---

Abstract Introduction The Role of the Budget Macroeconomic and Sectoral Policies Administrative and Regulatory Reforms Budget Planning: Freeing Resources for Poverty Programs Decentralization: Strengthening Local Government Institutions The 2003-2004 Budget    - Tax and Revenue Policies    - Recurrent Expenditures    - Development Spending The Role of the Donors Conclusion References

This is a revised version of a paper presented at a seminar organized by the Bangladesh Foundation for Development Research on June 20, 2003 at the CIRDAP Auditorium, Dhaka. I am grateful to a number of seminar participants for drawing my attention to various issues that required clarification.

---

“My experience working in the Grameen Bank has given me faith; an unshakeable faith in the creativity of human beings. It leads me to believe that humans are not born to suffer the misery of hunger and poverty. They suffer now as they did in the past because we turn our heads away from this issue. I have come to believe, deeply and firmly, that we can create a poverty-free world, if we want to.” -  Muhammad Yunus (1998) [1]

Abstract

This paper examines the constraints to developing a pro-poor budget in Bangladesh.  The forces that govern budget planning and implementation processes and the scope for reflecting the interests of the poor in the current budget framework are discussed, with special reference to the 2003-2004 Budget. It is argued that the current budget system is badly structured, very inefficiently managed, and (in the past couple of years in particular) lurching in the wrong direction. Pro-poor policies are far from being at the center of the budget process, despite the political rhetoric. There is a need to initiate the painstaking process of reviewing each budget proposal (revenue and expenditure), item by item, from a specifically poverty alleviation perspective.  In order to make the budget a more effective instrument for poverty alleviation, fundamental reforms are needed. These would include substantial devolution of service delivery functions to elected local governments and a shift to program budgeting or performance budgeting. Budget policies impact on the poor in four main ways: macroeconomic effects, revenue policies, the pattern of expenditures and deregulation. Depending entirely on higher growth and investment to yield “trickle-down” benefits for the poor is not a sound strategy given the social risks posed by higher inequality. It is therefore vital for the Government to intervene through redistributive policies to ensure that the benefits of growth accrue to disadvantaged groups. The growth forecasts which underpin the current poverty alleviation strategy are also questioned given the breakdown in business confidence and the negative effects of a declining Annual


 

Development Program. The role of the multilateral donors (notably the IMF and the World Bank) and their policy bias is also discussed, together with the need to ensure that poverty alleviation strategies reflect political realities and are developed on the basis of wide participation.

Key words and terms: Bangladesh, budget, local governments, devolution, program budgeting, performance budgeting, inequality, Annual Development Program, IMF, World Bank.  

1. Introduction

19.(1)“The State shall endeavor to ensure equality of opportunity to all its citizens”. - Constitution of the People’s Republic of Bangladesh.

 “[Equality of opportunity] obtains in so far as, and only so far as, each member of a community, whatever his birth, or occupation, or social position, possesses in fact, and not merely in form, equal chances of using to the full his natural endowments of physique, of character, and of intelligence”. - R. H Tawney (1961:106)  

The National Budget must not be viewed merely as a presentation of the accounts of the Government of Bangladesh, a summary of its past financial operations and estimates for the future. It is the most important statement of the policy direction and intent of the Government, with ramifications throughout the economy and society. Even in a country where total budgetary outlays as a percentage of GDP are relatively low, it can be a powerful instrument for the achievement of key policy objectives such as the eradication of extreme poverty, ensuring the provision of good quality basic social services and providing support for private economic initiatives. The case for state intervention to ensure that the benefits of economic growth are shared widely is a well-established one (Chenery et al 1974). The content and the presentation of budgets in Bangladesh remain disappointing from this perspective. This is partly due to structural weaknesses that prevent the redirect budget resources towards policy goals, but also a lack of political will.

Most of the criticisms in this paper apply to some degree to all recent budgets. However, the past two budgets do stand out in many ways. These budgets have demonstrated not only an inadequate commitment to some of the professed policy goals of the Government, but an alarming tendency to move macroeconomic management, revenue and spending policies in a direction inimical to the interest of the bulk of this nation’s electorate.      

Every government in this country has declared that its budgets have been formulated with the interests of the poor in mind. It is asserted that the preeminence of poverty alleviation as a policy objective is reflected in the design of the budget. The rhetoric about social justice and equality of opportunity are there. However, the widening inequality of incomes (and disparities in rates of investment in human capital), the slow progress in reducing the number of the ultra-poor, and the inadequate access of the poor to basic social services all demonstrate that the reality is quite different. The emptiness of these declarations is quite well recognized by the common people of this country, whose understanding of the realities of economic policymaking should not be underestimated. There is a pervasive sense of helplessness in a large section of the electorate as neither of the two major political parties seems willing to take the bold steps needed to address the problem of poverty in this country. A major factor has been the influence of key groups that benefit from the current flawed system of public resource allocation and management. However, another factor may be the limited public recognition of the potential role of the national budget, and the vital necessity of reforms to enable the budget to be used as a more effective policy instrument. [2]

This paper examines the constraints to developing a pro-poor budget in Bangladesh. [3] The forces that govern budget planning and implementation processes and the scope for reflecting the interests of the poor in the current budget framework are discussed, with special reference to the 2003-2004 Budget. The growth forecasts which underpin the current poverty alleviation strategy are also questioned given the breakdown in business confidence and the negative effects of a declining Annual Development Program. The role of the multilateral donors (notably the IMF and the World Bank) and their policy bias is also discussed, together with the need to ensure that poverty alleviation strategies reflect political realities and are developed on the basis of wide participation.

The current budget system is badly structured, very inefficiently managed, and (in the past couple of years in particular) lurching in the wrong direction.   It will be argued that far from being at the center of budget planning processes, programs specifically directed towards poverty alleviation are treated at best as discrete add-ons or supplements at the margin. Even where innovative pro-poor programs have been introduced (such as the modest old-age pensions introduced under the previous government) the funding associated with these programs is dwarfed by outlays on many non-essential activities that have much less significance for the welfare of the poor.

In order to make the budget a more effective instrument for poverty alleviation, fundamental reforms are needed. These would include substantial devolution of service delivery functions to elected local governments and a shift to program budgeting or performance budgeting. There is a need to initiate the painstaking process of reviewing each budget proposal (revenue and expenditure), item by item, from a specifically poverty alleviation perspective.

