling in many areas, as discussed in the previous chapter, informal chan- nels continue to hold a competitive advantage. Literature and field surveys indicate that households in recipient countries receive more local currency when informal channels are used. If 1,000 U.A.E. dirhams are sent to India and Pakistan using various official and unofficial methods, for example, the recipient household will collect different amounts (box 5.1). Box 5.1 Remittance Options and Comparative Exchange Rates: An Example Pakistan rupees (PRs) for U.A.E. dirhams (Dh) on December 8, 2003, in Dubai. Amount of PRs for 1,000 Dh Means of buying PRs 11,905 At draft/bank rate 12,391 At money exchange rate 12,660 At hawala rate Indian rupees (Rs) for U.S. dollars (US$) on December 9, 2003. Amount of Rs for US$1,000 Means of buying Rs 44,128 At draft or bank rate 45,510 At money exchange rate 46,259 At hawala rate Source: Author interviews. INFORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 49 Whereas a Pakistani remitter's family would receive 11,905 rupees for the thousand UAE dirhams through the bank route or via a bank draft (before postage, encashment, and other fees), it could collect 12,391 rupees by cashing its remittance at a money exchange house. Through hawala, the remitter's family would do even better, receiving 12,660 rupees--the next day. Similar rate advantages exist for Indian rupees sent through hawala. Speed Speed is an important consideration. With courier and hawala deliveries, funds are available immediately, whereas some formal alternatives may take many days (table 5.1). A case study in Bangladesh showed that on aver- age 15 days were required to clear a draft made on a bank. The same draft was liquidated instantly if made by an exchange house. Table 5.1 Time Spent Cashing Remittances through Various Channels Time spent cashing bank draft/ telegraphic transfer order Mean Bank Category of Current District Noakhali 16 draft household migrant Chittagong 13 household Comilla 7 Sylhet 12 Return District Noakhali 16 migrant Chittagong 15 emittancer household Comilla 3 Postal Category of Current District Comilla 15 getting order household migrant of Hundi Category of Current District Chittagong 3 household migrant Sylhet n.a. Means household Relatives/ Category of Current District Comilla 5 friends household migrant Sylhet -- household Sources: Murshid, Kazi, and Meherun (2001); Author interviews. 50 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Convenience and Accessibility The convenience of door-to-door service is appealing to all customers as well, especially in regions where transportation is poor, and to female ben- eficiaries in remote and traditional regions, where they may not be allowed to leave the house. In some instances, as field surveys indicate, hawaladars visit the workplaces of expatriates and collect money for remittances. The informality of hawala makes it attractive to many South Asians who find banks and other financial institutions uncomfortable to deal with. The lack of bureaucracy renders informal services accessible to illiterate cus- tomers, many of whom live in rural areas where alternatives are not avail- able in any case. Further, migrants without credit cards and those paid at the end of their contract rather than periodically are more likely than others to turn to informal remittance systems. The latter, especially those in the Persian Gulf countries, are more likely to carry cash personally, believing that they can save thereby on costs and exchange rates, or to purchase goods for resale at a higher price at home. Unfortunately, some migrants are robbed of their cash and goods in the process. Other times, legal migrants avoid the banking systems because of language barriers, misperceptions about the cost of formal banking transactions, or social conditioning regarding formal financial institutions. Anonymity Informal transfers to the region are said to be significant, particularly from countries where there are large groups of illegal migrants. It is particularly attractive for undocumented migrants, who are unable to open bank accounts or transact with formal institutions. More significantly, the anonymity of informal remittance systems renders them highly susceptible to processing proceeds of criminal activities such as drug trafficking, prostitution, corruption, and tax evasion. Generally, clients need not produce identification documents or explain the source of their funds before using an informal system to remit funds to a counter- part in another jurisdiction. Nor is the recipient required to produce iden- tity documents other than a pre-agreed code. Once the transaction and settlement are completed, records are quickly destroyed. Consequently, hawala transactions sometimes leave no audit trail for law enforcement agencies investigating predicate offenses of money laundering, tax eva- sion, corruption, or other illicit activity. The anonymity of informal systems also facilitates the process of trans- ferring illegally obtained money or investments through various jurisdic- tions and institutions to conceal its true source, thereby preventing law INFORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 51 enforcement from uncovering or confiscating the proceeds of crime or using those proceeds as evidence in a criminal prosecution. As is occasionally the case in the formal sector, transactions in the informal sector do not always include accurate documentation of the beneficiary of either the actual crim- inal proceeds or other property that might be subject to confiscation. Risks and Potential Abuse of Informal Remittance Systems Unfortunately, the hawala system's success--speedy transactions with min- imal or no documentation--has also been its vulnerability to abuse. The anonymity that is possible with transactions of this kind has long raised con- cern in the law enforcement community. In some countries such as India, the concern has resulted in prohibition.5 India is not alone. Concerned about the evasion of currency controls, countries in the region have long been wor- ried about informal remittance systems, and the September 11, 2001, terror- ist attacks in New York and Washington have served to heighten this concern. The interface of legal and illegal transactions is especially complicated in the settlement process. Each time a hawaladar sends payment instructions to a counterpart, he creates an informal debt or loan. Where transactions are reciprocal, the net difference needs to be settled. The closer the rela- tionships among hawaladars, the easier the settlement process. Kinship, family, and ethnic ties have made hawala work smoothly. In straightforward and small operations, a courier brings the cash from one party to another. This is a preferred method of moneychangers who need currency in their possession for their regular trades. Other common methods involve interfaces with formal institutions, for example, by sending monetary instruments (such as checks and money orders) or a wire transfer to a designated account maintained in a financial center such as Dubai, Hong Kong, New York, or Singapore. With payments made to multiple parties around the world, the hawala settlement process resembles the way formal banks go about their business. The main difference with hawala, however, is that hawaladars generally have been unrestricted by rules on how and with whom they conduct this process. Banks and money exchange companies must follow the laws of each coun- try in which they operate and deal with duly authorized people and insti- tutions. On the other hand, hawala is either outlawed or regulated in a nonpragmatic fashion in other developed economies. In effect, it continues to operate largely outside formal regulatory frameworks. Consequently, "informal economies" or black markets play a large role in the settlement process. Some interviewees from South Asia have gone so far as to argue that this is precisely what gives hawala a competitive edge over moneychangers and other financial channels. That claim may or may not be entirely accurate, but no one seriously questions the interface of 52 MIGRANT LABOR REMITTANCES IN SOUTH ASIA hawala with unauthorized dealers, "black money," or unaccounted fund transfers. In South Asian countries, it is not uncommon to hear the argument that a pivotal factor that enables hawaladars to offer attractive rates to retail cus- tomers is their nexus with illegal and criminal actors who are prepared to pay a relatively high premium. Murshid, Kazi, and Meherun (2001) observed that the use of hundi in Bangladesh is significantly higher in Chittagong than in Noakhali, Comilla, and Sylhet. Not only is international migration a common demographic characteristic in Chittagong, but Chittagong is also a port city infamous for illicit trade. A recent study in Bangladesh by Siddiqui and Abrar (2003), for example, argues that hawala operations are based on the following factors: · Demand for foreign exchange from racketeers who wish to finance smug- gling of various items, including gold · Demand from importers for foreign exchange from other sources in order to derive tax savings by underinvoicing imports · An alliance to promote the personal prosperity of officials of financial institutions, businesses, and hundi operators · Payment of financing charges of labor recruiters · The gap between the official and unofficial exchange rates · Quality and speed of service, and ability to reach clients both in desti- nation and source countries The different ways in which the hawala system is known to be abused in the region include the following, as