Poverty in Bangladesh

Poverty in Bangladesh

 I. Poverty: Variation on a Theme

A. Silent Ascent of the Poor

Successive studies have shown that the cost of eliminating endemic poverty within the shortest possible period is already within the reach of most developing countries. Nevertheless, the weight of global burden of poverty continues to be enormous, with 1.31 billion absolute poor persisting below the $1 a day line adopted for international comparison in 1993. About 40 per cent of them live in South Asia. Within South Asia, the incidence, depth, and severity of poverty are highest in the Eastern region (Ravallion and Chen 1996), and within the latter, Bangladesh has the largest absolute number of the poor. Why has this higher incidence of poverty persisted? What are the factors behind it? What policies are needed to reduce it? These are the most urgent questions, which need to be addressed today.

A number of changes have occurred in the poverty profile of Bangladesh over the years, with considerable successes in some respects, and notable failures in others. This paper highlights both the movements. The most visible change has taken place in the way the poor themselves have positively responded to whatever development has reached them. The silent ascent of the poor, including the women and the marginalized, has brought to the fore new opportunities as well as new challenges. An in-depth review of what caused these changes demands further scrutiny, but a few key factors are analyzed in this paper.

The paper reviews a broad range of questions and concerns relating to poverty. Chapter 1 (Introduction) outlines some basic concepts and measures of income-poverty and human poverty. Chapter 2 (Trends in Poverty over Time) describes and explains the trends in income-poverty and human poverty. It also brings out the underlying spatial variation in social indicators. Chapter 3 (Policies and Institutional Measures for Poverty Reduction) pin-points the constraints to poverty reduction in Bangladesh, and reviews and recommends policies and institutional measures for faster reduction in income-poverty and human-poverty.

B. Poverty: Concepts and Measures

The paper is premised on a multidimensional understanding of poverty. It articulates the need for appropriate factoring of the “voices” of the poor. It underscores the need for alliance between science and the people. Scientifically reasoned arguments and broad-based participatory review can be effectively combined to arrive at sustainable solutions.

1. Multidimensionality of Poverty

Poverty is a multi-dimensional concept. It focuses on various aspects of deprivation, both income and non-income. It reflects disempowerment, insecurity against shocks, lack of opportunities. It has manifold expressions and, indeed, many roots. It is about income deprivation, about shortfalls in consumption and inadequate supply of nutrition. It is about poor access to education and low physical asset bases. It is about risks, uncertainties, and vulnerabilities. It is about personal insecurity as much as it is about lack of food security. It is about crisis coping capacities. It is about self-development initiatives. It is about dismal state of health and health care access. It is static and dynamic, transient and chronic, sporadic and systemic. It is seasonal as much as it is spatial. It is inter-generational. It is about all known vicious circles, of low savings, low investment and low growth. It is about the quality of growth and being left out of growth. It is also about personal freedom, alienation, and social justice. It is expressed in each of these and all of these together. Given the multidimensionality, it is not difficult to see why all routes, income and non-income, should matter for poverty reduction. Nevertheless, it is useful to highlight the most important routes for attacking the multidimensional aspects of poverty.

There are four broad avenues that can help to think strategically about poverty. These are (i) lack of pro-poor economic growth; (ii) lack of human development; (iii) lack of social safety net; and (iv) lack of participatory governance. All the four avenues have important poverty effects, influencing both the income and non-income dimensions of poverty.

Lack of pro-poor economic growth points to the importance of making the process of economic growth broad-based. This requires attention to policies and measures to increase income, consumption, employment, and wealth of the poor. Lack of human development of the poor points to the need for basic capability-enhancing measures such as education and health.

Lack of social safety net focuses on the vulnerability dimension. Vulnerability to shocks may be triggered by many factors. There are shocks of covariate nature (such as, natural disaster which cuts across the poor and the non-poor groups), or of idiosyncratic nature (such as health-related hazards). Whatever the nature of the shocks, the poor are found more susceptible to shocks than the non-poor and they take a much longer time for recovery.

Lack of participatory governance (“voice and freedom”) has direct and indirect welfare effects. First, it is important in its own right. It captures the denial of basic human freedom, freedom of choice, freedom to live, act, think, and participate in the democratic process. Development is to be measured in terms of the degree of freedom it provides. Second, it has effects on economic and social variables and hence, is instrumentally important. For instance, lack of participatory governance may lead to the choice of inappropriate projects, or make decentralization efforts ineffectual, or even counter-productive. Greater “voice” would lead to transparency and help combat corruption. Effective governance requires devolution of power to the local bodies as well as a functioning local democracy since both are essential ingredients of the “voice” mechanism. However, both may be capacity-constrained. Appropriate policies would help develop the governance capacity of the local bodies, on one hand, and quality of the human and social capital essential for democratic graduation, on the other.

2. Plurality of Perspectives

Poverty as analytical construct cannot be reduced to quantitative aspects of measurement alone. Preferences of researchers may not always correspond to the preferences of the people: households classified as non-poor by some pre-determined “poverty-line” may actually be reckoned to be poor by the community at large, and vice versa. It is, therefore, important to recognize the heterogeneity of voices and perspectives within and beyond poverty, heterogeneity expressed in economic as well as cultural terms such as class, gender, caste, ethnicity, community, etc.

The existence of heterogeneity does not imply that broad-based consensus on policies is a mirage. However, consensus can only come if there is enough listening to the different perspectives on the same issue. One of the advantages of the participatory approach lies in the identification of issues that can be taken up by the researchers for subsequent analysis. Apart from problem-identification, a broad-based consultation process can also help identify concrete institutional solutions to problems for which “theory” (in its current level of sophistication) often may not have any ready answer. Participatory “hearings” of the poor would generally help to identify the key problems and test the validity of the policy solutions. Poverty measures and policies must be “owned” by the poor themselves. Without participation of the poor, the effective implementation of many of the policies and programs cannot be ensured.

3. Poverty Measures

In this paper, we focus on both income and non-income dimensions of poverty. The income dimension of poverty–termed as income-poverty–is defined by Foster-Greer-Thorbeck (FGT) class of poverty measures (see, box 1). Human Poverty Index (HPI), which was originally proposed by Anand and Sen (1996), may be viewed as the synthetic expression of non-income dimensions of poverty.It is in the latter sense we have used the term “human poverty” throughout this paper (see, box 2).

Box 1: Measures for Income-Poverty Calculations
The three poverty measures used in this study attempt to capture three aspects of poverty: its incidence, its depth, and its severity.The head-count index (H), given by the percentage of the population living in households with a consumption per capita that is less than the poverty line. This can be interpreted as a measure of the “incidence” of poverty. The measure has the advantage that it is easy to interpret, but it tells us nothing about the depth or severity of poverty.

The poverty-gap index (PG), defined by the mean distance below the poverty line as a proportion of that line (where the mean is formed over the entire population, counting the nonpoor as having zero poverty gap). One can interpret this as a measure of poverty “depth.” Its disadvantage is that it is unaffected by changes in inequality among the poor.

The squared poverty-gap index (SPG) of James Foster, J. Greer, and Erik Thorbecke, defined as the mean of the squared proportionate poverty gaps (again the mean is formed over the entire population, counting the nonpoor as having zero poverty gap). Thus the poverty gaps are weighted in aggregation, with greater weight given to larger gaps, and where the weights are simply the poverty gaps themselves. This simple change to the conventional poverty-gap index allows the index to reflect changes in the “severity” of poverty, in that it will be sensitive to inequality among the poor.

In the Bangladesh context, the poverty line corresponds to 2112 calories per person per day. There are several methods for setting the poverty line. Of these, the approach which is based on costing of a given food bundle (corresponding to 2112 calories per person per day)… the so-called fixed-bundle approach … is found to be better than other methods such as Food-Energy Intake (FFI). For details on the issue, see Ravallion and Sen (1996).