2. The Role of the Budget

“Great nations are never impoverished by private, though they sometimes are by public prodigality and misconduct. The whole, or almost the whole public revenue, is in most countries employed in maintaining unproductive hands.” - Adam Smith, The Wealth of Nations (1776) [4]

Budget policies impact on the poor in four main ways:

• through its impact on macroeconomic conditions (economic growth, inflation, investment and employment);

• expenditure policies; 

• revenue policies, and

• changes in the administrative and regulatory framework and policies that impact on the overall environment for private sector economic activity.

During the 1990s the Government’s scorecard on macroeconomic performance, although certainly better than many countries at a similar stage of development, has overall fallen short in a number of respects.  Even during the period of unprecedented growth in the agricultural sector, real GDP growth did not reach the 7 percent growth rate that has been as the critical minimum rate needed to ensure a substantial reduction in the numbers of absolute poor over the medium term (World Bank, 1998:21). It may be noted that even after a decade of remarkable progress in terms of social indicators the actual numbers of the poor remain virtually unchanged at the beginning of the new millennium compared to 1990, although the number of the hard-core poor declined by 2.5 million to 42.5 million over the same period (World Bank 2003a:i). 

Even if growth rates of 7-9 percent cannot be achieved in the medium-term, the Government can act to improve the condition of the poor. [5] This can be done through various redistributive policies, including taxation, targeted subsidies, social safety nets, and public goods provision. Depending entirely on higher growth has long been recognized as an unsound basis for a poverty alleviation strategy (Chenery et al, 1974).  The “trickle-down” effects of growth often fail to reach the most disadvantaged members of society or reach them partially, or with a considerable lag. Moreover, around the world, faster growth has been associated with an increase in inequality. [6]   There is considerable evidence that the same is true of Bangladesh. It is therefore important for the government to intervene through the budget to ensure that the poor fully share in the benefits of faster growth, and that inequality does not increase to levels that are unacceptable in country where many people remain deprived of the most basic necessities (Islam, 2003 and Khan, 2003).

In order to develop a “pro-poor” budget, both revenue and expenditure policies will need to be reformulated. The former will need to be revamped to make the tax structure more genuinely progressive while expenditure programs will need to be restructured to give primacy to the objective of poverty reduction. The incrementalism that still largely governs budget formulation (despite the considerable progress made under budget reform projects such as RIBEC over the last decade) must be gradually discarded. A thorough review of all major spending programs must be undertaken and decisive measures taken to curb unproductive spending.

3. Macroeconomic and Sectoral Policies

The slowdown in real GDP growth observed in 2001-2002 has apparently been at least partially reversed in 2002-2003. The 2003-2004 Budget presents relatively favorable growth projections for this year, largely on the basis of the resumption in growth of the agricultural sector. However, the deterioration in the overall business environment remains a cause for concern, with the impact of sharp falls in investment in the last two years likely to be seen in slower-than-expected growth in manufacturing and services. The reasons for the low confidence levels of both domestic and foreign investors are well-known: a breakdown of law and order (periodically addressed through such desperate measures as the deployment of the armed forces) and a confrontational political milieu. These important factors – given their influence over business confidence – are not directly susceptible to budgetary policy, and are likely to be among the main reasons (the dramatic fall in development spending is another important factor – see below) why the goal of faster real economic growth is unlikely to be achieved. In this environment, tinkering with tax or interest rate policies will not yield any results in terms of decisions to increase investment. This failure would have the most serious implications for the welfare of the poor and the rate of reduction in poverty, as demonstrated in the past decade (Islam 2003).

The external accounts remain vulnerable, as a sustained decline in export earnings in real terms (with a decline in nominal terms from about U.S. $6.5 billion in FY 2001 to U.S.$6.0 billion in FY 2002 and possibly U.S. $ 6.3 billion in FY 2003) has been offset by a surge in remittances that may not be sustained over the medium-term. Although there has been a rise in external reserves, the reasons for this are disquieting, as to a large extent it is the consequence of a slowdown in economic activity and lower levels of investment. Inflation stands at about 6 percent (compared to the 2 percent rate at the time the present government took office), well above the rate prevailing in our major trading partners, a factor that will put increasing pressure on external competitiveness. 

The decade of the 1990s was one of sustained economic progress and strong growth in real incomes for Bangladesh, despite severe shocks such as the 1997 Asian Economic Crisis and the devastating floods of 1998. Real GDP growth averaged 4.5% p.a. during 1991-95 followed by about 5.8 % p.a during the period of the previous government. The current budget reflects a realization that a more expansionary stance of financial policies is needed to restore the higher growth rate achieved during the second half of the 1990s, but there is a risk that growth will fall short of the target due to two factors: a failure to stimulate private investment activity adversely affected by the clouded political outlook and the law and order situation, and a failure to implement the ADP. In both these areas the 2003-2004 Budget Speech did not provide any indication of corrective measures being taken.     

The overall success of sectoral policies – particularly agriculture - can also have a major effect on the lives of the poor, and this is one area where the previous Government appears to have made a significant contribution. The sharp increase in foodgrain production (at least partly due to productivity gains) observed during the five years of the previous Awami League Government was associated with a decline in rice prices and a rise in the real incomes of the poor. The poor also benefited from the higher employment opportunities engendered by the substantial expansion in ADP outlays. [7]

The success achieved in this sector was the result of a combination of policies, including providing duty-free access for power tillers, tax incentives for agro-processing industries, improving fertilizer distribution, and substantially increasing agricultural loans (with a doubling of loan volumes compared to the previous government). The end result was the achievement of food-self sufficiency for the nation, and more pertinently, a large increase in the real earnings of the rural poor in terms of food (Chart 1). For large sections of the poor the availability of work, the daily wage rate and price of coarse rice are the critical determinants of well-being, and indeed, survival. The first of these factors is influenced - in rural areas without major industries in particular - by the level and composition of development spending (see 6.3 below). The latter two factors can be expressed in a ratio which indicates the amount of coarse rice that can be purchased with a day’s labor. The coarse rice value equivalent of one day’s earnings for unskilled labor rose from about 3.6 kg in 1994/95 to about 5.2 kg in 2001/2002. It is disturbing to note recent trends in this figure, with declines to 4.8 kg in 2001/2002 and an estimated 4.2 kg in 2002/2003. [8]   

4. Administrative and Regulatory Reforms

“Ordinary citizens have no expectation of assistance or cooperation, or even polite behavior, from government officials. When taxpayers without special clout are fortunate enough to receive the state services to which they are entitled, they are given the impression that they have received a favor from the civil servants involved…ordinary taxpayers have no sense of a right to services or, indeed, of ownership of the government for which they pay”.

