Bangladesh, a Middle Income Country by 2021: What Will It Take in terms of Poverty Reduction? 1 LEA GIMÉNEZ DEAN JOLLIFFE IFFATH SHARIF - PDF

Description
Bangladesh Development Studies Vol. XXXVII, March-June 2014, Nos. 1&2 Bangladesh, a Middle Income Country by 2021: What Will It Take in terms of Poverty Reduction? 1 LEA GIMÉNEZ DEAN JOLLIFFE IFFATH SHARIF

Please download to get full document.

View again

of 19
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Data & Analytics

Publish on:

Views: 23 | Pages: 19

Extension: PDF | Download: 0

Share
Transcript
Bangladesh Development Studies Vol. XXXVII, March-June 2014, Nos. 1&2 Bangladesh, a Middle Income Country by 2021: What Will It Take in terms of Poverty Reduction? 1 LEA GIMÉNEZ DEAN JOLLIFFE IFFATH SHARIF The Vision 2021 and the associated Perspective Plan , adopted by the Government of Bangladesh, lay out a series of development targets for Among the core targets identified to monitor the progress toward the Vision 2021 objectives is that of attaining a poverty headcount of 14 percent by The purpose of this paper is to answer the following question: Given Bangladesh s performance in poverty reduction over the last decades, can we expect the proportion of the country s population living in poverty to be 14 percent by 2021? Using data from the last three Household Income and Expenditure Survey, we examine changes in poverty rates during , estimate net elasticity of poverty reduction to growth in per-capita expenditure, and then project poverty headcounts into the future. Our poverty projections based on the last three HIES surveys suggest that Bangladesh will achieve its MDG goal of halving its poverty headcount to 28.5 percent by 2015 significantly ahead of schedule. Attaining the Vision 2021 poverty target of 14 percent by 2021, however, is less certain as it requires a GDP growth of at least 8 percent, or more than 2 percentage points higher than that observed in recent years. Keywords: Poverty, Poverty Decompositions, Bangladesh JEL Codes: O1, O4, I3 I. INTRODUCTION Over the 2000 to 2010 decade, Bangladesh experienced steady and strong GDP growth, averaging a growth rate of 6 percent per year. The Bangladesh Bureau of Statistics (BBS) reports that poverty rates have also demonstrated 1 Address: World Bank, 1818 H St. NW, Washington D.C ; (L. Giménez), (D. Jolliffe), (I. Sharif). The authors gratefully acknowledge generous support from the UK Department for International Development through the Joint Technical Assistance Programme (JOTAP). This paper was prepared as a background report for the Bangladesh Poverty Assessment report (World Bank 2013). The findings, interpretations, and conclusions of this paper are those of the authors and should not be attributed to the World Bank or its member countries. 2 Bangladesh Development Studies steady improvement during this period, with an average decline of 1.74 percentage points per year (BBS and World Bank 2012), a rate of decline that outperforms a majority of countries (Newman, Azevedo, Saavedra, and Molina 2008). In addition to the MDG goals on poverty reduction, the Government of Bangladesh has set its own goal on poverty reduction as part of its development strategy. For example, the Vision 2021 and the associated Perspective Plan lay out a series of development targets that must be achieved by When achieved, these targets would transform the socio-economic environment of Bangladesh from a low income economy to a middle income economy. Among the core targets that have been identified to monitor the progress toward the Vision 2021 objectives is that of attaining a poverty head-count rate of 14 percent by 2021, with an intermediate target of attaining a poverty head-count rate of 22 percent by Assuming population growth continues to decline at the same rate as during the period, achieving the Vision 2021 poverty target implies lifting approximately 15 million people out of poverty. Relatively less ambitious is the poverty Millennium Development Goal (MDG) for Bangladesh, which stipulates that the proportion of people living in extreme poverty that prevailed in 1990 (57 percent) must be reduced by at least one-half by the year Assuming population growth continues to decline at the same rate as during the period, achieving the poverty MDG implies lifting over 4.7 million people out of poverty. Our primary goal is to answer the following question: Given Bangladesh s performance in poverty reduction over the last decades, can we expect the proportion of the country s population living in poverty to be around 14 percent by 2021? As temporal comparisons are crucial to understanding how the poverty reduction process has evolved and qualitatively changed over time, we use data from the last three Bangladesh Household Income and Expenditure Surveys (HIES) to first analyze changes in poverty incidence taking place in the period. Next, we estimate Bangladesh's net elasticity of poverty reduction to growth in per-capita expenditure to project the poverty headcount index into the future. The last section summarizes our main findings and concludes the study. 