Box 2: Measures for Human-Poverty Calculations
The HPI was proposed in a paper by Sudhir Anand and Amartya Sen (1996), and implemented in global HDR 1997. Conceptually, the approach represented a step forward in capturing deprivations with respect to the key non-income dimensions of welfare. The analytical relationship between HDI and HPI is very much similar to the one that exists between economic growth and income-poverty. Two immediate differences between HDI and HPI may be highlighted. First, HDI seeks to highlight the aggregate social and economic achievements on an average for the entire population of the country. In contrast, HPI focuses on the deprived segments of the population. This is consistent with the standard practice of confining poverty measures exclusively to the deprived segments. HDI, in contrast, is an inclusive measure for the entire population. Second, ingredients going into HPI concentrates on health and education, while those going into HDI contain an additional component, namely, a transform of the purchasing power parity adjusted GDP per capita. HPI is composed of three indicators (one of which is a composite of three sub-indicators) capturing three types of deprivations: deprivation in health, deprivation in knowledge, and deprivation in overall economic provisioning. These dimensions are given equal weights in the construction of HPI .

Deprivation in health is indicated by vulnerability to death at a relatively early age as quantified in the percentage of people expected to die before the age of 40 years;

Deprivation in knowledge is captured by the percentage of adults who are illiterate;

Deprivation in overall economic provisioning is quantified in three variables, namely,
(i) percentage of people without access to safe drinking water, (ii) percentage of people without access to health services, and (iii) percentage of children under 5 who are moderately and severely underweight.

II. Trends in Income and Human Poverty

A. Trends in Income-Poverty

1. Long-term Change in Income-Poverty

Bangladesh made notable progress in income-poverty reduction since Independence. The percentage of rural people below the poverty line declined from 71 percent in 1973/74, declined to 51 percent in 1995/96 (Hossain and Sen 1992; Sen 1998). The rate of decline was remarkably fast in urban areas, where the incidence declined from a level of 63 percent to 26 percent during the same period. At the national level, this translates into a reduction from 70.6 percent to
46.5 percent. This gives a long-term poverty reduction rate of 1.55 per cent per annum. While this rate is much lower than the historically observed rates in the high-performing Asian economies (HPAEs) of East and South-East Asia, it was a commendable achievement, after decades of slow economic growth.

Opinions, however, differ on the extent to which poverty reduction has occurred during the entire period of post-Independence. Critics point out to the problems of making long-term comparison of poverty numbers for the period 1973/74 through 1981/82 with those relating to the later period of 1983/84 through 1995/96. We, therefore, estimate the rate of poverty reduction separately (Tables 1.1 and 1.2). Available evidence suggests that the rate of rural poverty reduction was higher during the period between 1973-81 compared with the subsequent period of 1983-95 (1.1 vs. 0.4 per cent per year). The matched difference between the two periods is negligible (2.97 as against 2.93 per cent). On balance, the incidence of national poverty declined at a faster rate in the first period than in the second period (1.29 vs. 0.92 per cent).

2. Slow Progress during the Eighties

Poverty comparisons are less problematic across the surveys carried out since the early eighties. Existing studies converge on this score (Khan 1990; Ravallion and Sen 1996; Mujeri 1999). Nevertheless, the very sharp fall of poverty between 1983/84 and 1985/86 appears to be problematic at best, hardly explainable in terms of standard economic variables. We, therefore, excluded 1985/86 from the present discussion, as it adds more heat than light in assessing the poverty trends since early eighties.

Two broad trends are discernible. First, the overall weight of evidence suggests marked instability in poverty reduction beginning with 1983/84. Rural poverty marginally declined between 1983/84 and 1988/89, slightly worsened between 1988/89 and 1991/92, and slowly improved thereafter (Table 1.2). Second, the overall gain over the entire 12-year period appears extremely modest. Thus, the head-count index of poverty declined at a rate of 0.4 percent per year during 1983-95.

The slow progress in poverty reduction is robust to the choice of poverty measures. All available studies indicate marginal decline in terms of more distributionally sensitive poverty measures, namely, poverty-gap and squared poverty-gap index. This is quite striking, given the persistent policy emphasis on “poverty alleviation” as the overarching goal both in the articulations of GoB, NGOs, and development partners during this period.

3. Recovery of the Nineties

The poverty reduction performance improved during the nineties. The rate of reduction in rural poverty during the first half of the nineties was in the order of 0.9 percent per annum which may be compared with the matched progress of 0.2 percent per year recorded during the immediately preceding period of 1983-91. The corresponding difference between the two sub-periods is more pronounced in case of urban poverty reduction. Thus, the rate of reduction in the urban poverty incidence was 2.2 percent per annum during 1983-91; it rose to 5.4 percent per annum during 1991-95.

In summing up the trends in poverty in Bangladesh over the last two decades two basic points need to be highlighted. First, the overall progress during the period between 1983/84 and 1995/96 was quite modest. Second, progress in income-poverty was considerably slower in rural areas than in urban areas. Given the fact that over 80 per cent of the national poor still live in rural areas, the slow progress in rural poverty appears to be the key challenge facing the country in the coming decade. In the next section, we shall try to provide some light on the processes that led to such a modest outcome.

B. Trends in Human Poverty

1. Progress in Human Development

The value of human development index (HDI) has roughly doubled in three decades, from 0.166 in 1960 to 0.309 in 1992, yielding an annual average rate of increase in the order of 2.7 percent per year (Table 1.4). The progress has been faster in the nineties: between 1992 and 1996/97, the HDI value has increased by 40 percent, with an implied average increase of 9.3 percent per year. Note that both over the longer-term as well as over the recent period the pace of human development was faster than the rate of economic growth.

2. Progress in Reducing Human Poverty

Human poverty focuses on three aspects of human deprivation: deprivation in longevity, deprivation in knowledge, and deprivation in economic provisioning. Methodology for constructing the human poverty index (HPI) is given in Table 1.5.

The results show that the incidence of human poverty has declined from 61.3 percent in 1981-83 to 47.2 percent in 1993-94, and dropped further to 40.1 percent in 1995-97. At the same time the incidence of income-poverty at the national level declined from 52.3 percent in 1983/84 to
46.6 percent in 1995/96.

Two aspects are worth noting in this context. First, progress on reducing human poverty was faster than the matched progress recorded in case of income poverty. The average rate of decline in HPI was about 3 percent per year during 1981-97. The matched figure for income-poverty reduction at national level was only 1 percent over 1983-95. Second, the progress in reducing human poverty was faster in the nineties than in the preceding decade.

C. Income vs. Human Poverty in Urban Areas

Analysis of various dimensions of urban poverty points to the asymmetry between income and non-income dimensions of poverty. While the incidence of urban income-poverty is about half the level of rural income-poverty, the urban poor appear to lag behind their rural counterparts in several important non-income dimensions such as schooling and immunization. The relative gap in human capital is particularly striking and would remain as a binding constraint in the way of further upward mobility on the part of the urban poor. The evidence also points to the presence of a significant gender dimension to urban poverty. Greater female work force participation responds to the compulsions of urban survival. Greater female work force participation may also entail further deterioration of the nutritional status of children. We do not have hard data on urban violence, but the incidence of such violence is widely regarded to be much higher in urban areas than in rural areas. All these show that reliance on the income route alone would not necessarily lead to higher social results, especially when migration to urban areas takes place under a variety of circumstances. There is a compelling need to act on both income and non-income fronts, with particular emphasis on social and environmental dimensions inflicting the urban poor.

D. Spatial Asymmetry between Income and Human Poverty

There is a general correspondence between the regional ranking of income-poverty and human-poverty. Rajshahi division, for instance, occupies the bottom position on both counts. However, the matching is not exact: there are very important omissions. Thus, Chittagong division has the lowest level of income-poverty but highest level of infant mortality rate (Table 1.6). In terms of life expectancy and adult literacy the otherwise most affluent division occupies the second lowest position. This suggests that there are asymmetries in the spatial distribution of income and human poverty.