World Bank, Government That Works (1996a:7)

There is an urgent need to reorient the entire government towards a genuine attitude of pubic service, built upon the principles of accountability and responsiveness to the needs of ordinary citizens. For the poor and less educated, obtaining access to the services ostensibly set up for them is fraught with difficulty and is sometimes relatively expensive as well. For example, one survey found that 30 percent of households whose members had been treated in government hospitals admitted making under-the-table payments in order to obtain care, poorer households paid larger bribes to obtain water connections in municipalities, and 55 percent of respondents experienced difficulties in obtaining textbooks that are supposed to be provided by the government free of cost (World Bank 2002, vii, ix, 27). This situation is observed in every area of public services. The establishment of customer service standards and the development of effective mechanisms for redress of complaints should therefore be a high priority for any government interested in improving the condition of the poor. The funding increases for social services that have taken place over the past decade must be welcomed, but they are not enough. It is vital that enhanced funding and expansion of programs are combined with efforts to enhance the effectiveness of spending, an issue discussed further in the next section.       

The quality of provision of basic social services (notably education and health) appears to have deteriorated in the past few years (World Bank 2002). For example the percentage of households with school-going children unhappy with the standard of education rose from 22 percent to 30 percent over a five year period, while the proportion of those dissatisfied with the standard of health care rose from 34 percent to 44 percent over the same period (World Bank, 25, 29). The reasons for such apparent declines in performance are unclear, but the available survey evidence suggests that this is a subject that needs to be investigated if corrective policies are to be taken.  

The decline in the performance of government agencies providing public services may have been aggravated by the situation with alternative service providers.  The apparent deterioration in the operating environment of NGOs after 2001 due to the less favorable attitude of the government and disagreements within the NGO movement is likely to have a negative effect on the overall welfare.

The poor do not have the option open to the rich of bypassing the effects of “government failure” by turning to private service providers. As a result they are more severely affected by poor access and low quality provision of public services in Bangladesh (World Bank, 1996a: 6-8).  Given the importance of investment in human capital, particularly with greater globalization and the emergence of a “knowledge-based” economy, the failures in the provision of education and other social services would have long-lasting   ramifications for future patterns of income distribution. It is therefore essential that the quantity and quality of public expenditures in basic social services be enhanced further, building upon the progress made in the 1990s (World Bank 2003b).      

The impact of the failure of economic and political governance – which has been especially pronounced in the past two years of the present BNP-Jamaat coalition government – appears to be placing a relatively larger burden on small-scale economic activities, which account for a large share of total employment. Small businesses, the self-employed and petty traders are least able to protect themselves from the breakdown of law and order and the increased toll-taking by criminals. This is in addition to rent-seeking behavior on the part of government officials. Regulations of any type tend to be selectively enforced in Bangladesh. The introduction of new regulations or initiatives to enhance compliance - however well-intentioned - often merely have the effect of providing a new source of revenues for those entrusted with enforcement. This adds significantly to start-up and operating costs of small businesses in particular, which appear to bear a disproportionately larger burden of such extortions. [9]   While some progress has been achieved in the area of deregulation over the past two decades, small businesses still face an enormous number of regulatory and administrative obstacles. These include: excessive amounts of paperwork, complicated official forms and regulatory conditions and sometimes the necessity of multiple official approvals for the most innocuous of private sector activities or the provision of public utilities. Together these constitute a daunting set of hurdles for those seeking to set up a small business in this country.  

5. Budget Planning:
Freeing Resources for Poverty Programs

…Government is doing to many commercial functions that others can do better, and too little – in quantity and quality of what it should be doing more, i.e., poverty alleviation, rural infrastructure, etc. It is overextended – with attendant regulatory intrusiveness, inefficient in most of its activities, and has too many people doing too little but more often obstructive initiative …its inefficiency affects the entire economy”. 

World Bank (1996a) Bangladesh: Government That Works (p. xvii)

The important part played by large scale and continuous investments in human capital in developing the foundations for self-sustained growth in East Asia exerted considerable influence on policymakers in developing countries, both directly and through the recommendations of multilateral donor agencies. [10]   In Bangladesh there was a considerable shift in the structure of the budget expenditures through the 1990s as expenditure on basic social services rose significantly  as a percentage of total budgetary outlays, from about 20.4% in 1991/92 to 25.8% in 1996/97 before and falling away slightly to 24% in 2002/03. [11] In general, the available evidence suggests that the poor have benefited significantly from this increased social sector spending (World Bank/ADB 2003b: 15-16).

The Bangladesh budget remains characterized by a number of severe shortcomings that prejudice its effectiveness as a leading vehicle for policy implementation (World Bank 1996a: 55-63). Some useful reforms were undertaken over the past decade with the support of external technical assistance (notably under the project for Reforms in Budgeting and Expenditure Control - RIBEC), but these have failed to address the fundamental weaknesses in the system. To a large extent these reflect wider governance and administrative problems. It would be best to first identify what Government should be doing, and then to turn to the issues of who should be doing them (local governments, alternative service delivery mechanisms) and then, finally, how they should be done. All three issues need to be addressed simultaneously: merely dealing with the third could lead to a situation where the wrong things are still being done, but done more efficiently.      

Budgets in this country still have the following weaknesses:

• a large degree of incrementalism (carry-overs of many spending programs without any thorough review of their value or rationale). As a result the bulk of the available funding may be pre-committed, allowing little room for a government to introduce its own expenditure priorities;

• a focus on inputs used rather than outputs and outcomes. As a result there are few real measures of performance available for public scrutiny;

• an absence of a consistent system of cost-analysis of programs (under the recurrent budget) and activities which be the basis for cost-effectiveness audits;

• a failure to develop mechanisms to improve service delivery and to enhance the responsiveness of agencies delivering public services to user needs. This is particularly important for the welfare of the ultra-poor, whose illiteracy and economic and social status (women and the elderly who live on their own in particular) make it difficult for them to obtain access to public services.

In order to use the budget as an effective instrument for a poverty-alleviation strategy it would be important to get the basics right and first create a well-functioning public finances management system that should have the following features:

1. A realistic budget implemented with few significant deviations from original estimates. The widening deviations from budget estimates in recent years is a cause for concern (see section 6.2 below);

2. A high level of transparency on public expenditures and the budget. There should be no hidden build-up of contingent liabilities or off-budget liabilities of the government. The large-scale lending of the nationalized banks to loss-making state-owned enterprises in Bangladesh violates this condition; 

3. The integrity of budget review procedures is maintained. There should be no off-budget expenditures or outlays that are not exposed to the standard review processes that ensure the contestability of claims for resources. In this regard the inclusion of sector-wise block allocations for 174 unapproved projects in the ADP for 2003-2004 is a cause for concern. Such allocations destroy the integrity of budget review and approval processes and should be discontinued;   

4. Public funds are used for authorized purposes. In this regard the present Government’s practice of having supplementary budgets that effectively violate the integrity of the budget approved earlier by Parliament should be a cause for alarm. Resources raised on the pretext of expanding development programs are later in the year quietly redirected to cover overruns in recurrent spending;

5. The use of the budget to validate expenditures;

6. Reported expenditure corresponds to actual expenditures (see item 3 above);

7. Reliable external or internal controls and audits. While expenditure monitoring has improved over the past decade, even where problems are identified follow-up is poor and there is no evidence that the budgetary framework is capable of making the necessary adjustment; [12]

8. Spending units have reasonable certainty as to funds that will be available. The preparation of the current budget within a Medium-Term Economic Framework (MTEF) is a step in the right direction; 

9. A management culture that promotes fidelity to formal rules; overruns in recurrent spending indicate that this is not the case in Bangladesh.

The widening deviations from the original Budget estimates over the years reflect a failure on the part of those involved in budget planning and implementation. On the planning side, consistently over-optimistic revenue forecasts and persistent expenditure overruns, suggest that a review of the budget preparation process is called for. In many countries these twin phenomena reflect political pressures that lead to the overriding of technically sound forecasting methods. Over-expenditures – particularly when they take place year after year - undermine the entire budget process, as Parliament is systematically deceived on the true financial position of the government. All MPs, including members of the ruling party, should take exception to this form of accounting sleight-of- hand  and demand greater accountability in spending. In this context it may be noted that on June 17, 2003 a Supplementary Budget of Tk. 2150 crores was passed by Parliament, with 88% of the excess spending (Tk. 1892 crores) being accounted for by non-development heads of expenditure. At the same time there was a sharp cutback in the size of the ADP from an originally budgeted Tk. 19,663 crores for 2002-2003 to perhaps less than Tk. 13,000 crores on current indications. The end result has been the raising of resources through higher taxation in the name of development that end up being diverted to other uses. This is unacceptable from both the perspectives of efficient budget management and democratic accountability.           

Clearly, achieving all the features of a well-functioning budget system in Bangladesh will take many years. What is needed is the development of results-oriented budgeting or what is sometimes termed as a Planning and Performance based budgeting system that creates incentives and mechanisms to support better performance and enhanced monitoring of outcomes or the impact of budgetary spending (World Bank 1998: Ch. 5). The absence of such a system should not be a cause for inaction, however.  There are a number of areas in which policy actions can be taken to move towards a pro-poor budget, even under the unreformed budget structures. Some of these are discussed in the sections that follow.

6.   Decentralization:
Strengthening Local Government Institutions

“The State shall encourage local Government institutions composed of representatives of the areas concerned and in such institutions special representation shall be given, as far as possible, to peasants, workers and women.”

Article 11, Constitution of the People’s Republic of Bangladesh

“…the practical principle in which safety resides, the ideal to be kept in view…may be conveyed in these words: the greatest dissemination of power consistent with efficiency; but the greatest possible entralization of information, and diffusion of it from the centre”… - John Stuart Mill, On Liberty (1859) [13]

Arguably one of the most important elements of a pro-poor budgetary framework is the establishment of a strong and well-funded local government system. Budgets have become ever more Dhaka-centric – not in terms of allocations but in terms of the disposition of budgetary resources. Large amounts of money are allocated for projects that would impact on the rural poor, but it is an accepted fact that only a small fraction of these resources actually reaches the target population, with substantial “margins” of various forms accruing to economic agents largely based in the capital. [14] This is reflected in the rising urban-rural disparity in per capita incomes in the past decade (Khan, 2003: 13). Supporting devolution and the creation of a large number of growth centers at the district level would help correct this tendency and would be desirable from both equity and poverty alleviation perspectives. 

The issue of developing an effective system of elected local governments is obviously a political one. The system of upazilas initiated by General Ershad in the 1980s had the potential of being one of the most innovative mechanisms for rural development seen in this country. For various reasons, the system failed to evolve as expected, perhaps due to the inherent contradiction of strengthening accountability and democratic institutions at the local level under the umbrella of a military dictatorship. It was unfortunate that the BNP Government that came to power in 1991 failed to examine the upazila system on its merits and chose to discard it as creation of its predecessors.  It must be noted that the Awami League Government that followed also failed to take the decisive actions needed to revive this initiative.  A larger minimum size of administrative jurisdiction would certainly be more desirable (perhaps in the range of the current districts) from the point of view of operational efficiency. Even a less-than-perfect-upazila system is to be preferred to the absence of any effective system of local government.

The proponents of centralized management of development activities have suggested that local elected governments would be prone to “capture” by local elites and would provide the scope for unchecked inefficiency and misuse of public funds. Even allowing for “government-failure”-type  problems under an elected local government system, it can be argued that the mere physical proximity of those carrying out development or poverty alleviation programs and the intended target or recipient groups would allow for greater accountability and transparency. It is virtually impossible to continuously monitor physical implementation, operational effectiveness and user satisfaction of rural programs by agencies based in Dhaka. A system of field visits and local representative offices operated by a Dhaka-based government agency is no substitute for a framework based on genuine devolution (in both an administrative and financial sense) that would allow effective decision-making to take place nearer the project implementation site and facilitate monitoring by stakeholders.

The reduction in the already inadequate level of allocations for local government in the 2003-2004 Budget may provide an indication of the current Government’s intentions in this area. Lump sum grants for “Upazila Development Assistance”, distributed as ADP grants to upazila parishads (UPs) has declined from Tk 200 crore to Tk 170 crore (0.86% of the total ADP). The position of the municipalities is no better: Pourashava Development Assistance has been reduced from Tk 120 crore to Tk 100 crore (0.49% of the ADP). The allocations for the six city corporations have also been reduced from Tk. 120 crore to Tk 100 crore. These grants were insignificant to begin with in the context of the ADP, but comprised a large share of the resources available to many local authorities. Such grants should be there to supplement local authorities’ own revenues – which should underpin their finances – but even here the policies of the current Government (and the previous Government) have not been positive. [15]      

The 2003-2004 Budget Speech is non-committal about the Government’s previously-expressed intention to move forward with a comprehensive program to strengthen the local government system in this country.. The recommendations of the Expenditure Review Commission to reduce the dependence of municipalities on the National Government by allowing them to raise their own revenue are merely noted as under government review in the 2003-2004 Budget Speech (part I, p.25). The failure to hold upazila elections as per the original schedule reflects deep divisions within the present government, with many MPs assuming that the upazila chairman would represent a serious rival source of political authority and a strong competitor in terms of influence over the allocation of development resources within a constituency. Reports of dissent within the cabinet would suggest that the prospects of real reform in this critical area are remote. In this context the recently-unveiled “gram sarkar” (village council) proposal of the BNP-Jamaat Government lacks the features that would allow any genuine devolution of power and authority to the local level.  

7. The 2003-2004 Budget

All past national budgets in this country may be criticized in terms of their failure to give sufficient weight to poverty reduction, and their inadequate pro-poor bias. However, the 2003-2004 Budget (like the previous budget) does stand out as a strongly retrograde document, containing vague (albeit donor-certified) statements about commitments to poverty reduction goals that are completely at odds with the actual stance of revenue and expenditure policies and the budgetary accounts. Projections of higher growth and investment (that would be essential to reducing poverty in the medium term) are put forward without any consideration given to the crippling breakdown of business confidence on the part of both domestic and foreign investors. Sharply higher expenditures – balanced by revenue measures falling disproportionately on the poor - are budgeted in the name of development. However as the experience of 2003-2003 demonstrated, due to an apparent weakening in management and implementation capacity and a likely repetition of overruns in recurrent spending that these resources are unlikely to be expended for any development activities that would benefit the poor of this country.       

The 2003-2004 Budget is presented as one of greater self-sufficiency, as the nation makes a larger contribution to funding its development spending, albeit with continuing strong support from the donor community. A surplus in the 2003-2004 Revenue Budget of Tk. 7,202 crores is estimated. However the experience of the previous year certainly gives cause to doubt this surplus projection: a projected revenue budget surplus of Tk. 9,112 crores for 2002-2003 melted away to one of just Tk.5,813 crores, reflecting revenue shortfalls and recurrent spending overruns. The macroeconomic implications are obvious: the Government would be forced to scale back the ADP or increase borrowing from the banking system. Given the likelihood of aid disbursements linked to Government borrowing ceilings, the former is the more likely scenario, once again leading to adverse impacts on the rural economy.

There has been considerable discussion of fiscal sustainability issues in recent years, with the multilateral institutions expressing concern at recent deficit levels. It is difficult to assess the current budget from this perspective – although the budgeted deficit remains quite large at the equivalent of 4.8 percent of GDP it is to be hoped that the outturn would be somewhat better unless the likely shortfall in ADP spending is offset completely by higher recurrent outlays. What is sometimes lost sight of in the public debate on this issue is while that the deficit level is certainly significant, the real problem is the poor quality (in terms of cost-effectiveness and efficiency) of public spending.

7.1  Tax and Revenue Policies

“If you drive a car, I’ll tax the street,
If you try to sit, I’ll tax your seat,
If you get too cold, I’ll tax the heat,
If you take a walk, Ill tax your feet…
Don’t ask me what I want it for,
If you don’t want to pay some more…
‘Cause I’m the Taxman…”
George Harrison , “Taxman” (1966) The Beatles – Revolver  

Taxation levels are actually quite low in Bangladesh, despite public perceptions to the contrary. The low levels of budget revenues relative to GDP in Bangladesh are a longstanding problem, hindering the expansion of public investment spending. [16] The Government has been facing increasing pressure from both bilateral and multilateral donors in this regard, but faces a number of critical problems. Raising tax rates is extremely transparent but politically unpopular, so the emphasis in recent years – particularly pronounced under the present Government – has been to keep rates generally stable while increasing revenues by extending the tax net and improving compliance. The structure of taxation is in a sense already skewed in favor of the rich, largely as the result of a combination of two factors: 1) the VAT introduced under the earlier BNP Government and 2) a failure to effectively administer income taxes that would ensure an element of progressivity in the tax system. The importance of the VAT – now the centerpiece of the domestic tax system – is undeniable. It is a relatively efficient tax in terms of administration and minimization of disincentive effects. However, while the VAT falls on a wide array of consumption goods and services that comprise a large part of the expenditures of the poor, the taxes imposed specifically on the upper-income groups are administered in a manner that allows the very rich (some groups of businessmen in particular) to minimize their tax liabilities.

Improving tax administration would certainly be desirable, but it is also necessary to deal with the underlying problems associated with the poor compliance rates, notably weaknesses in administration and a lack of confidence that tax monies are used judiciously. Moreover, there is a need to create a perception that tax resources are used efficiently and not subject to large-scale graft. No country’s citizens will be found to be enthusiastic about an increase in their tax burdens but public acceptance of higher and more effectively administered taxes will always be hindered by a lack of acceptance of the electorate of the legitimacy of expenditures and the probity of those entrusted with the public finances. The formation of a Public Expenditure Review Committee is no doubt a step in the right direction. The public and the media will need to be watching closely to see the impact of its findings and recommendations on the patterns and quality of public spending. It must be recognized that improvements are not likely to be achieved very quickly, and that groups within the Government and the business community that benefited from waste and misuse of public funds will put up strong resistance to necessary reforms.

The 2003-2004 Budget calls for an increase in the total of National Board of Revenue administered (NBR) taxes of 4,000 crores, and an additional Tk. 930 crores from other sources. A large part of this is to be achieved as a result of improved administration and growth. It must be observed that the fact that tax rates have not been raised in many cases is irrelevant. The reality is that the effective tax burden – the amount of taxes extracted from the people – rises significantly. The reduction in the income tax exemption limit from the ludicrously low Tk. 75,000 per year (which would have brought even rickshaw-pullers into the income tax net) to Tk. 90,000 per year is certainly to be welcomed. It should be remembered though, that before last year’s budget the limit was Tk.1,00,000 per year, so that there has been a substantial lowering of the tax exemption limit in absolute and real terms (taking account of inflation). Together with the proposed modifications to the self-assessment system, the net effect would be to increase the harassment of lower- and middle-income tax payers.   

Turning to a discussion of some specific revenue measures introduced under the 2003-2004 Budget, there are a large number that appear to be beneficial for the upper income groups in this country, following the policy direction set under the last budget. The discussion here is focused primarily on the items of interest to the poor, but a large number of items consumed by a fairly broad section of the population have attracted higher customs duties or supplementary duties (SD), including fresh fish, spices, biscuits, soft drinks, refrigerators/freezers, and toys. [17] Moreover what is incongruous is that the revenue losses from duty reductions on the consumer items of interest to the rich appear to be offset by imposing higher customs duties and/or supplementary duties on basic commodities upon which the survival of the poor depends. [18] These basic items include:

• Sugar (raw) (SD raised from 20% to 40% - serial number (sl.) 12, with an estimated revenue gain of Tk 230.83 crores);

• Salt (SD raised from 60% to 75%, sl. 26):

• Powdered milk, admittedly an item no longer in the reach of the very poor in this country, (sl. 2 and sl. 3, SD raised from 20% to 25% in retail packing, from 10% to 15%, in bulk).

Some items that are among the few luxuries that had become affordable for even less affluent households during the past decade have also been targeted under the present budget:

• radio cassette players (SD raised from 0% to 15% - sl. 110);

• dry cell batteries and lead acid batteries (SD raised 30% to 40% - sl. 109);

• bicycles and other cycles (SD raised from 0% to 40% - sl. 126).

Some of the above are already identified as unlikely to eventuate. However, there is the risk that that the Government will follow its earlier practice (common during the period 1991-95) of initially rolling back some revenue measures in the face of public protest and then quietly effectively re-imposing them (through Statutory Regulatory Orders – SROs) later during the year. Even if some of the revenue measures listed above are withdrawn, the overall impact of the revenue package would have a strongly negative impact on the real incomes of the poor, through its effect on general inflation, the adverse impact on employment in the already weakening construction sector (due to sharp increases in duties on a wide range of building materials), and most importantly, by raising the cost of commodities that figure significantly in the consumption baskets of the poor.

There are a large number of revenue measures under consideration, and the need to strengthen the revenue base is not denied. It is also understood that it is not generally possible to have any real public debate on proposed changes due to the risk that powerful interest groups would be given the opportunity to lobby against specific measures. This means that a large degree of responsibility falls upon the Ministry of Finance to develop a sound package of revenue measures (based on an analysis of tax incidence) that impose the least amount of additional tax burden on the less privileged groups in society. What the choice of revenue measures in the current budget demonstrates is that the Government has had no compunctions about placing a large share of the additional revenue burden on the backs of the poor.   

7.2  Recurrent Expenditures

The serious concerns raised by the repeated over-expenditures on the recurrent side (revenue budget) have been noted earlier (see section 3).  Revenue expenditures in 2002-2003 were 15% higher than last year’s revised budget, which in turn was 9 % higher than the original budget. For 2003-2004 revenue spending is projected at Tk. 28,969 crores, representing a 14.5 % rise over the previous year’s revised budget figure of Tk. 25,307 crores. This rate of increase in recurrent spending programs suggests the continuation of a dangerous trend. It must be kept in mind that unlike the ADP, there is generally no difficulty in meeting or exceeding recurrent outlay estimates. While part of the increase reflects higher-than-expected external interest payments, the failure to achieve offsetting savings should be noted. The excessive overruns reflect both a failure of budget planning (and forecasting) systems and a breakdown in budget discipline and expenditure controls. [19]

Some specific allocations in the 2003-2004 deserve comment:

• Rural development: Raised from Tk. 87 crores to Tk. 287 crores; 

• Chittagong Hill Tracts Affairs: A decline from Tk 98 crores to Tk. 76 crores, that may reflect a changed policy stance (note:the allocation for development spending has risen );

• Education and Science: A rise of just 3.6%, from Tk 4188 crores compared to last year’s revised budget figure of Tk. 4041 crores, representing a decline in real terms;

• Defence: The figures in the Budget Speech are at variance with the Summary Tables on Revenues and expenditures. In the Budget Speech the Minister of Finance has indicated a figure of Tk 3995 crores as gross outlays and earnings of Tk. 457 crores from UN Peacekeeping Operations. The Summary Tables present a net figure of Tk.3534 crores. In terms of presentation of budgetary figures, following the IMF’s Government Finance Statistics Handbook, outlays for all agencies should be shown in gross terms. All revenues of all agencies of the Government should be shown as part of consolidated revenues.

The claims of a budget focused on poverty remain unreal without each and every spending program being assessed (and prioritized) for its contribution to poverty alleviation goals. This is necessarily a time-consuming exercise and given the current budget cycle such a review must be initiated now to be complete in time for the framing of next year’s budget. In particular, we may note the large additional outlays in certain spending areas that cannot be justified from the perspective of poverty alleviation.

The Government has followed the lead of the previous government in providing for agricultural subsidies, despite the opposition to such subsidies from the donor community. [20] The 2003-2004 Budget also provides direct support for micro-credit programs directly and for the further development of certain innovative safety net programs initiated by the previous government:

• A total allocation of Tk. 345 crores is allocated for new micro-credit programs (there has been concern at the Government’s entry into what has hitherto been a well-functioning sector );

• Monthly allowances under the Old-Age Allowance Programme have been raised from Tk. 125/person to Tk. 150/person and by raising the number of beneficiaries from 5 lakh to 10 lakh, which would involve an outlay of Tk. 180 crores (compared to Tk. 105 crores last year);

• Widowed and Deserted Women’s Allowances have been increased from Tk. 125/person to Tk. 150/person and the number of beneficiaries raised from 2.7 lakhs to 5 lakhs, which would involve an outlay of Tk. 90 crores (compared to Tk. 50 crores under the previous budget).

The above represent steps in the right direction, but the scale of resources involved is still relatively small when compared to other recurrent outlays on lower priority activities. It should be noted that in addition to budgetary outlays the build-up of contingent liabilities in the form of borrowing from the nationalized banks by government-owned enterprises (and what is termed the quasi-fiscal deficit) needs to be taken into account. Despite the publicity given to the closure of the Adamjee Jute Mills, the present Government (like the ones before it) has failed to decisively tackle the problem of losses (or a rundown of assets) at state enterprises that pre-empt a substantial chunk of total budgetary resources (World Bank 2003:19-25).

7.3  Development Spending

The proposed Tk. 20,300 crores allocation for the Annual Development Programme in the 2003-2004 Budget has not met with the favorable reception anticipated by the Government. The reason for this is relatively simple: the figure proposed appears quite unrealistic given the experience of 2002-2003. In the 2002-2003 Budget an ADP of Tk 19,663 crores was proposed, that was subsequently slashed to Tk. 17,653 crores in the Revised Budget, and is likely to reach only about Tk 13,000 crores in terms of actual implementation. This stands in sharp contrast with the Tk.16,240  crores in actual ADP spending (89% of the budgeted figure of Tk 18,200 crores) achieved in 2000-2001, in the last year of the previous government. The previous government’s willingness and ability to support a large expansion in the ADP was reflected in the remarkable (and fairly steady) increases in ADP size from the level of Tk.10,303 crores which represents the highest level attained under the previous BNP government (GOB, Economic Survey 2003, Table 4.4, p. 32). Given the likely magnitudes of shortfalls in ADP implementation (of the order of 35% on the basis of the situation in 2002-2003) a detailed examination of much smaller percentage shifts in sectoral allocations for 2003-2004 appears futile.

The net effect of the large real declines in ADP expenditures (in the last two years) on rural incomes and future growth prospects in the rural economy are likely to be severe. Cutbacks in ongoing rural works and the cancellation of planned works has reduced the employment of the rural poor, already hard hit by sharp increases in coarse rice prices over the past two years. [21] The low implementation rates reflect an apparent weakening in implementation capacity that has not been adequately explained in the 2003-2004 Budget Speech or other documents.

As noted earlier, the electorate fails to obtain any comfort from the stated intention to use a large part of the additional revenues raised in the coming year for development purposes, given the past record of the Government. Hence, for the ADP figures to be credible, it would have been necessary for the Government to highlight the areas where corrective actions to improve implementation would be taken.

 8. The Role of the Donors

 It would be wrong to put the bulk – or even a large part – of the blame for the failure to adopt pro-poor budget policies on the multilateral (or bilateral) donors, as some observers here and abroad tend to do. In fact the core elements of the World Bank’s stated poverty reduction strategy around the world - achieving “broad-based” growth, developing human capital and creating safety nets for vulnerable groups – seem eminently reasonable (World Bank, 1996b). The Government’s Interim Poverty Reduction Strategy Paper (PSRP) reflects these views and presents generally unobjectionable targets. The problem with this document, noted by many others in this country, is the absence of any concrete actions to achieve the poverty reduction goals indicated. Much seems to depend on the achievement of higher growth and a modified trickle-down effect. 

Despite the laudable intentions enshrined in the PSRP, in practice the policies of the Government appear driven by rather different considerations, which appears puzzling to a public not privy to the formal and informal undertakings made to the donors by the Government. This partly reflects conflicting priorities among the multilateral agencies – in many countries governments find themselves being pressured to invest more in social sectors while the IMF is insisting on a scaling back of outlays to safeguard real and potential threats to macroeconomic stability. There is no international capitalist conspiracy at work here – just the expression of certain political realities (that are very much exposed following the recent Iraqi War) reflected in the role of the largest shareholders in the multilateral agencies and the operation of bureaucratic processes in the multilateral agencies. What many people of this country find objectionable is the acquiescence of the government to the multilateral agencies taking the lead role in setting the poverty alleviation agenda. In particular it is difficult to accept the presumption of the donors that the government of this country – which despite its undemocratic mores is supposedly a democratically elected one - is unable or unwilling to fulfill its role in articulating the interests of the less-advantaged sections of society.            

The focus of the most powerful of the multilateral donors – the IMF (and to a lesser extent the World Bank) - has been on areas such as trade liberalization, deregulation of capital markets, and maintenance of external viability (and capacity to service debt) rather than growth and employment creation (Stewart 1995, Stewart 2001). While these are no doubt important objectives, and there is often a trade-off involved in terms of growth and emergence of internal and external imbalances that could pose a threat to macroeconomic stability over the medium-term, the IMF would always prefer to err on the side of stability and even stagnation and rather than growth. [22] Although it may be staffed by well-trained economists one must not lose sight of the fact that the Fund is a bureaucracy, and one that is dominated by the major shareholders (namely the United States and its allies). Fund staff are never blamed at the Executive Board for pushing a country into recession (or forcing a devastating fall in the real incomes of the poor) through an excessively tight financial program, but will always be censured for a loose program that leads to the emergence of macroeconomic imbalances. The resultant policy bias should be obvious.

The record of the IMF around the world is certainly a cause for concern, with the impact of the restrictive financial policies imposed by the Fund as a condition for its lending falling disproportionately on the poor. Argentina, Egypt, Thailand and Jamaica have all had bitter experiences with Fund programs at different times, and it is difficult to find cases where Fund programs helped cushion economic downturns and created the preconditions for recovery. One must therefore regard Fund involvement in the design or implementation of a poverty reduction program with a certain amount of unease. The World Bank and the ADB are generally perceived as being more sympathetic to the objectives of sustaining growth and poverty alleviation in developing member states, but it must be noted that on the basis of their division of responsibilities the Fund retains control over the macroeconomic framework, and to a large extent, monetary and fiscal policies.

The Fund’s influence over macroeconomic policies is amplified by cross-conditionalities – always denied but quite prevalent – through which it uses “moral suasion” to discourage bilateral donors from providing assistance unless agreement is reached (even disbursements to ongoing projects can be affected). Partly this reflects a legitimate desire on the part of other donors to obtain the validation of the Fund for a country’s economic policies to reduce lending risk. However in practice it also reflects a leveraging of lending power on the part of the Fund (Kibria 2003).  

Hence while at first glance the multilateral donor agencies appear to be very favorably disposed to budgets that place emphasis on growth, employment creation and poverty reduction, their influence is at best mixed. In practice their main focus tends to be very different and their unspoken and self-assigned mandate to speak for the poor in this country must be questioned given their record around the world. While their assistance (on concessional terms) must be welcomed, the full range of conditionalities attached to such funding must be exposed (fully, not selectively) to the media and the public. In particular, the policy bias embedded in their lending programs must be taken into account when taking decisions that would impact on the level of economic activity and employment. It is vital that poverty alleviation programs are designed and developed domestically, rooted in political reality with extensive participation by the representatives of the people at the national and local levels. In this context it is important that the political leadership of both major parties recognize the electoral value of developing a pro-poor budget, but for this to happen it would be necessary for the electorate to first be made aware of the enormous potential impact of budgetary policy on the welfare of the poor.