2 See Part_1_ pdf. 3 That is, between 2010 and 2015, Bangladesh must reduce its poverty level by an average of 0.6 percent per annum, equivalent to a cumulative reduction of 3 percentage points over the course of this period. Giménez, Jolliffe & Sharif: Bangladesh, a Middle Income Country by II. POVERTY AND GROWTH IN RECENT YEARS Table I shows the Cost of Basic Needs (CBN) upper and lower poverty estimates for Bangladesh based on HIES data from 2000, 2005, and From 2000 to 2010, Bangladesh experienced a uniform and steady decline in poverty rates. Poverty rates demonstrated impressive and steady improvement during this period, with an average decline of 1.74 percentage points per year. During the period, the average decline in poverty rates was 1.78 percentage points per year; the analogous decline for the period was 1.7 percentage points. In 2000, 49 percent of the population was poor; by 2010, this number dropped to 31.5 percent. This reduction in the national-level poverty rate suggests that the series of shocks affecting Bangladesh in did not significantly slow down the speed of poverty reduction. 2.1 Trends in Poverty National, Rural, and Urban The national poverty headcount decreased by 17.4 percentage points over the period from 2000 to Across urban and rural areas, the rate of poverty reduction was comparable; in 2010, 35.2 (21.3) percent of the rural (urban) population was poor, compared to 52.3 (35.2) percent in 2000 (Table I and Figure 1). While the changes in poverty rates represent an outstanding 35.6 percent reduction over a ten-year span at the national-level (Table II), rural areas had only attained the decade-old poverty rate of urban areas in In general, the percentage change in poverty headcount rates for the period was larger in urban areas (39 percent) relative to rural areas (33 percent), and the gap in the speed of poverty reduction during the period between rural and urban areas (3 percentage points) widened over the period (5 percentage points). Extreme poverty continues to be a rural phenomenon. The national extreme poverty headcount decreased by 16.7 percentage points over the The 2000, 2005 and 2010 HIES questionnaires used for this analysis were designed as a poverty monitoring instrument by the BBS with support from the World Bank. In addition to being a multi-module survey that is nationally representative at the divisional level, the year round data collection modality followed helps to eliminate seasonal variations in income and expenditure in a year. The authors constructed all of the variables used in the analysis, including the income and consumption aggregates. The data from HIES was also triangulated with other similar instruments such as the Labor Force Surveys (2003, 2005, 2010), Demographic Health Surveys, InM Panel Household Survey 2011, and IFPRI s Bangladesh Integrated Household Survey, 2012 all of which show similar patterns in expenditure and other socio-economic indicators. 4 Bangladesh Development Studies period. In 2010, 21.1 (7.7) percent of the rural (urban) population was extremely poor, compared to 37.9 (19.9) percent in 2000 (Table I). That is, in 2010, 60 (36) percent of the poor in rural (urban) areas were also extremely poor. Furthermore, between 2005 and 2010, the rate of extreme poverty decline was 26 percent in rural areas and 47 percent in urban areas, compared to 25 percent in rural areas and 27 in urban areas between 2000 and TABLE I POVERTY HEADCOUNT RATES Poverty Extreme Poverty National Urban Rural Source: All estimates are CBN based on HIES 2005, updated for 2010, and back-casted for update: survey-based food prices and non-food allowance reestimated using upper poverty lines. Official Poverty Lines estimated for HIES (2000, 2005, and 2010). Figure 1: Poverty Trends Source: HIES 2000, 2005, and 2010. Giménez, Jolliffe & Sharif: Bangladesh, a Middle Income Country by TABLE II PERCENTAGE CHANGE IN POVERTY HEADCOUNT RATES Poverty Extreme Poverty National -18% -21% -36% -27% -30% -49% Urban -19% -25% -39% -27% -47% -61% Rural -16% -20% -33% -25% -26% -44% 2.2 Depth and Severity of Poverty The poverty headcount index measures the proportion of the population that is poor. This measure, however, does not indicate how poor the poor are. To accomplish this, we use two different indices. First, the poverty depth index (also known as the poverty gap index), which measures the extent to which individuals fall below the poverty line (poverty gaps) as a proportion of the poverty line. The sum of these poverty gaps over a population gives the minimum cost of eliminating poverty in that population, if transfers were perfectly targeted. Unlike the poverty depth index, the second index we use, the severity of poverty index (also known as the poverty gap square index), reflects changes in inequality among the poor. For example, a transfer from a poorer household to a poor household would increase the index. This index averages the squares of the poverty gaps relative to the poverty line and is one of the Foster-Greer- Thorbecke (FGT) class of poverty measures that allows varying weights to be placed on the