63050 Transfonning Bangladesh into a Middle Income Economy Transfonning Bangladesh into a Middle Income Economy Edited by SADIQ AH:MED MACMILLAN © 2005 The International Bank for Reconstruction and Developmentffhe World Bank 1818 H Street, NW Washington, DC 20433 All rights reserved. No part of this publication may be reproduced or transmitted, in any form or by any means, without permission. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. First published, 2005 Reprinted, 2005 MACMILLAN INDIA LTD. Delhi Chennai Jaipur Mumbai Patna Banglore Bhopal Chandigarh Coimbatore Cuttack Guwahati Hubli Hyderabad Lucknow Madurai Nagpur Pune Raipur Thiruvananthapuram Visakhapatnam Companies and representatives throughout the world ISBN 1403 927847 Published by Rajiv Beri for Macmillan India Ltd. 211 0 Ansari Road, Daryaganj, New Delhi 110002 Printed at Rajkamal Electric Press B-35/9, G.T. Kamal Road, Industrial Area, Delhi 110033 The finding, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colours, denominations, and other infonnation shown on any map in this work do not imply and judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work 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 Developmentffhe World Bank encourages dissemination of its work and will nonnally grant permission 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, www.conyrir:ht.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. Preface Between 1972 and 2003, Bangladesh raised its dollar per capita income fourfold, reduced poverty by more than a third, increased life expectancy by more than 40 per cent, and enhanced gross primary enrolment by over 80 per cent. These remarkable development achievements compare favourably with outcomes in less developing countries. This progress, fighting against many odds in the form of frequent natural calamities and generally poor governance, is a testimony to the resilience and determination of a dynamic young nation and gives hope that with continued determined efforts Bangladeshis can look forward to further gains with development. The challenges in this twenty-first century are enormous, notwithstanding past progress. The past improvements in performance were obtained from a very low base in 1972 when Bangladesh gained independence. Thus. with a 400 dollars per capita income, Bangladesh is still amongst the poorest country in the world. While the recent average per capita growth of 3.7 per cent is encouraging, even if this were raised to 6 per cent, a difficult challenge. it will take over 40 years of growth at this pace simply to reach the current per capita income level in Malaysia. Nevertheless, Bangladesh can and should aspire to become a middle income nation (using a cut-off level of dollar 1,000 per capita) over the next 20 years. This requires an overall annual growth rate of around 6 per cent, implying per capita growth rate of around 4.7 per cent. Along with this focus on growth. effort must also continue to ensure that the benefits of growth are more widely shared. While Bangladesh has done well in raising per capita income and improving human development, measured against comparable per capita income countries, its long term performance on poverty reduction is relatively weaker. What are the policy and institutional reform challenges for higher growth rates? What additional reforms are needed to increase the impact of growth on poverty? The answers to these questions require analysing and understanding past development performance, to look at what worked and what did not work and why. The objective of this book is to contribute to this analysis and understanding of policies and institutions in Bangladesh. The book vi Preface draws mostly on research done by current and former World Bank staff working in the South Asia Region. The only exception is the paper written DY Syed Ahsan of the University of Concordia in Canada. The book, a collected volume of papers, is organised thematically to maintain a logical flow. It is organised in four parts. The first part gives a summary of past development performance to provide the context for the detailed analysis in subsequent sections. Thus Part I consists of two papers. The starting paper by Sadiq Ahmed summarises the development achievements of Bangladesh since independence in 1972, identifies in a summary form the main policies and institutional reforms that helped Bangladesh to achieve these outcomes, explains why progress was difficult in a number of areas relating to institutions, and highlights the key challenges moving forward. The paper notes that good progress was made in achieving and preserving a stable macroeconomic environment; initiating the first round structural reforms in areas of trade, agriculture and banking; and in pursuing good human development policies, especially through budgetary allocations. Progress in the area of infrastructure and in improving institutions was weaker. The general governance environment has also become more difficult because of inadequate institutional reforms. Attention must shift to these areas for increasing growth and reducing poverty. The second paper by Rinku Murgai and Salman Zaidi looks more deeply at the poverty profile and progress during the 1990s. The paper notes the good progress with poverty reduction during the 1990s, following the stagnation in the mid-1980s. The paper, however, also shows that inequality increased during the 1990s. Finally, the paper draws attention to the regional dimensions of poverty, noting that the incidence of poverty varies from a relatively low level of40 per cent in Barisal to a high of 60 per cent in Rajshahi. Part II goes deeper into the analysis of GDP growth with a view to identifying ways to accelerate this growth. This consists of five papers. The section starts with a paper by Sandeep Mahajan that looks at the determinants of growth in Bangladesh and identifies associated constraints. The paper observes that productivity improvement has been a key driver of per capita GDP growth in Bangladesh, supported by policy reforms. Thus, policies contributed by maintaining macroeconomic stability, obtaining a sharp decline in population growth, and supporting higher rates of saving and investmen~. So, the potential for higher growth lies in further policy and institutional reforms. The next paper by Sadiq Ahmed and Zaidi Sattar looks carefully at the role of trade openness in stimulating growth and reducing poverty. This subject has been a source of considerable debate internationally and in Bangladesh. The paper's main conclusion is that trade openness has been good for growth and poverty reduction. It shows the progress on trade reforms and identifies the agenda for future reforms. Preface vII The third paper by Syed Mahmood deals with the all important issue of investment climate for private enterprise. International experience suggests that domestic investment climate can be a serious bottleneck for private investment and growth. The paper looks at the overall investment climate in Bangladesh as compared with competitors, and finds that there are indeed some serious constraints in terms of weak incentive and high cost of doing business that must be addressed in order to spur private investment and growth. Infrastructure constraints in particular are severe. The next paper by Sadiq Ahmed reviews progress with banking reforms and the challenges moving forward. International experience shows that financial depth and efficiency are important for the growth and competitiveness of the private sector. The paper notes that banking sector has traditionally been very inefficient but recent reforms have considerably improved performance of private banks, which have grown in relative importance as well. However, progress with reforming the public banks has been weak. Although the public banks are no longer dominant, their adverse effects on the banking sector performance remains substantial and must be addressed to put Bangladesh on a stronger growth path. The last paper in this part looks at the performance of state owned enterprises (SOEs). The paper here, written by Syed Nizamuddin, concludes that overall the SOEs have been a big drag on the economy. While some progress has been made in the recent years, the reform agenda remains substantial. Further progress in reforming SOEs will be necessary to consolidate the gains in macroeconomic stability, by reducing the adverse effects ofquasi-fiscal deficits and contingent liabilities, and for improving the investment climate for private enterprises. Moving on to Part III, the theme here is how to make growth work better for the poor. While over the longer term higher growth is necessary for faster progress with poverty reduction, the distribution of benefits of growth to the poor also matters. Part III has five papers, starting with a paper by Rinku Murgai and Salman Zaidi that analyses the main sources of assets and income for the poor based on the 2000 Household Income and Expenditure (HIES) data. The paper concludes that the poor derive their income from a variety of sources including agriculture and non-farm sources. It also notes that education, family assets, land ownership and availability of infrastructure are all correlated with poverty. In the following paper, Ndiame Diop reviews the role of agriculture in rural poverty reduction and suggests ways to boost farm productivity and income. The next paper by Forhad Shilpi takes up the subject of non-farm enterprises, identifies past performance and constraints, and suggests policy options for improving the investment climate for rural enterprises. The paper notes that non-farm enterprises have grown in importance in terms of both vIII Preface employment and income. Yet, the investment climate facing these enterprises can be much improved by enhancing infrastructure and financial services, by improving disaster management, and by improving law and order. The fourth paper in this part, written by Hassan Zaman, deals with the subject of finance for the poor. Bangladesh is justly famous for its innovative development of a variety of microcredit schemes. These schemes now benefit some 13 million poor households. What factors contributed to this success and what are the emerging challenges facing the industry? The paper describes the evolution of the micro finance industry in Bangladesh, identifies factors that contributed to the scale up progress, summarises evidence concerning the impact on poverty, and identifies key lessons and challenges for the future. The final paper in this part by Anil Deolalikar concerns the human development dimensions of the MDGs. Bangladesh has performed well by international standards in improving human development indicators. Yet many challenges remain. What are the prospects that Bangladesh will attain by the year 2015 the MDG goals related to human development? The paper looks quantitatively at how likely it is for Bangladesh to achieve MDG targets in the four areas of under-five mortality, child malnutrition, schooling enrolment and completion, and gender disparities, and what will it take to secure these MDGs. Part IV of the book picks up the theme of reforming public sector to promote growth and reduce poverty. A review of past experience shows that progress with reforming the public sector has advanced the least. Indeed, corruption and weak public sector are the most difficult development challenges facing Bangladesh today. Needless to say, these are interrelated and the corruption problem will be very difficult to tackle comprehensively without reforming the public sector. The required reforms are long term and progress will be slow, but certainly more progress is achievable with strong leadership. This part has four papers, starting with the one by Syed Ahsan that provides an analysis of the nature of the governance challenge in Bangladesh. The paper also identifies options that will likely help improve governance in the country. The paper correctly notes the political nature of the agenda and observes that associated reforms will need to encompass a wide variety of institutions. The next paper by Sadiq Ahmed focuses on arguably the most important public policy instrument: the role of the budget for growth and poverty reduction. The paper notes that while by and large Bangladesh pursued prudent fiscal management and good overall expenditure allocations, the budget management has fallen short in resource mobilisation, especially tax administration. The budget management is also relatively weak in terms of expenditure planning, assignment ofresponsibilities by levels ofgovernment, Preface Ix monitoring and evaluation, and fmancial controls and accountability. While some initial progress has been achieved very recently, these are areas where much more effort will be needed. In the following paper, William McCarten provides a detailed review of tax performance and suggests ways to improve this performance. The paper argues that a sustained programme oftax reforms, made up ofa set offeasible measures, would help Bangladesh improve efficiency, fairness and transparency of the tax system while also raising resources to finance development. The fmal paper by Elena Glinskaya, Eric Bell and Zahid Hussain looks more deeply at public spending issues. The paper notes the many positive features of expenditure management in Bangladesh in terms of its positive contribution to human development but also notes the scope for further improvement, especially as regards strengthening the institutional dimensions of expenditure management. Several people made this task possible. First and foremost, I will like to express my appreciation to all the authors who contributed to the book. In many cases, the work required time offfrom regular duties and accountabilities at the workplace. This was done cheerfully and on time with minimum prodding. Second, I received very competent research and processing support from Bala Bhaskar Naidu and Joyce Mormita Das, respectively. Both colleagues went out oftheir line ofduty to be helpful and patient in supporting the many requests I made in putting the papers together, checking on data and references, formatting the papers for consistency, and providing other support as called for. My special thanks to both. Finally, it would have been impossible to pull this work together without the understanding and support of my immediate family members. To Nusrat Ahmed, my wife, and our children Asif, Nabila and Saqib, lowe the deepest gratitude for letting me spend long hours on weekends and often weeknights for labouring on this task. The World Bank SADIQ AHMED Washington, DC Contributors Sadiq Ahmed is the Director of Poverty Reduction and Economic Management in the South Asia Region of the World Bank. Syed M. Ahsan is Professor and Head of Department at the Concordia University, Canada. Eric Bell is a Lead Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region of the World Bank. He is currently on leave of absence with the Australian Aid Agency in Canberra. Anil Deolalikar, formerly a Lead Economist in the Human Development Unit of the South Asia Region, is now Professor and Head of Department of Economics at the University of California, Riverside. Ndiame Diop is an Economist in the Development Research Group, World Bank. Elena Glinskaya is a Senior Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region, World Bank. Zahid Hussain is a Senior Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region, World Bank. Syed A. Mahmood is a Lead Economist in the Finance and Private Sector Unit of the South Asia Region, World Bank. William J. McCarten is a Senior Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region, World Bank. Rinku Murgai is a Senior Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region, World Bank. Syed Nizamuddin is a Senior Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region, World Bank. Zaidi Sattar is a Senior Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region, World Bank. Forhad Shilpi is a Senior Economist with the Agriculture and Rural Development Unit of the South Asia Region, World Bank. Hassan Zaman is a Senior Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region, World Bank. Salman Zaidi is a Senior Economist with the Poverty Reduction and Economic Management Unit of the South Asia Region, World Bank. Contents Preface v Contributors xi Part I Development Performance and Challenges: An Overview 1. Development Performance and Policy Reforms since Independence 3 Sadiq Ahmed 2. Poverty Trends During the Nineties 48 Rinku Murgai and Salman Zaidi Part II The Strategy for Accelerating Growth 3. Analysis of Growth Experience 91 Sandeep Mahajan 4. Trade Openness and Growth l30 Sadiq Ahmed and Zaidi Sattar 5. Investment Climate for Private Investment 166 Syed A. Mahmood 6. Banking Sector Performance and Reforms 201 Sadiq Ahmed 7. Reforming State Owned Enterprises 217 Syed Nizamuddin Part III Making Growth Work for the Poor 8. Main Income Sources and Assets of the Poor 243 Rinku Murgai and Salman Zaidi 9. Agriculture Development Strategy 261 Ndiame Diop 10. Improving the Rural Investment Climate for Non-Farm Enterprises 281 Forhad Shilpi xiv Contents 11. Scaling Up ofMicro finance 315 Hassan Zaman 12. Human Development Strategy to Achieve MDGs 334 Ani! Deolalikar Part IV Reforming Public Sector to Promote Growth and Poverty Reduction 13. Governance and Institutions 373 Syed M Ahsan 14. Managing the Budget for Growth and Poverty Reduction 413 Sadiq Ahmed 15. ReformingtheTaxAdministration 433 William. J. McCarten 16. Public Expenditure Management 464 Elena Glinskaya, Eric Bell, and Zahid Hussain Index 493 PART I Development Performance and Challenges: An Overview 1 Development Performance and Policy Reforms since Independence Sadiq Ahmed SECTION I: INTRODUCTION1 Background In the early 1970s, following independence, Bangladesh faced daunting development challenges. Per capita income was at a low of around 100 dollars, poverty level exceeded 70 per cent, schooling and health indicators were much below the average in South Asia and comparable to those found in Sub-Saharan Africa. Today, after 33 years of independence, the country can boast about some convincing progress with economic development (Table 1.1). Per capita income has expanded by four-fold, human development has progressed impressively, meeting or exceeding average levels in South Asia and much ahead of Sub-Saharan Africa, and the incidence of income/consumption poverty has been reduced to below 50 per cent. Yet, major challenges remain. At 400 dollar, per capita income is low even by South Asian standards; some 53 per cent of the rural population and 37 per cent of the urban population are still poor; and human development indicators are substantially below those in the East Asian economies. Indeed, without stronger efforts, Bangladesh may not be able to achieve all the Millennium Development Goals (MDGs), especially for reduction in income poverty.2 Importantly, against the backdrop of these 1 The paper draws extensively from an earlier paper titled 'The Political Economy of Development Experience in Bangladesh' presented to the BER-World Bank Conference in Dhaka, June 2003. 2 The millennium development goals or MDGs are defined in UN 2000. Table 1.1 Selected Development Indicators 1970 2002 Sub-Saharan East Asia Sub-Saharan East Asia Bangladesh South Asia Africa and Pacific Bangladesh South Asia Africa and Pacific GNP per capita (US $, current) 120 140 270 160 380 460 450 960 GOP per capita growth (%) (1973-2002) 2.2 2.8 -0.4 5.6 Life expectancy at birth, total (years) 44 49 44 59 62 63 46 69 Mortality rate, infant (per 1,000 live births) 145 129 141 85 48 68 103 32 School enrolment, primary (% gross) 54 71 51 89 98 a 95 b 87 a lllb School enrolment, secondary (% gross)19 23 6 24 47 a 48 b 26 c 66 b Notes: a. 2001; b. 2000; c. 1996. Source: World Bank, World Development Indicators Database, April 2004. Development Performance and Policy Reforms since Independence 5 challenges, major governance constraints have emerged which might threaten the sustainability of the past progress. Certain developments in the external environment, such as the phase-out of the Multi-Fibre Agreement (MFA) and surge in oil prices, also present significant downside risks. The development challenges for Bangladesh in the new millennium are indeed daunting, but by no means impossible. As the nation has shown, rising from the ruins of a war-devastated economy in the 1970s, with concerted efforts constraints can be overcome and progress with development is very much possible. What factors have contributed to the gains in development progress achieved by Bangladesh since its independence? What factors constrain progress in the twenty-fIrSt century? What will it take for Bangladesh to achieve the millennium development goals and become a middle income country? These are important questions and informed debate will be very useful for the policymakers. Past progress suggests that there have been many positive experiences with policy reforms. The paper will seek to show that much of these reforms have been in areas of relatively easy options where social and political constraints were not binding. Deeper reforms, which involve institutional change and matured political leadership, have been much more hesitant. In some cases, even when some of these reforms were agreed under donor pressure, implementation suffered from lack of political ownership and commitment. Indeed, it is now commonly acknowledged that the fundamental development constraint in Bangladesh is poor governance. 3 The governance problem is not unique to Bangladesh, but certain aspects such as corruption and law and order, unless addressed comprehensively, could substantially slow down investment and growth. Similarly, unless major institutional reforms aimed at improving public service delivery happen, the effectiveness of public expenditures to improve human development will be seriously impaired. Even so, reforms that were actually implemented yielded encouraging results. These responses have benefited from the dynamism of the people. Thus, while public entities failed to deliver good outcomes, private initiatives, including from the non-governmental organisations (NGOs), performed very well in rice agriculture, garments, human development and other services. Consequently, economic outcomes are better than what one would have expected from the generally weak governance. Thus, Bangladesh sustained an average pace of 2.3 per cent per capita economic growth per annum during 1972-2004. Encouragingly, the growth path shows a significant upward trend, rising from a low of 1.3 per cent in the 3 For a detailed review of the relationship between poverty reduction and governance in South Asia, see Ahmed 2002. 6 Transforming Bangladesh into a Middle Income Economy 1970s to 3.7 per cent in the 2001-04 period (Figure 1.1).4 These growth rates compare favourably with the average growth rate found in South Asia and low income developing economies. They also attest to the quality of the people and by implication to the missed opportunities for deeper reforms and institutional change. Thus, with matured political leadership leading to stronger reforms and improved public administration, Bangladesh can achieve much higher rates of economic growth, something comparable with rates achieved in East Asian economies. Along with policies to promote equity, it should be possible for Bangladesh to sharply reduce poverty and achieve the MOGs. 4 o 1973-80 1980-90 1990-2000 2000-04 TIme Period Figure 1.1 Bangladesh Per Capita GOP Growth, 1973-2004 SECTION II: DEVELOPMENT PERFORMANCE OVER THE PAST THREE DECADES Initial Situation At the time of independence, Bangladesh inherited a war-ravaged economy. Per capita income stood at less than 100 dollars and the incidence of poverty was large, estimated at over 70 per cent. 5 Human development indicators were very low and the anti-female bias was profound. Clearly, Bangladesh faced major development challenges in the early 1970s in terms of a large incidence of poverty as well as the additional burden of rebuilding an economy shattered by the fight for independence. How did it perform over the past three decades or so? This section will review this performance record. 4 Per capita GDP data is based on revised national accounts data prepared by the Bangladesh Bureau of Statistics. 5 The 70 per cent headcount poverty index for Bangladesh is for 1974 and is based on consumption data, see Sen 2000. Development Performance and Policy Reforms since Independence 7 Progress with Poverty Reduction and Human Development Arguably, Bangladesh is the most vulnerable ofthe South Asian economies in view of its extremely high population density (the highest in the world excluding Singapore) and the high incidence of natural disasters. Poverty incidence data compiled by World Bank staff and national researchers show that there was a substantial deterioration in the poverty situation in the 1970s, followed by rapid progress during 1978-86 (Figure 1.2A, Series A). 6 There was again a worsening of poverty in the 1986-92 period (Figure 1.2B, Series B). More recently (I992-2000), there has been an improvement in the poverty reduction effort (Figure 1.2B, Series B). Given the high concentration of the poor in the rural areas, the national poverty outcome is dominated by the poverty outcome in the rural areas. 7 100 l 80 !II ]l 60 C 40 § 'C 01 20 III :I: 0 1974 1977 1978 1979 1982 1984 1986 1989 Years Aural Urban National Figure 1.2A Bangladesh Poverty Trend Series A 1964 1986 1989 1992 1996 2000 Years -+- Aural - - Urban ---dr- National Figure 1.28 Bangladesh Poverty Trend Series B 6 Poverty numbers and data sources are provided in AppendicesTables Al.I and A1.2. 7 Discussions of the Bangladesh poverty estimates and issues are contained in Sen 2000, World Bank 1999 and World Bank 2002. A detailed analysis of poverty trends and profile during the 1990s is contained in an accompanying paper in this volume by Rinku Murgai and Salman Zaidi. 8 Transforming Bangladesh into a Middle Income Economy Although the two time series are not comparable, because of definitional and other measurement differences, some broad conclusions can be drawn. 8 First, overall, Bangladesh has made an impressive reduction in poverty since independence. Both the rural poor and the urban poor have benefited from this effort. Second, the reduction in poverty seems to have slowed down since the mid-1980s, due to limited progress with reduction of rural poverty. 1bis conclusion is still valid if one were to exclude the controversial poverty estimate for 1986 that shows a sharp reduction from 1984. Sen (2000) explains the nature of this controversy. Appendix Table A 1.2 shows that the important point is between 1984 and 1996, the incidence of rural poverty seems to have fluctuated between 50 and 60 per cent. Third, urban poverty has fallen faster, even though the pattern reversed between 1996-2000 when urban poverty increased and rural poverty declined. Fourth, the sharp fluctuations in the pattern of poverty outcomes suggest that many of the poor are positioned around the poverty line, leading to vulnerability to shocks especially in rural areas. Table 1.2 Bangladesh Progress with Human Development 1975-2002 Years Life Infant Total Access to Adult Primary Secondary expectancy mortality fertility drinking literary school school (years) rate (per rate water rate (%) enrol- enrol­ tODD) (%) (%) of ment (%) ment (%) population 1975-n 47 130 6.5 35 26 83 14 1985-88 51 118 5.5 46 33 59 17 1997--02 62 48 3.0 97 41 98 47 Source: World Bank 2000; World Development Indicators Database, April 2004. The picture regarding progress with human development (Table 1.2) shows that since independence Bangladesh made rapid strides from the low base levels in improving primary and secondary school coverage, reducing the population growth rate, improving access to drinking water, reducing infant mortality and raising life expectancy. Indeed, the progress on human development is more impressive than the progress in reducing income poverty and the social indicators are better than average for comparable per capita income countries. The performance in the area of popUlation management is truly remarkable, resulting in a rapid reduction in the rate of population growth from a high of 2.8 per cent per annum in the 1970s to 1.6 per cent per annum in the 1990s. The total fertility rate 8 Conclusions on poverty trends and directions are not affected by definitional and measurement issues. although levels are affected, see Sen 2000. Development Performance and Policy Reforms since Independence 9 declined from a high of over 6.5 per woman in the mid-1970s to 3.0 by 2001. Similarly, the ability to reduce infant mortality from a high of 130 in the mid-1970s to 48 by 2002 is a truly commendable achievement. The remarkable progress achieved by Bangladesh on human development was recently reviewed by Jean Dreze, who compared this performance with India and noted: 'Some of these estimates may not be accurate. Perhaps the ranking would be reversed, in some cases, if exact figures were available. But the general pattern, whereby Bangladesh is now doing better than India in terms of many aspects of social development, is unlikely to reflect measurement errors. The pattern is still the more striking as India used to fare better than Bangladesh in all these respects not so long ago in the early 1970s, when Bangladesh became independent.'9 Nevertheless, despite this progress, human development remains a major challenge. Nutrition is a serious problem, with over 60 per cent of the children suffering from malnutrition as compared with 49 per cent for the region and only 11 per cent for lower middle income countries. 10 While access to drinking water has increased, there are serious concerns about its safety due to contamination by arsenic poison. Only 44 per cent of the rural population has access to reasonable sanitation facilities. The adult literacy rate is still very low and secondary school enrolment is still below 50 per cent. Importantly, there are major concerns about the quality of education. 11 Progress with the Enabling Environment It is universally acknowledged that long term economic growth is necessary for poverty reduction. 12 The intuition underlying this conclusion is easy to see. Income growth creates the basis for providing productive employment, supports increases in real wages, and helps fmance public programmes for poverty reduction. However, it is also well recognised that the pattern of economic growth matters in poverty reduction. In the Bangladesh context, given the strong concentration of the poor in the rural areas and being a labour abundant country, the overall pace of economic growth, the performance of agriCUlture, and the overall labour intensity of production all matter for reducing income poverty.13 9 Dreze 2004. 10 See World Bank 1999a. 11 See World Bank 1999b. 12 WorldBank 2000; Dollar and Kraay 2000. 13 The paper bySandeep Mahajan in this volume looks in greater detail on the determinants of growth in Bangladesh. 10 Transforming Bangladesh into a Middle Income Economy What are the determinants of rapid, sustainable economic growth? There is a rich body of empirical literature that has analysed the determinants of economic growth. The role of physical and human capital and technology are generally well recognised. More recently, the importance of governance in affecting growth has been noted and emphasised. Governance affects the investment climate, which in turn is a determinant of the level and quality of private investment. The definition and measurement of governance, however, can take a variety of forms and there is no commonly used methodology. The main governance factors most commonly referred to by private investors as a deterrent to private investment in their response to survey questions are law and order, weak infrastructure, bureaucratic hassles and bribery,I4 Table 1.3 Bangladesh Progress with the Enabling Environment, 1973-2003 Indicators 1973 Growth of GOP (% p.a.) 4.6 Growth of per capita GOP (% p.a.) 2.3 GNP per capita ($) 120.0 380.0 Growth of agricultural GOP (% p.a.) 2.4 Investment rate (% of GOP) 8.7 23.2 National saving rate (% of GOP) 4.2 23.7 Domestic saving rate (% of GOP) 3.4 18.2 Extemal trade (% of GOP) 18.4 32.1 Total debt/GOP (%) 10.7 16 51.7 Inflation rate (% p.a.) 10.5 Source: Data for 1973 obtained from World Bank 1993. Data for 2003 obtained from World Bank 2003 and IMF staff reports. Table 1.3 shows the progress with the enabling environment for growth and poverty reduction in terms of a number of key economic variables that affect the pace, pattern and macroeconomic sustainability of growth. Governance issues are taken up later. As we saw earlier, Bangladesh made good ovel'aliiong term progress in increasing per capita income and that this performance compares favourably with the average per capita GDP growth in South Asia and low-income countries during the 1990s. The analysis of Table 1.3 suggests that the growth performance was based 14 World Bank: 2003. The link between governance and investment climate is reviewed in the paper on Investment Climate by Syed Akhtar Mahmood in this volume. 15 Growth rates refer to changes over the two terminal years. 16 Refers to 1975. Development Performance and Policy Reforms since Independence 11 on an expansion in the rate ofinvestment and supported by greater openness to international trade. 17 Exports grew faster than GOP, fueled by a rapid increase in labour­ intensive garment exports. Rapid expansion of food production also contributed significantly to higher growth. Importantly, much of this expansion in investment was fmanced by a rise in domestic and national saving rates. IS Following a brief period of serious macroeconomic instability reflected in high inflation during the early years of independence, Bangladesh was basically able to maintain good macroeconomic management until 1999. During 1999-2001, there was a significant slackening of macroeconomic management, especially fiscal policy. Fortunately, policy actions since then have helped correct this weakening. Macroeconomic management is now back on track. As against these positive developments, negative factors have been on an average a relatively slow pace of expansion in agricultural value­ added (notwithstanding the contribution of food production), an average long term inflation rate that has exceeded the international inflation rate (although the inflation rate was much lower in the 1990s and now in the decade ofthe 2000), and the rapidly increasing debt/GOP ratio. Even though the debt-servicing burden is low due to the high degree of concessionality of loans from official sources, this cannot be assured for the future due to the changing global environment for foreign aid and capital flows. Indeed, during the late 1990s Bangladesh increased exposure to more expensive suppliers credit to finance infrastructure and other development projects. 19 An added concern is the vulnerability of the Bangladeshi economy to the phase out of the MFA on 1 January 2005. Notwithstanding progress with trade reforms and the associated expansion of trade, Bangladesh exports remain narrowly focused on garments and a few other export items. Clearly much more needs to be done to diversify the export base. An important question is how conducive is the investment climate in Bangladesh and in particular what has been the impact of governance on the investment climate? A recent survey of investment climate suggests that access to efficient and reliable infrastructure (power, ports, telecoms, transport) is a major constraint to private investment.20 Importantly, there 17 A detailed empirical analysis of the implications of trade liberalisation and economic deregulation for growth and poverty reduction in Bangladesh is contained in another paper in this volume by Ahmed and Sattar. 18 The saving rate for 1973 is likely to be an underestimate and not directly comparable with data for the 1990s that are ba"ed on revised national accounts. Nevertheless, there is little doubt that national saving rate has increased substantially since independence. 19 World Bank 2000b. 20 World Bank. 2003a. Also the paper by Syed Mahmood in this volume. 12 Transforming Bangladesh into a Middle Income Economy is empirical evidence that poor governance has been an important constraint to private investment and growth in Bangladesh. The most signifiq.nt problem has been pervasive corruption. Using a cross-country econometric model, one study has concluded that corruption significantly reduces the growth of per capita GDP in Bangladesh.21 Another study has listed the large number of hidden costs (illegal payments), which a private investor has to incurto do business in Bangladesh,22 raising substantially the financial cost of doing business and weakening incentives. The same study also contains a long list of complaints and concerns expressed by the business community relating to over-regulation and bureaucratic hassles that also reduce the incentives for investment. An additional factor adversely affecting business incentives is the frequency of political turmoil and hartals (economywide strikes causing work stoppages). For example, in 1999 hartals disrupted work for as many as 30 working days. Over the period 1995-2000, the total number of workdays lost from hartals is estimated at about 200 days.23 Overall, the ability of Bangladesh to achieve a per capita income growth rate of 2.2 per cent per annum over the past thirty years and now growing at 3.7 per cent per annum (which is a better perfonuance than the average for all low income economies), despite these significant governance problems, is a testimony to the dynamism of its people. Although the public sector and non-food agriculture have been drag on the economy, a vibrant private sector has emerged. Thus, during the 1990s, the garment exports and private services, particularly construction sector, provided the impetus to this growth of income. Nevertheless, the missed opportunities and waste due to corruption and other governance problems had a tremendous cost in tenus of significantly lowering growth from its potential level. For example, the expansion in per capita income in Bangladesh in the 1980s and 1990s (3.0 per cent) still lagged behind the average growth of per capita income in India (4.5 per cent) and substantially behind the East Asian economies (around 6 per cent per annum). Not surprisingly, there has been much better progress with poverty reduction in these economies as compared with Bangladesh. Growth and Equity The aggregate results, however, hide some of the dynamics of the growth experience in Bangladesh and its implications for income distribution and poverty. A periodic breakdown of the experience is important to understand 21 World Bank 2000c. 22 World Bank 1996. 23 World Bank 2000c. Development Performance and Policy Reforms since Independence 13 why poverty reduction progress has slowed down since the mid-1980s. The pace and pattern of growth, corresponding to the various poverty progress periods, is shown in Table 1.4. A number of important points emerge from this breakdown. First, the early years of independence (1972­ 77) that saw a considerable increase in the incidence of poverty also witnessed slow overall per capita income growth and slow growth of agricultural income. In the second phase (1977-86), when a sharp reduction in the incidence of poverty in both rural and urban areas happened, overall per capita growth improved, notably, the pace of growth in agricultural value-added increased, and there was a rapid expansion in the exports of garments. In the third phase (1986-92), rural poverty increased despite an increase in the rate of overall per capita GOP. This happened partly due to a significant reduction in the pace of expansion of agricultural incomes. Urban poverty, however, continued to show a decline, benefiting from the continued good rate of expansion of urban incomes and the contribution of buoyant garment exports. Finally, in the fourth phase (1992-2000), poverty declined in both rural and urban areas, supported by an increase in the rate of growth of overall per capita GOP, an increase in rate of growth of agricultural GOP, the continued strong contribution of garment exports and expansion of non-farm activities. 24 Overall, on a long tenn basis, growth and poverty are negatively correlated suggesting that the upward movement of the growth rate of the last few years (2000-04) would likely lower poverty further in 2004. In particular, the pattern of growth is also favourable during the 2000 decade so far, with agriculture, garments and construction all showing solid growth rates. The breakdown of growth also shows a very important result. The impact of per capita income growth on poverty reduction has varied over time. Table 1.5 shows this result. The responsiveness of poverty to growth slackened considerably in the 1984-96 period. 25 Overall, the responsiveness of poverty reduction to growth has fallen sharply since the mid-1980s. What explains this? Part of the answer is the differential pattern of growth whereby the participation of the poor in the growth process has varied over time. Also, the poverty responsiveness to growth will vary over time 24 Salman Zaidi and Rinku Murgai show in an accompanying paper in this volume the sources of income for the poor. Evidence suggests that expansion of household income from agriculture and non-farm rural enterprises was an important factor for poverty reduction in the 199Os. 25 As noted earlier, there is some debate about the reliability of the 1986 poverty estimate that shows a marked reduction in poverty between 1984 and 1986. So, the 1986 poverty estimate is excluded from the calculation of growth responsiveness. Indeed, if one were to use 1986 as the base point for measuring poverty reduction since then, the outcome is one of negative progress with poverty reduction in the rural areas. 14 Transforming Bangladesh into a Middle Income Economy Table 1.4 Bangladesh Growth Pattern, 1973-2004 (% p.a.) Growth pattern 1973-77 1977-86 1986-92 1992-2000 2000-04 Per capita GOP 0.9 1.7 1.6 3.2 3.7 Agriculture 1.0 4.4 1.8 2.4 2.8 Garment exports 210 26 20 15 15 (current $ ) Construction 2.0 6.5 5.5 8.3 8.4 Source: Data obtained from various World Bank country reports and from World Bank 1993. Table 1.5 Bangladesh Poverty Responsiveness to Growth27 Years Per capita GDP Reduction Poverty responsive­ growth rate in poverty ness (2/1) (%p.a.) (%p.a.) 1977-84 2.5 7.4 3.0 1974-92 1.7 1.9 1.1 1984-96 2.4 1.0 0.4 1992-2000 3.1 4.3 1.4 Source: Growth data as in Table 1.4. Poverty data from Appendix Tables. based on the poverty line and the distribution around that poverty line. Nevertheless, evidence suggests that there has been rising inequality in both urban and rural areas in Bangladesh (see Figure 1.3). The incidence of inequality has been higher in urban areas than in rural areas. 28 [J National Gini Coefficient 0 Rural Gini Coefficient • Urban Gini Coefficient C 0.4 Q) '6 0.3 i: CI> 0 0.2 0 '2 0.1 a 0 1984 Years Figure 1.3 Bangladesh: Trend in Inequality 26 This high growth rate reflects an increase from a very low base of only dollar 3.0 million in June 1981. 27 Due to non-comparability of the various poverty series, the estimates of responsiveness are computed as follows. For 1977-84, series A in Appendix Table Al.l is used; for 1974-92, series B is used. For 1992-96 and 1984-96, poverty data in Appendix Table Al.2 is used. 28 Salman Zaidi and Rinku Murgai look in greater detail on this growth, inequality and poverty relationship in an accompanying paper in this volume. Development Performance and Policy Reforms since Independence 15 Participation and Social Development Achieving rapid economic growth and macroeconomic stability are important elements of a sound development strategy. Ensuring broad participation in the growth process is very important to ensure equitable distribution of this growth. The evidence of rising income inequality in Bangladesh is a matter of concern. Is rising inequality of income an indication of barriers to social mobility? Has the Bangladesh social and political environment evolved to allow broad-based participation in growth and development? One important element of this is the progress with improving the gender balance. In much of South Asia, women and girl child tend to fare quite badly relative to men and boy child. These are not only reflected in such indicators as access to education and income opportunities, but also in terms of basic indicators dealing with health and mortality. A particular strong indicator of the gender bias is the ratio of female to male. Dreze and Sen have argued that there is strong medical evidence that, given similar care, women tend to have lower age-specific mortality than men do. 29 So, a female to male ratio of below one is an indicator of the anti-female bias of the social policy. Summers also makes a strong case why investment in the girl child makes very good economic sense. 30 Bangladesh has made remarkable progress in reducing gender discrimination (Table 1.6). Apart from the rapid progress in increasing enrolment of girls in primary and secondary schools, reaching parity by 2001, the increase in the ratio of female to male from a pretty low level to around the same level as in advanced western countries is an indication of the strength of the progress in reducing the anti-female bias in the country. This is particularly notable because of the overall difficult social environment for women in South Asia. Table 1.6 Bangladesh Progress with Reduction in Anti-Female Bias Indicators 1972 1980 1990 2002 Female-male ratio 0.91 0.92 0.93 0.99 Female literacy (% of male) 35.6 43.1 53.4 62.4 Female primary school enrolment (% of male) 50.0 61.6 86.6 101.5 Female secondary school enrolment (% of male) 30.0 33.1 50.7 109.9 Source: World Bank 1997-2000; World Bank 2000a; Human Development Centre 1999. 29 Dreze and Sen 1995. See also Summers 1992. 30 Summers 1992. See also World Bank 200 la for evidence from international experience on the importance of gender balance for promoting development. 16 Transforming Bangladesh into a Middle Income Economy A second element affecting broad-based participation is ethnic balance. There are three main dimensions of this ethnic question: relationship with the Biharis, the relationship with the inhabitants of the Chittagong Hill Tracts (CRT), and relationship between Muslims and Hindus. During the early years of independence, the integration of the Bihari community with the mainstream Bengali community was a major issue. This, fortunately, proved to be a transitory problem. The immediate hardship and life threatening risks to the Bihari community after the independence of Bangladesh was brought to a reasonably quick resolution. 3l Those who chose to migrate to Pakistan were allowed to leave, subject to acceptance by Pakistan. Those who opted to stay gradually got assimilated in the main society. Episodes of Bengali-Bihari tension since the mid-1970s have been rare. Regarding the inhabitants of the CRT, there have been major difficulties in terms of conflict with the mainland settlers, leading to armed revolt by the CRT inhabitants. The government of Bangladesh had to deploy military force to contain this insurgency. In 1997, a peace accord was signed with the CRT inhabitants, which by and large seems to be holding. Another ethnic dimension relates to religion, particularly the relationship between the majority Muslims and the largest minority group of the Hindus. Again, following a large number of migrations to India during the war of independence, the general atmosphere has been one of peaceful coexistence. Reports of discriminatory policies and practices on grounds of religious beliefs are uncommon in Bangladesh. Overall, except in the case of the CRT problem, Bangladesh has been fortunate to avoid any prolonged period of ethnic conflict and tensions. This is a major social achievement in Bangladesh and successive governments should take credit for maintaining this harmony. In addition to investment in human capital, social capital mobilisation through the enabling role of the non-governmental organisations can be a major determinant of achieving broad-based participation. In Bangladesh, the strong role of the civil society has provided an important outlet for people's participation in economic and social decision-making. World-class NGOs, such as Grameen Bank, Brac and Proshika have provided opportunities for many ordinary citizens, who normally would have been by-passed by the government institutions, to participate in economic and social programmes thereby uplifting their economic status and contributing to the country's development. 32 3l There have been reports of loss of life and property immediately after 1972; also those who left for Pakistan may have lost property or other assets. See also, Khan 1983. 32 The dynamic role of these and some other NGOs are well-known internationally. Useful reviews of the positive role of NGOs are available in World Bank 1996a; World Bank 2000d; Khandker 1996 and Khandker and Pitt 1998. Development Performance and Policy Reforms since Independence 17 Progress with Management of the Natural Environment It is difficult to provide a quantitative framework of progress here. However, available evidence suggests that, by and large, progress has been limited. 33 In the area of disaster management, resulting from frequent floods and cyclones, there has been an improvement in providing cyclone precautionary information. Also, better shelter and other mitigating measures have helped reduce the loss of life. Better management of food distribution and provision of relief supplies has similarly reduced the loss of life from floods. Positive examples of this are the effective management of the 1998 and 2004 floods. In both instances, governments have been able to minimise losses of life and there were no reports of famine or health epidemics, unlike the situation in 1974. While this is a commendable achievement, progress with longer term flood control and mitigation measures has been more limited. 34 Other water related problems concern riverbank erosion and water drainage, which also remain serious problems. Generally, the vulnerability of the poor to the adverse effects of natural disaster still remains serious, especially in the rural areas due to loss of income from agriculture affected by the floods. An especially serious concern that has emerged in Bangladesh is water and air pollution. Water pollution is causing widespread illness and death due to waterborne diseases and toxicity. One study estimates that up to 80 per cent of all illnesses in Bangladesh are related to waterborne diseases. 35 The overall adverse health effects of poor water quality, sanitation and hygiene are estimated in the order of 3.6 million disability adjusted life years (DALY) with an associated financial cost of 2-4 per cent of GDP.36 More recently, arsenic poisoning of the groundwater has emerged as a serious health risk for Bangladesh. Air pollution has gotten worse over the years, especially in the urban areas. Available evidence shows that the air pollutant levels in Dhaka the capital city of Bangladesh - are substantially higher than the WHO standards. One estimate puts the loss of life from air pollution at 15,000 per year. 37 Many factors have contributed to the sharp deterioration in water and air quality in Bangladesh, including rapid and unplanned urbanisation without adequate capacity to provide basic urban services; heavy 33 Good reviews of the environmental management issues and progress in Bangladesh are available in World Bank 1991 and World Bank 1997a. 34 See World Bank 1998. 35 See World Bank 1997a. 36 World Bank 1997a. 37 World Bank 1997a. 18 Transforming Bangladesh into a Middle Income Economy concentration of the urban poor living in subhuman conditions; unplanned industrial locations with inadequate emission control standards, monitoring and enforcement; poor management of industrial and household wastes; inappropriate use of pesticides; and the use of two stroke engines in the transport sector. Generally, the citizens' awareness of environmental hazards to health has been as limited as the attention given to proper management of the environment by the government. SECTION III: PROGRESS WITH POLICY REFORMS Development outcomes are the results of the underlying policy and institutional environment. By and large, Bangladesh has pursued a development strategy that can broadly be classified as market oriented, although the mixture of state controls and state-owned enterprises relative to private sector enterprise has varied over time. Except for the early years of independence, Bangladesh has relied on economic growth and human development as the main instruments for poverty reduction. Since 1975, this focus on growth and human development has remained dominant in spite of the many changes in government. Correspondingly, the experiment with a controlled economy, comprising of state ownership and management of production and price controls, was short-lived and was basically reversed after the fall of the first government in 1975. Over the years since then, deregulation, liberalisation and focus on private enterprise have proceeded at different pace and pattern. To be effective, a market-based development strategy requires that policies and institutions are supportive of efficient functioning of markets. Government regulations need to provide an adequate enabling environment, so that competition from domestic and international enterprises leads to efficient allocation of resources based on demand. Forces of demand and supply determine prices. Market failures are corrected by government regulations to protect public interest. Distribution problems due to uneven initial conditions are resolved through proper state interventions, including regulations, taxation and public investment. Clearly, the role of the government is critical for achieving desired development outcomes even in a market-oriented economy, albeit the nature of this role is vastly different from a socialistic economy where resource allocation, distribution and pricing are all determined by the government. For analytical convenience, we will classify the policy framework for a market-based economy, such as in Bangladesh, under four categories: • Macroeconomic Policies • Policies for Private Sector Development Development Performance and Policy Reforms since Independence 19 • Policies for Human Development • Institutions and Governance It is important to note that this classification is somewhat artificial because policies in each category are often interrelated. In particular governance issues provide a common link to all categories. So, in the discussions, we will highlight the interrelationships whenever these are particularly important. Macroeconomic Management This was a problem area in the early years of independence of Bangladesh, but to a large extent became an area of strength after the late 1970s. Although some worrisome signs of strain on the macroeconomic balances emerged in the late 1990s, this was quickly corrected. So, on balance, after the difficult first few years of independence, macroeconomic management has been generally prudent allowing Bangladesh to reduce the pace of inflation, preserve the competitiveness of the real exchange rate, maintain positive real interest rates but without driving them too high, maintain low debt servicing ratio, and keep the fiscal cost of debt to a manageable level. Sound macroeconomic management has been a key contributor to growth performance in Bangladesh. The main problem areas though have been very low tax effort and the poor quality of public spending, both have adverse implications for poverty reduction. Exchange rate policy Over the longer term, the management of the exchange rate has been a positive aspect of Bangladesh economic management. The long term trend in real exchange rate is shown in Figure 1.4. A rapid pace of inflation in the 1972-75 period caused a sharp appreciation in the real exchange rate. This was reversed by 1980, and then was accompanied by a noticeable depreciation of the real exchange rate over 1980-85, aimed at securing the competitiveness of the real exchange rate. Since then to 1999, the real exchange rate showed a gentle upward drift indicating a slightly depreciating pattern. The management of the exchange rate suffered somewhat during 2000-01 from an appreciating trend. This has now been corrected with a switch to a flexible exchange rate system. 38 Overall, by 38 This major policy shift happened in late May 2003 despite substantial opposition from various intellectual groups and private business who feared instability. Not surprisingly, the response so far has been quite predictable with no destabilising influence. Critics did not recognise that exchange rate management was backed by a fiscal correction and substantial improvement in monetary policy management. 20 Transforming Bangladesh into a Middle Income Economy 12 j "0 10 Cl ~ c: 8 21 :; 6 U 4 !II: 2 W II: 0 1972 1980 1990 1997 2000 2002 1975 1985 1995 1999 2001 2003 Years Figure 1.4 Real Exchange Rate Movements aligning nominal exchange rate to reasonably competitive levels and avoiding significant periods of real exchange rate appreciation, Bangladesh was able to preserve export competitiveness to a significant degree. 39 This was an important factor underlying the good performance of the export sector, contributing positively to growth and poverty reduction.4D At the same time, by allowing a significant expansion in exports, this substantially increased the debt-servicing capacity of the economy, enabling the maintenance of a low debt-servicing ratio. Monetary policy In the 1970s, the management of monetary policy was a problem area for Bangladesh. Rapid monetary expansion, much in excess of the growth of real GDP, along with structural problems associated with a war-ravaged economy fueled a rapid pace of inflation in the early to the late 1970s, which hurt the poor considerably. Since late 1970s, the management of monetary policy improved significantly over the years and the pace of monetary expansion became more disciplined. Along with the sharp improvement in food supply, good monetary management contributed to a substantial slowdown in the pace of inflation to single digits in the 1990s (Table 1.7). Table 1.7 Money, Growth and Inflation IndicatorsiYears 1972-80 1980-90 1990-2000 200()-()4 Rate of growth of GOP 3.7 3.8 4.8 5.1 Rate of monetary growth (M1) 15.8 12.5 12.0 14.0 Rate of inflation (CPI) 19.3 10.4 5.4 3.2 Source: IMF and World Bank country reports. 39 One important concern though is the substantial appreciation of the Bangladesh currency vis-a.-vis India, a major competitor, thereby adversely affecting Bangladesh's export competitiveness. See Ahmed and Sattar 2003. 40 See Ahmed and Sanar 2003. Development Performance and Policy Reforms since Independence 21 Fiscal policy Although overall macroeconomic management has been good, the management of parts of fiscal policy has been weak (see Table 1.8). The main problems have been a low tax to GOP ratio and poor quality of spending, partly reflecting corruption in public spending. While fiscal deficits have generally been restrained, periodic outbursts of fiscal indiscipline have also happened. Expenditure management issues are discussed in greater detail in the review of the sectoral policies and in the section on human development. Table 1.8 Fiscal Developments in Bangladesh (per cent of GDP) Indicators 1972 1980 1990 2001 2002 2003 Fiscal deficit O.B 1.6 5.B 5.1 4.7 3.6 Interest cost 0.2 0.5 1.0 1.6 1.B 1.9 Tax revenue 3.2 5.2 7.2 8.1 B.3 8.9 Total revenue 3.B 6.6 6.B 9.0 10.2 10.4 Defence 0.5 0.9 1.5 1.3 1.2 1.1 Source: Data obtained from various government publications and World Bank country reports. Also, World Bank 1993. When Bangladesh became independent in 1972, it inherited a very difficult fiscal situation but also a number of advantages. These advantages related to low debt obligations and low military spending. By and large, while Bangladesh benefited from these initial conditions, it failed to fully capitalise on them. On the positive side, Bangladesh restrained defence spending and maintained a good balance between military and social spending. The relative priority given to social spending has served Bangladesh well and contributed to the notable progress made in human development over the past years. Yet, the ability to increase public spending on health and education to levels necessary to address the remaining substantial gap in human development noted earlier has been constrained by the very low tax effort (only 8.9 per cent of GOP - the lowest in the region). Tax compliance has been a serious problem in Bangladesh ­ estimated at only 6 per cent, the lowest in the regionY Large fiscal deficits in the late 1990s also emerged as a worrisome development. While the interest cost of debt is still small, this is only because of the past dominance of concessional foreign assistance, the continued availability of which is by no means assured. The growing debt to GOP ratio, rising from 11 per 41 See Ahmed 2002. A detailed review of tax performance and tax administration issues is contained in the paper by William McCarten in this volume. 22 Transforming Bangladesh into a Middle Income Economy cent in 1975 to 52 per cent in 2003, is an indication of the creeping effects of large external borrowings to finance the fiscal deficits. Efforts are now underway to reduce fiscal deficit by a combination ofexpenditure cutbacks and revenue mobilisation measures. Policies for Private Sector Development Following the early years of experimentation with a socialist pattern of economic management, Bangladesh embarked upon a more market-based economy. The incentive policies for private sector became the key determinant of economic growth. The pace and pattern of liberalisation and deregulation varied considerably in scope and significance by sectors and over time. Progress with deregulation and liberalisation advanced most in the area of trade policy, following the big push in the early 1990s. Deregulation in the industrial sector has been of mixed effectiveness because of the inability to address many of the deep-rooted governance problems of this sector. Liberalisation in agriculture, focused mostly on the input markets, yielded good results for rice production. Deregulation in the area of infrastructure has not advanced much, although encouraging beginning has been made recently in reforming the financial sector. Trade policies Despite a late start, by and large, progress with trade policy reforms has been an area of strength for Bangladesh. It has made steady progress in the 1990s in reducing trade barriers, both tarifr and non-tariff (see Table 1.9).42 Thus, the maximum tariff rate was sharply cut back from a high of about 350 per cent in 1992 to 25 per cent at present. The unweighted average tariff rate came down from 94 per cent in 1989 and 57 per cent in 1992 to Table 1.9 Bangladesh Progress with Tariff Rationalisation (average percentages) Tariffs FY1992 FY1996 FY1997 FY2002 FY2003 FY2004 FY2005 ------.~---- Average CD (unweighted) 57 22 22 17 16 16 13 Average protective tax 61 74 26 27 22 9 8 Top CD rate 350 50 45 37 32 30 25 Source: Estimates based on customs data. 42 The Trade paper by Ahmed and Sattar in this volume reviews trade policy issues in greater detaiL Development Performance and Policy Reforms since Independence 23 15 E ::> <: <: < 10 ~ C ~ 5 & 0 B'desh Pal< Low Inc Wor1d o 1973-80 1m 1960-90 .1990-97 Figure 1.5 Growth of Export Volume 13 per cent at present. The number of banned items has been substantially cut back. As a result, nominal and effective rates of protection have been reduced noticeably.43 In addition, special bonded warehouses and export processing zones have provided a big boost to the exports of the garments industry, taking advantage of preferential access to markets through the MFA. Owing to these trade reforms and good exchange rate management, the anti-export bias has been substantially reduced and the efficiency of domestic production has increased. 44 Overall, progress with trade policy reforms has been an important factor in supporting the expansion of exports (Figure 1.5) and economic growth in Bangladesh. 45 Nevertheless, there is still a substantial unfinished agenda and trade protection remains high by world standards. 46 This is partly because Bangladesh has continued to use a large number of supplementary duties and other forms of trade protection, but also because other countries in the developing world have moved faster with trade reforms. So, proceeding with the next phase of trade reforms, including reduction in the dispersion of tariffs and elimination of most other remaining trade bans, will be important for securing a further increase in the efficiency of domestic production, promoting export diversification and increasing the rate of economic growth. Industrial policy After the massive nationalisation of the industrial sector in 1972, there has been a slow process of privatisation during the 1980s and the 1990s. 43 See World Bank 1996. 44 See Sattar 1997 and World Bank 1999c. 45 See Ahmed and Sattar 2003. The concern that the expansion of garment exports was mainly due to favourable access to markets from the MFA and that this will be lost once the MFA phase out happens in 2005, is overstated. It is true that the MFA created a favourable environment for garment exports, but supply response was greatly facilitated by incentives. Bangladesh has favourable cost competitiveness in the low-end of garment products it exports, especially from cheap labour. 46 See World Bank 2003b. 24 Transforming Bangladesh into a Middle Income Economy Progress with industrial deregulation, however, gained momentum since the 1980s. Nevertheless, a number of quasi-controls still remained as of late 1990s, requiring business enterprises to seek permits, certifications, etc., before they can import machinery and equipment for setting up and running manufacturing enterprises. For example, the CCI&E still issues import permits for banned/restricted items, and the BOI issues permits for importation of machinery at concessional rates of tariff. More importantly, even though the number of procedures involved with starting business in Bangladesh has been sharply reduced, the implementation of these procedures takes a long time due to inefficiency and corrupt practices. Not surprisingly, the overall response of the manufacturing sector to this deregulation effort has been weak. 47 The main exception has been the garments subsector, where deregulation has been most intense, and it benefited substantially, contributing handsomely to poverty reduction through higher incomes and employment for the poor. The reason for the weak supply response from the non-garments enterprises is the combined effects of remaining regulations and high transactions cost of doing business, deep­ seated governance problems of the nationalised large-scale manufacturing sector, problems in the financial sector, weak infrastructure, and law and order problems. The policy of nationalisation of the early 1970s left a very difficult legacy of labour militancy, overemployment, corruption, and management inefficiency. all contributing to heavy financial losses and a severe overhang of debt in the industrial sector from which the country is yet to recover. Despite past efforts to reform the public enterprise sector, there is still a large involvement of the public sector in the management of industrial enterprises, many of which are in the red and do not have a ready market for pri vatisation due to the overhang of indebtedness and labour problems. 48 For example, the consolidated losses of state-owned manufacturing enterprises stood at taka 7 billion in 1999, increasing from around taka 6 billion in 1998. 49 As well, the depth of the manufacturing sector remains low owing in a large part to inadequate foreign investment. Indeed, Bangladesh is amongst the lowest recipient of foreign direct investment in 47 See Bakht and Rahman 1998 for a detailed analysis of the industrial sector constraints and performance. The paper by Syed Mahmood in this volume provides a detailed review of the constraints faced by Bangladesh enterprise sector. 48 The paper by Syed Nizamuddin on Public Enterprises in this volume provides a detailed review of public enterprise performance and reform strategy. 49 The deficits of the consolidated state-owned non-financial enterprises stood at taka 8 billion in 1999. Total SOE debt stood at taka 470 billion or around 27 per cent of GOP as of June 1997. Development Performance and Policy Reforms since Independence 25 South Asia, except Nepal. Among the negative factors for low foreign investment and shallowness of the industrial sector have been corruption, poor infrastructure, poor law and order situation, and political instability. 50 Fortunately, there is now a renewed effort to reform the state enterprises. Significant progress has been made over the 2002-04 period in stemming the loses of publicly owned manufacturing sector through a combination of reforms including privatisation, closures and downsizing. Data suggests that manufacturing enterprises losses have come down from taka 8 billion in 1998 to taka 2 billion in 2004. The government has also taken steps to reduce the social cost of these actions by providing adequate compensation to retrenched labour. Agricultural policies Bangladesh began with a heavy dose of controls over both the product and input markets for agriculture. Deregulation in agriculture started in early 1980s.51 This involved liberalisation ofthe fertiliser and irrigation equipment markets, and the reform of the public marketing of food grains. While much of the reform focused on agricultural inputs, on the output side the main noteworthy reform was the abolishment of most forms of food rationing and the monopoly in the import and export of food grains. The overall effect of this deregulation was positive, contributing to an expansion in agricultural productivity and value-added. 