9. Conclusion

Despite the enormous obstacles to the development of a pro-poor budget in this country, the foregoing discussion should not be a cause for despondency. Greater awareness of the possibilities of designing such a budget – among the public and policymakers – will help the development of a coalition of interests that will be able to force major changes. The problem is that this may take a relatively long time, a delay that would condemn another generation to live in the type of inhuman poverty that exists in few other places on this planet. Some changes may be relatively easy – such as shifting the incidence of new revenue measures away from the poor or providing them with compensatory payments (as done in Australia for example, when it was realized that the VAT would impact heavily on the poor). Other changes would be extremely difficult, given the influence of powerful interest groups that have long enjoyed a disproportionate level of access to State resources, such as business houses that have been able to garner various direct and indirect financial subsidies. Some changes would need several years to implement effectively as they would require a shift in the mind-set of the bureaucracy and political leadership and a rethinking of the role of the budget, refocusing spending programs based on an assessment of their contribution to poverty alleviation objectives. 

However, it is quite possible that the political leadership of one of the two major parties will tackle these obstacles head-on, buoyed by strong electoral support. Despite the retrograde and generally anti-poor stance of budget policies reflected in the 2003-2004 Budget, one should not discount the possibility that even the leadership of the current government may be capable of taking the difficult steps needed to develop a strongly pro-poor budget, if they became convinced of the electoral value of such a move. Such a budget would involve a major restructuring of expenditures, decisive measures to ensure greater efficiency in government spending and a high degree of administrative decentralization. Budget reform over the medium-term would need to be combined with shifts in the pattern of budget allocations and the structure of revenues. These measures would not only help develop the basis for the sustainable, broadly-based and faster growth needed to lift the poorer half of our population out of poverty, they would transform the economic landscape of this country and create the equality of opportunity upon which any just society must be founded. 


References

Anand, S., and Kanbur, S.M.R.(1991), “Public Policy and Basic Needs Provision: Intervention and Achievement in Sri Lanka” in Sen, A. and Dreze, J. (1991), The Political Economy of Hunger: Endemic Hunger, vol. 3, pp. 59-92. 

Chenery, H., Ahluwalia, M.S., Bell, C.L.G., Duloy, G.H., and Jolly, R. (1974), Redistribution with Growth, OUP, Oxford.

Government of Bangladesh, Budget Speech: Annual Budget 2003-2004, Dhaka.

Government of Bangladesh (2003), Economic Survey 2003, Dhaka.

Hossain, M., Islam, I. and Kibria, R. (1999), South Asian Economic Development: Transformation, Opportunities and Challenges, Routledge, London.

International Herald Tribune, June 17, 2003.

IDPAA/Proshikha (2000), The Budget and the Poor, UPL, Dhaka

IDPAA/Proshikha (2002), The People’s Budget: An Illustrative Exercise Using Participatory Tools, UPL, Dhaka.

IDPAA/Proshikha(2002), The Budget-Making Process, UPL, Dhaka.

Islam, I. and Chowdhury, A. (1997), The Asia-Pacific Economies: A Survey, Routledge, London and New York.

Islam, I. (2003), “Human Development, Democracy and Economic Growth: Bangladesh in the 20th Century”, in Hossain, M., Nurun Nabi, A.K.M. and Islam, I. (2003), (eds.), Bangladesh’s Development Agenda and Vision 2020: Rhetoric or Reality? , UPL, Dhaka.

Khan, A.R. (2003) “What Can Be Done to Contain Rising Inequality in Bangladesh?” (mimeo), paper presented at  a Seminar on Accelerating Growth and Poverty Reduction in Bangladesh, Dhaka June 26-27, World Bank and Bureau of Economic Research, Dhaka University.

Kibria, R. (2003), “The Political Economy of Reform: Designing, Initiating and Sustaining Public Sector Reform in Developing Countries”, in Hossain, M. Nurun Nabi, A.K.M., and Islam, I. (2003) (eds.), Bangladesh’s Development Agenda and Vision 2020.UPL, Dhaka.

Mahmud, Wahiduddin (2003), “Strategy for Pro-Poor Growth in Bangladesh”, (mimeo), p. 14., a paper presented at the Seminar on Accelerating Growth and Poverty Reduction in Bangladesh, Dhaka June 26-27, World Bank and Bureau of Economic Research, Dhaka University.  

Majumdar, B.S.(2003), ““Reduction of budget for local government: What does it imply?”, Daily Star, July 7, 2003, p. 5

Mandal, M.A.S., and Palmer-Jones, R.W. (2003), “Agricultural Growth and Rural Poverty in South Asia: Lessons Learnt and Issues to be Addressed”, in Hossain, M. Nurun Nabi, A.K.M., and Islam, I. (2003) (eds.), Bangladesh’s Development Agenda and Vision 2020.UPL, Dhaka.  

Mill, John Stuart (1859), “On Liberty”, Utilitarianism (1974), Collins/Fontana, Glasgow.

Sen, A. and Dreze, J. (1998), Hunger and Public Action, OUP, Oxford. 

Smith, Adam (1776), The Wealth of Nations, Everyman’s Library Edition (1991), London

Stewart, F. (1995), Adjustment and Poverty: Options and Choices, Routledge, London.

Stewart, F. (2001), “Adjustment and Poverty in Asia: Old Solutions and New Problems”, in Mahmud, W. (ed.), Adjustment and Beyond: The Reform Experience in South Asia (Proceedings of the IEA Conference held in Dhaka, Bangladesh), IEA Conference Volume No. 128, Palgrave, UK.

Tawney, R.H. (1961), Equality, Capricorn Books edition, New York, quoted in Becker, Gary S., (1993)  Human Capital: A Theoretical and Empirical Analysis with Special Reference to Education, University of Chicago Press, Third Edition, Chicago, p. 137.

Taylor, L. (1983), Structuralist Macroeconomics; Applicable Models for the Third World, Basic Books, New York.

World Bank (1996 a), Bangladesh: Government That Works: Reforming the Public Sector, UPL/World Bank, Dhaka.

World Bank (1996b), Poverty Reduction and the World Bank: Progress and Challenges in the 1990s, Washington DC, World Bank.  

World Bank (1998), Public Expenditure Management Handbook, World Bank, Washington D.C.

World Bank (2002), Bangladesh: Improving Governance for Reducing Poverty, World Bank, Washington DC.

World Bank/ADB (2003a), Poverty in Bangladesh: Building on Progress, World Bank, Washington DC.

World Bank/ADB (2003b), Bangladesh: Public Expenditure Review, World Bank/ADB, Washington D.C.

Yunus, Muhammad (with Alan Jolis) (1998), Banker to the Poor: The Autobiography of Muhammad Yunus, Founder of Grameen Bank, UPL, Dhaka.

 

---