Second, the regional variability is considerably higher for some human poverty indicators than others. Divisional data show considerable variability in adult literacy, immunization rate, and income-poverty. As for other indicators such as life expectancy and infant mortality the variability is much lower. This suggests that there is a greater need for developing a spatial focus in designing policies when it comes to tackling the issues of primary education, primary health and income-poverty reduction.

A third aspect relates to the presence of rather strong geographic effects on poverty. This is in spite of the fact that the country is relatively small in terms of area coverage and relatively homogeneous in terms of ethnic composition. One needs to go beyond the division or district to identify the pockets of severe distress, i.e., areas which are more vulnerable to widespread starvation and intensified destitution during bad agricultural year and/or during the routine lean period even during a normal agricultural year. Thus, a significant differentiation in poverty may be observed even within the Rajshahi division. Bogra, for instance, represents the fastest growing area in that division on account of both crop agriculture and small-scale industries, putting the district on top of the list in terms of higher average affluence and lower income-poverty. In contrast, areas located in the river-erosion belts of Kurigram, Sirajganj, and Gaibandha (all falling within the Rajshahi division) have the highest income-poverty level in Bangladesh. Poverty mapping developed by the 1991 GoB Task Force Report on Poverty Alleviation, and, subsequently by the World Food Programme (WFP), can be consulted to identify the areas of extreme distress to allow more spatial focus in the policies for addressing poverty reduction.

The upshot of the above is to underscore the need for acting simultaneously on both the fronts, income and non-income. Even in cases where the income route is instrumentally important in achieving the non-income (human-poverty) outcomes, the latter cannot simply be seen as derivatives of the income factor.

E. Progress in International Perspective

The continuing trend of improvement in the human poverty situation in Bangladesh in the nineties compares favourably with the matched progress of LDCs as a group as well as South Asia (Table 1.7).

Two aspects are noteworthy. First, Bangladesh performed relatively well in the broad area of human development. Better performance in social indicators comes through even in long-term comparison. Key social indicators such as life expectancy at birth, infant and under-five mortality rate display faster rate of improvement compared with average LDC over 1970-97. The rate of improvement was also better than in its South Asian neighbours. In respect of one crucial indicator of social change–“total fertility rate”–Bangladesh’s long-term progress was considerably better than its South Asian neighbours, and was way above the corresponding average progress in LDC. Note that in making assessment of progress in social indicators we are only talking about the rate of change, and not about the level, where Bangladesh still considerably trails behind its more affluent neighbours. Nevertheless, the trend of “catch-up” is noteworthy. Second, on a comparative scale, Bangladesh’s efforts are seriously lagging behind in the area of accelerating per capita income growth as well as faster reduction in income-poverty.

Statistics on faster pace of human development combined with slower progress in economic growth are striking. It may mean that potential gains to be derived from human development failed to translate commensurately into additional growth possibilities. This happens when the stock of human capital is of relatively poor quality, or when there are not enough market opportunities to effectively utilize the existing stock of human capital. In both cases the return to physical investments falls, and with rising incremental capital output ratio (ICOR)–caused by distorted factor prices–the rate of growth declines. The overall situation is aggravated further by a public culture that thrives on misgovernance, reducing the effectiveness of market and non-market institutions.

III. Policies and Institutional Measures for Poverty Reduction

This chapter is about policies and institutional measures that can reduce poverty. Here we draw broad policy implications of the findings presented in the previous three chapters. We also consider new and potential areas of policy interventions that can lead to faster poverty reduction in Bangladesh. National poverty reduction cannot be achieved only via national policies. A number of anti-poverty policies have transnational dimensions and would require international assistance, coordination and support. Some of the policies have regional dimensions as well. They are likely to succeed better if regional and sub-regional initiatives can be undertaken. In line with the framework adopted in this paper, we discuss policies not only in terms of their impacts on the income-dimension of poverty, but also on crucial non-income dimensions of well being and agency.

A. Pro-Poor Economic Growth, Human Development, Social Safety Net, Participatory Governance: A Policy Framework

Poverty can be reduced in different ways. This is because poverty is caused by many factors. Despite the diverse nature of causes of poverty, one can group them into some broad policy-relevant categories. As discussed in Chapter 1, there are four broad avenues that can help us to think strategically about anti-poverty policies and institutional reforms. The first set of policies expands scope for pro-poor economic growth for increasing income and employment of the poor. The second set of policies fosters human development of the poor. The third set of policies provides social safety net to the poor against various anticipated and unanticipated income (or, consumption) shocks. The fourth set of polices favourably influences participatory governance and enhances voice of the poor by strengthening women’s empowerment, by improving the performance of the existing anti-poverty institutions and by removing the institutional hurdles that stand in the way of social mobility of the poor.

Note that the above policies help to reduce poverty at all levels: both inside the family and at the level of household, and at successive aggregations–locally, regionally, nationally, and globally. Considerable synergies exist among the four sets of policies. For instance, policies that empower the poor by strengthening local-level democracy can indirectly lead to greater bargaining power in the wage market. It can also lead to better health delivery by exerting popular pressure for improving the management of local health centres. Thus, greater voice can lead to greater income-earning opportunity. Better health delivery, in turn, can reduce income-shocks associated with the sudden illness of breadwinners.

Before we proceed to discuss the specific issues under each set of policies, it is useful to recognize the synergies that exist among them. These policies should not be seen as a random wish-list of all possible and potential efforts. Not all of these policies can be implemented at the same time because available resources are limited. Nor is it feasible as some policies (such as asset distributive reforms) are usually more contentious, with winners and losers, and therefore constrained by political economy. Strategic thinking would suggest that often it is useful to start from the issues that are of common concern, or issues that are likely to yield maximum possible support. This, therefore, suggests the need for policy sequencing.

B. Lessons from the Urban-Rural Contrast in Poverty

The contrasting experience of rural and urban development may serve as the central theme in thinking about policies and institutions for faster poverty reduction. In urban areas, income-poverty declined at a faster rate than in rural areas. The rural poor, however, appears to be better-off in terms of non-income indicators such as education and health than their urban counterparts. Access to microcredit, which has important social effects especially on women empowerment, is much lower among the urban poor women than in case of rural poor women. The crux of the contrast is “higher income growth-lower social development” for urban poverty, while it is “lower income growth-higher social development” for rural poverty.

Can the “best practices” in rural and urban developments be combined? The answer is—yes, and may be outlined in the following directions. First, even though there has been a sharp increase in relative inequality in the nineties, the rate of urban poverty reduction was substantially higher than in rural areas (5.43 per cent per annum vs. 0.85 per cent per annum; see Table 1). This implies that urban growth was more than offsetting the adverse effects of increase in inequality. Indeed, the growth in urban average income as captured by the consumption data was 5.7 per cent per annum compared with 1.6 per cent recorded for the rural areas. The moot question is—how to replicate the urban high-growth experience in case of rural areas without, however, losing the existing targeted focus by way of greater allocations for the social sectors and microcredit for the rural poor (more about the rural growth stimuli later).

Second, as pointed out earlier, urban poor residing in slums trail behind their rural counterparts in respect of a number of indicators such as school enrollment, infant mortality, morbidity and the like. This is because slum dwellers do not have legal status as settlers in urban areas, which explains their marginal access to public services. This has adverse health and environmental implications for the urban poor. Rural targeted programs such as microcredit, preventive vaccination, food-for-education, incentives for girl’s education (to name the few) are virtually missing in the urban areas The relatively poor social outcomes for the urban poor show that the path of income growth-mediated security holds out little promise in the context of Bangladesh. Social sector public programs need to be undertaken to supplement what may not be achieved through the route of private purchasing power alone.

C. The Regional Dimension

Earlier we noted considerable spatial variation in poverty rates not only across the rural-urban divide, but also across districts. Relatively deprived regions are often also the natural habitat of the relatively deprived communities. The persistent regional difference can be corrected through judicious public interventions. The most poverty-stricken districts need to be prioritized in allocating investments and in targeting assistance. The unit cost of delivering one dollar of assistance tends to be higher in infrastructurally less developed areas (such as CHT) than in case of more developed areas. However, the financial rate of return criteria should not be the only criteria in judging the merit of particular investment. It is important to factor into the analysis of public policies and public investments some normative considerations of social justice. Poor regions, like poor people, must get priority when it comes to designing policies and programs for reducing national poverty.

D. Policies for Increasing Pro-Poor Economic Growth

From the viewpoint of encouraging pro-poor economic growth opportunities for the poor, a number of macro and sector policies may be identified. The review below attempts to take into account what has been done in the past in a particular policy area and draws attention to the possible broadening of the scope of anti-poverty interventions.

1. The Role of Macroeconomic Stability

The strategy of non-inflationary pro-poor economic growth requires a stable macroeconomic framework. This would create an enabling environment for economic sector policies crucial to accelerating the pace of poverty reduction. In this sense, macroeconomic policies matter for poverty reduction. Indeed, inappropriate macro policies can be as disastrous (if not more) for poverty reduction than the shortfall in any of the sector policies discussed below. The poor, however, is not a static category, as there is a continuous movement in and out of poverty. Wrong macro policies can easily lead to the swelling of the ranks of the poor. This has been amply demonstrated by the adverse social effects of hyperinflation in a number of Latin American countries in the eighties, and also borne out by the recent East Asian Financial Crisis.

A fiscal policy that creates inflation, crowds out private investment, and undermines social sector allocations would have adverse implications for the poor. A wrong exchange rate policy (as with the markedly overvalued exchange rate) would have balance-of-payment implications over and above the point that long adherence to such policies would lead to rising ICOR through distorted factor prices and, hence, lower long-term growth and less scope for poverty reduction. Lack of prudent regulations in the financial sector in the face of volatility would hurt the small investors, as suggested by the recent episodes of financial crisis in several Asian countries.

The cross-country literature on the policy determinants of long-term growth and poverty reduction points to the importance of factors such as openness and inflation (Werner and Sachs 1995 for evidence on openness; Ravallion 1997 on inflation). Inflation in particular has a direct bearing on the welfare of the poor (most of whom are net consumers) through food prices. In short, a stable macro policy framework for maintaining internal and external balances is a necessary pre-condition for higher long-term growth and sustainable poverty reduction.

2. New Technology for Higher Agricultural Growth

The rate of rural poverty reduction is positively correlated with the rate of agricultural growth. Agriculture merits attention because of its importance as a source of employment. Agricultural growth may be encouraged through various policies, ranging from new technology to credit for the small farmers. Agricultural productivity may be increased by pro-poor agrarian institutions, as in the case of tenancy reforms. The issues of farm credit and appropriate rural institutions are taken up later. In this sub-section we shall concentrate on new technology in agriculture, especially in crop agriculture, which still constitutes about 75 percent of the agricultural value added.

New HYV rice technology—the so-called Green revolution—helped to accelerate agricultural growth. During the first phase green revolution progressed mainly in the heartland where small peasantry dominated. This resulted in better distribution of the benefits from increased production and hence magnified the favourable impact of green revolution on rural poverty. Mid-1980s onward, the green revolution expanded into areas dominated by larger landholdings, with less encouraging results in terms of poverty reduction. The other challenge is to spread HYV technology into areas with ecological vulnerability. Nevertheless, the progress achieved was by no means inconsequential. In fact, production of cereals in Bangladesh during the period between early eighties and early nineties grew at about the same rate as West Bengal (2.4% per annum), which has registered one of the fastest growth in this respect among the Indian states.

The policy regime within which this success was achieved points to the importance of both state and the market though some issues such as de-subsidization as well as unbridled privatization of the mode of delivery of agricultural inputs remain contentious. Whatever success was achieved in accelerating food production would not be possible without the pro-active role of the state in the dissemination of information as well as subsidized provision of inputs, especially in the first phase of green revolution. The role of subsidy declined in the eighties, mostly under the pressure of fiscal adjustment, with increasingly more emphasis on the market for distribution of agricultural inputs. Farm investments in irrigation also benefited from the liberalized import of agricultural machineries, which has led to some spectacular recovery in food production in the turn of the nineties. However, most observers feel that this effect was of transient nature and soon disappeared as the price regime of low rice prices set in (Abdullah and Shahabuddin 1997 and 1998). In the second half of the nineties more explicit targeting mechanisms were used for the delivery of agricultural inputs along with the restoration of some measures for farm subsidies.

Although agricultural growth has accelerated visibly in the second half of the nineties the sustainability of such growth in future remains questionable. The question of appropriate mix of incentives is important for the farmers, but incentives per se would not produce higher growth. The latter would critically depend on the ability to adopt new technology. The average yield level achieved in Bangladesh in respect of HYV rice was low compared with other Asian comparators to begin with. It has been declining further in course of the adoption process. There has not been a visible breakthrough, however, in global rice research in many years since the advent of green revolution. The pace of progress has slowed down in the face of declining funds for national and international rice research for the disadvantaged flood-prone ecosystem that characterizes countries such as Bangladesh. While this will require long-term solutions, efforts should be made without delay to reap the full advantage of the recent advances in science and technology (including biotechnology) for raising the existing yield levels.

The issue at stake is not a hypothetical one. Excluding the city-states, Bangladesh has one of the highest population density in the world. Already between the two censuses (1983/83 and 1996), agriculture has lost about 1 million hectare of arable land under population pressure and urbanization (in which, again, it ranks among the highest in the developing world). Although population is growing at a slower rate than twenty years ago (the current rate is 1.8 percent), the pressure of absolute number is on increase. This combined with the fact of rapid urbanization means that agriculture has to feed a growing population (and future urban towns) with rapidly declining arable land. This cannot be done without substantial increase in the yield level.

3. Policy Support for Agricultural Diversification

With rise in per capita income, demand for non-rice (such as vegetables and pulses) and non-crop products (such as livestock, poultry, and fishery) crops increases. Production of these products for markets would increase further if the share of urban population rises at a rapid pace. Fishery and livestock/ poultry sectors have grown fast during the nineties, partly in response to these factors. However, domestic production has not been able to meet the domestic demand for these products: an increasing share of the market is being accrued to the agricultural trade with the neighbouring country, especially through cross-border channels. A recent Bangladesh Bank study shows that higher yield level of non-rice crops in the neighbouring country is mainly attributable to the relatively superior institutional performance in the areas of research, extension, marketing, and storage facilities. The failure of agricultural diversification is particularly telling, especially since for many of the non-rice products the financial rate of return is higher than in case of HYV rice (Rahman et al 1993; Zohir 1994).

There is something of a paradox here in that, despite successive policy emphasis, crop diversification with considerable potentials for employment generation has failed to take off in Bangladesh. Given the relatively fast development of road infrastructure one would expect that the production of perishable goods would be encouraged. While there has been some improvement in the marketing situation following the construction of Jamuna bridge, this is still far below the desired level. One proximate explanation for this failure relates to the fact that the entire system of agricultural extension has been in a state of shamble for a long time, and no serious effort is discernible to redress this pathetic situation. For example, basic soil testing for judging the suitability of particular crop choice has not been done in at least half the territory of Bangladesh. This is one of the reasons why farmers had to go through a painful learning process just to find out what grows well on a particular soil and what combination of inputs is necessary to maximize the yield in that soil. Besides extension, the lack of adequate marketing, storage and transport network remains a critical bottleneck in the diversification process. An efficient system of extension is a must in the context of liberalized agricultural trade, and priority attention needs to be given to this aspect of the agricultural policy.