52 Much of the positive impact happened in rice production the dominant agriculture activity. Deregulation enabled rapid adoption of the high-yielding varieties (HYV) of rice, causing paddy production to increase at a faster pace than population and Bangladesh achieved rice self-sufficiency by the early 1990s. Yet, the agriculture sector did not show overall dynamism in terms of long term growth and diversification. The performance of the agriculture sector is heavily dependent upon the contribution of rice, which seems to have become constrained by slowdown of demand due to low income elasticity of demand and the slowdown of the population growth rate. Given the importance of agriculture for poverty reduction, it has been a subject of considerable analysis, debate and discussion. The upshot of the main results of this intensive research is provided below 53 : 50 World Bank 2003. 51 See Abdullah and Shahabuddin 1997 and Ahmed 1998. 52 Abdullah and Shahabuddin 1997; Ahmed 1998. 53 A good summary of these issues is contained in Abdullah and Shahabuddin 1997; Mahmud 1998; and Faruqee 1998. 26 Transforming Bangladesh into a Middle Income Economy • Despite progress, agriculture continues to suffer from many controls over output pricing, marketing, and input supply; removal of these controls will benefit farm production and value-added. • Deregulation has focused mostly on input markets; on the demand side, low expansion of the pace of domestic demand and inadequate export orientation continue to limit the incentives for production; • Diversification of agricultural production away from the heavy reliance on rice is important to raise agricultural value-added; this in tum will require policies to boost domestic demand along with policy support for pushing agricultural exports. • The prospects for non-rice agriculture exports are also limited by the relatively low productivity. So, policies for promoting technological progress in non-rice agriculture are very important. • Inadequate agricultural infrastructure is another constraint on farm production and productivity. • Overall water management and flood control policies are deficient, suffering from serious management problems, including O&M. • Land markets function inefficiently due to inadequate land administration policies, including the inadequacy of the legal process for land tenure arrangements and land sales. 54 • Finally, the inadequacy of the farm credit market remains a constraint on farm production. Infrastructure policies Progress with infrastructure policy reforms has been weak in Bangladesh. Much of the infrastructure provision (power, telecommunications, railways, ports, road-network) and maintenance has been in the public sector. Except for bus services, where an active private sector has operated, incentives for attracting private investment in infrastructure have been inadequate and a proper regulatory framework for private provision of infrastructure has been lacking. Consequently, substantial deficiencies in the quality and quantity of infrastructure services have prevailed, thereby increasing the cost of doing business in Bangladesh. Most serious problems have been in the power sector. Difficulties in the power sector have included huge losses, supply constraints, low reliability of service and corruption. 55 Thus, power losses, estimated in the range of 35 to 50 per cent of output during the 1990s, were to a large part a reflection of power theft and poor collection of power bills. 56 Power sector's financial losses were estimated at taka 5.6 54 See World Bank 2000d. 55 See World Bank 1998a. estimates for power losses are 35 per cent. However, more detailed scrutiny 56 Official suggests that these losses may be as large as 50 per cent. Development Performance and Policy Reforms since Independence 27 billion in 1999. Recently, faced with acute power shortage, Bangladesh has sought to attract foreign private investment in power generation. Some efforts have also been made to improve performance by improving billing collection, reducing power losses and price adjustments. Overall, the agenda for sector reforms has to go much further to ensure efficient and affordable power supply. Ports in both Chittagong and Chalna have been plagued with inefficiencies and labour problems, resulting in high shipping costs. For example, the cost of moving a container through Chittagong is estimated at dollar 600 as compared with norms of dollar 150-300 in neighbouring country portS. 57 Railways has suffered from serious financial losses and poor service. Inland waterways have not been adequately regulated, causing serious safety hazards. In all these areas serious reforms have been lacking. Telecommunications network has been inadequate, inefficient and expensive. 58 Recently, competition from private wireless providers has improved availability and lowered service cost. However, the scope for further improvement is enormous. Progress in expanding the road network has been more encouraging, including a rapid expansion in private bus operators. Yet, road traffic has remained seriously congested, and has suffered from unduly high incidence of accidents and poor maintenance. Financial sector policies Historically, this has been another major area of concern. Public banks with seriously infected portfolios have dominated the banking sector. Even in the private banks, the portfolio infection has been pretty high. Due to weak management, constant political interference, and problems of corruption and directed lending. As of 2000, the four Nationalised Commercialised Banks (NCBs), holding 63 per cent of the industry's net assets, were technically insolvent, with the ratio of their non-performing loans to total loans estimated at 46 per cent. The court system could not be relied upon to protect creditor rights, which further damaged lender incentives and contributed to misallocation of credit. While the issue of non-performing loans is fundamentally a problem of political economy, there were many associated management problems in Bangladesh relating to the inadequacy of the legal framework for banking control, banking supervision, inefficient management and overstaffing in public banks, and poor service standards in many banks, particularly in public banks.59 57 See World Bank 1998. 58 See World Bank 1998. 59 See paper by Sadiq Ahmed on banking reforms in this book. 28 Transforming Bangladesh into a Middle Income Economy Reform efforts over the past several years have sought to improve the regulatory and 1egal environment. Much of this initial effort was focused on the private banks. More recently, over the past two years, attention has been given to the public banks. These efforts are now yielding results. Thus, the share of non-performing loans in both private banks and NCBs is declining sharply, especially in private banks, and the share of private banks in total deposit and lending is rising (Table 1.10). Thus private banks now account for 49 per cent of deposits as compared with 34 per cent in 2000. Importantly, their share of net advances has accelerated to 52 per cent as compared with only 37 per cent in 2000. Related institutional reforms are also progressing well. The regulatory framework has improved and the capacity as well as effectiveness of the Central Bank in supervising banks has improved. Progress has also been made towards the application of international standards of loan classification and provisioning. Overall, the efficiency and soundness of the banking sector is better now than 3-4 years ago. Nevertheless, the vulnerability ofthe banking sector remains a concern and continued efforts are needed to complete the long term goal of establishing a market-based banking sector. 60 One area of the financial sector where better long term progress has been made in Bangladesh concerns the management of the interest rate. Figure 1.6 shows the trend in real deposit rates. In the early years, rapid inflation along with control over interest rates, caused deposit rates to become highly negative until 1975. High inflation of the early 1980s also caused deposit rates to become negative. However, since 1983, deposit Table 1.10 Banking Sector Performance FY Non-performing loans Deposits and advances (% of advances) (taka bllion) NCBs PCBs NCBs PCBs Gross Net Gross Net Gross Net Gross Net 2000 46 38 27 19 400 282 213 166 2001 39 32 22 14 441 310 267 226 2002 39 33 18 10 484 333 328 277 2003 33 28 15 9 510 357 420 338 2004 544 387 515 420 Source: Data from Bangladesh Bank. banking reforms and the case for a market -based 60 A review of South Asian experience with system as opposed to regulations system is contained in Ahmed, Ramachandran and Uy 2002. Development Performance and Policy Reforms since Independence 29 10 8 l 6 J 4 2 I 1 -2 1976 \ \,. 1980/ 1985 1990 1995 1997 1999 2001 2003 11: \ l -4 1978' J 1984 1987 1993 1996 1998 2000 2001 \/ Years Figure 1.6 Bangladesh Real Deposit Rates rates have remained largely positive. They fluctuated a bit though, reflecting yearly variations in inflation rates. Following years of control, interest rate policy was gradually deregulated in the 1990s. Recent reforms of government saving instruments have further reduced distortions and also helped calm the fear of rising real interest rates by reducing the cost of deposit mobilisation that is closely linked to returns on government saving instruments. So, the reform of taxes on financial instruments, the proper management of returns on long term government debt instruments (saving certificates, etc.) and the maintenance of positive real interest rates have all played an important role in spurring financial saving while reducing the bias against the stock market. Thus, the ratio of broad money (M2) to GDP surged from a low of only 20 per cent in 1980 to 40 per cent in 2003. Policies for Human Development Emphasis on human development has been a hallmark of the development strategy in Bangladesh. In addition to stronger funding support through the budget (Table 1.11), a notable feature of the Bangladesh experience is a successful partnership between the government and the NGOs in the delivery of basic human services. Indeed, this aspect of the Bangladesh experience with human development has important positive lessons for other South Asian countries. In particular, this has facilitated good progress with reducing the anti-female bias of development policy. Successive governments in Bangladesh placed strong emphasis on funding human development through the budget while also involving NGO participation, mainly through donor funding, for service delivery. As a result, significant increase in budgetary funding over the depressed levels 30 Transforming Bangladesh into a Middle Income Economy Table 1.11 Public Spending for Human Development in Bangladesh ---- Spending priority (% of GDP) 1972 1990 2001 2003 Human development 1.9 2.4 4.3 4.0 Defence spending 0.5 1.4 1.4 1.1 Ratio of human to defence spending 3.8 1.6 3.1 3.6 Source: Date collected from various Ministry of Finance publications. of the early years of independence happened. Even so, the level of funding has been inadequate relative to needs due to a severe public resource constraint imposed by the weak tax effort. As well, cost recovery policies for higher levels of education and tertiary health services have been inadequate. Even though the overall effectiveness of public spending on human development has been better than in other South Asian countries, especially Pakistan, there is still significant scope for improving the efficiency and equity of these services. Population management policies For Bangladesh, a particularly notable area of policy success in human development has been the population control programme. 61 A combination of education, social marketing of popUlation control materials, and technical advice based on family health workers meshed into an effective delivery system has enabled Bangladesh to sharply cut back the rate of population growth, declining from 2.8 per cent in the early 1970s to 1.4 per cent in the late ] 990s. Correspondingly, the fertility rate has been cut back from 6 in 1975 to only 3 in 2003. Prevalence of the contraceptive rate is amongst the highest in the developing world (Figure 1.7). Partnership between government, private sector NGOs and the donor community has also been exemplary. Arguably, this is amongst the most important policy success in the field of human development and shows that a concerted effort coupled 80 c:: Q) E 60 0 ~O> ~I c::­ 40 Q) u 20 ~ Il.. 00 Bangladesh Pakistan South Asia Low Income Middie Income Figure 1.7 Contraceptive Prevalence Rate 1999-2001 61 See Barkat et al., 1997 for a review of the popUlation control progress in Bangladesh. Development Performance and Policy Reforms since Independence 31 with strong partnership with concerned agents can provide the basis for attaining the desired goals of human development in Bangladesh and other South Asian countries. Health policy reforms Progress with health reform policies in Bangladesh has been mixed. Since independence there has been a gradual increase in public spending on health. Yet, at only 1 per cent of GDP, public spending on public health services is grossly inadequate. Given this tight resource situation, careful spending priorities are essentiaL On the positive side, emphasis on child immunisation and control of communicable diseases has yielded very good results leading to a substantial reduction in infant mortality.62 Partnership with NGOs has been instrumental. As well, there has been progress in regulating the pharmaceutical industries and import of drugs to ensure afford ability of basic medicines and to improve public safety of drug usage. Even so, the agenda on the health side remains daunting, reflected in low life expectancy and still high rates of child and maternal mortality rates. Of particular concern is the high incidence of child malnutrition, not withstanding past progress. 63 Along with a larger allocation of public resources for health services based on higher resource mobilisation, public expenditures needs to focus even more strongly on basic health care as opposed to tertiary care which should mostly be left to the private sector. A major policy weakness is the lack of an effective regulatory framework for private health care. As a result, while a flourishing private health care service has emerged in response to growing demand, there are serious problems of quality control, accountability and afford ability. Additionally, pricing policies for public tertiary care are inappropriate, providing huge subsidies without regard to incomes. Education policies Bangladesh has rightly recognised the importance of education for development. The main policy elements for education have been a combination of public financing and provision, and partnership with the NGOs for service delivery. Public spending for education expanded noticeably, reaching 2 per cent of GDP in 1990 and further rising to around 3 per cent in 2001. Yet, this level offunding is inadequate to meet both the quality and quantity requirements, particularly at the secondary level. 64 62 See World Bank 2002. 63 World Bank 2002. 64 See World Bank 1999b. 32 Transforming Bangladesh into a Middle Income Economy Regarding service delivery, Bangladesh has made remarkable progress in developing a strong partnership with the NGO community in the delivery of primary and secondary education. This has been the main factor underlying rapid increase in gross enrolment rates. 65 Another striking positive feature of Bangladesh education policy has been the strong emphasis on girl's education. The stipend programme for girls' secondary education has been a particularly successful policy, enabling the rapid enrolment of girls in secondary education and the elimination of the gender gap.66 The main problem areas in the education sector concern the poor quality of education at all levels, the inadequacy of enrolments at the secondary level despite progress from very low levels, inequity of public education spending, and inadequate policy framework for private provision. While the progress in raising enrolments at the primary level, particularly in increasing education participation of the girl child, should be celebrated, these difficulties in the education sector continue to pose a substantial policy challenge for reducing long term poverty as well as providing the skill base for supporting higher economic growth. Poor education quality is reflected in low education achievements, high dropouts, and the weak link of education with the job market. Quality problems relate to a whole host of factors in the entire education chain involving the curriculum, teacher quality, inadequacy of education materials, examination and testing standards. There are also governance problems relating to absentee teachers in public schools and examination and testing related corruption. As well, the efficiency and equity of public spending is of concern. These problems are not insurmountable, but require a thorough overhaul of education management. Recently efforts have been initiated to address the challenges in secondary education. These efforts will need to be sustained. Water supply, waste management, and management of air pollution Public spending along with supportive NGO role has enabled a sharp increase in the access to safe drinking water in Bangladesh. Also, one positive development in air pollution management has been the banning of leaded gasoline use in 1998. However, as noted earlier, water quality, including the problem of arsenic poisoning, is a serious health hazard. Actions taken by the govemment to mitigate this risk are welcome, but the arsenic problem remains a challenge. Similarly, heavily polluted air poses a severe health risk. The problems here largely reflect weakness of the 65 See World Bank 1999b. 66 See World Bank 2002. Development Performance and Policy Reforms since Independence 33 underlying institutions and corruption. 67 In particular, the ability of urban municipalities to handle urban waste and other basic urban services is severely constrained by poor governance. Additional policies, standards and monitoring of water and air pollution are constrained by weak progress with environmental management institutions. 68 Overall, pollution management is a major area of policy failure in Bangladesh leading to serious environmental health problems from air and water pollution. Progress with Institutional Reforms There is now a strong recognition that institutional reforms are necessary to sustain rapid growth and poverty reduction in developing countries. For example, the World Development Report (WDR) 2002 of the World Bank focused on 'Building Institutions for Markets' as the core theme of the report. There has been emphasis on certain aspects of institutions in other WDRs as well, including the WDR 1997 on the Role of State, the WDR 2003 on Environment, and the WDR 2004 on Service Delivery. Each of the WDRs builds on the work ofearlier WDRs and provides further analysis of the institutional context of the development challenge in concrete terms to make associated reforms implementable. This is indeed a laudable effort and provides a fairly rich analytical base over which one can build on to move the institutional reform agenda forward. Institutions have substantial political dimensions. However, to keep the focus on economic management issues, the paper will not dwell on political matters and look at only those institutions that have direct impact on economic management and related performance. For the context of this paper, and drawing on the work of the various WDRs, following dimensions of the institutional reform agenda would appear important: • Management of the budget • Anti-corruption policies and institutions • Independent regulatory bodies to protect public interests • Decentralization and devolution of decision-making • Civil service reforms Needless to say, institutional arrangements underlying each of these themes and associated issues are interrelated. Yet, it will be helpful to focus on each of these important themes separately in order to have clarity of underlying issues, measure specific areas of progress and identify areas where further attention is needed. 67 See World Bank 1997a. 68 See World Bank 1997b. 34 Transforming Bangladesh into a Middle Income Economy Management of the budget The government budget is probably amongst the most fundamental institution that has a detennining influence on development perfonnance. 69 A detailed exposition of past progress, issues and refonn agenda is provided in an accompanying paper in this volume. 70 In several respects, the management of the budget has been a positive development in Bangladesh. In particular, the emphasis placed on human development in budgetary allocations is a substantial achievement. However, in a number of areas, budgetary management has not progressed adequately. One key problem is tax administration. Bangladesh did make some progress in improving the tax structure by introducing the VAT and reducing the reliance on trade taxes, but the overall tax effort is weak due to both low coverage and low tax compliance. 71 Modernisation of tax administration is a major institutional challenge for Bangladesh. In expenditure management, while broad allocations are fine, there are several concerns relating to quality of public spending and corruption. The budgetary process in Bangladesh suffers from a number of weaknesses that reduce the effectiveness of the budget. These include: lack of strategic focus and clarity on sectoral priorities; inadequate infonnation on the cost of policies, programmes, and services; short tenn horizon for budget decision-making that fails to account for the long tenn costs and benefits; an artificial separation of development and recurrent budgets, weakening the integrated, strategic content of the budget; and end of year spending incentives that reduce the efficiency of spending. These drawbacks have often led to inefficient spending decisions. For example, the separation of development and recurrent spending has meant inadequate attention to the recurrent cost implications of capital spending. Indeed, operation and maintenance (O&M) has typically tended to be short­ circuited causing low returns from capital spending. Lack of good costing analysis of public programmes and policies has similarly meant inappropriate or inadequate funding of many programmes, leading to poor efficiency of these programmes. Many public programmes are based on political expediency rather than on a careful analysis and choice of competing priorities, thereby contributing to low returns from such spending. Improving budget fonnulation to link policy planning and budgeting is necessary to embark on strategic planning initiatives such as the Poverty Reduction Strategy (PRS). The government aims to move gradually towards 69 Ahmed 2004. 70 See paper by Sadiq Ahmed on Budgetary Management for Growth and Poverty Reduction in this volume. 71 The paper by William McCarten on Tax Administration in this volume looks at these issues in greater detail. Development Performance and Policy Reforms since Independence 35 a Medium Term Expenditure Framework (MTEF). The government has already reclassified around 50 development projects and aligned them with the overall budget classification. 72 It plans to continue until all projects have been reclassified and the revenue and development budgets have been fully realigned. With regard to budget strategy formulation, the government plans to proceed gradually by using selected ministries (starting with the education sector) as pilots for the preparation of sectoral medium term expenditures. The goal is to extend the approach to other sectors and then to the entire budget. These are important positive steps, but there is a long way to go. Among the issues that need more effort and resolution include: assigning clear responsibilities between different levels of governments and line ministries in the formulation of the MTEF; capacity to do detailed costing of programmes and policies; establishing sectoral and programme priorities; setting realistic programme goals and targets; and strengthening the relationship between the PRS and the annual budgets. Another key factor underlying the low effectiveness of public ex­ penditure in Bangladesh has been a dysfunctional expenditure monitoring system. The traditional monitoring emphasis has been on spending of budgetary allocations rather than on outcomes. Thus, the success of budget implementation is judged on the basis of how closely actual spending matched the allocated budget. The tracking of expenditure has also been constrained by inadequate attention to data quality and timeliness. Very recently, as a part of the PRS, an effort is now underway to establish a fairly detailed mechanism to monitor poverty-related expenditure, along with their intermediate indicators and outcomes, in relation to the specific human development and poverty targets in the PRS. This mechanism, when fully functional, can provide a powetful analytical tool to assess the effective­ ness of public spending in achieving stated outcomes. The success, however, requires a long term effort to increase monitoring and evaluation capacity and strengthen the underlying database in terms of quality, comprehensive­ ness and timeliness at the district level and moving upwards. Although on paper Bangladesh has a reasonably sound financial accountability set-up, in practice weak financial accountability has been a fundamental institutional constraint on public spending (World Bank 2000t). Major concerns include: (i) ineffective parliamentary and executive control of the budgetary spending audit reports on annual accounts are heavily qualified and accounts do not currently meet the expectations of the Constitution and associated laws and regulations; for example, the audit reports focus only on irregularities in individual transactions without 72 Overall budget classification refers to the classification of expenditures into administrative units, functional, subfunctional and economic categories according to the General Financial Statistics or UN internationally recognised standards. 36 Transforming Bangladesh into a Middle Income Economy sufficient attention to significance and systematic weaknesses of financial management system. Similarly, the legislative oversight of the accountability system is weak. The accountability of the executive is not sufficiently supported by information that would enable it to become focused on results and outcomes; (ii) the separation of audit and accounts took a long time to come; even now the process is yet to be completed; (iii) although a system of penalties exist, these sanctions and associated implementation is inordinately cumbersome and lengthy; (iv) the quality of audit suffers both from quality and timeliness of data as well as from lack of automation and low staffing quality. To address some of these concerns, since 2002 Bangladesh has initiated important reforms to improve financial management. The auditing and accounting functions have been separated. The Controller General of Accounts (CGA) no longer reports to the Controller and Auditor General (C&AG). Six Divisional Controller of Accounts (DCA) offices have replaced 20 regional accounting offices. Unions and districts now send their monthly accounts and report directly to the DCAs. A central reconciliation unit is working in the office of CGA. Budget Monitoring Committees have been formed in a number of ministries. Financial Management Units (FMUs) are being set up in five ministries in addition to seven previously done. The authorities have recently accepted an Inception Report for a Financial Management Reform Programme (FMRP), funded jointly by DFID and the government of Norway. The FMRP will help develop these FMUs, depending on the size of their ADP budgets. The Public Accounts Committee has been reestablished and is functioning. Further reforms are still needed to improve data reliability and timeliness, the comprehensiveness of government accounts, reporting and disclosures following International Public Sector Accounting Standards (IPSAS), quality of audit reports, and effectiveness of parliamentary oversight and other watchdog institutions. Anti-corruption policies and institutions For f0ur years in a row, Bangladesh has been labelled by the Transparency International (TI) as the most corrupt country in the world. While one can debate the precision of this ranking and quibble with methodology, there is no doubt that corruption has become a major development problem in Bangladesh. The observed coexistence of serious corruption and good development performance is seen by many as a puzzle. A closer look will show that this is not indeed a puzzle. A combination of good policies and private sector initiatives, including the positive role of NGOs, explain the good performance of Bangladesh on the development front. On the other hand, corruption and poor governance explain the weak performance of Development Performance and Policy Reforms since Independence 37 the public sector, which is a drag on the economy. Starting from a low level of economic activities and social indicators, it has been possible for Bangladesh to move forward rapidly with solid progress with growth, poverty reduction and human development. Nevertheless, Bangladesh remains a very poor and vulnerable economy, and long term sustained progress will be needed to reach the next stage of the MOOs. Increasingly this progress is being threatened by the pervasive problem of corruption. The investment climate is adversely affected by the associated high cost of doing business resulting from corruption. The missed opportunities for higher level of investment and growth are enormous. The enormity of the corruption problem is recognized by the government and a strategy is evolving to tackle many of the problems at source. The strategy involves efforts at both economywide level as well as at the sectoral level. At the aggregate level efforts are being made to improve procurement policies, financial management policies, better sharing and access to information,' improving budget management including tax administration, and establishing an Anti-Corruption Commission. At the sectoral level, the focus is on deregulation to reduce the scope for public intervention in commercial decision-making, reducing public ownership in commercial activities and establishing better regulatory policies and oversight in banking, energy and infrastructure. This effort is yielding results in a number of areas in terms of better management of public expenditures, reduction in non-performing loans of banks, and higher collection of utility bills. This progress is encouraging but there is a long, long way to go. The implementation of these reforms has begun only recently and implementation of some of the economywide reforms such as the Anti­ Corruption Commission has barely moved, demonstrating the highly political nature of these reforms and weak commitment. Needless to say, the battle here is long term and sustained political will at the highest level is essential for success. Independent regulatory bodies to protect public interest The corruption problems partly emanate from the absence of proper regulatory bodies that reduces the accountability of the concerned agency, public or private. Important examples of these institutions are: an independent Central Bank; an autonomous tax administration; an independent Audit Body; Public Accounts Committee (PACs); utility regulatory bodies, etc. The mileage that one could get from these institutions even in the context of a poor developing country like Bangladesh is illustrated by the recent progress in the banking sector. To a significant extent, this progress has been possible because of the recent empowerment of the Central Bank to regulate more rigorously the banking sector without 38 Transforming Bangladesh into a Middle Income Economy excessive interference from the government. The experience of Pakistan's banking sector with this approach is similarly illustrative of the good outcomes that strong institutions can deliver. Bangladesh is slowly moving in this important area. In addition to the Central Bank, some progress has also been made with strengthening the audit functions and a start has been made to improve the effectiveness of the tax administration. This progress, however, is very small and gradual. The agenda here is long and arduous. Once again strong, sustained political will is necessary. Decentralisation and devolution of decision-making Public administration in Bangladesh is heavily centralized. Within the civil administration, almost all authority is exercised by the head of the government and the cabinet. Local governments are very weak, with little administrative and financial authority (World Bank 19%a; World Bank 1997b; World Bank 2000e; UNDP 2002). There have been a number of attempts to establish a stronger system of local governments. These have had very limited success due to lack of strong political commitment at the top. Consequently, the setting of expenditure priorities, allocation of resources, procurement of goods and services, and the implementation of projects are largely centralised at the ministry level in the capital city of Dhaka. District administrations are run by the civil servants with little independent authority. So, local level involvement is limited to running public facilities at the district level and maintaining law and order. Additionally, local level civil servants are not accountable to the local governments but to the ministries at the central level. The political progress on decentralisation has suffered in Bangladesh from the un supportive attitude of both the large political parties (Awami League and the BNP), even though the rhetoric has been different. At the heart is the contentious issue of division of power between the national legislators and the local governments. While similar conflicts arise in other countries as well, in Bangladesh this has become particularly complicated because of the small physical size of the country, the homogenous nature of the people and the relative ease of physical mobility. Consequently, national legislators have tended to argue that they can take care of their constituencies without the need for an intermediary political agent (elected and empowered local governments). Civil servants have also found this convenient since this has given them more authority without accountability, especially at the local level. What has been the outcome of this heavily centralised administration for development performance? Even though the jury is out on the international evidence on the role of decentralised governments in service Development Performance and Policy Reforms since Independence 39 delivery (World Bank 2003a), a priori it can be argued that the centralised system of public service delivery is likely to be an important contributor to the poor service quality in Bangladesh and needs to be revisited. The example from Pakistan in the delivery of social services is one negative experience of the ineffectiveness of a heavily centralised service delivery model. Recognising this, Pakistan has embarked upon an ambitious programme to decentralise basic service delivery to the district level. Although, the changes are still ongoing and it is too early to judge whether or not there has been a major breakthrough in service delivery, initial evidence seems encouraging. In the case of Bangladesh, this is perhaps best illustrated by the experience with delivery of basic social services (education, health, water supply) through NGOs who tend to work very closely with the local communities as compared with centralised public services with poor accountability. It is generally acknowledged that on average health and education services delivered by NGOs involving community participation has been significantly better than comparable services provided by public facilities (see for example Ahmed and Nath 2004 for a review of BRAC experience in delivery of primary education). This is not surprising. The NGO facilities have typically involved heavy participation of local community members in the design and delivery of these services and with better accountability by NGO officials, whereas on average public services have suffered from lack of community participation and accountability of service providers. Centralised management has meant that there is no functioning mechanism to ensure accountability of the local public service providers. Thus, for example, the teacher or the health worker continues to get paid whether or not the person shows up because these service providers are not hired by the community or get punished for non­ performance. Civil service reforms At both the central and local levels, day to day general administration is run by the civil servants. This system was inherited from the British colonial times and has changed very little in terms of basic attitude and accountabilities. In the early days after the independence of India and Pakistan from the British, the standard for the selection of civil servants was tough leading to high quality of officers. The high quality and the prestige value of the job sometimes motivated officers to take a missionary zeal and make an effort to contribute to local development. Over time, though, the quality of civil service weakened. The pay and benefits also fell drastically in real terms. Currently, the civil service is 40 Transforming Bangladesh into a Middle Income Economy plagued with a plethora of problems including low quality, poor remuneration, weak accountability and corruption. Consequently, service quality is low and bureaucracy is often a serious hurdle to the efficient functioning of the private sector and the economy (World Bank 1996a; World Bank 2000e). Reform efforts in this area have been weak. There have been many studies of the civil service in Bangladesh, the most recent of these being the 2000 Report of the Public Administration Reform Commission (PARC). Generally, the recommendations in these reports are sound and sensible. But very little of the recommendations from the PARC report, or from other reviews and commissions, have been implemented. The major reason for the lack of action in this area is the absence of political will. While a systematic reform effort has not been made, some partial reforms have happened. First, the government has implemented a form of merit-based promotion for deputy secretaries and above. Second, to rebalance the pyramid nature of staff with heavy concentration at the lowest level (class IV), the freezing and shifting towards a 'no work no pay' contract model for new hires at the class IV employee level has been maintained for a number of years now. Third, to support the new policy of separating the judicial and executive functions, the government has very recently separated the judicial cadre from the normal civil service. It is clear that civil service reforms lack political champion in Bangladesh, which is very unfortunate. This constraint will continue to impose a serious burden on development effectiveness in Bangladesh by reducing the quantity and quality of public services, as well as by reducing the progress in tackling corruption. Public pressure and donor effort to create ownership for this important reform will help. Further deregulation and reducing public sector role in delivering commercial services will also help. Similarly, greater devolution of responsibilities to publicly elected local bodies will help create the enabling environment for civil service reforms. The agenda is a long term one and progress is likely to be gradual. SECTION IV: SUMMARY AND CONCLUSIONS In this paper we drew on the basic principle emphasised by Sen 73 and Stiglitz 74 that development is much more complex than only securing rapid economic growth. Key elements of our analytical frame­ work for measuring development progress in Bangladesh included: progress with poverty reduction and human development; the pace and pattern of economic growth; the distribution of the benefits of economic 73 Sen 2000. 74 Stiglitz 1998. Development Performance and Policy Reforms since Independence 41 growth; and social development and participation. The overall record of Bangladesh on development performance since independence is fairly impressive. Yet substantial challenges remain. The review also shows a rich and fascinating pattern of policy successes and failures. The paper showed that a number of positive socio­ political-economic factors contributed to past progress. These include: (i) favourable initial social conditions such as common language, moderate religious influence, a relatively positive social attitude towards women's empowerment; (ii) favourable regional security environment enabling military spending to be kept low; (iii) low initial debt obligations; (iv) a tradition of political consciousness and civil society activism leading to political competition and some degree of accountability; and (v) an abundant and hardworking labour force willing to take risks. As a result of these factors: (a) broad spending priorities were favourable to human development; (b) a strong role of NGOs/CBOs emerged in the delivery of social services, partially substituting for the failures of public service delivery; (c) an enterprising private sector emerged in most spheres of economic and social activities, again partly substituting for the weakness of public sector performance. Additionally, much of the gains came from relatively non-controversial policies where political trade-offs were either not serious or where these trade-offs were manageable. Generally, the pattern of policy reforms has followed a path of least controversy. These include policies such as: investment in seed-fertiliser technology, building up road infrastructure, investment in water, investment in education of girls, establishing free trade zones for the garment sector, and maintaining good monetary and exchange rate policies. Even so, the weak governance environment reduced the effectiveness of some of these policies, for example the low effectiveness of public spending due to corruption. While appreciating the good progress with development, one should note that this progress is measured from very low initial levels. Consequently, notwithstanding the progress, Bangladesh remains one of the poorest country in the world with a per capita income of only 400 dollars. Not surprisingly, the incidence of poverty is still very high, nearly 50 per cent, and there is still substantial vulnerability of the poor to shocks. Human development has been impressive, but remaining challenges are substantial. Addressing these concerns will require growth to be raised to 6-7 per cent per annum by improving the investment climate, while social inclusion and participation in the growth process needs to be strengthened through further improvements in human development. Securing these goals will require moving to the second and more complicated phase of reforms that involve substantial political trade-offs. Without matured political leadership and improvement in public administration, such reforms will not be possible. For example, problems include low resource mobilization, 42 Transforming Bangladesh into a Middle Income Economy inefficiency of the financial sector, the generally weak investment climate due to high cost of doing business, low quality of education and health services, and inadequate social protection. Reforms in each of these area involve substantial political trade-offs. The political nature of the agenda can be seen from the following illustrations: • Stronger resource mobilisation will require bringing the powerful elites to the tax net and collecting electricity and other utility bills; • Financial sector improvements will require taking the public sector banks out of the lending business and assigning this to private banks with full servicing and repayment of loans, which in tum requires the powerful business and others to repay their loans and eliminate the rents of politicianslbureaucrats from their influence over public banks; • Improvement in investment climate will require removal of all hurdles to private investment, improving law and order, and strengthening infrastructure services. All of this require strong political will to tackle the many vested interests that are a constraint presently; • Finally, improvement in the quality of human development and stronger social protection will require substantial resources as well as a sea change in the delivery of public services to ensure better accountability and transparency, including decentralisation of public services to link them better with users. The agenda here is clearly political in nature. The basic message is that reforms are intimately linked with politics. Indeed, the link between politics, administration and policy performance is well demonstrated by the review of the experience of Bangladesh. Progress in policies happened in areas where political trade-offs were low and vested interest was not over-powering. But deeper reforms, needed to push ahead the poverty reduction and human development agenda beyond the easy options will require stronger political commitment and better representation of the common citizen in the decision-making process. REFERENCES Abdullah, Abu and Quazi Sahabuddin. 'Critical Issues in Agriculture: Policy Response and Unfinished Agenda' in The Bangladesh Economy in Transition' edited by M.G. Quibria, Dhaka: University Press Limited, 1997. Ahmad, Monawaruddin and Mohabbat Khan. 'Dimensions of Governance' in The Bangladesh Economy in Transition, edited by M.G. Quibria, Dhaka: University Press Limited, 1997. Development Performance and Policy Reforms since Independence 43 Ahmed, Raisuddin. 'Assessment of Past Agricultural Policies' in Bangladesh Agriculture in the 21 st Century, edited by Rashid Faruqee, Dhaka: University Press Limited, 1998. Ahmed, Sadiq. 'Inflation in Bangladesh, Causes and Consequences', Ph.D. Thesis (Microfilm), Boston: Boston University, 1984. - - - . 'Explaining Pakistan's High Growth Performance Over the Past Two Decades: Can it be Sustained?' Policy Research Working Paper, No. 1341, Washington, DC: World Bank, 1994. - - - , . 'The Political Economy of Poverty Reduction in South Asia: Role of Good Governance', South Asia Discussion Paper Series No. IDP-183, Washington, DC: World Bank, 2002. - - - . 'The Political Economy of Development Performance in Bangladesh', Paper presented to the BER-World Bank Conference on Bangladesh, Dhaka: June 2003. - - - . 'Budgetary Reforms for Growth and Poverty Reduction in Pakistan', The Iqbal Memorial Lecture, 19th Annual General Meeting and Conference of the Pakistan Institute of Development Economics, Islamabad: 2004. Ahmed, Sadiq, Marilou Uy and S. Ramachandran. 'South Asian Banking and Finance: Growth With Festering Problems', South Asia Region Internal Discussion Paper No. IDP-186, Washington, DC: World Bank, 2003. Ahmed, Sadiq and Zaidi Sattar. 'Trade Liberalisation, Growth and Poverty Reduction in Bangladesh' in Annual Bank Conference on Development Economics 2003, South Asia Component edited by Sadiq Ahmed and Suman Berry, New Delhi: NCAER, 2004. Ahmed, Salehuddin and Samir R. Nath. 'Public Service Delivery in Education: The BRAC Experience' in Annual Bank Conference on Development Economics 2003, South Asia Component edited by Sadiq Ahmed and Suman Berry, New Delhi: NCAER, 2004. Balcht, Zaid and Mashiur Rahman. 'Constraints to Industrial Development: Recent Reforms and Future Directions', in The Bangladesh Economy in Transition, edited by M.G. Quibria, Dhaka: University Press Limited, 1997. Barkat, Abul et al. Family Planning Unmet Need in Bangladesh, Dhaka: University Research Corporation (Bangladesh), 1997. Dollar, David and Art Kraay. Growth is Goodfor the Poor, Mimeo, Washington, DC: World Bank, 2000. Dreze, Jean and Amartya Sen. Economic Development and Social Opportunity, New Delhi: Oxford University Press, 1995. Dreze, Jean. 'Bangladesh Shows The Way', The Hindu, 17 September, 2004. Faroqee, Rashid. 'Linking Agricultural Growth With Macroeconomic Policies and Performance' in Bangladesh Agriculture in the 21 st Century, Dhaka: University Press Limited, 1998. Gray, Cheryl and Daniel Kaufman. 'Corruption and Development', Finance and Development, Number 35, March, 1998. Hossain, Akhtar and Salim Rashid. 'Financial Sector Reform' , in The Bangladesh 44 Transforming Bangladesh into a Middle Income Economy Economy in Transition, edited by M.G. Quibria, Dhaka: University Press Limited, 1997. Hossain, Mahabub. Green Revolution in Bangladesh: Impact on Growth and Distribution of Income, Dhaka: University Press Ltd., 1987. Human Development Center. South Asia Human Development Report 1999: The Challenge of Humane Governance, Karachi: Oxford University Press, 1999. Kaufman, Daniel, et aL 1999. 'Governance Matters', Mimeo, Washington, DC: World Bank, 1999. - - - . 'Governance Matters II: Updated Indicators for 2000/1', World Bank Working Paper, Washington, DC: 2002. Khan, Zillur R. Leadership in the Least Developed Nation: Bangladesh, Maxwell School, New York, Syracuse University, 1983. Khan, Mohammad Mohabbat and Monawwaruddin Ahmad. 'Dimensions of Governance' in The Bangladesh Economy in Transition, edited by M.G. Quibria, New Delhi: Oxford University Press, 1997. Khandker, Shahidur, et al. Credit Programs for the Poor: Household and Intra­ household Impacts and Program Sustainability, Bangladesh Institute of Development Studies and World Bank, Washington, DC: Dhaka, 1996. Khandker, Shahidur and Mark Pitt. 'The Impact of Group-Based Credit Programs on Poor Households in Bangladesh: Does the Gender of Participants Matter?' Journal of Political Economy, number 106, 1998. Kumar, Praveen and Shahid Yusuf. 'Developing the Non-Farm Sector in Bangladesh: Lessons From Other Asian Countries', World Bank Discussion Paper Number 340, Washington, DC: 1996. Mahmud, Wahiduddin. 'Agricultural Development Strategy' in Bangladesh Agriculture in the 21 st Century, edited by Rashid Faruqee, Dhaka: University Press Limited, 1998. - - - . Bangladesh Economy into the 21 st Century, Mimeo, Presentation to the Annual Bangladesh Economic Association, Dhaka: 2000. Rodrik, Dani. 'Where Did All the Growth Go? External Shocks, Social Conflict and Growth Collapses' National Bureau ofEconomic Research Working Paper, number 6350, Boston: 1998. - - - . 'Institutions for High Policy Growth: What They Are and How to Acquire Them?' NBER Working Paper Number 7540, Mass: Cambridge, 2000. Rose-Ackerman, Susan. 'Corruption and Development', Paper prepared for the Annual World Bank Conference on Development Economics. Washington, DC: World Bank, 1997. Sattar. Zaidi. 'Non-Traditional Exports and Economic Performance', in 'The Bangladesh Economy in Transition' edited by M.G. Quibria, Dhaka: University Press Limited,1997. Sen, Amartya. Development as Freedom, Knopf and Oxford, New York: Oxford University Press, 1999. - - - . 'Globalization And Its Discontent'. Paper presented to the Annual Development Performance and Policy Reforms since Independence 45 Conference on Development Economics, Paris: World Bank, 2000. Sen, Binayak. Bangladesh Poverty Analysis: Trends, Policies and Institution, Mimeo, Dhaka: BIDS, 2000. Sobhan, Rehman et a1. Growth or Stagnation: A Review of Bangladesh Development, Center for Policy Dialogue, Dhaka: University Press Limited, 1996. - - - . Crisis in Governance: A Review of Bangladesh:S Development in 1997, Center for Policy Dialogue, Dhaka: University Press Limited, 1998. Sobhan, Rehman. 'Restoring Justice to Development: The Challenge Before the Economists of Bangladesh', paper presented to the 13 th Biennial Conference of the Bangladesh Economic Association, 10-12 August, Dhaka: 2000. Srinivasan, T.N. Growth, Poverty Reduction and Inequality, Mimeo, new Haven: Yale University, 2000. Stiglitz, Joseph. Towards a New Paradigm ofDevelopment, The 1998 Raul Prebisch Lecture, Geneva: UNCTAD, 1998. Summers Lawerence H. 'Investing in All the People', Papers and Proceedings, Part 1, Eight Annual General Meeting of the Pakistan Society of Development Economists, Islamabad: The Pakistan Development Review, 1992. Thomas, Vinod et a1. The Quality of Growth, World Bank, New York: Oxford University Press, 2000. United Nations. Human Development Report, New York: Oxford University Press,1999-91. United Nations, IMF, OECD and World Bank. A Better World for All, Communications Development, Washington, DC: World Bank and London: Grundy & Northledge, 2000. UNDP. Role of UNDP in promoting Local Governance and Decentralization in Bangladesh, Report of the initial SPPD scoping Mission, New York. 2000. Wei, Shang-Jin. Corruption and Poverty: New Evidence, Mimeo. Development Economics Department, Washington, DC: World Bank, 1999. Westergaard, Kirsten. State and Rural Society in Bangladesh, London: Curzon Press, 1985. Wodon, Quentin T. 'Growth, Poverty, and Inequality: A Regional panel for Bangladesh' , Policy Research Working Paper Number 2072, Washington, DC: World Bank, 1999. World Bank. Bangladesh Environment Strategy Review, Report Number 9551­ BD, Washington, DC: 1991. - - - . World Tables 1993, Baltimore: Johns Hopkins University Press, 1993. - - - . Bangladesh Privatization and Adjustment, Report Number 12318-BD, Washington, DC: 1994a. - - - . Bangladesh Trade Policy Reform For Improving the Incentive Regime, Report Number 15900-BD, Washington, DC: 1996. - - - . Bangladesh Government That Works, Report Number 15182-BD, Washington, DC: 1996a. - - - . Bangladesh Labour Market Policies for Higher Employment, Report 46 Transfonning Bangladesh into a Middle Income Economy No. 13799-BD, Washington, DC: 1996<:. - - - . Bangladesh Public Expenditure Review, Washington, DC: 1997. - - - . A Review of Key Environmental Issues in Bangladesh, Mimeo, Washington, DC: 1997a. - - - . Bangladesh Municipal Finance Management Study, Infrastructure Operations Division, Country Department 1, South Asia Region, Washington, DC: 1997b. - - - . World Development Indicators, Baltimore: lohns Hopkins University Press, 1997-2000. - - - . Water Resource Management in Bangladesh: Steps Towards a National Plan, Report Number 17663-BD, Washington, DC: 1998. - - - . Bangladesh Energy Sector Strategy Note, Washington, DC: Draft Report, 1998a. - - - . Bangladesh From Counting the Poor to Making Them Count, Washington, DC: World Bank Country Study, 1999. - - - . Bangladesh Breaking the Malnutrition Barrier is Key to Development, Washington, DC: Draft Report, 1999a. - - - . Bangladesh Education Strategy, Washington, DC: Draft Report, 1999b. - - - . Bangladesh Key Challenges for the Next Millennium, Washington, DC: 1999c. - - - . Bangladesh Country Procurement Assessment Report, Washington, DC: 1999d. - - - . World Development Report 2000/1, New York: Oxford University Press, 2000. - - - . Bangladesh Recent Economic Performance, Document for the Bangladesh Development Forum 2000, Washington, DC,2000a. - -.•loint BIS-IMF-OECD-World Bank Statistics on External Debt' , Mimeo, Washington, DC: 2000b. - - - . Bangladesh Paying for Corruption, South Asia Poverty Reduction and Economic Management Department, Washington, DC: 2000c. - - - . Bangladesh lilnd Study, South Asia Rural Development Department, Washington, DC: 2000d. - - - . Taming Leviathan: Reforming Governance in Bangladesh An Institutional Review, Washington, DC: Draft Report, 2000e. World Bank and UNDP. Bangladesh Country Financial Accountability Assessment, Washington, DC: Draft Report, 2000f. World Bank. World Development Indicators, Washington. DC: 2001. - - - . Engendering Development. Washington, DC: A World Bank Policy Research Report, 200la. ---,. Poverty in Bangladesh; Building on Progress, Report No. BD-24299, Washington, DC: 2002. - - - . Improving the Investment Climate in Bangladesh, Washington, DC: Draft Report, 2003. - - - . Making Services Work for the Poor, World Development Report 2004, New York: Oxford University Press, 2003a. ---,. World Development Indicators, Washington, DC: 2004. Development Performance and Policy Reforms since Independence 47 APPENDICES Table A 1.1 Bangladesh Poverty Incidence, 1973--92 Years Headcount (A)75 Headcount (Bf6 Rural Urban National Rural Urban National 1974 60.3 52.0 59.4 1977 78.9 67.7 77.4 n.a. n.a. n.a. 1978 77.4 68.4 75.4 n.a. n.a n.a 1979 65.8 54.9 64.1 n.a. n.a. n.a. 1982 55.3 45.5 54.2 65.3 48.4 63.3 1984 46.3 37.1 45.1 53.8 40.9 52.1 1986 37.3 26.4 35.9 45.9 30.8 43.9 1989 43.4 37.4 41.2 49.7 35.9 47.9 1992 n.a. n.a. n.a. 52.9 33.6 50.4 1996 n.a. n.a. n.a. 51.1 26.3 46.4 Source: Sobhan et al. 1996 and Sen 2000. Table A 1.2 Bangladesh Progress with Poverty Reduction, 1984-200077 Years Poor (headcount index) Vel}' Poor (headcount index) Rural Urban National Rural Urban National 1984 60 50 59 43 28 41 1986 53 43 52 36 20 34 1989 59 44 57 44 22 41 1992 61 45 59 46 23 43 1996 55 30 51 38 14 34 2000 53 37 49 37 19 33 Source: Wortd Bank 1999 and World Bank and Asian Development Bank 2003. 75 Based on distribution data ranked by per household expenditure. 76 Based on distribution data ranked by per capita expenditure. 77 These estimates in some sense are based on best practice methodology, being derived from the cost ofbasic needs method and careful attention is paid to deriving price deflators. 2 Poverty Trends During the Nineties Rinku Murgai and Salman Zaidi SECTION I: INTRODUCTION1 The performance of the Bangladesh economy in the past decade has been relatively strong, with annual growth in gross domestic product (GDP) averaging about 5 per cent during the 1990s. Between 1991 and 2000, real GDP increased by 52 per cent in real terms, with gross output in agriculture, services, and the industrial sector increasing by about 33 per cent, 50 per cent and 86 per cent, respectively. Given the widespread interest in linkages between growth, equity, and poverty reduction, investigating the extent to which this impressive growth performance translated into reduced incidence of poverty in the country is an important one. The household expenditure surveys (RES) series conducted by the Bangladesh Bureau of Statistics (BBS) are the main data source for estimation of poverty in Bangladesh. These surveys are designed by BBS to be comparable over time (i.e. in terms of methodology, questionnaire content, interviewing procedures, etc.), and have been carried out in Bangladesh at regular intervals. This paper presents the main findings of the analysis of the 2000 household income and expenditure survey (HIES), as well as of earlier rounds of the RES series (i.e. the 1991-92 and 1995-96 surveys) to assess changes in poverty incidence in Bangladesh during the past decade. The analysis presented was carried out in close collaboration with BBS. Trends in poverty and inequality in Bangladesh during the 1990s are presented in Section II, which also outlines the various steps followed to derive these estimates. Section ill compares selected findings from the various RES datasets with other data sources such as the National Accounts. 1 This paper was prepared as a background paper for the Bangladesh Poveny Assessment. Poverty Trends during the Nineties 49 This section includes a discussion of the extent to which the main HES findings are corroborated by these data sources, as well as highlights areas where the two present conflicting trends. Section IV presents a brief discussion of the extent to which the three HES datasets are comparable with one another. In addition, estimates of poverty obtained by following alternate methodologies are also presented in this section, along with trends in other selected indicators of living standards. Section V contrasts the pace of poverty reduction in Bangladesh with its neighbouring countries in South Asia and East Asia. Finally, Section VI concludes by summarising some of the main findings of the paper, as well as outlining areas where further work and research might prove fruitful. SECTION II: TRENDS IN POVERTY INCIDENCE DURING THE NINETIES Methodology BBS and the World Bank used the cost of basic needs (CBN) method to derive poverty lines and poverty measures from the 1991-92 and 1995-96 RES (BBS 1997; World Bank 1999). To summarise briefly, the CBN approach entailed three main steps: First, a food bundle yielding 2,122 kcal per day per person was chosen comprising rice, wheat, pulses, milk, mustard oil, beef, fresh water fish. potato, other vegetables. sugar, and bananas. Purging reported unit values in the survey data of possible variation due to differences in the quality of items consumed, the prices of the various food items in this bundle were estimated for 14 different geographic regions to ascertain the total cost of consuming this bundle in different parts of the country. 2 The second step was then to estimate the cost of basic non-food needs. Following the approach proposed by Ravallion (1994), two non­ food allowance components were calculated: the first obtained by taking the amount spent on non-food items by those households whose total consumption was equal to their regional food poverty line (corresponding to the lower poverty line), while the second was obtained by taking the amount spent on non-food items by those households whose food consumption was equal to the regional food poverty line (corresponding 2 The 14 regions used comprised: 1. Dhaka SMA, 2. Other urban areas of Dhaka division, 3. Rural areas ofDhaka and Mymensingh, 4. Rural areas of Faridpur, Tangail, and Jamalpur, 5. Chinagong SMA, 6. Other urban areas of Chittagong division, 1. Rural areas of Sylhet and Comilla, 8. Rural areas ofNoakhali and Chinagong, 9. Urban areas of Khulna division, 10. Rural areas of Barisal and Pathuaka1i, 11. Rural areas of Khulna, Jessore, and Kushtia, 12. Urban areas ofRajshahl, 13. Rural areas of Rajshahl and Pabna, and 14. Rural areas of Bogra, Rangpur, and Dinajpur greater districts. 50 Transforming Bangladesh into a Middle Incame Economy to the upper poverty line). The third step in calculating the lower and upper poverty lines for each region entailed simply adding up the cost of purchasing the food bundle in each region to the respective non-food allowance components. The lower poverty lines thus incorporated a minima1 allowance for non-food goods (the typical non-food spending of those who could just afford the food requirement) while the upper poverty lines made a more generous allowance (the typical non-food spending of those who just attained the food requirement). In assessing trends in poverty over the decade, we hold fixed in real terms the poverty lines estimated by the CBN method at the beginning of the period, i.e. 1991-92, and update in subsequent years each region's base year poverty line for changes in the cost of living using a region-specific price index. The methodology used to derive these regional price indices and related estimates of poverty trends are discussed in this section. Alternative estimates of poverty trends obtained using poverty lines derived through other methodologies are discussed in Section N. Adjusting Poverty Unes for Changes In Cost of Uvlng There are several data sources that could potentially be used for estimates of cost of living increases, needed to update the 1991-92 poverty lines. For instance, we could (i) rely on estimates of inflation from official sources such as the consumer price index (CPI) or the GOP deflator series, or (ii) use price indices derived from the lIES datasets themselves using information on unit values of various consumption items collected in the surveys.3 In some sense, the CPI is the natural choice for updating the poverty lines, as it is the standard yardstick used by most to assess changes in the cost of living over time. However. in the case of Bangladesh, the official CPI suffers from two main shortcomings: (i) it is based on a set of weights that have not changed since 1985-86, and may, therefore, be quite out of date in relation to current consumption patterns, and (ii) the national index, which is derived by aggregating urban and rural price indices, may be a poor proxy for changes in price levels in different regions. By contrast, an important advantage of using the lIES data to derive price indices is that not only do the surveys report unit value information relating to actual transactions, i.e. rather than prices listed or reported by shops - but also that these data permit one to calculate region-specific 3 This, for instance, was the approach used by Deaton and Tarozzi (2000), who used similar data from the Indian National Sample Survey Organisation (NSSO) datasets to derive inflation rates over time as well as across regions for their analysis of poverty trends in India. Poverty Trends during the Nineties 51 indices to take into account differential rates of inflation across various parts of the country. However, one drawback of using data from surveys is that they rarely have information on prices of non-food items, and thus provide only a partial picture of the change in the aggregate price level. Food and non-food items (mainly fuels) for which unit values can be calculated from the HES surveys account for approximately two-thirds of total household expenditures.4 The budget shares not covered in urban areas tend to be higher, which is a reflection of the relatively greater importance for urban consumers of goods such as housing and transportation for which we have no information on unit values. If the prices of these non-covered items change at a rate different from those items included in the index, then the price indices derived from the HES data may not fully capture changes in the cost of living over time. As both the above alternatives - the CPI or HES based price indexes each have their advantages as well as shortcomings, we combined the two into a composite index so as to capitalise on the relative strengths of both approaches. The HES based price indexes were derived in four steps. First, expenditures on various items in the HES were divided into 14 groups. These groups were chosen so as to retain as much disaggregation as possible (to minimise heterogeneity within categories) as well as to be comparable across the three survey years. 5 Second, unit values (by dividing expenditures by quantity) of the most commonly consumed item within each of the expenditure groups were calculated for each household. For each group, the median of the unit values within each geographic region was calculated.6 Using the price of the most commonly consumed item within each group and medians (which are more robust to outliers as compared to means) for the summary region-specific unit values helped minimise the problem that the calculated unit values are contaminated by choice of quality rather than providing information on market price alone. Third, average budget shares of the 14 main expenditure groups were calculated for each survey year. Finally, region-specific Tornqvist price indexes were then calculated using budget shares of the expenditure groups along with median prices of the selected items.7 The Tornqvist price indices for each region k were calculated as follows: 4 Budget shares are presented in Appendix Table A2.1. 5 Appendix Table A2.2 lists the relative budget share weights of each group in the overall RES price index for each year. 6 The median values of the unit values for the three surveys are reported in Appendix Table A2.3. 7 We used the chained Tornqvist price index in preference to the Laspeyres or Paasche indexes because it uses budget shares averaged between consecutive years, and therefore allows for changes in consumption patterns over time. 52 Transforming Bangladesh into a Middle Income Economy In [pi}] PO j where pTk denotes the Tornqvist price index for region k, 1 and 0 denote the two years of comparison, w'\and wkOj are the respective budget shares, and p\ and ~Oj are the respective prices for good j in the two years of comparison. Once the RES based price indexes for each region had been derived from the survey data, we took a weighted average of these and the non­ food component of the official CPI (disaggregated by urban and rural sectors) to derive region-specific cost of living indexes for 1995-96 and 2000, the relative weights being the budget shares of covered goods in each region for the RES price index, and balance (i.e. one minus these budget shares) for the non-food CPl. The composite price indexes were then used to update the 1991-92 CBN poverty lines to 1995-96 and 2000. 8 The derived composite price indexes show cost of living in Bangladesh to have increased by, on average, about 16 per cent between 1991-2 and 1995-96, and by about 12 per cent between 1995-96 and 2000. Note that the overall 30 per cent increase in price level between 1991-92 and 2000 implied by these indexes is somewhat lower than the 35 per cent increase in the GDP deflator over the same period, and much lower than the 52 per cent increase in the overall CPI. We will return to the implications for poverty trends of this difference between the composite price index and the CPI in Section IV. Poverty and Inequality Trends Readcount rates based on both the upper as well as lower poverty lines show poverty in Bangladesh to have declined considerably during the nineties (Table 2.1). In 2000, 50 per cent of Bangladesh's population was poor (as measured by the upper poverty line) as compared to 59 per cent in 1991-92. Similarly, 34 per cent of the country's population was very poor (i.e. below the lower poverty line) in 2000 as compared to 43 per cent in 1991-92. Thus, according to both the upper and lower poverty estimates, the incidence of poverty in Bangladesh declined by about 9 percentage points over the course of the decade. Throughout the decade, poverty in rural areas remained higher than in urban areas; however, the overall decline in poverty incidence over time was roughly equal across the two sectors. 9 8 The composite price indexes, as well as the eBN poverty lines for each region derived using these, are presented in Appendix Tables A2.4 and A2.5 respectively. 9 During the 1990s, the overall decline in poverty in Bangladesh as a whole (9.0 per cent) was greater than in either urban (8.3 per cent) or rural (8.2 per cent) areas because (0 the Poverty Trends during the Nineties 53 Table 2.1 Trends in CBN Poverty Measures Upper Change Lower poverty line (upper line) poverty line ~ - , ~ .9(,0 .9 Q) S~ . ~ - (,0 , 0) ~ ~~ 8 ~ ,.:.~ ~8 ~ - ­ Q)~ ~ ~ - 0) - - 0)0) 0)- '<::;) ~C\J 0) .S So 0 Q) ~ 0) Headcount rate (Po) National 58.8 51.0 49.8 -7.8 -1.2 -9.0 42.7 34.4 33.7 Urban 44.9 29.4 36.6 -15.5 +7.2 -8.3 23.3 13.7 19.1 Rural 61.2 55.2 53.0 -6.0 -2.2 -8.2 46.0 38.5 37.4 Poverty gap (P 1) '---.-~- .. ..- - - . - - - -.. National 17.2 13.3 12.9 -3.9 -0.4 -4.3 10.7 7.6 7.3 Urban 12.0 7.2 9.5 -4.8 +2.3 -2.5 4.9 2.6 3.8 Rural 18.1 14.5 13.8 -3.6 -0.7 -4.3 11.7 8.6 8.2 Squared poverty gap (P2) National 6.8 4.8 4.6 -2.0 -0.2 -2.2 3.9 2.5 2.3 Urban 4.4 2.5 3.4 -1.9 +0.9 -1.0 1.5 0.7 1.2 Rural 7.2 5.3 4.9 -1.9 -0.4 -2.3 4.3 2.8 2.6 The poverty gap (P t ) estimates how far below the poverty line the poor are on average as a proportion of that line. The squared poverty gap (Pz> takes into account not only the distance separating the poor from the poverty line, but also inequality among the poor. Trends in these measures broadly mirror those observed with the headcount rates. Both measures confirm that urban poverty remained lower than rural poverty throughout the decade. In addition, however, these measures also suggest that rural areas experienced greater reductions than urban areas in the depth and severity of poverty. Plotting the cumulative distributions for monthly real per capita expenditures (peE) in Bangladesh (national, urban, and rural, respectively) for the three years confirms that trends in poverty between 1991-92 and 2000 (as well as between 1991-92 and 1995-96) are robust to the choice of share of population living in urban area~ increased significantly during the period, and (ii) the incidence of poverty in urban areas was considerably lower than in rural areas. 54 Transforming Bangladesh into a Middle Income Economy National, by Year Urban, by Year -1991-92 -1995·96 - 2000 -1991·92 1995-96:!~;:;;::=~....:== '0 c: o .8 ~ .8 ~~ .6 oc: £:.~ .6 CD:; CD:; ~ !itA !it.4 .2: ,l!!a. ~ ~ 0 - - ,14') ,0 0 ~ 0) 0) ~~ -0) .t; - - 0) 0) Ql 0) 0) <:\j 0)0) 0)­ 0) eSC National 56.8 49.3 44.2 -7.5 -5.1 -12.6 45.5 34.8 30.6 Urban 56.5 28.3 31.2 -28.2 3.1 -25.3 41.6 14.0 16.8 Rural 56.9 53.5 47.5 -3.4 -6.0 -9.4 46.1 38.9 34.1 interesting to see that - as was the case with the DCI as well as CBN poverty estimates the CBN income-based estimates also show urban poverty to have increased between 1995-96 and 2000. In sum, while the DCI poverty estimates provide mixed support for the CBN estimates, the CBN methodology as well as CPI based estimates shows virtually no progress in poverty reduction in Bangladesh over the nineties. Not only does this run counter to the expected drop in poverty suggested by the National Accounts which show a 3 per cent rise in per capita incomes per year, but these alternate poverty estimates are also sharply at odds with other indications within the same datasets showing considerable improvement in living conditions in Bangladesh during this period. These findings are presented and discussed in the following section. Other Evidence of Changes in Living Standards One ofthe main drawbacks of the DCI method is that it makes no allowance for improvements in composition of the food bundle consumed. Analysis of the average quantities of different food items consumed reveals that per capita consumption of virtually all major food groups (with the notable exception of rice, wheat, and pulses) increased substantially during this period (Figure 2.3). For instance, between 1991-92 and 2000, per capita consumption of fish increased by 9 per cent, meat increased by 48 per cent, poultry increased by 120 per cent, milk increased by 55 per cent, cooking oil by 26 per cent, while sugar consumption increased by 11 per cent. While per capita wheat consumption dropped considerably in 2000,20 20 Part of the decline in wheat consumption is probably due to two important factors: (i) sharp rise in the relative price of wheat in relation to rice. and (ii) lower distribution of wheat through the various food assistance programmes. During a period of bumper rice production and large rice stocks in the Public Food Distribution System (PFDS). rather than curtail procurement, in some instances the government resorted to distributing rice instead of wheat. 66 Transforming Bangladesh into a Middle Income Economy 16000 I 14000 . 12000 10000 8000 6000 4000 2000 o U R U2 R2 o U R U2 R2 Rice FIsh 1200 1600 1000 1400 800 1200 1000 600 800 400 600 400 200 200 o o o U R U2R2 o U R U2R2 .J1l1k Wheat 600 700 500 600 500 400 400 300 300 200 200 100 100 o o o U R U2 R2 o U R U2 R2 Meat 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 00 U R U2R2 Vegetables Poultry 2500 800 700 2000 600 500 1500 400 300 1000 200 500 100 o o OUR U2 R2 o U R U2R2 Pulses Potatoes .1990-91 .1995-96 02000 0: Overall; U: Urban; R: Rural; U2: Urban: Bottom 2 Quintiles; R2: Rural: Bottom 2 Quinliles Figure 2.3 Average Quantities Consumed (grams per capita per month) Poverty Trends during the Nineties 67 and rice consumption declined marginally (by about 3 per cent), consumption ofpotatoes increased by 25 per cent during this period. Similarly, the 13 per cent decline in consumption of pulses is probably due to substitution away from pulses towards higher-value sources of protein. Consumption by the poor also increased considerably over the nineties. Figure 2.3 also reports per capita food consumption for households in the bottom two quintiles in urban and rural areas, and shows that the trend reported above was not confmed to upper income groups only­ consumption of fish, meat, pOUltry, and milk by the poor also increased considerably over the nineties. In addition to confirming improvements in living conditions in Bangladesh during the nineties, the graphs also support the earlier finding ofrapid progress during the first half ofthe decade, but then modest decline in poverty between 1995-96 and 2000. For instance, while there are indications ofconsiderable improvement in consumption patterns between 1991-92 and 1995-96, the pattern over the latter half of the nineties is much murkier. Finally, the graphs also help vividly illustrate the difference in living conditions between urban and rural areas. As indicated by the CBN poverty estimates presented earlier, virtually all households in the bottom two quintiles in urban and rural areas fall below the poverty line. However, across these two groups, there are striking differences in consumption patterns: levels of consumption of rice, wheat, vegetables, and pulses for the urban and rural poor are fairly similar; however, on average the urban poor consume considerably more high-value food items like meat, fish, poultry, milk, oils, and sugar. In contrast to what the DCI poverty estimates suggest, the poor in urban areas, at least in terms of the range and quantities of different food items they consume, appear to be considerably better off than their counterparts in rural areas. SECTION V: COMPARING BANGLADESH TO SOUTH ASIA AND EAST ASIA Bangladesh's pace of poverty reduction compares favourably with its South Asian neighbours (see Box 2.1 regarding comparability of poverty estimates). The decline in income poverty of about one percentage point per year during the nineties is in sharp contrast to the virtual stagnation Bangladesh experienced during the eighties. 21 The reduction in poverty in World Bank estimates show poverty in Bangladesh to have been stagnant at 59 21 Earlier per cent between 1983-84 and 1991-92 (World Bank, 1999). Similarly, RavaIlion and Sen 1996 estimate that rural poverty in Bangladesh declined only marginally from 54 per cent in 1983-84 to 53 per cent in 1991-92. 68 Transforming Bangladesh into a Middle Income Economy Box 2.1 Are Poverty Estimates Across Countries in South Asian Comparable? While trends in poverty reduction are comparable across countries in South Asia, estimates of poverty incidence levels are not. The national statistical offices in India, Pakistan, and Bangladesh all prepare poverty estimates using data from fairly similar national household surveys conducted on a regular basis in their respective countries. In India, the NSSO Consumer Expenditure Survey Series is used to estimate the incidence of poverty in the country. The Government of India Planning Commission's latest estimates using the 1999­ 2000 survey show poverty in India to be 26.1 per cent. In Pakistan, the Household Income and Expenditure surveys (HIES) conducted by the Federal Bureau of Statistics are used to estimate the incidence of poverty. Using data from the 1998-9 HIES, the incidence of poverty in the country was estimated to be 32.6 per cent. In Bangladesh, two measures of poverty are estimated by BBS, corresponding to the upper and lower poverty lines. Using the upper poverty line, poverty was estimated to be 49.8 per cent in 2000, while the lower poverty line yielded estimates of extreme poverty of 33.7 per cent. Is poverty in Bangladesh so much higher than in India or Pakistan, as indicated by these estimates? Cross-country comparisons of poverty are fraught with complex measurement and comparability issues and,amongst other factors, depend on the yardstick used to assess poverty levels in the respective countries. However, comparing the poverty lines used across these three countries suggest that part of the reason why poverty estimates in Bangladesh are so much higher than in either India or Pakistan is that a considerably higher poverty line is used to assess poverty (see table below). In US dollar terms, the upper poverty lines in use in Bangladesh are considerably higher than those in India and Pakistan (at prevailing exchange rates, not PPP-adjusted like the often used dollar 1 per person per day line) Poverty Lines in India, Pakistan, and Bangladesh Country Year Poverty line per Poverty line US $ Ratio to upper capita per month prevailing line (in US $) in (local currency) exchange rate) Bangladesh India Urban 1999-2000 Rs 454 9.88 0.62 Rural Rs 328 7.14 0.57 Pakistan Urban 1998-9 Rs 665 13.27 0.84 Rural Rs 589 11.76 0.95 Bangladesh Upper line Urban: 2000 Tk832 15.85 1.00 Rural Tk 652 12.42 1.00 Lower line Urban Tk628 11.96 0.75 Rural Tk 549 10.46 0.84 Poverty Trends during the Nineties 69 Bangladesh during the nineties also compares favourably with other countries in the region. In India, where the economy grew at about 6 per cent per annum during the nineties, consensus is emerging that poverty declined by roughly 5-10 percentage points over a 6 year period between 1993-94 and 1999­ 2000, a magnitude similar to that observed in Bangladesh. However, in Pakistan where the rate of GDP growth has slowed down considerably in the latter part of the nineties, recent evidence suggests that poverty has more or less stagnated over the nineties. And in Sri Lanka, poverty declined at a considerably slower pace, by 6 percentage points between 1985 and 1995. How do non-income indicators of living standards in Bangladesh compare to other countries? Using measures of stunting, wasting, and underweight children from Demographic and Health Surveys carried out in India and Bangladesh in 1998-99 and 1999-2000 respectively, Bangladesh compares favourably with India (Table 2.9). The comparison with Pakistan Table 2.9 Bangladesh and South Asia: Comparison of Selected Indicators of Child Nutrition Nutrition status Bangladesh India Pakistan Sri Lanka Indicator _ _. ... 1999-2000 1998-99 -1990-91 ....- -~ . 1987 Stunting (height for age) % below 2 std. Deviations 50 57 57 34 % below 3 std. Deviations 20 32 36 ~--~--- .. ---~~-----~~--~---. -~---~-~---. "' .. ~--~---. Wasting (weight for height) % below 2 std. Deviations 9 13 10 13 % below 3 std. Deviations 1 2 Underweight (weIght for age) % below 2 std. Deviations 56 58 46 48 % below 3 std. Deviations 17 24 19 Source: Various DHS reports. For comparability, comparison limited to children 24-35 months (24-36 for Sri Lanka). Table 2.10 International Comparisons of Selected Development Indicators Indicator Bangladesh China India Pakistan Thailand Vietnam GNP per capita (US$) 370 780 450 470 1,960 370 Population growth (%) 1.6 1.1 1.8 2.5 1.2 1.8 Urban population (% of total) 24 32 28 36 21 20 Health Male life expectancy at birth (years) 58 68 62 61 70 66 Infant mortality (per 1.000 live births) 73 31 70 91 29 34 Under 5 mortality rate (per 1.000 (96) 36 83 120 33 42 Access to water and sanitation (% of population with access) Access to improved water source 84 90 81 60 89 36 Access to sanitation 35 21 16 30 96 21 ~--------- Literacy and Education Male illiteracy (% of age 15 & older) 49 9 33 42 3 5 Female illiteracy (% of age 15 & older) 71 25 57 71 7 9 Net primary school enrolment 75 100 77 88 100 Notes: Estimates are from 1999. or most recent estimates reported in the Database. Source: World Development Indicators. Poverty Trends during the Nineties 71 and Sri Lanka is mure mixed. While Bangladesh has lower rates of stunting and wasting than Pakistan, the percentage of underweight children is far greater. Comparisons of other development indicators show that Bangladesh, with a lower GNP per capita, has done reasonably well on some dimensions but lags with respect to others when compared with other South Asian countries (Table 2.10). It has lower population growth and mortality rates than both India and Pakistan. Access to improved water supply is better in Bangladesh, although this success is being threatened by the problem of arsenic contamination ofgroundwater. Adult literacy remains a problematic area relative to other countries, although Bangladesh has made significant strides in improving gender parity in enrolments. While cross-country comparisons always require some care, the recent experiences of Vietnam, a country with the same GNP per capita as Bangladesh, may point to what is possible. Between 1993 and 1998, Vietnam experienced a 21 percentage point drop in poverty, spurred in large part by an ambitious reform programme that included land reform, liberalisation of agricultural input and output markets, freeing up the informal sector, and equitable investments in human capital. 22 Between 1992 and 1998, the average annual GDP growth rate in Vietnam was a spectacular 8.4 per cent, with agricultural growth averaging 4.5 per cent, industrial growth 13 per cent, and the services sector growing by 8 per cent per annum. In addition to progress in reducing consumption based poverty, Vietnam has also achieved substantial progress in educational and health outcomes, which are now comparable to those of other East Asian countries that have much higher income levels. Vietnam's experience suggests that the poverty reduction payoffs to further reforms and institutional development in Bangladesh could be substantial. SECTION VI: SUMMARY OF POVERTY TRENDS ANALYSIS Both survey-based CBN poverty estimates as well as those based on the National Accounts show that the nineties were a period of declining poverty in Bangladesh. The proportion of the very poor declined from 43 per cent in 1991-92 to 34 per cent in 2000, while the proportion of the poor fell from 59 per cent to 50 per cent. Poverty in rural areas continues to be higher than in urban areas, but has declined at a fairly rapid rate in both sectors during the nineties. The improvement in living conditions is evidenced not only by increases in PCE, but also by a shift in composition of the food bundle consumed by the poor towards more high value items. 22 See World Bank 2000. 72 Transforming Bangladesh into a Middle Income Economy While the incidence of poverty has fallen considerably during the decade, examination of the total number of individuals living below the poverty line reveals a more sobering picture the total population living below the upper poverty line in 2000 remained virtually unchanged (at about 63 million) compared to 1991-92, while the population living below the lower poverty line declined somewhat from 45.2 million in 1991-92 to 42.5 million in 2000. The analysis carried out has also brought to light a number of puzzles that warrant further attention: Discrepancy with National Accounts The discrepancy between the HES and National Accounts series relates to the pattern of growth over the decade, with the HES surveys indicating much more modest progress at poverty reduction over the latter half of the decade as compared to the NA statistics. Assessing which of the two gives the correct picture of poverty trends is problematic as there exists supporting evidence for both standpoints. 23 On the one hand, since 1995-96, the Bangladesh economy as a whole (agriculture in particular) has performed quite well. Inflation has remained within single digits, the price of rice is virtually unchanged in real terms, and per capita availability of essential food items has improved considerably. Secondary data on wages and agricultural incomes also points toward improvements in living standards through the nineties. On the other hand, HES data suggest that much of the increase in PCE as well as improvement in composition of the average food bundle consumed has taken place over the ftrst half of the nineties. Similarly, while the share of total expenditures allocated to food has gone down considerably over the decade, most of this decline took place over the ftrst half. Finally, enrolment rates derived from the two datasets suggest that the proportion of primary school age children attending school has declined between 1995-96 and 2000. One important contra-indication to the otherwise bleak picture painted by the HES series for the latter half of the nineties is the 34 per cent incr:ease in wages (in nominal terms) and 23 per cent increase in median crop revenues per capita between 1995-96 and 2000, in contrast to the much smaller 15 per cent increase in mean consumption. Amongst the possible reasons the latest HES dataset may underestimate improvement in living standards could be that the constructed welfare measure excludes important items for which expenditure has increased considerably in recent years (purchase of livestock and other assets damaged or destroyed by the 1998 (e.g. differences in items included in NA consumption 23 In addition, there are several reasons v. survey consumption measures, and various sources of measurement errors in both the NA and survey data) why it is not surprising to find a discrepancy between National Accounts and household survey based consumption estimates (Deaton, 2000). Poverty Trends during the Nineties 73 floods), that it does not fully capture improved access to publicly provided goods and services, or that it does not fully account for increase in household savings rates, etc. 24 Further analysis of shifts in employment patterns indicated by the most recent Labour Force Survey, analysed in conjunction with the poverty profile yielded by the 2000 HIES survey (in particular, the relationship between poverty and sources of household income) may be helpful in better understanding the relationship between aggregate growth and poverty during the latter half of the decade. Discrepancy with the CPI A second important issue worth noting concerns the discrepancy in inflation estimates from the survey data and the official CPI series. As pointed out earlier, the Thnqvist indices derived from the survey data suggest that the price level in Bangladesh increased by about 20 per cent between 1991-92 and 2000 while the CPI series show a rise of about 52 per cent ov;r the same period. Part of the discrepancy relates to the fact that the CPI is a Laspeyres index, which tends to overestimate inflation over long time periods since budget-share weights are fixed at the base year level. In Bangladesh, these weights have not been revised since 1985-86, and may be quite out of date in relation to current consumption pattern. Given the widespread use of the CPI, updating these weights merits serious consideration by BBS. Impact of Rural-Urban Migration on Poverty Estimates Finally, the last issue we'd like to draw attention to concerns differentials in living standards between the urban and rural sectors. Notwithstanding the observed stagnation in urban poverty rates in recent years, living standards appear to be considerably higher in urban as compared to rural areas (as suggested by greater consumption of higher-value food items by the urban poor compared to the rural poor). The influx of migrants from rural to urban areas of Bangladesh appears to have continued unabated through the nineties. Results from the 'quick-count' carried out for the recent population census suggest that the share of the country's urban population has risen from around 14 per cent in 1991-92 to over 20 per cent in 2000. A final question we would like to pose is whether the rural-urban cost of living differential implicit in our choice of poverty lines may have led to a slight underestimation of the decline in poverty. Recall that our 24Unfortunately. since comparable data on wages and incomes are not available in the 1991­ 92 HES, it is not possible to investigate the same trends over the 1991-95 period. 74 Transforming Bangladesh into a Middle Income Economy choice of poverty lines for the two sectors was tied to the rural-urban differential embedded in the 1991-92 poverty lines which we updated to 1995-96 and 2000 using region-specific cost of living indices. These lines imply that the cost of living is any where up to 41 per cent higher in urban as compared to rural areas. However, what if this overestimates the difference in cost of living across the two sectors? Consider the case of a person who is just above the poverty line in the rural sector, and who moves to the urban sector where he obtains a job generating a real income gain less than the difference in poverty lines across the two sectors.25 Though that person may be better off in his new residence, the poverty measures used will show an increase in both urban as well as rural sectors (there is one less non-poor person in rural areas, and one more poor person in urban areas). Further investigation into the extent and nature of migration trends in Bangladesh (What types of individuals moved? What jobs were they engaged in before moving to urban areas? What types ofjobs did they take up in -their new residence? etc.) will undoubtly be an important topic for future research. ACKNOWLEDGEMENTS We would like to thank Faizuddin Ahmed, Syed Nizamuddin, Zahid Hussain, Kapil Kapoor, Martin Ravallion, Zaidi Sattar, Shekhar Shah, Binayak Sen, and participants of the Bangladesh Poverty Assessment workshops for useful comments and suggestions. REFERENCES Bangladesh Bureau of Statistics (BBS). Summary Report of the Household Expenditure Survey 1995-6. Dhaka, 1997. ---,1999 Statistical Yearbook of Bangladesh. Dhaka, 2001. Bidani, B. and M. Ravallion. 'How Robust is a Poverty Profile?' The World Bank Economic Review, 8, 1994, pp.75-102. Deaton, A. Counting the World's Poor: Problems and Possible Solutions, Princeton University: New Jersy, 2000. Deaton, Angus and A. Tarozzi. Prices and Poverty in India. Research Program in Development Studies, Princeton University, New Jersy, 2000. Ravallion, M. Poverty Comparisons. Switzerland: Harwood Academic Press, 1994. Ravallion, M. and B. Sen. 'When Method Matters: Monitoring Poverty in Bangladesh,' Economic Development and Cultural Change, 44: 1996, pp.761-792. World Bank. From Counting the Poor to Making the Poor Count. South Asia 25 The argument outlined follows the one outlined in Ravallion 1994. Poverty Trends during the Nineties 75 Poverty Reduction and Economic Management Unit, Washington, DC, 1999. - - - . Vietnam Development Report 2000: Attacking Poverty. Washington, DC, 2000. 76 Transforming Bangladesh into a Middle Income Economy APPENDICES Table A 2.1 Budget Shares of Items with Unit-Value Information in the HES Region Budget shares (%) ~-~ ... 1991-92 1995-96 2000 SMA Dhaka 65.3 53.4 52.2 OU Dhaka 74.8 61.0 58.7 R. Dhaka 75.9 72.3 71.5 R. Faridpur Tangail Jamalpur 82.4 74.8 72.2 SMA Chittagong 60.6 63.2 55.8 OU Chittagong 71.8 62.2 58.4 R. Sylhet Comilla 78.0 76.0 66.6 R. Noakhali Chittagong 74.2 71.5 63.1 U. Khulna 68.6 64.9 58.3 R. Barisal Pathuakali 80.3 73.1 66.3 R. Khulna Jessore Kushtia 75.3 70.4 68.3 U. Rajshahi 71.8 61.5 60.8 R. Rajshahi Pabna 76.7 70.2 71.4 R. Bogra Rangpur Dinajpur 78.7 71.4 68.5 Table A2.2 Relative Weights of Items Covered in the Price Index Food grains Vegetables Pulses Fish Eggs Region 91 95 00 91 95 00 91 95 00 91 95 00 91 95 00 SMA Dhaka 35.2 29.7 29.4 9.4 8.4 8.5 4.3 3.5 3.6 12.3 13.9 13.5 1.4 2.1 2.2 au Dhaka 44.9 37.8 33.3 8.0 7.8 7.7 3.6 3.4 3.4 8.1 10.6 11.0 1.3 1.7 1.6 R. Dhaka 52.8 44.9 39.6 6.7 8.4 8.0 2.9 2.3 3.2 7.0 10.0 11.4 0.7 0.8 1.8 R. Faridpur Tangail Jamalpur 55.5 49.9 43.0 6.7 8.0 8.2 3.0 2.2 2.6 6.9 8.3 9.5 0.7 0.9 1.1 SMA Chittagong 41.7 35.0 32.8 7.8 8.4 9.0 2.6 2.4 2.9 10.2 11.1 11.0 0.6 1.1 1.3 au Chittagong 41.0 38.2 32.1 6.6 7.7 9.7 3.5 2.7 3.1 11.3 11.3 13.0 1.0 2.0 0.8 R. Sylhet Com ilia 44.4 45.2 36.3 8.0 7.7 9.3 3.0 2.5 2.7 9.9 10.7 13.0 1.1 0.7 0.7 R. Noakhali Chittagong 45.4 41.3 37.9 7.3 8.9 9.3 2.6 2.1 3.2 9.7 12.7 13.4 0.7 0.8 0.7 U. Khulna 42.8 39.5 31.9 8.1 8.6 8.8 3.2 3.0 3.0 9.8 11.6 13.0 1.1 1.7 1.9 R. Barisal Pathuakali 49.2 45.6 38.8 6.2 7.5 9.8 3.9 3.4 3.1 8.3 11.7 13.5 0.9 1.0 1.2 R. Khulna Jessore Kushtia 53.1 47.8 42.8 8.3 9.0 9.2 2.2 2.0 2.1 7.7 9.5 10.0 0.7 0.9 1.1 U. Rajshahi 42.4 39.0 36.1 9.1 8.6 9.4 3.0 2.8 2.5 7.8 8.4 9.0 1.0 1.3 1.7 R. Rajshahi Pabna 54.2 48.9 45.0 8.2 9.3 8.9 1.9 2.3 2.6 4.8 6.7 8.3 0.6 0.9 1.0 R. Bogra Rangpur Dinajpur 55.7 50.7 47.6 8.6 8.8 9.8 2.3 1.7 1.5 5.8 6.5 7.1 0.7 0.9 1.1 (Contd. Table A2.2) Meat Salt& spices Milk Sugar Cooking oils 91 95 00 91 95 00 91 95 00 91 95 00 91 95 00 SMA Dhaka 4.9 6.9 8.4 6.5 5.0 6.4 2.1 3.6 3.7 2.2 2.4 2.1 4.4 4.4 3.9 au Dhaka 2.1 4.9 5.9 6.8 4.9 5.8 3.2 3.0 4.4 2.2 1.4 1.9 3.9 3.7 3.5 R. Dhaka 2.0 2.9 4.2 6.8 5.4 6.9 1.9 3.4 3.3 1.3 1.7 1.3 3.2 3.2 3.6 R. Faridpur Tangail Jamalpur 1.6 2.1 3.5 6.3 5.6 7.0 1.8 2.5 2.7 1.3 1.2 1.5 3.2 3.3 3.2 SMA Chittagong 3.7 4.8 7.0 4.6 4.1 6.9 3.9 3.0 2.7 2.0 1.9 2.6 2.8 2.8 3.3 au Chittagong 3.5 4.6 6.1 4.6 4.4 7.0 2.6 2.7 2.5 1.9 1.2 2.0 3.4 2.7 2.9 R. Sylhet Comilla 3.0 2.5 3.9 7.7 5.1 6.9 1.8 2.2 2.7 1.6 1.5 2.0 3.8 2.9 3.2 R. Noakhali Chittagong 3.4 3.9 6.1 6.0 5.1 6.4 1.3 1.4 1.7 1.9 1.5 1.2 2.6 2.7 2.9 U. Khulna 3.7 5.1 62 6.4 4.3 6.4 2.2 2.6 2.6 1.6 2.1 2.0 4.6 3.9 4.1 R. Barisal Pathuakali 1.9 2.4 3.1 6.5 6.2 5.8 1.3 1.8 2.2 1.6 2.4 1.2 3.5 3.1 3.7 R. Khulna Jessore Kushtia 2.2 3.4 4.2 5.7 4.4 6.3 0.7 1.7 2.2 1.5 2.2 1.7 4.3 4.2 3.7 U. Rajshahi 4.8 5.8 7.0 5.8 5.1 7.8 2.3 3.0 2.8 1.5 2.0 1.6 4.2 4.5 3.9 R. Rajshahi Pabna 2.8 3.6 4.1 5.8 4.7 8.4 1.4 2.3 1.9 1.3 2.0 1.4 3.0 3.7 3.8 R. Bogra Rangpur Dinajpur 3.0 4.1 4.2 5.5 4.9 6.1 1.3 2.1 2.1 1.8 1.5 1.4 2.9 2.7 2.9 (Contd.) (Contd. Table A2.2) Fruits Soft drinks Betel& Tobacco Fuels 91 95 00 91 95 00 91 95 00 91 95 00 SMA Dhaka 1.6 3.5 4.0 2.2 3.1 0.6 5.1 4.8 1.9 8.6 8.9 11.8 au Dhaka 0.7 2.6 5.0 1.5 1.2 0.4 4.4 4.8 3.2 9.2 12.2 12.8 R. Dhaka 1.0 2.0 2.8 0.3 0.8 0.1 4.8 4.7 3.1 8.7 9.7 10.7 R. Faridpur Tangail Jamalpur 1.0 1.4 2.0 0.3 0.4 0.1 5.3 5.1 3.2 6.3 9.0 12.2 SMA Chittagong 0.9 2.6 2.7 3.1 5.7 1.2 5.0 7.6 4.0 11.1 9.5 12.5 au Chittagong 1.7 2.5 2.7 2.5 4.1 0.9 7.2 6.8 4.6 9.3 9.1 12.7 R. Sylhet Com ilia 1.5 1.7 2.4 1.2 2.2 0.5 6.5 6.6 5.2 6.5 8.6 11.0 R. Noakhali Chittagong 1.1 1.5 1.8 3.3 3.3 0.4 8.2 6.8 4.1 6.5 8.1 10.8 U. Khulna 1.2 2.1 2.6 1.6 1.4 0.4 4.7 3.6 2.0 9.0 10.5 15.0 R. Barisal Pathuakali 0.9 0.9 2.6 1.4 1.1 0.2 5.6 5.7 3.7 8.7 7.3 11.0 R. Khulna Jessore Kushtia 1.0 1.4 1.6 0.3 0.3 0.2 4.2 3.6 1.9 8.0 9.5 13.1 U. Rajshahi 1.0 1.2 2.1 1.4 2.4 0.3 4.0 3.8 2.1 11.9 12.1 13.7 R. Rajshahi Pabna 0.3 1.8 1.2 0.4 0.5 0.2 5.1 3.2 1.6 10.1 10.2 11.7 R. Bogra Rangpur Dinajpur 0.5 1.0 1.9 0.5 0.8 0.1 3.8 4.1 2.8 7.5 10.3 11.4 Table A2.3 Selected Unit-Values (tk/unit) from the Surveys Coarse ice (kg) Masur (kg) Puti (kg) Hen's egg (per 10) Beet (kg) Region 91 95 00 91 95 00 91 95 00 91 95 00 91 95 00 SMA Dhaka 12 15 14 29 40 40 30 50 67 23 35 35 60 60 80 au Dhaka 12 14 12 28 40 40 20 40 50 19 30 30 50 60 70 R. Dhaka 12 13 12 28 40 40 20 25 40 13 30 30 48 60 70 R. Faridpur Tangail Jamalpur 12 12 12 28 40 40 20 32 40 13 25 30 48 50 70 SMA Chittagong 13 14 13 30 40 40 24 48 50 23 35 40 60 70 80 au Chittagong 12 14 13 28 40 40 23 40 50 23 30 40 60 70 80 R. Sylhet Comilla 12 13 12 28 40 40 20 27 40 13 25 30 60 60 80 R. Noakhali Chittagong 12 13 12 29 40 40 20 40 40 17 30 30 54 60 80 U. Khulna 12 13 12 28 40 40 22 30 44 19 30 30 48 60 70 R. Bansal Pathuakali 12 13 12 28 40 40 20 30 40 13 25 30 40 60 70 R. Khulna Jessore Kushtia 11 12 11 27 30 40 20 30 40 13 20 25 40 50 67 U. Rajshahi 12 12 12 28 40 40 22 30 40 18 30 30 40 50 63 R. Rajshahi Pabna 10 12 12 26 40 40 20 24 40 13 20 25 40 40 60 R. Bogra Rangpur Dinajpur 11 11 11 26 35 40 20 24 40 13 20 25 39 40 60 (Contd.) (Contd. Table A2.3) Potato (kg) Milk (kg) Sugar (kg) Mustard oil (Itr.) Salt (kg) -------~~ 91 95 00 91 95 00 91 95 00 91 95 00 91 95 00 SMA Dhaka 8 8 10 20 22 26 31 32 40 50 60 60 10 10 10 au Dhaka 7 8 10 13 16 20 31 36 32 50 60 60 10 10 10 R. Dhaka 7 8 8 12 16 16 30 32 32 55 60 60 8 10 10 R. Faridpur Tangail Jamalpur 7 8 8 12 13 16 31 40 32 53 60 60 7 8 7 SMA Chittagong 7 8 10 14 16 20 30 32 33 55 75 80 7 10 10 au Chittagong 7 9 10 14 16 24 32 30 36 52 67 50 8 10 10 R. Sylhet Comilla 8 8 10 12 16 20 30 32 36 59 60 60 8 10 10 R. Noakhali Chittagong 8 8 10 12 15 20 30 32 34 51 60 60 8 8 8 U. Khulna 7 8 8 14 16 18 30 30 32 58 60 60 8 10 10 R. Barisal Pathuakali 6 8 8 10 12 16 30 32 30 54 60 60 8 10 10 R. Khulna Jessore Kushtia 6 8 8 10 12 12 30 30 30 53 60 60 7 10 9 U. Rajshahi 7 7 8 12 14 16 30 30 32 50 60 60 9 10 8 R. Rajshahi Pabna 6 8 8 10 12 14 30 28 32 50 60 50 8 8 7 R. Bogra Rangpur Dinajpur 7 8 6 10 10 12 30 30 31 51 60 60 8 8 6 (Contd.) (Contd. Table A2.3) Bananas (kg) Coke/Fanta (no.) Prepared betel Kerosene (ltr.) 91 95 00 91 95 00 91 95 00 91 95 00 SMA Dhaka 15 20 20 7 8 15 0.5 0.5 1.0 18 18 16 au Dhaka 10 20 20 6 7 11 0.5 0.5 1.0 18 16 16 R. Dhaka 10 14 16 7 7 15 0.5 0.5 1.0 18 16 18 R. Faridpur Tangail Jamalpur 10 16 20 6 4 17 0.5 0.5 1.0 18 15 16 SMA Chittagong 9 20 27 7 8 15 0.5 0.5 1.0 18 16 16 au Chittagong 10 20 20 7 8 12 0.5 0.5 1.0 18 16 16 R. Sylhet Comilla 8 20 20 7 7 17 0.5 0.5 1.0 16 16 16 R. Noakhali Chittagong 10 20 20 6 7 15 0.5 0.5 1.0 16 16 16 U. Khulna 9 16 15 6 8 15 0.5 0.5 1.0 20 16 16 R. Barisal Pathuakali 8 12 20 5 8 15 0.5 0.5 1.0 16 16 15 R. Khulna Jessore Kushtia 10 12 12 6 8 19 0.5 0.5 0.5 20 20 16 U. Rajshahi 10 16 13 7 6 15 0.5 0.5 1.0 18 16 16 R. Rajshahi Pabna 8 16 10 7 7 15 0.5 0.5 0.5 20 17 18 R. Bogra Rangpur Dinajpur 10 12 13 6 7 14 0.5 0.5 0.5 20 16 16 Table A2.4 Composite Price Indices: 1991-92, 1995-96 and 1995-96 - 2000 1991-92 to 1995-96 1995-96 to 2000 Food Covered Covered HES budget sh. Non-food Composite FoodHES budget sh. Non-food Composite Region index (%) CPt price index index (%) CPt price index SMA Dhaka 1.20 59 1.20 1.20 1.10 53 1.16 1.13 Other Urban Dhaka 1.20 68 1.20 1.20 1.03 60 1.16 1.08 Rural Dhaka 1.12 74 1.26 1.16 1.07 72 1.20 1.11 Rural Faridpur Tangail Jamalpur 1.08 79 1.26 1.12 1.07 74 1.20 1.12 SMA Chittagong 1.20 62 1.20 1.20 1.09 59 1.16 1.12 Other Urban Chittagong 1.20 67 1.20 1.20 1.09 60 1.16 1.12 Rural Sylhet Com ilia 1.12 77 1.26 1.15 1.11 71 1.20 1.15 Rural Noakhali Chittagong 1.17 73 1.26 1.19 1.06 67 1.20 1.11 Urban Khulna 1.12 67 1.20 1.14 1.06 62 1.16 1.10 Rural Barisal Pathuakali 1.17 77 1.26 1.19 1.05 70 1.20 1.10 Rural Khulna Jessore Kushtia 1.16 73 1.26 1.19 0.98 69 1.20 1.05 Urban Rajshahi 1.07 67 1.20 1.11 1.08 61 1.16 1.12 Rural Rajshahi Pabna 1.13 73 1.26 1.17 1.04 71 1.20 1.10 Rural Bogra Rangpur Dinajpur 1.04 75 1.25 1.10 1.01 70 1.20 1.09 Table A2.5 CBN Poverty Lines: Updating 1991-92 Lines with the Composite Price Index (taka per person per month) 1991-92 1995-96 2000 ZL ZU ZL ZU ZL ZU SMA Dhaka 480 660 574 791 649 893 Other Urban Dhaka 399 482 480 580 521 629 Rural Dhaka 425 512 492 593 548 659 Rural Faridpur Tangail Jamalpur 432 472 484 529 540 591 SMA Chittagong 523 722 627 867 702 971 Other Urban Chittagong 517 609 619 730 694 818 Rural Sylhet Comilla 432 558 499 644 572 738 Rural Noakhali Chittagong 438 541 522 645 582 719 Urban Khulna 482 635 552 727 609 803 Rural Barisal Pathuakali 413 467 494 558 546 616 Rural Khulna Jessore Kushtia 420 497 499 592 527 624 Urban Rajshahi 446 582 496 647 557 726 Rural Rajshahi Pabna 459 540 535 630 586 690 Rural Bogra Rangpur Dinajpur 426 487 468 535 510 582 Note: ZL is the lower poverty line; ZU is the upper poverty line. Amounts are in Tk per person per month. Table A2.6 Poverty Lines: Reapplying the CBN Methodology to Each Dataset (taka per person per month) Region Ratio: 1995-96 to Ratio: 2000 to 1991-92 1995-96 2000 1991-92 1995-96 --------­ ZL ZU ZL ZU ZL ZU ZL ZU ZL ZU SMA Dhaka 480 660 613 950 746 1,236 1.28 1.44 1.22 1.30 au Dhaka 399 482 584 931 724 979 1.46 1.93 1.24 1.05 R. Dhaka 425 512 523 661 575 712 1.23 1.29 1.10 1.08 R. Faridpur Tangail Jamalpur 432 472 521 604 581 703 1.21 1.28 1.12 1.16 SMA Chittagong 523 722 561 749 646 937 1.07 1.04 1.15 1.25 au Chittagong 517 609 564 704 659 842 1.09 1.16 1.17 1.20 R. Sylhet Com ilia 432 558 515 584 614 848 1.19 1.05 1.19 1.45 R. Noakhali Chittagong 438 541 548 638 628 858 1.25 1.18 1.15 1.34 U. Khulna 482 635 541 779 646 872 1.12 1.23 1.19 1.12 R. Barisal Pathuakali 413 467 522 639 600 748 1.26 1.37 1.15 1.17 R. Khulna Jessore Kushtia 420 497 481 563 556 694 1.15 1.13 1.16 1.23 U. Rajshahi 446 582 499 628 576 752 1.12 1.08 1.15 1.20 R. Rajshahi Pabna 459 540 480 582 522 624 1.05 1.08 1.09 1.07 R. Bogra Rangpur Dinajpur 426 487 457 570 529 671 1.07 1.17 1.16 1.18 Note: ZL is the lower poverty line; ZU is the upper poverty line. Amounts are in Tk per person per month. Table A2.7 Share of Household Budget Allocated to Food Items Nominal Tk per Overall population Bottom 40% person per month 1991-92 1995-96 2000 1991-92 1995-96 2000 peE 550 764 876 326 427 473 peE Food 353 432 463 236 287 305 peE Non-Food 197 332 413 89 140 168 Share of peE on food 64.2 56.5 52.8 72.6 67.2 64.6 Table A2.B Poverty Lines: Updating 1991-92 Lines with the CPt 1991-92 1995-96 2000 Region ZL ZU ZL ZU ZL ZU SMA Dhaka 480 660 590 812 729 1004 OU Dhaka 399 482 491 593 607 733 R. Dhaka 425 512 523 630 647 779 R. Faridpur Tangail Jamalpur 432 472 531 580 656 717 SMA Chittagong 523 722 643 889 794 1098 OU Chittagong 517 609 635 750 785 926 R. Sylhet Comilla 432 558 532 686 657 848 R. Noakhali Chittagong 438 541 539 665 666 822 U. Khulna 482 635 593 782 732 966 R. Barisal Pathuakali 413 467 509 574 628 709 R. Khulna Jessore Kushtia 420 497 516 612 638 756 U. Rajshahi 446 582 549 715 679 884 R. Rajshahi Pabna 459 540 564 665 697 821 R. Bogra Rangpur Dinajpur 426 487 524 599 648 740 Note: ZL is the lower poverty line; ZU is the upper poverty line. Amounts are in Tk per person per month. Table A2.9 Poverty Lines: Updating 1991-92 Lines with the HES-TP (taka per person per month) 1991-92 1995-96 2000 Region ZL ZU ZL ZU ZL ZU SMA Dhaka 480 660 577 795 641 883 OU Dhaka 399 482 485 585 501 605 R. Dhaka 425 512 479 576 512 617 R. Faridpur, Tangail, Jamalpur 432 472 469 513 501 548 SMA Chittagong 523 722 634 876 691 956 OU Chittagong 517 609 624 736 682 805 R. Sylhet, Comilla 432 558 486 627 540 697 R. Noakhali, Chittagong 438 541 512 632 544 671 U. Khulna 482 635 543 717 574 757 R. Barisal, Pathuakali 413 467 487 550 512 578 R. Khulna, Jessore, Kushtia 420 497 490 580 479 568 U. Rajshahi 446 582 480 626 517 674 R. Rajshahi, Pabna 459 540 523 616 544 641 R. Bogra, Rangpur, Dinajpur 426 487 447 510 453 518 Note: ZL is the lower poverty line; ZU is the upper poverty line. Amounts are in Tk per person per month. PART II The Strategy for Accelerating Growth 3 Analysis of Growth Experience Sandeep Afahajan SECTION I: INTRODUCTION Growth matters because it affects the poor. GDP growth is almost universally acknow ledged to be the most potent economic force in the fight against poverty. Across the globe, examples abound of countries that achieved high GDP growth rates over sustained periods and made visible dents in their poverty rates; conversely, nations that experienced low or negative growth saw minimal reductions and, in some cases, even increases in poverty. China, with an average annual per capita growth of 8.5 per cent during 1985-98, saw a reduction in its rural poverty rate (per cent of rural population living under dollar 1 a day) from 45 per cent to 24 per cent over the period. Average per capita growth in Thailand was an impressive 5 per cent p.a. during 1981-98, • a period over which its national poverty rate fell from 26 per cent to almost nil. In Indonesia, the share of the population living in poverty in 1993 was a quarter of its level in 1984, while per capita growth averaged 5.1 per cent over these nine years. l Even in Bangladesh, the highest rates of poverty reduction and GDP growth since independence coincided during the 1990s. On the other hand, in Niger, where per capita growth averaged -2.3 per cent during the 1991-96 period, the poverty rate increased from 59 per cent to 73 per cent. Testing the growth-poverty relationship across countries, Dollar and Kraay (200 I) convincingly show that higher income growth, on average, is equally shared by the poor. More specifically, on average, a one per cent increase in per capita GDP corresponds to a one per cent increase in the income of the bottom quintile of the population. In Bangladesh, in fact, since both Thailand and Indonesia poverty rates rose considerably after the 1997 crisis, further I In substantiating that income growth and poverty are correlated. 92 Transforming Bangladesh into a Middle Income Economy independence the share of the bottom quintile in total income growth has been more than even one, i.e. for a one per cent increase in income, on average, the income ofthe bottom quintile increased by 1.25 per cent. Figure 3.1 shows the change in per capita income of the bottom quintile for Bangladesh and some comparator countries over the 1972-2000 period. Income growth accounted for about 65 per cent of the increase in the incomes of the poor in Bangladesh, 71 per cent in Thailand, and 84 per cent in Indonesia, the rest being accounted for by changes in the bottom quintile's share in national income. In the case of China and India, the share of the bottom quintile fell, as a result of which about a third of the gains due to income growth were reversed in China and about a tenth in India. Even then, the per capita income in the bottom quintile in China increased by about 225 per cent between 1972-2000, almost double the gain seen in Bangladesh. 2000 1500 Change Due to Income Growth Effect ~- -;;CI. 1000 15 It E..fl ~'I: rJl c. 500 ::> 0 -500 Bangladesh China India Indonesia Thailand Income of the poor is measured as the income of the average person in the bottom quintUe of the income distribution Figure 3.1 Trends in GOP Per Capita Level and Growth Source: Author's calculations. These compelling stylised facts point to the enormous development possibilities that improved and sustained growth performance present for Bangladesh. Higher GDP growth is also needed if the government's own poverty reduction objectives are to be met. The government's estimates, as presented in its Interim Poverty Reduction Strategy Paper (J-PRSP), show that GDP growth of about 7 per cent would be needed to meet the country's stated poverty reduction objectives. It appears important, therefore, to better understand the growth process in Bangladesh and to identify the economic factors that are most critical in unleashing its growth potential. This paper takes a step in that direction. Analysis of Growth Experience 93 SECTION II: SOME STYLISED FACTS Economic Growth in Bangladesh has been on a Rising Trend since Independence Per capita growth increased from an annual average of -0.7 per cent during the 1970s, to 1.1 per cent during the 1980s, and further to 3.0 per cent during the 1990s and 3.3 per cent during 2001-03. As seen in Figure 3.2, there was a sharp structural break in the growth process in 1990. Annual per capita growth averaged 1.1 per cent during 1973-89 and almost triple that much during 1990-2003. 6 4 160 ~ :II -..j(/J 2 150 ~!!!. c: o " CD () 140 3 !!l II> () ~ 0 130 ::-@l Q,. 2 120 6" !iiii' Ci) ~O 110 ~ " 4 100 Figure 3.2 Trends in GDP Per Capita Level and Growth Source: Author's calculations . . . . Which has been Increasingly Stable The growth process has also become increasingly stable. The standard deviation of per capita GDP growth fell from 3.9 during the 1970s to 1.1 during the 1980s and further to 0.6 during the 1990s. Volatility of growth during the decades of 1980s and 1990s was among the lowest in the world. Not surprisingly, volatility and GDP growth have shown a highly significant negative relationship (Figure 3.3). That lower volatility ofgrowth contributes to higher growth performance (in a causal sense) has now been well documented in the literature (Ramey and Ramey 1995). Also, Bangladesh is among a handful of countries that have managed to avoid even a single year of negative per capita growth since the early 1990s - remarkable feat for an economy that is persistently hit hard by weather related adverse shocks. 2 2 Bangladesh will need to remain vigilant against future growth crashes. A remarkable feature of international growth experience has been the lack of persistence in growth (Rodrik 1999). 94 Transforming Bangladesh into a Middle Income Economy 4 3 2 o ~ Per Capita Growth _ Std. Dev. of Per Capita Growth Figure 3.3 Volatility of GDP Growth and Growth are Negatively Correlated Note: Volatility of per capita growth in anyone year is calculated as the standard deviation of the per capita growth rates over five years; spanning that year and the 4 years preceding it. Source: Author's calculations and World Bank Data Bangladesh under a Comparative Lens ... Growing Faster than Most Bangladesh has consistently improved its growth performance relative to the rest of the world. According to the World Bank's World Development Indicators (WDI) database, its annual per capita growth during 1973-2000 was slightly better than the performance of the median country in the world and significantly better than the median low-income country (LIC) (Table 3.1). Bangladesh has outperformed the median country in the world by a considerable margin since the early 1980s. Among the countries for which comparable data are available, 45 per cent (59 out of 130) of all countries grew at a pace faster than Bangladesh over the 1973-2000 period. This percentage drops to 32 per cent for the 1981-2000 period and further to 21 per cent for the 1991-2000 period.:! Among the LIes, Bangladesh has Few countries have sustained strong growth over long periods. There are numerous examples ofcountries growing strongly over a brief period only to see growth crash. Per capita growth in Algeria and Cameroon averaged 2.6 per cent and 4.7 per cent during 1971-85 before crashing to -1.9 per cent and - 4.7 per cent over 1986-95; average per capita growth in Argentina crashed from 1.4 per cent during 1970s to -2.8 per cent during the 1980s; Burundi saw its growth crash to -3.8 per cent, after having grown at an average of 1.7 per cent over 1976-90. 3 Only those countries are considered in this comparison for which more than 20, 10, and 5 annual observations on per capita growth are available for the 1973-2000, 1981·-2000, and 1991-2000 periods. Table 3.1 Cross-Country Per Capita Growth Comparisons Bangladesh Median Median India Nepal Pakistan Sri Lanka Bangladesh's Bangladesh's country low-income rank rank in World country (LlC) in World among LIes 1973-2000 1.8 1.7 0.2 3.0 1.9 2.5 3.5 60/130 7139 1981-2000 2.1 1.0 0.2 3.6 2.5 2.5 3.5 54/168 14/58 1991-2000 3.0 1.5 -0.1 3.6 2.6 1.4 3.9 39/182 11/62 Source: World Bank Data, BBS, and author's calculations. 96 Transforming Bangladesh into a Middle Income Economy consistently perfonned in the eightieth percentile. Within the South Asia region though it was outperformed by all other countries (not considering Afghanistan) between 1973-90. Pakistan and Nepal lagged Bangladesh during the 1990s while India and Sri Lanka did better. Per capita income in Bangladesh, as per the Summers-Heston database (which is expressed in constant 1996 US dollar and adjusted for international purchasing power differences), ranked 91 among 102 countries in 1972, improving to rank 78 within the same group by 2000. In the region, India improved its rank from 87 to 71, Pakistan from 89 to 74, Sri Lanka from 74 to 67, and Nepal from 94 to 82. China, Thailand, and Korea recorded the most impressive improvements over the period: China improved from 93 to 60, Thailand from 71 to 41, and Korea from 54 to 26. China and Nepal are the only non Sub-Saharan countries that ranked lower than Bangladesh in 1972. Also, all countries that were overtaken in ranking by Bangladesh are Sub­ Saharan countries. The Power of Growth Compounding Sustained high GDP growth over long spells can produce dramatic results. As per the WDI database. real per capita income in Bangladesh increased by about 65 per cent between 1972-2000. If Bangladesh had grown at the same rate as Niger, its income per capita would have been less than a third of the actual realisation in 2000 (Figure 3.4). Conversely, if Bangladesh had grown as fast as China, the average Bangladeshi in 2000 would have been more than four times richer. Clearly, Bangladesh has done well in avoiding the pitfalls faced by most of Sub-Saharan Africa and some other LDCs. At the same time, the magnitude of its lost potential is also significant. Either way, important lessons are to be learnt. 800 700 Real Per Capita GOP if Growth had Equaled China's <1l 600 :1: 0­ ttl u 500 Qj a. 400 a. 0 (!' Real Per Capita = 300 li a:: GOP if Growth had Real Per Capita 200 Equaled Niger's GOP in Bangladesh 100 0 ...... =£ \ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ <0 f::! ~ ~ Figure 3.4 Half Empty ... or Halt Full? 1972 Normalised to 100 Source: Author's calculations. Analysis of Growth Experience 97 If Indonesia Could Do It " " " At the time of independence in 1971, Bangladesh shared a number of similarities with Indonesia. Per capita GDP in Bangladesh stood at dollar 920 (in constant 1996 PPP prices as per the Summers-Heston database) compared with Indonesia's dollar 1,200, secondary enrolment rates were 19 and 20 per cent respectively in the two countries, and life expectancy in Bangladesh was 45 years compared to Indonesia's 49 years. Population was growing at about 2.5 per cent in each country. Yet, by 2000, Indonesia's per capita GDP had tripled to dollar 3,640 while Bangladesh's GDP had grown by 80 per cent to dollar 1,680 (Figure 3.5). This is another example of opportunities not fully availed of by Bangladesh. /~-"'t _..I L,.­ 3500 Indonesi~'~ ~3000 .-,'-­ .-< ~ 2500 _ 2000 Figure 3.5 Tale of Two Countries (real GOP per capita,1996US$) Source: Author's calculations. SEC1"ION III: GROWTH ACCOUNTING TO IDENTIFY SOURCES OF GROWTH Here we develop a growth accounting framework to estimate the contributions of capital (physical and human), labour, and total factor productivity (TFP) toward long term growth. Let us assume that the standard neoclassical constant-returns-to-scale Cobb-Douglas production function describes overall production in Bangladesh; (1) Yt = AlqiE I (I-4} Y symbolises real GDP, K stands for physical capital stock, E for effective labour, A is the Solow residual and represents TFP, a is the share of capital stock in output under perfect competition. E equals L *H, where L is the total labour force and H, a measure of human capital based on education stock and returns on education, adjusts for the quality of the labour force. 4 Following 4 Previous detailed studies, e.g. Young (1994) have shown education to be by far the most important element in accounting for differences in labour quality. A more accurate proxy for human capital would also include learning by doing, but lack of data precludes this here. 98 Transforming Bangladesh into a Middle Income Economy Ghosh and Kraay (2000), we define H = eO IS, where S is the average number ofyears of schooling per worker, and 0.1 is the returns on a unit increase in S. This assumed value of S is consistent with estimates for Bangladesh and with estimates found in the literature for other countries (e.g. Klenow and Rodriguez­ Clare, 1997). The initial stock of capital in 1973 is derived using the capital-output ratio in 1973 from Nehru and Dhareshwar (1993). The capital stock data are then extended through 2000 using the perpetual inventory method. (2) K/ = (1 ~ geometric depreciation rate) K{ 1 + Gross Capital Formation (I 1). Data on gross capital formation are from the WB's SIMA database. The data on S, the average years ofschool attainment by population aged 15 years or more, are from the updated Barro-Lee (2000) database on education attainment. The frequency of the Barro-Lee database is every 5 years and their numbers for 2000 are projections. We fill in each five year period using the assumption of a constant geometric growth rate within that period. From (1), the growth rate of A (TFP) may be written as (3) g(TFP) = g(Y) a*g(K) (1 ~ a)g(E), where g(X) indicates the annual growth rate of variable X. Using this accounting framework, and setting the value of a equal to 0.4,5 Figure 3.6 and Table 3.2 present estimates of the contribution of capital, effective labour, and total factor productivity to GDP growth. TFP growth in Bangladesh has remained low fluctuating mostly within the -1.0 to 1.0 per cent range since 1981. Capital stock growth has been robust and shows an 8.0 7.0 6.0 5.0 1: 4.0 ~ 3.0 OJ 2.0 !l. -1.0 ' ', -2.0 L:;;;;:::"1"O..,'"""'' "' ' ....,...,'' '"' ' '.,-,'' IC"""i'i' ' r7'IXlPr'' ....-::;;;'""'' ' '"' ' ' '""'' '....',.......,.'' .........,'' ::->O'l' '...,.,'' ' ' '' m~m~~mmmmmm~m~mmmmm~ - GOP -.a-. Cap Stock .~... Human Capi- - 1FP Growth Growth tal Growth Growth Figure 3.6 Factor and TFP Growth Rates Source: Author's calculations. 5 Other country studies have found the value of 'a' to be in the 0.3-0.6 range. Qualitatively, the results for Bangladesh are not sensitive to the value of a chosen within this range. Analysis of Growth Experience 99 Table 3.2 Factor and TFP Contributions of Growth GOP Capital stock Effective labour TFP growth growth growth growth A. Average growth rates 1981-5 3.72 3.36 2.70 1986-90 3.74 4.61 2.60 1991-5 4.39 5.06 3.64 1996-2000 5.21 6.74 3.44 1981-2000 4.27 4.94 3.10 B. Contribution to output growth (TFP = solow residual) 1981-5 3.72 1.34 1.62 0.75 1986-90 3.74 1.84 1.56 0.33 1991-5 4.39 2.02 2.19 0.18 1996-2000 5.21 2.70 2.06 0.45 1981-2000 4.27 1.98 1.86 0.43 Source: Author's calculations. increasing trend. Effective labour showed some improvement in the 1990s, reflecting slightly higher growth in both the labour force and the human capital stock. Within the region, Bangladesh recorded the lowest average growth during the 1980s, mainly due to the weakest performance in the region in capital stock growth and weaker performance than India and Pakistan in TFP growth (Table 3.3). During the 1990s, despite recording the highest rate of growth in its effective labour stock and the second highest rate of growth (after Nepal) in its physical capital stock, average GDP growth in Bangladesh was the second lowest in the region. This was due to weak TFP growth lower than in all other countries except Nepal. Explaining Growth Over Time: Primacy of Factor Productivity Over Factor Accumulation At first brush, it seems that capital and labour accumulation can explain almost entirely the growth process in Bangladesh. Given that physical capital growth during the 1980s and 1990s averaged 4.9 per cent per year (Table 3.3), and assuming the share of capital in production to be 40 per cent, the contribution of capital to growth would be 2.0 percentage points (46 per cent ofGDP growth). Similarly, effective labour grew at an average rate on. I per cent over 1981-2000 period, which, given its assumed share of 60 per cent in output, implies a 1.9 percentage point contribution to growth. Then, the contribution ofthe residual (TFP growth) in total GDP growth would be 0.43, or 10 per cent of total GDP growth. Table 3.3 Regional Comparisons of Factor. TFP. and GOP Growth Growth rates of GDP and its components 1984 -85 1986-90 1991-95 1996 -2000 1981-2000 GOP Growth Bangladesh 3.7 3.7 4.4 5.2 4.3 India 5.4 6.2 5.3 5.6 5.6 Nepal 5.0 4.5 5.3 4.9 4.9 Pakistan 6.8 5.8 4.6 3.3 5.1 Sri Lanka 5.0 3.5 5.4 5.0 4.7 Cap. Stock Growth Bangladesh 3.4 4.6 5.1 6.7 4.9 India 4.3 5.0 5.0 6.3 5.2 Nepal 7.7 7.5 6.8 7.2 7.3 Pakistan 6.0 6.0 5.5 4.3 5.4 Sri Lanka 10.3 6.1 4.5 5.0 6.5 Effective Labour Growth Bangladesh 2.7 2.6 3.6 3.4 3.1 India 2.7 2.8 3.1 3.4 3.0 Nepal 2.4 2.5 3.2 3.4 2.9 Pakistan 3.0 6.9 2.5 3.0 3.8 Sri Lanka 2.7 2.3 2.8 3.2 2.7 TFP Growth Bangladesh 0.8 0.3 0.2 0.4 0.4 India 2.0 2.6 1.4 1.1 1.8 Nepal 0.5 0.0 0.6 -0.1 0.3 Pakistan 2.6 -0.7 1.0 -0.2 0.6 Sri Lanka -0.8 -0.4 1,9 1.1 0.5 Source: Authors calculations. Analysis of Growth Experience 101 Based on similar calculations for the rapidly growing East Asian economies, Young (1995) too is led to conclude that factor accumulation accounted for most of their GDP growth. But economists (e.g. Barro and Sala-l-Martin, 1995 and Klenow and Rodriguez-Clare, 1997) have since criticised Young's approach on the grounds that it does not account for the fact that capital accumulation responds to technological progress - that higher the rates of return, the larger the investments in physical and human capital can be expected to be. If this effect is taken into consideration, a more significant contribution of TFP growth is likely to emerge. Is it the case that factor accumulation in Bangladesh responds to TFP growth? There is some evidence to support this for the 1975-2000 period. As seen in Table 3.4, the null hypothesis of no causation between TFP growth and physical capital growth cannot be rejected at the 5 per cent level in either direction, i.e. higher TFP growth appears to cause faster capital accumulation, and faster accumulation appears to result in productivity gains. The causal arrow in the relation between TFP growth and (first-differenced) effective labour growth flows from the former to the latter, i.e. faster productivity growth causes faster accumulation of effective labour (education stock and labour force) in Bangladesh.6 To assess more accurately the impact of TFP on growth - in a way that incorporates the fact that physical and human capital respond to technological progress - Klenow and Rodriguez-Clare (1997) recommend looking at the covariance between TFP growth and GDP growth divided by the variance of GDP growth. According to this measure, TFP growth explains 80 per cent of GDP growth between 1975-2000 and 61 per cent of GDP growth between 1981-2000. This is consistent with similar calculations for other countries (See Ghosh and Kraay, 2000, for the case of Korea) and with cross-country studies (e.g. Klenow and Rodriguez-Clare, and Easterly and Levine, 2000) that show that most of the cross country differences in GDP growth can be explained by differences in TFP growth. Labour Force in Bangladesh has Become more Productive An average Bangladeshi worker is about 30 per cent more productive today than in 198 J.1 This is because of a two and half time increase in the stock of physical capital, 7 per cent increase in human capital stock, and 8 per cent increase in TFP which combined to overcome the diminishing returns to labour 6 We use the first difference of the effecti ve labour growth because unit root tests predict strong presence of non stationarity in it. 7 Labour efficiency for period t is calculated as oY t I 04. where Y is as defined in equation (I) and L is the size of the labour force. This calculation for years 2000 and 1975 shows the former to be 30 per cent higher than the latter. 102 Transforming Bangladesh into a Middle Income Economy Table 3.4 Granger Causality Tests (using two lags), 1975-2000 Null hypothesis p-vafues TFP growth does not Granger cause physical capital growth 0.01 Physical capital growth does not Granger cause TFP growth 0.05 TFP growth does not Granger cause (differenced effective labour growth 0.05 Effective labour (differenced) growth does not Granger cause TFP growth 0.84 Source: Auhor's calculations. that would have arisen from the 70 per cent increase in the labour force over this period. Admittedly, this provides only a low-end estimate of labour productivity since learning-by-doing gains in human capital, which can be expected to have been substantial in the garments industry in Bangladesh are not accounted for here. SECTION IV: GROWTH, INVESTMENT AND SAVING IN BANGLADESH Figure 3.7A shows the investment, gross domestic saving, and GDP growth rate patterns in Bangladesh since 1981. The saving and investment series show an increasing trend and are highly correlated. The correlation coefficient between the two series is 0.92 - even after removing non-stationary in each series the correlation coefficient remains quite high at 0.68. 8 This is consistent with the cross-country findings of Feldstein and Horioka (1980) and highlights the lack of capital account integration in Bangladesh as in majority of other countries. GDP growth is highly correlated with both the investment series (correlation coefficient =0.53) and the saving series (correlation coefficient = 0.55), although this result deteriorates considerably when the stationary saving and investment series are used: the detrended saving series has a lower but still significant correlation with growth (correlation coefficient = 0.27), while the first-differenced investment series has a insignificant correlation with growth (correlatio:1 coefficient = 0.05). Figure 3.7B shows the breakdown of investment by its private and public components. The 1980s saw offsetting changes in public and private investment rates the correlation coefficient between the two series for the 1981-90 8 Unit root tests for the GDP growth, investment, and saving series show growth to be stationary and non stationarity in the investment and saving series. Unit root tests that include a time trend show the saving series to be stationary, i.e. the detrended saving is stationary. Similarly, first-differenced investment series is found to be stationary. To detrend the saving series, it is regressed on a constant and a time trend. The residual from the regression then provides the detrended saving series. Analysis of Growth Experience 103 24~---------------------------------, 20 Q. C (!J '0 16 8 - Total Investment _ Private Investment - Public Investment Figure 3.7A Saving-Investment Trends 20 5 0.. c 16 (!J E '0 4 '" <> Q; E 0.. " 'Q; <> 12 Q. 3 8 82 84 86 88 90 92 94 96 98 00 02 --=- GDP Growth (right axis) _Investment Rate (left axis) - - Savings Rate (left axis) Figure 3.7B Public-Private Breakdown of Investment Source: World Bank data, BSS, and author's calculations. period is 0.81 and as a result total investment remained virtually unchanged over the decade. Since the early I 990s, the total investment rate has increased significantly, almost entirely on account ofincreases in the private investment rate. The total investment rate increased from 17 per cent to 23 per cent between 1990-2002, with the share ofprivate investment in total investment increasing from 57 per cent to 72 per cent. 104 Transforming Bangladesh into a Middle Income Economy SECTION V: ASSESSING BANGLADESH'S GROWTH PERFORMANCE IN A CROSS-COUNTRY FRAMEWORK This section draws on the vast cross-country growth literature, that has evolved mainly since the early 1990s, to help identify the main determinants of GDP growth in Bangladesh. The intent is to carefully select from among the large number of explanatory variables in the literature (over 50 by some accounts (Levine and Renelt, 1992) a set that helps explain Bangladesh's performance vis-a.-vis, the rest of the world and a set of comparator countries in particular. As in the section on growth accounting, this section looks at whether it is factor accumulation or technology growth and more efficient resource allocation that fundamentally explains growth in Bangladesh. The answer to this has important implications for policy formulation should the focus of the policymakers be on accumulation of capital (saving and investment) or on technology infusion via R&D, FDI, and higher quality labour force? It's More than Just Factor Accumulation We begin with the simple (human-capital-augmented) neoclassical cross­ country growth framework, first tested by Mankiw, Romer and Weil (1992) (MRW). Specifically, Regression 1 (see Table 3.5) regresses average annual per capita GDP growth during 1981-2000 on initial income (real per capita income in 1980 in constant PPP US dollar), average years of schooling (proxy for human capital), gross domestic investment rate (proxy for domestic saving), and population growth. 9 The framework is purely cross-sectional, with all variables except initial income averaged over 1981-2000 for each country, giving us one observation per country.1O Regression 1 results closely resemble those of MRW. All coefficients have predicted signs and are highly significant. The coefficient on initial income is negative, indicating conditional convergence. On average, population growth is seen to hurt per capita growth while investment and education attainment are beneficial for growth. The broad interpretation is that the growth differentials across countries are due to out of steady state dynamics. Each country is converging to its unique steady state income determined by its exogenous rates of saving in human and physical capital and population growth. The pace at which the convergence takes place, i.e. the growth rate in each country depends on the gap between the existing levels of inputs (physical and human capital stocks) and their steady state levels - the shorter the inputs are of the steady state, the higher is their marginal returns and faster is growth. In this sense, the growth differentials are transitory, only until convergence to 9 The regression analysis of Mankiw, Romer and Weil covers the period 1960--85 and uses secondary education as a proxy for human capital. 10 All regressions in this section use White's heteroskedasticity-corrected standard errors. Table 3.5 Cross-Country Regressions. Dependent Variable-Average Per Capita Growth (1981-2000) Regression 1 Regression 2 Regression 3 Coeff. t-stat Coeff. t-stat Coeff. t-stat Regressors Constant -3.32 -1.01 8.22** 2.45 18.37*** 3.71 Initial Income -1.04*· -2.48 -1.64*** -4.44 -2.34*** -4.67 GDI (% GOP) 3.31*'* 3.82 3.12*** 3.84 2.3,?·** 3.35 POP -0.88*·* -3.69 -0.22 -1.11 -0.35* -1.82 EDU 1.88*" 2.64 0.59 0.96 CONC -1.55* -1.66 -13.67*· -2.37 CONC" Initial Income 1.52** 2.38 FDI 0.04 1.00 0.55·" 2.28 FDI* Initial Income -0.06** -2.38 INFLAT -0.13** -1.97 INFLAT· (LiC + MIC) -0.22** -3.32 INFLAr HIC 0.02 -0.13 TOT -0.56** -1.99 TOr (LiC + MIC) -0.6]*· -2.33 TOT" (UMIC + HIC) -0.50" -1.82 RULE 1.43*" 2.26 1.12** 1.98 COST -0.22 -1.37 -0.31** -1.99 # of Observations 78 78 78 Adjusted R Squared 0.43 0.62 0.68 Predicted value for Bangladesh 0.71% 1.87% 2.02% * reflects significance at the 10% level;"· reflects significance at 5% level;··· reflects significance at 1% level. Source: Authors calculation. 106 Transforming Bangladesh into a Middle Income Economy the steady state takes place. After convergence, all countries would grow at the same rate that is determined by the growth of a shared technological frontier. While Regression 1 explains cross-country differences in growth reason­ ably well- the adjusted R2 equals 0.43 and happens to equal that obtained by MRW for their 'intermediate sample' - it does not give a good fit for Bangladesh. The fitted per capita growth for Bangladesh is 0.71 per cent p.a., significantly lower than the 2.06 per cent actual performance during the 1981-2000 period. Clearly, to understand why growth in Bangladesh differs from that in other countries we would need to look at factors beyond capital accumulation. Specifically, we need to incorporate cross-country differences in the efficiency with which factors of production are used - something that Regression 1 omits.11 This would also be consistent with the analysis of Easterly and Levine (2000), who argue that productivity differences are far more important in explaining cross-country growth differences than the factor accumulation differences. To capture productivity differences, we include 6 additional regressors - each regressor may be viewed as a proxy for a major policy - institutional area that the literature has identified as being important for efficiency of investment. • The first variable, CONC, measures the degree of concentration in the banking system. Cetorelli and Gambera (2001) find that a higher value of CONC, i.e. a less competitive banking sector, tends to dampen productivity growth. • The second variable, FDI, captures the spillover effects of foreign direct investment. The FDI variable interacts gross FDI inflows into a country (adjusted by its GDP) with the education stock and the phone density in the economy. The latter two variables are meant to capture the absorptive capacity of the economy, which Borensztein, Gregorio, and Lee (1998), show to be important in capturing the growth effects of FDI.12 • The RULE variable is from Kaufman, Kraay, and Zoido-Lobaton (2002), and summarises the extent to which agents have confidence in and abide by the rules of the society. Kaufman and Kraay (2002) find this to be an important determinant of long-run growth. • The next two variables, INFLAT and TOT, are meant to capture macroeconomic stability. INFLAT measures the standard deviation of II Implicit in this MRW framework is the assumption that technology is an international public good and, therefore, the rate of technological progress is the same for each country (even though the initial levels of technology may differ across countries). 12 Borensztein et al. interact FDUGDP only with the education variable. We add the phone variable in the interaction under the assumption that physical infrastructure (which the phone variable is meant to be a proxy for) is also an important component ofthe absorptive capacity. Analysis of Growth Experience 107 annual CPI inflation and is included as a proxy for the degree of macro policy stability. Fischer (1993) shows this to be negatively correlated with growth. TOT measures the volatility of annual terms of trade growth in a country (normalised by the country's trade to GDP ratio) and acts as a proxy for the level of external shocks a country is exposed to. Rodrik (1999) shows this to have a negative impact on growth. • Finally, the variable COST measures the cost of starting a new business (adjusted by the GNI) and, a priori, can be expected to be negatively correlated with both the rate and quality of investment. 13 Results are presented in Regression 2. All coefficients have the predicted signs and the additional regressors, with the exception of FDI and COST, are all significant at least at the 10 per cent leveL The adjusted R2 improves considerably to 0.62 and the regression gives a much improved predicted value of 1.87 per capita growth for Bangladesh. However, the coefficients on popUlation growth and education stock lose significance and the lack of significance of the coefficients on FDI and COST is of concern. To further improve the fit, we test for non-linearities in the framework. In particular, we test how the impact of these variables changes with the income level. First, we add a regressor that interacts CONC with the level of initial income. This is included for two reasons - one, the banking sector in more developed economies can be expected to be more contestable, i.e. free entry by new banks. If this is the case, a given concentration ratio in a more developed economy would be associated with a more competitive banking sector than in a poorer country, and the interaction would have a negative coefficient. Second, the literature (Diamond, 1984 and Allen 1990) predicts important economies of scale in the banking business, which higher bank concentration would help exploit. 14 As long as the banking sector consolidation is a result of market forces capitalising on the economies of scale, an increase in bank concentration can be viewed positively. There is evidence, however, that this process is more evident in developed economies - in emerging economies bank concentration is typically due to other forces, such as direct government intervention and public and private responses to crises. IS To the extent that the positive forces of bank consolidation are correlated with the development level of an economy, a negative sign on the interaction variable is to be expected. 13The COST variable is taken from the WB's new Doing Business 2003 database. 14 Proponents of the recent spate of mergers in financial services in developed countries also argue for economies of scales in financial intermediation. 15 'Fragile, handle with care', A survey of banking in emerging markets by The Economist, 12 April 1997. 108 Transforming Bangladesh into a Middle Income Economy Similarly, since the FDI spillovers effect measures flow of technology from more to less developed economies, its impact can be expected to diminish with income level. This is what the interaction variable between FDI and initial income tests for. We also interact the INFLAT and TOT variables with income group dummies to test whether the higher income group countries, with presumably deeper institutions for economic management, are better able to cope with economic uncertainty. Finally, we drop the EDU variable as a separate regressor, since, as results of Regression 2 show, it is no longer significant in the presence of the additional productivity regressors. 16 Results of the revised regression are presented under Regression 3 in Table 3.5. The adjusted R2 improves further to 0.68 and the predicted value for Bangladesh from the regression is 2.02 per cent, which is remarkably close to the actual performance of2.06 per cent. All variables are now highly significant - most variables are significant at the 5 per cent level. The negative impact ofbank concentration is seen to ameliorate with income level, perhaps confirming our prior that the offsetting scale effect and contestability in the banking sector become more pronounced with higher income. Inflation volatility is seen to harm the groups of low-income and middle-income countries but not high-income countries. TOT volatility negatively impacts growth in all income groups, although the effect is statistically larger for the groups oflow-income and lower-middle-income countries than for the groups ofupper-middle-income and high-income countries. The FDI spillover effect has a positive impact on growth, although it diminishes with income level. The rule of law variable is seen to have a positive and significant impact on growth, confirming earlier results in the literature. The new cost of starting a business variable appears to have a negative and significant impact on growth. Bangladesh Relative to Comparators The results of Regression 3 capture well Bangladesh's relative growth performance during the 1980s and 1990s. In this section we benchmark Bangladesh's performance against a group ofcomparator countries. The group consists mainly of countries that have outperformed Bangladesh in the hope identifying reform areas for even higher growth. Niger is included as an example ofa poor performer to understand why Bangladesh was able to avoid the fate of most of the Sub-Saharan Africa. As seen in the Table 3.6 below, Regression 3 does a fairly good job ofpredicting growth in all ofthe comparator countries. Figure 3.8 compares Bangladesh to the benchmark countries on the 16 Easterly(200 I) argues that a higher education level does not in itself improve growth: the returns on education are high only when the business environment enables good use of the education stock. Regression 2 results appear to support this argument as the addition of business environment variables result in the EDU variable losing significance. - ~ ~ I\J Per cent Per cent o I\J .I> 0> g.fl 0 .I> N 0> 0 o I\J '" -. '" w &l~wO'l""'(() 8 8 8 8 !~I! 8~~_ 8888 8 8 8 8 o 8 8 8 8 8 ~ 8888888 ." ce' c $ w -il nl i, ~!I 'g: ~I g~1 !S:r Gi i~ol Thailand Chile... Malaysia Chin "II! !2:' I I I I Bangladesh Iia en or ~ China Sri Lanka Thailand E jIiiII , )10 i <, 91 c§ i. )10: China Thailand Vietnam' I I ~ . "1 ~ '~ -:;;c iil ~~Ii Q, Chile ~ 5: MalaYSia~' ~ "CI §~~I Sri Lan I (Q C i , ~"CI Cl Zlll ';111 ~.. -&1 6'1 Indonesia=. ~ !!t.,. Indonesia! ~~ S. I "CI, on m III , ~ Indonesi iii II ~.g 111 , 0 ~ ~ India. 1 Chile -" 3 ilI ChinaE;J~ i Ban~::::==.... I' Lanka~ chilejll_ _ S' ~ i'. So Vietnam < s· III, India G) Q) ::J ::!.I ~l a en a BangladeshCIIjjjffffijlI Nigel ;;! Niger i MalaYSia - I' Sri Bangladesh J ,::I i G') Niger J Bangladesh Indonesia . NigerL_. .-~~ I __N_ig_er-=~========~~ ~ =r 00000 o 0 0 0 ~ I\J '" .I> • (1996US$) g~~~~g 8 ~ ~ ~ ~ OI\J.I>O> 00000 s;!~g88 (") a . 88888 88888 ..._-- '1 » nl <:5 8 0 0 0 'I m m I 0 1 5[ I Chile Thailand.1_ __ I China i~ ~hina . 0 8 8 8 88 8 ;. )II ::J Q) ~ (") o III III CD fI India ~5 Thailand I G) Bangladesh 1W d ~ iii' c: ::I ::I' ""-I Malaysia.1• • • ~ 8!, >I a ~ a ::J ~ -Ql ::1­ en gl III -'I India.I• •_ !iI. Chile...... Bangladesh s~ ~I Malaysia Indonesia ;gIl: iil C c "CI G') ~ 3~ ~ (") o ~)II I -I»: China. . . . 8 1 l BangladeSh~~ ~I 3 ::Ii Thailand 0 Sri Lanka IIl!CDi. "'Ill n =r "C Q) :::!. ::II ~ I, 5' Indonesia il I g. ~ i III ~ "C m o :::J China 1'1' Vietnam G) ~I Nige i' 81 iii ~!. :i' ..... .., CD Cir m Sri L a n k a E Vietnam I Bangladesh i Indonesia ~ II Malaysia il. Bangladesh ~ .~ 21 ! ~ ::J @ Niger L .............._...... J _. Niger ::I aI ! Sri L a n k a : r i Vietnam 'J .. Niger -Ii» ....J S' . Chile _____.. ..... = CD 110 Transforming Bangladesh into a Middle Income Economy Table 3.6 Average Per Capita GOP Growth, 1981-2000 Actual Predicted Bangladesh 2.06 2.02 Chile 3.63 3.17 China 8.37 6.24 India 3.61 4.42 Indonesia 3.69 2.89 Malaysia 3.84 3.60 Niger -2.26 -1.94 Sri Lanka 3.49 2.65 Thailand 4.87 4.35 Source: Author's calculations. growth determinants identified in Regression 3. Key features of the comparisons are as follows: • Bangladesh had the lowest initial income among the group, which, ceteris paribus, should have produced higher growth. Clearly, given Bangladesh's relative standing in the group on growth performance, other factors outweighed any advantage from lower initial income level. • A partial explanation of Bangladesh's relative position is its pace of capital accumulation - slower than in the better growth performers and faster than in Niger. • The relative rates of population growth within the group help fit another piece in the puzzle. Average population growth in Bangladesh was higher than comparators except Malaysia and Niger. • Bangladesh appears to have been a mediocre performer on banking sector, although, as suggested by the regression results, the growth impact of this may have been more severe in Bangladesh due to its low development level. • Bangladesh has been an impressive performer on macro stability - as measured by the inflation and TOT volatility measures. This, in fact, appears to be a major reason why its growth rate has surpassed most other LICs' and compares fairly well in most global comparisons. • Relative inability to attract FDI and benefit from its spillover effects has been a major area of deficiency for Bangladesh. Even Niger outperforms Bangladesh on this measure. • The quality of economic governance has been quite weak in Bangladesh it lags all comparators barring Indonesia and Niger on this measure. This comparison is consistent with the priors that weak performance in this area has hampered Bangladesh's growth performance. • Cost of starting a business in Bangladesh is relatively high exceeded only by Niger and its ranking on this measure is consistent with its ranking on growth. Analysis of Growth Experience 111 SECTION VI: A CLOSER LOOK AT THE MAIN STRUCTURAL AREAS This section takes a closer look at the structural areas identified in the previous section as important determinants of Bangladesh's growth performance. Broadly, the areas of relative weakness in Bangladesh are - policies and institutions to enhance and benefit from global integration, quality of financial intermediation, quality of economic governance, and regulatory and administrative burdens that impact cost ofbusiness. Weaknesses in these areas offset to some extent the advantages offered by the lower initial income, relatively good performance on macroeconomic stabilisation, and the decline in population growth. Macroeconomic stability has been the cornerstone of Bangladesh's economic performance. CPI inflation has not touched the two digit mark since the mid-1980s, and has been more stable than in any other low- or lower­ middle-income country, as measured by its standard deviation. 17 Exposure to external shocks has also been limited, mainly due to a low trade base though. Terms of trade volatility has been high - standard deviation of the terms of trade growth over the 1980s and 1990s was the highest in the comparator group - but when adjusted by the trade to GDP ratio, Bangladesh's external vulnerability appears to be quite low. Partly in response to low exposure to macro policy and external shocks, GDP growth has been remarkably stable ­ the volatility of GDP growth in Bangladesh over the previous two decades was among the lowest in the world. To be sure, the economy remains vulnerable to weather related shocks, floods and hurricanes in particular, that frequent Bangladesh due to its unique topography. However, the capacity to manage and cope with natural calamities has improved tremendously. For example, the spread of defensive infrastructure, such as coastal and river embankments, has helped contain natural disasters, while improved ability to evacuate people from disaster zones and provide shelter, improved access to micro-finance, implementation of government-sponsored employment programmes during the lean season, and liberalisation of private rice imports have collectively helped mitigate the impact of disasters. Economic Governance The problem of weak economic governance in Bangladesh is severe, and is a major impediment to faster economic growth. IS 17 Among the benchmark countries in this paper, only Malaysia had a more stable CPI inflation process during the 1981-2000 period. 18 Economic governance here refers to the quality of those aspects of public service delivery and public-private dealings that directly affect the efficiency of the private business sector. Table 3.7 Cost Doing Business - Bangladesh v. Comparators Starting a Business Enforcing Contracts Number of Duration Cost Cost (% of Number of Duration Cost Cost (% procedures (days) (US$) GNI per capita) procedures (days) (US$) GNI per capita) Bangladesh 7 30 272 75.5 15 270 973 270.3 Chile 10 28 494 11.6 21 200 626 14.7 China 11 46 134 14.3 20 180 301 32 India 10 88 239 49.8 11 365 456 95 Indonesia 11 168 103 14.5 0 225 1910 269 Malaysia 8 31 959 27.1 22 270 687 19.4 Niger 11 27 759 446.6 29 365 97 57.1 Sri Lanka 8 58 154 18.3 17 440 64 7.6 Thailand 9 42 144 7.3 19 210 586 29.6 Vietnam 11 63 129 29.9 28 120 37 8.5 Median Country in Global Sample 10 45 541 24.3 21 225 352 12 Source: DOing Business Database. Analysis of Growth Experience 113 The problem of endemic corruption Bangladesh has received a lot of negative attention lately due to perceptions of widespread corruption in public-private dealings. Most prominently, for four years in a row now Transparency International, the Berlin based private rating agency, has placed Bangladesh last in its cross-country corruption rankings. Such a blanket indictment ofBangladesh may not seem fully justified since it reflects subjective perceptions that are not meaningful in cross-country comparisons. Nonetheless, the problem ofendemic corruption in Bangladesh and the harm it inflicts on the economy is undeniable. In a recent survey, about 60 per cent ofthe 1,001 firms surveyed in Dhaka and Chittagong viewed corruption as a major constraint to business operations and growth second only to problems related to power supply. Breakdown of law and order Extortions and other criminal activities, feeding on an environment of widespread corruption in police and lower courts, are a major governance problem. This has seriously hurt public confidence in enforcement ofthe rule of law and impartial protection of property rights, vitiating the business environment and harming long-run growth. Regulatory environment and the cost of doing business According to the World Bank's Doing Business 2003 database, on average, 7 procedures are involved in starting a business in Bangladesh (Figure 3.9). Bangladesh .~.... ·so .I.,--_~r---'" 70 (") ~ '00 i ~ 20 Cost h---F=:I---=' 40 ~; 50 ~ (right axis) (') t 15 Time : 30 i ~ i= 10 (left axis) .20 '" ~ 5 ..----r4=::::F=:::::I--...J . 10 (D O~~~2~~3~~4--~5~~6~~7~O Procedure Figure 3.9 Procedures Involved with Starting a New Business in Bangladesh Notes: 1. Buy non-judicial stamps 2. Verify the company name 3. A lawyer verifies the Memorandum and Articles of Association 4. Pay the registration fee 5. File documents with the Registrar of Joint Stock Companies 6. Make a company seal 7. Register with the tax authority Source: World Bank's Doing Business Database. Table 3.8 Country Rankings on A.T. Kearney/Foreign Policy Globalisation Index Measures Economic integration Personal contact Technology Political engagement t: .2 0) .... 0) ..... )Jl t: O!/!~ e! !'2 e! -(I.) co t: .S: .... ~ Q. t: .2 S ~ §:2 (I.) ~ (I.) Q) e! Q) ~~ 'lb g ~ Ii ~ ~ co t: E ..2 'lb Q ~ (I.) ~ ~ t: ~ ~ ~ t:- ~ 5l ~<1l ~ EJ!! ~ ~ 0 Q Q E (I.) i Q) Q) Q) ~ ,~ co ::::: ~ t: e! t: ~ Jg'lb <.> t: 0))0<; Q "5 Q E E ~ .... t: co ~ Q) ~ j.;;: ::. ~ Q) 1:) Llj cf ~ Cl f::: E t: '- Q .i!! .i!! 2 .i!!~ Q) j.;;: Q ..... ~~ t: >;;. Q. (!) .S: Q. .£ .£ ~ -0 <: :::J Bangladesh 56 62 43 62 35 58 60 61 62 61 60 11 62 62 59 28 29 30 41 Chile 37 29 50 27 39 36 17 26 23 35 44 54 26 32 30 37 46 9 33 China 57 37 59 49 56 44 47 22 55 56 55 49 46 51 55 54 35 44 58 India 61 61 53 55 57 59 56 55 60 58 61 30 55 57 53 28 58 53 51 Indonesia 59 47 61 51 53 33 51 50 32 57 57 53 51 49 51 37 59 30 40 Malaysia 20 8 14 26 46 2 42 17 15 28 7 12 20 36 35 32 43 30 45 Sri Lanka 51 41 34 56 60 22 54 43 53 49 52 7 58 50 47 51 55 44 49 Thailand 48 28 48 40 58 8 44 53 37 53 42 32 37 41 42 45 45 53 52 Analysis of Growth Experience 115 This compares favourably relative to other benchmark countries. On average, it takes about 30 days to clear these procedures in Bangladesh, which again is relatively good - only Niger and Chile among the comparator countries have a shorter duration. However, the cost of even these relatively few procedures is extremely high in Bangladesh. In US dollar terms, the cost of starting a business is higher in Bangladesh than in China, India, Indonesia, Sri Lanka, Thailand, and Vietnam. As a share of per capita GNI, the cost of startbg a business in Bangladesh is lower than only Niger within the comparator group. As seen in Figure 3.9, the most costly procedures, in terms oftime and money spent, are filing documents with the registrar of Joint Stock Companies and registering the company with the tax authorities. Similarly, while the number ofprocedures involved in getting a business contract enforced is relatively low in Bangladesh, the cost ofclearing these is quite high - the average cost of enforcing a contract in Bangladeh is almost three times its per capita GNI, the highest ratio within the comparator group (Table 3.7). Global Integration Bangladesh's economic integration with the global economy is low, by most measures. The 2004 globalisation index prepared by the management consulting firm A.T. Kearney and the Foreign Policy magazine ranks Bangladesh at 56 out of62 countries (Table 3.8). Bangladesh ranks last among the 62 countries in the area of economic integration, and within that category, Bangladesh ranks last on investment income, second to last on FDI, third to last on portfolio flows, and fifth from last on trade openness. A commonly used measure of trade openness is the ratio of total trade (exports + imports) to GDP. By this measure, Bangladesh ranks 175 among the 182 countries for which the World Bank's WDI database provides comparable data for the 1990s. The trade to GDP ratio has increased considerably in Bangladesh in recent years from 20 per cent in 1990 to 34 per cent in 2002, reflecting significant reductions in trade tariffs and quantitative restrictions and considerable progress on exchange liberalisation (Ahmed and Sattar, 2003). However, since trade liberalisation was a common phenomenon across the globe during the 1990s, Bangladesh's global rank improved only marginally over the decade. In terms of openness to FDI, as measured by the gross FDI inflows to GDP ratio, Bangladesh ranks 137 among the 141 countries for which comparable data for the 1990s are available in the WDI database. Only Nepal, Iran, Republic of Congo and Samoa rank low. This is despite the fact that Bangladesh has a relatively liberal FDI regime - the most liberal in the South Asia region (World Bank 1999). Clearly, foreign investors are deterred by the licensing requirements for private activity in the energy and the telecom 116 Transforming Bangladesh into a Middle Income Economy sectors, as well as other investment climate problems identified by the recent Investment Climate Assessment (lCA). The perception of high corruption likely also dampens FDI inflows, as shown by Smarzynska and Wei (2000) for a cross section of countries. Not only are the levels ofFDI into Bangladesh quite low, its absorptive capacity to benefit from these is also weak due to low levels of infrastructure development and human capital. Infrastructure Supply Infrastructure bottlenecks in the areas of power, ports, gas, and telecommunications are severe. Only 31 per cent of the population 80 per cent in urban areas and 19 per cent in rural areas has access to power. Inadequate access to electricity was the most frequent complaint among firms surveyed for the Bangladesh ICA. Lack of access to power is a lot more constraining for rural enterprises. A recent survey found only 32 per cent of rural firms with access to electricity compared to 60 per cent in urban areas. 19 For those fortunate enough to have a utility connection, reliability is a major issue - power outages and surges are all too frequent. This again affects the rural enterprises disproportionately since only 1.8 per cent of these own a generator. 20 Poor port conditions extract a hefty price in terms of lost economic potential. The Chittagong port, which handles nearly 85 per cent of the country's merchandise trade, is plagued by labour problems, poor management, and lack of equipment. Its container terminal handles only 100-105 lifts per berth a day, well below the UNCTAD productivity standard of 230 lifts a day. Ship turnaround time is 5-9 days, higher than the one day standard of more efficient ports. In addition, handling charges for a 20 foot container are estimated at dollar 640 (of which dollar 250 is for unofficial tolls) compared to dollar 220 in Colombo and dollar 360 in Bangkok. 21 Infrastructure related problems at the ports are exacerbated by cumbersome customs procedures that are also a source for bureaucratic discretion. Bangladeshi firms reported an average 11.7 days to get their last shipment through customs, compared with Malaysia (3.4 days), China and India (both 7.5 days), Philippines (10.2 days), and Pakistan (17.9 days). Customs procedures at airports, too, are fraught with multiple requirements for documents and approvals. Despite a relatively high density ofits road network, poor road conditions and lack of transportation services seriously impair private activity. Factors contributing to this are: poor construction of roads and bridges; lack of J 9 The World Bank, Promoting the Rural Non-Faun Sector in Bangladesh, (2004). 20 Ibid. 21 Bangladesh Enterprise Institute, Reducing the Cost ofDoing Business in Bangladesh (2004), pp.99-105. Table 3.9 Infrastructure Indicators: Cross-Country Comparisons Electricity generating Telephone mainlines Internet users (per Total road network Roads, paved (% of capacity (kW per (per 1,000 people), 1,000 pop) (per sq. km. area), total roads), 1999 capita), 2002 2001 1999 Bangladesh 0.03 8.3 1.9 1.6 9.53 Chile 0.63 574.8 201.4 0.1 18.9 China 0.25 247.7 26.5 0.1 22.4 India 0.11 43.8 6.8 1.1 45.7 Indonesia 0.10 65.7 19.1 0.2 46.3 Malaysia 0.58 509.9 273.1 0.2 75.8 Niger 2.1 1.1 0.0 7.9 Sri Lanka 0.09 79.9 8.0 1.5 95.0 Thailand 0.34 221.9 57.8 0.1 97.5 Vietnam 0.06 53.0 12.7 0.3 25.1 Source: Wond Bank's WDI database and US energy information agency. 118 Transforming Bangladesh into a Middle Income Economy maintenance; and a lack of integration of different modes of transportation (which makes long-haul transport very difficult). Poor road conditions and lack of transportation are especially constraining for enterprises in far-flung rural areas where, in a recent survey, 36 and 18 per cent of firms, respectively, reported them as severe problems. 22 Public spending on road maintenance falls well short of what is needed. For example, expenditure by the local government engineering department (LGED) on rural road maintenance was taka 1.7 billion in FYOO against an assessed requirement oftaka 2.4 billion. 23 Another source of the weak investment climate in Bangladesh is communications services. Tele-density (fixed line and cellular) in Bangladesh is far lower than in comparator countries (Table 3.9). The vast majority of rural enterprises do not have access to fixed line phone service only 1.5 per cent of rural firms reported access to phone in the recent NPSEB survey. Bangladesh has one public call office (PCO) per 32,000 population (India has one per 1,000), which gives only 31 per cent of the rural population access to a fixed line (compared with 70 per cent in India). Even in regions within reach of the telephone network, access is made difficult by the extremely high cost involved in getting the initial connection. Average cost of obtaining a fixed line is dollar 171 in Bangladesh, compared to dollar 18 in India and dollar 28 in Nepal, while the average waiting period is 130 days in Bangladesh, compared to 42 days in Pakistan and 16 days in China. Human Capital Development and the Innovation System As seen in Table 3.10, despite significant improvements since independence, Bangladesh still lags comparator countries on most human development indicators. 24 This significantly dampens labour productivity and, consequently, growth and wage levels. Bangladesh also lags comparator countries in terms of innovation potential and information technology (IT) infrastructure. The cross-country comparisons in Table 3.11 on four telling measures of a country's innovation system (the first four indicators) and two indicators of information infrastructure exemplify this. Financial Sector Performance Bangladesh has a relatively shallow financial sector. In 2002, bank credit to private sector was just 29 per cent of GDP and broad money was 37 per cent ofGDP (Table 3.12). Historically, deep-rooted institutional weaknesses have drastically restricted the efficacy of the banking sector. Due to weak management, political interference, directed lending and corruption, the four 22 The World Bank, Promoting the Rural Non-Farm Sector in Bangladesh (2004). 23 The World Bank, Bangladesh Public Expenditure Review. (2003) 24 The Bangladesh Development Policy Review (2003) contains an in-depth assessment of human development issues. Table 3.10 Human Development Indicators: Cross-Country Comparisons Prevalence of child Matemal mortality malnutrition (% of Average yrs of edu rate (per 1,00,000 children under 5), (2000) Life expectancy Fertility rate live births) weight for age 1975 2000 1975 2000 1975 2000 1990-1998 1993-2001 Bangladesh 1.3 2.6 46.0 61.2 6.6 3.1 440 48 Chile 5.6 7.6 65.7 75.6 3.2 2.2 20 1 China 4.4 6.4 64.7 70.3 3.4 1.9 55 10 India 2.7 5.1 51.8 62.8 5.4 3.1 410 53 Indonesia 3.0 5.0 51.3 66.0 5.0 2.5 450 25 Malaysia 4.4 6.8 64.4 72.5 4.6 3.0 39 20 Niger 0.5 1.0 39.9 45.7 8.0 7.2 590 40 Sri Lanka 4.8 6.9 66.2 73.0 3.9 2.1 60 33 Thailand 4.0 6.5 60.6 68.8 4.6 1.8 44 18 Veitnam 55.9 69.1 5.7 2.2 160 34 Sources: World Bank's WDI database and Barro-Lee (2000) Note: Data are for the most recent year available. Table 3.11 Innovation Potential and Information Technology Infrastructure R&D Number of Patents granted by Scientific & Personal computer Number of expenditure researchers the US patent & technical journal penetration (per internet (%ofGDP) (per million pop.) trademark office articles (per 100 pop.) users (per (per million pop.) million pop.) 10,000 pop.) Bangladesh 0.03 51 0.00 1.2 0.3 15 China 0.7 551 0.21 9.3 2.8 460 India 0.7 158 0.17 9.2 0.7 159 Malaysia 0.2 154 2.35 18.3 14.7 3197 Pakistan 78 0.01 2.1 0.4 103 Philippines 0.2 156 0.19 2.2 2.8 440 Source: WBI, 2004 Knowledge Assessment Basic Scorecard, World Bank. Table 3.12 Financial Sector Indicators: Cross-Country Comparisons (2002) Bank Stock market Lending rate Real Private concentration capitalisation minus deposit interest rate M2IGOP credit/GOP ratio to GOP rate Bangladesh ·37.02 29.07 0.43 0.02 7.34 12.95 Chile 44.73 66.41 0.23 0.75 5.70 5.03 China 167.58 139.67 0.52 0.50 3.60 6.55 India 57.99 31.74 0.48 0.21 7.61 Indonesia 53.68 22.28 0.53 0.08 3.07 10.99 Malaysia 100.27 145.23 0.30 1.32 3.29 2.59 Niger 9.02 5.01 1.00 Sri Lanka 36.76 28.25 0.52 0.05 8.38 5.51 Thailand 98.54 102.92 0.31 0.26 4.71 6.09 Vietnam 49.90 43.10 0.71 4.12 5.56 Source: World Bank Data. 122 Transforming Bangladesh into a Middle Income Economy nationalised commercial banks (NeBs), holding almost halfofthe industry's net assets, are technically insolvent, with the ratio of their non-performing loans to total loans estimated at 34 per cent. The court system cannot be relied upon to protect creditor rights which further vitiates lender incentives and contributes to misallocation of credit. It is hardly surprising then that instead of contributing to economic growth the banking sector has dragged growth down and poses serious risks to financial stability. Recognising this, a 1998 World Bank report commented that 'The three institutional pillars of banking a strong regulatory system, well-managed banks, and an effective court system - have crumbled to such a point that banking institutions cannot be relied on to ensure the safety of deposits and efficiently allocate credit, their two most important functions. In a more difficult economic environment, the banking system could become Bangladesh's Achilles heel.' Responding to the difficulties the government over the past few years has embarked on a significant reform of the banking sector. Some positive developments have happened since then. In particular, the condition of private banks has improved markedly in response to improved banking supervision and governance. The performance and financial positions of the NeBs, however, has not improved much. Greater eompetition from private banks along with a reform of administered interest rates on government saving instruments such as National Saving Certificates and postal saving schemes have helped lower the real lending rates, which have been rather high until recently. Other elements of the financial sector are fairly underdeveloped. The size of the capital market is small - market capitalisation equals only about 2.4 per cent of GDP. There are no derivative or over the counter markets in Bangladesh. Insurance company assets equal only about 1.3 per cent ofGDP and those of finance companies another 0.7 per cent ofGDP. - Governance 6.00 - Cost of Starting 4.00 Business - Inflation Volatility C 2.00 ~ - TOT Volatility ~ 0.00 D FDI Spillovers -2.00 -4.00 - GDI t;; !!! E E'" '" c: '0 1: .!2 (J) . 'Qi OJ "''c" " -0 c: .. UJ,f{ E ~ ~ I- Figure 3.10 Growth that could have Been Note: Each coloured box bar shows incrementally the change in per capita growth if Bangladesh's performance during 1981-2000 equaled that of a benchmark country. Source: Author's calculations. Analysis of Growth Experience 123 Growth that could have been Based on Regression 3 results, Figure 3.10 above shows how much growth would have been higher (or lower) if Bangladesh had matched the performance of its benchmark countries along the main policy dimensions. Higher FDI combined with better absorptive capacity appears to have been the area with the highest potential payoff. This is followed by higher investment levels, a better financial sector, better governance, and lower cost of business. SECTION VII: ASSESSING THE FUTURE GROWTH POTENTIAL Growth Potential Based on Current Policy Performance Figure 3.11 below shows current growth potential of Bangladesh based on the most recent performance measurements of the right hand side variables in Regression 3. According to these calculations, current policy and institutional • Reduction in Pop Gr 4,00 3,50 D FDI Spillovers 3.00 2.50 D Increse in Domestic C 8 2,00 Investment Rate ~ '" a. 1.50 • Financial Sector 1.00 Improvements 0.50 • Average Growth 0,00 (1981-2000) Figure 3.11 Future Growth Dividends from Key Areas Note: Each coloured box bar shows incrementally the amount of additional growth that can be obtained due to improved performance in 2002 (or latest data) vis-a-vis the 1981-2000 average. Source: Author's calculations. performance in Bangladesh appears to be consistent with annual per capita growth of 3.7 per cent. This is remarkably close to the actual performance in Bangladesh seen over the last couple of years. It appears, therefore, that the current policy stance should be able to sustain GDP growth in the 5.0-5.5 per cent range. 124 Transforming Bangladesh into a Middle Income Economy Growth that can be... Reform Areas that Hold Most Promise for Growth Figure 3.12 below uses Regression 3 results to assess the growth dividends if key growth-oriented policies and institutions are improved (or worsened in some cases) to equal those of the comparator countries. For instance, if Bangladesh's investment climate performance across the five structural areas had equaled that of Chile, its fllture per capita growth would rise from 3.5 to II per cent, or if it had equaled that of China, it would rise to 8 per cent. The highest pay-off areas in terms of improving growth appear to be attracting more FDI and increasing the economy's capacity to benefit from its spillovers; improving the quality of and access to physical infrastructure and education; financial sector reforms; improving economic governance especially by strengthening the rule of law; and streamlining the regulatory and administrative environment in order to reduce the cost of starting (and doing) business in Bangladesh. An assessment of the required rates of saving, investment, and TFP growth Scenario 1. Assume that physical investment grows at an annual rate of 9 per cent (investment grew at about 8.5 per cent during the 1990s), that the GOP growth path reaches a steady state 7 per cent, and labour force grows at 3.1 per cent (the same as in the 1990s). Average years of schooling are assumed to grow at the same rate as in the 1990s. Then, the implied growth path of 8.00 • Governance 6.00 a Cost of starting 4.00 Business C '" (.) 2.00 Q) a FOI Spillovers 0­ 0.00 -2.00 • GOI -4.00 a Financial Seelor -8.00 t;; .!!! E <1l c <1l '6 <1l '0; <1l '0; 01 ~ c -0 C E <1l '" '" e.g> () E () .f: 'c:: .!l! 0 " .,>- Z <1! ..J ! S '" E ­ <1! '" (/)~ -0 .f: ::E (;5 F :> Figure 3.12 Growth that can be Note: Each coloured box bar shows incrementally the change in future per capita growth if Bangladesh's performance in 2000 equaled that of a benchmark country. Source: Author's Calculations. Table 3.13 Alternative Growth, Investment and Productivity Scenarios Scenario 1 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 GOP growth 5.5 5.5 6.0 6.5 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 Growth rate of investment 7.8 8.0 8.5 9.0 9.0 9.0 9.0 9.0 9.0 9.0 9.0 9.0 Investment rate 25.6 26.2 26.8 27.5 28.0 28.5 29.0 29.6 30.1 30.7 31.3 31.9 Capital stock growth 7.3 7.4 7.4 7.6 7.7 7.8 8.0 8.1 8.2 8.3 8.4 8.4 Effective labour growth 11 3.5 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 TFP growth 0.4 0.4 0.9 1.3 1.8 1.7 1.7 1.6 1.6 1.5 1.5 1.4 Scenario 2 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 GOP growth 5.5 5.5 6.0 6.5 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 Growth rate of investment 7.8 8.0 9.5 11.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 Investment rate 25.6 26.2 27.1 28.2 29.5 30.9 32.4 33.9 35.5 37.1 38.9 40.7 Capital stock growth 7.3 7.4 7.4 7.7 8.0 8.5 8.9 9.2 9.6 9.9 10.1 10.4 Effective labour growth 11 3.5 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 TFP growth 0.4 0.4 0.9 1.3 1.6 1.5 1.3 1.1 1.0 0.9 0.8 0.7 Scenario 3 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 GOP growth 5.5 5.5 5.8 5.8 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 Growth rate of investment 7.8 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 Investment rate 25.6 26.2 26.8 27.3 27.8 28.4 28.9 29.4 30.0 30.6 31.1 31.7 Capital stock growth 7.3 7.4 7.4 7.5 7.6 7.6 7.6 7.7 7.7 7.7 7.8 7.8 Effective labour growth11 3.5 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 TFP growth 0.4 0.4 0.7 0.7 0.8 0.8 0.8 0.8 0.7 0.7 0.7 0.7 Note: 1/Effective labour growth is a composite of labour force growth and average education stock growth. Source: Author's calculations. 126 Transforming Bangladesh into a Middle Income Economy TFP is as shown under scenario 1 in Table 3.13. Note that under this scenario annual TFP growth would reach 1.8 per cent in 2008 and would need to average 1.4 per cent during 2005-15. In comparison, in the 1990s TFP growth averaged 0.3 per cent. Note also that under the given assumptions the investment rate will increase from 26 per cent in 2004 to 29 per cent by 2010 and 32 per cent in 2015. A tall order overall to get GDP growth of up to 7 per cent. Scenario 2 below presents an alternate set of assumptions, where GDP growth path remains at 7 per cent but investment growth reaches 12 per cent p.a. This scenario has two main implications - one, the investment rate would jump to 32 per cent by 2010 and to 41 per cent by 2010, implying that additional savings/external financing of 6 per cent of GDP would be needed by 2010 and of 15 per cent by 2015. Two, the implied average TFP growth rate for 2005-15 is still about 1 per cent p.a., about 3 times higher than the 1990s. Scenario 3 seems the most plausible. A GDP growth path of 6 per cent from 2008 on and investment growth of 8 per cent would imply an average annual TFP growth of 0.7 per cent during 2005-15 (more than double of the 1990s). The investment rate with this GDP growth path would reach 29 per cent by 2010 and 32 per cent by 2015. Bottom Line Ambitious growth in both investment and its productivity is needed to achieve a growth target of 6 per cent p.a. Even then, performances on both fronts would need to be far superior than what has been seen so far. This of course also assumes that if the garment sector is adversely affected by the MFA phase-out in January 2005 then capital and labour from that sector would be shifted efficiently and costlessly to other equally or more productive sectors. A net write-off of capital stock and loss of employment would push growth lower. Overall, given the substantial demands on investment and its efficiency, major improvements are needed in the investment climate. For productivity gains of the magnitude needed, structural measures are necessary in the areas identified above - improving financial intermediation (to improve the allocative efficiency of investment), increasing FDI inflows, accelerating human capital and infrastructure development, increasing private participation (private investment has been far more productive than public sector investment), improving economic governance, and reducing regulatory and administrative burdens that increase cost of business. At the same time, continued macroeconomic stability, lower barriers to entry for new firms, and improved access to finance are critical in stimulating investment growth. Even then, financing of the required higher investment levels will remain a challenge Analysis of Growth Experience 127 and will require fiscal prudence (cutting budget deficits, lowering SOE losses, etc.) and substantial donor support. ACKNOWLEDGEMENTS Author acknowledges useful feedback from Kapil Kapoor, MaIjin Verhoeven, Sadiq Ahmed, and Tercan Baysan. The views are those of the author alone, and should not be attributed to the World Bank, its Board of Governors, or any of its members. REFERENCES Ahmed, S., and Z. Sattar. 'Trade Liberalisation, Growth and Poverty Reduction: the Case of Bangladesh', South Asia Region Internal Discussion Paper (IDP-190), Washington, DC: The World Bank, 2003. Allen, F. 'The Market for Information and the Origin of Financial Intermediation.' Journal ofFinancial Intermediation' volume 1, 1990. Bangladesh Enterprise Institute. Reducing the Cost of Doing Business in Bangladesh, Dhaka: Draft Mimeo, 2004. Barro, R., and l.W. Lee. 'International Data on Educational Attainment Updates and Implications.' NBER Working Paper Series no. 7911, 2000. Barro, R., and X. Sala-i-Martin. Economic Growth, McGraw-Hill, New York: 1995. Borensztein, E., l. De Gregorio, and J.W. Lee. 'How Does Foreign Direct Investment Affect Economic Growth?' Journal of International Economics, no. 45, 1998. Cetorelli, N., and M. Gambera. 'Banking Market Structure, Financial Dependence and Growth: International Evidence from Industry Data' The Journal ofFinance, no. LVI (2), 2001. Chen, S., and M. Ravallion. 'What Can New Survey Data Tell Us About Recent Changes in Distribution and Poverty.' World Bank Economic Review, no. 11/2, 1997. Diamond, D. 'Financial intermediation and delegated monitoring.' Review of Economic Studies no. 51 (3), 1984. Dollar, D. and A. Kraay. 'Growth is Good for the Poor' Journal ofEconomic Growth, no. 7 (3), 2001. Easterly, W. The Elusive Quest for Growth, Cambridge, Massachusetts: MIT Press, 2001. Easterly, W., and R. Levine. It's Not Factor Accumulation: Stylized Facts and Growth Models, Mimeo, Washington, DC: World Bank, 2000. 128 Transforming Bangladesh into a Middle Income Economy The Economist. 'Fragile, handle with care: A Survey ofBanking in Emerging Markets', 12 April, 1997. Fischer, S. 'The Role of Macroeconomic Factors in Growth'. Journal of Monetary Economics no. 32, 1993. Feldstein, M., and C. Horioka. 'Domestic Saving and International Capital Flows'. Economic Journal, no. 90, 1980. Foreign Policy. March/April, 2004. Ghosh, S., and A. Kraay. 'Measuring Growth in Total Factor Productivity' PREM Note # 42. Washington, DC: World Bank, 2000. Heston A., R. Summers, and A. Bettina. Penn World Table version 6.1, Center for International Comparisons at the University of Pennsylvania (CICUP), Pennsylvania, 2002. Kaufman, D., A. Kraay, and P. Zoido-Lobaton. 'Governance Matters II ­ Updated Indicators for 2000-01' World Bank Policy Research Department Working Paper no. 2772, Washington, DC: 2002. Kaufman, D. an4 A. Kraay. Growth Without Governance, Mimeo, Washington, DC: The World Bank, 2002. Klenow, P., and A. Rodriguez-Clare. 'The Neoclassical Revival in Growth Economics: Has it Gone Too Far?' NBER Macroeconomics Annual no. 12, 1997. Levine, R., and D. Renelt. 'Sensitivity Analysis of Cross-Country Growth Regressions.' American Economic Review no. 82, 1992. Mankiw, G., D. Romer, and D. Weil. 'A Contribution to the Empirics of Economic Growth' Quarterly Journal o/Economics, no. 107, 1992. Nehru, V., and A. Dhareshwar. 'A New Database on Physical Capital Stock: Sources, Methodology and Results' Revista de Analisis Economico, (Chile) no. 8, 1993. Ramey, G., and V. Ramey. 'Cross-Country Evidence on the Link Between Volatility and Growth' American Economic Review, volume 85, 1995. Rodrick, D. 'Where Did All the Growth Go? External Shocks, Social Conflict, and Growth Collapses' Journal ofEconomic Growth, no. 4, 1999. Smarzynska, B., and S. Wei. 'Corruption and the Composition of Foreign Direct Investment: Firm-level Evidence' NBER Working Paper Series no. 7969,2000. Young, A. 'The Tyranny ofNumbers: Confronting the Statistical Realities of the East Asian Growth Experience' Quarterly Journal ofEconomics, no. 110, 1995. - - - . 'Lessons from the East Asian NICs: A Contrarian View' European Economic Review, no. 38, 1994. World Bank. Foreign Direct Investment in Bangladesh: Issues ofLong-Run Sustainability, Washington, DC: Country Study, 1999. Analysis of Growth Experience 129 - - - . Bangladesh Development Policy Review: Achievement and Challenges, Report Number 26154-BD, Washington, DC: 2003. - - - . Bangladesh Public Expenditure Review, Report Number 24370­ BD, Washington, DC: 2003. - - - . Promoting the Rural Non-Farm Sector in Bangladesh, Washington, DC: Draft Report, 2004. 4 Trade Openness and Growth Sadiq Ahmed and Zaidi Sattar SECTION I: INTRODUCTION1 Few policy issues evoke as contentious a debate in developing economies as trade liberalisation, openness and globalisation. Critics comprising of politicians, labour unions, business and intellectuals see the world economy as inherently unequal, made worse by advocates of openness to trade and finance. Such openness allegedly exposes the poorer countries to the whims of the advanced countries who benefit more from this open access leading to unequal exchange and exploitation. Much of this debate is often fueled by perceptions rather than a careful review of the underlying analytical framework or empirical foundations. Like in other developing economies, this is a matter of lively debate in Bangladesh where trade liberalisation was accelerated since the early 1990s. A populist view is that trade liberalisation and economic integration has contributed to inadequate progress with industrial development, and sometimes even leading to deindustrialisation. Consequently, growth and employment prospects have suffered. Cheap imports have flooded Bangladeshi markets, making it impossible for domestic enterprises to grow and survive. This view prevails despite the fact that industrial performance under trade protection has been quite dismal. Critics also fail to recognise the gains to consumers and producers from access to lower cost consumer, intermediate and capital goods owing to greater trade openness. Not surprisingly, there is much less recognition that, despite major reduction in trade barriers since the mid-1980s, trade protection still remains a significant source of distortion in Bangladesh. 1 This is an updated version of a paper published in the Economic and Political Weekly 2004. Trade Openness and Growth 131 South Asian Scene The past decade has seen significant though varying degrees of trade liberalisation in the South Asia region known for its long history of high protection to import substituting industries (World Bank 2004). Until the last decade, with the lone exception of Sri Lanka, few countries in the region acknowledged the fact that trade could be an engine of growth and poverty reduction. What is seldom recognised in policy and business circles is that existing trade regimes across South Asia discriminated more against intra­ regional trade than they did against trade with the rest of the world. As a consequence, officially recorded intra-regional trade, which was around 20 per cent in 1947, fell to under 2 per cent by 1974, and showed only modest rise to about 4 per cent as a result of trade liberalisation in all the countries during the 1990s. That the volume of potential trade is suppressed by trade barriers between countries can be discerned from the prevalence of massive amounts of informal trade. Curiously enough, it was the multilateral liberalisation of the 1990s that served as a catalyst for increasing the volume of intra-regional trade. Two broad trends in the trade policy area can be observed in the South Asia region: (i) First, during the last decade, five countries in South Asia­ India, Pakistan, Sri Lanka, Bangladesh and Nepal- have been implementing trade policy reforms, gradually moving their economies away from protectionism toward greater trade openness and global integration, (ii) Second, all the South Asian countries are to deal with a number of major trade policy developments which are related to the new international trading system and, in particular, to the WTO, which have important implications for their economic prosperity. The combination of these two developments has resulted in significant liberalisation of trade and exchange regimes in the whole region. Nevertheless, sufficient barriers to trade remain to suggest that trade restrictiveness in the region is relatively high compared to other regions of the world. Despite significant opposition from vested groups, the considerable progress made in Bangladesh in recent years in opening markets and redirecting incentives away from import substitution toward export competition demonstrated guarded optimism on the part of policymakers in the proposition that trade openness was good for growth and poverty reduction. The present paper makes an attempt to document how the country has . benefitted from these trade reforms ofthe past decade, noting that the country did not complete the full range of possible reforms. Table 4.1 Summary of Trade Regimes in South Asia Policies India" Pakistan Bangladesh" Sri Lanka Nepal ---------------------­ Exchange rate Unified Unified Unified Unified Unified Exchange rate determination Free float Free float Free float Free float Pegged to Indian Rp. Payment convertibility Current account Yes Yes Yes, limited Yes Yes Capital account Yes, limited No No No No Import restrictions (trade reasons) General import licensing No No No, but some No No restrictions Some QRs on imports Yes Yes Yes Yes, minor Yes, minor State import monopolies (exel POL) Yes No . No No No Tariff structure May 03 Top normal CD rate 30 25 25 25 25 Average CD rate 22.2 18.2 13.5 9.6 13.7 Average CD + other protective taxes 22.2 18.2 21.4 12.5 16.2 Number of normal CD slabs 7 4 4 5 5 Uses anti-dumping Yes Yes No No No Export policies Some export QRs Yes No Yes No Yes Some export taxes Yes Yes Yes No Yes Some direct export subsidies Yes No No Yes No Indirect export subsidies Yes Yes Yes Yes Yes Trade openness: trade-GOP ratio (%) 2000 19 33 33 77 44 (*) Traiff data for Bangladesh reflect changes in FY05 I:udget. For India, changes of February 2004. 134 Transforming Bangladesh into a Middle Income Economy SECTION II: BANGLADESH TRADE REGIME IN SOUTH ASIAN CONTEXT Although the Bangladeshi economy now is much more open than it was in 1990, international trade is still subject to a host of tariff and non-tariff barriers which makes the trade regime quite restricted in the global context. Table 4.1 provides an overview of the trade regimes in major South Asian countries. While quantitative restrictions (QRs) have largely been removed in the rest of South Asia, Bangladesh is the only country that still maintains traditional QRs with the explicit purpose of protecting local industries. The most important restrictions are on the import of a range oftextile products. It has also retained general administrative controls over imports which, depending on how they are implemented, can amount to a form of import licensing. Tariffs are now the principal means by which the South Asian countries protect their domestic industries. While Sri Lanka reduced tariffs in the late 1970s, all the other four South Asian countries steadily reduced their tariffs in the 1990s, starting from very high and in many cases prohibitive levels (India, Pakistan and Bangladesh). In about 1997, in India and Bangladesh, these tariff reductions started to bite and reached the point where, for an increasing number of domestically produced products, much or most of the previous 'water' or tariff redundancy had been eliminated. This occurred at the same time as a slump in the world prices of many commodities and manufactured products happened, which accompanied the Asian financial cnSlS. This partly explains a five year period between 1997 and 2001 during which there was backtracking on tariff reforms in both India and Bangladesh. During 2002, tariff reforms in India and Bangladesh resumed, with reductions in Bangladesh's 2002 budget and successive reductions in India's 2002 and 2003 budgets. In addition, India has announced that its general maximum customs duty rate (but only for non-agricultural products) will go down to 20 per cent in 2004. As of May 2003, the general maximum customs duties (which apply to most but not all products since there are products that are subject to higher customs duties than these generally applied maxima) in the major South Asian countries were as follows: India, 30 per cent; Pakistan 25 per cent; Bangladesh 32.5 per cent; Sri Lanka 25 per cent; and Nepal 25 per cent. 2 2 These maxima are those in force in April 2003, and in the case of India are the new rates promulgated its 2003/04 budget which announced a new reduced 'maximum' Customs Duty of25 per cent. However, all agricultural tariffs were excluded from the reduction and many of these as well as a large number of non-agricultural tariffs remain at 30 per cent, and others at rates well above this. 30 per cent therefore has been treated as de facto general maximum rate. Trade Openness and Growth 135 Other Protective Import Taxes Except in Pakistan, customs duty rates alone give a misleading impression of actual protection rates for domestic industries, since India, Bangladesh, Sri Lanka and Nepal also employ other protective import taxes which are applied across the board on top ofcustoms duties. In addition to this, India in particular and the other four countries to a lesser extent, use specific tariffs which can correspond to very high ad valorem equivalent rates, depending on import prices, and Bangladesh employs two additional protective taxes for selected products. It is not possible to quantify the overall impact ofthe specific tariffs, but estimates of the unweighted averages of the combined protective effect of customs duties and other taxes are shown in Table 4.1. After allowing for these, it is evident that tariffs are still very high in India and Bangladesh: in fact, compared with average tariffs in 105 developing countries on all products , and agricultural products, India currently ranks sixth and fifth, and Bangladesh ranks fifth and seventh. By contrast, Pakistan is now a low to moderate tariff country with average tariffs about the same as Nepal's, while Sri Lanka remains overall a low tariff country (average protective import duties and taxes in manufacturing around 10 per cent), with the important proviso that its tariffs are very high in agriculture. SECTION III: EFFECTS OF TRADE OPENNESS Why Trade Openness Might be Good for Growth and Poverty Reduction Although the trade regime in Bangladesh is restrictive when judged by international standards, it is nevertheless true that considerable liberalisation has taken place over the past decade or so. The paper would show that this has benefited the economy both in terms of growth performance and poverty reduction. At the theoretical level, it can be argued that since Bangladesh is relatively heavily endowed with labour, within the framework of the well known Hecksher-Ohlin model (Ohlin 1933; Hecksher 1949) expansion of international trade opportunities will likely support the growth of labour intensive export industries, promoting employment and incomes, as predicted by the Stolper-Samuelson Theorem (Stolper and Samuelson 1941). More generally, following Krueger (1998) the arguments why greater trade openness is likely to contribute to higher growth and poverty reduction in Bangladesh might include: • Improved productive and allocative efficiency: Unlike the protectionist regime of import substitution, trade openness will likely direct scarce resources into areas ofBangladesh 's comparative advantage, promoting specialisation and resulting in higher levels of output. 136 Transforming Bangladesh into a Middle Income Economy • Factor accumulation and investment: Trade liberalisation will accelerate investment by allowing access to bigger markets, permitting scale economies, and encouraging imports of previously unavailable or cheaper capital goods and intermediate inputs. • Knowledge spillovers: Greater trade openness will produce knowledge spillovers through technical innovation that is embodied in traded goods and machinery as well as FDL • Improved income distribution: Open economies enhance the returns to a country's relatively abundant factor of production - unskilled labour, in the case of Bangladesh - thus augmenting real wages of unskilled labour, thereby contributing to improved income distribution. • Government policy improvements: As the economy is opened up to global competition, there is greater pressure on policymakers to ensure macroeconomic stability and enhance deregulation. • The paper will show, using qualitative and quantitative analysis and a case study approach, that Bangladesh's increasing global integration based on trade liberalisation and other economic deregulation, especially since the early 1990s, contributed significantly to the acceleration ofper capita income growth and poverty reduction. This is not to suggest that trade liberalisation alone was responsible for these positive developments. It is but only a necessary condition for removing the constraints to rapid growth. Other complementary measures that were taken, such as restoring macroeconomic stability, and removal of many burdensome controls on business and investment, provided the supportive policy environment for bringing about these outcomes. International Evidence on Trade Openness and Growth The most compelling argument for greater trade openness comes from the effects of this on economic efficiency which promotes private investment and economic growth. Higher growth in tum helps lower poverty by increasing employment and real incomes of the poor. Empirical evidence on the positive effects oftrade openness on growth is quite abundant (Dollar 1992; Frankel and Romer 1999; Dollar and Kraay 2001; Bhagwati and Srinivasan 200 I; Wacziarg 1998) although there are some critics who dispute these findings on methodological ground (Rodrick, 1997; Rodriguez and Rodrick 1999). Even so, Rodriguez and Rodrick caution that their main intention is to challenge the over-enthusiasm on the questionable outcomes of many research showing strong positive correlation between openness and growth, and not to convey the message that they think trade protection is good for growth. Indeed they admit that they are not aware of Trade Openness and Growth 137 any credible evidence, at least for the post 1945 period, which suggests that trade restrictions are systematically associated with higher growth rates. The most wellknown recent study that provides evidence on trade liberalisation, growth and poverty reduction is that ofKraay and David Dollar (2001). The study concludes that a third of the developing countries of the world, described as 'rapid globalisers', did extremely well in terms ofincome growth and poverty reduction over the past two decades or so. These countries, which include Bangladesh, India and Sri Lanka in South Asia, have experienced large increases in trade and significant reduction in tariff and non-tariff barriers. Bangladesh, for instance, saw its trade GDP ratio almost double (during the course of the 1990s decade). In contrast, the remaining two-thirds of the developing world, with a large concentration in Africa, that did not experience trade expansion due to a lack of sufficient outward orientation, performed poorly both in terms ofgrowth and poverty reduction. Perhaps the most useful insights emerge from a review of country experiences. These also dispel any nagging doubts about the robustness of the results coming from cross-country regressions. Amongst the well known examples of how trade liberalisation has contributed to growth and poverty reduction include Chile, China and India. Chile liberalised the trade regime unilaterally over the 1974-2000 period. Chile initially suffered from adjustment costs in terms ofloss ofemployment in protected sectors. However, between 1985-2000, Chile grew by almost 7 per cent per annum and inflation declined to around 6 per cent p.a. Importantly, Chile was able to cut poverty by more than half between 1987 and 1998. China's trade liberalisation policy started in late 1978. Within the pace of20 years, China became one ofthe most open economies in the developing world with the GDP share of trade growing from only 11 per cent in 1978 to over 40 per cent in 1998. Average GDP grew by over 7 per cent per annum and exports by 20 per cent per annum. Although there has been rising income inequality, poverty fell at rapid pace with the incidence of poverty declining from 30 per cent in 1978 to less than 15 per cent in 1998. India is yet another example of how economic liberalisation including trade liberalisation spurred growth of exports, investment, income and contributed to poverty reduction. Compared with the famous 3 per cent per annum so called Hindu growth rate over the 1950-80 period, India's GDP expanded at almost double the pace between 1980 and 2000, fueled by economic deregulation and trade liberalisation. Trade to GDP ratio grew from only 10 per cent in 1980 to 25 per cent in 2000. Poverty declined substantially, falling from 45 per cent in 1980 to 26 per cent in 2000. The progress with poverty reduction continued in the 19908 despite evidence ofgrowing income inequality. 138 Transforming Bangladesh into a Middle Income Economy These diverse country examples are a reassuring sign that trade liberalisation can be used as a positive force to push economic growth and reduce poverty. Ofcourse, there are other associated factors that have helped secure this progress. Invariably, private investment has increased in response to better investment climate and incentives. Supportive government policies in terms of investment in basic social services and infrastructure have helped. Attention to rural development and agriCUlture (in China) was an important determinant of poverty reduction. Furthermore, generally sound macroeconomic policies kept inflation low and real exchange rate at a competitive level. But the main point is that in all these cases there was a significant departure of policies from inward looking import substituting industrialisation strategy to an outward looking export promotion strategy. Overall policy framework, including macroeconomic management, exchange rate management and other policies, were supportive of this outward orientation. There are also interesting differences in country experiences that have important bearing on the design of outward-oriented public policy. • Both China and Chile gained tremendously from inflow ofdirect foreign investment following economic liberalisation. Thus, the volume ofDFI in China surged from only dollar 5 billion in 1978 to dollar 40 billion in 1998. In Chile, this grew from dollar 0.2 billion in 1974 to dollar 3.5 billion in 1998. But, in India the role of DFI has been very modest, growing from dollar I billion per annum in 1980 to dollar 3.5 billion in 2000. This is largely because supportive reforms (behind the border reforms) to enable trade liberalisation to work fully have been lagging behind significantly as compared with Chile and China. Consequently, the investment climate for doing business remains less conducive in India as compared with Chile or China (Dollar and Goswami 200 I). • The pace and magnitude oftrade liberalisation have been lower in India than China and Chile. India's tariff level is still quite high relative to the average for other developing countries. As well, other forms of trade protection, especially anti-dumping, is hurting the growth of trade in India (World Bank 2002). Consequently, the lost opportunity in terms of greater gains from trade has been large in India (Stem 2002). • Growing income inequality was seen in China and India along with trade liberalisation and rapid growth. Interestingly, however, growth has been relatively more pro-poor in China than in India (Ravalli on 2002). This suggests that there is scope for improving public policy to support more rapid poverty reduction in both countries, especially in India, by putting in place processes and policies for the disadvantaged to catch up. Trade Openness and Growth 139 Summary of International Evidence on Trade Openness, Growth and Poverty The above analysis provides strong evidence that greater trade openness is good for growth and poverty reduction over the longer term. It also suggests that there may be short term costs in terms of falling real wages of unskilled labour and or initially declining employment as greater competition drives out inefficient firms from business. Although these transition costs do not represent a credible case against trade openness as the longer term benefit would likely offset these short term cost, they need to be tackled through proper compensatory policies aimed at mitigating such costs. This is particularly important in order to manage the political economy of reforms. Since the loss is almost immediate while gains are more longer term, losers are likely to have a strong vested interest to fight the change. Additionally, in many cases, these losers are a strong political force and can provide formidable opposition to change. Research provides considerable insights on how such compensatory policies could be formulated (Bannister and Thugge 2001; Cranfield, Hertel and Preckel 2000; Edwards and Lederman 1998; Harrison and Tarr 200 I). Also, as Winters (2000) has rightly pointed out these trade-offs and associated mitigation measures need to be tackled on a case by case basis because individual country circumstances will likely be quite different. SECTION IV: LlBERALISATION EFFORTS IN BANGLADESH Since independence in December 1971 and until late 1975, Bangladesh followed a development strategy of state intervention and controls (Ahmed 2002). This first phase was characterised by massive nationalisation of most production entities, heavy trade controls, and other forms ofstate interventions as Bangladesh experimented with a socialist type economic framework. An early phase of economic deregulation and denationalisation started in 1976, although it took some years to have a clear direction for this to work out. This second phase ran up to 1990 with a mix of denationalisation, economic deregulation encompassing both goods and services sector, and limited trade liberalisation. The third phase, starting in 1991, saw continued progress with deregulation and privatisation but most importantly witnessed fairly rapid trade liberalisation compared to the past. The major progress in trade policy reform came in this last phase starting in 1991 with a substantial scaling down and rationalisation oftariffs, removal of trade-related QRs and elimination of import licensing, unification of exchange rates and the move to a more flexible exchange rate system. In 1994, Bangladesh complied with the IMF's Article VIII, making most current account transactions convertible. Unfortunately, by the middle of the 1990s, 140 Transforming Bangladesh into a Middle Income Economy trade reforms slowed, and was even reversed somewhat during the last years of the decade, on the popular notion that earlier reforms had been 'too much too fast'. Outward orientation in Bangladesh in the 1990s covered, inter alia, three major areas: • Liberalisation of imports through removal of quantitative restrictions (QRs) Reductions in nominal and effective tariffs, and • Adoption of a unified and moderately flexible exchange rate regime. It should be noted that, in contrast to the piecemeal and partial reforms ofthe 1980s, liberalisation ofBangladesh's trade regime since 1991 was more systematic and comprehensive. The overriding objectives of these policies were to promote competition and achieve neutrality of incentives between production for exports and import substitutes, while gradually making trade facilitation the centerpiece of customs administration. Liberalisation of Imports The liberalisation ofimports was done primarily via removal ofQRs in phases (see Table 4.2). Significant progress was made in removing QRs, as compared with the 1980s, to the point where a small but sensitive list of items remain restricted, mostly for non-trade reasons. Whereas nearly 26 per cent of all HS-4 digit codes were subject to QRs in 1900, now only 122 items (or 10 per cent of total items) covering about 2 per cent of imports remain restricted for trade and non-trade reasons. However, the progress towards increasing liberalisation that was evident up until the Import Policy Order of 1995-97 seems to have been haIted ­ even modestly reversed in the IPO 1997-2002, though some progress is now evident in the latest IPO 2003-06. Overall, since 1990, trade-related QRs have been progressively removed, leaving some 2.2 per cent of total HS 4­ digit tariff lines (and 0.5 per cent of import value) subject to trade-related prohibitions or bans as of 2002. These restrictions are now limited to only three categories: agricultural products (chicks, eggs, salt), packaging materials and textile products. Nearly 40 per cent of all QRs apply to textile products that enjoy the heaviest protection. Although the readymade garment sector imports woven fabrics and gray cloth duty-free under bonded warehouse facilities, the system is cumbersome and susceptible to corruption (through leakage into the protected domestic market). Some ofremaining ban - restrictions on imports is on grounds ofhealth, religion, environment, culture and so on. Yet, a review of all the items in this Table 4.2 Phased Removal of Import Restrictions IPO IPO IPO IPO IPO IPO 1987-91 1991-93 1993-95 1995-97 1997-2000 2003-06 Number of items in the control 315 193 111 120 122 63 list at the HS 4-digit level (26.1%) (15.6%) (9.0%) (9.7%) (9.8%) (5.1%) ----­ Number of trade-related items in the 253 79 19 27 27 24 control list at the HS 4-digit level (21.1%) (6.4%) (1.5%) (1.9%) (2.2%) (1.9%) Source: Ministry of Commerce. import policy orders for various years. 142 Transforming Bangladesh into a Middle Income Economy group reveals that many ofthe prohibitions or restrictions cannot be justified on these grounds, and are presumably included for protection purposes. Thus, in the interest of economic efficiency and predictability of impacts, at a minimum, it would still be more meaningful to replace the QRs with equivalent tariffs. While removal ofQRs and elimination ofimport licensing helped remove anti-export bias of the trade regime, export expansion, especially the spectacular growth of the readymade garment sector, would not have been possible without the supplementary facilitating policies such as Special Bonded Warehouse, Duty Drawback and back-to-back LCs, all of which helped ensure world-priced inputs to exporters. Tariff Rationalisation During the 1990s, Bangladesh significantly reduced its tariff rates and rationalised the structure, progressively moving towards the goal ofsimplicity and transparency ofcustoms tariff. The top CD rate came down from 350 per cent in 1990 to 25 per cent today (see Table 4.3). The average (unweighted) customs duty declined to 17 per cent in FY 02 as compared with 57 per cent in FY 92, and 100 per cent in 1985. Considerable rationalisation of the tariff structure occurred and progress was made towards achieving a degree of uniformity and removing some tariff anomalies that existed due to higher tariffs on intermediates than final products (see Table 4.4). Table 4.3 Progress in Tariff Rationalisation Tariffs FY92 FY96 FY97 FY02 FY03 FY04 FY05 Average CD (unweighted) 57 22 22 17 16.5 15.7 13.5 Average protective tax 61 74 26 27 22 9.0 7.9 Top CD rate 350 50 45 37.5 32.5 30.0 25 Source: Author estimates from customs data. Table 4.4 Summary Indicators of MFN Tariffs in Bangladesh, 1992-2003 Indicators 1991-92 1996-97 2003-04 Unweighted avg. of customs duty 57.23 21.87 15.7 Agricultural products (HS Ch 1-24) 71.80 28.86 22.01 Industrial products (HS Ch 25-97) 55.62 20.76 14.32 Standard deviation of CD 40.85 16.22 10.72 Unweighted avg. of all import taxes 82.57 39.08 37.42 Avg. collection rates (aU imp. taxes) 37.41 34.39 25.89 Avg. protective tax 60.92 25.66 24.47 Source: Author estimates from customs data. Trade Openness and Growth 143 Notwithstanding this progress, there is still a substantial way to go forward with further tariff rationalisation. First, the application of various levies, surcharges, supplementary duties (and even VAT applied discriminatingly against imports) on imports resulted in an average nominal protection of 27 per cent in FY 02, well above the protection afforded by customs duty. Lately, more than a third of the protective effect carne from these para tariffs. In fact, there was a modest increase in average nominal protection since the mid-1990s in contrast with the sharp decline experienced since 1991-92. Second, the level of effective protection afforded by the tariff structure is influenced by the level of average tariffs as well as its dispersion. Since both these indicators have shown a substantial downward trend since 1991, the presumption is that ERP has also declined over the years. However, enterprise level effective protection estimates are not available for a long period, except for FY 02. The fact that the average ERP, at 78 per cent in FY 02, is significantly higher than the average nominal protection of 27 per cent suggests the existence of tariff escalation (Table 4.5), whereby average applied tariffs on basic raw materials (11.3 per cent) and intermediate goods (11.9 per cent) are substantially less than that of final consumer goods (25.6 per cent). 3 Furthermore, there is wide variability in ERP across sectors, ranging from 23 per cent for manufacture of machinery and equipment to 258 per cent for chemical and chemical products. Also, it is worth noting that although Bangladesh has moved, in principle, to the WTO's transactions value system, in practice, discretionary valuation practices leave room for further protection and unpredictability of transaction costs. Finally, other notable features of the tariff structure that could lead to potential loss of efficiency and possible abuse of the system are: tariff concessions granted by end-use, existence of multiple rates for similar products (at HS 4-digit level), and rise in zero rates. Figure 4.1 gives the distribution of MFN tariffs for fiscal year 2003-04. Table 4.5 Tariff Escalation in Bangladesh (2003-04) Indicators Unweigh ted Customs duty Unweighted Collection rate, customs duty collection rate total taxes all taxes Basic raw material 11.33 9.49 25.26 23.47 Intermediate inputs 14.51 14.92 34.76 36.49 Capital goods 7.85 6.34 20.15 12.23 Final consumer 22.24 10.93 52.55 24.91 Source: Author estimates from customs data. 3 Review of Relative Protection 2002, Bangladesh Tariff Commission. 144 Transforming Bangladesh into a Middle Income Economy 27.53% 15 CD Rate Figure 4.1 Distribution of MFN Tariffs, 2003-04 (4-digit level) Exchange Rate Liberalisation As in the case of trade liberalisation, following the difficult period of 1972-75, exchange rate management improved gradually and by and large the adverse effects of overvaluation were avoided for most of the period since then. However, exchange rate reforms gained momentum in the 1990s. Importantly, the multiple exchange rate regime was replaced by a unified exchange rate in 1992 and the flexibility of the exchange rate system was enhanced to strengthen the competitiveness of exports. In 1994, the taka was made convertible for current account transactions, and since May 2003, the taka has been floating. Trends in Real Effective Exchange Rates (REER) Compared with India and Pakistan, Bangladesh's REER has been remarkably stable for the most part, since 1980 (Figure 4.3). During the 1970s, real exchange rate trend (Figure 4.2) is characterised by the 40 per cent depreciation in ] 975-76 and a modest depreciation4 in the remainder of the decade. Except for a period in the mid-1980s, during which the REER appreciated rapidly for three years but was then devalued sharply in 1985, the real effective exchange rate has moved within a fairly small band of about 10 per cent around its trend, which was slow devaluation from 1980 up to about 1998, followed by a final episode ofmodest depreciation starting from about 2002. The strength of the taka during this period is in part due to the rapid growth of ready made garment exports, which increased from US dollar 116 million in 1985 to nearly US dollar 5 billion in FY 04, and increasing remittances, both through formal channels and unrecorded, from Bangladesh workers outside the country. Together, these increases more than offset the decline in 4 IMF statistics on REER do not extend beyond 1980. Hence, earlier RER trends had to be drawn from S. Rahman, 1992. aid inflows relative to GDP, and were sufficient to balance whatever increases in imports resulted from the trade liberalisation measures that were implemented during the 1980s and 1990s. However, one consequence of the relatively stable taka alongside the massive devaluation of the Indian rupee between 1985 and 1992, was that the bilateral real taka/rupee rate appreciated by about 30 per cent during the same period (Figure 4.4). This appreciating trend of the real taka/rupee rate continued at a slower pace until, in 1999, the total real appreciation of the taka relative to its value during the mid-1980s was about 40 per cent. A perceptible turnaround in this trend could only be seen since 2002. Combined with Bangladesh's gradual removal of QRs and tariff reductions over the same period, this led to a rapid growth of imports from India, which during the 1990s became Bangladesh's largest single supplier, accounting for between 15 and 18 per cent of its total recorded imports. In addition, studies of informal trade suggest that there has been similar large increase in unrecorded imports from India, with a total value perhaps as large as the total value ofrecorded imports. If this is correct, India is supplying about one-third of Bangladesh's total imports. These developments have made the trading relationship with India a key concern in Bangladesh, with some groups using the import competition from India and the bilateral trade deficit to argue against further general trade liberalisation, while others are pushing for further trade liberalisation combined with reform of the customs service so as to divert illegal imports into legal channels. 1­ , .-..... ~ - Figure 4.3 Bangladesh-India Real Exchange Rate 1980-2003 146 Transforming Bangladesh into a Middle Income Economy Other Deregulation While the main focus ofthis paper is on outward orientation, it is well recognised that trade deregulation alone will not have the full beneficial effects unless supported by other deregulation in the domestic economy (the so-called 'behind the border' issues). A fullerreview of the progress in Bangladesh in this area is available in Ahmed (2002). Progress in deregulation and market orientation might be summarised as follows: • The period of the 1972-75 was characterised by heavy administrative controls on the means ofproduction and pricing. This control was pretty pervasive, including massive nationalisation of manufacturing enterprises, banking, public utilities, and distribution of agricultural inputs. State Trade Corporations handled a large chunk of international trade and domestic distribution of commodities. Price controls were pervasive, including manufacturing goods, exchange rate, interest rates, agricultural inputs and outputs, utility rates and house rents. • This control-oriented economic management changed quite dramatically starting around 1976. In this phase, the main focus of deregulation was to move towards a market-based economy by removing price controls, initiating denationalisation and reducing regulatory barriers to private investment. Much ofthe focus of domestic deregulation was on freeing up agriculture, domestic trade and services. Progress with privatisation was limited, even in manufacturing sector. Also, much of the banking and public utilities remained under public management, although the period saw the gradual growth of private sector banking. • The third phase, which began in 1991, saw a more rapid pace of deregulation in manufacturing, including more substantive progress with privatisation in the manufacturing sector. Private banking also grew significantly, even though the large public banks remained in place. In infrastructure, private telecommunications services (wireless services) emerged while some limited private generation ofprivate power emerged. However, reforms ofbanking, public utilities, and ports generally lagged behind. SECTION V: IMPACT OF ECONOMIC LlBERALISATION ON GROWTH AND POVERTY REDUCTION IN BANGLADESH It is clear from the above that Bangladesh experienced progressive liberalisation ofits economy and the trade regime since about 1976. Outward orientation gained particular momentum in the 1990s, although important foundations were laid in the 1980s, especially in the garments sector. What has been the impact of this progressive outward orientation on economic growth and poverty reduction? Trade Openness and Growth 147 Average growth and poverty reduction performance over the past 30 years, broken down by several distinct phases of outward orientation are shown in Table 4.6. The key results are as follows: • First, trade data confirm the substantial opening up of the economy, progressively over the early years. Thus, the Bangladesh trade GDP ratio a standard measure of trade orientation of the economy - rose from an average of 11 per cent in 1972-75, to 17 per cent in 1990, and further to 30 per cent in 2000. Both imports and exports grew as a share of GDP, with imports growing from an average of 8 per cent of GDP in 1972-75 to 18 percent in 2000, while exports expanded from 3 per cent of GDP in 1972-75 to 12 per cent in 2000. • Per capita GDP growth moved from (-)0.1 per cent during phase 1 to 1.6 per cent in phase 2, rising to 3.1 per cent in phase 3. • Inflation, which is key determinant of poverty, declined from a rapid pace of 47 per cent per annum in the first phase to 9 per cent in the second phase to 6 per cent in the third phase. The incidence of poverty (headcount index) fell from 88 per cent in 1974 to 59 per cent in 1992 and further declined to 50 per cent in 2000.5 • Importantly, the progress has continued since 2000 in terms of greater openness, stronger macroeconomic stability and higher growth. Although the overall growth and poverty outcomes presented above are the results of a wide variety of economic and social policies, it is fair to conclude that greater outward orientation reflected in trade openness and a supportive macroeconomic management played a key role. Indeed, as argued by Sriniva.;;an and Bhagwati (2002), outward orientation is generally supported by sound macroeconomic management that helps maintain low inflation to preserve the competitiveness of the exchange rate. This is also true of Bangladesh. The rapid control over inflation during phase 2 and 3 was not an isolated development. It was aimed at supporting the management of the external balances and the real exchange rate. In addition, progressively improving exchange rate management and supportive prudent macroeconomic management were complemented by other market-oriented reforms such as deregulation of industries and a liberal investment regime that encouraged investment. While these aggregative results give comfort that outward orientation has on average yielded positive outcomes in terms of better growth and lower poverty, it would be reassuring to obtain more direct linkage between 5 There are some problems of comparability of these point poverty estimates over the various years, but there are no doubts about the general trend of these numbers. See Ahmed 2002. Table 4.6 Growth Perfonnance and Poverty Reduction, 1972-2004 (per cent) Period Average GOP Average per capita Average Poverty headcount, Trade-GOP ratio, growth GOP growth inflation rate end period end period 1972-75 2.75 -0.10 47.14 88.1* 11.28 1976-90 3.89 1.58 9.12 58.8** 17.4 1991-2000 4.80 3.09 5.64 49.8*** 30.1 2001-04 5.12 3.70 3.74 31.3b Note: * refers to 1973-74. ** refers to 1991-92. *** refers to 1999-2000. 8 average for 1972-75. baverage for 2001-04. Source: Bangladesh Bureau of Statistics and author estimates. Trade Openness and Growth 149 outward orientation and poverty reduction by looking at sectoral performance: employment and income generation in manufacturing -large and small-scale, in agriculture and non-farmloff-farm activities, and within the broad category of service activities, that could have been affected by the changes brought about by greater trade openness. For Bangladesh, a labour abundant economy, growth to be pro-poor must also be based on labour-intensive production. Trade openness promotes production in accordance with a nation's comparative advantage, which in Bangladesh's case, would be in those production activities that have a relatively high labour to capital ratio. Sector Growth Performance Table 4.7 provides a summary and comparison ofgrowth performance in key sectors over the three phases. 6 (a) Agriculture In the first phase, Bangladesh began with a pretty heavy dose ofcontrols over both the product and input markets for agriculture. Following the debacle of the phase I, agriculture growth picked up considerably in phase 2, fueled by input-output deregulation as well as the supportive policies of the green revolution (the seed-fertiliser technology). Deregulation in agriculture started in early 1980s. This involved liberalisation of the fertiliser and irrigation equipment markets, and the reform of the public marketing of food grains. While much of the reform focused on agricultural inputs, on the output side the main noteworthy reform was the abolishment of most forms of food rationing and the monopoly in the import and export offood grain. The overall effect of this deregulation was positive, contributing to an expansion in agricultural productivity and value-added. Much of the positive impact ' happened in rice production - the dominant agriculture activity. Deregulation enabled rapid adoption ofthe high-yielding varieties (HYV) of rice, causing paddy production to increase at a faster pace than population and Bangladesh achieved rice self-sufficiency by the early 1990s. Direct effects of trade liberalisation came in the form of imports of agricultural inputs and implements in the mid-1980s (coupled with deregulation of domestic fertiliser markets) providing the initial impetus to 6 Comparison ofsectoral perfonnance is a bit complicated by the revision af national accounts in (1995). The revised series do not go far enough and stop at 1980. For this study, we use the old series to calculate growth indicators for phases I and 2 and the new series for phase 3. This could bias the magnitude of the growth estimates somewhat but does not affect the direction of change as indicated by the comparison of growth estimates for 1980-90 in the two series. 7 Abdullah and Shahabuddin 1997; Ahmed 1997. Table 4.7 Sectoral Growth Performance (per cent) Average Share in Average Share in Average Share in Average Share in SUb-sector growth 1975 growth 1990 growth 2000 growth 2004 (1972-75) GOP (1976-90) GOP (1990-2000) GOP (2001-04) GOP 1.02 43.09 2.58 25.15 2.60 19.49 2.65 15.89 Crops & horticulture 0.98 39.34 2.69 19.34 2.85 14.59 2.09 11.70 Animal farming 1.64 3.75 2.10 3.73 2.51 3.02 4.13 2.43 Fishing 0.36 5.00 2.35 4.37 8.21 6.09 0.92 4.66 Manufacturing -10.92 10.88 4.99 12.52 6.90 15.40 6.58 16.05 Large-scale -22.36 4.81 4.94 8.91 6.95 11.01 6.26 11.33 Small-scale 0.87 6.07 5.15 3.61 6.78 4.39 7.39 4.73 Construction -19.44 2.81 6.02 6.01 8.31 7.84 8.42 8.12 Wholesale & retail trade 12.24 8.65 4.50 12.17 5.67 13.35 6.41 13.75 Hotels & restaurants n/a n/a 4.12 0.59 5.49 0.63 7.05 0.68 Transport. Storage & Communication 3.66 10.70 4.57 9.32 4.89 9.20 6.99 11.2 Other services 18.87 29.87 28.0 27.16 Source: Statistical Yearbook of Bangladesh 2000. BBS and National Accounts. Trade Openness and Growth 151 agricultural growth in the late 1980s, which was then sustained in the 1990s. In agriculture, therefore, trade liberalisation with supportive liberalisation of the input market, provided impetus to rice (a labour-intensive import substituting activity) that expanded through higher area coverage, intensity of cultivation, and adoption of high-yielding technology. Moreover, there is now a growing segment in agriculture - vegetables and horticulture that is geared towards export markets. Export volumes for such products, though modest in relative terms (about dollar 40 million in FY02), has been rising rapidly in the recent past (exports in FY 00 were five times those of FY 90). A major export­ oriented agriculture sector that benefitted from trade liberalisation is the fishing industry. Fishing as an economic activity grew at a record pace during the 1990s, largely driven by the export-oriented shrimps production that responded to exchange liberalisation and other trade incentives (Table 4.8). Liberalisation of imports allowing duty-free inputs most needed in commercial fish farming also helped strong growth. Table 4.8 Shrimp Production and Exports Year Production Value added Export of frozen shrimp (M7J (US$ million) (M7J (US$ million) FY90 70,776 121.85 17,505 126.9 FY91 80,384 138.42 17,985 128.0 FY92 98,419 164.69 16,730 119.7 FY93 123,231 225.73 19,224 155.5 FY94 100,542 196.63 22,054 197.7 FY95 113,366 244.75 26,277 260.7 FY96 101,655 267.24 25,225 270.5 FY97 112,634 286.44 25,742 279.2 FY98 146,356 368.85 18,630 260.4 FY99 157,371 375.41 20,086 242.2 FYOO 170,788 386.36 28,514 322.4 FY01 n/a n/a n/a 349.8 Source: Bangladesh Bureay of Statistics, Exports Promotion Bureau. Yet, overall the agriculture sector did not show dynamism. The perfor­ mance of the agriculture sector remains heavily dependent upon the contribu­ tion of rice, which seems to have become constrained by slowdown of domestic demand due to low income elasticity and the slowdown of the popUlation growth rate. Given the importance of agriculture for poverty reduction, it has been a subject of considerable analysis, debate and discussion. s Despite 8 A good summary of these issues is contained in Abdullah and Shahabuddin 1997; Mahmud 1998; and Faruqee 1998. 152 Transforming Bangladesh into a Middle Income Economy progress, agriculture continues to suffer from many controls over output pricing, marketing, and input supply; removal of these controls will benefit farm production and value-added. Deregulation has focused mostly on input markets. On the demand side, low expansion of the pace of domestic demand and inadequate export orientation continue to limit the incentives for production. So, diversification of agricultural production away from the heavy reliance on rice is important to raise agricultural value-added. This in turn will require policies to boost domestic demand along with efforts to remove constraints to agricultural exports. The prospects for non-rice agriculture exports are also limited by the relatively low productivity. So, policies for promoting technological progress in non-rice agriculture are very important. Inadequate agricultural infrastructure is another constraint on farm production and productivity. Overall water management and flood control policies are deficient, suffering from serious management problems, including O&M. Land markets function inefficiently due to inadequate land administration policies, including the inadequacy ofthe legal process for land tenure arrangements and land sales. 9 Finally, the inadequacy of the farm credit market remains a major constraint on farm production. Growth of shrimp sector. A notable exception in non-rice agriculture is the shrimp sector. The shrimp sector has expanded rapidly over the 1990s, thanks to the rise in the volume of exports. Labour-intensive in character, it now employs over half a million rural poor in various stages of processing and shrimp culture. Shrimp farming is now a major economic activity in the coastal regions of Bangladesh. Bangladesh accounted for 4.1 per cent of global production of commercial shrimp in the mid-1990s, and after RMG, the shrimp sector is now the second largest export industry. The production of shrimps doubled in the last decade, while exports multiplied three times within the same period of time. Its share in fishing sub-sector rose from II per cent to 14 per cent by the end of the decade, but it made up 90 per cent of export earnings from fish. Export-oriented shrimp has expanded from a US dollar 10 million business in 1975 to US dollar 77 million in 1985, and is running close to US dollar 400 million today (Table 4.8). Over the last decade, the industry has registered a robust growth rate of about 9 per cent per annum (in dollar terms). Except for the readymade garments industry, no other industry in Bangladesh has been able to match this performance. (b) Manufacturing Much of the debate on the possible adverse effects of trade liberalisation, denationalisation and deregulation policies has concentrated on the impact 9 See World Bank 2000. Table 4.9 Growth of the RMG Sector Year Export volume Export value Share in total Employment (million Number of garment ('000 dozen) (US$ million) exports (%) workers) factories 1985-86 4,762.6 131.5 16.1 0.2 594 1990-91 30,566.6 866.8 50.5 0.4 834 1995-96 72,005.0 2,547.1 65.6 1.3 2,353 1999-2000 111,905.8 4,349.4 75.6 1.6 3,200 2001-02 140,444.6 4,583.8 76.6 1.8 3,618 2002-03 152,013.0 4,912.1 75.1 2.0 3,760 2003-04 161,343.0* 5,043.1* 74.6* 2.0 3,962 ,. July 2003-May 2004 Source: Bangladesh Garment Manufacturers and Exporters Association & Export Promotion Bureau. 154 Transforming Bangladesh into a Middle Income Economy of these policies in manufacturing. For years, the trade regime in Bangladesh was geared to providing high protection to the manufacturing sector, in general, and some so-called thrust industries like textiles, in particular. While effective rates of protection (ERP) in agriculture were either very low or even negative, average ERP in manufacturing was in excess of 100 per cent for much of the 1970s and 1980s. Recently, after all the liberalisation of the 1990s, Bangladesh Tariff Commission's Review of Relative Protection 2002 found average ERP in a large sample of manufacturing enterprises to be 78 per cent. The review also found that although average nominal protection rates had come down significantly, the sector stubbornly resisted reduction of effective rates by pressing for and obtaining reduction in tariffs of intermediate inputs as well as the continuation of tariff concessions by end-use for imported capital machinery and many industrial raw materials. What has been the effect of this trade liberalisation on the manufacturing sector? Since much of the impetus to trade liberalisation came in the early 1990s, it is possibly most appropriate to compare the performance of the manufacturing sector in phase 2 and phase 3, with the caveat that even phase 2 benefited from significant deregulation and the rapid expansion of the export­ oriented garment sector based on establishment of the free trade zones (see below). On average, the share of manufacturing in GDP rose by almost 3 per cent of GDP in the 1990s, from 12.5 per cent in the 1980s. This fact, in itself, belies the contention of critiques who argue that trade liberalisation of the 1990s resulted in deindustrialisation (Zaid Bakht 2001). Nevertheless, it is accepted that tariff reductions and QR removal introduced a substantial degree of import competition in the local manufacturing sector, forcing enterprises to restructure and raise productive efficiency. Many did, such as ceramics, textiles (new spinning capacities), RMG accessories, electrical goods, etc. Those that failed to adjust had to close down and layoff workers [see World Bank (1999), for a detailed assessment of the impact of trade liberalisation on the manufacturing sector]. In this category, were a large number of state­ owned manufacturing enterprises which ended up becoming sick (unviable), and could only continue operation with regular budgetary subsidy. In this group were a large number of jute and cotton textile mills, apart from paper and chemical products. Private enterprises which were beneficiaries of high protection for long but failed to adjust following liberalisation had no other option but to close down. Some metal products; light engineering, and rubber products are in this group. Such outcomes are predictable and the natural consequences of trade liberalisation. Manufacturing growth averaged 8.2 per cent per annum (only 4.3 per cent for non-RMG) in the first half of the 1990s during the peak of the liberalising period, but tapered off to an average of 5.6 per cent in the latter half, to end the decade with an average of 6.9 per cent growth, compared to Trade Openness and Growth 155 5 per cent in the 1980s. These averages mask the fact that it was the RMG enterprises (in the medium and large scale group) that grew by over 20 per cent and drove manufacturing growth, while there was stagnation and even decline in some import-substituting industries a predictable impact of trade liberalisation in a manufacturing sector that enjoyed high protection for a prolonged period. Most notable was a sustained average growth of 6.8 per cent in the small-scale enterprise sector, compared to only 5 per cent growth in the previous decade. This is anything but surprising. Small and rural or peri-urban entrepreneurs typically lack access to licenses, permits or exemptions that their larger compatriots extricate from the controlling authorities. Tariffs and QR regime served as constraints to the availability of raw materials and intermediate inputs for small firms. Thus, import liberalisation and the abolition of import licensing improved access of small enterprises to capital machinery, raw materials and implements that could now be purchased readily and at low cost. Import liberalisation thus acted to eliminate supply constraints in the sector. Moreover, their products being poor substitutes for imports, small-scale enterprises benefited more from the removal of supply constraints than they suffered from increased import competition, unlike their large-scale counterparts (Bakht 2001). Growth of RMG industry. The phenomenal growth of the readymade garment sector in Bangladesh is a major success story of an export-oriented activity, giving a major boost to domestic manufacturing and exports, while creating direct and indirect employment and investment opportunities in backward and forward linked activities. Notwithstanding the fact that this sector's emergence and expansion is the direct outcome of the global MFA regime, there is no denying that it has had a stellar impact on overall economic growth, income generation and poverty reduction in Bangladesh. Table 4.9 gives a summary picture of the evolution of this sector from humble beginnings in the late 1970s. From a minuscule share of about 4 per cent in total exports in the early 1980s, garments now constitute 76 per cent of total exports from Bangladesh, raking in nearly dollar 5 billion of foreign exchange, out of total export earnings of dollar 6.8 billion in 2001-02. Net domestic value addition hitherto a weak point, on account of the heavy dependence of the sector on imported fabrics, yam and accessories - has risen substantially, so much so that nearly 60 per cent of the required inputs are now domestically sourced (Table 4.10), as compared to a mere 36 per cent in 1991-92. This sector has been unique in its ability to create jobs, particularly for the women workforce. From 2,00,000 workers in FY 85-86, RMG industry now directly employs 2.0 million workers - almost 90 per cent of them women in its 3,962 factories. This was the fastest employment creation experienced by any sub-sector in Bangladesh. BGMEA, the industry association, claims (perhaps rightly since no serious Table 4.10 Value Addition in the RMG Industries (US$ million) FY92 FY93 FY94 FY95 FY96 FY97 FY98 FY99 FYOO FY01 FY02 FY03 FY04* Exports of: RMG 1062 1231 1292 1835 1985 2238 2843 2985 3083 3364 3125 3255 3116 Knitwear 119 205 264 393 598 763 940 1035 1270 1496 1459 1652 1859 Specialised textile & household linen 18 29 32 33 41 52 58 76 96 117 126 128 182 Total 1199 1465 1588 2261 2624 3053 3842 4097 4448 4977 4710 5036 5156 Imports against back-to-back UCs: Textile fabrics & accessories 746 878 1034 1524 1434 1581 1886 1729 1992 2142 1927 2031 2056 Domestic content: US$mill 453 587 554 737 1190 1472 1956 2368 2456 2835 2783 3005 3100 Percentage 37.8 40.1 34.9 32.6 45.4 48.2 50.9 57.8 55.2 57.0 59.1 59.7 60.1 .. July 2003-May 2004 Source: Bangladesh Bank and Export Promotion Bureau. Trade Openness and Growth 157 data has been compiled on indirect employment effects) that a similar number of workers might be employed in a host of support activities (backward and forward linkages, e.g. courier, packaging) that have also emerged alongside the expansion ofRMG. The largely female workforce ofRMG industries are drawn from the vast reserves of unskilled labour amongst the poor rural communities. Surely, this has made an enormous contribution to lifting several million rural families out of poverty over two decades, quite apart from advancing the cause of women's empowerment in a big way. Another example is the case of leather manufacturing. From modest exports ofdollar 16 million in 1973, exports ofleather and leather goods rose to nearly dollar 140 million by 1989, and to dollar 250 million in FY 02, making it the fourth largest export category in Bangladesh. Although export growth has been robust, this sector is miles behind its competitors, Pakistan and India, in terms of quality and productivity. While goat hides are reputed to be the best in the world, cow hides are the worst in quality, and the sector has been unable to attract the technology and investment required to raise quality standards to become a significant player in the world market for leather or footwear. Nevertheless, it accounts for 12,000 jobs directly and employment has been remarkably stable. Because of the polluting nature of this industry, there are indications that this is a sunset industry in Korea and Taiwan, needing re-Iocations to such places as Bangladesh; but the investment climate here has not been judged as favourable by prospective foreign investors in the sector. (c) Construction and other services The construction sector benefited in both phases 2 and 3 from the growth of income and from remittance inflows. Investment in housing provided the impetus in the second phase fueled by the rapid growth ofremittances, while other construction activities provided the lead in the third phase. Additionally, on the supply side, liberalisation of the import regime, and complementary policies ofmarket orientation and deregulation ofthe investment regime helped the sector to expand. A liberalised import regime that improved the availability ofimplements and construction materials at internationally competitive prices supported the growth ofthis sector by reducing cost. At the close ofthe 1990s decade, the share of construction in GDP had risen by nearly 2 per cent of GDP, spurred by a significantly higher annual rate of growth (8.3 per cent compared to 6 per cent in the 1980s). Besides construction, the largest expansion of activity in the 1990s decade came in retail and wholesale trade (average growth up by +2 per cent in 1990s), hotels and restaurants (up by 1.2 per cent) and transport and communications (up 0.3 per cent) - activities that are clearly linked to the level and growth of trade (trade-GDP ratio). The stimulus to these activities 158 Transforming Bangladesh into a Middle Income Economy 2,500 2,000 ,., * Q i5. E Ul 1,500 1,000 § 500 ~ ~ ~ ~ ~ ~ ~ ~ ~ 8 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Figure 4.4 RMG Employment FY 1982-2004 of course come from the spurt in production and investment that occurs in the agriculture and industry. Export-oriented manufacturing was the main driver of this growth, followed by readjusted lSI enterprises, which had to be financially and managerially nimble to remain competitive. Import liberalisation facilitated not only export-oriented firms but also the vast majority of small and large firms feeding the domestic market. Some 75-80 per cent of Bangladesh's imports continue to be made up of intermediate inputs, industrial raw materials and capital goods. Aggregate production in Bangladesh is still import-intensive. Import liberalisation, by removing various control mechanisms, had the general effect of eliminating supply constraints and generating a supply response that contributed to higher GDP growth. Impact of trade liberalisation on employment. The lack of a consistent set of data from periodic LFS, makes it difficult to draw conclusive assessments of the impact of trade liberalisation on aggregate employment. While acknowledging that intensive liberalisation and opening up of the economy that occurred in the early 1990s would have adversely impacted production and employment, especially for highly protected import-substituting firms, it is also true that trade openness and removal of anti-export bias led to the establishment of new industries and robust expansion of export industries, such as RMG. Hence, in this paper, we limit our discussion to the impact on employment in the RMG sector for which empirical evidence was readily available. Employment in RMG. During the 1990s, the fastest rate of employment creation was in the export-oriented RMG sector (Figure 4.4) which directly employed 1.5 million workers in 2000 (and 2.0 million in 2004) growing from only 200 thousands in 1990-91, 80 per cent of whom were females. It is important to note that the RMG sector helped create jobs in complementary industries or services, such as accessories, packaging, toiletries (demanded by newly employed female RMG workers), courier, finance, transport and telecommunications services, etc. BGMEA claims, quite rightly, that the RMG sector creates as many jobs in these complementary enterprises as there are in Trade Openness and Growth 159 2,600 2360 2,400 2,200 2,000 ;; 1,800 t:.. 1,600 Q) CBNWage 1ii 1,400 c:: 1,200 Equivalence Q) OJ 1,000 en ~ Io "Iii () '" Co u:: -gS: t'CI en c: ~ III c: 0 .s w 0 () en lL Figure 5.11 Corruption: Unofficial Payments to Government Agencies Source: World Bank and Bangladesh Enterprise Institute, June 2003. SECTION IV: THE EFFECTS OF BANGLADESH'S INVESTMENT CLIMATE ON FIRM PERFORMANCE The deficiencies in Bangladesh's investment climate discussed above have a significantly negative impact on the performance of Bangladesh's enterprise sector. The Costs of Poor Logistics As discussed above, the quality of logistics is critical for the enterprise sector, not only for exporting firms but also for domestic-oriented firms because most of them have to rely significantly on imported inputs. Even for firms that source primarily from, and sell largely to, the domestic market logistics is an important determinant of performance. Deficiencies in the logistics system thus affect Bangladeshi firms in many ways. Logistics is particularly important for perishable goods. An example is 192 Transforming Bangladesh into a Middle Income Economy the shrimp industry, one of the major export industries of Bangladesh. While the outbound supply chain, i.e. from the shrimp processing factories up to the ports has improved over time, especially with the use of refrigerated trucks, there are problems in the input supply chain. Here, logistics linking the hatcheries to the shrimp farms is often poor. Numerous middlemen operate in the supply chain without necessarily adding much value. A consequence of this is the high mortality rate of shrimp fries. Fresh vegetable is largely exported by air using Bangladesh Biman carriers. Biman has a monopoly on cargo export, it subsidises air freight for vegetable exports but has limited capacity. This generates excess demand for air freight, especially for the Middle Eastern markets. Cargo space is allocated through quotas which serve as a constraint on the large, perhaps more efficient, exporters. Inefficiencies in the air freight system generate considerable uncertainty for vegetable exporters (see Box 5.1). Since air freight costs comprise more than 80 per cent of the total costs of delivering fresh vegetable produce to a foreign, high-end supermarket, the level of efficiency in this part of the supply chain is critical. As the vegetable export business grows, it will also be necessary to convert to frozen vegetables and to utilise ocean freight in order to remain cost competitive. The footwear industry also faces handling, transport and other inefficiencies, which decreases their global competitiveness significantly. As discussed above, lead times with regard to input supply and output delivery have become an important determinant of competitiveness in the garment industry. Port performance is an important determinant of lead times for Bangladeshi industry. Delays in clearing imported inputs through the Chittagong port adversely affect the garments industry with fIrms often missing delivery dates. Regression analysis carried out for the Bangladesh investment climate assessment suggest that the waits at the port can be costly to fIrms. Regression analysis controlling for industry and firm characteristics suggests that import delays are associated with lower profIts, while customs delays for exports are correlated with slower growth in sales and employment and lower investment. 17 While imports are typically delayed longer than exports, firms tend to be hurt more by export delays. Indeed, each day that exports are delayed in customs is associated with a 0.3 percentage point reduction in investment and a 0.2 percentage point reduction in sales and employment growth. Interpreted causally, these figures suggest that the average wait of nine days for exports reduced the three-year average for sales and employment growth by nearly 2 percentage points and investment by 2.7 percentage points. Among the fIrms for which all the relevant data are available, sales growth averaged around 11 17 See World Bank and Bangladesh Enterprise Institute, 2003. Investment Climate for Private Enterprise 193 Box 5.1 How Poor Logistics Affect Vegetable Exports It is reported that, in about five per cent of the cases, Bangladesh Biman rejects fresh vegetable shipments because of insufficient air cargo capacity. The rejected consignments are then sold in the local Dhaka market at much lower prices than could have been obtained through exports. Apart from lost revenues, this practice generates uncertainty, especially for those exporters that would like to expand their operations and enter into larger supply contracts with vegetable producers. Another source of uncertainty is the numerous flight delays caused by rescheduling and plane repairs and maintenance. This has huge costs because there are no cold-storage facilities at the Dhaka airport and fresh vegetables delivered to the airport have to stay in un­ refrigerated air freight containers on the tarmac, often for hours. The quality of fresh vegetables thus deteriorate rapidly; in some cases, foreign buyers have rejected up to 30 per cent of a Bangladeshi fresh vegetable shipment. 80% 2.5% 70% ., (!) 2.0% Ol 60% S0 ~ (j; £ ~ 50% 1.5% ~ Q. (!) £: (!) 40% 2 (!) Ol .r:; :> '0 .~ 1.0% \91 \) \# \91\) \91\) \# \' .4,0:1'" 4,~ .4,0},'\ 4,0},'0 1..0:10}, 4,~ 4,(;;)" 4,(;;)'1.- 4,(;;)".) 4,~ ~ ~ ~ ~ «' ~ «' ~ ~. ~ ~ « « « FIgure 7.1 Direct Value Added Contribution of SOEs to GOP 1400 1200 1000 j .el 800 ~ I'" 800 @. 400 200 0 I q," OJ".) .4,q,":> .4,~ .4,,g; ~4: ~4: «, «, «, ,<;:J" .f(..... «.,<;:J".) .. 'I .-M-an-U-fa-ctu-n-'ng-o-N-on-.M-ar--l-ufa-ct-u-rin-g-1 Figure 7.2 Trends in SOE Assets need to respond aggressively to the challenge of rapid and efficient sector development so as to enhance economic growth and speed up the creation of economic opportunities for the poor and pace of poverty reduction in this decade. Bangladesh has lagged behind other developing countries, including some in the region, in undertaking SOE reforms to deal with the underdevelopment of key sectors, and the inefficiencies of service and fiscal costs associated with SOE operations. Public sector involvement in Bangladesh in a number of areas of manufacturing, energy, transport and utility services through state-owned entities has been inefficient and costly to the economy. This bas been manifested in various fonns: minimal return on resources; poor dinect contribution to GDP and economic growth; low coverage of services, particularly for the poor; unsatisfactory quality and high transaction cost of sel'"Vices; inefficient and inequitable subsidies; high fiscal costs of SOE deficit financing; and substantial contingent liabilities for the government. The cost of inefficient public services in energy, telecommuni­ cations, railways, ports, urban water supply and banking, in tenns of increased cost ofdoing business, economic opportunities foregone and consumer welfare lost, has been high. The SOEs employ substantial resources (size of assets equal to 36 per Reforming State Owned Enterprises 219 cent of GDP). The economy's growth has been limited to some extent by their low efficiency of services in key areas and on account of their poor financial returns. Their direct contribution to GDP in terms of value addition has declined to 1 per cent in nominal terms in recent years, while their contribution to growth has been minimal in most years. The inefficient operation of SOEs has led to a massive resource transfer, estimated to be equal in magnitude to 4.8 per cent of GDP annually during FY1991-2004, assuming a warranted return on assets of 13 per cent. A part of this is the fiscal resource transfer on government loans and equity provided to SOEs on account of subsidised capital, and poor recovery of interest and dividend ­ conservatively estimated at around 2.8 per cent of GDP annually during FY 1997-2000. The true fiscal resource transfer was higher than this on account of budgetary subsidies, grants, default in debt amortization, conversion of loans to equity or grant, erosion of equity due to accumulated losses, and recapitalization ofbanks against classified loans owed by SOEs. In addition, debt service liabilities owed by SOEs to the government had been underestimated in the past, and the revised estimate could add up to 0.8 per cent of GDP to the above estimate of fiscal resource transfer. The public financial institutions have performed poorly, suffering from significant and understated losses and capital shortfall. A recent special evaluation of 4 NCBs, based on international accounting standards (lAS), found their true losses to be much higher than audited books for 2002. Their true capital shortfall has also risen since the last evaluation in 1996. The liabilities and capital shortfall of the SOEs, arising from their poor fmancial performance, represent a major source ofcontingent liability for the exchequer, a part of which has been realised. Given that SOEs are involved mostly in commercial activities, the rationale for 'public' provision has been weak or largely missing in many areas. Because ofpersistent losses and poor returns and associated fiscal costs, the 'financial' rationale for public ownership has also been largely missing. Complacency about the outcome of public provision and its impact on the economy and public finances contributed to a status quo on the SOE reform agenda for many years which has been costly, and it is imperative that the welcome reform initiatives since FY 02 be scaled up significantly while adequately addressing the social costs ofadjustment. Complacent views about SOE performance and reservations about structural adjustment reforms in some areas, including corporatisation, functional unbundling, manpower rationalisation, private participation and divestiture, have been expressed within government, SOEs, interest groups, academic community and other sections of the society. 2 In the context of the high costs of SOE operations, 2 Such a view was expressed in the SAPRl Report led by Center For Policy Dialogue 2002. See also Bhattacharya et al. 200 I. 220 Transforming Bangladesh into a Middle Income Economy postponing the unfinished reform agenda would be costly in terms offoregone growth and poverty reduction. The urgency of a decisive stance on implementing the unfinished SOE reform agenda - particularly in the context of the backlog of sector underdevelopment, limited access to services by the poor, high costs of public provision, emerging opportunities for private participation, and the need to prioritise the use of budgetary resources for provision of public goods to promote faster growth and poverty reduction could hardly be overemphasised. The measures initiated since FY 02 to address poor performance in various areas are steps in the right direction but much more needs to be done to implement the substantial unfinished reform agenda. The reforms will need to be carried out with a human face, in consultation with relevant stakeholders, taking into account their legitimate concerns, and supported by safety nets to address the social costs for those likely to be affected adversely by the reforms. Role of SOEs in the Economy The SOEs have been involved in a wide range ofcommercial activities in key sectors of the economy, employing significant resources. Growing at an average annual rate of 9 per cent in nominal terms, the SOEs' assets today are more than threefold their levels in the early 1990s (equal in magnitude to 36 percent ofGDP). The domain ofactivities has included jute goods, textiles, steel and engineering, fertiliser, paper, newsprint, wood products, sugar, power, gas, petroleum, telecommunications, railway, ports, shipping, aviation, inland water and urban road transport, urban water supply, urban land development, hotels, tourism, trading and banking. The entities have held near monopoly position or high market shares in many of these activities. Until FY 02, there has been little restructuring or downsizing of the SOEs in Bangladesh in the last decade, unlike in most other countries. FDI inflows that could help substitute for public investment and provision have been relatively small, being limited to a joint-venture fertiliser plant, a few power purchase and gas production sharing contracts and mobile phones. Contributing less than 2 per cent ofGDP, the SOEs have claimed on average 8 per cent ofGDI in the last decade. They currently account for around 5 per cent ofmanufacturing sector employment. The market share of the PFIs has come down from around 70 per cent ofbank loans and credit in the early nineties to 51 per cent at present. This largely reflects the growth of private sector banking, rather than restructuring or privatisation of these public financial entities. The BTTB has operated as a state monopoly in urban fixed telephone services while BR is a state monopoly in rail transportation. Reforming State Owned Enterprises 221 Governance and Management of SOEs The corporate governance and management of the SOEs has not been conducive to their efficient commercial operations and institutional development. Much of the shortfall in their performance can be linked to ineffective or unresponsive governance and management. This is attributable to a variety of factors, including their limited operational autonomy, unclear mandates, conflicting policies, low incentives and accountability, lack of corporate structure and culture, and inadequate commercialization of many entities. Overstaffing and skill mismatch, particularly dearth of competent technical and managerial expertise, have compounded their problems. This is linked to their inappropriate compensation system and the process of selecting technical personnel and managers, which have not been competitive. Private sector representation on the Board of Directors has been limited to two SOEs (e.g. DWASA and BWDB), but its impact on improving governance and operational performance appears to have been limited so far. In the case ofthe NCBs and SDBs too, such representation did not result in a noticeable improvement in their governance and performance. A welcome initiative has been the initiation of efforts to restructure the NCBs and improve their operational performance with assistance of private management advisors. Some parastatals have been required to carry out quasi-fiscal activities, with little or no budgetary compensation. The NCBs and SDBs for many years have enjoyed little autonomy, and political or partisan considerations could have been at work in high-level personnel decisions and in shaping their senior management, precluding the participation of the most qualified and experienced personnel available in the profession. These problems were exacerbated by the exclusion of NCBs and SDBs from the purview of Bangladesh Bank's regulatory control, until a recent amendment ofthe relevant legislation. Interference of employees and trade unions in the operational and personnel matters ofthe entities has further constrained their management. SECTION II: FINANCIAL PERFORMANCE OF SOEs The financial performance of SOEs has generally been quite unsatisfactory in the last decade, leading to a significant resource transfer on account of their operations, and some welcome measures initiated since FY 02 have contributed to their loss reduction. Gross SOE losses (counting the loss­ making entities only), averaged taka 15 billion annually during FY1991~ 2002, while net annual losses of all SOEs averaged taka 9 billion annually. The BPDB, DESA, BJMC, BCIC, BTMC and BSEC contributed most to these losses, while BPC lost heavily during FY 2000--02. Largely as a result ofrestructuring, mill closure, manpower rationalisation, system loss reduction and pricing measures, the net annual SOE losses were contained during FY 222 Transforming Bangladesh into a Middle Income Economy 2003-04 at taka 3.5 billion (Figure 7.3). This outcome largely reflects the lower losses of BPDB, DESA, BJMC and BCJC, and overall losses would have been lower without BPC's continued high losses. The pre-interest rate of return on assets (ROA) of SOEs for the period FYI 991-2004 as a whole averaged a mere 0.25 per cent annually, falling significantly short of the warranted ROA, assumed in this study to be 13 per cent. The implied shortfall in SOEs' returns on assets, and hence resource transfer on account of public provision through the entities, was equal to 4.8 per cent ofGDP annually on average during FY1991-2004 (Figure 7.4). This is likely to understate the true shortfall because of various reasons, including weaknesses in financial accounting and asset valuation. 350 300 Assets Equity + Uability 250 • Equity 0 Uability 200 150 100 50 o+-'---'-""""--'-r-'---'-r­ --!iO -100 Figure 7.3 Equity. Liability and Assets of Major SOEs: FY 04 2 0 CL C C:l -1 '0 -2 il <> G; --3 .eo --4 -0 -6 - - - Shortfall in Returns as per cent of GOP _ _ _ Overall ROA (per cent) Figure 7.4 Rate of Return on Assets (ROA) of SOEs and Shortfall in Returns as % of GDP (FY 91-04) Reforming State Owned Enterprises 223 The net worth of SOEs, currently around one-third of total assets, has been eroded by significant accumulated losses. The net worth and asset build up of SOEs would have been much lower without the substantial budgetary equity fmancing, conversion ofgovernment loans into equity or grants, explicit and hidden subsidies, loss financing and asset revaluation by some agencies, partly offset by nominal dividend payments, contributed by a few profit­ making entities. Their net worth (or equity) is, therefore, not a true reflection of their financial performance. The BTTB has contributed a modest revenue surplus, largely due to its state monopoly in fixed telephones in urban areas, rather than efficient operation. Its net contribution to the exchequer, taking investment under the ADP into account, has been much more limited. Bangladesh Railway has been losing heavily because of a variety of factors, including overmanning and unresponsive management, linked to the absence of a corporate structure and commercial orientation. Largely as a result of a high proportion of non-performing loans, and resulting fmanciallosses, the net worth ofPFls' has become negative, and their viability has been threatened by a significant capital shortfall. Some of the PFls have been officially reporting profits for many years but their true performance, taking loan loss provisioning requirements into account, was a significant cumulative loss ­ estimated at taka 66 billion (2.2 per cent ofGDP) during 1991-2003. Implications for Service Provisions The cost to the economy on account ofinadequate and unsatisfactory quality of public provision in energy, telecommunication, railway, ports and other public utilities and banking, in terms of increased cost ofdoing business and economic opportunities and consumer welfare lost due to limited access and inefficient services has been high. Power outages and voltage fluctuations, shortage ofgas supply, limited and inefficient telephone services, inadequate urban water supply, and the high transaction costs ofaccessing these services (e.g. cost of obtaining new connections) and using them (costs of billing anomalies and paying) have imposed considerable burden on entrepreneurs and households alike. The cost of inefficient cargo operations at the seaports has been particularly high, thus affecting the external competitiveness of the economy. The cost ofsystem loss and poor revenue collection and high arrears for power, gas, urban water supply and telephones has been substantial. These financial losses have affected the viability of the entities and precluded a faster expansion of coverage and improvement of services. The inefficiency costs of parastatals' operations have been passed on to households and entrepreneurs - as effectively higher user charges due to high transaction costs, low quality ofutility, infrastructure and banking services, and foregone access to services by the large majority ofthe poor - and to taxpayers through rising budgetary liability on account of financing SOE deficits. 224 Transforming Bangladesh into a Middle Income Economy Equity Implications Public provision of services through SOEs has contributed little to equity objectives because ofthe limited access by the poor to much ofthese services and associated transfers. Because of the low coverage of most services, the substantial subsidies associated with them have not been equitable. The coverage of power, piped gas and water supply, and telephones has been largely limited to the urban non-poor. The large majority of the poor have remained outside their coverage, and so associated subsidies have not benefitted them. Access to power is limited to around 31 per cent of the population, with 80 per cent in urban areas and just 19 per cent in rural areas. Access to electricity by the poor has been very limited - only 12 per cent of the bottom fifth of the income distribution and 21 per cent of the second­ lowest quintile while nearly all households in the top two quintiles are connected to the public distribution network. Access to piped natural gas for domestic use has been limited to 4 per cent of households, mostly living in urban areas. Subsidy on household gas consumption has been significant, with gas tariffs being about a quarter of the cost of alternative fuels, mainly kerosene and LPG, but these subsidies have been benefitting a relatively small minority, mostly the non-poor urban residents. Gas subsidies to the power sector, using almost half the gas produced, do not benefit the large majority of the poor because of their very limited access to power. The urban poor, including slum dwellers, have paid a much higher price for informal access to electricity and water, compared to the prices paid by the affluent, subsidised well below cost of provision and benefits. Urban residents have paid a much lower price for power than some categories of rural consumers served by Palli Bidyut Samities. For many years, until 2001, urban consumers enjoyed a generous initial block ofcheap electricity, not available to the rural consumers in Bangladesh or urban consumers in other countries. Entrepreneurs, including those operating small and medium enterprises, have effectively cross subsidised other consumers of power and gas, including the urban non-poor, rather than consumption being subsidized through the budget. The burden of cost recovery has been shifted non-transparently and inequitably, raising the taxed groups' cost ofdoing business, and encouraging malpractice and revenue leakage. Consumption of kerosene, diesel, sugar and telephone services (until mid-FY 02), and supply of sugarcane by farmers have been taxed considerably. Sugarcane farmers and gur consumers, largely the rural poor, and sugar consumers households and food-processing enterprises alike - have been taxed in order to keep the public sector sugar mills running. The very low access to fixed telephones, together with high charges until FY 02, has denied the public, particularly the poor, the right to communicate cost effectively. The annual gross losses ofthe SOEs peaked at 38 per cent of annual public spending on education and health in FY 01 and Reforming State Owned Enterprises 225 declined to 11 per cent in FY 03 but rose to 19 per cent in FY 04. Gross SOE losses have been on average equal to around 27 per cent of annual public spending on education and health since FY 91 (Figure 7.5). In addition, maintaining jobs in unprofitable activities has been costly. The net loss per employee in nine major SOEs averaged taka 53,200 per employee per annum, or 91 per cent of their average annual wage per employee during FY 9l-OV The high cost of maintaining employment suggests that the employment argument for continuing many loss-making activities is weak. In the context ofthe country's massive underemployment and poverty, and backlog of social underdevelopment, the direct and indirect budgetary financing of SO Es' deficit has meant a high opportunity cost in terms of foregone priority investment in the social and infrastructure sectors that could have been financed with these resources to achieve a significantly greater employment and poverty reduction impact. 150 j 100 $ r:: ~ a. 50 o _ Social Spending (billion taka) - - t - Ratio of Gross SOE Losses to Social Spending (%J Figure 7.5 Comparison of Gross SOE Losses and Budgetary spending on Social Sectors Fiscal Implications of SOE Operations Budgetary financing ofSOEs' deficits has contributed much to the government budget deficit, necessitating higher domestic borrowing. There was a noticeable deterioration in terms of a rise in domestic deficit financing to high levels during FY 2000-0 I, linked partly to a rise in financing of SOEs deficit. The SOEs' gross deficit or financing gap widened to 2.8 per cent of GDP during FY 2000-02 This contributed to some extent to the rise in government's domestic borrowing, largely to finance parastatals' investment under the ADP, averaging 2.4 per cent of GDP. In addition, direct financing ofloss-making entities, through subsidies or working capital loans, and indirect 3 The entities are BJMC, BIMC, BSEC, BSFIC, BCIC, BPDB, DESA, BOGMC and BPC. 226 Transforming Bangladesh into a Middle Income Economy financing, in tenns ofshortfalls in recovery ofDSL, dividends and taxes, have contributed to the rising domestic borrowing. The heavy reliance on the exchequer for deficit financing on account of the poor perfonnance of parastatal entities is a fiscal issue, particularly for energy entities, while this has also been a banking sector issue for manufacturing entities. A welcome development is the recent decline in the gross SOE deficit, coming down to 1.8 per cent of GDP in FY 04. Budgetary resource transfer on account of financing SOEs' deficit has been taking place systematically and on a significant scale for many years and a welcome development is that actions were initiated in FY 02 to rationalize funding and tighten the budget constraint for SOEs. Direct budgetary subsidy to SOEs has been limited to around taka 4 billion annually in recent years. The budgetary cost of supporting safety nets in tenns of manpower separation costs of SOEs since the early nineties has been substantial, particularly on account ofsizable manpower rationalization during FY 03-04. Budgetary financing of SOEs' investment under the ADP has averaged taka 33 billion annually while investment ofdepartmental enterprises averaged taka 7 billion per annum. Total budgetary resource transfers, including hidden subsidies, arising from default on DSL payments owed to government, for instance, have been much larger than direct budgetary transfers. The budgetary resource transfer on loans and equity financing provided to SOEs on account ofinterest subsidy and poorrecovery of interest on loans and dividend return on equity has been significant estimated at around 2.8 per cent ofGDP annually during FY1997-2000 (Figure 7.6). Actual resource transfers to SOEs has, however, been higher due to explicit and hidden subsidies, grants, default on loan amortization payments, conversion ofloans to equity and grant, erosion ofequity, and recapitalization ofbanks against classified loans owed by SOEs. The total stock ofDSL owed by SOEs and PFls to the government has been recently estimated at taka 482 billion (18 per cent ofGDP). This revision implies a higher level ofbudgetary resource transfer, by up to 0.8 per cent of GDP annually, than the above 0 0 -10 0.5 -1 ~ '6' S -1.5 9l c ,g -2 2 e -50 ~ -2.5 .§! -3 ,... (XI 0) 0 -70 0) 0) 0) 0 -3.5 it it it it c:::::J Net Transfer on equity _ Net Transfer on loans --+- Total Transfer/GOP ("!oj Figure 7.6 Implicit Resource Transfer on Government Lending and Equity to SOEs Reforming State Owned Enterprises 227 estimate. Most ofthis is reportedly owed by BPDB, DESA, BCIC, BOGMC, REB and BSEC. The limited direct subsidy to SOEs, reported in the budget, therefore, represents the tip ofthe iceberg. The liabilities and capital shortfall ofparastatals arising from their substantial losses represent a major source of contingent liability for the exchequer. The budgetary constraint on SOEs has been tightened in recent years. Manufacturing enterprises, for instance, are not being provided budgetary financing for meeting their working capital loans (e.g. raw jute purchase by BJMC) and now rely entirely on commercial bank credit for this purpose, without sovereign guarantee by the government. Manpower separation is the only recurrent purpose for which budgetary financing is being provided because this is crucial for restructuring or privatising the entities, which are financially unable to meet these liabilities. SECTION III: PERFORMANCE AND STRATEGY IN PUBLIC MANUFACTURING The shortfall in returns as a result ofmanufacturing SOEs' operations averaged 1.6 per cent ofGDP annually during FY1991-2004, leading to a significant resource transfer (Figure 7.7). Substantial downsizing and closure ofunviable mills and manpower separation, particularly in FY 03 and to a lesser extent in FY 04, contributed to a reduction in manufacturing sector losses in the last two years. Within manufacturing, the shortfall in return on assets has been highest for BCIC, followed by BJMC. The unsatisfactory performance of manufacturing entities is linked to a variety of factors - inefficient capacity, excess overhead costs, overmanning, increased external competition, weak external demand, shortages in input supply, distortions in input and output pricing, and rising debt burden of accumulated losses and liabilities. BJMC's poor operational performance and weak external competitiveness continued until FY 02 because oflimited actions to address the constraints identified in the early nineties. The closure ofthe Adamjee Jute Mill and other inefficient and high-cost mills has helped to lower BJMC's losses. Because of efficient adjustments within the jute manufacturing sector, production and exports have not been affected by the sizable downsizing ofBJMC's inefficient capacity. 1.6 1.4 Ii:: 1.1 0 BCIC (!) 1 15 (0.78) E 0.8 CD t) 0.6 li; S, 0.4 BSRC 0.2 (0.07) 0.0 Figure 7.7 Shortfall in Annual Returns for Manufacturing SOEs: FY 1991-2004 228 Transforming Bangladesh into a Middle Income Economy Further restructuring and divestiture should lead to efficiency gains through competitive private participation and help improve viability. The viability of the public textiles and steel'and engineering sectors have been poor and further mill closure and manpower rationalization would facilitate their divestiture. The poor viability of the public sugar mills has imposed high inefficiency costs on other stakeholders - particularly sugarcane farmers, and gur and sugar consumers. Experience in Bangladesh and other countries suggests that the public sector is ill-suited to run sugar mills profitably, and divestiture seems to be a prudent option worth pursuing. BCIC's financial performance has not been satisfactory because of industrial weaknesses and domestic policies. Urea has been under-priced considerably, by 45 per cent below import/export parity price on average, since the early nineties but the fertiliser plants have not been compensated, unlike in India, for instance. It is imperative that any investment decision relating to expansion ofurea production capacity takes into account all relevant factors, including developments in the global fertiliser market, options for FDI and cost effective imports, and the economics of gas use for urea production, vis-a-vis, other uses. Economic pricing of natural gas and urea, and transparent budgetary accounting of subsidies, if warranted, and compensation to fertiliser producing entities, would facilitate efficient resource allocation decisions in fertiliser manufacturing and fertiliser application in agriculture. The rationale for active public sector involvement in manufacturing in the last decade appears questionable and costly. These are commercial activities that could have been carried out by the private sector under appropriate sector strategy and incentive regimes. Public involvement and associated sector policies discouraged development ofvarious manufacturing sub-sectors. In addition, deficit financing of manufacturing entities through budgetary transfers and bank borrowings has been costly. Consequently, there has been little or no financial rationale for the government's involvement in carrying out most of these commercial activities in the public sector and assuming the associated risks. Divestiture of manufacturing enterprises is desirable to move government out of these business activities, that are best left to the private sector. Closure of unviable mills and manpower rationalization initiated since FY 02 has facilitated this process. Liberalization of sector policy regimes, including efficient input and output pricing, would promote competitive private participation in manufacturing. SECTION IV: PERFORMANCE AND STRATEGY IN PUBLIC ENERGY PROVISION Public provision of energy has been a significant source of resource transfer. The shortfall in returns as a result of inefficient public provision of energy is estimated conservatively to be equal to 1.9 per cent of GDP annually during Reforming State Owned Enterprises 229 FY1991-2004 (Figure 7.8). This shortfall has been highest for BPDB, followed by DESA, BOGMC and BPC. The inefficiency ofpublic provision ofenergy has been linked to, among other factors, low efficiency of generation and refining, high system loss, inadequate tariffs, and poor billing and revenue collection. Power sector performance has improved to some extent and losses have declined through improvements in operational performance -lowering of system losses and account receivables for public sector consumers - and tariff measures. Unbundling and corporatisation in power distribution and selective use of private management have been pursued but there has been little progress in privatisation. Performance in petroleum continues to be unsatisfactory. A surge in world oil prices and inadequate price adjustment have impacted adversely on BPC's finances. Price adjustments in line with BPDB 1.2 (1.14) c::- Cl 0.8 " '0 C 0.6 G) <> ! 0.4 0.2 0+--­ Figure 7.8 Shortfall in Annual Returns for Energy SOEs: FY 1991-2004 changes in cost of supply and restructuring are key to performance recovery for the petroleum sector. Comprehensive reforms are needed to improve efficiency of provision of commercial energy, which is vital for driving the wheels of growth and reducing poverty. This is also critical for improving the financial viability of the public energy entities, most of which have been in an acute financial state. Natural gas has remained significantly under-exploited and underpriced, considerably below long-run marginal cost (LRMC) and netback value of gas, for various categories of consumer. This has led to significant resource transfers to the power and fertiliser sectors and urban households. These resources could instead have been mobilised through realistic pricing to support investment within the sector as well as worthwhile development activities in the rest of the economy. The challenge for Bangladesh is to exploit its gas reserves effectively, expand its gas market and optimally price gas, so as to make the most profitable use ofits only abundant mineral resource and enhance its contribution to the country's enormous development fmancing needs. 230 Transforming Bangladesh into a Middle Income Economy Inefficient power sector operations have imposed a high cost on the economy through inadequate access, poor quality and high transaction cost of services, while budgetary financing of the large deficits of energy entities has led to substantial fiscal resource transfers. The urgency of continued implementation of power sector reforms - so as to enhance cost recovery by curbing system loss and pricing power adequately, expand coverage and improve quality of services, and contain the massive resource transfer and the associated contingent liabilities and fiscal risks could hardly be overemphasised. State monopoly in petroleum refining, import and marketing and administered pricing have proved to be inefficient and costly. BPC's mounting losses in recent years have demonstrated the weaknesses of the sector strategy, and exposed the government to substantial liabilities for the exchequer. Liberalisation of the downstream petroleum market and pricing would lead to efficiency gains through competitive private participation, with favourable impact on growth as well as petroleum revenues, without exposing the government to the financial risks associated with petroleum operations. Actions will need to continue in a number ofareas: completion ofunbundling of operations in all energy sectors; instituting effective commercialisation in all entities; effective measures to curb pilferage and recover arrears; promotion of private participation in distribution through policy, legal and regulatory reforms; and implementation ofeconomic or market-based pricing ofall types ofcommercial energy, with automatic periodic adjustments, linked to changes in the cost ofsupply, as in many other countries. Following the passage ofthe Energy Regulatory Commission Act, the Energy Regulatory Commission is in the process of being made functional, with the appointment of key functionaries. The challenge now will be to operationalise the energy regulatory entity with adequate capacity and independence, so as to enable it to play an effective role in promoting development of the energy sectors, including protecting consumers' interests and safeguarding quality ofservices. The effective implementation ofthe energy pricing framework for electricity, gas and petroleum products that was adopted last year would help depoliticise the process of tariff adjustments and improve the poor financial viability of the entities and relieve government from the associated contingent liabilities. SECTION V: PUBLIC SECTOR PERFORMANCE AND STRATEGY IN TELECOMMUNICATIONS AND RAILWAY Bangladesh's telecommunications sector has lagged considerably behind that of most other countries, and the sector strategy has been inward looking, constraining economic activities and overall growth and lowering consumer welfare. Bangladesh has one of the lowest tele-densities in the world. The unsatisfactory outcome has been manifested by restricted access to telephone connections, uncompetitive pricing, up to mid-FY 02, and poor quality of Reforming State Owned Enterprises 231 services, linked to BTTB's public monopoly in fixed telephones, and inadequate competition and private investment. The sectoral inefficiencies and underdevelopment have imposed a very high cost on the economy, by limiting economic opportunities, raising the cost of doing business, and constraining growth and reducing consumer welfare. Inadequate investment has been linked to restrictive sector policies, absence of a regulatory regime, until FY 02, and severe interconnection constraints for mobile phones. The urgency ofliberalising the telecommunications sector and effective functioning of the regulatory body, with adequate operational autonomy, could hardly be overemphasised. This is vital to attract adequate private investment, particularly in fixed lines, that is crucially needed to improve the very low telephone access and quality of services. While the coverage ofthe fixed line network operated by BTTB has grown very slowly, the mobile phone network has grown noticeably, partly reflecting the liberalisation initiated in recent years. A welcome development in recent years has been the reform initiatives in the telecommunication sector designed to enhance competition and encourage new investment and better improve performance. Initiatives are underway to restructure and corporatise BTTB. This could lead to possible partial privatisation ofthe entity - ideally in strategic partnership with a foreign investor, as implemented in Sri Lanka and planned in Pakistan, for instance­ in order to pave the way for the much-needed development of the telecommunications sector. Following the establishment of the Bangladesh Telecommunications Regulatory Commission (BTRC), activities have been undertaken to enhance its capacity to function effectively within a reasonable time period. Other initiatives include: liberalisation of the licensing regime, including licensing of new private sector PSTN operators and imminent licensing ofVOIP operators; approval and establishment ofan interconnection framework; and reduction of long distance and international access charges. Bangladesh Railway's losses have continued to impose a high fiscal cost (Figure 7.9), and its potential to contribute to the economy through efficient transportation remains largely under-exploited. Its poor performance is attributable to a variety of factors - including overstaffing and skill mismatch; public service obligations (PSO), like operation on unviable routes and inadequate passenger fares; pilferage of revenue and other assets; and generally weak governance and unresponsive management. These problems are linked partly to the absence of a corporate structure, which has limited its operational autonomy and commercial orientation. Reforms ofBR are needed to improve the efficiency of railway services, and contain its highly unfavourable budgetary impact. Key elements ofthe needed reform are BR's restructuring, corporatisation, commercialisation, and promotion of private participation in various areas ofrailway investment and operations, as in many other countries. There has, however, been an apparent lack of political will 232 Transforming Bangladesh into a Middle Income Economy for many years to undertake reforms needed to transform BR into an efficient commercial entity. Such a transformation ofBR will take more time. 4 In the meantime, efforts are underway to achieve modest improvements in BR's autonomy in day to day operational matters by amending the Railway Act (1996), while the Ministry of Communications will continue to retain control on substantive policy issues. One major constraint has been that railway fares for containers and other freight as well as passengers have remained unadjusted for long periods, despite rising operational costs, further compounded by ticketless travelling by passengers. Adequate compensation for PSO has not been provided in the budget to BR, partly on account ofaccounting weaknesses, and establishment of improved commercial accounts is, therefore a high priority. It has been recently decided that BR will set up a Freight Business Unit. SECTION VI: PERFORMANCE AND STRATEGY IN PUBLIC BANKING Performance in public sector banking has generally been quite unsatisfactory in the last decade (Figure 7.10), and governance of the PFIs has been considerably unresponsive to their unfavourable performance outcome. The true financial performance outcome of the PFIs, adjusted for shortfalls in loan loss provisioning, has been significant cumulative losses, amounting to taka 65 billion (2.2 per cent ofGDP) during 1991-2003. The NCBs and SDBs continue to suffer from high and significantly understated capital shortfall, 20 Railway 15 10 3 3 If C» 2.5 2.5 .SS ..c:: ~ 2 2 'ia e'6 1.5 C!J C 1.5 'iii! "'x cW ~ 0.5 0.5 O~~~-r-r~~~~ 0+-r-.--.r--1---...................-......... a 1020 30 40 50 60 70 so 90100 a 10 20 30 40 50 60 70 so 90 100 Rural Area Urban Area Figure 8.4 Rural and Urban Growth Incidence Curves, 1991-92 to 2000 this analysis, total household income earned has been classified into income from daily wage work, salaried employment, agriculture (including fisheries and livestock), non-farm self-employment activities, and 'other' income (remittances, rental of property, investments, etc.). Per cent of Population Ranked By Per Capita Expenditures Comparing sources of income across the income distribution shows that households in Bangladesh pursue a broad range of activities to eam their livelihoods. Even within any single income decile, not only do earning strategies vary extensively across households, but even in the same household different members engage in various different activities over time and across seasons (Figure 8.5). The data show that 54 per cent ofhouseholds get income from more than one source, defined broadly as agricultural wages, non-farm wages, family business, and agricultural self-employment. Among households for whom agricultural or non-farm wage labour is the main source ofincome, about 40 per cent also derive some income from non-wage agriculture. Gear 100"10 9O".-b Other SO% Self-employment 70% 60"10 50"10 40% 30% 20'% 10% 0% 1 2 3 4 5 6 7 8 9 10 Figure 8.5 High Dependence of the Poor on Daily Wage Income 248 Transforming Bangladesh into a Middle Income Economy differences in the importance ofalternative income generation strategies also mark the behaviour of rich and poor. The starkest difference is in the tremendously high reliance on daily wage labour among the lower income groups and by contrast, the negligible share that such work contributes to income among the relatively better-off. For the poorest-fifth ofthe population, daily wage income and salaries account for close to half of total income. The importance of daily wage labour for the poor is evident in both rural and urban areas (Figure 8.6). About 70 per cent of the rural population earns income from agriculture. However, the share of income earned from this source is fairly constant across income deciles. While only about 46 per cent of the population residing in rural areas derive income from daily wage work, the poor in particular are highly dependent on this income source. Income from salaries and self-employment activities contributes a relatively larger share of income for the more affluent; about 20 per cent and 28 per cent respectively of the overall rural population derives some income from these two sources. In urban areas, close to half the population (48 per cent) rely upon salaried employment as a source of income, while about one­ quarter (27 per cent) earn income from daily wage work. However, as in rural areas, labour income from daily wage work is a particularly important source for the poor. About two-fifths of the population (42 per cent) earn income from various self-employment activities, though these tend to be relatively more important income sources for the non-poor. Rural Areas 2 3 4 5 6 7 8 9 10 Urban Areas 100"10 90% 80% 70% 6O"k 50"10 40% 30"10 Nk 10% 0% 1 2 3 4 5 6 7 8 9 10 Figure 8.6 Sectoral Differences in Sources of Income Main Income Sources and Assets of the Poor 249 Agriculture provides a significant share ofincome for the poorer groups, so clearly policies and actions that help raise income from agriculture are likely to play an important role in reducing income inequality. However, a noteworthy feature in the income profile is the importance of the non-farm sector, i.e. all economic activities except agriculture, livestock, fishing, and hunting - as a source of income for households, including in rural areas and for the poorest income groups. Even for the poorest 10 per cent of the population in rural areas, about 40 per cent of income is derived from non­ agricultural sources, either as wages earned in the non-farm sector, family business income, or remittances and other income (Figure 8.7). 10Cfk~""""""""""""""""" 90% Other 80% r-.. . . .- ---.. . . . - --.--__ Non-ag Self-employment 10% 60% Non-agWage 500/. 40% 30% 10% 0% Figure 8.7 Agricultural versus Non-Farm Sources of Income, Rural Areas Profile of Household Asset Portfolios Income differences between the rich and poor in turn reflect severe disparities in asset endowments across households. Regardless ofthe type of physical or financial asset, poorer households invariably tend to own less of it than the more affluent. 2000 HIES data indicate that the distribution ofthe taka value of various personal, financial and productive assets has an overall Gini of 0.7, far higher than either per capita incomes (Gini 0.4) or per capita expenditures (Gini 0.3). Regardless of the type of physical or financial asset, poorer households invariably tend to own less of it than the more affluent. A profile of household assets by per capita expenditure deciles reveals that poor households in Bangladesh do not look much different from the poor in other parts of South Asia. They tend to have low levels of education and limited access to land and to be highly concentrated in low-paying and physically demanding occupations. In both urban and rural areas, the poor have less access to modern amenities and services and also commonly reside in houses of inferior quality. A brief profile ofhousehold assets by per capita expenditure decile or poverty status is presented below: 250 Transforming Bangladesh into a Middle Income Economy Education The 2000 HIES data shows considerable disparities within Bangladesh in the highest educational attainment amongst the population aged 15 years and 9 8 7.7 7 5.9 6 5 4.8 4 3.3 3.7 3 2.9 2.3 2 1.5 1.8 o~~~~~~~~~~~~~~~ 4 5 6 7 8 9 10 Per Capita Expenditure Decile Figure 8.8 Average Years of Schooling older ranked by per capita expenditure deciles (Figure 8.8). On average, individuals in the top expenditure decile had seven times as many years of education as compared to individuals from the lowest decile. Examining poverty status by educational attainment of the household head shows that close to three-quarters of the poor population lives in households where the head is illiterate (Table 8.2). The incidence ofpoverty Table 8.2 Poverty Incidence by Level of Education of Household Head Highest Incidence of Percentage educational poverly of attainment of head c:: c:: :::: ~ ... It! 1: e :::J e ~ .$! :::J 0 & :;::, a:: 0 13­ a.. Not literate 64.2 64.0 64.1 57 73 Less than primary 40.6 41.5 41.3 5 4 Completed Primary 31.1 43.6 40.9 15 12 Completed middle 22.4 34.1 30.7 9 6 Completed secondary 12.5 29.0 24.0 6 3 Completed higher 2.8 13.2 8.2 7 1 Level Overall 36.6 53.0 49.8 100 100 Source: 2000 HIES. Main Income Sources and Assets of the Poor 251 declines as household heads are progressively more educated, such that the incidence ofpoverty falls by halfin urban households whose heads completed primary school and by almost one-third in similar rural households as compared to those headed by illiterates. Access to land As with educational attainment, the 2000 HIES shows distribution ofland in Bangladesh to be highly unequal. Close to two-thirds of the popUlation in the bottom quintile in rural areas owned less than 0.05 acres ofland (Table 8.3). The incidence oflandlessness falls steadily with income level, while average landholdings increases; the top-fifth ofthe popUlation was almost 10 times as Table 8.3 Land Ownership and Poverty: Rural Bangladesh Per % Population by amount of capita land owned (in acres) expenditure quintile < 0.05- 0.50- 1.50- 2.5+ Total 0.05 0.49 1.49 2.49 acre 1 64.9 14.7 13.6 3.7 3.1 100 2 55.0 13.5 17.5 8.3 5.7 100 3 43.7 14.2 19.5 10.4 12.2 100 4 35.2 12.0 20.3 12.3 20.2 100 5 32.3 8.5 16.7 13.6 28.9 100 Overall 48.0 13.0 17.5 9.2 12.4 100 Source: World Bank staff estimates from 2000 HIES, rural sample. likely to own 2.5 acres or more as compared to the bottom-fifth. Apart from obvious gains due to larger area under cultivation, larger landholdings also allow greater scope for diversification into higher value-added crops, as well as permit greater economies of scale with regard to investments in irrigation and other technological innovations. Housing conditions and infrastructure Housing conditions and infrastructure for the poor in Bangladesh are also generally worse than for the rest ofthe population and in rural as compared to urban areas (Table 8.4). While 7 per cent of overall population has access to tapped water supplies in Bangladesh, only 2 per cent of the poor had access to this drinking water source. The vast majority of the rural population relies on tubewells for drinking water supply. Only 38 per cent of the poor lived in dwellings with proper toilet facilities compared to 71 per cent of the non­ poor. Poorer access to drinking water supply and sanitation in turn make it more likely that the poor suffer from worse health than the non-poor. Access 252 Transforming Bangladesh into a Middle Income Economy to services and amenities such as electricity and phones is much lower among the poor compared to the non-poor, as well as among rural residents compared to those in urban areas. Not only do the poor own less, they also have substantially different asset portfolios from richer households. In particular, poorer households hold a greater share of their asset endowments in relatively more liquid form such as livestock and financial assets (Figure 8.9).2 The poor are much less likely Table 8.4 Housing Conditions by Sector and Poverty Amongst the population Amongst the: Urban Rural Overall Non- Poor poor Average 2.4 2.3 2.4 2.8 1.9 number of rooms in dwelling Average size 419 364 375 462 287 of dwelling (sq. feet): Population 32 0.3 7 11 2 with supply water for drinking (%) Population in 93 78 81 90 72 dwelling with hard roof. ** (%) Population 83 47 55 71 38 with proper toilet facilities:*** (%) Population 81 20 33 48 17 living in electrified dwelling (%) Population 8 0.3 2 3 < 0.1 living in dwelling with telephone: (%) Notes: * Excluding bottom and top one per cent of the population. **i.e. cement, corrugated iron sheets, wood, tiles etc. *** i.e. excluding those using temporary latrines or open fields. Source: 2000 HIES. 2 The figure excludes owner occupied housing and agricultural land which in terms of total value are by far the most important assets for households across the wealth distribution. Main Income Sources and Assets of the Poor 253 Urban Areas 100% Land or Property 90"/. 80% 7fY'1. 60% Family Business Assets 50"10 40% 30'% 20% 10% 0% 1 2 3 4 5 6 7 8 9 10 Wealth Decile Rural Areas 100% 9O"k 80% 70% 6f1'/o 50% 40% 3O"k 20% 10% fY'I. 2 3 4 5 6 7 8 9 10 Wealth Decile Figure 8.9 Importance of Livestock & Financial Assets for the Poor to own business assets. These differences in portfolios reflect not only differences in access to investment opportunities (e.g. due to credit constraints), but also the compulsion on poor households to seek safer but lower-returns, to refrain from profitable but non-divisible investments, and to rely on costly strategies for smoothing consumption (e.g. through distress sales of liquid productive assets). The varied range ofinc orne generation strategies and asset holdings has several implications for designing an effective poverty reduction strategy. First, it emphasises that no silver bullet will cut poverty. Rather, development interventions need to take different contexts and classes of households into account at the same time that they bolster an environment that opens opportunities to the poor and lets them tum their own specific endowments and circumstances to advantage. Secondly, it is precisely because of this heterogeneity in the context of asymmetric information between households and the government that decentralisation and participation - which enable people to reveal their demands for intervention and hence shape the 254 Transforming Bangladesh into a Middle Income Economy development process - are important. And finally, heterogeneity illustrates why growth in itself may not assure poverty reduction. Growth will always be a precondition for the elimination of poverty, but policymakers need to also take into account the nature of growth and the variegated potential of households and individuals to participate in and benefit from an expanding economy. SECTION IV: ANALYSING THE CORRELATES OF POVERTY In order to better understand the role ofkey private and public assets that are essential for poverty reduction, we use a regression model to explore how various household and community characteristics are correlated with per capita expenditures, separately for rural and urban areas. The regressions include geographical identifiers trat control for differences across communities, leaving the household variables to account for differences in expenditure levels within communities. At the outset, it is important to note that although theory predicts that many of the variables included in the analysis do indeed contribute to poverty reduction, the statistical relationships should be interpreted as correlates and not as determinants since causality can run both ways for some variables. In addition, the analysis is limited by the variables available at the household and community level in the 2000 HIES. Education The higher the educational attainment of the household head, the less likely the household is to be poor. For instance, in rural areas, expenditures of people living in households in which the head had completed primary education were 10 per cent higher than those who lived in otherwise similar households whose head was uneducated (Figure 8.10). However, given the non-linearity in returns to education, the poor derived lower returns from education. As the figure shows, secondary and tertiary education had an even stronger association than primary education with expenditure levels. A rural household whose head had completed secondary education had 31 per cent higher per capita expenditure than a household with an uneducated head. Also note that gains from education were larger in urban than in rural areas, possibly reflecting the differences in income generation opportunities for skilled workers across the two sectors. Land Comparing expenditure levels across different land ownership categories, after controlling for other household characteristics, we find a strong association between land ownership and poverty. Compared to a landless rural household, a household with less than half an acre had 7 per cent higher Main Income Sources and Assets of the Poor 255 Rural o Higher level • Class 6-9 • Up to Class 5 50 60 Percentage Change in Income Figure 8.10 Gains from Education: 2000 HIES. expenditure, and a household with more than 2.5 acres had 44 per cent higher per capita expenditures (Figure 8.11). Expenditure gains in urban areas also rise as the size of the landholding changes, but as one would expect, the rate of increase across different landowning categories is not as sharp. Above 1.5 acres, households have 27 per cent higher expenditure than the landless. Family Business Assets Compared to rural households with family business assets less than taka 5,000 (or no family business), households with a business worth taka 5 to 100,000 had 17 per cent higher per capita expenditures, while those with a business worth taka 100,000 or more had 38 per cent higher expenditure levels. Returns to business assets were found to be of similar magnitude in urban areas as well. Overall, self-employed households in Bangladesh that had business assets worth taka one hundred thousand or more had per capita expenditures almost Rural c; 2.50 a;es or more o 1.50-2.49 acres r• 0.50-1.49 acres : II < 0.50 acres o 10 20 30 40 50 Percentage Change in Income Figure 8.11 Gains from Land Ownership: HIES 2000 256 Transforming Bangladesh into a Middle Income Economy -10 o 10 20 30 Percentage Change in Income Figure 8.12 Characteristics of Household Head three times as high as self-employed households with business assets less than taka five thousand. Disparities in business asset ownership between the poor and non-poor are perpetuated across generations as a large share ofthese business assets tend to inheritit, rather than purchase with credit or from the household's own savings. Marital Status In both urban and rural areas, per capita expenditure levels in households headed by females who were currently married were significantly higher (between 18 and 20 per cent) than in household headed by males (Figure 8.12). This could potentially be due to receipt ofremittances from the migrant spouse in such households. However, not all female headed households were better off. In rural areas, per capita expenditure levels in female headed households, where the woman was either widowed, divoreed or separated from her spouse were about II per cent lower than levels in a household headed by males. Such households in urban areas, on an average, did not have significantly higher expenditure levels than male headed households. Distressed female-headship, therefore, appears to be a useful proxy for targeting poverty interventions in rural areas. Geographic Effects Geographic identifiers in the per capita expenditure regressions attract significant coefficients, even after controlling for household characteristics. Differentials in per capita expenditure due to geographic effects are considerable, suggesting that the same, observationally equivalent, household is poor in one place but not another. High payoffs could potentially be reaped through investments designed to improve area characteristics. Main Income Sources and Assets of the Poor 257 For the rural sample, as we have information on various community­ level characteristics, we can also examine what aspects ofarea characteristics give arise to geographic effects. Table 8.5 presents the percentage increase in per capita expenditure associated with the community characteristics, after controlling for household characteristics. The strongest correlates of expenditure levels and community characteristics are indicators of infrastructure. Even after controlling for household characteristics, individuals living in communities that have electricity and phones had between 6 to 12 per cent higher expenditure than those living in villages without such services. The degree of isolation, as Table 8.5 Per Capita Expenditures and Community Characteristics Per cent change Variables in per capita expenditures Infrastructure and access to markets Electricity 5.7 Phone 12,4 Distance to bus stop (km) -0.5 Distance to Dhaka (km) Ns Access to natural assets Beel 3.3 Forest Ns Khas 2.3 Grazing Ns Diversification Main economic activity females: NAG 3.8 Main economic activity: NAG Ns One crop land most predominant 5.2 Presence of banks, NGOs and cooperatives Krishi bank -14.8 Grameen Ns BRAC -3.9 Proshika -5,4 Farmers cooperative Ns B.S.S. 3.9 Note: The estimates measures the precentage increase in per capita expenditure associated with each variable, controlling for household characteristics. Ns means not significant at 10 per cent. Source: World Bank Staff Estimates from 2000 HIES Rural Sample. 258 Transforming Bangladesh into a Middle Income Economy measured by the distance to the nearest bus stop, accounted for a significant but small part of the differences in expenditure levels across communities. Access to natural assets such as beels (ponds) and khas land was found to have a positive impact on expenditure levels. Such resources can be an important source oflivelihood for the poor as has been extensively highlighted in the literature on poverty in Bangladesh. For example, the Consultations with the Poor study (Nabi et al. 2002) reports that in two ofthe villages covered, access to khas land was highlighted by respondents as an important means of moving out of poverty. Similarly, the extent of diversification into non-farm activities also appeared to matter: people living in villages where the main activity offemales was outside agriculture had 3.5 per cent higher expenditures than those living in villages where agriculture was the predominant income generation activity. Comparing the importance of the non-farm sector as a source of livelihood across communities shows that availability of basic infrastructure and non­ farm sector work are highly correlated in Bangladesh. For instance, localities where commercial electricity connections were available were more than four times as likely as others to report small cottage industries as one of the main economic activities. Likewise, the transport and services sector were also more likely to be reported amongst the main economic activities in communities with relatively better infrastructure. The strong effects of these community characteristics on incomes indicate that investments designed to improve infrastructure and foster non-farm growth should play an important role in poverty reduction. External Institutions and Indigenous Cooperatives Service provision by agricultural banks and NGOs appears to be well-targeted to poor areas communities in which services were provided by the major NGOs and the agricultural banks tended to have lower per capita expenditure levels. By contrast, indigenous cooperatives for the poor, such as the B.S.S. (Asset-less Cooperative Association) were associated with higher expenditure levels, suggesting that such organisations that help the poor may be less likely to be formed in the poorest areas. SECTION V: SUMMARY AND CONCLUSIONS This paper reviews the evidence on poverty in Bangladesh during the 1990s and the main sources of income and assets of the poor based on the 2000 HIES data. The main conclusions of the paper are: • Households in Bangladesh pursue a broad range of activities to earn their livelihood. Thus, some 54 per cent ofhouseholds get income from Main Income Sources and Assets of the Poor 259 more than one source, broadly defined as agricultural wages, non-farm wages, family business and agricultural self-employment. • Among households for whom agricultural or non-farm wage labour is the main source of income, about 40 per cent also derive some income from non-wage agriculture. • Lower income groups exhibit a very high reliance on daily wage incomes. Thus, for the poorest-fifth of the population, daily wage incomes and salaries accounts for close to half of the total income. • About 70 per cent ofthe rural population eams income from agriculture. • A noteworthy feature in the income profile is the importance of non­ farm sector. Even for the poorest 10 per cent of the population in rural areas, about 40 per cent of income is derived from non-agricultural sources, either as wages earned in the non-farm sector, family business income or remittances and other income. • Income differences between the rich and poor mainly reflect severe disparities in asset endowments across households. Regardless of the type of physical or financial asset, poorer households invariably tend to own less of it than the more affluent. • The 2000 HIES data show considerable disparities within Bangladesh in the highest educational attainment amongst the population aged 15 years and older ranked by per capita expenditure deciles. On average, individuals in the top expenditure decile had seven times as many years of education as compared to individuals from the lowest decile. • As with educational attainment, the 2000 HIES shows distribution of land in Bangladesh to be highly unequal. Close to two-thirds of the population in the bottom quintile in rural areas owned less than 0.05 acres ofland. The incidence oflandlessness falls steadily with income level - while average landholdings increases, the top-fifth of the population was almost 10 times as likely to own 2.5 acres or more as compared to the bottom-fifth. • Housing conditions and infrastructure for the poor in Bangladesh are also generally worse than for the rest of the population and in rural as compared to urban areas. While 7 per cent of overall population has access to tapped water supplies in Bangladesh, only 2 per cent of the poor had access to this drinking water source. Only 38 per cent of the poor lived in dwellings with proper toilet facilities compared to 71 per cent ofthe non-poor. Access to services and amenities such as electricity and phones is much lower among the poor compared to the non-poor, as well as among rural residents compared to those in urban areas. • Putting together these variables in a regression model shows the level of education, land ownership, level of family assets and availability of infrastructure (roads, electricity, phones) are all correlated with poverty reduction. 260 Transforming Bangladesh into a Middle Income Economy Given the above, policies and programmes that create employment, improve education for the poor, improve their land holdings and other assets, improve agriculture, and improve infrastructure will all have a positive impact on poverty reduction. REFERENCES Government of Bangladesh. Household Income and Expenditure Survey, 2000, Bangladesh Bureau of Statistics, Dhaka, 2001. Nabi, Datta, and Chakrabarty. 'Bangladesh Wavers of Disaster' in Narayan and Petesch (eds.) Voices of the Poor: From Many Lands, New York: World Bank and Oxford University Press, 2002. World Bank. Poverty in Bangladesh, Building on Progress, Report Number 24299-BD, Washington, DC, 2002. 9 Agriculture Development Strategy Ndiame Diop SECTION I: INTRODUCTION Following years of slow growth, Bangladesh's agriculture performed well in recent years, with real agriCUlture GDP increasing by an average of 5.1 per cent per year from 1996 to 2001. This growth reflects relatively high growth rates in crops and horticulture (4.6 per cent), forest activities (4.6 per cent) and fisheries (8 per cent) and is largely attributable to improved access to inputs, facilitated by domestic market liberalisation and increased factor productivity. The analysis of sources of income for the poor in the previous paper showed that agriculture is a major source of income for the rural poor. It is not surprising, therefore, that rural poverty declined more rapidly in the 1990s than during the mid-1980s. Thus, the rural poverty rate declined by 8 percentage points over the 1990s, from 61.2 per cent in 1992 to 53 per cent in 2000 as compared with stagnation between 1986 and 92 (World Bank 2002). Nevertheless, rural poverty is still pervasive as the number of rural people sti1lliving below the poverty line is 53 million. This represents 85 per cent of the total number of poor in Bangladesh. Aggregate data, however, cannot be entirely relied upon to pinpoint the specific policies and investments needed to alleviate rural poverty through agriculture growth and productivity. The incidence of rural poverty varies significantly by regions, which requires looking at this sources of disparity. This paper presents a regionally disaggregated analysis ofagricultural growth and rural poverty trends in Bangladesh and examines the linkages between these two factors. To pinpoint the specific role of agriculture in rural poverty dynamics, the paper analyses the sources of rural income (especially farm versus non-farm), the extent of agricultural growth and the distribution of income in each region. Having identified the regions where agricultural growth 262 Transforming Bangladesh into a Middle Income Economy has the greatest poverty impact, the paper then proceeds to identifY the key drivers of agricultural growth. Econometric investigation suggests that macroeconomic stability, human capital, irrigation, mechanisation and environment factors (floods) are the maj or determinants ofagricultural growth in Bangladesh. The next section examines the profile and evolution ofrural poverty in Bangladesh. The third section looks at the role of agriculture in the observed poverty trends. The fourth section analyses the determinants of agricultural growth in Bangladesh while the last section derives the policy implications of the study. SECTION II: PROFILE AND EVOLUTION OF RURAL POVERTY IN BANGLADESH Poverty in Bangladesh has declined significantly since the early 1990s, but it is still pervasive, with about half of the population living below the 'food­ based' poverty line and about one-quarter subsisting in extreme povertyl in the year 2000 (Table 9.1). The national poverty incidence declined from 58.8 to 49.8 per cent between 1992 and 2000, reflecting an 8 percentage points of both urban and rural poverty. 2 This notwithstanding, poverty remains a predominantly rural phenomenon in Bangladesh, with 53 per cent ofthe rural population living in poverty. This represents a population of 53 millions or 85 per cent of the total number of poor in Bangladesh. There are, however, wide variations in rural poverty incidence across Bangladeshi geographical regions. 3 Despite a significant reduction in the I Poor households are defined as those consuming less than 2122 cal/capita/day. Households consuming less than 1800 cal/capita/day are considered extremely poor. 2 The headcount index, PO, is the proportion of population with per capita expenditure below the poverty line. The poverty-gap, PI, estimates the average distance separating the poor from the poverty line, as a proportion of that line. Finally, the squared poverty-gap, P2, takes into account not only the distance, but also inequality among the poor. 3 To capture some ofthe regional diversities, six broad agro-ecological regions were constructed. These regions were created in accordance with several major criteria: consistency with broad agro-ecological zones, hydrological regions, division boundaries, and the need to limit the number ofregions to get a statistically representative sample from the Household Income and Expenditure Survey (HIES 2000) for the region. The six regions defined are: (i) the northwest region which corresponds with Rajshahi division; (ii) the south Ganges flood plain region which contains Barisal district, as well as parts of Dhaka and Khulna divisions areas south of the Padma (Ganges) river; (iii) the coastal region which is a complex and fragile natural ecosystem comprising the districts ofSatkhira, Khulna, Bagerhat, Perojpur, Jhalokati, Borguna, Bhola and Patuakhali. This region faces major problems of water drainage; (iv) north central region, defined as the parts ofDhaka division north ofthe Padma, a region similar to Rajshahi division, though with marketing advantages arising from its proximity to Dhaka city; (v) The eastern part of the country (Chittagong division) is split between the Meghna flood plain (the western part of Chittagorig division) and the eastern hills regions. Agriculture Development Strategy 263 1990s, the northwest area remains Bangladesh's poorest region - the rural poverty incidence fell by 11 per cent over the 1990s but still remains 18 per cent higher than the national average (Table 9.2). Similarly, rural poverty fell by 13.7 per cent in the north central, making a significant dent on the incidence of poverty but still leaving 49 per cent of the rural population in this region below the poverty line. By scrapping poverty by almost one-third in the 1990s, however, the coastal region has now the lowest rural poverty incidence in the country (37 per cent). In sharp contrast with these three regions, south Ganges, the second poorest region (15 per cent of national average), enjoyed only a 1.7 per cent reduction in the poverty incidence over the 1990s while the Meghna flood plains region saw a stagnant poverty rate. The only region where poverty increased is the eastern highlands region. The small sample sizes in the latter make, however, this trend uncertain. Table 9.1 Trends in Bangladesh's Poverty Measures 1991-2 1995-6 2000 Headcount rate (PO) National 58.8 51.0 49.8 Urban 44.9 29.4 36.6 Rural 61.2 55.2 53.0 Poverty gap (P1) National 17.2 13.3 12.9 Urban 12.0 7.2 9.5 Rural 18.1 14.5 13.8 Squared poverty gap (P2) National 6.8 4.8 4.6 Urban 4.4 2.5 3.4 Rural 7.2 5.3 4.9 Source: BBS and World Bank staff estimates. In sum, analysis of rural poverty dynamics shows that despite an 8 percentage points decline over the 1990s, rural poverty remains pervasive, with 53 per cent of the rural population leaving below the poverty line in 2000. But trends in rural poverty have been uneven across regions, leading to a contrasted regional profile of rural poverty in the beginning of the new millennium: • Rural poverty remains very high in the northwest, north central and south Ganges regions, despite a significant reduction in the first two regions; • The coastal region now enjoys the lowest rural poverty incidence in the country thanks to an impressive reduction of poverty in the 1990s; • While rural poverty stagnated in Meghna flood plains and increased in Table 9.2 Regional Trends and Profile of Rural Poverty in Bangladesh South North Meghna Eastern Coastal Northwest Gangesh central FPLN hills area FP Headcount rates, UPL'", 1991 (%) 73.7 62.4 63.0 49.3 29.4 66.5 Headcount rates, UPL, 2000 (%) 63.2 60.8 49.4 48.8 46.6 37.0 Changes in headcount rates, UPL, 1991-2000 (%) -11 -1.7 -13.7 -0.5 17.1 -29.6 Changes in poverty gap, 1991-2000 (%) -4.33 -2.8 -2.53 0.33 1.9 -4.46 Changes in the square of poverty gap, 1991-2000 (%) -5.83 -2.73 -3.45 0.98 3.6 -7.98 Source: Calculated using HES 1991 and HIES 2000. Note: '"UPL: Upper poverty line; For details regarding the calculation of the UPL, see World Bank 2002. Agriculture Development Strategy 265 the eastern hills, poverty rates in these regions remain below the national average. What has been the specific role of agriculture in the observed rural poverty trends? In principle, the impact of agriculture on poverty is strongest the larger the share of agriculture in total income, the higher the rate of agricultural growth. However, (total) income distribution effects can offset the impact of agricultural growth as the poor may fail to capture the benefits of growth. To pinpoint the role of agriculture in rural poverty dynamics, it is thus necessary to analyse the sources of rural income, the extent of agricultural growth and the distribution of income by region. SECTION III: ROLE OF AGRICULTURE IN THE OBSERVED POVERTY TRENDS Agriculture is a major source of income in rural Bangladesh. According to Household Income and Expenditure Survey data (HIES 2000), agriCUltural income, i.e. sales of crops, livestock, and fisheries, and agricultural wages, accounts for 40 per cent of average rural household income. But the prominence of agriculture as a source of income and livelihood is more patent for the poor. Households in the bottom 20 per cent of the income distribution derive 48 per cent of their income from agriculture. For households in the top 20 per cent of the income scale, agricultural income accounts for 27 per cent of total income (World Bank 2003). The share of income from the agricultural wages component declines even more dramatically across expenditure quintiles, accounting for 29 per cent of the incomes of the poorest, but only 3.5 per cent of incomes of the top quintile. The importance of agriculture in rural household incomes varies, however, considerably across regions (Table 9.3). Rural households are more dependent on agriculture for income in the northwest and south ganges regions, the two poorest regions, than in the other regions. Income from agriculture accounts for 50 per cent and 43 per cent of total income respectively in northwest and south Ganges. In contrast, rural households in the eastern hills and Meghna flood plains regions are far more dependent on non-farm activities for income. They derive respectively 40 and 38 per cent of their income from non-farm activities, against only 24 per cent from agriculture. Rural households in the coastal and north central regions also predominantly derive their income from non-farm activities even if agriculture provides a sizeable share of total income. The importance of agriculture in rural household incomes varies, however, considerably across regions (Table 9.3). Rural households are more dependent on agriculture for income in the northwest and south Ganges regions, the two poorest regions, than in the other regions. Income from Table 9.3 Sources of Rural Household Income by Region (per cent) South North Meghna Eastern Northwest Ganges Coastal central flood plain hills Agricultural income 50.51 43.27 32.04 36.84 23.93 23.91 of which: agricultural wages 22.52 17.97 14.09 15.63 10.54 11.2 of which farm income 27.99 25.3 17.95 21.21 13.39 12.71 Non-farm income 29.84 34.17 36.52 38.46 37.68 39.88 of which: non-agricultural wages 17.52 17.97 22.34 21.95 22.29 27.59 of which: enterprise income 12.32 16.2 14.18 16.51 15.39 12.29 Other income 19.65 22.55 31.44 24.7 38.4 36.21 Total 100 100 100 100 100 100 Poverty rate (headcount) 62.8 60.8 37 49.4 48.8 46.6 Note: Other income includes rental income, local and foreign remittances, imputed value of house, and other transfers. Agriculture Development Strategy 267 agriculture accounts for 50 per cent and 43 per cent of total income respectively in northwest and south Ganges. In contrast, rural households in the eastern hills and Meghna Flood Plains regions are far more dependent on non-farm activities for income. They derive respectively 40 and 38 per cent of their income from non-farm activities, against only 24 per cent from agriculture. Rural households in the coastal and north central regions also predominantly derive their income from non-farm activities even if agriculture provides a sizeable share of total income. Thus, while agriculture is of critical economic importance for the poor in Bangladesh, there are wide disparities in the dependence on agriculture for income across Bangladeshi regions. The dependence on agriculture for income is the highest in northwest and south Ganges regions. In contrast, the coastal, eastern hills, Meghna Flood Plains and north central regions are more dependent on non-farm activities for income, even though income from agriculture is significant (above 30 per cent) in the north central and coastal regions. Regional Agricultural and Total GOP Growth Are the poverty trends of the 1990s consistent with the growth dynamics of the decade? We analyse here in some detail the regional patterns in agricultural growth and total growth. Table 9.4 summarises the trends of these indicators in the 1990s. 4 All regions grew at a rate in excess of 3 per cent per year, except for the coastal region, which grew at only 1.7 per cent per year. South Ganges Flood Plain registered the highest in real agriculture growth, with an annual growth of 4.1 per cent in the period 1993-94 to 1997-98. Agricultural growth was above national average for Northwest, South Ganges Flood Plain and North Central Regions, where rural households derive a significant share of their income from agriculture. The pattern of growth rates of total GDP by region is much different, however. Indeed, limited non-agricultural growth in the northwest and south ganges flood plain regions kept total GDP growth in these regions to only 4.4 and 5.4 per cent, respectively, in spite of rapid growth in agriculture. In contrast, spurred by rapid non-agricultural growth (both urban and rural) total GDP in the coastal region grew by 5 per cent, only slightly below the national average of 5.4 per cent (Table 9.4). Similarly, the eastern hills experienced a rapid non-agricultural growth leading to the highest growth in total GDP (6 per cent). 4 Dataavailability has constrained the time frame of this analysis. Because regional total GDP growth rates were available only until 1998, we calibrated the time frame accordingly for consistent comparisons. Table 9.4 Regional Difference in Agricultural and Total GOP Growth South North Meghna Eastern Coastal Northwest Ganges central FPLN hills area Overall FP Aver. Ann. growth in boro rice: 1988-98 (%) 9.6 11.7 6.6 5 4.8 9.6 7.7 Real agriculture GOP growth: 1993-98 (%) 3.4 4.1 3.7 3.1 3.8 1.7 3.3 Total GOP growth: 1993-8 (%) 4.4 5.3 5 5.7 5.8 6 5.3 Real agri. wage ann. Growth: 1992-2001" 1.03 1.02 1.02 0.9 1.01 1.03 1.00 Note: " Wage rates are average daily real wage rate of agricultural male labour without food. Source: Bangladesh Bureau of Statistics. Agriculture Development Strategy 269 Distribution of Income Growth Across Rural Household Quintiles To what extent poor in different regions have captured the fruits ofagricultural growth in the 1990s? To respond to this question, we follow Ravallion and Chen (2001) and construct 'regional growth incidence curves' (Figure 9.1). These curves show the changes in real per capita expenditure over the 1990s for different groups ranked by level of income. 5 As Figure 9.1 shows, all the segments of the population in all regions experienced an increase in real per capita expenditure over the 1990s, the rate of change varying between 6 and 40 per cent. However, the distribution ofreal income across quintiles was uneven across regions. In northwest, north central and south Ganges regions, the curves are U-shaped, indicating that the lowest quintile as well as the highest quintile in these regions had increased their real per capita expenditure relatively more than the middle class. With an average per capita increase of 13 per cent, the poorest quintile also fared better than the middle class in the coastal region. In sharp contrast, real per capita distribution was biased in favour of the richer parts of the population in the Meghna flood plains and eastern hills regions. The crucial but insufficient role of agriculture in reducing rural poverty Because most ofthe poor are still employed in agriculture, this sector plays a central role in reducing rural poverty in Bangladesh. In the northwest, north central and south Ganges regions where sales of agricultural products and agricultural wages provide a large share of total income, robust agricultural growth has allowed a significant reduction in poverty. The fact that the lowest quintiles in these regions have increased their income relatively more than the middle class has reinforced the poverty effect of agricultural growth. The still high level ofpoverty in these regions only reflects their initial conditions: the poverty levels were extremely high and deep in the early 1990s. It is clear, however, that agricultural growth is insufficient to significantly attack poverty in some other regions of Bangladesh. The coastal region is a vivid example of this. Despite limited role of agriculture, spurred by rapid non-agricultural growth, total GDP in the coastal region grew by 5 per cent and allowed the largest rural poverty reduction among the regions. Similarly, in spite of strong agricultural growth, poverty stagnated in Meghna flood plains and increased in the eastern hills. In these two regions, where rural households also derive the bulk oftheir income from rural non-farm activities, the impact of agriculture was further limited by a skewed distribution of 5 See Ravallion and Chen (200 I ) for details regarding the mathematical derivation and properties of these curves. 270 Transforming Bangladesh into a Middle Income Economy Figure 9.1 Distribution of Household Income Growth in the 1990s: Regional 'Growth Incidence Curves' income, since the real per capita distribution was biased in favour of richer parts of the rural population. SECTION IV: BOOSTING AGRICULTURAL GROWTH IN BANGLADESH: WHAT ARE THE KEY DRIVERS? As shown by Table 9.4, Bangladesh's agricultural growth record has been impressive in the 1990s. We have also seen that agricultural growth has greatly contributed to reducing poverty in Bangladesh's poorest regions. Given the still pervasive poverty in these regions, it is critical to identify the key drivers of agricultural growth in order to carefully implement policies that have the highest potential for boosting growth and reducing poverty in rural Bangladesh. This section looks at this issue. We briefly review literature on the determinants of agricultural growth and undertake an econometric analysis of the case of Bangladesh. Determinants of Agricultural Growth The empirical literature on the factors behind agricultural growth is abundant. Early researches have identified physical capital, infrastructure, human capital, research, extension, and rural population density as the major determinants ofagricultural supply (Binswanger et aL 1987). Many recent studies ofgrowth of agricultural output find similar results, with much focus, however, on the different contribution of factor inputs and total factor productivity (TFP) or technical change. Changes in factor inputs (land, labour, capital, etc.) appear to account for only about a halfof agricultural output (Mundlak 1999). Thus, Agriculture Development Strategy 271 about 50 per cent ofchange in agricultural growth is explained by TFP growth. Hence, a great deal of attention is paid to factors determining TFP. Confirming the earlier findings of Antle (1983), many recent studies identify education, research and infrastructure as major determinants ofTFP. In a recent monograph, Evenson, Pray and Rosegrant (1999) estimate the contribution of various factors to India's TFP and found that public research and extension are the two most important factors accounting for TFP growth, with irrigation coming next. In particular, investments in agricultural research and extension accounted for nearly three-fourths ofthe growth in productivity. Similarly Fan, Hazell, and Thorat (1999) show, using Indian data, that public expenditures for research and extension have had the largest impact on agricultural productivity growth, followed by rural roads, education and irrigation. In the same vein, Fan, Zhang and Zhang (2002) found that in China the largest contribution to agricultural productivity has come from research and extension, followed by education, rural telephones, rural roads, and electricity. The relationship between factor inputs and TFP has also received much attention in recent years. Mundlak (1999) suggests that the empirical evidence points to the fact that the major way technology is incorporated into agricultural production is through physical capital. The same result is found in Mundlak, Larson and Butzer (1997). The underlying idea is that adoption of more advanced techniques (e.g. HYV seeds) takes time and resources. McGuirk and Mundlak (1991) show that the pace of adoption ofthe modern wheat and rice varieties in Punjab was largely related to the pace ofexpansion ofirrigation facilities, infrastructure and fertilisers, all of which require resources. As such the pace of adoption is related to the scarcity or availability of capital. Without accompanying infrastructure (irrigation, roads), there could be an important lag between technological innovations in agriCUlture and overall TFP improvements. Murgai's (1999) finding for Punjab agriculture during and after the green revolution illustrates this point. The various estimates in the literature are, however, hampered by the limited or inaccuracy of aggregate data for inputs such as labour, capital and infrastructure variables. Similarly, the interpretation of the contribution of several variables can be problematic, since for instance technological innovation may increase the elasticity of output to modern inputs, and TFP figures do not capture these effects (Murgai 1999). This notwithstanding, researchers like Mundlak have recently proposed innovative ways to pinpoint the drivers of agricultural growth. Drivers of Agricultural Growth in Bangladesh To assess the sources of agricultural growth in Bangladesh, we estimate a Mundlak-type Cobb-Douglas production function (Mundlak, 2000). The 272 Transforming Bangladesh into a Middle Income Economy originality ofthis model is the recognition that aggregate output is the sum of outputs produced by more than one technique, since adoption of more advanced techniques takes time and resources. The economic problem faced by producers involves then a decision on what techniques of production to employ in addition to their decision on the level of inputs. Optimisation under these circumstances, and after approx~mation, yields a Cobb-Douglas-like production function, which has as arguments the major inputs and a set of variables called 'state variables'. State variables are variables such as human capital, infrastructure, and incentives that condition an effective adoption of technologies. The technical problems related to specification of estimation of a production function under this specific context are discussed in Mundlak (1988,2000). The dependent variable is the logarithm of agricultural value added, measured in 1995 constant prices. We introduced four categories of explanatory variables, expressed in logarithmic form: • Incentive variables Real exchange rates, vis-ii-vis, India (RER) and Bangladesh's GDP deflators (inflation) are introduced to capture evolution in the incentive structure brought about by the broader macroeconomic environment. • Input variables namely, the labour force in agriculture, net irrigated area and number of tractors, used as a proxy of capital. • Human capital and infrastructure Human capital is proxied by adult illiteracy rates (education) and infant mortality rates (health). As for infrastructure, because of limited data, only production of electricity (availability of electricity nationwide) is introduced in the regression. • Environmental variables capturing the effects ofweather shocks (dummy for lower than normal years of rainfall) and natural calamities (dummy for flood and cyclone years). Table 9.5 shows the results of the econometric estimations. The major drivers of aggregate agriculture output in Bangladesh appear to be irrigation, mechanisation, education, infrastructure, inflation and occurrence offlooding. The positive and significant impact of irrigation and mechanisation is consistent with the findings of the broad literature as well as that of some recent studies on crop production growth in Bangladesh. For instance, Asaduzzaman (2000) found a strong and positive impact of mechanisation of tillage on cropping intensity and crop production while Mandai (2000) highlighted the key role of irrigation on crop expansion. The positive impact of the education is consistent with findings in the literature. Education is critical to increase productivity and agricultural value added. Indeed agricultural growth requires change, including adoption ofsophisticated new technologies and marketing practices, hence the critical role of education. Agriculture Development Strategy 273 As expected, a stable macroeconomic environment (captured here by a low inflation) contributes significantly to agricultural growth. Finally, the negative impact of flooding on agricultural growth reflects the fact that about 80 per cent of Bangladesh is a flood plain and despite progress on coping mechanisms, flooding significantly and frequently disrupts production. A surprising result is the negative (though insignificant) impact on agricultural labour on aggregate production, but this may suggest the existence of important disguised unemployment (labour surplus) in agriculture. Table 9.5 Determinants of Aggregate Agricultural Growth Variables Coefficient t-ratio Labour -0.004 -0.44 Irrigation 0.36**· 2.88 Mechanisation 0.49· 1.69 Illiteracy -0.96**· -2.74 Infant Mortality -0.20 ns -0.97 Electricity 0.22 ns 1.49 Inflation -0.26 00 • -3.01 Flood -0.12 ns -0.82 Note: The variables, in log, are expressed in first difference. •••. Significant at 1 per cent •• Significant at 5 per cent o Significant at 10 per cent These results are consistent with that of studies focused on rice. For instance, Hossain (1989) attributes the growth in rice production to the increase in the application of major inputs (HYV seeds, irrigation). According to Ahmed (2001), the availability ofHYV seeds combined with increased small­ scale private irrigation and fertiliser has been key in boosting productivity in the rice sector. In particular, these have allowed a seasonal shift of area in favour of high-yielding, irrigated and fertiliser-intensive boro rice grown during the dry (winter) season. Indeed, gradually, area under pre-monsoon rice (aus rice), susceptible to drought, and that under deepwater aman rice, susceptible to floods, have been reduced to give way to boro rice cultivation. Between 1989-90 and 1998-99, the share of boro in total rice area increased from 24 per cent to 35 per cent. Consistently, the share of boro in production increased dramatically from 35 per cent to 53 per cent in the same period. TFP in rice has reportedly increased by 1 per cent annually for the last two decades (Ahmed 2001). 274 Transforming Bangladesh into a Middle Income Economy SECTION V: POLICY IMPLICATIONS The above analysis allows us to derive several policy implications: • In attempting to enhance the income earning opportunities for the rural poor and specifically to remove the impediments to agricultural growth, one should be cognizant of regional differences in dependence on agriculture versus non-agriculture and regional differences in productivity constraints. As shown above, rural households in the northwest and south Ganges regions are highly dependent on agriculture for income. In contrast, rural households in Meghna flood plain, the eastern highlands and the coastal regions derive a significant share of their income from rural non-farm activities. Thus, a given rate of growth in agriculture is likely to have a bigger impact on income and poverty in northwest and south Ganges than elsewhere. • Given large numbers of non-farmers in rural areas, agricultural growth alone will be insufficient to rapidly reduce poverty. In fact, the largest reductions in incidence and depth of rural poverty have occurred in the coastal regions that have enjoyed substantial non-agricultural income growth and only very modest gains in agricultural incomes. Further analysis of non-agricultural income sources, constraints and growth is crucial for understanding policy options to accelerate poverty reductions in the future. • To further boost agricultural growth, continued investments in irrigation, water control, mechanisation, and human capital are key factors. As in the past, a rapid spread of green revolution technology for rice and wheat will be more favourable in the south Ganges, northwest and north central regions, where agro-ecological conditions are more favourable. The spread of green revolution technology through irrigation and mechanisation will be limited in Meghna flood plains and the coastal regions because the risk of flooding - cyclone is higher and the drainage relatively difficult. • Furthermore, investment in better education is particularly important in the agriculture-dependent regions (northwest, South ganges and to a lesser degree, the north central region) to increase productivity and boost diversification towards non-rice crops, livestock and fisheries. 6 A recent 6 To date, Bangladesh agriculture (particularly the crop sector) is characterised by weak diversity. Rice remains the dominant crop, accounting for 77 per cent oftotal value of production. For many individual fann households, rice is their sole source ofcrop incomes. This large share of agricultural GDP coming from rice to a large extent reflects the comparative advantage of rice given the agro-cIimatic characteristics of Bangladesh. where approximately 80 per cent of the country is a flood plain. Rice, thus virtually dictates the performance of the agricultural sector. Agriculture Development Strategy 275 Bank report (World Bank 2003) has shown that education is an important determinant of diversification in Bangladesh. • Because agricultural growth may translate into lower real prices received by net seners, an export strategy should be devised to expand markets and increase income. In Bangladesh, growth in rice production in the 1990s has been accompanied by declining real prices (Appendix Figure A9.1). This real price decline unambiguously benefitted net buyers of this staple, including the poor landless and marginal farmers. However, since rice is not yet exported (due to inferior quality), real price declines lead to lower income for net rice sellers unless productivity increases dramatically. 7 Exports can help prevent collapses in real prices as a result of growth. • Bangladesh has enormous potential of plugging its small producers of vegetables, spices and poultry into some market niches (Singapore, Thailand, some middle-eastern countries) and into larger and more demanding markets (e.g. European Union). • One finding of this report is the negative effect of inflation on aggregate agricultural growth. Price stability, as during the 1990s, would contribute to agricultural growth in Bangladesh. A sound macroeconomic policy is particularly crucial in the context ofgreater integration ofBangladesh to the world economy. • Finally, in developing the strategy to promote agricultural and rural non­ farm growth, it should be recognised that region-specific rural development strategies and public expenditure plans would be required, so as to take into account the differential regional potentials and initial conditions as illustrated in Appendices A9.1 and A9.2. REFERENCES 1. Antle, 'Testing the Stochastic Structure ofProduction: A Flexible Moment­ Based Approach', Journal ofBusiness and Economic Statistics, vol. 1, 1983. M. Asaduzzaman, Frontiers of Rural Changes in Bangladesh: Natural Resources and Sustainable Livelihood in Medium- Term Future, Paper prepared for DFID, Dhaka, Bangladesh, 2000. Raisuddin A hmed, Retrospect and Prospects of the Rice Economy of Bangladesh, Dhaka, University Press Limited, 2001. H. Binswanger, M.e. Yang, A. Bowers and Y. Mundlak, 'On the Determinants 7 The real price ofthis staple is an important channel through which agricultural growth affects poverty in Bangladesh. With a per capita rice consumption of 190 kg this eountry is the third world largest consumer of rice, after Burma and Vietnam (FAOstat). The staple provides more than 70 per eent of total calorie intake of the 140 million Bangladeshis. At the same time, two-thirds of Bangladesh popUlation is engaged in livelihood activities related to rice. 276 Transforming Bangladesh into a Middle Income Economy of Cross-country Aggregate Agricultural Supply', Journal of Econometrics, voL 36, 1987. P. Dorosh, 'Trade Liberalization and National Food Security: Rice Trade Between Bangladesh and India', World Development, No.2, 2002. R.E. Evenson, C.E. Pray and M.W. Rosegrant, Agricultural Research and Productivity Growth in India, IFPRI, Research Report No. 109, Washington, DC, 1999. S. Fan, P. Hazell and S. Thorat, Linkages between Government Spending, Growth and Poverty in Rural India, IFPRI, Research Report No. 110, Washington, DC, 1999. S. Fan, L. Zhang, and X. Zhang. Growth and Poverty in Rural China: the Role ofPublic Investments, Research Report 125, International Food Policy Research Institute, Washington, DC, 2002. Mahabub Hossain, Green Revolution in Bangladesh: Impact on Growth and Distribution ofIncome, Dhaka: University Press Limited, 1989. M.A.S. MandaI, 'Dynamics of irrigation water market in Bangladesh' in M.A.S. MandaI (ed.) Changing Rural Economy of Bangladesh, Dhaka: Bangladesh Economic Association, 2000. A. McGuirk, and Y. Mundlak. Incentives and constraints in the transformation of Punjab agriculture. Research report no. 87. International Food Policy Research Institute (IFPRl), Washington, DC, 1991. Y. Mundlak, Agriculture and Economic: Growth Theory and Measurement, Cambridge: Harvard University Press, Mass, 2000. - - - . 'The Dynamics ofAgriculture', Elmhirst Memorial Lecture, in G.H. Peters and 1.Von Braun (eds.), Food Security, Diversification, and Resource Management: Refocusing the Role of Agriculture? Proceedings of the Twenty- Third International Conference of Agricultural Economists, Aaldershots and Brookfield, Ashgate, 1999. Y. Mundlak, D.F. Larson, and R. Butzer. 'The Determinants of Agricultural Production: A Cross-Country Analysis', Policy Research Working Paper 1827, Washington, DC: World Bank, 2000. R. Murgai, 'The Green Revolution and the Productivity Paradox: Evidence from Indian Punjab' Policy Research Working Paper 2234, Washington, DC: World Bank, 1999. M. Ravallion and G. Datt. 'When is Growth Pro-Poor? Evidence from Diverse Experiences of India's States', mimeo, Development Economics Department, Washington, DC: World Bank, 2000. Ravallion and Chen. 'Measuring Pro-Poor Growth', Policy Research Working Paper, Washington, DC: World Bank, 2001. The World Bank. Poverty in Bangladesh: Building on Progress, Report 24299­ BD, Washington, DC, 2002. - - - . Bangladesh: Agricultural Growth and Rural Poverty Dynamics: A Regional Perspective, Washington, DC, 2003. APPENDICES FIGURE AND TABLES ~~------------------------------------------------------------------------, ~!I----~-------------------------------------------------, 24 I R 221 Fl _ _ Ill I 20 18 t=' ~ 16 ~ 14 § ~ 12 10 8 6 4 2 0 ~ ~ t3 l8 txi ~ ~ Sl c;; Sl Sl CI; >­ ~ >­ l§l >­ I"­ 0> >­ g: s: 8 ~ .., ~ .., .i!:' .., .i!:' .., .i!:' .., :::I .i!:' .., :::I .i!:' .., .i!:' .., ~ .., ~ .., ~ .., ~ .., .i!:' .., .i!:' .., .., "3 .., "3 .., "3 .., "3 ~ .., .i!:' .., .i!:' .., " " " " " " " " Figure A 9.1 National Average Real Wholesale Price of Rice and Wheat in Bangladesh. 1980-2001 Note: Prices are deflated using the non-food Dhaka middle-income Cost of Living Index (and the national CPI after June 1998). Source: Dorosh (2001); FPMU data. Table A9.1 Some Socio-Economic Indicators by Rural Regions (2000) South North Meghna Eastern Coastal Northwest Ganges central FPLN hills area Overall FP Natural constraints: Vulnerability to flood/cyclone/other disaster Low Medium Medium High High High (flood) (flood) (steep (cyclones) slopes) Other constraints: -Access to irrigation (% of irrigated area in total cultivated area) 64 47.5 54.4 40.9 35.4 11.8 47 -Density of rural roads (km square km) 0.05 0.09 0.09 0.11 0.06 0.03 -Availability of electricity in community* 13.8 10.9 25 26.4 16.8 15.6 19 Human capital development: Education 38 39 34 44 43 54 40 Health 16 12.6 11.7 13.9 21.1 17.7 14.5 Note: *Percentage of household living in a community that has electricity. Source: BBS and HIES 2000. Table A9.2 Some Socio-Economic Indicators by Rural Regions (2000) South Northwest Ganges Coastal North Meghna Eastern Overall flood region central flood hills plain plain Education Percentage of individuals aged 15-70 38 39 54 34 44 43 40 that can read Percentage of individuals aged 7-70 that 17 17 23 15 17 21 15 have primary level of education Percentage of children aged 6-10 76 72 82 71 74 79 74 currently enrolled in school Health Percentage of individuals reported being 16.0 12.6 17.7 11.7 13.9 21.1 14.5 sick Percentage of sick that sought medical 14.3 13.9 14.9 14.0 21.2 18.7 16.3 care Average distance (minute) to: - satdllite clinic 35 52 29 34 22 35 - Govt. district hospital 96 100 94 112 132 106 - Thana health complex 54 65 47 62 77 54 (Contd.) (Contd. Table A9.2) Housing conditions Share of rural household with adequate 99.0 98.25 91.4 99.6 97.7 81.4 97.1 supply of drinking water Share of rural household with proper 28.1 48.5 65.8 52.4 55.8 44.9 47.4 total facilities* Share of rural household living in 15.2 13.1 16.7 26.3 27.1 18.9 20.3 electrified dwelling. Note: *excluding those using temporary latrines or open fields. Source: HIES 2000. 10 Improving the Rural Investment Climate for Non-Farm Enterprises Forhad Shilpi SECTION I: INTRODUCTION The rural economy in Bangladesh has made remarkable strides during the last three decades. The widespread adoption of Green Revolution technology, facilitated by liberalisation of input and output markets during the I 980s and early I 990s, resulted in the cereal (mainly rice) production more than doubling since 1975. This impressive productivity gain in agriculture not only eliminated the threats of hunger and famine, but also allowed some diversification into other higher-value crops, livestock, and fisheries, and fueled a rapid expansion in rural non-farm (RNF) activities during the 1990s. The strong performance of both rural sectors during the 1990s led to an annual growth of rural income of about 2.2 per cent, and to an annual decline in rural poverty of about 1 per cent. Despite this progress, about 53 per cent of rural population still lives in poverty and 37 per cent in extreme poverty. The Interim Poverty Reduction Strategy Paper (I-PRSP) for Bangladeshl estimates that a per capita rural consumption growth of at least 4 per cent per annum is needed to reduce rural poverty by half by 2015 - as envisioned by the Millennium Development Goals (MDGs). Such strong growth in per capita consumption, in turn, requires a doubling of the rural income growth rate from the 2.2 per cent trend level registered during the 1990s. The doubling of rural growth rate can be achieved by significantly enhancing the performances of the agriculture and RNF sectors, and by creating a virtuous circle of growth between these sectors through multitudes of inter-linkages. The demographic, geographic, and economic reality of Bangladesh I Government of Bangladesh, Ministry of Finance (2003). 282 Transforming Bangladesh into a Middle Income Economy assigns a central role to the RNF sector in generating employment and income in rural areas. With 1,000 people per square kilometer (sq. km.), population density in Bangladesh is among the highest in the world more than three times the density of India. As a result, nearly all available cultivable land is already under cultivation. Agricultural land is increasingly being diverted to competing uses such as housing, roads, and industrial development. The total cultivated area declined from 8.15 million hectares (ha) to 7.20 million ha between 1984 and 1996,2 and will continue to decline in coming years. Cropping intensity has almost reached its limit, further restricting agriculture's ability to create incremental employment. Despite a significant decline in population growth rate, there are nearly a million new entrants into the total labour force every year. The urban areas currently employ only about 18 per cent of the country's total labour force of 58 million people, and even a very significant rise in growth rate in urban areas will not be able to absorb a large proportion of the new entrants. In fact, the majority of the new entrants (80 per cent) will need to find a job in rural areas. Given the limited opportunity for employment expansion in agriculture, the labour force joining the RNF sector will continue to swell in the foreseeable future. Only stimulated productivity growth in the RNF sector can ensure that those joining the sector are able to escape poverty. During the] 990s, growth in the RNF sector was triggered primarily by rapid growth and transformation ofagriculture. In particular, the robust growth ofpump irrigation (15 per cent per year for shallow tube-wells since the early 1970s) and power tillage technologies stimulated various manufacturing, repair, and other service activities at the local level. At the same time, accelerated growth in the fisheries and livestock sector helped develop the feed supply chain. The forward link ofthe supply chain, involving collection from farmers, processing, storing, and distributing to urban consumers, is now in place for most agricultural products, but its level of development remains uneven across commodities. Indeed, development of marketing and processing facilities, an important segment ofRNF economy, is essential for supporting intensification and diversification in agriculture. The rising productivity and wage levels resulting from the exodus of labour from agriculture and into RNF sector, and concomitant tightening ofthe rural labour market will facilitate creation ofa virtuous circle of broad-based rural growth and poverty reduction. However, for such broad-based rural growth and poverty reduction to occur, the growth in RNF sector has to be accelerated much beyond its current rate of 4 per cent. The strength and ability of the RNF economy to meet the daunting challenge of employment and income generation in rural areas will depend 2 Bangladesh Bureau of Statistics (BBS) ( 1999). Improving the Rural Investment Climate for Non-Farm Enterprises 283 crucially on the environment in which rural enterprises emerge and operate. The analysis of rural investment climate thus becomes essential in identifying the policy, institutional and behavioural constraints that affect enterprise dynamics and hence the potential for generating gainful employment in this sector. Moreover, the relevance of rural investment environment extends well beyond rural areas. Historical experience suggests that rural areas often act as the breeding grounds for entrepreneurs in urban areas (Hayami 1998). This paper investigates the rural investment climate using a large-scale survey of micro, small and medium-sized enterprises in Bangladesh conducted during March-June 2003. 3 The survey, known as the National Private Sector Enterprise Survey of Bangladesh (NPSESB 2003), collected information on 10,000 micro, small and medium enterprises from a nationally representative random sample. Out of the total sample of 10,000 enterprises, some 8,135 enterprises are engaged in non-farm activities where non-farm activities are defined to include all activities other than primary agricultural production (crop, fishery, livestock and forestry). The investment climate analysis in this paper is based on the responses of these 8,135 non-farm enterprises. In order to provide the context ofthe rural investment environment, the contribution and profile of the micro, small and medium (MSM) enterprises are briefly discussed in Section II. Main results on rural investment climate based on the direct responses of the MSM firms are presented in Section III Section N elaborates on the state of rural physical and financial infrastructure, the most pressing constraints identified by the enterprises in starting and operating enterprises. Section V concludes the paper. SECTION II: THE MICRO, SMALL AND MEDIUM ENTERPRISES IN BANGLADESH ECONOMY Contribution of the Micro, Small and Medium (MSM) Non-farm Enterprises The contribution of the MSM non-farm activities is substantial both in terms of employment and value added. 4 There are about 4.25 million MSM enterprises in Bangladesh (Table 10.1) and nearly 70 per cent of them are 3 For details on the National Private Sector Enterprise Survey of Bangladesh (NPSEB) 2003 done by International Consulting Group and Micro Industries Development Assistance and Services. See ICG & MIDAS (2003). See Daniels (2003) for details on sampling. 4 In the NPSEB 2003 survey. information on employment and value added are collected for all firms employing less than 100 workers. Note that the survey covers only a part of the non­ farm activities as it does not include mobile enterprises (e.g. rickshaw pullers). The survey is likely to under-represent household based activities (not considered as enterprises) such as domestic services. 284 Transforming Bangladesh into a Middle Income Economy located in rural areas. Another 11 per cent are located in smaller urban centers and rural towns. Noting that overwhelming majority of the enterprises in Bangladesh employ less than 20 workers, the importance of rural enterprises in generating employment in the non-farm activities becomes even more striking. The MSM enterprises generate employment for about 12.8 million people, 73 per cent of this employment is generated in rural areas. The contribution of MSM enterprises to GDP is also substantiaL Total nominal GDP in Bangladesh for 2003 is estimated to be taka, 2,996 billion. The MSM enterprises' contribution to GDP is estimated to be taka 531 billion, accounting for 17.7 per cent of total GDP.5 The share of MSM enterprises located in rural areas is again substantial- they account for 76 per cent of value added generated by the MSME sub-sector as a whole. Another 8 per cent of MSM enterprises' value added is generated in the smaller urban centers and rural towns. Table 10.1 Contribution of Micro, Small and Medium Non-Farm Enterprises to Bangladesh Economy, 2003 Strata _.. __ __. ­ Enterprise .. __. _ _ Employment ­ .. Value added (Million) (%) (Million) __ Taka (%) (%) __...Billion...__ _ ­ .. .. Rural 2.953 69.46 9.375 73.19 4050 76.33 Metropolitan 0.472 11.11 1.258 9.82 467 8.80 Other urban 0.455 10.69 1.204 9.40 427 8.05 Peri-urban 0.292 6.88 0.669 5.23 255 4.81 Industrial 0.009 0.21 0.034 0.27 10 0.20 Commercial cluster 0.067 1.59 0.233 1.82 86 1.62 Commercial house 0.001 0.02 0.001 0.Q1 0 0.01 BSIC estate 0.002 0.04 0.034 0.27 11 0.20 Total 4.25 100.00 12.81 100.00 5306 100.00 Source: Author's estimate based on NPSESB (2003). Dynamics of Non-Farm Activities During the 1990s The decade of the 1990s has witnessed impressive growth of the number of non-farm MSM enterprises and their employment. Bangladesh Bureau of Statistics conducted a survey ofall non-farm activities in 1989-90 (Integrated Annual Survey of Non-Farm Activities - IASNF 1989-90) using restricted definitions ofthe non-farm activities. 6 Using this restricted definition ofnon­ farm activities, the estimates from the IASNF and NPSEB surveys show that number ofnon-farm enterprises has increased from 2.1 million in 1989-90 to 5 This assumes that total nominal GDP already included value added generated by the MSM enterprises. Note also that the estimate of GDP generated by MSM enterprises is likely to be underestimate because of large number of non-responses to profit questions. 6 For detail on this definitional issues, see Shilpi 2003. Improving the Rural Investment Climate for Non-Farm Enterprises 285 4.1 million in 2003, implying an annual average growth rate of 5.3 per cent (Table 10.2). Employment has increased by an annual average rate of 3.8 per cent, from 5.3 million to 9 million during the same time. More interestingly, the enterprises in permanent structure, which tend to have higher productivity, experienced impressive growth in employment (4.2 per cent). The number of enterprises in permanent structure has also grown (2.6 per cent) implying an extensive growth. Importantly, employment growth has exceeded growth in the number of enterprises implying an increase in the average size of the enterprises in permanent establishments. Such extensive (growth in number of enterprises) and intensive (growth in size) growth highlights the inherent strength of the sector in transforming itself into a more dynamic and efficient sector. The process of structural transformation into more productive and dynamic sector is also evident from the fact that household based enterprises experienced much higher growth in number than employment. The process of shift from household to permanent establishment is clearer in manufacturing sector where household based enterprises experienced a slight decrease in employment. In rural areas, the growth of the number of enterprises. and of employment have been slightly slower compared with national trends, yet the process of structural transformation is more evident there. Manufacturing sector in permanent establishment in rural areas has experienced impressive growth in numbers (6.1 per cent) and employment (10.4 per cent), exceeding the growth in urban areas. As observed at the national level, household manufacturing experienced decline in employment (0.85 per cent) in rural areas. Table 10.2 Dynamics of Non-Farm Enterprises Number of enterprises Employment Permanent HH Total Permanent HH Total ---~-----~-~- ..- -. .. -~---.--- ... -~-----~---.--~- _ - (National) Growth- _.. ... __ __ __ (%) (%) .. . .. (%) (%) .. (%) __. _ - - _ (%) ..­ 1989/90-2002103 Industry 5.84 3.38 4.01 7.03 -<:J.11 1.69 Services 2.15 8.82 5.68 3.51 5.48 4.66 Total _ _ _ _ _ _M _______ 2.63 7.29 5.32 4.17 3.64 3.83 . -----~~----- .. -----.--~---.------- Growth (Rural) (%) (%) (%) (%) _-_ _ -(%) .. _ .. .. (%) 1989/90-2002/03 Industry 6.13 2.24 2.91 10.36 -<:J.85 0.62 Services 2.53 8.18 5.85 3.42 5.02 4.51 Total 2.98 6.25 5.11 4.47 2.86 3.28 Source: Author's estimate based on NPSESB 2003. 286 Transforming Bangladesh into a Middle Income Economy Profile of the Micro, Small- and Medium-Sized Non-farm Enterprises Consistent with evidence from other developing countries, the non-farm activities in Bangladesh consist of widely heterogeneous activities. The sector is dominated by wholesale and retail trade across the geographical spectrum with its share ranging from 60 per cent in smaller towns to 54 per cent in metropolitan areas (Table 10.3). The share of trade in total enterprise population is about 57 per cent in rural areas. Interestingly, a quarter of all enterprises in rural areas are engaged in manufacturing activities, higher than that observed even in metropolitan areas (18 per cent). Trade appears to be particularly important in the smaller urban areas and towns. The predominance of trade in non-farm sector signifies its relatively early stage of development. Historical evidence indicates that at later stage of RNF development, the predominance ofrelatively low productive trading is replaced by more skilled services activity and manufacturing. Table 10.3 Sectoral Composition of Micro, Small and Medium Non-Farm Enterprises (per cent) Rural Other Peri-urban Metropolitan Bangladesh areas urban areas areas areas Manufacturing 25.1 17.2 14.1 18.4 22.6 Construction 2.1 1.3 1.1 0.7 1.8 Wholesale & retail trade 57.4 59.8 55.1 53.9 57.3 Hotels & restaurants 3.5 9.9 18.8 17.8 6.9 Transport, storage & communications 2.2 0.9 2.6 1.3 2.0 Real estate, renting, business activity 5.1 3.3 2.9 2.1 4.4 Education 0.0 0.3 0.0 0.1 0.1 Health and social work 1.1 1.5 0.2 0.4 1.0 Other service activities 3.5 5.8 5.2 5.3 4.1 Total 100.0 100.0 100.0 100.0 100.0 Source: Author's estimate based on NPSESB (2003). The enterprise profile indicates that a typical enterprise is small in size, fairly young in age, employs mainly unpaid family labour, provides self­ employment for the owner, yet it is not engaged in one of the low productive residual type activities. The average size of a typical enterprise is about 3 workers (Table 10.4). More than a third of the firms/enterprises are less than 3 years old. The age distribution suggests higher rate of enterprise start­ ups in the large urban cities, particularly in peri-urban areas compared with Improving the Rural Investment Climate for Non-Farm Enterprises 287 rural areas and towns - about 26 per cent of enterprises in peri-urban areas are less than two years old relative to 17 per cent in rural areas. More than a third ofthe rural enterprises are household based enterprises. Urban enterprises are more frequently based in permanent structure such as shops. Employment data for enterprises show that nearly all ofthe owners work in their enterprises (99 per cent) implying that these enterprises also generate self-employment for the entrepreneurs themselves. Yet, they do not seem to be activities undertaken just to earn some income during slack seasons in agriculture. Most enterprises are operated year around (11 months in a year and 27 days in a month), and tend to have long working days (11 hours in a day). Though number of months for which the enterprises are operated is slightly smaller in rural areas (10.9 months), the rural enterprises are by no means operated as stop-gap activities, which follow agricultural cycles. Detailed age and size profiles of the enterprises indicate that non-farm activities provide rich breeding grounds for entrepreneurs in both urban and rural areas. The MSM enterprises in non-farm sector lacks access even to basic factors ofproduction and services. About 40 per cent ofthe enterprises reported having utility connections (Table 10.5). 7 There are striking differences between Table 10.4 Basic Profile of Micro, Small and Medium sized Non-Farm Enterprise Rural Other Peri-urban Metropolitan Bangladesh areas urban areas areas areas Average size 3.2 2.7 2.3 2.7 3.0 Average age 9.3 9.0 6.5 7.9 8.9 Enterprise type Home 37.0 22.3 31.0 23.8 32.9 Traditional market 6.2 3.1 4.7 4.5 5.6 Shop 28.6 51.3 43.1 45.7 34.8 Roadside/riverside 11.8 16.3 12.6 13.9 12.6 Mobile enterprise 15.8 6.5 8.4 11.2 13.6 Industrial site building 0.1 0.0 0.0 0.8 0.2 Enterprise operation Months/year of operation 10.9 11.6 11.8 11.8 11.1 Days/month of operation 27.7 27.7 28.5 28,2 27.8 Hours/day of operation 10.2 11.8 13.0 12.7 10.9 Source: Author's estimate based on NPSESB (2003). 7 The estimate of utility cOlll1ection may be an under estimate as it is derived from responses of the enterprises about their expenses for utilities. 288 Transforming Bangladesh into a Middle Income Economy urban and rural MSM enterprises in non-farm sector itself. In rural areas, less than a third of firms have utility connections compared with more than 60 per cent in urban areas. MSM firms in both rural and urban fare worse in terms of ownership of generators - only 3 per cent of the firms reported owning it relative to 72 per cent of larger firms located in Dhaka and Chittagong (Investment Climate Assessment (ICA) Survey, 2003).8 Access to phone is equally dismal in rural areas. In terms of access to finance, less than a third borrows from formal sources including micro-finance institutions. In the ICA 2003 survey, 65 per cent of the larger firms have a line of credit/overdraft facility with formal banks. Within MSM non-farm sector itself, rural enterprises fare slightly better than other enterprises primarily due to access to micro-finance. Among MSM enterprises, less than a fifth are registered or members of business associations. This limits their ability to access market and technical assistance through these channels. Table 10.5 Access to Factors and Services (per cent) Rural Other Peri-urban Metropolitan Total urban Enterprises with access to services .. __ __ __. _ - ­ ... .. Utility connection 31.49 59.65 57.05 59.88 40.00 Generator 1.81 3.96 3.40 4.15 3.00 Phone 1.51 5.94 4.54 9.79 4.00 Registered 13.20 37.93 22.05 41.78 21.11 Association member 16.21 25.08 16.04 21.95 18.41 .. - -...- -.. _ __ __ __ __ _ _ __ __ Access to credit ... .. .. .. .. .. .. .. None 41.25 47.03 54.62 49.06 43.74 Informal 22.23 20.96 27.23 32.86 23.71 Formal* 24.62 20.02 9.89 11.02 21.45 Both 11.90 11.99 8.27 7.06 11.10 Source: Author's estimate based on NPSESB (2003). Note: *Formal includes micro~'ending by NGOs. Productivity of the MSM Non-Farm Enterprises The productivity of the MSM firms, measured by total factor productivity (TFP) and labour productivity, varies widely across locations with firms 8 The ICA 2003 survey collected information on relatively larger firms located in the Dhaka and Chittagong, two main metropolitan cities in Bangladesh. Improving the Rural Investment Climate for Non-Farm Enterprises 289 Table 10.6 Productivity of the MSM Non-Farm Enterprises Productivity Variance of productivity Labour TFP Labour TFP Strata Taka Taka Rural 27978 8.8 1476976 0.003 Other urban 34454 9.2 3474341 0.005 Peri-urban 41377 9.3 9915979 0.009 Metropolitan 46064 9.5 5628973 0.006 Sector Manufacturing 24479 8.7 1.39E+06 0.011 Construction 27048 8.6 3.02E+07 0.047 Wholesale & retail trade 34611 9.1 1.47E+06 0.001 Hotels & restaurants 30422 9.1 5.28E+06 0.008 Transport, storage & communications 41794 9.3 1.59E+07 0.011 Real estate. renting. business activity 34963 9.1 1.61E+07 0.010 Health and social work 37909 9.5 5.00E+07 0.028 Other service activities 35033 9.2 6.12E+06 0.006 All 32314 9.0 931.3 0.039 Source: Author's estimate based on NPSESB (2003). located in rural areas and smaller towns having somewhat lower productivity than their counterparts in urban areas (Table 10.6). Similar variations are evident across sectors, age categories and establishment types. TFP is lower in rural areas, in home-based enterprises, in manufacturing and construction, and in firms of age higher than 50 years. The average TFP is higher in urban areas, health and social work, in smaller firms, and in firms based in the industrial sites. Significant differences in productivity exist within each size category of firms (Figures 10.1 and 10.2), within each sub-sector, each type of firms, and in each region. For instance, among firms of size equal to one, labour productivity varies from approximately taka 1 to about taka 180 thousand. Variation in TFP is much smaller, yet significant. Such large variations are observed for firms in each sub-sector (manufacturing, trade, etc.), for each type of firms (home based/with permanent structure, etc.).9 In a competitive and efficient market with free entry and exit, the force ofcompetition is likely to push the inefficient fIrmS out of business causing exit and hence wide productivity differences across firms are not likely to persist. The persistence ofsuch huge productivity differences across wide spectrum offirms seems to 9 For details, see Shilpi (2003). 290 Transforming Bangladesh into a Middle Income Economy suggest that lot of non-farm activities are either dealing with highly differentiated niche products or that there are significant constraints in acquiring technology and services which can allow firms to improve their performances or that entry and exit costs are too high. Indeed, a big bang in the growth of non-farm income and value added can be achieved simply by allowing inefficient firms to upgrade their technology and products. This will become feasible if critical bottlenecks in firm's access to factors and services as well as entry and exit constraints are removed. 200,000 • Lowess smoother, bandwidth .8 • • •• • l::­ 150,000. •• • • • • :~ •• •• .. . • • •• • • •••'. • '0 100,000 e :::l "0 • •• a. • 50,000 ___ 9 • • • • • • • • 0 lill ll ·:,'·. 0 5 10 15 20 tot # of workers adding up categ. Figure 10.1 Labour Productivity by Firm Sizes 15 • Lowess smoother, bandwidth .8 • • • •• • • • • 10 •• ~ • ••• • ••• • • • • • •• • • 5 • o ~------~~------.--------r------~~ o 5 10 15 20 tot # of workers adding up categ. Figure 10.2 TFP by Firm Sizes SECTION III: CONSTRAINTS TO ENTRY AND OPERATION: THE VOICES OF THE ENTREPRENEURS The NPSESB 2003 collected information on the constraints faced by entrepreneurs at the start up and in running a firm using two different methods. First, the entrepreneurs were asked to indicate what were the two most important problems encountered when they started the firm and in operating Improving the Rural Investment Climate for Non-Farm Enterprises 291 an existing finn.lo Second, responses on a list of possible constraints were gathered on a scale of I ('no problem') to 3 ('severe problem') only after responses for the open - ended questions were recorded. We utilise responses from both ofthese two sets ofquestions to identify possible constraints faced by the entrepreneurs. From the responses of the finns to open-ended questions about main constraints, access to finance comes out to be the most pressing constraint in starting and operating an enterprise (Table 10.7). More than half of the finns reported lack of access to finance as the biggest problem in starting an enterprise. About 44 per cent ofentrepreneurs noted it as the biggest problem in running a finn. A finer analysis of the entrepreneurs reporting finance problem indicates that an overwhelming majority of the entrepreneurs (89 per cent) identified lack of investment funds as the biggest problem for starting up a finn and about 60 per cent identified lack of working capital as the biggest problem in operating a finn. Interestingly, about 35 per cent ofthe entrepreneurs reported having no problems in starting a finn and another 26 per cent reported no problems in running it. Table 10.7 Constraints to Start-up and Operation of a Non-Farm Firm:Open-Ended Questions Constraint type ---------------_._------­ Starting-up (%) Current (%) Finance 53.35 44.34 Tools/Machinery 0.32 0.45 Market 2.91 15.05 Government/Regulatory 0.57 0.92 Shop/Rental space 1.81 1.96 Input problem 0.22 1.22 Transport 1.11 2.16 Labour 0.41 0.66 Utilities 0.42 1.21 Technical 1.15 0.28 Misc. 1.99 3.39 Agriculture 0.96 1.58 No Problem 34.76 26.77 From a fixed response questionnaire, entrepreneurs stated flood and disaster, electricity, road condition, access to finance, transport to market as five most severe problems in running an enterprise. Figure 10.3 plots the percentage of finns reporting a given problem as a 'severe problem' for finns 10 Theenumerators were specifically instructed not to make any suggestions in order to avoid creating a bias towards the suggested problems. 292 Transforming Bangladesh into a Middle Income Economy located in rural areas and metropolitan cities. Among the rural firms, flood and natural disasters ranked first with 37.1 per cent of firms indicating it as a severe problem. Nearly an equal percentage of rural firms (37 per cent) reported electricity availability as a severe problem. Road conditions rank a close third (36 per cent reporting, it as a severe problem) followed by access to finance (34 per cent), transportation to market (18 per cent), and crime, theft and disorder (10 per cent). There are some interesting differences between rural and urban metropolitan firms. Among the urban fmns, electricity is ranked as a severe problem by largest number of firms (24 per cent), followed by flood and natural disasters (21 per cent), access to finance (20 per cent), road condition (18 per cent) and too many competitors (15 per cent). Among major constraints identified by both rural and urban firms such as electricity, flood and natural disaster, road conditions and so on, the percentage offmns reporting these as severe constraints are significantly larger in rural areas the percentage of firms complaining, for instance about road condition in rural areas, is double the percentage of firms in urban areas. On the other hand, more urban firms complained about competition and anti-competitive practices (20 per cent vs. 9 per cent), water availability and sewer and rubbish disposal (23 per cent vs. 11 per cent), corruption, political influence and taxes (11 per cent vs. 4 per cent). 40 35 I_ Ru~1 : 30 • SMA . 25 20 15 10 5 0 'Of!! &,1D .;£e; 1i>~ C< ~ 'f: ::n= c.0:1 ~~ 0 ~g ¥a~ 0:1'li) o~ "" .... '" g.!Il "'0:1 '5 C "'::!: ~.9 aia $a Eo:1 U 0:1.­ t'.l~ _0 u.~ ~ 0 0 '0 0:1 0 .....g E-o .'" C 0'" '" '" § i .... e 4) firms, confirming one of the long held view about 'missing middle' in credit delivery. Flood and natural disasters affect micro and medium sized (20+ employees) firms more adversely. Crime and theft is more ofa problem for the small firms (size 2 to 5). Similar to pattern observed in the case of access to finance, the micro and medium sized firms complain less about electricity. Such non-linear pattern is consistent with expectations - the micro firms do not complain so much, for instance about electricity, as the scale oftheir operation and focus oftheir activity are such that electricity may not be as important for them. This may also be due to the fact that these finns have already adopted a line operation which would not require electricity as much simply because they do not have access to electricity. In terms of 'severe problems' Barisal and Rajshahi divisions come out to be the areas where firms face disproportionate share of problems by most measures. Firms located further away from urban markets have greater propensity to face problems with accessibility to services and factors. In the regression analysis ofa firm's satisfaction with the state ofdifferent services and factors (electricity, road, finance), the accessibility to urban centers, measured by distance to poles, comes out to be statistically significant with a negative sign after controlling for a wide range of firm, sector and region specific characteristics. The peril of the missing middle is evident in the cases of access to finance and electricity. Firm size is correlated significantly with an entrepreneur's satisfaction with access to finance and electricity. In the case offinance, firms with size bigger than one employee but less than or equal to 10 employees complained much more and there appears to be a concave response curve with firms ofsize equal to 4 complaining the most (Figure 10.5). Almost similar pattern is observed in the case of electricity. Tackling the Constraints: The Challenges and Options According to the rural investment climate analysis presented in the previous section, for rural entrepreneurs, access to bare minimum services (e.g. infrastructure) remains simply inadequate and even when access is not the issues, reliability ofservices is questionable. The policy debate in Bangladesh's development arena has recently been dominated by concerns of large urban firms which are disproportionately affected by the governance issues as reflected in their responses in the investment climate assessment survey. Rural entrepreneurs, however, face different set of constraints and hence a different set ofpolicy and investment instruments, perhaps specific to rural space, will be needed to address those constraints. A better understanding ofthe roles of public, private and other stakeholders in promoting this sector calls for a deeper 296 Transforming Bangladesh into a Middle Income Economy 0.7 , . . . . - - - - - - - - - - - - - - - - - , 0.6 +--­ 0.5 0.4 0.3 0.2 0.1 o 2 3 4 5 6-10 11-20 III Electricity • Finance I Figure 10.5 Complaints About Access to Electricity and Finance by Firm Size analysis of the constraints faced the rural entrepreneurs. 13 In this section, the state of rural physical and financial infrastructure is reviewed in detail to identify the specific policy and investment options. SECTION IV: RURAL INFRASTRUCTURE: INADEQUATE ACCESS AND UNRELIABLE SERVICES Transport Demand for transport services will continue to grow in Bangladesh. making it even more important for the continued expansion of agriculture and non­ farm activities. Providing transport services raises challenges at many levels. At local scale, there is a need for reliable, year-round rural feeder roads that allow farmers and rural non-farm producers to bring their products to local markets or trading centers cheaply and quickly. At the other extreme.. it requires a national transport system that allows farmers and producers in the hinterland to ship products to major national markets and to the major export hubs (ports and airports). Overall, demand for transport services is likely to grow strongly in Bangladesh as the share of trade in the country's GDP increases. For instance, the 1998 Bangladesh Integrated Transport Study estimates that a transport sector growth rate of 8-9 per cent is required to sustain GDP growth of 7 per cent per year as targeted in the Fifth Five-Year Plan. Road density per square kilometer (sq. km.) in Bangladesh is among the highest in South Asia, yet the rural entrepreneurs lack access to reliable, timely, and year-round transportation services. Transportation in Bangladesh relies on three main modes: road, which accounts for 73 per cent of passenger 13 The high degree of heterogeneity and spatial dispersion of the rural entrepreneurs along with their small scale of operation makes delivery of some critical services such business development services a daunting issue. The broader issue of rural service delivery, particularly institutional option, is not covered in this chapter and can be found in World Bank (2004). Improving the Rurallnvestment Climate for Non-Farm Enterprises 297 kilometers and 63 per cent of freight transport; raiJ (13 per centl7 per cent); and inland waterways (14 per centi30 per cent). 14 The share of road transport continues to increase for both passengers and freight, while the importance of railways is declining and inland water transport remains fairly stable. Bangladesh's primary road network consists ofnational and regional highways and Zila roads (Table 10.8). It has a length of 12,400 km and is managed by the Roads and Highways Department. Another 92,000 km make up the secondary road network - Upazila and Union roads - that is managed by the Local Government Engineering Department (LGED). The remaining tertiary village roads represent over half of the total road network and are managed by local government institutions. As already noted, Bangladesh ranks high in road density, and except for investing in filling the critical gaps in existing network, the return to investment in new road construction is likely to be small and diminishing. Instead, the critical challenges for the road transport sector include. (i) poor maintenance of rural roads making them available only for part of the year, (ii) lack of integration of transport modes, and (iii) lack of complementary transport and trading services. ls Despite the extensive road network in Bangladesh, poor construction and maintenance of roads, particularly rural roads, impose high costs on the users. A large proportion of rural roads were constructed through the 'Food for Work' and other rural development programmes. Because ofthe emphasis on labour - intensive technology, and lack of attention to engineering design, nearly 80 per cent of the roads are of poor quality. Only a relatively small proportion of the total road network consists of paved roads. For instance, among the Upazila roads (Feeder Road B), only about one-third are paved Table 10.8 Bangladesh Road Network Length .. _ . _ . _.. Category km Primary National highways (NH) 3,086 Regional highways (RH) 1.751 lila roads (lR) (former feeder roads A) 7.543 Secondary Upazila roads (UlR) (former feeder roads B) 23,434 Union roads (UR) (former Rural Roads 1) 68.639 Tertiary Village roads (VR) (former rural roads 2 and 3) 114,126 Not classified 1,960 Total 220,539 Source: RHO and LGEO. 2003. 14 Bangladesh Integrated Transport Study 1998. IS This section builds on PPIAF (2003) which details the problems in the transport sector and outlines possible options to increase efficiency in each transport area. 298 Transforming Bangladesh into a Middle Income Economy Table 10.9 Quality of Rural Upazila Roads Type of upazi/a road surface Bitumen carpet Herring bone bond Water bound Earth macadam Per cent of all upazila roads 32.0 6.7 0.8 60.6 of which (%) Good 79.5 55.3 70.1 2.6 Fair 7.8 8.4 12.0 95.8 Poor 12.7 36.2 17.9 1.6 Source: Calculations based on LGED database, 2003. roads (Table 10.9). Of those, 80 per cent are considered in good condition. Of the earth roads, 96 per cent are in fair condition. Almost none of the Union roads are all weather roads. In addition to the roads themselves, there is also a large need for bridges and other infrastructure to cross the huge number of streams and rivers. Many of roads themselves largely need bridges and other infrastructure to cross the huge number of streams and rivers. Many of those are in poor repair. In addition to making roads impassable during monsoon, poor maintenance imposes high costs on users as vehicle operating costs increase strongly when road conditions deteriorate. Indeed, road maintenance projects have been shown to have a higher rate of return than transport or other infrastructure projects in general (Gwilliam and Shalizi 1999). In the presence of chronic shortage of public funds for road maintenance, as is the case in Bangladesh, a second generation or commercially managed road fund provides an important alternative path to sustainable road maintenance financing. Improved financing approaches have also to be paired with better targeting of maintenance of the secondary and tertiary feeder road network. With economic activity outpacing road construction and lack of proper maintenance and upgrading, major highways have become overly congested, contributing to major delays and higher transport costs. A large proportion of the major highway network carries a heavy volume of traffic that approaches or surpasses its capacity. For instance, the most heavily used national road (Tongi-Joydevpur road, with four traffic lanes) carries about 34,000 vehicles per day. Along with traffic delays, road death has emerged as a major source of accident-related death among young and adults. Long haul transport in Bangladesh remains underdeveloped because of poor integration of different modes of transportation. Major challenges remain in the integration of all different modes of transportation such as road, ferry, river transport, railways, and air transport. Improving the Rural Investment Climate for Non-Farm Enterprises 299 Road-Ferry The major rivers- especially the Padma (Ganges) and Jamuna (Brahmaputra) effectively divide the country into three zones. River crossings by ferry delays freight by many hours. Bridges across these vast rivers, which are characterised by shifting riverbeds, is expensive, yet they can significantly contribute to increased inter-regional trade. The Jamuna Bridge linking Tangail and Sirajganj zila, for example, has reduced travel time from Bogra to Dhaka by about four hours. Anecdotal evidence suggests that farmers in Bogra district, for instance, have started to produce higher value crops that are shipped to Dhaka overnight by truck. However, currently the tolls charged by the Jamuna Multipurpose Bridge Authority concessionaire are high compared to the cost of ferries, and utilisation of the bridge remains low. Clearly, tolls are needed to be reassessed if the bridge is to develop its full economic benefit to the regions it connects. Road-river transport Only about 3,500 km of waterways are available for navigation in the dry season, while more than 6,000 km are navigable in the wet season. Major barriers to using waterways more efficiently are the lack of dredging and poor management and maintenance of landing infrastructure such as jetties. While inland water transportation is widely used for long-distance travelling, especially among the poorest population groups, use for cargo transport may well be below potential. In addition to the formal inland water transport system, there are an estimated 2,80,000 country boats operating. 16 These are increasingly motorised (see Box 10.2). But their potential in addressing transport needs has diminished, because waterways are often obstructed by other infrastructure development. Road-rail Because of the inefficiency and resulting high cost of the rail system, rail transport has lost a considerable share ofboth passengers (from 20 per cent to 13 per cent) and freight (from 17 per cent to 7 per cent) between 1985 and 1998. 17 This trend is likely to continue without majorrevamping ofthe system. On the other hand, for export oriented RNF products, an efficient container shipment system could provide a flexible and cost - effective way to export products through the shipping hub in Chittagong. Currently, rail is used to ship less than a third of container cargo unloaded in Chittagong. 16 In 1999-2000: Motorised: 65,000; Passenger: 1,38,000; Cargo: 74,000 (Statistical Yearbook 2000). 17 Planning Commission, Bangladesh Integrated Transport System Study, June 1998 (World Bank 2003c). 300 Transforming Bangladesh into a Middle Income Economy Box 10.2 Unintended Effects of Trade Liberalisation Promote Rural Water Transport Increased use of motorised water transport is, in part, a consequence of the liberalisation of agricultural inputs in Bangladesh during the 1980s. Farmers imported a large number of small diesel engines that were used for pumps or shallow tube wells. During the off-season these engines could be used to propel river boats or to power rice dehullers. These uses were not anticipated, but had a tremendous effect on increasing rural transport and income opportunities in the rural non-farm sector. By some estimates, this process has contributed to the establishment of 30,000 small rural dehuller mills (Chowdhury and Haggblade 2000, Haggblade et al. 2002). Road-shipping (containers) With rapidly rising demand for port services, the main port at Chittagong is unable to supply sufficient capacity and efficient transactions to meet the needs of exporters and importers. Delays at the port due to inefficient management and lack of investment in new loading/unloading technology, cause high turnaround times of6-9 days for a standard container ship compared to 1-2 days in Colombo or Singapore. Bangladesh moves a container for US dollar 600, compared with US dollar 150--300 in neighbouring ports such as Bombay and Shanghai. The system of container transport has yet to become important because ofa lack ofmodern high-capacity trucks that could provide point to point shipments to and from the ports. Expansion oftrade will require long-overdue reform of container port management in Chittagong and a removal of barriers to the adoption of modem container trucks. Road-air (export) Air freight is used to ship horticultural products such as fresh produce to overseas markets. The total amount ofair freight from Bangladesh is currently over 85,000 tons of which 17.5 per cent are fresh produce exports mainly of ethnic vegetables to Gulf States and Europe. The structure of current freight charges in Biman Bangladesh Airline flights currently the only air-freight option available to fresh produce exporters - leads to a preferential access to freight space of readymade garments at the expense of other freight such as horticultural or high value agricultural products. There is a consensus that the main bottleneck to increasing these exports is the shortage ofcargo space for perishables, the lack of dedicated handling services (e.g. specialist cargo consolidators, cold storage), and the fragmented nature ofproducers and export agents. Investment in these services can be expedited by encouraging foreign carriers to enter the business of transporting perishable goods, providing incentives for additional flights from and to Dhaka, and promoting productivity Improving the Rural Investment Climate for Non-Farm Enterprises 301 gains and complementary services (packaging, marketing, etc.) in the horticultural sector (Dixie 2002). Road only The road transport is characterised by lack ofcomplementary services lacking options both at the larger and smaller ends. Most cargos in Bangladesh are transported by 8 ton trucks, which are manufactured in India and assembled in Bangladesh. So far, modem high capacity trucks, including the ones with refrigeration capacity, are not widely used in Bangladesh, limiting the prospect of developing a container shipping system especially for perishable goods. Smaller trucks, on the other hand, could address transport needs between bicycle rickshaws and medium size trucks and could serve as a more efficient link between local producers and regional markets. The use of large trucks for local transport also puts a large strain on rural roads that are not built for heavy loads. The limited use of small trucks, which could be purchased with smaller investment, is caused partly by high import tariffs to protect the domestic assembling operations for larger trucks. Achieving a more diversified transport fleet mix among private service providers will require the lowering or removal of taxes, tariffs and regulations that currently prevent import or local production of such vehicles. Access to Power Despite recent improvements in electricity coverage, access to power remains a major issue in rural Bangladesh. Only 31 per cent of all households in Bangladesh and 19 per cent in rural areas have access to electricity (HIES 2000). Similarly, only about 40 per cent of enterprises located in rural areas are reported to have electricity access. Bangladesh ranks low among South Asian countries in terms of electricity access. Access to electricity in Bangladesh remains less than half of that of India, and per capita use is less than a third (WDI 200 I). Ofthe total energy use in the country of 12.7 billion kWh, 41 per cent is used by households (mostly in urban areas), 47 per cent by industry and 7 per cent by commercial users. Irrigation consumes about 4 per cent. The lack of access to electricity imposes large costs on rural economy and rural household's welfare. Both the household and enterprise level analyses in Chapters 3 and 5 already pointed out the positive correlation of electricity connections (at household and village level) to RNF employment and performances ofthe RNF enterprises. A recent impact evaluation study (Barkat et at. 2002) also confirms that rural households in Bangladesh with access to electricity tend to be better offthan households in villages without electricity. The study estimates that the average annual income of households with electricity is 64.5 per cent higher than that in the households ofnon-electrified 302 Transforming Bangladesh into a Middle Income Economy villages. is For shops and trading activities, the study estimates that rural electricity's contribution to sales turnover is 17.3 per cent for retail and 11.2 per cent for wholesale businesses. The off-farm income produced by electrified households was estimated to be 33 per cent higher than households in non­ electrified control villages and 66 per cent higher than non-electrified households in villages with access to electricity. In the industrial sector, firms with electricity are estimated to generate about eleven times more employment than the firms without. Over the past five years, output (measured by weight) in electrified firms has increased almost 80 per cent compared to 8 per cent in firms without electricity. Even for those enterprises and households with access, unreliability of electricity supply causes major disruptions, leading to its under-utilisation in productive activities. In the Barkat et aL (2002) study on rural electricity, about 85 per cent of respondents complained about irregular supply, with 72 per cent stating that power interruptions occur almost daily. In rural areas, electricity is often cut-off in the evening hours, when it would be most beneficial to households (World Bank 2001). These findings are consistent with those reported in Chapter 5 and investment climate assessment study by World Bank (2003). Unreliability is perhaps one ofthe major reasons for low utilisation of electricity in production activities in rural areas. To stimulate and sustain rural growth, Bangladesh will need to ensure a much higher rate of rural electrification and greater reliability of electricity supply in rural areas. If current new connection rates of about 2,00,000 customers per year continue, it will take until 2010 to double the proportion of rural households with access (World Bank 2001). Even if the rate of new connection could be doubled, it would take 25 years to reach halfof the rural population. Given the shortage ofpower throughout the country, it is unlikely that new power plants will benefit the rural areas in the short-run as investment in power plants is likely to occur in or near large agglomerations. More promising avenues for increasing access to electricity in rural areas quickly include decentralised generation systems and reliance on distribution through independent micro grids. The inadequate access in rural areas and unreliability of power supply in both rural and urban areas are results ofan existing institutional mechanism, dominated by the public sector, for generating, transmitting, and distributing power. The power sector is dominated by the Bangladesh Power Development Board (BPDB), a public sector utility that accounts for 81 per cent of 18 This study does not discuss possible endogeneity in that electricity lines may have been preferentially placed in already better-off areas where customers are more likely to afford the connection charges. But at least some of the difference in household welfare is likely due to access to electricity. Improving the Rural Investment Climate for Non-Farm Enterprises 303 generation, 55 per cent oftransmission, and 43 per cent ofdistribution. While transmission and distribution are entirely controlled by public utility companies, the Independent Power Providers (IPPs) account for 17 per cent of power generation. The power sector suffers from huge under-investment because of limited private sector participation, particularly in downstream activities, and because of the financial fragility of the public sector. Private sector participation has remained well below its potential because of the absence ofa coherent strategy, including an appropriate regulatory framework, and the entrenched presence of public utility agencies. 19 The public sector has been unable to generate surplus for required investment because of high system loss, poor financial management (e.g. inadequate pricing, poor billing and collection), and ineffective governance. 20 Bangladesh has to build on the success of Rural Electrification Board (REB) and Palli Bidyut Samities (PBS) in ensuring wider access to electricity in rural areas. Power distribution in rural areas is coordinated by the Rural Electrification Board (REB), which negotiates prices with the Bangladesh Power Development Board (BPDB) and the five Independent Power Producers (IPPs). Distribution systems are owned and operated by 67 consumer cooperatives (Palli Bidyut Samitis, PBS), which currently reach about 20.5 million people. In contrast to the performance problems associated with BPDB's urban distribution system, the rural distribution system is well managed. Although tariffs in rural areas are below purchase costs, they are higher than that in urban areas and high compared to rural power prices elsewhere in South Asia. The main problems faced are reliance on external power suppliers, e.g. BPDB cuts supplies to PBSs if there is an increase in urban demand, and infrastructure problems in supplying customers in low density rural areas. With a reasonable distribution system in place, improving coverage of REB and PBSs and removing the unfair load shedding policy will not only improve household's access but also encourage more productive and efficient enterprises to emerge and sustain in rural areas. Given higher capital costs of supplying power from the grid in rural areas, alternative power generation and distribution by private sector should also be encouraged. New innovative approaches ofdeveloping Remote Area Power Supply Systems (RAPSS) seem to be more promising for Bangladesh as well. These are community based programmes that encourage private 19 World Bank: (2003b). 20 The system loss on account ofpower pilferage and inadequate maintenance has been as high as 35 per cent of power generation and purchase in 2000. There are significant delays in bilI collection, resulting in large account receivables (12 months of sales revenue for BPDB in 2000). In addition, low power tariffs compared with marginal costs amounted to large subsidies to consumers (households and agricultural use), and contributed to BPDB's financial difficulties. 304 Transforming Bangladesh into a Middle Income Economy suppliers to develop local generation and grid infrastructure. Renewable off~ the~grid technologies, such as photovoltaic systems, are options for households with limited electricity needs particularly in remote areas, but they are still too expensive for most customers and tend to be insufficient for industry or commercial activities. Access to Telecommunication With increasing dependence on markets beyond local areas, the need for access to telecommunication by rural producers has been rising sharply, yet actual access remains inadequate in rural Bangladesh. Despite the recent widely publicised increase in cell phone coverage, Bangladesh continues to have a low tele-density (number of telephone lines per hundred person) of 0.8 per cent, compared to India's 4 per cent. Tele-reach (the proportion of the population with access to a telephone) is higher. For instance 'Grameen Mobile's Village Phone' programme is now available in 20,000 of 68,000 villages. According to the HIES (2000) data, 31 per cent of rural non-farm enterprises had a telephone in their village compared with only 14 per cent among farm enterprises. India has achieved a tele-reach of about 70 per cent in large part thanks to a programme of making public call offices (PCOs) available in rural areas. Bangladesh has one PCO per 32,000 population, while India has one per 1,000. In addition, cost of telephone services are very high in Bangladesh. The cost of a fixed line is USD 171 compared to India's USD 18 or Nepal's USD 28 (World Bank 2003a). The mobile phone network has expanded rapidly in rural areas in recent years, and will continue to be the main source oftele-coverage in the coming years. There are now almost a million cell phone subscribers compared to 6,05,000 fixed line connections. 21 Among four cell phone providers, Grameen and Aktel have extended their network significantly into rural areas. Coverage is highest in major cities and along major transport routes, while some major gaps remain in the rural areas in the northeast, south and center. There remains significant unmet demand for telecommunications, due to slow growth in the coverage of private mobile phone networks caused by regulatory constraints, and even slower progress in the expansion of fixed­ line networks due to poor performance of the Bangladesh Telephone and Telegraph Board (BTTB), a public sector telecom agency. The unmet demand for basic telephone services is estimated to be around 3.5 million, and is expected to grow to 10 million within next three years. The BTTB has a de facto monopoly on domestic fixedline public and international telephony. Lack of investment in capacity expansion by BTTB has restricted growth in 21 There are an additional 26,000 lines operated by rural service providers, Bangladesh Rural Telecom Authorities and Sheba Telecom-Rural (see PPIAF 2003). Improving the Rural Investment Climate for Non-Farm Enterprises 305 the supply of telecom services. A major problem of cellular phone networks is that the integration between the cellular and landline networks is limited ­ more than half of the cell phone subscribers are unable to connect to telephones on the BTTB fixedline network. This lack of interconnection capacity is one of the greatest bottlenecks in the expansion of tele-networks managed by the private sector. The current interconnect environment is based on revenue­ sharing agreements between BTTB and the mobile operators. These agreements are based on retail prices and asymmetric payments (BTIB does not compensate mobile operators for call termination, but mobile operators pay BTTB for termination). 100 , . - - - - - - - - - - - - - , eo 1--------: 60 40 20 Other Urban Peri· Metropolitan Area urban (sma) • Informal D Both D Formal Figure 10.6 Borrowing Status of the MSM Enterprises 100 90 eo 70 1: 60 Q) 0 50 Q; 40 a. 30 20 10 0 Rural Other Urban Peri- Metropolitan Area urban (sma) .. Supplier Soredit • Microfinance C Bank Figure 10.7 Sources of Formal Credit The Rural Financial System ­ 'Reaching the Missing Middles' A well functioning rural financial market plays a critical role in both agriculture and rural non-farm growth by mobilising savings and financing working capital and long term investments in farm and non-farm enterprises. The rural financial markets in Bangladesh are characterised by the coexistence of formal and micro-finance institutions and informal sources and continue to be inadequate 306 Transforming Bangladesh into a Middle Income Economy to meet the demand for financial services by a large proportion of the rural population. Because of expansion of microcredit in rural areas, the percentage of firms borrowing for working capital and investment are higher in rural areas compared with rural towns, peri-urban and urban areas (Figure 10.6). For instance, nearly 60 per cent ofrural micro, small and medium (MSM) borrow from formal and informal sources compared with around 50 per cent in urban areas. The percentage of firms not borrowing at all is highest in the peri­ urban areas. Interestingly, access to formal loans including micro-finance is much higher in rural areas whereas firms in urban areas appear to be relying disproportionately on informal loan. Among the firms located in rural areas, smaller urban towns and peri-urban areas, Micro Finance Institutions (MFIs) are the sources of credit for nearly 60 per cent of all firms borrowing from formal sources compared with 38 per cent in urban SMA (Figure 10.7). Nearly half of all firms borrowing from formal sources and located in urban SMAs borrow from the formal banks. Among different sources of informal finance, interest free loan from family predominates (60 per cent of all informal) in all different geographical locations and for firms of different sizes. The micro, small and medium (MSM) enterprises rely mainly on borrowing from the MFIs, whereas access to formal bank is much higher among larger enterprises (Figure 10.8). Indeed, as clear from Figure 10.8, the importance of MFIs declines steadily with an increase in firm size and the opposite pattern is observed in the case of access to bank loans. The MFIs are the main source of finance for about 60 per cent of the firms of size one compared with 20 per cent for firms ofsize 20 and over. This pattern is observed in other developing countries as well. The lack of information on loan sizes precludes any analysis on the size of loan advanced by MFIs and the formal banks. The loan size for MFls are usually small with most loans ranging between dollar 100 dollar150. The average loan size from formal banks is much larger. Thus, while the MFIs play an important role in providing credit 45 40 35 30 25 20 15 10 5 o I_ Supplier's Credit _ Microfinance a Bank I Figure 10.8 Formal Credit by Enterprise Size Improving the Rural Investment Climate for Non-Farm Enterprises 307 to the smaller firms, their overall contribution to total rural credit supply is likely to be much smaller than their share in the total number of borrowing firms. The credit constraint seems to affect the small and medium sized firms disproportionately. While there are no significant differences in the percentage of firms not borrowing across firm sizes, it does not necessarily imply that the non-borrowing firms are credit constrained. Similarly, firms borrowing from both formal and informal sources can be credit constrained even though they had access to finance. In order to find out the extent of credit constraint firms, direct responses of the firms to questions about the severity of the problem of accessing finance are used. The regression results from ranking of the severity of finance problem suggest that small and medium sized firms tend to be more credit constrained in the sense they complain disproportionately more about it. A detailed analysis of data on reason for not borrowing also indicates that a large fraction of firms are indeed credit constrained, as about 60 per cent of those not borrowing would like to borrow. The transaction costs of providing and receiving financial services in rural areas are high, from the point of view of both lenders and borrowers, thus contributing to thinness of rural financial markets. From the point of view of the banks, including formal banks, reaching out to rural customers involves high transaction costs because of smaller loan sizes, more dispersed geographical coverage, lack of information about potential borrowers, high risk of default, and difficulty of enforcing contracts in the case of default. Bangladesh's vulnerability to frequent natural disasters such as floods, droughts, and cyclones also increases the income volatility in rural areas, raising the risk of default. From the borrowers' side, a significant proportion of the firms surveyed 22 reported that they did not borrow because of uncertainty about their credit rating, their ability to repay, complicated bank forms, high interest rates, lack of collateral, and lack of knowledge about finance opportunities (Figure 10.9). For instance, 16 per cent of the firms did not borrow because either they did not know where to go for borrowing or because bank forms were too complicated. About 12 per cent were priced out because of lack of collateral and too high interest rate. Interestingly, a larger share of the small firms (size of 4 and 5) complained about high interest rate (12 per cent) compared with about 6 per cent of the firms of all sizes complaining about it. This confirms the findings from other studies that it is this 'missing middle' who are often priced out by MFIs because of high interest rate. The performance of public sector banks catering the rural areas has been poor, raising serious doubts about their long term sustainability in the absence of significant reforms. Bangladesh Krishi Bank (BKB), and Rajshahi 22 ICG and MIDAS (2003). 308 Transforming Bangladesh into a Middle Income Economy 100.0 90.0 80.0 70.0 60.0 50.0 40.0 ~iiii.i!iiiiiiiiiiiii.iiiii J 30.0 20.0 10.0 o+-----~~----~----~ 2 3 4 5 6-10 11-20 o Don't need credit • No collateral o Don't know where to apply o Bank forms too complicated • Don't apply, afraid • Sure wouldn't get credit can't pay back • Interest rate/charges 100 high Figure 10.9 Reasons for Not Getting Credit, by Enterprise Size Krishi Unnanyan Bank (RKUB), two public sector banks, have a large rural network, with 1,155 branches in 2000, comparable to the branch network of Grameen Bank (1,160). These banks, along with other private and public banks, disbursed about dollar 1.9 billion in rural areas in 1998-99,23 though the expansion of rural credit was slower (5.8 per cent rate) than overall credit expansion (8.5 per cent) during the second half of the 1990s. Weak internal governance, persistent political interference, loan write-offs (undermining credit discipline), lack of a legal and regulatory framework for enforcing contracts and loan recovery, poor management systems, and lack of investment in human capital have all compounded the financial fragility of these public sector banks. These banks have been used more as a window for disbursing credit than as institutions to promote financial intermediation. The government of Bangladesh has committed to reforming the nationalised commercial banks, but the existing reform programme does not include these two agricultural banks. As opposed to the public sector agricultural banks, the outreach of the MFIs and microcredit programmes administered by the NGOs has grown considerably, but they are not still able to meet the need of an important segment of rural entrepreneurs, i.e. the middles. The microcredit programmes, including the two MFIs (Grameen Bank and BRAC 24 Bank), have about 15 million active borrowers/members. However, the average size of loans from the MFls remains low between dollar 100 dollar] 50. The repayment schedule is also rigid - repayment starts within a week of a loan disbursement. Moreover, interest rates charged by the MFls, though not sufficient to cover their costs, are high enough to discourage larger and long term borrowing. Because of its 23 The disbursement figure also includes disbursement by Grameen Bank. 24 BRAC was formerly known as Bangladesh Rural Advancement Committee. Improving the Rural Investment Climate for Non-Farm Enterprises 309 targeting ofpoorer households, medium-sized enterprises and marginally richer households are often not eligible for these loans. However, some ofthe MFIs, notably BRAC Bank, Grameen Bank, and ASA, are moving toward flexible loan instruments with longer duration and larger size, in order to reach small and medium enterprises. Despite this progress, the sustainability of a large number ofMFIs and lack of a legal framework to regulate these institutions are two major issues that need to be resolved. Investing in Flood and Natural Disaster Management Most of Bangladesh lies in the gangetic floodplain which is characterised by low elevation. Three-quarters ofthe country lie below 10 meters - even areas far inland along the upper Meghna river and only about 5 per cent rises above 100 meters. Almost all ofthese higher elevations are in the southeastern Chittagong hills. In a normal year, 30 per cent ofthe country gets flooded, but often floods are much more severe. Sixty per cent ofthe country experiences a flood every hundred or so years. The most flood-prone areas are along the Ganges-Brahmaputra-Meghna (GBM) river system, the northeastern region just west ofSylhet, and the low lying coastal regions that are subject to tidal surges and cyclones. Bangladesh suffered from major floods in 1954,1955,1974,1987,1988 and 1998. The major flood in 1998 inundated two-thirds of the country and affected some 30 million people, and caused over 1,100 deaths. The flood also severely damaged an estimated 15,000 kilometres of roads, 14,000 schools, hundreds ofbridges and culverts, and close to 5,00,000 homes. Like, floods, cyclones occur annually in low lying coastal regions and cause widespread devastation and loss of life. Major cyclones have occurred in November 1970, May 1985, November 1988, and Apri11991. The cyclone of 1991 caused 1,40,000 deaths and severe damage to the eastern coast of the Bay of Bengal. Because of the loss oflife and severe damage to infrastructure, crops and the overall economy, management and mitigation of natural disasters (floods and cyclones) is a major priority in Bangladesh. Since Independence in 1971, the Bangladesh government, aided by its development partners, made large investments in both structural and non-structural measures to protect from floods and cyclones. Structural measures include aspects such as embankments, compartmentalisation with controlled flooding, and polders in coastal areas. Non-structural measures include aspects such as improved flood forecasting and warning systems, evacuation, education, flood proofing such as village mounts, better construction, flood insurance and floodplain zoning. Such investment in flood and cyclone protection measures have been effective in limiting the damages from these disasters in recent floods and 310 Transforming Bangladesh into a Middle Income Economy cyclones. Yet, natural disaster management requires continuous efforts from the government as embankments and other physical infrastructure need continuous maintenance and rehabilitation. The inadequate provision for operation and maintenance, along with the institutional inefficiencies of the Bangladesh Water Development Board (BWDB), threaten the sustainability of embankments and other physical infrastructure, which require frequent rehabilitation. Past structural projects have not involved local stakeholders in any major way, leading to local opposition and often resulting in physical damage to the new installations by aggrieved groups. In addition, environmental considerations have not received adequate attention in water management particularly in the construction and maintenance of flood management structures. SECTION V: CONCLUSIONS The MSM enterprises engaged in non-farm activities have made significant contribution to Bangladesh economy both in terms ofemployment and income. While most of these enterprises are small, generating self-employment for their proprietors, and are mostly engaged in trading activities, they are not engaged in residual activities providing supplementary employment following agricultural production cycle. Most enterprises are in fact full time engagement for their owners and employees. The age and size profiles of the enterprises suggest that non-farm activities provide rich breeding grounds for entrepreneurs in both urban and rural areas. The sector has shown strength over the last decade, both employment and number ofenterprises have grown strongly. While average labour productivity in the sector is higher than agricultural wage, there are large variations in labour productivity and total factor productivity implying existence of constraints in technological upgrading and large entry and exit costs. The investment climate analyses indicate that these smaller enterprises are concerned more with bare bone services and infrastructure. Among many different constraints, enterprises predominantly identified electricity, road and transport, flood and natural disasters, finance and crime and theft as five major problems. Bangladesh's vulnerability to frequent floods and other natural disasters severely hampers operations of more than a third of rural firms. The next most important constraint to RNF growth is the lack ofaccess to electricity, which is available to only 19 per cent of rural households (as compared to 31 per cent of all households). Third, Bangladesh ranks quite high in terms ofroad density, however, because of poor construction ofroads and bridges, lack ofmaintenance of roads and waterways, lack of integration ofdifferent modes oftransportation due to inefficiencies at the container port and in the rail system road conditions and transportation to markets are reported to be severe problems by 36 and 18 per cent ofrural firms, respectively. Improving the Rural Investment Climate for Non-Farm Enterprises 311 In addition, inadequate access to investment finance and to working capital disproportionately affects small and medium-sized fums (the 'missing middles '). The lack ofaccess to telecommunications adversely affects the start up, growth, and performance ofthe micro, small and medium sized (MSM) firms. The effects of the constraints are more severe for rural firms, which accounts for three-quarters of all enterprises in non-farm sector. This ranking ofthe constraints stands in contrast with the ranking observed for larger firms located in two metropolitan cities (Dhaka and Chittagong); the Investment Climate Analysis survey 2003 results suggested a much more damaging role of corruption and governance, with finance ranking well behind corruption and governance. The MSM enterprises in rural areas (and in towns and cities as well) are less concerned with governance issues partly because they remain 'informal' and outside the realm of regulatory and other governance institutions, and partly because most lack access to minimum infrastructure. Indeed, lack of access to basic physical and financial infrastructure seem to impose disproportionate cost on these enterprises particularly those in rural areas. The demographic, geographic and economic reality ofBangladesh puts the onus of generating employment and reducing rural poverty squarely on the growth ofnon-farm activities. In view of the investment climate analysis ofMSM enterprises, the policy options for stimulating future growth of the non-farm sector are clear. Bangladesh has to start providing the very basic physical and financial infrastructure to enterprises in the non-farm sector. Specifically, an enabling rural investment climate will require actions to: • ensure access to basic physical infrastructure, by investing in rural electrification, removing regulatory obstacles to expansion of telecommunications coverage in rural areas, ensuring adequate funding for maintenance of rural road infrastructure, reforming the container port management in Chittagong port, and ensuring integration of inter­ modal transport links; • reform the rural financial system, along with the overall financial sector reforms, to mobilise rural savings and to improve delivery of credit particularly to small and medium sized firms while ensuring credit discipline and the sustainability of financial institutions. Legal reforms will also be necessary to enable greater financial intermediation, the sustainable expansion of microfinance programmes, and the securitisation of transactions; • sustain progress in natural disaster management through continued investment and maintenance and institutional reforms. This set of measures assigns a pivotal role for the government in maintaining and enabling an investment climate in rural areas. Without these 312 Transforming Bangladesh into a Middle Income Economy institutional and policy reforms, the dynamism in the RNF sector observed during the last decade risks being eroded. ACKNOWLEDGEMENTS The author would like to acknowledge excellent and timely assistance from Eric Knepper. The author would also like to thank Lisa Daniels (Team leader for the National Private Sector Enterprises Survey 2003) for her quick and patient responses to many queries about the database, and Frank Matseart (DFID) for all the help in getting access to the database. This paper benefitted from comments from the fellow team members ofthe AAA report and seminar participants. at World Bank. REFERENCES Bangladesh Bureau of Statistics (BBS). Household Expenditure and Income Survey 2000, Dhaka, 2000. - - - . Census ofAgriculture 1996, vol. I, Dhaka, 1999. Barkat, A., et at. Economic and social impact evaluation study ofthe rural electrification program in Bangladesh, Human Development Research Centre (HDRC) for NRECA International Ltd., Rural Power for Poverty Reduction Program, USAID, Dhaka, 2002. Binswanger Hans P. 'Agricultural and Rural Development: Painful Lessons' in Carl K. Eicher and John M. Staatz (eds.) International Agricultural Development, 3rd edition, Baltimore: The Johns Hopkins University Press, 1998. Chowdhury, N. and S. Haggblade. 'Evolving rice and wheat markets', in Out ofthe Shadow ofFamine: Evolving Food Markets and Food Policy in Bangladesh, R. Ahmed, S. Haggblade and T. Chowdhury, Baltimore: Johns Hopkins University Press, 2000. Daniels, Lisa. National Private Sector Survey ofEnterprises in Bangladesh, 2003, Report written for the UK Department for International Development, United States Agency for International Development, Swiss Agency for Development and Cooperation Swedish International Development Cooperation Agency, Dhaka, 2003. Dixie, G. 'Review ofcurrent and future airfreight situation from Bangladesh', Final Report, Accord Associates for Hortex Foundation, Dhaka, 2002. Government of Bangladesh. Bangladesh: A National Strategy for Economic Growth and Poverty Reduction, Ministry of Finance, Economic Relations Division Dhaka, 2003. -----. Bangladesh Integrated Transport Systems Study, Planning Commission, Dhaka, 1998. Improving the Rural Investment Climate for Non-Farm Enterprises 313 ---,. National Water Policy, Dhaka, 1998. Gwilliam, KM. and A. Kumar. 'Road funds revisited: A preliminary appraisal of the effectiveness of "second generation" road funds', Discussion Paper, Transport Division, Washington, DC; World Bank, 2002. Gwilliam, K.M. and Z. Shalizi 'Road funds, user charges and taxes', World Bank Research Observer, 14, no. 2, Washington, DC, 1999. Haggblade, S., P. Hazell and T. Reardon. 'Strategies for stimulating poverty­ alleviating growth in the rural nonfarm economy in developing countries' , EPTD Discussion Paper No.9 2, International Food Policy Institute, Washington, DC, 2002. Hayami Y, edited, Toward the Rural-based Development ofCommerce and Industry: Selected Experiences from East Asia, World Bank, Washington, DC, 1998. Henderson, V. 'Urbanization in developing countries', World Bank Research Observer, volume 17, no. 1, Washington, DC, 2002. Hossain, M. 'Rural non-farm economy in Bangladesh: A view from household surveys', Background Technical Paper prepared for the dialogue on Promoting Rural Non-farm Economy: Is Bangladesh Doing Enough, Center for Policy Dialogue, Dhaka, 2003. International ConSUlting Group and Micro Industries Development Assistance and Service. National Private Sector Survey of Enterprises in Bangladesh, Dhaka, 2003. MandaI, M.A.S. and Asaduzzaman, M. Rural Non-farm Economy in Bangladesh: Characteristics and Issues for Development', BIDS­ DFID Workshop Paper, 22-23, Dhaka, 2002. Shilpi, F. Understanding the Rural Investment Climate: Evidence from Enterprise and Household Surveys, Background Paper, World Bank, 2003. Stem, Nicholas, A Strategy for Development, Washington DC, The World Bank,2002. World Bank. Bangladesh: The Non-Farm Sector in a Diversifying Rural Economy, Report No. 16740-BD, 1997. - - - . Reforming the power sector, Policy Brief#13, Dhaka: World Bank, 2001. - - - . Improving the Investment Climate in Bangladesh, Preliminary Draft. Washington, DC, 2003a. - - - . Bangladesh: Public Expenditure Review, Report no. 24370-BD. Washington, DC, 2003. "-"---. Private Solutions for Infrastructure in Bangladesh, Mimeo, Washington, DC, 2003c. - - - . Bangladesh: Development Policy Review, Report no. 26154-BD. Washington DC, 2003d. 314 Transforming Bangladesh into a Middle Income Economy World Bank. Promoting Agro-Enterprise and Agro-Food Systems Development in Developing and Transition Countries :Towards an Operational Strategy for the World Bank Group, Agriculture and Rural Development Family, Washington, DC, 2003e. 11 Scaling Up of Microfinance Hassan Zaman SECTION I: INTRODUCTION1 The fact that the microfinance industry has been able to provide access to credit, currently, to nearly thirteen million poor households in Bangladesh is truly remarkable. There are around twelve hundred micro finance institutions (MFI) in Bangladesh (CDF 2002) but the industry is dominated by four large MFls (BRAC, Grameen, ASA and Proshika) that serve around 11.5 million or ninety per cent of all clients. 2 After the 'big four', the next largest NG03 has 0.7 million clients and there are only ten NGOs who have more than 100,000 borrowers. The bottom line is that the majority of the MFIs are small (less than five thousand borrowers) and that the bulk of the access to micro­ credit is supplied by the four large MFls. As such, the experiences of scaling up discussed here draw primarily upon these large MFls. This paper is structured as follows. Section II describes the evolution of the microfinance industry over the past three decades in Bangladesh. Section III proposes several factors that contributed to its scaling-up. Section IV summarises the evidence on the impact of microfinance on poverty, vulnerability and female empowerment. Section V concludes by discussing key lessons from the Bangladesh experience. 1 An earlier version of this paper was written as a background paper for the' Scaling Up Poverty Reduction' conference held in Shanghai, China in May 2004. 2 Latest figures indicate that BRAC has 3.5 million borrowers, Grameen Bank has 3.1 million clients, Proshika 2.9 million and ASA 2.1 million clients. 3 Swamivar Bangladesh. 316 Transforming Bangladesh into a Middle Income Economy SECTION II: THE EVOLUTION OF MICROCREDIT INDUSTRY IN BANGLADESH The growth in access to credit by the poor took place in several distinct phases over the last three decades. The origins ofthe current microcredit model can be traced back to action-research in the late 1970s, carried out by academics as well as practitioners in organisations that were created to deal with the relief and rehabilitation needs ofpost-independence Bangladesh. The 1980's witnessed a growing number of non-governmental organisations (NGOs) which experimented with different modalities ofdelivering credit to the poor. The various models converged in the beginning ofthe 1990s towards a fairly uniform 'Grameen-model' ofdelivering microcredit. It sparked a sharp growth in access to microcredit during this decade. In recent years the standard 'Grameen-model' has undergone more refinements in order to cater to different niche markets as well as to different life cycle circumstances. The Seventies Experimentation in providing credit to households considered 'unbankable' by the formal financial system has its origins a few years after Bangladesh's independence war in 1971. The independence movement gave rise to a new generation of young activists who were keen on contributing to the reconstruction ofthis war-ravaged country. The new government and a myriad ofaid agencies that arrived on the scene were unable to cope with the scale of de<;titution non-governmental organisations emerged to meet the challenges. The early years of the NGO movement in Bangladesh focused on relief and rehabilitation with an emphasis on community development. However, by the mid-1970s, two of the NGOs that would subsequently expand in scale, BRAC and Proshika, found that 'elite capture' was a serious impediment to their development objectives. As a result, a separate focus on the poor through a 'target-group' approach was introduced. Moreover, an ideological debate within both these organisations began to brew, between those who favoured 'economic tools' (credit, savings, etc.) to support poverty reduction and those who believed that social mobilisation against existing injustices would suffice and financial services were unnecessary. Around the same time a team of researchers at Chittagong University, led by Professor Yunus, began an action-research programme that provided loans to poor households in a few villages. Borrowers were mobilised in 'peer groups' composed offour-five individuals who were jointly responsible for each other's repayment. Several of these small 'peer monitoring groups' were organised into a larger unit which would meet weekly with the primary purpose of repaying loan instalments. The process oftrial and error included combining males and females in the same credit group and then changing this to separate groups divided by gender. It also included forming Scaling up of Microfinance 317 'occupational groups' but this was dropped in favour ofvillage-based groups. The demand for loans grew rapidly and Professor Yunus enlisted the support of the Bangladesh Bank and commercial banks to provide the 'Grameen Project' - as it was then called - with resources. The success of this experimentation paved the way for the establishment of the Grameen Bank under a special ordinance in 1983. The Eighties In the early 1980's several NGOs experimented with different ways of delivering credit. One important mode tested was the efficacy of providing loans for group projects compared to offering loans to individuals with peer monitoring. The broad lesson was that the latter was more effective due to incentives and 'free-rider' problems compared with lending to a group. Hence by the late 1980s the predominant model became one ofproviding individual loans to a target group of poor households, with peer monitoring and strong MFI staff follow-up. The Association for Social Advancement (ASA) is a classic example of this shift. Its initial emphasis was on forming 'peoples organisations' mobilised for social action against oppression. It changed to the 'target group' approach and then towards the provision of financial services in the late 1980s. Now ASA is the fourth largest MFI in Bangladesh in terms ofclients and its unique low-cost credit delivery mechanism is being replicated in several other countries. ASA keeps paperwork requirements to a minimum, has decentralised most decision-making to the field and overall has a very lean operation (Choudhury 2003). The 1980's and early 1990's was also important in the development of management capacity within several of the large MFIs which allowed them to expand their microcredit programmes. What is particularly interesting is that the development of the know-how and confidence to implement large programmes arose, in some cases, from the experience of scaling-up programmes not related to microcredit. For instance in the case ofBRAC, the first major experience with a nationwide programme came about when it implemented an oral rehydration programme to combat diarrhoeal disease. Thirteen million women were trained to use a simple but effective rehydration solution and BRAC staff were paid based on how many oftheir trainees used and retained this knowledge. 4 4 In addition to this innovative staff incentive system, a detailed evaluation ofthe oral rehydration experience also point to a number of other 'success factors' - (i) systematic recruitment and training of staff, (ii) an effective feedback loop and the willingness of senior management to learn from the lessons from the field and (iii) support from government, donors and professional experts. (Chowdhury and Cash 1996). 318 Transforming Bangladesh into a Middle Income Economy Early to Mid-1990s The early 1990s was the period of rapid expansion of the Grameen-style microcredit approach (Ahmed 2003). The growth was fueled largely by a 'franchising approach' whereby new branches replicated the procedures and norms that prevailed in existing branches. The product that was offered to the client at the time was fairly narrow, focusing mainly on a standard microcredit package offered to all clients. The view was that it was easier to recruit new staff and train them quickly in a simple product during a phase when branches were opened at a rapid rate. This growth was clearly aided by the high population density and relative ethnic, social and cultural homogeneity in Bangladesh. A notable shift that took place during this expansion phase was a greater emphasis on individual borrower accountability for loan repayment and less reliance on peer monitoring. Staff follow-up of loans became more rigorous and professional with the use of computerised management information systems. Donor funds also contributed to expanding the revolving loan funds for MFls during this expansion phase. Moreover this period saw the emergence ofPKSF as a wholesale financing institution (see Section III). Following this expansion, a geographical mapping ofmicro finance suggests that all districts in Bangladesh have microcredit services, though there are many smaller pockets with little or no coverage (e.g. Chittagong Hill Tracts). A closer look shows that there is somewhat greater coverage of poor households in the central and western districts. The south-east and pockets ofthe north-east are areas with room for more expansion (PKSF 2003).5 Mid-1990s Onwards Feedback from the field, academic research and international experience contributed to an increasing emphasis on providing diversified financial services for different groups of households from the mid-1990s onwards. The benefits of a narrow focus on microcredit during the 'expansion phase' was that it kept costs low, operations transparent and relatively straightforward management oversight. However, it became clear that the standard Grameen model ofproviding microcredit with fixed repayment schedules, with standard floors and ceilings on loan sizes, was not sufficient to meet the needs of the extreme poor or the 'vulnerable non-poor' group. Moreover, existing microcredit borrowers also required complementary financial and non-financial services. The standard practice for MFls until the late 1990s was to collect compulsory weekly savings from their clients, holding the money as a de facto lump sum 'pension' returned when a client left the organisation. Access to these deposits was otherwise limited, which curtailed 5 PKSF (2003) 'Maps on Microcredit: Coverage in Bangladesh' Dhaka. Scaling up of Microfinance 319 a potentially important source ofconsumption-smoothing. Recognising these limitations an increasing number of MFIs in Bangladesh have introduced an open access current account scheme in addition to the fixed deposit scheme. Moreover, many MFls have life insurance products whereby, outstanding microcredit debts are written off and other benefits are paid following the death ofa borrower. Non-credit services can also take the form ofinput supply, skill training and marketing support for micro-entrepreneurs. 6 A complementary package to microcredit can also take the form of providing education for the children of borrowers. Grameen Bank for instance has a scholarship programme for female secondary education and a student loan programme for tertiary education. Similarly, many MFls have community health programmes, legal literacy training and provide information on accessing local resources. MFls began to experiment with catering to new niche markets as the traditional microcredit business became standardised (and horizontal expansion slowed) and required less attention. For instance several NGOs began providing larger loans to 'graduate' microcredit borrowers and in some cases to households who were not part of the microcredit system but which wanted a micro-enterprise loan. These loans typically range from taka 20,000 (around dollar 320) to taka 200,000 (dollar 3,200). Innovative solutions are also emerging to address the problem of access to finance for the small enterprise sector. For instance, BRAC established a separate financial institution, BRAC Bank, that focuses on lending to the 'smaller end' of the small enterprise sector, with loans averaging taka 400,000. Moreover, evaluation studies pointed out that extreme poor households were struggling to benefit from the standard microcredit model, even if they joined the programmes. The main reasons were: (i) minimum loan floors for a first loan that sometimes exceeded their own perceived needs; (ii) fixed weekly loan repayments could be difficult to commit to in light of sharp seasonality of income; (iii) other members of peer-monitored groups sometimes do not wish to guarantee loans for extreme poor households, etc.; (iv) residing in remote or depressed areas. Programmes that have been developed to cater for these constraints include: (i) introducing more flexible repayment schedules such as ASA's Flexible Loan Programme; (ii) lowering first loan floors so that amounts as small as taka 500 (dollar 9) can be borrowed; (iii) Grameen's programme that offers zero interest loans to beggars; (iv) the Resource Integration Centre's programme that specialises in offering loans to a specific vulnerable group 6 For instance in the sericulture sector BRAe supplies the eggs to the silkworm rearer, plants the mulbery trees, trains the entrepreneur in silk rearing, arranges for extension services by a BRAe rearing specialist, purchases the cocoons from the rearer from her homestead and supplies these to a BRAe silk reeling centre. 320 Transforming Bangladesh into a Middle Income Economy the elderly poor; (v) various programmes that combine food aid with microcredit and training, e.g. BRAC's IGVGD programme; (vi) targeting remote areas through, for instance, ASA's cost-effective mini-branch system and Integrated Development Foundations work in the Chittagong Hill Tracts. SECTION III: FACTORS THAT LED TO SCALING UP Institution Building - Leadership. Staff Incentives and Learning by DOing It is unquestionable that the vision and persistence ofthe leaders ofthe NGO/ MFI movement are a key factor behind the success ofthe microfinance industry in Bangladesh. Leadership skills were instrumental at the initial stages in persuading a skeptical public that providing credit to the poor could become a viable and replicable proposition. These skills were equally important during the process of scaling up. These included being able to recruit and motivate staff, decentralising authority away from the centre, building management information systems and internal controls as well as having the humility to learn from mistakes. Moreover, once the leaders ofthese organisations showed that scaling up is possible, this view became part ofthe staff culture and other staff had the confidence to expand the programmes within their own jurisdiction. Staff recruitment, motivation and retention are particularly important for large organisations - BRAC for instance employs around 28,000 staff in its various programmes, Grameen has around 12,000 in its microcredit programme and ASA's microcredit programme employs around 8,000 staff. A critical element in this process is an objective performance evaluation system for staff that is linked to career mobility and a myriad of other incentives for staff to perform well both individually and in their teams. For instance Grameen Bank has introduced a system ofrating branch offices based on the achievement ofspecific targets which not only include standard loan recovery, but also factor in social indicators such as the proportion ofGrameen children going to school. Staff motivation is also enhanced by decentralising significant amounts of responsibility to the lower tiers of the administrative structure. ASA is the best example ofa lean credit delivery structure with high levels of decision-making authority given to field offices, from loan sanctioning decisions to staff human resource issues. Moreover, the structure within field offices is relatively flat with a branch manager who works with individual field workers to resolve problems and typically shares living quarters with other field staff (Jain and Moore 2003). Effective internal controls are also important in ensuring effective staff performance. First of all the fact that financial transactions are carried out publicly, in the weekly meetings and in the branch offices, is a major check Scaling up of Microfinance 321 against any form of discretionary behaviour by field workers. Many NGOs, particularly the ones that have successfully expanded in scale, have developed measures that include frequently rotating staff within and between branches, regular field visits by senior management, a strong internal audit team and annual external audits. A fundamental part of the scaling up of Bangladesh's NGOs and more specifically the microfinance movement, has been the ability to learn from experiences and adapt programmes accordingly. This learning process takes place both through informal feedback by field staff during regular interactions with management as well as through a formal monitoring and evaluation process. BRAC's Research and Evaluation Division has around twenty professionals whose key function is to evaluate BRAC's multi-dimensional programmes and give timely feedback to programme staff and management. This process offeedback occurs through longer term research as well as 'quick­ turnaround' assessments. The shift to more flexible financial services that took place in recent years is largely based on adapting to client feedback and analysis of the limitations of a uniform microcredit model. Many of these lessons have also been shared with the rest of the micro­ finance industry as there are many examples of smaller MFIs being managed by former employees ofthe large MFIs. A Constructive Donor-client Relationship External resources played an important part in the experimentation process, subsequent growth in outreach and institutional strengthening of the micro­ finance industry. At the same time, the large microfinance institutions have been successful in 'managing donors'. Northern NGOs, such as the Ford Foundation, Oxfam and the Aga Khan Foundation, played an important role at the initial stages of the NGO-MFI industry in Bangladesh. The subsequent expansion and consolidation stage was funded largely by official bilateral agencies? and later by multilateral agencies once northern NGOs could not match the growing resource requirements of the larger MFIs. A large part of these donor investments went to the capitalisation of MFIs loan funds, crucial to the rapid expansion that took place in the 1990s, as well as into developing institutional capacity through management information systems and human resource development. Finally, the 1990s have seen dependence on donor resources progressively declining for the large MFIs Grameen Bank, ASA and BRAe do not receive any grant financing for their microcredit operations. Moreover, out ofBRAC's ? The Department for International Development (DFID) has been one of the largest of the bilateral donors having provided around dollar 130 million over a twenty year period to MFIs in Bangladesh. 322 Transforming Bangladesh into a Middle Income Economy total dollar 160 million expenditure on development programmes in 2002, more than 80 per cent was financed from its own resources, through the interest income on microcredit as well as surplus from its commercial enterprises. Two facets of these trends are worth highlighting. First, the decisions to subsidise these operations were not free from controversy. The advocates for funding these loan funds had to argue their case with officials within their own agencies who believed that the capital base for loan operations ought to be enhanced only by savings mobilisation or borrowing from commercial sources. In retrospect, these decisions to help contribute to MFIs loan funds were by and large correct as almost all of the MFIs that received this support have either reached financial self-sufficiency or are well on their way to doing so. Donors also invested in organisational systems and MFI staff training in order to strengthen the capacity to administer these growing programmes. Second, large NGOs in particular have been reasonably successful in 'managing donors'. Investing in the capital base of expanding MFIs with a strategic sustainability plan required donors with a long term vision but crucially required MFls who were able to persuade donors of the soundness of this strategy. In other countries MFIs have not been able to provide this type of strategic view on the best use of donor resources with the result that these resources have often been used to either subsidise final on-lending rates or support MFIs without a solid business plan. BRAC, with its large multi­ faceted programmes, has had a long history of working with donors and the evolution of this relationship is worth highlighting. Donors who have their own incentives to commit resources and demonstrate results on the ground, have been eager to provide resources to implementing organisations with a proven track record. Hence the likes ofBRAC, have had to deal with multiple donors where each wanted to fund specific projects. These uncoordinated donor missions and disparate disbursement and reporting arrangements taxed BRAC's internal capacity and led to its management proposing changes for how its donors ought to operate. In the early nineties, donors shifted their approach from financing specific BRAC projects to financing BRAC programmes. Donors also formed a 'consortium' that pooled funds, negotiated jointly with BRAC and had common reporting requirements. An important part of the consortium funding arrangement and the move towards programme funding has been to improve the predictability of resource flows for instance BRAC secured financing for its Rural Development Programme for a five year period from the donor consortium. Moreover, the establishment of a donor liaison office for BRAC also acts as a buffer between BRAC staff and the various visitors, consultants and evaluators. Scaling up of Microfinance 323 An Enabling Macro-economic and Regulatory Environment The early experimentation and later scaling-up of the microfinance industry in Bangladesh was helped by an appropriate 'enabling environment'. First the macro-economy has been, by and large, soundly managed and one should not underestimate the significance ofthis. The rate of inflation has been kept to single digits and economic growth over the past decade has averaged around 5 per cent per annum, thereby creating economic opportunities for microcredit financed investments. Second, the government of Bangladesh has thus far maintained a balanced approach towards regulating and supervising the activities of the NGO sector. This has been critical in ensuring the operational flexibility that is the cornerstone of service delivery by NGOs. While this long relationship has not been free from tensions on both sides, the government ofBangladesh has thus far been able to place the interests of the poor foremost in its mind while dealing with NGO issues. A less charitable view is that the scaling up ofNGOs went largely unnoticed and once this took place the combined clout of large NGOs and donors has led to the government taking on a largely laissez-faire approach. Ultimately the relationship between government and NGOs also depends on individual personalities and social ties (Hossain 2003) as there have always been widely varying views regarding NGOs within the civil service and the cabinet. Individuals in key positions within government have time and again proved instrumental in facilitating the growth of the microcredit sector. The early development of the Grameen project, its registration as a bank and the decision to grant it managerial autonomy are clear examples (Yunus 1999), as was the establishment ofPKSF with a strong autonomous board. The prevailing consensus position is supportive ofNGOs though accusations of involvement in party politics by a handful of NGOs has strained the overall government-NGO relationship of late. Looking forward, it is clear that both the overall umbrella legislation governing NGO activities and the regulatory framework for micro finance needs to be strengthened, particularly in light ofthe large amounts ofdeposits mobilised for the poor. The Central Bank, PKSF and representatives ofMFIs are currently working to produce a set of guidelines and standards that will strengthen the regulatory framework for microfinance. Population Density, Ethnic Homogeneity and Religious Tolerance Aside from the Chittagong Hill Tracts area, Bangladesh is ethnically a relatively homogenous country, with a high population density and good communication networks. Moreover, this largely homogenous market is also very large in absolute terms. The contrast with Nepal for instance is striking 324 Transforming Bangladesh into a Middle Income Economy and the difference in microcredit outreach between these two countries is partly due to these factors. It is also striking how in Pakistan, Afghanistan, Egypt and certain other Muslim countries, MFIs have found religion to be a factor that has led to a relatively lower demand for microcredit and MFIs who are more cautious about expanding. 8 In contrast, even conservative religious forces in Bangladesh have been largely tolerant of microfinance activities and the greater economic empowerment and mobility of women. A professional Apex Body for Microfinance The Palli Karma Sahayak Foundation (PKSF) was created in 1990 and is governed by a board composed of both public and private sector representatives. It is a pUblic-private apex body that channels funds for micro­ finance to MFIs has been critical in the expansion and improved professionalism of the microcredit industry in Bangladesh. PKSF's core functions include: (i) lending money to MFIs which meet certain eligibility criteria to expand their microfinance operations, (ii) capacity building and hands-on assistance to strengthen MFIs and move them towards financial sustainability, (iii) advocacy on microfinance issues and helping develop an appropriate regulatory framework for the industry. PKSF has been instrumental in contributing to the sharp increase in access in microcredit that took place in the 1990s by expanding the capital base for MFIs to on-lend to the poor. For instance, as of December 2003, PKSF loans constitute around 30 per cent of ASA's current revolving loan fund. PKSF is also widely credited for sharpening the focus of many MFIs towards financial sustainability and also in setting appropriate standards that will ease the way for a strengthened regulatory structure for microfinance. Moreover, PKSF's funding arrangements are in line with the MFls cashflow requirements and the transaction costs of dealing with PKSF are kept to a minimum. There is a growing experience with setting up apex institutions worldwide, e.g. PPAF in Pakistan, RMDC in Nepal, FONCAP in Argentina, LID in Bosnia-Herzegovina and MISFA in Afghanistan. One of the fundamental factors behind the success or failure ofan apex is the underlying retail capacity in a particular country. The overall strength of the MFls in Bangladesh has been key to PKSF's success. An overestimation ofthe capacity to absorb funds by the MFIs on the ground is likely to lead to a failure of an apex body. However, ifa realistic assessment ofthe underlying retail capacity is made, then apexes offer many benefits such as the ability to screen MFIs on standard criteria and creating a 'level playing field'. 8 This point was made by Stephen Rasmussen in a personal communication. Scaling up of Microfinance 325 SECTION IV: THE IMPACT OF MICROFINANCE IN BANGLADESH The evidence on the impact of microcredit can be assessed from two inter­ related angles. Firstly, who does credit reach and secondly, how does it affect the welfare of different groups of individuals and households? Land ceilings, occupational criteria and asset valuations are standard targeting tools used by microcredit providers in Bangladesh in order to direct resources to the rural poor. These indicators have been shown to be relatively accurate correlates ofpoverty by programme administrators who do not have the time, resources or expertise to carry out more sophisticated calculations of poverty for each household in their targeted area. In practice, the land criterion is the one that is more closely adhered to in the field. A large proportion of extremely-poor households, measured by initiallandho Iding size, join microcredit programmes Zaman 1999, Khandker 2003). For instance in Khandker's sample, sixty per cent of the sampled members have less than twenty decimals of land. However, several studies also show that between 15 and 30 per cent of microcredit members are from 'non-target' households measured in terms of land (Mustafa et al. 1996, Montgomery et al. 1996, Zaman 1999, Khandker 1998).9 However, these households are typically marginal farmers and can be considered part of the 'vulnerable non-poor' group, prone to transient bouts of poverty. There is also evidence which suggests that households who join microcredit programmes a few years after the village group has been established tend to be less poor compared to the members who join at the start of the programme (Matin 1998). This feature of better-off households joining over time has also been noted as a general rule of thumb in many targeted anti-poverty programmes worldwide (Lipton 1996). The presence of wealthier households do not appear to affect the credit supply to poor households though there is evidence to suggest that poorer households use a larger share of their loans for consumption purposes compared to better-off households. The bottom line is that the literature on targeting suggests that microfinance programmes are reasonably successful at reaching the poor, and that those households who fall above the stipulated landholding criterion tend to be marginally above the poverty line and are susceptible to transient poverty in certain years. The literature broadly supports the hypothesis that access to microcredit contributes to poverty reduction in Bangladesh though the evidence is not 9 It is interesting to compare this figure with Copestake 's (1992) evaluation ofIndia's Integrated Rural Development Project (lRDP) where the proportion of non-poor households ranged up to 36 per cent. 326 Transforming Bangladesh into a Middle Income Economy entirely clear-cut. lo For instance, data collected by the World Bank in 1992 have been used to show widely varying results depending on the methodology chosen to assess impact. Khandker( 1998) estimates that for every taka 100 lent to a woman, household consumption increases by taka 18, interestingly the figure is taka 11 if the same amount was lent to a man. Moderate poverty falls by around 15 per cent and ultra-poverty by 25 per cent for households who have been BRAC members for up to three years controlling for other factors according to the author. Similar results are found for Grameen Bank and Bangladesh Rural Development Board (BRDB) members. On the other hand, using the, same data and a different way of correcting for selectivity bias, Morduch (1998) finds that microcredit does not have a significant impact on consumption levels and therefore, on income poverty. Consumption data from 1,072 households in one district of Bangladesh is used to show that the largest effect on poverty occurs when a moderate-poor BRAC client borrows more than taka 10,000 (dollar 200) in cumulative loans (Zaman 1998). In other words, there may be a threshold level ofcredit above which a household gains most in terms of increases in income. Recent evidence from are-survey ofthe same households suggests that microcredit has significantly contributed to reducing poverty (Khandker 2003). Somewhat surprisingly, the impact appears to be greater for households who started off extremely poor (18 percentage point drop in extreme poverty in seven years) compared to moderate poor households (8.5 percentage point drop). These results differ from earlier evidence that pointed to moderate poor borrowers benefitting more than extremely poor borrowers due to the fact that the poorest have a number of constraints (fewer income sources, worse health and education, etc.) which prevent them from investing the loan in a high-return activity (Wood and Sharif 1997). This feature of better-off households benefitting more was also borne out by detailed case-study evidence (Farashuddin et al. 1998) and by comparing participants of credit programmes who cater to different socio-economic groups (Montgomery et al. 1996).11 10 The methodological problems associated with impact assessment ofmicrocredit are complex. The literature typically uses 'control groups', usually 'eligible non-members' or 'recently joined members' in order to address the problem of the counter-factual. There have been attempts to cater for the 'selectivity bias' problem but with varying degrees of success. t I Montgomery et al. compare the performance of BRAC borrowers with the borrowers from a government-run microcredit scheme, the Thana Resource Development and Employment Programme (TRDEP). TRDEP's borrowers' initial endowment conditions is shown to be higher than BRAC's (average pre-loan landholding is 46 and 30 decimals for TRDEP and BRAC members respectively and the percentage of income derived from daily labour is 5 per cent and 32 per cent respectively) whilst the credit-delivery mechanism and average loan size are broadly speaking very similar. The typical TRDEP borrower's increase in assets and income during the course of the most recent loan is higher than BRAC's giving rise to the author's contention that better-off borrowers benefit more than poorer borrowers. Scaling up of Microfinance 327 The Bangladesh Institute of Development Studies (BIDS) carried out an extensive study of the impact ofPKSF POs microcredit programme using longitudinal data of three thousand households between 1997 and 2000. One of the key findings was that microcredit has a positive and significant effect on poverty status o/the programme households .... (BIDS 2001 page 155). The study also finds that microcredit members are less vulnerable when struck by crises. Moreover improvements in other social indicators (child immunization, use ofsanitary latrines, contraceptive prevalence) are also more noticeable for microcredit programme members compared to non-members. There has been limited work on the aggregate poverty reduction impact of microcredit at the local or national level in Bangladesh. Khandker (2003) uses the panel data discussed above to suggest that there is some positive externality due to microcredit programmes but that the overall spillover benefits are somewhat limited. For instance, the net contribution ofmicrocredit on moderate poverty for non-participants is a small decline of 1.1 percentage points between 1991-92 and 1998-99 compared to a decline of8.5 percentage points by borrowers in the same village. The impact on extreme poverty is estimated to be somewhat greater - 4.8 percentage points for non-borrowers and 18.2 percentage points for borrowers over this seven year period. There is strong evidence that microcredit contributes to reducing household vulnerability. Morduch shows that consumption variability is 47 per cent lower for eligible 12 Grameen households, 54 per cent lower for eligible BRAe households and 51 per cent lower for eligible BRDB households compared to a control group.13 This consumption smoothing is driven by income smoothing as evidenced by the significantly lower labour supply variability experienced by microcredit members compared to the control group.14 The importance of this result cannot be over-emphasised given the fact that seasonal deficits playa key part in the poverty process in Bangladesh (Rahman 1995). Essentially Morduch's results indicate that programme participants do not benefit in terms of greater consumption levels, but they participate because they benefit from risk reduction. Asset creation is important to reduce household vulnerability to various livelihood risks. The findings of an impact assessment of ASA borrowers conducted in 2003 suggests that the average value ofphysical assets increased by 127 per cent in rural areas and grew by about 150 per cent in urban areas over a five year period. Moreover, the average increase in cash savings rose by 133 per cent and III per cent in rural and urban areas respectively over 12 Morduch only includes households who fulfil the targeting criteria ofthe three organisations and labels them 'eligible households'. 13 These results are statistically significant at the 95 per cent level. 14 Morduch's estimates of labour supply variability is 39-46 per cent lower for microcredit members compared to a control group. 328 Transforming Bangladesh into a Middle Income Economy this same five year period. Similar evidence is found in studies of BRAC, Grameen and PKSF's partner organisations. Another pathway by which microfinance appears to reduce vulnerability is through the emergency assistance provided by many microfinance organisations during periods ofacute natural disasters such as the recent floods in Bangladesh. The fact that these organisations tum into de facto relief agencies is crucial in sustaining these households in the immediate aftermath of a natural disaster. Moreover, the post-disaster rehabilitation assistance, in terms ofboth financial and other services, is also highly valued by microcredit clients. The pathways by which microcredit reduces vulnerability, that have been discussed here, relate to income and consumption smoothing and asset building. However, the impact of credit on female empowerment, or a reduction in 'female vulnerability' has also received considerable attention. female empowerment in Bangladesh, can be viewed against the backdrop of 'patriarchy', defined by Cain et al. (1979) as a set ofsocial relations with a material base that enables men to dominate women and hence can be thought of in terms of an improvement in intra-household gender relations (Naved 199' , Hashemi et at. 1996). Moreover, given the institution ofpurdah (loosely translated as 'veil'), a pervasive social construct which restricts the female sphere within a typical Bangladeshi household, 'female empowerment' can a also be view~d in terms of woman's interactions outside the homestead and the acquisition of skills, knowledge and confidence that such interactions can bring (Amin et al. op.cit., White 1992, Mahmud 1994). Amin et at (1994) work in thirty-six villages in Bangladesh showed that membership in microcredit programmes positively affected a woman's decision-making role, her marital stability, her control over resources and mobility but has less impact on her attitude regarding marriage and education of their daughters. Naved (1994)15 finds that the women credit-programme participants in her sample felt their status had improved due to the fact that they were seen as income earners for the family through their access to credit. Hashemi et at. (1996) develop an 'empowerment index' based on eight empowerment indicators. Their analysis establishes that contributing to her household's income is a significant contributing factor to a woman's own empowerment. However, Hashemi et al. also show that credit programmes can empower women independently of whether they contribute to family income or not, after controlled for other factors. The focus ofthose skeptical about the empowering effect of microcredit has been on the issue of women's control over loans. Goetz et al. (1996) used 15 Naved (1994) uses Participatory Rural Appraisal (PRA) techniques to identify the effect of participation in Save The Children's savings and credit programme in Manikanj. Scaling up of Microfinance 329 a sample of 253 female borrowers from four rural credit providers in Bangladesh. Their investigation of loan histories led the authors to conclude that: About 63 per cent ofthe cases fall into the three categories ofpartial, very limited or no control indicating a fairly significant pattern of loss of direct control over credit. The authors disaggregated their data in terms of loan activity and concluded that investing in traditional women's work increased their chances of being able to control the loan. Montgomery et al. (1996) also have reservations about the 'empowering effect' of microcredit. Their argument is based largely on secondary sources and a small field survey focusing on the issue of control over loans. Whilst the authors admit that their sample is small, they on balance support Goetz et al. (op.cit.) view that microcredit reinforces existing gender patterns and inequalities by promoting traditional income generation activities, which they believe do little to alter the social status quo. On the whole, the evidence presented by those who argue that microcredit improves female status within a household appears more convincing than that argued by the 'skeptics' camp. There are two main reasons for this contention. First, the underlying thread ofthe 'positive' argument, that access to an important household resource (credit) enhances a female's status within the household is both intuitively appealing and resonates with the theoretical literature on bargaining models ofthe household (Lundberg and Pollak 1993). Second, the focus on female control over loans as a key component of the 'skeptics' argument fails to recognise that credit enters the overall household income pool and that household members jointly participate in the loan investment. The role of several family members in managing the loan-fmanced investment is now explicitly recognised by MFls who now give loans to women for 'male-activities' such as rickshaw loans, that many did not permit in earlier days. SECTION V: CONCLUDING LESSONS AND FUTURE CHALLENGES Lessons This paper proposes six lessons from Bangladesh that could be relevant to micro finance growth and impact in other countries. First, is the importance of an 'enabling environment' for microfinance. A critical part is maintaining a stable macroeconomic environment with both interest rates and inflation kept at reasonable levels. The lack ofmacro-stability has seriously constrained the growth of microfinance in several countries, e.g. Malawi. Government regulations and policies are also crucial in creating the appropriate environment for the growth ofthe sector. These policies need to strike a balance between protecting the interests of depositors, in micro­ 330 Transforming Bangladesh into a Middle Income Economy finance institutions that collect savings, and not regulating the sector excessively with unnecessary red tape. The scaling up in Bangladesh benefitted significantly from good communications networks - hence government investments in road networks is important to reduce the transaction costs of microcredit. A second lesson is that microcredit may be a more effective remedy against poverty and vulnerability if it is complemented with other interventions. These interventions may be particularly appropriate for the poorest households which face the greatest risks of income fluctuations and have the greatest need for a range of financial and non-financial services. Moreover, whilst the provision of micro credit can enhance a woman's status in the eyes of other household members, social mobilisation and legal education interventions in conjunction with credit are likely to have a more significant effect than credit alone. However, this does not imply that micro­ finance institutions ought to provide these services. In many cases organisations may prefer to specialise in providing micro finance and facilitate linkages to providers of other non-credit interventions. Third, there is a role for donor financial assistance in expanding the capital base in emerging micro finance institutions as well as in developing technical capacity that leadS-to organisational sustainability. Hence, subsidies can be justified to support 'infant' microfinance institutions as long as there is a viable route to institutional sustainability. The duration ofthese subsidies would vary according to local conditions and level of poverty of the clients. A fourth lesson is that while visionary leadership cannot simply be 'franchised', the systems and formal rules that govern the successful micro­ finance industry in Bangladesh can to an extent be replicated. These vary according to the size ofthe organisation but by and large, these organisations delegate significant decision-making authority away from head-offices, are able to monitor individual staff performance and have linked staff incentives with programme targets. Client feedback and programme monitoring are also crucial. As organisations grow, the willingness to change products based on this feedback and to tailor products for niche markets is critical for success. A fifth lesson from the Bangladesh experience is that the creation of a micro finance wholesaler has the potential to play an important role in expanding access, developing professional standards and in advocacy for MFI issues. However, apex bodies are not a panacea and a rigorous analysis of the underlying retail capacity and demand for funds needs to be carried out before they are established. The Bangladesh experience suggests that ifan apex is to contribute significantly to scaling up then it needs to ensure that the flow of funds is synchronised with the needs of growing MFls, that it should have clear requirements for MFls to implement solid business plans and that it should not overburden MFIs with high transaction costs. Sixth, it is clear that the bulk of the scaling up took place through four Scaling up of Microfinance 331 institutions which currently serve ninety per cent ofall microcredit borrowers. Hence another lesson from the Bangladesh experience could be that it is not necessarily a sound strategy to support many different institutions, and risk spreading resources thinly, in order to reach large numbers of poor people. This is particularly in light ofthe fact that the leadership skills and professional capacity to go to scale are in limited supply in most countries. However, this needs to be balanced by the risks of having a concentrated market structure such as the industrywide consequences of poor performance within one institution. A related point is that the 'franchising' model ofsetting up virtually identical field offices was possible due to the relatively simple credit delivery system that was offered at the time. The introduction of more diversified products, catering to different niche markets, was only introduced after the rapid expansion phase slowed down. This deliberate sequencing strategy for scaling up services is a notable lesson from the Bangladesh experience. Challenges While this paper focused on the lessons from the scaling up ofthe microfinance industry, and its impact on the poor, the story would be incomplete without mentioning the future challenges that the industry faces in Bangladesh. First, the microfinance sector needs to build a stronger domestic constituency that understands the economics ofmicrofinance and in particular the reasons for why interest rates are higher than commercial banking rates. Demands for lowering and capping interest rates have gained ground in recent months and greater public debate around these issues need to be fostered. In turn the apex microfinance body, PKSF, has a role in improving transparency by publishing information on interest rates, operating costs, profit margins, etc. Second, the industry needs to accept that their current operating spreads will need to shrink as donor-grant funds for micro finance diminish and they move towards more commercial sources of funding. Hence, MFI managers will need to balance greater efficiency with their ultimate objective which is to provide quality financial services to the poor. The reduction in spreads will also imply less cross-subsidisation of social programmes that many NGO­ MFIs operate which in turn requires a financing strategy for these programmes. A third challenge is developing an appropriate regulatory framework for the industry. Recent progress on this, led by the Central Bank, PKSF and industry representatives, needs to be built on. Given the large variety of institutions that exist, a tiered regulatory structure, as in the Philippines, is likely to be appropriate for Bangladesh. A related issue is the fact that governance arrangements of MFIs need to be strengthened and lines of accountability made more explicit. Finally, as the competition for new microcredit clients intensifies, MFIs 332 Transforming Bangladesh into a Middle Income Economy will need to further refine the services that they offer and cater to niche markets. The capacity to identify these markets and design appropriate products will require investments in institutional capacity building as well as in publicly accessible market information and research. ACKNOWLEDGEMENTS Comments from Salehuddin Ahmed, Sadiq Ahmed, Syed Hashemi, Naomi Hossain, Kristin Hunter, Wahiduddin Mahmud, Frank Matsaert, Iffath Sharif and Stephen Rasmussen are gratefully acknowledged. REFERENCES Ahmed S. 'Microcredit and poverty: new realities and strategic issues' in Attacking poverty with micro-credit, UPL, Dhaka, 2003. Amin S. and A. Pebley. 'Gender inequality within households: the impact of a women's development programme in 36 Bangladeshi Villages' in The Bangladesh Development Studies Special Issue on Women, Development and Change voL XXII, no. 2&3, edited by S. Amin, BIDS, Dhaka: 1994. Cain M., S.R. Khanum and S. Nahar. 'Class, patriarchy and women's work in Bangladesh' Population and Development Review, voL 5 (3), 1979. Chowdhury M. and R. Cash. A Simple Solution: Teaching Millions to Treat Diarrhoea at Home, UPL, Dhaka: 1996. Choudhury S.H. 'Financing the poor: ASA experience', The Daily Star, 13 March Dhaka: 2003. Credit and Development Forum. CDF Microfinance Statistic, Dhaka: 2002. Farashuddin F. and N. Amin. Poverty Alleviation and Empowerment: An Impact Assessment Study of BRAC's RDP - ten qualitative case studies, mimeo, BRAC Research and Evaluation Division, Dhaka: 1998. Goetz A. and R. SenGupta. Who takes the credit? Gender, power and control over loan use in rural credit programmes in Bangladesh', World Development, voL 24, no. 1, 1996. Hashemi S., S. Schuler and I. Riley. 'Rural credit programmes and women's empowerment in Bangladesh', World Development, voL 24, no. 4, 1996. Hossain N. Elites and Poverty in Bangladesh, D. Phil Thesis, University of Sussex, 2003. Jain P. and M. Moore. 'What makes microcredit programmes effective? Fashionable fallacies and workable realities', IDS Working Paper 177, University of Sussex, 2003. Scaling up of Microfinance 333 Khandker S. Fighting Poverty with Microcredit, Oxford University Press, 1998. - - - . 'Microfinance and poverty: Evidence using panel data from Bangladesh', Policy Research Working Paper 2945, Washington, DC: World Bank, 2003. Lipton, M. 'Successes in anti-poverty', Issues in Development Discussion Paper no. 8, International Labour Office, Geneva: 1996. Lundberg S. and R. Pollak. 'Separate spheres bargaining and the marriage market', Journal ofPolitical Economy vol. 101, no. 6, 1993. Mahmud S. 'From women's status to empowerment: The shift in the population policy debate', The Bangladesh Development Studies vol. XXII, no.4, BIDS, Dhaka: 1994. Matin I. 'Mis-targeting by the Grameen Bank: A possible explanation', IDS Bulletin vol. 29, no. 4, 1998. Montgomery R., D. Bhattacharya and D. Hulme. 'Credit for the poor in Bangladesh: The BRAC rural development programme and the government than a resource development and employment programme' in Hulme D. and P. Mosely Finance against Poverty, vols. 1 and 2, London: Routledge, 1996. Morduch J. Does Microfinance Really Help the Poor: New Evidence from Flagship Programs in Bangladesh, Department of Economics and HIID, Harvard University and Hoover Institution, Stanford University, 1998. Mustafa S., I. Ara, D. Banu, A. Kabir, M. Mohsin, A. Yusuf and S. Jahan. Beacon of Hope: An Impact Assessment Study of BRAC's RDP, Research and Evaluation Division, BRAC, Dhaka: 1996. Naved R. 'Empowerment of women: Listening to the voices of women' in The Bangladesh Development Studies Special Issue on Women, Development and Change, vol. XXII, no. 2&3, edited by S. Amin, BIDS, Dhaka, 1994. PKSF. Maps on Microcredit: Coverage in Bangladesh, Dhaka, 2003. Rahman H. 'Mora kartik: Seasonal deficits and the vulnerability of the rural poor' in Rahman H. and M. Hossain (eds) Rethinking Rural Poverty: Bangladesh As A Case Study, UPL, Dhaka, 1995. Wood G. and I. Sharif (eds). Who Needs Credit? Poverty and Finance in Bangladesh, London: Zed Books, 1997. Yunus M. Banker to the Poor, UPL, Dhaka: 1998. Zaman H. 'Assessing the poverty and vulnerability impact of microcredit in Bangladesh: a case study ofBRAC' ,Policy Research Working Paper Series 2145, Washington, DC: The World Bank, 1999. 12 Human Development Strategy to Achieve MDGs Anil Deolalikar SECTION I: INTRODUCTION' Since the launch of the Millennium Development Ooals (MDOs) at the Millennium Summit held in New York in September 2000, the MDOs have become the most widely-accepted yardstick of development efforts by governments, donors and NOOs. The MDOs are a set of numerical and time­ bound targets related to key achievements in human development. They include: halving income-poverty and hunger; achieving universal primary education and gender equality; reducing infant and child mortality by two­ thirds and maternal mortality by three-quarters; reversing the spread of HIVI AIDS; and halving the proportion of people without access to safe water. These targets are to be achieved by 2015, from their level in 1990. Almost all the countries in the world, including Bangladesh, have committed themselves to attaining the targets embodied in the Millennium Declaration by 2015. Unfortunately, there is little understanding of whether Bangladesh will be able to attain all of the MDOs, and whether there are some MDOs that Bangladesh will be able to attain. There is even less understanding of what it will take - by way of economic growth, infrastructural investments, and social-sector interventions - to attain the different MDOs. This paper focuses on the attainment of four major human development­ related MDOs by sub-national units in Bangladesh - under-five mortality, child malnutrition, schooling enrolment and completion, and gender disparities in schooling. It is mainly concerned with answering the questions: how likely 1 The paper is based on a World Bank report on the subject led by the author. See World Bank 2004. Human Development Strategy to Achieve MDGs 335 is Bangladesh to attain the MDGs in these four areas, and what will it take to attain these MDGs?2 Data, Methodology and Caveats Virtually all of the analysis in this paper is based on three sets of national household surveys. First, data from three rounds of the nationally representative Bangladesh Demographic and Health Survey (BDHS), which were collected in 1993-94, 1996-97, and 1999-2000, are used to analyse the levels and correlates of infant and under-five mortality and malnutrition. Second, unit record data from the 2000 Child Nutrition Survey (CNS) conducted by the Bangladesh Bureau of Statistics (BBS) are also used to analyse the levels and correlates of child malnutrition. Finally, unit record data from the 2000 Household Expenditure and Income Survey (HIES), also conducted by the BBS, are used to analyse the levels, patterns, schooling enrolment and completion, and gender disparity in schooling. The methodological approach adopted in this paper consists of applying econometric estimation techniques to household survey data in order to analyse the socio-economic and policy correlates of the selected MDG indicators. These estimates are then used to simulate the likely trajectory of the MDG indicators under alternative scenarios of change between 2001 and 2015. By its very nature, any empirical analysis is predicated on assumptions about data quality and measurement, inferences of causality between variables, and potential biases of statistical and econometric estimates. The analysis presented in this paper is not immune to these same concerns. It is, therefore, important to note at the outset that while the results and simulations presented in this paper may give an impression of precision, they are not that. 3 They should be treated as indicative of possible broad trends, and could usefully be complemented with other analyses using different methodological approaches. As long as the results are used with this understanding, they can be helpful in 'rough-order' planning for MDG attainment. Finally, it is important to note an important limitation of the simulations perfonned in this paper. The simulations are based on statistical analysis of household survey data. By its very nature, such analysis tends to over­ 2 The selection of these MDGs for detailed analysis was based in large part on the availability of reliable sub-national data. For example, reliable data on disease prevalence at the district or divisional level are simply not available, and this hampers useful sub-national analysis of the communicable disease-related MOO. The same is true of another important MDG indicator - maternal mortality. 3 In addition to lack of precision, the estimates presented in this paper, like other econometric estimates, may be subject to systematic biases arising from measurement errors in the independent variables and from the omission of important variables and unobserved heterogeneity from the analysis. 336 Transforming Bangladesh into a Middle Income Economy emphasise readily-measurable variables, such as household income or consumption, adult schooling levels and access to infrastructure, and under­ emphasise qualitative variables, such as the quality of institutions, governance, and empowerment. Obviously, this does not imply that the latter variables are irrelevant to the MDG indicators; indeed, institutional reform and good governance are critical to the attainment of the MDGs. It is, therefore, important to view the messages of this paper as complementing those from the numerous qualitative (and detailed) studies of health, nutrition, schooling and poverty that have been conducted in the past. Source: See Appendix Table A 12.2 Figure 12.1 Infant Mortality Rate, 1911-99 SECTION II: INFANT AND UNDER-FIVE MORTALITY The mortality of children is often seen as the criteria of 'success and failure of nations' (Sen 1998). It is an important indicator of well-being in its own right, as recognised by its inclusion among the MDGs. The mortality of children not only represents an enormous waste of human resources, but also a major cause of suffering in the population. The millennium development goal for Bangladesh is to reduce the infant mortality rate to 3 1 by20l5. Trends The historical trends in infant mortality, culled from various sources and surveys, are shown in Figure 12.1. The IMR appears to have dropped sharply in the early 1900s, but barely dropped from 168 infant deaths per 1,000 live births to 161 deaths during the two decades between 1951 and 1971. In the immediate aftermath of Bangladesh's independence, the IMR actually Human Development Strategy to Achieve MDGs 337 increased to 173. But since then the IMR has fallen secularly and rapidly, reaching a level of 125 by 1984-85,80 in 1994-95, and 66 for currently. It is only after 1989 does one see a definitive and a faster trend of decline. Two data sets provide much of the recent information on infant mortality. One is the vital registration survey (VRS) data of the BBS and the other is the BDHS data. Both suggest dramatic improvements in infant mortality in the 1990s. The VRS data of the BBS represents the longest series on IMR based on a single source. The VRS data show virtually no improvement in infant mortality during 1980-88 (and an increase in 1980-82) (Figure 12.2). In 1988, the IMR still stood at 116, but by 1995 it had dropped to 75. The rate fell even faster during the late 1990s and early 2000s to 57 by 1998, 53 by 2000 and 51 by 2002. While the very low infant mortality rate of 51 estimated by the VRS for 2002 is probably the result of a death registration system that is not complete, the rate of decline in infant mortality in recent years suggested by the BBS data is confirmed by the BDHS data, which also show the IMR halving from its levels in the last decade (Figure 12.2). 130 129 122 122 120 1 8 112116113116 117 117 112 110 100 ~ 96 94 91 90 88 84 87 82 00 n ~ n87 66 :~~~~~~~~~~~~~~~lyi~~j51~~~~~~ i~i~~~~~~~~~~~~~~~~ ii;;;;; Estimates from the Vital Registration System Estimates from BDHS Figure 12.2 Infant Mortality Rate Estimated from the Vital Registration System and the Bangladesh DHS, 1980-2002 International Comparisons How does Bangladesh's performance at infant mortality reduction compare to that of other countries in the region? Over the period 1970-2000, infant mortality has fallen by anywhere from 2.6 to 5.6 per cent annually in the countries shown in Figure 12.3, with South Korea and Sri Lanka being the stellar performers. Bangladesh has, however, done very well, managing to reduce its infant mortality rate at a rate comparable to that of Thailand and much faster than that of India. Indeed, what is surprising is that the level of infant mortality is now lower in Bangladesh than in India - a country whose per capita GDP is about two times that of Bangladesh's. 338 Transforming Bangladesh into a Middle Income Economy -3.3% 145 [] III • 1970­ 1980­ 1994­ 125 75 85 2000 -3.9% 100 105 Annual Rate of Change in IMR, 1970--2000 85 -3.1% 65 60 -4.3% 45 -5.6% 44 33 25 51-'-~ __.,.....L--,"", S. Korea Sri Lanka Thailand Indonesia India Bangladesh -15 Figure 12.3 Infant Mortality Rate, 1970-2000, Selected Countries in Asia Role of Family Planning and MCH Programmes The importance of family planning interventions in bringing about infant and child mortality decline cannot be discounted. It is well known that fertility decline and mortality decline often go hand in hand with each other. Bangladesh has had one of the most successful family planning programmes in the developing world. The programme has achieved extraordinary results by building an extensive network of health and family welfare clinics throughout the country, training thousands of female workers to take family planning advice directly to women, and using mass media campaigns to create awareness about family planning in the population. The programme has enjoyed strong political commitment from the government, grassroots-level partnership with NGOs, and generous and coordinated assistance from donors. Indeed, Bangladesh's experience has shown that it is possible to bring about fertility and mortality decline in poor countries even in the absence of strong economic growth and improving socio-economic conditions. Some of the best evidence of the role of effective family planning and MCH interventions on infant mortality decline in the developing world comes from the Matlab area of Bangladesh, where the Maternal Child Health and Family Planning (MCH-FP) Project has been operating since 1977. This project has provided more accessible and better quality family planning services to a 'treatment' area in comparison to those offered in nearby 'control' areas. The more accessible and better quality family planning services have included more frequent visits from female welfare assistants who provide counselling and deliver contraceptives, as well as closer access to a network of family planning sub-centres operated by the International Centre for Diarrheal Diseases Research (ICD DR, B). Figure 12.4 suggests that the MCH­ FP project has contributed to a decline of 10 to 30 per cent in infant mortality since its inception. Human Development Strategy to Achieve MDGs 339 Projections to 2015. The Bangladesh DRS data suggests that the decline in infant mortality in Bangladesh between 1979-83 and 1995-99 has averaged an impressive 3.6 per cent annually. The decline during the 1990s has been even more rapid - about 4.7 per cent annually. Figure 12.5 suggests that if the rate of infant mortality decline experienced between 1979-83 and 1995-99 continues into the future, infant mortality rate in Bangladesh could be expected to reach a level of34 in 2015 - just slightly above the MDG level of31. If the future rate of decline remains at the (higher 4.7 per cent) rate experienced in the 1990s, the infant mortality rate could decline to 29 by 2015. Thus, on the surface, it would appear that Bangladesh could expect to attain the infant mortality MDG - or come very close to attaining it - if it simply continues the trend it has seen in the recent past. 180 160 A