income (or expenditure) level of the poorest members in society (Haughton and Khandker 2009). The ratio of the depth of poverty to headcount (6.5/31.5) in 2010 indicates that, on average, the poor fell nearly 21 percent short of the poverty threshold (i.e. the poor consume at a level equal to only 79 percent of the cost of basic needs). The same ratio was 26 percent in 2000 and 23 percent in At the national level, the depth of poverty was nearly halved over the period (Table III). This rapid decline in the depth of poverty allowed Bangladesh to attain its Millennium Development Goal (MDG) target about five years ahead of schedule (the depth of poverty had been 16 percent during the 1990s, and the goal was to reduce this to 8 percent by 2015). The decline in poverty depth was larger in urban areas (52 percent) relative to rural areas (46 percent). The difference in poverty depth reduction between urban and rural areas widened over the decade. Like the poverty headcount rate, the difference in the speed of poverty depth reduction between rural and urban areas that existed in the period (less than 0.5 percent) widened over the period (10 percent). A similar pattern is observed for the severity measure. 6 Bangladesh Development Studies Significant improvements occurred with respect to the incidence of poverty, the severity of poverty, as well as the depth and inequality of poverty among the poor over the last decade. Overall, a clear narrative emerges: over the last decade, poverty has continued to decline in both rural and urban areas in Bangladesh. In general, fewer people are below the poverty line, and variation in the severity of poverty among the poor has significantly narrowed, primarily due to decreasing numbers of individuals who are extremely poor. Nevertheless, poverty in rural areas continues to be relatively more pervasive and extreme, and the gap in the speed of poverty reduction between urban and rural areas has, in fact, widened over that last five years. TABLE III DEPTH AND SEVERITY OF POVERTY Poverty Depth Severity National Urban Rural Consumption Growth and Distributional Changes We now turn to analyzing changes in real per-capita consumption, the welfare measure that underlies the poverty indices. In terms of levels, Table IV shows that average real per-capita consumption increased by 20 percent over the last decade, 60 percent of which took place over the first part of the decade. While real per-capita consumption for the year 2010 remained about 26 percent lower in rural areas relative to urban areas, the average annual growth in real percapita consumption was twice as large in rural areas (2.1 percent) relative to urban areas (0.9 percent) throughout the decade. TABLE IV MEAN REAL PER-CAPITA MONTHLY CONSUMPTION Per-capita Consumption Cumulative Change (%) Average Annual Growth (%) National Urban Rural Note: The base is the national poverty line for 2005. Giménez, Jolliffe & Sharif: Bangladesh, a Middle Income Country by In Figure 2.A, we observe that the distribution of per-capita real expenditure has shifted down and to the right for both the and periods. These shifts suggest that real per-capita expenditure has increased for the entire population. According to the cumulative distribution of per-capita real expenditures displayed in Figure 2.B, for the relevant range of the poverty line the poverty rate in 2005 is below that of 2000, regardless of how high the poverty line is set. The same is true for the year 2010 relative to both 2005 and In other words, irrespective of the poverty line level, the official poverty estimates indicate that poverty has declined in 2005 relative to 2000 and in 2010 relative to Nevertheless, it is important to note that, while these reductions in poverty indicate a positive trend, individuals who are no longer classified as poor may nevertheless be vulnerable to poverty. For instance, the percentage of non-poor people consuming less than 1.5 times the national poverty line was 28 percent (or about 36 million people) in By 2010, about 35 percent of the population, or 52 million non-poor people, consumed more than the poverty line and less than 1.5 times the national poverty line. Figure 2: Distribution of Per Capita Real Expenditure by Survey Year A. Density B. Cumulative Distribution Note: The vertical lines represent the mean real per-capita expenditure for each survey year (µ). The base is the national poverty line for Figure 3 depicts qualitative differences in the distribution of per-capita real expenditure between the first and the second part of the decade. In particular, during the period, the increase in per-capita consumption benefited both the rich and the poor, particularly those in the upper (the extremely rich) and lower (the extremely poor) tails of the consumption distribution relative to 5 However, we note that first order stochastic dominance holds only for the year 2000 relative to 2005, and it fails to hold at high levels of real per-capita expenditure for the year 2005 relative to 2010. 