4. Expanding Credit Access

The poor have limited access to credit because they lack collateralisable assets. Imperfections of the credit market have adverse implications for poverty reduction both in the short and the long term. Due to lack of credit the poor cannot cope with the short-term exigencies arising out of sudden income (or consumption) shocks (or can cope only through less desired coping methods such as disinvestment of assets and forward sale of labour). In most cases they have to absorb the shocks, with implications for a drop in the food entitlements and deterioration in the nutritional status of children. Limited access to credit has implications for long-term intergenerational poverty. If the poor had access to credit they could have financed the educational expenses of their children as a way of climbing out of poverty. It is in the context of this bigger picture that one needs to assess the policy significance of Grameen Bank-type innovative schemes of targeted microcredit.

There is a need for building upon the success of microcredit achieved and sustained till date to service the credit needs of the various segments of the rural poor. The mechanism for delivery of microcredit may be suitable for banking with the women folks from the middle-poor households, but may need to be modified for meeting the needs of the small farmers who are at the threshold poverty line. Credit needs would vary for the farm and non-farm borrowers, with different implications for seasonality in credit demand. The same may be said of the extremely poor households who have very different risk-bearing capacity. These are new institutional challenges that require innovations in the mode of delivery. However, it is important not to undermine the value of what has already been achieved by way of microcredit.

Expansion of non-crop and non-farm employment played a compensatory role in mitigating the effects of slow agricultural growth during the first half of the nineties. The role of this factor is strengthened further in the in rural areas second half of the nineties. This momentum needs to be sustained and expanded to the depressed districts (including CHT) as well as urban poor settlements. A possible target could be to have at least one MFI in every village of Bangladesh so that the poor can get access to credit even as the formal credit institutions gradually withdraw from the rural areas. It would be possible to apply eventually, if not in the near future, new financial technologies (such as credit card) in the system of microfinance to better service the needs of the borrowers.

Urban microfinance operations warrant special attention, given the importance of credit access in urban informal sectors where most of the urban poor are employed. It is distressing to see how the diligent entrepreneurs of the Dholai Khal are starved of working capital loans and even at exhorbitantly high interest rates from the informal money market, while the banks accumulate large arrears on lending to the richer sections of society. Microenterprise loans for urban small-scale industries modeled on some variation of the principles of microcredit would help to reduce urban inequality and foster higher growth.

A related issue is whether some large MFIs with established reputation should be encouraged to act as conduits for mobilizing savings from the non-members of the village, and channel that savings into the financing of pro-poor projects both in rural and urban areas. Such an act will help financial integration and deepening, especially in view of the inadequate presence of the formal banking system in rural areas. Such a step would require modification of the legal framework that regulates MFI operations in Bangladesh.

4. Increased Rural Non-Farm Activities and Forging Urban-Rural Links

While technological progress in agriculture would accelerate output growth, the same cannot be said about employment dynamics. Experience of green revolution in India and elsewhere shows that technological progress in agriculture is labour-absorbing in the first phase and labour-displacing in the subsequent phase, with a rise in the opportunity cost of labour, and hence capital intensity. Green revolution thus fits into the overall transition dynamics, involving shift of labour force from farm to non-farm sectors. In response to this, non-farm activities need to expand in a corresponding manner so that it can provide enough incentives for the movement of labour from farm to non-farm sector. This, in turn, requires a steady growth in productivity in the non-farm sector.

As pointed out earlier, rural non-farm activities have played an important role in generating new sources of employment for the poor, but the productivity growth in this sector has been rather modest. In many cases, these activities persist as the source of part-time employment. This is particularly true of the microcredit-financed activities, which generally correspond to the lower end of the productivity spectrum. The way-out from this situation requires some degree of upscaling, with improved technology and marketing support. In recent years, partly in recognition of low-productivity and market saturation problems, the loan portfolio of the MFIs has become diversified, with a discernible shift towards moderately capital-intensive projects. However, this may not be enough to sustain this process.

This process of upscaling of rural non-farm activities would be further stimulated if urban-rural links can be forged in design, production, and marketing stages. It is possible to conceive a decentralized industrial process built-around the secondary towns and the peri-urban economy with backward and forward links with the surrounding rural economy. Agro-industries as well as modern industrial ancillaries can be sustained in this way. This is in sharp contrast to the earlier (largely failed) strategy of rural industrialization where the focus was on the village, rather than on secondary towns and peri-urban economy.

The process of forging increased urban-rural links would be beneficial to both rural and urban economies. For instance, relocation of modern activities to the secondary towns will help relieve urban congestion, on the one hand, and help create jobs for the commuters from the rural areas, on the other. The same applies to export-oriented industries such as readymade garments (which are concentrated in Dhaka and Chittagong) as well as EPZs. One of the important social results of this process would be the much-needed slow-down in the flow of migrants to the metropolitan cities. The process of migration has been quite rapid in the nineties, which is putting enormous pressure on the already fragile urban infrastructure, with resulting deterioration in the quality of urban environment. Such a strategy of decentralized industrialization involving secondary towns and peri-urban economy through forging increased urban-rural links would, however, require matching investments in the entire range of infrastructure such as telecommunication, internet, road, electricity, water, and sanitation.

5. Road, Power and Telecommunication: The Intervention Package

The impact of roads on poverty is well documented in the Bangladesh literature. Roads can influence poverty by creating new employment opportunities (especially in the non-farm sector) as well as by changing input and output prices. Improved road access in rural areas can encourage the cultivation of perishable products, thereby exerting favourable influence on crop diversification.

With the spread of literacy, improvement in the road network, and higher household savings via microcredit operations, the availability of electricity is going to be a binding constraint in promoting private investment in rural areas. Electricity access can impact through a number of channels. In the rural Bangladesh context, a large part of its impact is realised through its cost-reducing effects on the use of irrigation machines. Average cost for electricity operated irrigation pumps is substantially lower than in case of diesel operated machines; the difference is substantial with respect to LLP and DTW. Electricity can directly impact on the modernization of rural industry, contribute to longer working hours for commercial enterprises, and often has favourable influence on social development.Electricity is already having a strong growth impact on the poultry sector, especially in the nineties, and represents a successful example of increased urban-rural linkages.

In short, investments in roads and electricity would be extremely beneficial for the poverty reduction process. In areas where road communication is available, more attention needs to be given to electrification. In general, power sector investments must get priority over other investments in physical infrastructure in the next few years. This is partly because of the quite considerable latent demand for electricity that already exists in rural and urban areas because of availability of better roads, higher human capital, and greater household saving. Higher current allocations to the power sector are also necessary to compensate for the relative neglect of past investments, especially during the first half of the nineties.

Telecommunication is the third critical element in the infrastructural package for pro-poor growth. The relative importance of telecommunication has increased manifold following the revolution in information technology. While more documentation regarding the poverty impact of the new technology needs to be done in this area—Grameen mobile phone operated by the poor women in rural areas is a case in point—its impact is already being felt everywhere. Telecommunication would help market integration regionally, increase the effectiveness of the early-warning system for preventing disaster, and help improve the system of administrative governance. It is clear that much of the future growth in rural and urban areas would come through IT where access to telecommunication is critical.

6. Need for a Technology Policy

All these call for a comprehensive technology policy (especially IT and bio-technology) for purposes of eradicating poverty in the shortest possible time. New technology is going to be critical in developing new seeds for agriculture and in the adoption of these seeds for the adverse agro-ecological environments. It will help linking hitherto disintegrated national and local markets with global markets, and possibly lead to the discovery of new medicines for eradicating malaria, tuberculosis, and HIV Aids, to name only a few. Some of these items are in the nature of global public goods, have strong positive externalities, and merit international assistance.

The present institutional infrastructure in Bangladesh for the delivery and adoption of new technology, however, appears quite inadequate. Take the case of the current telecommunication monopoly, which has not been able to ensure wider coverage and quality service even in urban areas, not to mention rural areas. This sector must allow more competition and foreign investments in order to reap the full advantage of the potentials that such technology has in fostering higher growth and faster poverty reduction.

Enhancing technological capability in the public sector and removing the barriers in doing so in the private sector remains the strategic challenge for Bangladesh economy, be it in the area of growth, poverty reduction, or good governance. Development of quality human capital is one necessary aspect of the matter. Innovation and adoption of technology requires skills that come through higher, vocational, and technical education which is a much greater requirement than the level of basic education that enters into the human development index. However, education alone cannot resolve the issue of technological progress. As a late-starter we have to rely mainly on “borrowed technology”. There is much to be derived in this regard from “learning by doing” in an inter-dependent world, and for that Bangladesh needs to open up and locate itself in the globalized supply chain of production.

Regional cooperation within South Asia in science, technology, and information may also be important to resolve issues of common concern, especially in agriculture, environment, health, and education.

7. Regional and Sub-regional Solutions

The pace of poverty reduction cannot be maximized by seeking solutions only within the given national boundary. Greater emphasis needs to be given to regional and sub-regional issues. The role of transboundary factors has increased over the nineties across the globe, and Bangladesh is not an exception to this. Considerable scope exists today for sub-regional cooperation in the areas of water, energy, transit, trade, tourism, and environment, to name a few obvious categories. The principle of inter-dependence–projects executed in one country may have implications for another country, as in the classic case of Farakka barrage–is yet to be built into planning process of the member countries of the region. A climate of trust needs to be built around the resolution of common problems, such as degradation of environment, food security, and nuclear menace. The indirect effect of peace in the region would release additional resources for poverty alleviation.

E. Policies for Increasing Human Development of the Poor

The policies outlined in the previous section influences the income-earning environment of the poor. In this section we shall discuss policies that have a direct bearing on the basic capability (and hence, functionings) of the poor.

1. Caring for Human Development

The importance of human capital for increasing long-term economic growth is well known. In that capacity, development of human capital has strong income-poverty reducing effects. However, it is useful to emphasize here on the concept of “human development” as it allows a broader approach to poverty reduction.

As pointed out earlier in Chapter II (section 3), Bangladesh has achieved considerable progress in human development over the last two decades and the rate of human poverty reduction was faster than in case of income-poverty. This progress needs to be sustained in future.

Four aspects need to be highlighted in this regard. First, human development has three distinct components: education, health, and nutrition. While there has been a considerable quantitative expansion of education, the dimensions of health and nutrition have remained relatively neglected. Thus, there has been a considerable increase in allocations for education as proportion of GNP, but the matched figure for health remained stagnant. Within the health sector, some success has been achieved in child vaccination program. The same cannot be said of the curative health, however. Only a small proportion of poor population has access to public health care services till date. The performance of outpatient facilities in Metropolitan areas seems to have worsened in recent years. The degree of confidence on the quality of curative care is fast declining, with rising number of the rich and the middle class patients taking recourse to treatment abroad, which was not the case even 15 years ago. New health policy—Health Population Sector Programme (HPSP)–adopted by the government is expected to improve the situation, though much would depend on the extent to which it is implemented. In short, addressing the pro-poor concerns in health remains an unfinished task and the sector must get the priority it deserves.

Second, as discussed earlier, the level of malnutrition in Bangladesh remains one of the highest in South Asia and in the developing world, despite some improvement in the last decade. The percentage of underweight children below five-years was assessed at about 60 per cent in 1996. About 53 percent of the same age group children are stunted in rural areas. The situation does improve somewhat for the urban children, but even here the corresponding level remains unacceptably high (the rate of underweight and stunting being 46 and 43 per cent, respectively). Such a high degree of undernourishment has very adverse implications for future poverty reduction. This also shows that one cannot rely only on the growth process to redress the particularly adverse situation on the nutritional front. There is a need for public intervention. Steps have been taken recently in that direction through the introduction of National Nutrition Programme (NNP) to address the malnutrition of children under two years as well as pregnant and lactating mothers through the provision of food supplements, nutrition and health counseling. These programs need to be strengthened institutionally, with special focus on the poor areas and communities. Existing food-targeting programs such as Vulnerable Group Development (VGD) and Rural Maintenance Programme (RMP) may be used to reach out to the poorest and the most vulnerable with the nutrition-support package.

Third, even in basic educational indicator such as enrollment, the country is yet to achieve full-enrollment at the primary level. The dropout rate (including non-enrollment) is estimated to be
35 per cent at the primary level, with poverty being the most proximate cause for it. Incentives that are currently provided for enrollment of children from the poor households as well as for girl’s education need to be maintained, again with a focus on the regional dimension.

Fourth, the issue of quality cuts across all three components of human development. Some recent studies, which carried out aptitude tests, indicate that the level of knowledge, as implied by the 5th grade, seems to be achieved only by the 8th graders (Alam 1997). This implies that mere graduation from the primary level would not give the desired results. Given the quality of education, emphasis should be given to encourage study at least up to the 8th grade in order to have some impact of education on the skill level. At the secondary level of education more emphasis needs to be given to vocational and technical education, including dissemination on improved agricultural practices.

2. Social Sector Investments for the Urban Poor

As mentioned earlier, the level of human development in case of the urban poor (especially those living in the slums) is even worse than that observed for the rural poor. This is because urban poor living in slums lack legal residence status, which, in turn, deprives them of the basic urban amenities such as electricity, safe drinking water, and sanitation. More importantly, incentives for educating children of the type existing for the rural areas are missing in case of the urban poor. For them there is no “food for education”, no “free education up to the 10th grade for girls”, no “free distribution of books”. Note that although the access to public health is limited in rural areas, there is still some physical infrastructure for rural health care, with a referral system (though inadequately implemented) that links the union health center and thana (sub-district) level health centres to the district hospitals. No such referral system exists for the urban poor. In Dhaka metropolitan area there are 75 wards, but there are only 5 ward-level dispensaries.

Because of the extremely limited access to public education and health, the prospects for future social mobility remain bleak for the majority of the urban poor. Public health and educational policies thus indirectly aggravate the dualism between urban formal and informal sectors. These policy distortions need to be removed with more explicit targeting of the health and educational policies to the urban poor. These need to be supplemented by policies that in general provide wider access to urban amenities, thereby leading to improvement of the living as well as working conditions of the urban poor. As discussed earlier, urban policies require some re-thinking, with enhanced emphasis on the development of secondary towns and peri-urban economies (pourashavas). The quality of basic infrastructure is, however, even worse in the latter contexts given the poor governance capacity of these smaller urban bodies. This matter merits priority attention.

F. Policies for Strengthening Social Safety Net to Reduce Vulnerability

Poverty reduction is not just about income generation. It is about preventing the risks of income erosion as well. One can broadly identify three sets of policies to improve the security dimension of the anti-poverty measures, helping the poor to cope better with the various income (consumption) shocks. These policies directly or indirectly help the crisis-coping capability of the poor, and are more widely known as the risks-insurance policies. First, providing credit access to the poor in times of emergency needs would help to ease the burden of the shocks, reduce distress sales and some such “negative” methods of coping. Second, ensuring good public health (with possible public-private-NGO partnership) will reduce the health hazard related income and consumption shocks for the poor family. Third, strengthening of the disaster preventing and mitigating mechanisms at national, district, sub-district, and union levels would help to enhance the coping capability of the poor in times of natural disasters. This is important from the perspective of averting large-scale entitlement failure, which may result in as a consequence of severe natural disaster, especially in the most vulnerable regional pockets of distress.

We have already discussed the first two aspects in the previous two sections. The main focus is, therefore, on the third aspect of the security-enhancing policy. However, before we proceed to discuss the third aspect some comments on the importance of credit access and health interventions from the viewpoint of crisis coping strategy would be in order.

1. Credit as Insurance against Risks

On the credit access, the new point to be noted here is that the conventional formal market as well as non-conventional microcredit do not directly deal with the problem of access to emergency loans. The issue of meeting short-term credit demand of the poor is also important from the angle of the efficiency consideration. Thus, both the programs have not been particularly successful in devising a reliable way of meeting seasonal working capital demand—a drawback that partly explains why an efficient credit program for the small farmers is yet to emerge. In recent years some of the MFIs such as the Grameen Bank and BRAC have introduced insurance schemes for their borrowers. This is a step in the right direction. However, the same mechanism needs to be extended to the non-borrowers (including the poor and the non-poor) as well.

2. Health Care as Insurance against risks

On the health access, the additional point to be noted here is that a good system of public health care is the best social insurance against risks and vulnerability. This is especially true for the poor who usually have to part away about 10-15 per cent of their incomes for meeting private health expense (Sen 1998). Health hazard may relate to adult breadwinner’s ill health caused by illness; it also pertains to child and maternal health. Health risks may be caused by diverse sources. It can be caused by the lack of adequate public health measures against infectious diseases such as malaria, by the deteriorating environment such as air-and sound-pollution, which have serious consequences for child health. It may also be created by the indiscriminate adoption of ground water technology leading to public health concerns such as arsenic contamination, by occupational hazards as in the case of relatively high degree of morbidity among the readymade garment workers, mostly female, working under the most unfavourable work conditions. Measures of prevention of health risks, therefore, have to discriminate between the rural and urban poor, and would vary according to age-sex characteristics. Whatever the source of risks, it is clear that a variety of preventive, promotive, and curative health care measures needs to be undertaken to ease the burden of ill-health on the poor. The new health policy, as reflected in HPSP, needs to be made sensitive to the health need of the poor and the most vulnerable, especially poor women and children.

3. Coping with Disaster

Experience of the 1998 flood points to several policy implications. The disaster once again showed the essential vulnerability of the poverty reduction process. But, it also showed the resilience capacity of the people that were affected by the flood. The aggregate production shortfall during the flood-year of 1998 was less than the absolute amount of shortfall during the flood-year of 1988 (which was also devastating in nature). What is, perhaps, more important is the difference that the combined effort of GoB, NGOs and the people can make in fast overcoming what was termed as the worst flood of this century. The increased crisis-coping capacity of GoB was also better appreciated during the 1998 flood and its aftermath. Policies should continue to support schemes which would take into account the “flood factor” in the designing stage of the project over and above its involvement in projects explicitly aimed at flood control and drainage. More attention also needs to be paid to dredging and strengthening of the capillaries in order to reduce the intensity of flooding in the major rivers. There is also a greater need for policy support in ensuring better disaster preparedness, and, in general, developing a more people-centered approach in managing disaster and in accelerating the rehabilitation program.

G. Measures for Enhancing Participatory Governance

Policies and institutional measures for broadening participatory governance and enhancing the “voice” of the poor may be classified into two distinct action arena, requiring policy as well as institutional interventions. The first group of measures focuses on empowering the women. This is crucial because of its intrinsic value as a welfare goal. But it is also an important policy as instrument for bringing about favourable social and economic change. The second group of measures pertains to strengthening of the system of good governance, especially decentralization at the local level (the issue of effective local governance) where the impact of poverty reduction is immediate. The third group of measures recognize the need for building and/or coalescing grassroots level initiatives—outside the domain of local government—to create a demand-driven, receiving mechanism “from below” and also to act as a pressure mechanism on the quality of local governance. Such popular initiatives would improve the quality and functioning of local democracy. In this section we shall mainly focus on the first two measures, while recognizing the general importance of the third issue.

1. Closing the Gender Gaps

Women’s empowerment has important social and economic effects. Although opinions differ as to the precise measure of women’s empowerment, there is some operational consensus on the issue (Murthi et al 1995; Sen 1999). Higher literacy, greater property rights, higher work force participation, access to credit, greater scope for political participation at various levels of representation are important measures of raising the level of women’s empowerment. These measures will not only help to increase the well being of women, but also strengthen their role as active agent of social and economic transformation (the agency aspect).

Bangladesh has made considerable progress over the last two decades in respect of some of the above measures, contributing to greater women’s empowerment. The strengthening of these measures would further contribute to the decline in fertility rate (where the progress has been significant in the past), improvement in child and maternal nutrition (where the incidence of malnutrition is still very high) and greater welfare for the women themselves. The latter, like income-poverty reduction, is a goal to be pursued in its own right, not to be mixed up with its instrumental (or consequential) importance for bringing about other desirable social and economic results.

While the gender-gap is closing in Bangladesh for most of the social indicators such as mortality, life expectancy, and enrollment, the overall level of empowerment measured in terms of literacy, work force participation, property rights, and credit access leaves much to be desired. In recent years, the incidence of violence against women seems to be on the increase. This needs to be addressed as a matter of priority if the trends of past improvement on the empowerment front are to be sustained in future. A related institutional issue is to increase the political voice of women, especially poor women, which will further enhance their agency role and hence, contribute to the faster progress in the well being of children and women. An important step has been taken by the government to reserve for women about one-third of the seats at local government levels, whereby the women candidates would be elected through direct voting. The same principle needs to be taken up for representations at higher tier of government, including the national parliament.

2. Strengthening Local Government

Public action can have maximum impact on income-poverty as well as human poverty with the support of strong and effective local government. While Bangladesh made some considerable progress in this area, much remains to be accomplished (Sobhan 1998; Khan 2000). This is partly because of the lack of policy continuity on the issue of local governance, often generating conflicting signals associated with the change in regimes. An important step has been undertaken to empower the local governance process by holding the Union Parishad (UP) election in 1999 (which corresponds to the lowest tier of the local governance structure). Another election is scheduled to take place this year to elect local representatives at the Thana (sub-district) level. But, the issue at stake goes beyond the aspect of local representation. It involves significant devolution of financial and administrative power from the central authority to local bodies. Unfortunately very little progress was made in that direction. With fiscal federalism as the ultimate goal, the matter may make a modest beginning by undertaking steps towards expenditure decentralization, followed subsequently by the decentralization of the taxing power. Decentralizing without capacity building at the local level would not produce desired results, however. Appropriate measures may be undertaken to build up that capacity following the “best practice” examples in South Asia. The slow progress in the area of local governance is, perhaps, due to the steep opposition coming from vested interest groups, both in the polity and in bureaucracy.

H. Policies and Institutions for Reducing Inequality

The concerns expressed in the earlier discussions regarding the sharp rise in inequality in the nineties need to be addressed. When growth is accompanied by an increase in inequality opportunities are missed for poverty alleviation. When such a rise in inequality takes place in the absence of rapid growth this becomes a part of wider policy concern.

The way-out from the problem of rising inequality does not always lie in the traditional asset distribution program such as through land re-distributive program. In any case Bangladesh has limited physical scope for land re-distribution, because a ceiling of even a 5 acre would not provide enough land to the tillers. However, this does not imply that other measures for reducing inequality cannot be undertaken. The poor can get access to land through the tenancy market and benefit considerably from it through the enforcement of better terms and conditions for tenancy. Recent studies using cross-state data for India have pointed to significant productivity gains with tenancy reform leading to greater security for the tenants (see, for instance, Burgess 1999). Bangladesh already has a legal framework for such reforms in the form of 1984 Land Reform Ordinance. However, many of the measures stipulated by the ordinance are yet to be implemented.

The other way of approaching the problem of reducing inequality is to think in terms of a broad asset access framework. There are several such assets potentially available for distribution (or, re-distribution) to the poor. These are physical assets, human assets, financial assets, natural assets, social assets, and political assets. If the access to physical capital is technically constrained by the limited availability of physical assets such as land, then the access to other assets on the part of the poor may be a good alternative. Providing access to human assets such as basic education and higher level of skills will help the poor to move up the income ladder faster than possibly through any other means. Ensuring expanded access to financial assets via microcredit would help the poor to grow fast. The cooperatives of the poor can also earn income by investing in the bond market. The prospects of the latter type of interventions would increase with the spread of IT in rural areas. Access to natural assets such as common property resources would help the poor in mitigating risk in times of distress. Access to social assets (so-called social capital) can be ensured by building grassroots organizations of the poor at the village level, leading to better networking capacity, which would again have the role of risks-insurance. And, finally, the access to political assets—greater political empowerment—would help the poor to ensure their legitimate pie in the allocations of public resources and in the benefits of the development process.

Table 1.1

Trends in Rural and Urban Poverty, 1973/74 – 1981/82

Head-Count Index, P(O)71.365.363.248.4
FGT Poverty Gap, P(1)25.620.221.114.9
Distributionally Sensitive Measure, P(2)

Source: Hossain and Sen (1992).

Table 1.2

Recent Trends in Rural and Urban Poverty, 1983/84-1995/96


Index (%)Poverty-Gap

Index (%)Squared Poverty-Gap

Index (%)Urban:   1983/8440.911.44.41988/8935.98.72.81991/9233.68.42.81995/9626.36.01.9    Rural:   1983/8453.815.05.91988/8949.713.14.81991/9252.914.65.61995/9651.114.15.5

Source: Ravallion and Sen (1996); Sen (1998).

Table 1.3

Summary Statistics on Growth and Inequality, 1983/84 – 1995/96


MeanPoverty Line (Tk/Person/Month)Consumption (Tk/Person/Month)Mean/Poverty Line (%)Gini Index (%)Urban:    1983/84301.72396.5313129.81988/89453.65695.1915332.61991/92534.99817.1215331.91995/96642.581233.0619236.7     Rural:    1983/84268.92284.8410624.61988/89379.08435.3911526.51991/92469.13509.6710925.51995/96567.00658.4511628.8

Source: Ravallion and Sen (1996); Sen (1998)

Table 1.4

Trend in Human Development Index (HDI), 1960 – 97


Source and Note: The 1996/97 figure is own estimate. All other figures are from the Human Development Reports.

Table 1.5

Trend in Human Poverty Index, 1981 – 97

  1. Deprivation in Longevity (P1)
  1.       II.            – Probability of dying before age 40







  • Deprivation in Knowledge (P2)
  • Adult illiteracy (weight: 2/3)

– Child aged 6-14 years not attending school (weight: 1/3)















  1. Deprivation in Economic Provisioning (P3)
  1. Public Provisioning
  1. Share of population without access to health

services proxied by a composite indicator of:

– children not fully immunized

– % of deliveries not in the institutions


  1. Percentage of population with out access to

safe water

  1. Percentage of population not living in

electrified houses

  1. Private Provisioning
  • Percentage of children under 5 years of age

who were malnourished









































(1996/97)VI. Human Poverty Index*61.347.240.1

Notes: * HPI Index is calculated as follows:
HPI = [ 1/3 ( P13 + P23 + P33 )]1/3
Source: Sen and Ali (1999).

Table 1.6

Spatial Variation of Social Statistics





RateIMR (per 1000 live births)Life


at BirthImmunization

(12-23 months)Child Death Rate 1-4


Index of Poverty 199519951995199519951995/96Division:      Barisal56.476.657.264.610.859.9Chittagong41.281.957.072.28.944.9Dhaka43.078.358.352.710.852.0Khulna47.272.458.481.39.551.7Rajshahi35.279.956.554.58.662.2Sector:      Rural36.683.357.161.310.256.7Urban60.060.860.676.37.735.0National42.677.757.965.49.753.1

Source: Estimated from the 1995 Health and Demographic Survey.

Table 1.7

Inter– Country Statistics on Growth, Human Development, and Income–Poverty

BangladeshIndiaPakistanNepalSri-lankaSouth AsiaLDCDeveloping Countries
Per capita GNP Growth, 1975-95
(% per year)
Total fertility rate
% change per year-2.1-1.5-1.1-1.1-1.8-1.5-0.9-1.6
Life expectancy at birth (year)
% change per year1.
Infant mortality rate

(per 1000 live births)        197014813011815665131149111199781719575177210464% change per year-1.6-1.6-0.7-1.9-2.7-1.6-1.1-1.5Under- Five mortality rate (per 1000 live births)        197023920618323410020724217019971091081361041910616294% change per year-2.0-1.7-0.9-2.0-3.0-1.8-1.2-1.6Head-count index of income-poverty        Early 80s52.346.5029.142.527.345.4na33.9Early 90s47.037.3626.345.022.443.1na31.9% change per year-0.84-1.87-1.370.53-3.59-0.84na-0.98

Source and Note: 1. Per-capita GNP growth data are from 1999 HDR.

2. Information on fertility, infant mortality, under-five mortality and human development index are from 1999 HDR; adult literacy rate is taken from World Development Indicators (CD-ROM version). For Bangladesh, adult literacy data for 1995 are taken from the Fifth Plan document. Bangladesh HDI value for the later year relates to 1996/97 and represent own estimate.

3. Aggregate poverty estimates for South Asia and developing countries are from Ravallion and Chen 1996).


Anand, Sudhir and Martin Ravallion, “Human Development in Poor Countries: on the Role of Private Incomes and Public Services”, Journal of Economic Perspectives, Vol. 7, No. 1, 1993, pp. 133-150.

Anand, Sudhir and Amartya Sen, ” Human Poverty: A Note”, Background Paper Prepared for the 1997 Human Development Report, 1996.

Asian Development Bank (ADB), Fighting Poverty in Asia and the Pacific: The Poverty Reduction Strategy, Manila, November 1999.

Dreze, Jean and Amartya Sen, India. Economic Development and Social Opportunity, Oxford University Press, 1995.

Hossain, Mahabub and Binayak Sen, “Rural Poverty in Bangladesh: Trends and Determinants”, Asian Development Review, Vol.10, No. 1, January 1992, pp. 1-34.

Khan, Azizur Rahman “Poverty as a Cause and a Consequence of Economic Growth“, Bangladesh Development Studies, 1990, pp. 221-253.
Khan, A.Z.M. Obaidullah, Bangladesh: Governance Issues, Bangladesh Centre for Advanced Studies, February 2000 (Draft).

Mundle, Sudipto and V.B. Tulasidhar, Adjustment and Distribution: The Indian Experience, ADB Occasional Paper Series No. 17, Manila, 1998.
Sen, Binayak,”Poverty and Policy” in Rehman Sobhan (Ed.), Growth or Stagnation? A Review of Bangladesh’s Development 1996, Centre for Policy Dialogue and University Press Ltd, Dhaka, February 1997, pp. 115-160.

Sen, Binayak, “Politics of Poverty Alleviation”, in Rehman Sobhan (Ed.), Crisis in Governance. A Review of Bangladesh’s Development 1997, Centre for Policy Dialogue and University Press Ltd, Dhaka, June 1998, pp. 159-182.

Sen, Binayak and Zulfiqar Ali, “Poverty and Human Development in Bangladesh”, Background Paper Prepared for the Bangladesh Human Development Report, BIDS/ Planning Commission, Dhaka, June 1999 (Draft).

Sobhan, Rehman, “Overview”, in Rehman Sobhan (Ed.), Crisis in Governance. A Review of Bangladesh’s Development 1997, Centre for Policy Dialogue and University Press Ltd, Dhaka, June 1998.

World Bank, Bangladesh. From Counting the Poor to Making the Poor Count, Poverty Reduction and Economic Management Network, South Asia Region, World Bank, Washington, D.C., April 29, 1998.

Poverty in Bangladesh