described in El-Qorchi, Maimbo, and Wilson (2003: 12­13). · Circumventing capital and exchange controls: Countries facing shortages of foreign exchange reserves often have imposed capital controls and cre- ated tax barriers for imports. Individuals and businesses seek alterna- tive means to make international fund transfers through the reverse hawala route, from countries under exchange controls to other, usually more developed, economies without any documentary requirements. The incentive to use informal mechanisms to externalize funds is even higher if the capital controls exist in a country where political and eco- nomic uncertainties create an exchange-rate risk.6 · Customs, excise, and income tax evasion: Importers sometimes resort to making partial payments to an overseas exporter through informal fund transfers, particularly when customs, excise, and income taxes are high. To avoid paying customs duties, importers ask the overseas exporter to underinvoice the goods. The difference between the actual price and the invoiced amount is then remitted to the overseas exporter through the hawala system. Similarly, when a government grants subsidies based on INFORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 53 export receipts to encourage exports, exporters may resort to overin- voicing to maximize their gains. · Smuggling. Recent literature attributes part of the growth of the hawala network to gold trading and smuggling operations in South Asia in the 1960s and 1970s. To avoid restrictions on gold imports, traders and smug- glers used boats to ship gold from places such as the Persian Gulf region to South Asia. To remit funds back to their countries of origin or in order to purchase more gold, traders and smugglers (importers) found a solu- tion in the growing population of South Asian nationals working in the Persian Gulf countries. To settle their liabilities, hawaladars in Dubai, for instance, would finance gold exports to their counterparts and clients in South Asia. The remitting workers received better rates because hawaladars charged higher fees to smugglers, who made substantial profits from the gold trade. Thus the smuggling activities benefited from, and enhanced, the existing systems of fund transfers used by expatriates in the Middle East, Southeast Asia, the United Kingdom, and even in North America. This network, it is argued, formed the base for the large-scale hawala operations that exist to this day. · Money laundering. Both the formal banking sector and informal fund transfer systems are vulnerable to abuse. The number and variety of methods used to launder the proceeds of criminal and illegal activities and finance terrorist acts grow ever more complex. The methods are diverse and can employ both banking and nonbanking channels, including bureaux de change, check cashing services, insurers, brokers, and nonfi- nancial traders. · Terrorist financing. The anonymous transfer of funds through informal fund transfer systems has occasioned concern about their potential use as a conduit for terrorist funds. Because there is no requirement for iden- tification documents or for an indication of the source of funds, an infor- mal dealer can initiate or facilitate many transfers through his network in different jurisdictions, thereby concealing the true origin of funds. The recipient can use the funds to commit a terrorist act. Once the transac- tion is completed, all customer identification documents, codes, or ref- erences are probably destroyed, except, perhaps, those required for settlement purposes. It has also been reported that hawala has been used to finance trafficking in gold, precious stones, human organs, and even human beings, as well as to commit credit card fraud and violate embargoes and sanctions. However, some formal institutions--travel agencies, merchants, non- governmental organizations, money service businesses, banks, correspon- dent accounts, brokerages, import/export businesses, corner stores, and so on--also allegedly play intermediary roles in the settlement processes that facilitate illicit deals. 54 MIGRANT LABOR REMITTANCES IN SOUTH ASIA Conclusion Informal transfer systems remain prevalent in the South Asia region because of their speed, convenience, low cost, and accessibility. Steeped in long- held historical trade practices they continue to exist today, facilitating a host of both legal and illegal trade activities, notably remittances, trade, and money laundering. Because of the lack of transparency associated with informal flows, efforts are being made to encourage migrants to choose formal over informal remit- tance channels such as courier services, in-kind remittances, and hawala. Addressing the risks of financial abuse that the potential anonymity of informal remittance systems provides requires a two-pronged approach. A recent 2003 World Bank­IMF study7 advised that in the majority of countries where informal systems exist alongside a functioning conventional banking sector, it is recommended that hawala dealers be registered. In these sys- tems, additional efforts should be made to improve transparency by bring- ing informal dealers closer to the formal financial sector without attempting to alter their specific nature. Simultaneously, the regulatory response must address weaknesses that may exist in the formal sector. Presently, formal and informal financial systems benefit from their mutual deficiencies, with each tending to expand when the condition of the other is impaired. High transaction costs, delays in transferring remittances, exchange controls, and overly bureaucratic policies and procedures for simple money transfers are major incentives for the existence of the informal financial system. To face the challenge, the formal sector should tackle its deficiencies and enhance its competitiveness. The regulators should consider the adverse effects on the formal industry of excessive administrative requirements. Notes 1. See Abella (1989) and Alburo and Abella (1992). 2. See Kazi (1989). 3. For more details on informal remittance systems and how they work, includ- ing their economic and legal implications, see Passas (2000), El-Qorchi, Maimbo, and Wilson (2003); and Maimbo (2003). 4. A chiti is a piece of paper (for example, half a banknote) that a beneficiary presents to receive funds. A Hindi word, it was introduced into China by the English and denotes (with "chop shop") the Chinese system of alternative remittance (Passas 1999a). The term hawala has acquired a wide, negative connota- tion in India. Associated with illegal payments to politicians, it is used by companies for a variety of advances, payments, and transfers. The famous Jain Hawala scandal of the early 1990s, which involved bribes to politicians, had repercussions throughout the decade. The case is recounted in Kapoor (1998). INFORMAL FINANCIAL SECTOR INFRASTRUCTURE FOR REMITTANCES 55 5. Under both the Foreign Exchange Regulation Act (FERA, 1973) and its succes- sor, the Foreign Exchange Management Act (FEMA, 2000), hawala-type transactions have been explicitly prohibited. The number of institutions (notably, "authorized persons" such as banks) permitted to deal in foreign exchange has been closely defined, and the kinds of permitted transactions (travel, medical treatment, acquisi- tion of foreign assets, and so on) have been set forth in regulations that have been frequently revised. The recent FEMA wording specifically addresses hawala-type transactions by prohibiting Indian residents from entering "into any financial transaction in India as consideration for or in association with acquisition or creation or transfer of a right to acquire, any asset outside India by any person." Similarly, one of the mandates of the Directorate of Enforcement has been to prevent remittances of Indians abroad other than through normal banking channels (i.e., through compensatory payments). 6. Hawaladars may be able to avoid capital controls in the short term without any difficulties in settling their external accounts. However, if these controls persist, they may experience difficulties in settling their accounts, especially if the volume of fund transfers requires the use of the formal banking sector. Different settlement mechanisms (discussed later) may have to be devised, including the smuggling of physical cash. 7. See El-Qorchi, Maimbo, and Wilson (2003). 6 Conclusions and Recommendations With globalization, international migration will continue to drive increas- ing levels of remittances to developing countries. The South Asia region is no exception. A recent World Bank study concluded that "while long term and settlement migration are still predominant in most developed countries, migrant flows are now more diverse and complex, with migrants moving back and forth more readily and rapidly. Temporary movement, in particu- lar by high skilled workers has seen the largest growth in the past decade."1 The study identified five forces governing international migration patterns: · Wage and opportunity gaps between rich and poor countries · Regional conflicts and political instability in developing countries · The relative share of young adults in the populations of sending and receiving countries · Numbers of migrant people residing in receiving countries2 · Reductions in the cost and inconvenience of travel All these factors apply to the South Asia region. The income and oppor- tunity gaps between the subcontinent and the Persian Gulf region and Europe could not be more disparate. Regional conflicts have led to the migration of many people. The rising share of young people in South Asia, as in other parts of the developing world, is complementary to the shrink- ing share of young adults in the developed countries. The high number of foreign-born South Asians in developed countries makes it easier for others to join them. And the declining cost of international travel enables young people to embark on migration. These trends have a significant impact on remittance flows. Governments in the region have rightly expressed a strong interest in macro- and microim- pacts of remittance flows. Enhancing the Development Impact of Remittances The South Asia region has much to gain from the increased flow of migrants--employment and higher wages for the migrants and the remit- tances that they send to their families. This study has shown that international 56 CONCLUSIONS AND RECOMMENDATIONS 57 migration and remittances are important to poverty and economic devel- opment in South Asia. International remittances (official and unofficial) reduce poverty in South Asia, and international remittances do lead to asset accumulation--mainly land accumulation--in rural areas. Although the sending countries need to balance the loss of skilled workers (who may have been educated at public expense) and the benefits they accrue, there is an argument to be made for accepting the inevitability of present migration trends and enhancing the development impact of the remittances that the recipients are already receiving. Chapter 2 of this study showed that there are two key determinants to enhancing the development impact of remittances--the education level of the migrants and the investment opportunities available to the migrants. Using data from rural Pakistan, this study concludes that it seems wrong to claim that international migrants do not invest. Migrants do invest; they actually exhibit a higher propensity to invest than do their nonmigrant counterparts. Migrants in the case study from rural Pakistan tend to invest their remittance earnings in land, either irrigated or rain-fed. From the standpoint of the individual migrant, land represents a good investment. Not only does the value of land tend to keep pace with the rate of inflation, but for most peasant farmers it also represents the best type of investment available. To see a greater share of remittances invested in business activities, gov- ernments need to improve the rates of return on small business activities by improving infrastructure--better roads, telephone systems, and electri- cal grids--to encourage commerce and trade. Providing small business loans to new firms also would do much to improve the business climate for new mercantile activities in remittance-receiving areas. The central point is that migrants do invest their remittance earnings, and these investments tend to go into areas where financial rates of return are the greatest. At present, financial rates of return in remittance- receiving areas are highest in land investment. However, with a little cre- ative policy work by the government there is no reason why rates of return could not be raised in the area of small business activities, thereby creat- ing new opportunities for investment by international migrants and their families. In addition to improving physical infrastructure in recipient communi- ties, governments need to provide career opportunities to maximize the use of job skills learned by migrants working abroad. Education in business development, particularly financial planning and management skills, would be particularly effective. In light of the opportunities highlighted above, it is further recommended that more research work be done to clarify the impact of international remit- tances on poverty and economic development in South Asia. Such work should be focused in three areas. 58 MIGRANT LABOR REMITTANCES IN SOUTH ASIA · Data: Governments should make greater efforts to count the amount of remittance monies that are currently transmitted through informal, unof- ficial channels. Since it is likely that poor people remit more through informal channels, a full and complete accounting of the impact of inter- national remittances (official and unofficial) on poverty in South Asia needs more accurate data on the large amount of unofficial remittances returning to that region. · Research methods: More household panel surveys need to be undertaken in South Asian countries to collect information on the expenditure pat- terns of international migrant households. These surveys need to be conducted on nationally representative samples of urban and rural house- holds, and should include modules designed to measure the second- and third-round effects of international remittances on wages and employ- ment. Afull and complete accounting of the impact of international remit- tances on local economic development requires more detailed and complete household-level data. · Investment climate: More research and policy work needs to undertaken to encourage policy makers in South Asia to improve the investment cli- mate for returning migrant workers. If, for example, policy makers would like to see a larger share of international remittance monies invested in business, then they need to undertake the type of infrastructure improve- ments that would encourage expanded commerce and trade. Providing small business loans to new firms would also do much to improve the business climate for returning migrants. The point to emphasize here is that international migrants do invest their remittance earnings, and they tend to invest these earnings in areas where financial rates of return are the greatest. Public Infrastructure for Remittances Recognizing the importance of remittances, governments in Bangladesh, India, Pakistan, and Sri Lanka already have in place the public and private infrastructure for an active remittance industry. The high volume of remit- tances already going to the four countries testifies amply to that fact. Each country has a ministry of labor with a bureau or department dedicated to overseas workers, and a national association for migrant workers to safe- guard their welfare. With varying degrees of success, governments in the region have estab- lished public institutions to facilitate migration of their nationals, the wel- fare of their citizens abroad, and the remittance of earnings to their home countries. However, more can be done to improve the collective effective- ness and efficiency of the public remittance infrastructure. There is still room to improve the quality of training provided to potential migrants, the role of workers' associations in providing financial management information CONCLUSIONS AND RECOMMENDATIONS 59 to their members, and links between state bank branch networks and the private, generally more innovative financial institutions. Formal Financial Infrastructure for Remittances Although the region is inundated with a dense, formal state commercial bank branch network, it does not provide quality financial service, espe- cially to the rural poor. The extensive state bank branch network provides a relatively dense infrastructure for remittance services. However, gaps remain in exploiting the existing infrastructure for affordable payment sys- tems attractive to migrant workers. Poor access to formal finance has resulted in a heavy reliance among poorer rural households on informal finance, mostly moneylenders and shopkeepers, for credit and deposit services, including remittance services. Also, while the cost of international remittances has fallen drastically in recent years, the cost (and convenience) of in-country money transfers-- the last portion of the remittance transaction--remains unnecessarily high compared with the actual cost of technology, labor, and currency-exchange commissions. Though remittance fees in the South Asia region are generally low compared with those in other regions where, according to Ratha (2004), it is not uncommon for fees to be as high as 20 percent for small transfers, more needs to be done to further improve the quality of access and level of outreach of financial institutions. Unless there is a concerted effort to improve the level of access for the rural poor to financial services in general, the following goals will not be maximized with respect to improving their access to remittance services: · Easy, nearby, convenient access to electronic small-value payment ser- vices, particularly in rural areas · Reliable, rapid, low-cost remittances to support the transition from infor- mal to formal channels · Basic financial services, including savings and credit facilities · Affordable access to information and communication The above goals are all essential ingredients for attracting remittance clients. To achieve these goals, the following possible policy options should be considered across the spectrum of remittance options open to the public. · Competition among money-transfer agents: It is often argued that the high costs of remittance reflect the large investments required to enter the formal money transfer market, including a widespread branch network in both source and recipient countries. High fixed costs, it is argued, impede would-be entrants to the market, allowing existing financial insti- tutions to charge above the marginal cost of transactions. Nevertheless, 60 MIGRANT LABOR REMITTANCES IN SOUTH ASIA there is some evidence that competitive forces in other regions are having an impact on reducing remittance fees.3 In South Asia, the competitive process would be strengthened if smaller firms could enter into agree- ments with the state banks or the national post offices to use their branch networks to remit funds. Although many state banks lack the efficiency required to successfully link remittance services, they could dedicate specific staff or branches to the remittance business and thereby make better and faster use of technology to raise their efficiency. · Shared payments systems platforms: In each country, the vast network of rural financial institutions should be converted into a single shared pay- ments platform through which remittances could easily be channeled to any part of the country regardless of the owner of the physical building at the final remittance outlet. Presently, banks in the region are in the process of upgrading their IT systems and networks to provide electronic financial services in a very proprietary manner. Except for the payments clearing system, there appears to be minimal cooperation in integrating or creating shared financial services networks (such as ATMs). Consequently, the investment in this technology may not be achieving its maximum remittance utilization potential. To rectify this, investment in payments information technology that supports an open architecture payment system should become the norm. Recipients must be able to get remittance value from any state or private bank or nonbank finan- cial institution whether or not they hold an account at that outlet. The expected benefits of standardization of payments systems and shared platforms and networks are immense. · Public-private partnerships: Achieving the level of integration suggested above requires strategic public-private partnerships between state insti- tutions (banks and post offices) with their vast branch networks, and pri- vate institutions (banks, nonbanks, and Internet service providers) with the technology and capital for investment. Partnerships across infra- structure, products, and services will yield the types of gains attained in Latin America and East Asia with respect to remittance costs and con- venience. Fundamentally, remittance service providers should cooper- ate on infrastructure and compete on services, but not both. For example, the 4,600 offices of the Sri Lankan post office provide a basic infrastructure for rural financial services that, hitherto, remain strongly underutilized. Partnering with the post office immediately increases the reach of any bank that chooses to do so. International expe- rience shows that postal networks can--as a niche player--be the front office for broad-based access to payments, deposits, and other financial services, and also for channeling informal remittances into formal channels.4 · Cross-selling: In the medium to long term, the profitability of the remit- tance business will continue to decline. Advancements in technology CONCLUSIONS AND RECOMMENDATIONS 61 will ensure that banks receive less income from cash floats. Better macro- economic management has already whittled away foreign exchange prof- its, and competition will force banks to charge lower fixed commissions on transfers. It is therefore essential that banks continue to build on cross- selling other financial products and services (as many have already started doing). Encouraging recipients to open and maintain bank accounts that might eventually lead to short- and long-term loans should be part of the business strategy of banks in the region. · Credibility: Finally, besides investing in technology and creating new public-private partnerships, formal financial institutions, particularly state banks and post offices, need to rebuild their credibility. People use informal remittance systems because they are not only cheaper, but they are also often more reliable than the formal institutions within their reach. Informal Financial Infrastructure for Remittances In Bangladesh, India, Pakistan, and Sri Lanka, informal remittance systems are commonplace. Originally developed to facilitate trade between distant regions at a time when conventional banking instruments were either absent or weak, their speed, low cost, ethnic and cultural convenience, versatility, and anonymity led to their use for various legal and illegitimate remittance purposes, including money laundering and terrorist financing. Clearly, the development of various informal fund transfer systems over many years and across many countries points to the important role that these systems can play in the absence of a robust and efficient formal finan- cial sector. However, the informal nature of these systems, particularly anonymity and lack of records, presents risks of misuse. If the formal finan- cial sector is to respond to the legitimate market demand for hawala-type transactions, and if hawala operators are to be effectively regulated, countries will need sound and sustainable macroeconomic policies, a well-developed payments system, and a healthy financial sector. The progress made by the formal sector in expanding its activity at the expense of informal activity could easily be reversed. Poorly functioning financial systems or the dete- rioration in financial or macroeconomic conditions could pave the way for greater recourse to informal payments systems. Meanwhile, setbacks in financial and exchange liberalization or increases in the spread between official and parallel market exchanges will spur greater use of informal funds transfer systems. For long-term financial sector development, addressing the risks of financial abuse and criminal activity requires a two-pronged approach. In the majority of countries, where infor- mal systems exist alongside a functioning conventional banking sector, it is recommended that hawala dealers be registered. In these systems, addi- tional efforts should be made to improve transparency by bringing informal dealers closer to the formal financial sector without attempting to alter their 62 MIGRANT LABOR REMITTANCES IN SOUTH ASIA specific nature. Simultaneously, the regulatory response must address weak- nesses that may exist in the formal sector. Presently, formal and informal financial systems benefit from their mutual deficiencies, with each tending to expand when the condition of the other is impaired. High transaction costs, long delays in transferring remittances, exchange controls, and overly bureaucratic policies and procedures for simple money transfers are major incentives for the existence of the informal financial system. To face the challenge, the formal sector should tackle its deficiencies and enhance its competitiveness. In conflict-afflicted countries with no functioning bank- ing system, imposing requirements beyond basic registration may not be feasible because of lack of supervisory capacity. Conclusion Overall, remittances to the South Asia region are substantial and, given today's migration trends, will continue to grow. Harnessing the development impact of these flows with an efficient and effective formal public and private remittance infrastructure in the face of an active parallel informal structure is today's principal policy challenge for governments in the region--a challenge governments in Bangladesh, India, Pakistan, and Sri Lanka are taking seriously. Notes 1. See World Bank (2004a: 143). 2. The World Bank estimates that foreign-born persons now account for 24 percent of the total population in Australia, 17 percent in Canada, 10 percent in the United States, 5 percent in Europe, and 1 percent in Japan. 3. Ratha (2004) reports that some nonprofit credit unions affiliated with the World Council of Credit Unions and its International Remittance Network are able to provide the same service--for example, sending US$1,000 to Mexico from the United States--for a fixed fee of US$10­$15, substantially lower than the US$50­$76 charged by major commercial money-transfer agents. South Africa's Teba Bank and Canada's Meli Melo Transfert are able to send cross-border transfers at a fixed fee of US$3 for amounts up to US$400. In Hong Kong, cutthroat competition among money-transfer agents has brought the cost of sending remittances to the Philippines down to a fixed US$2.50 per transaction. 4. See Ratha (2004: 165). Appendix 1 Summary of Dataset on Poverty, Inequality, International Migration, and Remittances Official Poverty Migration Official remittances headcount Squared as share of remittances as share of Survey (US$1/ Poverty poverty Gini country (US$ GDP (constant Country year Region person/day) gap (%) gap (%) coefficient population million) 1995 US$) Algeria 1988 Middle East, 63 North Africa 1.75 0.64 0.48 0.414 2.77 379 0.97 Algeria 1995 Middle East, North Africa 1.16 0.23 0.094 0.353 2.01 1,101 2.63 Bangladesh 1984 South Asia 26.16 5.98 1.96 0.258 0.04 527 2.19 Bangladesh 1986 South Asia 21.96 3.92 1.07 0.269 0.04 497 1.9 Bangladesh 1989 South Asia 33.75 7.72 2.44 0.288 0.05 771 2.68 Bangladesh 1992 South Asia 35.86 8.77 2.98 0.282 0.06 848 2.55 Bangladesh 1996 South Asia 29.07 5.88 1.6 0.336 0.09 1,217 3.16 Belarus 1988 Europe, Central Asia 0 0 0 0.227 0 0 0 Belarus 1993 Europe, Central Asia 1.06 0.13 0.03 0.216 0 0 0 Belarus 1995 Europe, Central Asia 2.27 0.71 0.46 0.287 0 29 0.27 Bolivia 1990 Latin America 11.28 2.22 0.6 0.42 0.47 2 0.03 Botswana 1985 Sub-Saharan Africa 33.3 12.53 6.09 0.542 0 0 0 Brazil 1985 Latin America 15.8 4.69 1.82 0.595 0.05 25 0.01 Brazil 1988 Latin America 18.62 6.78 3.22 0.624 0.05 19 0.01 (Continues on the following page.) Appendix 1 (continued) Official Poverty Migration Official remittances headcount Squared as share of remittances as share of Survey (US$1/ Poverty poverty Gini country (US$ GDP (constant Country year Region person/day) gap (%) gap (%) coefficient population million) 1995 US$) Brazil 1993 Latin America 18.79 8.38 5.01 0.615 0.08 1,123 0.17 Brazil 1995 Latin America 13.94 3.94 1.46 0.6 0.09 2,891 0.41 Brazil 1997 Latin America 5.1 1.32 0.5 0.517 0.11 1,324 0.17 Bulgaria 1989 Europe, Central Asia 0 0 0 0.233 0.2 0 0 Bulgaria 1992 Europe, Central Asia 0 0 0 0.308 0.2 0 0 Bulgaria 1995 Europe, Central Asia 0 0 0 0.285 0.2 0 0 Burkina Faso 1994 Sub-Saharan Africa 61.18 25.51 13.03 0.482 0 80 3.83 64 Central African Rep. 1993 Sub-Saharan Africa 66.58 40.04 28.56 0.613 0 0 0 Chile 1987 Latin America 10.2 2.25 0.66 0.564 0.4 0 0 Chile 1990 Latin America 8.26 2.03 0.73 0.56 0.42 0 0 Chile 1992 Latin America 3.91 0.74 0.23 0.557 0.44 0 0 Chile 1994 Latin America 4.18 0.65 0.15 0.548 0.46 0 0 Colombia 1988 Latin America 4.47 1.31 0.57 0.531 0.8 448 0.65 Colombia 1991 Latin America 2.82 0.75 0.32 0.513 0.86 866 1.14 Colombia 1995 Latin America 8.87 2.05 0.63 0.574 1.02 739 0.8 Colombia 1996 Latin America 10.99 3.16 1.21 0.571 1.06 635 0.67 Costa Rica 1986 Latin America 12.52 5.44 3.27 0.344 1.43 0 0 Costa Rica 1990 Latin America 11.08 4.19 2.37 0.456 1.41 0 0 Costa Rica 1993 Latin America 10.3 3.53 1.79 0.462 1.58 0 0 Costa Rica 1996 Latin America 9.57 3.18 1.55 0.47 1.71 122 1.03 Côte d'Ivoire 1985 Sub-Saharan Africa 4.71 0.59 0.1 0.412 0 0 0 Côte d'Ivoire 1987 Sub-Saharan Africa 3.28 0.41 0.08 0.4 0 0 0 Côte d'Ivoire 1993 Sub-Saharan Africa 9.88 1.86 0.54 0.369 0 0 0 Côte d'Ivoire 1995 Sub-Saharan Africa 12.29 2.4 0.71 0.367 0 0 0 Czech Republic 1988 Europe, Central Asia 0 0 0 0.194 1.73 0 0 Czech Republic 1993 Europe, Central Asia 0 0 0 0.266 1.53 0 0 Dominican Republic 1989 Latin America 7.73 1.51 0.42 0.504 4.89 301 2.91 Dominican Republic 1996 Latin America 3.19 0.71 0.26 0.487 7.08 914 7.13 Ecuador 1988 Latin America 24.85 10.21 5.82 0.439 1.38 0 0 Ecuador 1995 Latin America 20.21 5.77 2.27 0.437 1.92 382 2.13 Egypt, Arab Rep. of 1991 Middle East, North Africa 3.97 0.53 0.13 0.35 0.15 2,569 4.99 Egypt, Arab Rep. of 1995 Middle East, North Africa 5.55 0.66 0.13 0.283 0.18 3,279 5.45 El Salvador 1989 Latin America 25.49 13.72 10.06 0.489 9.06 228 3.39 El Salvador 1996 Latin America 25.26 10.35 5.79 0.522 11.67 1,084 11.22 Estonia 1988 Europe, Central Asia 0 0 0 0.229 0 0 0 65 Estonia 1993 Europe, Central Asia 3.15 0.91 0.51 0.395 0 0 0 Estonia 1995 Europe, Central Asia 4.85 1.18 0.39 0.353 0 0 0 Ethiopia 1981 Sub-Saharan Africa 32.73 7.69 2.71 0.324 0.07 0 0 Ethiopia 1995 Sub-Saharan Africa 31.25 7.95 2.99 0.399 0.09 0 0 Gambia, The 1992 Sub-Saharan Africa 53.69 23.27 13.28 0.478 0 0 0 Ghana 1987 Sub-Saharan Africa 47.68 16.6 7.81 0.353 0.11 1 0.02 Ghana 1989 Sub-Saharan Africa 50.44 17.71 8.36 0.359 0.12 6 0.11 Ghana 1992 Sub-Saharan Africa 45.31 13.73 5.61 0.339 0.18 7 0.12 Ghana 1999 Sub-Saharan Africa 44.81 17.28 8.71 0.327 0.32 26 0.37 Guatemala 1987 Latin America 47.04 22.47 13.63 0.582 2.09 0 0 Guatemala 1989 Latin America 39.81 19.79 12.59 0.596 2.34 69 0.59 Honduras 1989 Latin America 44.67 20.65 12.08 0.595 2.11 35 1.05 Honduras 1992 Latin America 38.98 17.74 10.4 0.545 2.74 60 1.65 Honduras 1994 Latin America 37.93 16.6 9.38 0.552 3.23 85 2.23 (Continues on the following page.) Appendix 1 (continued) Official Poverty Migration Official remittances headcount Squared as share of remittances as share of Survey (US$1/ Poverty poverty Gini country (US$ GDP (constant Country year Region person/day) gap (%) gap (%) coefficient population million) 1995 US$) Honduras 1996 Latin America 40.49 17.47 9.72 0.537 3.66 128 3.12 Hungary 1989 Europe, Central Asia 0 0 0 0.233 2.02 0 0 Hungary 1993 Europe, Central Asia 0 0 0 0.279 1.75 0 0 India 1983 South Asia 52.55 16.27 n.a. 0.32 0.04 2,311 1.25 India 1986 South Asia 47.46 13.92 n.a. 0.337 0.06 2,105 0.99 India 1988 South Asia 47.99 13.51 n.a. 0.329 0.07 2,402 0.98 India 1990 South Asia 45.95 12.63 n.a. 0.312 0.09 1,875 0.68 66 India 1995 South Asia 46.75 12.72 n.a. 0.363 0.11 7,685 2.17 India 1997 South Asia 44.03 11.96 n.a. 0.378 0.12 10,688 2.7 Indonesia 1987 East Asia 28.08 6.08 1.78 0.331 0.01 86 0.07 Indonesia 1993 East Asia 14.82 2.98 0.39 0.317 0.05 346 0.2 Indonesia 1996 East Asia 7.81 0.95 0.18 0.364 0.08 796 0.39 Indonesia 1998 East Asia 26.33 5.43 1.69 0.315 0.1 959 0.4 Iran, Islamic Rep. of 1990 Middle East, North Africa 0.9 0.8 n.a. 0.434 0.63 1 0.01 Jamaica 1988 Latin America 5.02 1.38 0.67 0.431 17.03 76 1.73 Jamaica 1990 Latin America 0.62 0.03 0.01 0.418 19.07 136 2.75 Jamaica 1993 Latin America 4.52 0.86 0.29 0.379 21.8 187 3.39 Jamaica 1996 Latin America 3.15 0.73 0.32 0.364 24.4 636 11.46 Jordan 1987 Middle East, North Africa 0 0 0 0.36 0.87 939 16.72 Jordan 1992 Middle East, North Africa 0.55 0.12 0.05 0.433 0.93 843 14.46 Jordan 1997 Middle East, North Africa 0.36 0.1 0.06 0.364 0.94 1,655 23.08 Kazakhstan 1988 Europe, Central Asia 0.05 0.02 0.01 0.257 0 0 0 Kazakhstan 1993 Europe, Central Asia 1.06 0.04 0.01 0.326 0 0 0 Kazakhstan 1996 Europe, Central Asia 1.49 0.27 0.1 0.354 0 10 0.05 Kenya 1992 Sub-Saharan Africa 33.54 12.82 6.62 0.574 0 0 0 Kenya 1994 Sub-Saharan Africa 26.54 9.03 4.5 0.445 0 0 0 Kyrgyz Republic 1988 Europe, Central Asia 0 0 0 0.26 0 0 0 Kyrgyz Republic 1993 Europe, Central Asia 22.99 10.87 6.82 0.537 0 2 0.1 Kyrgyz Republic 1997 Europe, Central Asia 1.57 0.28 0.1 0.405 0 3 0.17 Latvia 1988 Europe, Central Asia 0 0 0 0.225 0 0 0 Latvia 1993 Europe, Central Asia 0 0 0 0.269 0 0 0 Latvia 1995 Europe, Central Asia 0 0 0 0.284 0 0 0 Latvia 1998 Europe, Central Asia 0.19 0.01 0 0.323 0 3 0.05 67 Lesotho 1987 Sub-Saharan Africa 30.34 12.66 6.85 0.56 0 0 0 Lesotho 1993 Sub-Saharan Africa 43.14 20.26 11.84 0.579 0 0 0 Lithuania 1988 Europe, Central Asia 0 0 0 0.224 0 0 0 Lithuania 1993 Europe, Central Asia 16.47 3.37 0.95 0.336 0 0 0 Lithuania 1996 Europe, Central Asia 0 0 0 0.323 0 2 0.03 Madagascar 1980 Sub-Saharan Africa 49.18 19.74 10.21 0.468 0 0 0 Madagascar 1994 Sub-Saharan Africa 60.17 24.46 12.83 0.434 0 11 0.35 Mali 1989 Sub-Saharan Africa 16.46 3.92 1.39 0.365 0 76 3.49 Mali 1994 Sub-Saharan Africa 72.29 37.38 23.06 0.505 0 103 4.43 Mauritania 1988 Sub-Saharan Africa 40.64 19.07 12.75 0.425 0 9 1.04 Mauritania 1993 Sub-Saharan Africa 49.37 17.83 8.58 0.5 0 2 0.2 Mauritania 1995 Sub-Saharan Africa 30.98 9.99 4.59 0.389 0 5 0.47 Mexico 1984 Latin America 12.05 2.65 0.78 0.54 1.86 1,127 0.47 Mexico 1989 Latin America 16.2 5.63 2.75 0.551 4.66 2,213 0.87 (Continues on the following page.) Appendix 1 (continued) Official Poverty Migration Official remittances headcount Squared as share of remittances as share of Survey (US$1/ Poverty poverty Gini country (US$ GDP (constant Country year Region person/day) gap (%) gap (%) coefficient population million) 1995 US$) Mexico 1992 Latin America 13.31 3.23 1.04 0.543 6.1 3,070 1.07 Mexico 1995 Latin America 17.9 6.15 2.92 0.537 7.39 3,673 1.28 Moldova 1988 Europe, Central Asia 0 0 0 0.241 0 0 0 Moldova 1992 Europe, Central Asia 7.31 1.32 0.32 0.344 0 0 0 Morocco 1985 Middle East, North Africa 2.04 0.7 0.5 0.392 4.38 967 3.81 Morocco 1990 Middle East, 68 North Africa 0.14 0.02 0.01 0.392 4.02 1,336 4.24 Mozambique 1996 Sub-Saharan Africa 37.85 12.02 5.42 0.396 0 0 0 Namibia 1993 Sub-Saharan Africa 34.93 13.97 6.93 0.743 0 8 0.25 Nepal 1985 South Asia 42.13 10.79 3.75 0.334 0 39 1.43 Nepal 1995 South Asia 37.68 9.74 3.71 0.387 0 101 2.3 Nicaragua 1993 Latin America 47.94 20.4 11.19 0.503 4.38 25 1.47 Niger 1993 Sub-Saharan Africa 41.73 12.43 5.29 0.361 0 13 0.74 Niger 1995 Sub-Saharan Africa 61.42 33.93 23.66 0.506 0 6 0.32 Nigeria 1997 Sub-Saharan Africa 70.24 34.91 n.a. 0.505 0.09 1,920 6.37 Pakistan 1988 South Asia 49.63 14.85 6.03 0.333 0.11 2,013 4.56 Pakistan 1991 South Asia 47.76 14.57 6.04 0.332 0.16 1,848 3.62 Pakistan 1993 South Asia 33.9 8.44 3.01 0.342 0.18 1,562 2.78 Pakistan 1997 South Asia 30.96 6.16 1.86 0.312 0.22 1,409 2.19 Panama 1989 Latin America 16.57 7.84 4.9 0.565 3.53 14 0.25 Panama 1991 Latin America 18.9 8.87 5.48 0.568 3.55 14 0.21 Panama 1995 Latin America 14.73 6.15 3.39 0.57 3.61 16 0.2 Panama 1997 Latin America 10.31 3.15 3.67 0.485 3.67 16 0.19 Paraguay 1990 Latin America 11.05 2.47 0.8 0.397 0 43 0.56 Paraguay 1995 Latin America 19.36 8.27 4.65 0.591 0 200 2.21 Peru 1985 Latin America 1.14 0.29 0.14 0.457 0.33 0 0 Peru 1994 Latin America 9.13 2.37 0.92 0.446 0.89 472 0.96 Peru 1997 Latin America 15.49 5.38 2.81 0.462 1.03 636 1.08 Philippines 1985 East Asia 22.78 5.32 1.66 0.41 1.26 111 0.21 Philippines 1988 East Asia 18.28 3.59 0.94 0.407 1.49 388 0.64 Philippines 1991 East Asia 15.7 2.79 0.66 0.438 1.69 329 0.49 Philippines 1994 East Asia 18.36 3.85 1.07 0.429 1.86 443 0.62 Philippines 1997 East Asia 14.4 2.85 0.75 0.461 2 1,057 1.28 Poland 1987 Europe, Central Asia 0 0 0 0.255 1.89 0 0 Poland 1990 Europe, Central Asia 0.08 0.027 0.02 0.283 1.84 0 0 Poland 1992 Europe, Central Asia 0.08 0.031 0.02 0.271 1.81 0 0 69 Romania 1989 Europe, Central Asia 0 0 0 0.233 0.62 0 0 Romania 1992 Europe, Central Asia 0.8 0.34 0.31 0.254 0.77 0 0 Romania 1994 Europe, Central Asia 2.81 0.76 0.43 0.282 0.88 4 0.01 Russian Federation 1994 Europe, Central Asia 6.23 1.6 0.55 0.436 0.34 0 0 Russian Federation 1996 Europe, Central Asia 7.24 1.6 0.47 0.48 0.35 0 0 Russian Federation 1998 Europe, Central Asia 7.05 1.45 0.39 0.487 0.36 0 0 Senegal 1991 Sub-Saharan Africa 45.38 19.95 11.18 0.541 0 105 2.54 Senegal 1994 Sub-Saharan Africa 26.26 7.04 2.73 0.412 0 73 1.71 Sierra Leone 1989 Sub-Saharan Africa 56.81 40.45 33.8 0.628 0.18 0 0 South Africa 1993 Sub-Saharan Africa 11.47 1.83 0.38 0.593 0.14 0 0 Sri Lanka 1985 South Asia 9.39 1.69 0.5 0.324 0.06 292 3.45 Sri Lanka 1990 South Asia 3.82 0.67 0.23 0.301 0.12 401 4 Sri Lanka 1995 South Asia 6.56 1 0.26 0.343 0.3 790 6.06 Tanzania 1991 Sub-Saharan Africa 48.54 24.42 15.4 0.59 0 0 0 (Continues on the following page.) Appendix 1 (continued) Official Poverty Migration Official remittances headcount Squared as share of remittances as share of Survey (US$1/ Poverty poverty Gini country (US$ GDP (constant Country year Region person/day) gap (%) gap (%) co-efficient population million) 1995 US$) Tanzania 1993 Sub-Saharan Africa 19.89 4.77 1.66 0.381 0 0 0 Thailand 1988 East Asia 25.91 7.36 2.73 0.438 0.17 0 0 Thailand 1992 East Asia 6.02 0.48 0.05 0.462 0.21 0 0 Thailand 1996 East Asia 2.2 0.14 0.01 0.434 0.24 0 0 Thailand 1998 East Asia 0 0 0 0.413 0.25 0 0 Trinidad and Tobago 1992 Latin America 12.36 3.48 n.a. 0.402 10.5 6 0.12 Tunisia 1985 Middle East, 70 North Africa 1.67 0.34 0.13 0.434 3.12 271 2.11 Tunisia 1990 Middle East, North Africa 1.26 0.33 0.16 0.402 3.01 551 3.71 Turkey 1987 Europe, Central Asia 1.49 0.36 0.17 0.435 4.18 2,021 1.56 Turkey 1994 Europe, Central Asia 2.35 0.55 0.24 0.415 4.13 2,627 1.66 Turkmenistan 1988 Europe, Central Asia 0 0 0 0.264 0 0 0 Turkmenistan 1993 Europe, Central Asia 20.92 5.69 2.1 0.357 0 0 0 Uganda 1989 Sub-Saharan Africa 39.17 14.99 7.57 0.443 0 0 0 Uganda 1993 Sub-Saharan Africa 36.7 11.44 5 0.391 0 0 0 Ukraine 1989 Europe, Central Asia 0 0 0 0.233 0 0 0 Ukraine 1992 Europe, Central Asia 0.04 0.01 0.01 0.257 0 0 0 Ukraine 1996 Europe, Central Asia 0 0 0 0.325 0 0 0 Uruguay 1989 Latin America 1.1 0.47 0.4 0.423 0 0 0 Uzbekistan 1988 Europe, Central Asia 0 0 0 0.249 0 0 0 Uzbekistan 1993 Europe, Central Asia 3.29 0.46 0.11 0.332 0 0 0 Venezuela 1981 Latin America 6.3 1.08 0.25 0.556 0.08 0 0 Venezuela 1987 Latin America 6.6 1.04 0.22 0.534 0.14 0 0 Venezuela 1989 Latin America 8.49 1.77 0.49 0.557 0.19 0 0 Venezuela 1993 Latin America 2.66 0.57 0.22 0.416 0.29 0 0 Venezuela 1996 Latin America 14.69 5.62 3.17 0.487 0.36 0 0 Yemen, Republic of 1992 Middle East, North Africa 5.07 0.93 n.a. 0.394 0 1,018 28.49 Yemen, Republic of 1998 Middle East, North Africa 10.7 2.42 0.85 0.344 0 1,202 23.77 Zambia 1991 Sub-Saharan Africa 58.59 31.04 20.18 0.483 0 0 0 Zambia 1993 Sub-Saharan Africa 69.16 38.49 25.7 0.462 0 0 0 Zambia 1996 Sub-Saharan Africa 72.63 37.75 23.88 0.497 0 0 0 Zimbabwe 1991 Sub-Saharan Africa 35.95 11.39 4.56 0.568 0 0 0 Sources: All poverty and inequality data are from the World Bank's Global Poverty Monitoring database. Migration data are from the U.S. Census 71 Bureau and OECD, Trends in International Migration. Remittance data are from the IMF, Balance of Payments. Bibliography Abella, M. I. 1989. "Policies and Practices to Promote Remittances" Philippine Labour Review 13(1): 1­17. ABN AMRO. 2005. ATM Network. Available at http://www.abnamro. com.pk/consumerbnk/atmnw.htm. Adams, R. Jr. 1991. "The Effects of International Remittances on Poverty, Inequality and Development in Rural Egypt." Research Report 86, International Food Policy Research Institute, Washington, DC. . 1998. "Remittances, Investment and Rural Asset Accumulation in Pakistan." Economic Development and Cultural Change 47(1): 155­73. . 2003a. "Economic Growth, Inequality and Poverty: Findings from a New Data Set." Policy Research Working Paper 2972, World Bank, Washington, DC. . 2003b. "International Migration, Remittances and the Brain Drain: A Study of 24 Labor-Exporting Countries." Policy Research Working Paper 3069, World Bank, Washington, DC. Adams, R. Jr., and J. Page. 2003. "The Impact of International Migration and Remittances on Poverty." Paper prepared for the World Bank, Department for International Development (DFID, UK), and International Migration Policy Programme (IMP) Conference, "The International Conference on Migrant Remittances: Development Impact, Opportunities for the Financial Sector and Future Prospects," London, October 9­10. Alburo, F. A., and D. I. Abella. 1992. "The Impact of Informal Remittances of Overseas Contract Workers' Earnings on the Philippine Economy." New Delhi: ARTEP (mimeo). APG (Asia Pacific Group against Money Laundering). 2001. "Report for the APG Money Laundering Methods and Typologies Workshop." Singapore. . 2002. "Outline of Anti-Money Laundering Laws Passed or Under Consideration in the Asia/Pacific Region." APG Secretariat, Sydney. Azad, A. K. 2005 (forthcoming). "The Importance of Migrant Remittances for the Bangladesh Economy." In Remittances: Development Impact and Potential, ed. S. Maimbo and D. Ratha. Washington, DC: World Bank. Bagai, S. S. 1991. Black Money in India. Bombay: Bagai Tax Law Services. Bank of Ceylon. 2005. Our Strength.Available at http://www.bankofceylon. net/boc/strength.htm. Bevan, D., P. Collier, and J. Gunning. 1989. Peasants and Governments: An Economic Analysis. Oxford: Clarendon Press. 72 BIBLIOGRAPHY 73 Birdsall, N., and J. L. Londoño. 1997. "Asset Inequality Matters: An Assessment of the World Bank's Approach to Poverty Reduction." American Economic Review 87(2): 32­37. Central Bank of Sri Lanka. 2003. Annual Report 2002. Colombo: Central Bank of Sri Lanka. Central Bank of Sri Lanka, Payment and Settlements Department. 2004. Payments Bulletin 4(2). Chandavarkar, A. G. 1980. "Use of Migrants' Remittances in Labor-Exporting Countries." Finance and Development 17(2): 36­39. Chopra, O. P. 1985. Tax Ethics, Unaccounted Income: Some Tax Reforms, or Black Money, The Norm of the Day. New Delhi: Inter-India. Collier, P., and J. Dehn. 2001. Aid, Shocks, and Growth. Policy Research Working Paper No. 2688, World Bank, Washington, DC. The Economic Times (Channai). January 10, 2004. El-Qorchi, M., S. M. Maimbo, and J. F. Wilson. 2003. "Informal Funds Transfer Systems: An Analysis of the Informal Hawala System." IMF Occasional Paper 222, International Monetary Fund, Washington, DC. FATF (Financial Action Task Force). 1997. 1996­1997 Report on Money Laundering Typologies. Paris: OECD. . 1999. 1998­1999 Report on Money Laundering Typologies. Paris: OECD. . 2001. 2000­2001 Report on Money Laundering Typologies. Paris: OECD. . 2002. 2001­2002 Report on Money Laundering Typologies. Paris: OECD. Foreign Exchange Management Act [India] (FEMA). 2000. Foreign Exchange Regulation Act [India] (FERA). 1973. G-10 Committee on Banking Regulations and Supervisory Practices. 1988. "Prevention of Criminal Use of the Banking System for the Purpose of Money Laundering." Basel Statement of Principles, Bank for International Settlements, Basel, Switzerland. GAO (U.S. General Accounting Office). 1994. "Money Laundering: U.S. Efforts to Fight It Are Threatened by Currency Smuggling." GAO/GGD- 94-73, Washington, DC. Gaur, K. D. 1987. Taxation, Black Money, and the Law. New Delhi: Deep & Deep. Gupta, S. B. 1992. Black Income in India. New Delhi: Sage. Habib, A. 1985. "Economic Consequences of International Migration for Sending Countries: Review of Evidence from Bangladesh." Ph.D. thesis, University of Newcastle, Australia. Hugo, G. 2003. "Migration and Development: A Perspective from Asia." IOM Migration Research Series 14, International Organization for Migration, Geneva. IMF (International Monetary Fund). Various years. Balance of Payments Statistics Yearbook. Washington, DC: IMF. . 2001a. "Financial System Abuse, Financial Crime, and Money Laundering." Background Paper, IMF, Washington, DC. Available at: www.imf.org/external/np/ml/2001/eng/021201.pdf. 74 BIBLIOGRAPHY IMF and World Bank. 2003. "Global Dialogue on Money Laundering and Terrorist Financing." October 23­24, 2002. International Monetary Fund and World Bank Global Dialogue Series. Washington, DC: World Bank. International Organisation for Migration (IOM). 2003. "Labor Migration in Asia: Trends, Challenges and Policy Responses in Countries of Origin." Vienna: IOM Publicatons. Iqbal, Z., S. K. Qureshi, and R. Mahmood. 1998. The Underground Economy and Tax Evasion in Pakistan: A Fresh Assessment. Islamabad: Pakistan Institute of Development Economics. Jost, P., and H. S. Sandu. 2000. The Hawala Alternative Remittance System and Its Role in Money Laundering. Lyon: Interpol. Kaplan, D. E., and A. Dubro. 1986. Yakuza: The Explosive Account of Japan's Criminal Underworld. Reading, Mass.: Addison-Wesley. Kapoor, S. 1998. Bad Money, Bad Politics: The Untold Hawala Story. New Delhi: South Asia Books. Kazi, S. 1989. "Domestic Impact of Overseas Migration: Pakistan." In To the Gulf and Back: Studies on the Economic Impact of Asian Labor Migration, ed. R. Amjad, 167­96. New Delhi: ILO/ARTEP. Kochi, J. 1999. "Hawala Rates for Dollar Surge." FT Asia Intelligence Wire/ Business Line. January 7. Mahmud, W. 1989. "The Impact of Overseas Labour Migration on the Bangladesh Economy." Bangladesh Development Studies 8(3): 1­28. Mahmud, W., and S. R. Osmani. 1980. "Impact of Emigrant Workers on the Bangladesh Economy." Bangladesh Development Studies 8(3): 1­28. Maimbo, S. M. 2003. "The Money Exchange Dealers of Kabul: A Study of the Informal Funds Transfer Market in Afghanistan." World Bank Working Paper 13, World Bank, Washington, DC. Maimbo, S., and N. Passas. 2004. "The Regulation and Supervision of Informal Remittance Systems." Small Enterprise Development 15(1): 53­62. MCB (Muslim Commercial Bank). 2005. MCB Virtual Internet Banking Web site. Available at: http://www.mcb.com.pk/. Miller, M. 1999. "Underground Banking." Institutional Investor 24(1): 32­37. Murshid, K. A. S., I. Kazi, and A. Meherun. 2001. Migrant Workers from Bangladesh: Remittance Inflows and Utilization. Dhaka: Bangladesh Institute of Development Studies. Naqvi, J. 1994. "India Says Economic Crime Rising, Plans Crackdown." The Reuter Asia-Pacific Business Report, BC Cycle. October 24. Naylor, R. T. 2002. Wages of Crime: Black Markets, Illegal Finance, and the Underworld Economy. Ithaca, N.Y.: Cornell University Press. OECD (Organisation for Economic Co-operation and Development). Various years. Trends in International Migration: Annual Report. Paris: OECD. Passas, N. 1991. "Frauds Affecting the Budget of the European Community." Report to the Commission of the European Communities, Brussels. BIBLIOGRAPHY 75 . 1999a. "Informal Value Transfer Systems and Criminal Organizations: A Study into So-Called Underground Banking Networks." Report pre- pared for the Netherlands Ministry of Justice, The Hague. . 1999b. "Globalization, Criminogenic Asymmetries, and Economic Crime." European Journal of Law Reform 1(4): 399­423. . 2000. "Facts and Myths About `Underground Banking.' " In Cross- Border Crime in a Changing Europe, ed. P. C. van Duyne, V. Ruggiero, M. Scheinost, and W. Valkenburg, 192­208. Tilburg and Prague: Tilburg University and Prague Institute of Criminology and Social Prevention. ------. 2003. "Informal Value Transfer Systems, Money Laundering, and Terrorism." Report prepared for the National Institute of Justice and Financial Crimes Enforcement Network, Washington, DC. People's Bank [Sri Lanka]. 2005. "Our Branch Network." Available at: http://www.peoplesbank.lk/Templates/Branchnetwork.htm. Pohit, S., and N. Taneja. 2000. India's Informal Trade with Bangladesh and Nepal: A Qualitative Assessment. New Delhi: Indian Council for Research on International Economic Relations. Puri, S., and T. Ritzema 1999. "Migrant Worker Remittances, Micro-Finance, and the Informal Economy: Prospects and Issues." Working Paper 21, Social Finance Unit, International Labor Organization, Geneva. Ratha, D. 2003. "Workers' Remittances: An Important and Stable Source of External Development Finance." In Global Development Finance 2003 (chapter 7). Washington, DC: World Bank. . 2004. "Enhancing the Developmental Effect of Workers' Remittances to Developing Countries." In Global Development Finance 2004 (appen- dix A). Washington, DC: World Bank. Ravallion, M. 1997. "Can High-Inequality Developing Countries Escape Absolute Poverty?" Economics Letters 56: 51­7. Ravallion, M., and S. Chen. 1997. "What Can New Survey Data Tell Us about Recent Changes in Distribution and Poverty?" World Bank Economic Review 11(2): 357­82. Reserve Bank of India (RBI). 2004a. Bulletin Mumbai (September). . 2004b. Annual Report. Mumbai: RBI. Russell, S. 1986. "Remittances from International Migration: A Review in Perspective." World Development 14(6): 677­96. Sander, C., and S. Maimbo. 2003. "Migrant Labor Remittances in Africa: Reducing Obstacles to Developmental Contributions." Africa Region Working Paper Series 64, World Bank, Washington, DC. Seddon, D., J. Adhikari, and G. Gurung. 2002. "Foreign Labor Migration and the Remittance Economy of Nepal." Critical Asian Studies 34 (1): 19­40. Siddiqui, T., and C. R. Abrar. 2003. "Migrant Worker Remittances and Micro- finance in Bangladesh." Report prepared for the International Labour Office, Refugee and Migratory Movement Research Unit, Dhaka. 76 BIBLIOGRAPHY Sri Lanka Bureau of Foreign Employment, Research Division. 2004. Statistical Handbook on Migration 2003. Colombo: Sri Lanka Bureau of Foreign Employment. Sri Lanka Ministry of Employment and Labour. 2004. National Employment Policy. Colombo: Ministry of Labour. StateBankofPakistan.2002."MandatoryConnectivitytoTwoATMSwitches." SBP Circular No. 20, Karachi. Taylor, J. E., and I. Adelman. 1996. Village Economies: The Design, Estimation, and Use of Villagewide Economic Models. Cambridge: Cambridge University Press. Thieme, S. 2003. "Savings and Credit Associations and Remittances: The Case of Far West Nepalese Labor Migrants in Delhi, India." Working Paper No. 39, Social Finance Unit, International Labour Office, Dhaka. U.A.E. Central Bank. 2002. "Abu Dhabi Declaration on Hawala." Abu Dhabi. United Nations, Department of Economic and Social Affairs, Population Division. 2002. International Migration Report 2002. New York: United Nations. U.S. Census Bureau. U.S. Department of State. 2002. "International Narcotics Control Strategy Report." Washington, DC: U.S. Department of State. U.S. Department of the Treasury. 2002. A Report to the Congress in Accordance with Section 359 of the Uniting and Strengthening America by Providing Appropriate Tools Required to Intercept and Obstruct Terrorism Act of 2001. Washington, DC. http://fincen.gov/hawalarptfinal11222002.pdf. U.S.HouseofRepresentatives,CommitteeonBankingandFinancialServices, Subcommittee on Financial Institutions and Consumer Credit. 2000. H.R. 240--The Bulk Cash Smuggling Act of 1999: Hearing before the Subcommittee on Financial Institutions and Consumer Credit of the Committee on Banking and Financial Services, U.S. House of Representatives, 106th Cong., 2d sess. May 15, 2000. Wimaladharma, J., D. Pearce, and D. Stanton. 2004. "Remittances: The New Development Finance?" Small Enterprise Development 15(1): 12­20. World Bank. 2001. World Development Report 2000/01. Washington, DC: World Bank. . 2004a. Global Economic Prospects 2004: Realizing the Development Promise of the Doha Agenda. Washington, DC: World Bank. . 2004b. "India: Scaling-Up Access to Finance for India's Rural Poor." Finance and Private Sector Development Unit, South Asia Region, Report No. 30740-IN, World Bank, Washington, DC. Wyss, S. 2003. "Organisation and Finance of International Labor Migration: ACase Study of Sainik Basti, Western Nepal." Master's thesis, University of Zurich. Zdanovicz, J. S., W. W. Welch, and S. J. Pak. 1995. "Capital Flight from India to the United States through Abnormal Pricing in International Trade." Finance India 9(3). Index Boxes, figures, notes, and tables are indicated by b, f, n, and t. Access to Finance Survey (World bank drafts, 31 Bank), 39 Bank of Ceylon (Sri Lanka), 32 accountability. See transparency of banks. See formal financial informal remittance industry infrastructure for remittances Africa, 1, 31, 43 bills of exchange, 46­47 "alternative remittance systems." branches, bank See hawala See also formal financial Andhra Bank (India), 37 infrastructure for remittances anonymity of informal financial Arabian Gulf States, lack in, 31 sector infrastructure, 50­51 ATM-networked branches in anti-money laundering and rural communities, 36 counter-terrorist financing of foreign commercial banks, 33 (AML/CTF) standards, 33­34 of Sonali Bank (Bangladesh), 32 Arabian Gulf region. See Middle of Sri Lankan state commercial East; specific country banks, 32 automated teller machines of state banks, 5 (ATMs), 32, 34­35, 60 capital convertibility, 24­25 Bangladesh caps on remittances, 24 circumventing capital and cash floats, reduction of, 60­61 exchange controls, 52 cashier's checks, 32 comparing official and unofficial cash pools in hawala, sources of, remittances, 13t, 13­14 43­44 foreign commercial banks, 33 cash transactions, 42­43, 50 hundi, 43, 46­47. See also hawala charities and remittances in informal financial sector Pakistan, 27 infrastructure, 6, 37, 49 commercial banks, xiii, 31­36 local commercial banks, 34 commissions on remittances, 33 overseas development assistance conflict, areas of and informal (ODA) trends, 2­3, 3f remittance systems, 47 public infrastructure for convenience and accessibility of remittances, 21­23, 23t, 58 informal financial sector infra- state commercial banks, 32 structure, 23, 50 77 78 INDEX costs dual citizenship and Indian bank drafts, 31 migrants, 24 foreign commercial banks, 33 of formal financial East Asia region, 1, 31, 60 infrastructure for remittances, e-banking and e-services, 34, 37 5­6, 40­41, 59 embargoes and sanctions, 53 of informal financial sector Emigration Act of 1983 (India), infrastructure, 48b, 48­49 23­24 remitters' fees, 33 Emigration Ordinance of 1979 telegraphic money transfers, 31 (Pakistan), 25, 26 wire transfers, 32 "ethnic banking." See hawala counter-terrorism, 33­34 Europe, 28, 43, 45, 47 courier systems, 42­43 exchange companies, 40­41 credibility of formal financial exchange rate, profit on, 33 infrastructure, 61 credit card fraud, 53 fees paid by remitters, 33 credit unions, 62n fei-ch'ien (China), 43 cross-selling, 60­61 See also hawala C-Sam, 37 foreign commercial banks, xiii, custom duties, 52­53 33­34 foreign exchange controls, 47, 55n dataset on international formal financial infrastructure for remittances and poverty, 7­10, remittances, xii­xiv, 31­41 9t, 58, 63­71 competition among money- demand drafts, 32 transfer agents, 59­60 development assistance and costs associated with, 5­6, official aid, 2­3, 3f 40­41, 59 development impact of credibility, 61 remittances, x­xi, xf, 2­4, 7­20 cross-selling, 60­61 analyzing, 4, 7­8 econometric results, 10­12, 11t econometric results, 10­12 excessive administrative estimating and simulation requirements, adverse effects functions, 12­14, 13t of, 54, 62 investment and economic goals for, 59­60 development, 16­17 money service businesses, 36­37 new dataset for, 7­10, 9t, 63­71 NRFC (nonresident foreign official and unofficial currency accounts) (Sri Lanka), remittances and poverty, 29­30 10­12, 11t, 14, 15t Pakistan remittances and, 27 recommendations for, 56­58 payment card infrastructure, domestic workers 34­36, 35t physical abuse of, 24 post office networks, 37­39, 38b from Sri Lanka, 28 poverty and, 5, 14, 15t INDEX 79 private foreign and local foreign commercial banks, 33 commercial banks, xiii, 33­36 formal financial public-private partnerships, 60 infrastructure, 40 recommendations, 59­61 hawala. See hawala shared payments systems historical background of platforms, 60 informal remittance systems, state, xii, 5, 29, 31­32, 59, 60 46­47 hundi, 46­47 growth-poverty model, informal financial sector 10­12, 11t infrastructure, 6, 37, 47, 48, Gulf States. See Middle East 48b, 51 Internet services, 37 hawala local commercial banks, 34 circumventing capital and money service businesses, exchange controls, 52 36­37 Indian law against, 51, 55n nationalization of banks, migrant incentives for using, 31­32 47­51 nonresident Indians (NRIs), recommendations, 61­62 incentives for, 24­25, 25t registration of dealers, 54, 61 overseas development assistance risks and potential abuse of, (ODA) trends, 2­3, 3f 51­53 post office networks, 36­37, workings of, 43­46, 44f 38, 38b headcount index, 9, 9t, 14, 63­71 public infrastructure for historical background of informal remittances, 23­25, 25t, 58 remittance systems, 46­47 regional disparities in financial household data, 4, 7, 8, 16, 58 services, 40 hui kuan (Hong Kong), 43 state commercial banks, 31­32 See also hawala Western Union in, 36­37, 38b hundi (Pakistan and Bangladesh), India Post, 36­37, 38b, 39 43, 46­47, 52 informal financial sector See also hawala infrastructure for remittances, xiv­xv, 42­55 illegal businesses and remittance See also hawala systems, 6, 51­53 anonymity, 50­51 import-export businesses, 47 convenience and accessibility, 50 incentives for remittances, 23­27, costs, 48b, 48­49 25t, 47­51 courier systems, 42­43 income tax evasion, 52­53 data on, 7, 58 India foreign exchange controls, 47 ATM machines, 35 historical background of, 46­47 comparing official and unofficial IMF estimates, 8 remittances, 13t, 13­14 in-kind remittances, 43 80 INDEX law enforcement concerns and, Latin America, 1, 31, 60 6, 50­53, 61­62 Living Standard Measurement migrant incentives for using, Survey (LSMS), 16 47­51 loans, predeparture, for Sri Lankan moneylenders, 5 migrants, 29 informal financial sector infra- local banks, xiii, 34 structure for remittances (continued) Malaysia, 24 poor households and, 5, 59 methodological problems in poverty and, 14, 15t analyzing effects of recommendations, 61­62 remittances, 4 risks and potential abuse of, Middle East 51­53 See also specific country shopkeepers, 5 bank branches, lack of, 31 speed, 49, 49t, 51 cash transactions, 42­43, 50 trade and, 46­47 courier services, 43 transparency and accountability, formal financial institutions, 6, 54, 61­62 40­41 travel agents, 37, 43 informal financial sector trust, 48 infrastructure, 50 types of systems, 42­46 migration to, 10, 17, 24, 26t, in-kind remittances, 43 26­27, 28, 28t insurance, predeparture, for money exchange companies, 32 Sri Lankan migrants, 29 physical abuse of domestic International Monetary Fund workers, 24 (IMF) estimates, 8 migration from South Asia region, international money orders, 38b 9­10, 56 Internet and fund transfers, 34, 37 See also Middle East investment, 16­17, 27, 56 Bangladesh, 22­23, 23t investment climate, improving, 58 India, 23­24, 30n Inward International Money Order Pakistan, 25­26, 26t Service (India), 38b skilled migrants, 33 Sri Lanka, 28t, 28­30 Jain Hawala scandal, 54n welfare funds for migrants, 26, 30n Kashmir, 47 working condition of Kuoni Travels (India), 37 migrants, 24 Kuwait, 24 "money bazaars," 43 money exchange companies, 32 land MoneyGram, 36 and in-kind remittances, 43 money laundering, 53 as remittance investment in moneylenders, 5 rural Pakistan, 17, 57 money orders, 38b INDEX 81 money service businesses (MSBs), People's Bank (Sri Lanka), 32 xiii, 36­37 Persian Gulf. See Middle East phei kwan (Thailand), 43 National Employment Policy See also hawala (Sri Lanka), 30 Philippines, 43 Nationwide Funds Transfer post office networks, xiv, 36­39, System (NEFT, India), 35 38b, 60 Nepal, 13t, 13­14 Post Office Savings Bank (POSB) nonresident foreign-currency (India), 38b accounts (NRFC, Sri Lanka), poverty 29­30 See also development impact of remittances overseas development assistance growth-poverty model, 10­12 (ODA) trends, 2­4 new dataset on international remittances and, 7­10, 9t, padala (Philippines), 43 63­71 See also hawala official and unofficial Pakistan remittances and, 10­12, 11t, ATM machines, 35 14, 15t comparing official and unofficial poverty gap, 9, 9t, 12, 14, 63­71 remittances, 13t, 13­14 premigration assistance courier services, 43 Bangladesh, 22 development impact of India, 23­24 remittances, 4 Pakistan, 26 foreign commercial banks, 33 Sri Lanka, 28 hawala, 13. See also hawala private commercial banks, 33­36 informal financial sector See also formal financial infrastructure, 6, 37, 47, 48b, infrastructure for remittances 48­49 profits on financial services and investment by migrants from, products, 33 17, 56, 57 promissory notes, 46 local commercial banks, 34 Protectors of Emigrants (India), 23 nonresident Pakistanis public infrastructure for (NRPs), 27 remittances, xi­xii, 21­30 overseas development assistance efficiency and effectiveness of, 5 (ODA) trends, 2­3, 3f recommendations, 58­59 public infrastructure for public-private partnerships, 60 remittances, 25­27, 26t, 58 Punjab National Bank (India), 37 rural households, study of, 16­17 real estate market, 17, 43, 57 payment card infrastructure, recommendations, 56­62 34­36, 35t enhancing development impact payment system technology, 40 of remittances, 56­58 82 INDEX formal financial infrastructure, volume of remittances, 1 59­61 Southeast Asia informal financial sector domestic workers in, 24, 28 infrastructure, 61­62 Special Electronic Funds Transfer public infrastructure, 58­59 System (SEFT, India), 35 regional disparities in financial speed of informal financial sector services, 40 infrastructure, 49, 49t, 51 registration of hawala dealers, 54, 61 squared poverty gap index, 9, 9t, See also hawala 63­71 Remit2India, 37 Sri Lanka remittance business model, 33 ATM machines, 32, 34­35 remittance modes, 32 comparing official and unofficial Reserve Bank of India, 37 remittances, 13t, 13­14 rural communities foreign commercial banks, 33 outreach in, 39 hawala. See hawala rural financial systems, 6, 16­17, informal financial sector 31­32, 36, 40 infrastructure, 6, 37, 47 post office networks, 37­38, 38b Internet services, 34, 37 local commercial banks, 34 shared payments systems money service businesses, 37 platforms, 60 overseas development assistance shopkeepers and informal (ODA) trends, 2­3, 3f remittance system, 5 post office network, 38, 60 smuggling, 53 public infrastructure for Sonali Bank (Bangladesh), 32 remittances, 28t, 28­30, 58 South Asia region state commercial banks, 32 See also migration from South Sri Lanka Bureau of Foreign Asia region; specific country Employment (SLBFE), 28­29 comparison of official and State Bank of India (SBI) unofficial remittances, 13t, accounts, 32 13­14 State Bank of Pakistan (SBP), 35 dataset on. See dataset on state banks, xii, 5, 29, 31­32, 59, 60 international remittances and Sudan, 43 poverty econometric results, 10­12, 11t taxation estimating unofficial remit- income tax evasion, 52­53 tances, 12­14, 16 informal financial sector hawala. See hawala infrastructure, 47, 50 historical background of Pakistan, remittances not subject informal remittance systems, to, 27 46­47 technology, effects of, 34, 37, 60­61 poverty, 9, 9t See also specific method or service use of remittances, study of, telegraphic money transfers, 31 16­17 See also Western Union INDEX 83 telex transfer, 41 "underground banking." See terrorist financing, 53 hawala Times of India, 37 TimesofMoney.com (India), 37 very small aperture terminals trade and remittance business, 6, (VSAT) connections, 38b 46­47 volume of international trafficking, 53 remittances, 1, 2t training of Sri Lankan migrants, 29 transaction costs. See costs Wage Earners Bonds transfer agents, competition (Bangladesh), 23 among, 33, 59­60 Wall Street Finance Limited transparency of informal (India), 37 remittance industry, 6, 54, welfare funds for migrants, 26, 30n 61­62 Western Union, 36­37, 38b travel agents and remittance wire transfers, 32 business, 37, 43 working condition of migrants, 24 trust and informal financial sector World Development Report, 2000/01 infrastructure, 48 (World Bank), 8 South Asia has experienced a sharp increase in remittances since the mid-1990s. According to recent World Bank research, Bangladesh, India, Pakistan, and Sri Lanka are all among the world's top 20 receivers of remittances, with estimated receipts of US$3.2 billion, US$17.4 bil- lion, US$4.0 billion, and US$1.3 billion, respectively. Building on this research, Migrant Labor Remittances in South Asia focuses on the regional characteristics that distinguish South Asia from other remittance-active parts of the world: · A large migrant population of semi-skilled and unskilled workers largely concentrated in Persian Gulf countries · The presence of dedicated public institutions and government finan- cial incentives aimed at facilitating remittance flows and providing incentives for temporary migration · The existence of large state bank branch networks with immense potential to create more effective and efficient remittance financial markets · The widespread use of trade-related informal remittance channels by both legal and illegal migrants. The book discusses key issues regarding the development and imple- mentation of remittance industry policies, processes, and infrastructure in South Asia to foster development-oriented transfer of financial resources between migrants in developed economies and their families in the region. It will be of special use to financial sector policymakers and legislators, particularly those responsible for ensuring compliance with international money-transfer codes and standards, as well as to researchers and providers of remittance services. ISBN 0-8213-6183-X