8 Bangladesh Development Studies the 40th to 80th percentiles. The pro-poor growth rate of per-capita consumption over this period (2.27 percent) was virtually equal to the mean growth rate of per-capita consumption (2.28 percent). 6 During the period, growth was relatively more pro-poor. In particular, the increase in per-capita consumption was higher than average for those in the 10th to 80th percentiles relative to those in the upper and lower tails of the consumption distribution. Those below the 70th percentile of the per-capita consumption distribution experienced the largest increases in per-capita consumption. The pro-poor growth rate of per-capita consumption over the second half of the decade (1.76 percent) was higher than the mean growth rate of per-capita consumption (1.41 percent). The same was true for the pro-poor growth rate over the decade (2.01 percent) relative to the mean growth rate (1.84 percent). Figure 3: Growth Incidence Curve Growth Rate Mean Median Percentile Pro-poor Source: Authors own calculation using HIES 2000, 2005, and Note: The base is the national poverty line for Next, we use the Datt and Ravallion (1992) decomposition to separate the change in poverty headcount into its growth and redistribution components. In particular, Datt and Ravallion (1992) observe that poverty measures (Pt) may be fully characterized by the poverty line (z), the mean of the distribution of economic welfare (μ), and relative inequality, as represented by the Lorenz curve (L), such that: 6 Here, pro-poor is defined as growth that reduces poverty. A more precise definition is provided by Ravallion and Chen (2003): Pro-poor growth is the ordinary growth rate in the mean scaled up or down the ratio of the actual change in the Watts index to the change implied by distribution-neutral growth. Giménez, Jolliffe & Sharif: Bangladesh, a Middle Income Country by Then, the overall change in poverty from base period 0 to end period 1 can be written as follows: where is known as the growth component, is the redistribution component, and is the residual or unexplained component. Following this methodology, we first create a counterfactual distribution of real per capita expenditure. This counterfactual shares the same distributional properties as the actual distribution, yet it assumes that the growth of real percapita expenditures was the same among all households between 2000 and Under these assumptions, the difference in poverty rate between the two distributions of real per capita expenditure, the actual and the counterfactual, is credited exclusively to economic growth between 2000 and Similarly, since the counterfactual distribution and the 2010 distribution share the same mean expenditure, the difference in poverty rates implied by these distributions is credited exclusively to a change in inequality between 2000 and The residual component is eliminated by undertaking the decomposition twice, forward and backwards, and taking the average of the two. The results, presented in Figure 4 and Table V, show that in the period, the reduction in the poverty headcount ratio was fully explained by the growth component. Furthermore, the redistribution component had a negative effect on poverty headcount. However, during the second half of the decade, the redistribution component complemented the growth component. This decomposition suggests stark differences in the underlying components of poverty decline between the first and the second halves of the decade. Over the period, both the growth and redistribution components moved in the same direction, with the former being the predominant driving force for poverty reduction. Figure 4: Growth and Redistribution Components of Changes in Poverty Note: The results are obtained by taking the average of the two decompositions with 2000 and 2005 as base years. 10 Bangladesh Development Studies Period Area Total poverty reduction of the period TABLE V DATT AND RAVALLION (1992) GROWTH DECOMPOSITION METHOD Grosse up base poverty (holding distribution constant) Actual poverty rate in base year Forward Difference Grosse down end of period poverty (holding growth constant) Actual poverty rate in base year Difference Residual Nation Rural Urban Nation Rural Urban Nation Rural Urban Backward Period Area Total poverty reduction of the period Actual poverty rate in base year Grosse down end of period poverty (holding growth constant) Difference Actual poverty rate in base year Grosse up base poverty (holding distribution constant) Difference Residu al Nation Rural Urban Nation Rural Urban Nation Rural Urban Giménez, Jolliffe & Sharif: Bangladesh, a Middle Income Country by TABLE V DATT AND RAVALLION (1992) GROWTH DECOMPOSITION METHOD (CONT.) Average of forward and backward decompositions Period Area Growth Redistribution Total Residual Nation Rural Urban Nation Rural Urban Nation Rural Urban Source: Authors own calculations using HIES 200, 2005, and 2010. 12 Bangladesh Development Studies III. PROJECTING RECENT TRENDS IN GROWTH, INEQUALITY, AND POVERTY INTO THE FUTURE In this section, we use data from the last three HIES surveys to estimate Bangladesh's net elasticity of poverty redu
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks