Transcript Slide 1

Poverty in Bangladesh:
Creating Opportunities & Bridging the EastWest divide
Ambar Narayan, Hassan Zaman
(based on World Bank Poverty Assessment for Bangladesh 2008)
April 20, 2009
1
Poverty trends in recent years
Poverty headcount rates (%)
Upper PL
Lower PL
2000
2005
2000
2005
National
48.9
40.0
34.3
25.1
Urban
35.2
28.4
19.9
14.6
Rural
52.3
43.8
37.9
28.6
Notes: Using official Poverty Lines estimated for HIES (2005, 2000)
“Consumption” poverty: inability to purchase what is necessary to satisfy “basic
needs”, including minimum calorie needs and non-food items
 Poverty rate reduced from 49% to 40% during 2000-2005
• Similar reduction for extreme poverty rate (34% to 25%)
• Significant reduction in both urban and rural poverty
 No. of people in poverty fell by 6 million, in extreme poverty by 8 million
 But nearly 56 mn. still in poverty, including 35 mn. in extreme poverty
2
Long-term poverty trends (91-92 to 2005)
Low er Poverty Line
60
Headcount rate (%)
Headcount rate (%)
Upper Poverty Line
40
20
0
60
40
20
0
1991/92
1995/96
Rural
2000
Urban
2005
1991/92
National
1995/96
Rural
2000
Urban
2005
National
• Poverty rate fell from 57 to 40%, extreme poverty rate from 41 to 25%
– Highest reduction in poverty during 2000-2005
• Consumption growth has benefited the poor
– Stable relative inequality (Gini index for consumption at 0.31) since 1995-96
– During 2000-2005, consumption growth among the bottom 30% higher
than the average
3
Rich-poor gaps
Ratio of percentiles of real per capita exp: 2000 and
2005
Absolute differences between percentiles of real
per capita exp: 2000 and 2005
1,600
Taka (constant 2005 rural
Dhaka prices)
Taka (constant 2005 rural
Dhaka prices)
4
3
2
1
0
1,200
800
400
0
p90/p10
p90/p50
p50/p10
p75/p25
p75/p50
Ratios of pctiles of real per capita exp
2000
2005
p50/p25
p90-p10 p90-p50 p50-p10 p75-p25 p75-p50 p50-p25
Gaps betw een pctiles of real per capita exp
2000
2005
• Ratios of percentiles of per capita exp have remained mostly
unchanged between 2000 and 2005 (e.g. p90:p10, p50:p10) – little
change in relative inequality
• But gaps between percentiles have increased (e.g. p90-p10, p50-p10)
– increase in absolute inequality
4
Bangladesh compares well with South Asian
countries in poverty reduction…..
5.0
% values
4.0
3.0
2.0
1.0
Annual grow th of per capita gdp
Sri Lanka
(1991-02)
Pakistan
(1999-05)
Nepal
(1996-04)
India
(1994-00)
Bangladesh
(2000-05)
0.0
Annual rate of poverty reduction
Poverty reduction in Bangladesh among the highest in the region since
1990s
•
• Growth in Bangladesh more pro-poor than in other SA countries except
Nepal
5
…. But less so with East Asia
Average annual rate (% ) of poverty reduction
(in comparison to East Asian Countries)
Banglades hChina (90- Ko rea Rep Malays ia Thailand Vietnam
(00-05)
01)
(90-01)
(90-99) (90-02) (93-02)
0
(%)
-2
-4
-6
-8
-10
• East Asian countries have reduced poverty much faster – higher growth
rates and comparable responsiveness of poverty to growth
•If Bangladesh attains similar growth rates, it would match the pace of
poverty reduction in East Asia
― In 1990 both Bangladesh and Vietnam had poverty rate of 58%; in 2005
Vietnam’s poverty rate was 20%, half that of Bangladesh’s
6
Poverty projections and MDGs
• IF the 2000-05 GDP growth is maintained (annual average 5.3%), MDG
target (halving poverty rate from 57% in 91-92) will be met
• These projections assume (based on recent history)
– Stable inequality: if inequality were to rise, less poverty reduction will occur
for same GDP growth
– Continued fall in fertility: if household size had not fallen (from 5.2 to 4.9)
between 2000 and 2005, poverty reduction would have been cut by half
• Whether the projections are met also depends on
– Frequency of shocks (e.g. food prices, natural disasters) can reduce GDP
growth and responsiveness of poverty to growth
• Had there been no rice price shock, there would be around 4.2 million fewer poor
people in Bangladesh in 2008
• Estimated poverty reduction during 2005-2008 would have been 5 pct points;
instead ~2 percentage points with the rice price shock (i.e. from 40 to 38%)
• Rice price shock is also estimated to have raised poverty gap (average deficit of
the poor’s consumption relative to the poverty line) by 30%
– How deep and long the current global recession will be (impact on
remittances and RMG exports)
7
Rise in non-income indicators of welfare
% increase between 2000 and 2005
120
100
80
60
40
20
0
Livestock
ownership
Cement/CI
sheet wall
Cement/CI
sheet roof
Safe latrine
use
Electricity
connection
All households Bottom 30%
• Many non-income indicators improved more significantly for
the extreme poor than for the overall population
• Consistent with inequality in consumption not worsening
8
What explains poverty reduction during
2000-2005?
Significant social and economic transformation
• Economic transformation – closely related to strong GDP
growth (>5% annually) and urbanization
– rising returns to human and physical assets, especially wages
– shift from low return agricultural labor to nonfarm employment
(mainly services) in urban areas
– growth in remittances and exports
• Forces emerging from social transformations over time
– A fall in the number of dependents in a household, linked to past
reductions in fertility
– Increases in labor force participation and educational attainment,
particularly among women
9
Economic transformation: labor markets
•
Poverty reduction attributed to rising labor productivity and wages, and (to a lesser extent)
to shift from agriculture to non-farm employment
– Increase in labor productivity in industry contributed most significantly to income growth
– Share of agriculture in employment fell from 51% to 46%, that of services grew from 27% to 31%
– Overall job creation during 2000-2005 kept pace with population growth
•
Poverty reduction by sector
– Highest within services due to employment growth (5.4% annually, compared to 0.7% in
agriculture and 3.9% in industry)
– Substantial in agriculture due to small productivity growth and the large share of population
employed in agriculture
– Within industry driven primarily by productivity growth
Decomposition of Changes in
GDP per Capita into Components
Contribution to changes in GDP per capita
Productivity by Sector
Poverty Changes by Sector
Contribution to grow th in GDP per
w orker, by sector
Share of
w orking age
population
Change in poverty associated with
poverty changes w ithin sector
inflow /outflow to/from sector
Agriculture
by broad sector
1%
0%
Industry
Productivity
-1%
Services
Employment
(rate)
-20%
-2%
-3%
Intresectoral
flow s
0%
20%
40%
60%
-4%
80%
0%
Source: HIES 2000, 2005
10%
20%
30%
40%
Agriculture
Industry
10
Services
Demographic transition creates opportunities and
challenges in the labor market
Population (LHS) and
its grow th (RHS)
250
200
95-99
90-94
85-89
80-84
75-79
70-74
65-69
60-64
55-59
50-54
45-49
40-44
35-39
30-34
25-29
20-24
15-19
10-14
5-9
0-4
2005
2.5%
2.0%
150
1.5%
100
1.0%
50
0.5%
0
0.0%
Women
Men
-8
-6
-4
-2
0
2
4
6
8
2050
2045
2040
2035
2030
2025
2020
2015
2010
2005
2000
1995
1990
1985
1980
1975
1970
milion-10
total
w orking age
total
w orking age
% per
year
Although population growth is now
1.5% per year, the working age
population is growing at 2.5-2.8 %
million
•
10
• Large cohorts will enter the labor market; ~22 mn. new entrants will
need to be absorbed into the labor market between 2005 and 2015
• Annual rate of job creation has to double in the years up to 2015,
compared to the rate during 2000-2005
11
Rising contribution of women in the labor market
• Women’s labor force participation rates, working hours,
education, income have increased more than those of men
between 2000 and 2005
• Women finding jobs increasingly in
– Urban areas, public sector
– Self-employment outside agriculture and using formal financing
(micro-credit)
• Women’s labor income is important for poverty reduction
but its full potential remains unexploited
– women’s labor market participation still too small to make a
significant dent in poverty
– growth in women’s participation and incomes largely concentrated
among the nonpoor
12
Remittances and poverty
• International remittances grew at 20% annually during 2000-05;
Bangladesh among the top 10 remittance receiving countries
• Poverty rate among households receiving remittances from abroad is
17% compared to 42% among the rest
• Areas with higher incidence of remittances less likely to be poor; 24%
of households in Chittagong division and 16% in Sylhet received
remittances, compared to <5% in the rest
• CGE simulations attribute a little above 15% of the poverty reduction
to the effect of growth of foreign remittances
• However, likely global recession poses clear risks to remittance inflow
– 63% of remittances come from the Gulf where construction industry is
likely to take a downturn
13
Emerging regional divide in poverty
57
51
53 52
47
50
40
46
32
45 46
34
42
34
30
20
2000
Syllhet
Khulna
Barisal
Rajshahi
0
Chittagong
10
Dhaka
Headcount rate (%)
60
2005
• Rising inequality between the East and West
– Dhaka and Chittagong contributed 79% of national poverty
reduction with just over half the population; Khulna and Barisal
with 20% of total population had no contribution
– Till 2000, the largest difference was between Dhaka and the rest of
the country; from 2000 to 2005, divergence between East and West
• East-West gap in poverty rate doubled from 2000 to 2005
14
Growth poles and the East-West economic divide
•
Regions differentiated by access to
growth poles
– Dhaka and Chittagong: growth
poles (shares in population,
economic activities)
East
– Padma and Jamuna rivers obstacles
to access to the growth poles
•
East-West gap widened from 2000
to 2005
– Average consumption in East 9%
higher in 2000, 17% higher in 2005
West
– Much faster consumption growth
for Eastern poor than Western poor
– Gaps in both endowments and
returns contributed to the widening
East-West consumption gap
– Gap in endowment widened for all;
gap in returns widened for the poor
15
Labor markets – East and West
•
Real wages growing robustly in the East but stagnating in
the West
–
–
•
Total labor income growth in the East was twice that of the West
Slow growth of wages in the West mostly explained by stagnation
in the urban areas
Labor markets and the types of jobs they offer are also
different between regions
–
–
In the East, salaried jobs dominate; in the West, farming remains
important, with high share of low-paid daily waged workers
43% of labor income in the West from agriculture, compared to
25% in the East
•
Higher returns to education in the East than West;
difference especially large for urban areas
•
Gender difference and public sector premium are lower in
East, likely pointing to better labor market integration
16
Economic concentration increasing over time
Changes in employment in firms with TPE>10
(2003-2006)
Changes in employment in Dhaka city and
surrounding areas (2003-2006)
Note: Employment in firms with TPE of 10+; green indicates positive change, with darker green indicating higher values; red
indicates negative change, with darker red indicating higher absolute values.
•
•
Clustering of formal sector employment around Dhaka and Chittagong
Large increase in formal sector employment in East from 2003 to 2006
– In areas to the north and west of Dhaka City Corporation, but not the city center
•
Rapid economic expansion to the north and west of Dhaka city (Gazipur, Savar)
– cheaper land and labor, less congestion, while allowing firms to exploit the spillovers
17
and positive externalities from Dhaka city
Why has the East-West economic gap expanded?
• Empirical evidence suggests the following story
– Increasing agglomeration of high-return economic activities at
growth poles have led to strong spillover effects in surrounding
areas and higher incomes within East
– East-West differences have expanded because West is handicapped
by the absence of growth poles, poor connectivity with urban
centers, and deficient public infrastructure and markets
– While the better-endowed households from West can respond to the
economic opportunities in East by migrating, the poor are mostly
unable to overcome barriers to their mobility
• Poverty incidence also varies widely within East or West
– Agro-climatic factors affecting specific areas; e.g. “Chars”,
Chittagong Hill Tracts, salinity of land in south
– Vulnerability to natural disasters/seasonal shocks: areas with higher
risk of cyclone or monga more likely to be poor
18
Human development: current challenges
Malnutrition continues to be strikingly high
• High malnutrition
compared to countries of
same income level
• Fortification of basic
staples in nutritional
program may be key to
further nutritional gains
1996-97
Percentage of Children Under 5
• Between 1996-97 and
2006, nutritional gains
plateauing in terms of %
of underweight children
Trends in Nutritional Status of
Children Under Five 1996 to 2007
60
50
40
1999-2000
2004
2007
56
55
45
48 48
43
46
36
30
18
20
10
13
16
10
0
Stunting
(height for
age)
Wasting
Underw eight
(w eight for (w eight for
height)
age)
Source: Based on Preliminary Bangladesh
DHS 2007 report
19
Inequities persist in health service utilization
Assisted delivery by m edically trained
person, by w ealth quintile
• Several indicators show increasing
equity divide in health service
utilization despite gains overall, e.g.:
Percentage
– Immunization
– Ante-natal care from a medical
trained provider
– Deliveries at both public and private
facilities
Poorest
Richest
50
40
30
20
10
0
1996/97
1999/00
2004
Source: DHS data (different years)
Im m unization, by w ealth quintile
– Absenteeism of staff
– Ratio of nursing staff to doctors
remains low
Richest
100
Percentage
• Access to care hindered by
challenges in human resources,
despite extensive network of
facilities
Poorest
80
60
40
20
0
1996/97
1999/00
2004
Source: DHS data (different years)
20
Mixed progress in education since 2000
• Little change in primary gross enrollment since 90 %
enrollment rate attained in 2000
– But pattern of late entry into school: over 30% of children of
primary school going age not attending primary
• Substantial growth in secondary enrollment (52% to 62%
from 2000 to 2005).
– However, completion rates declining with rising rich-poor gap:
only 6-7% of the poor have completed secondary schooling
• Public education expenditure per student significantly
lower than other countries in SAsia
– Appears to be compensated by high private tuition expenditures
which worsen inequalities
– Room to make expenditures more pro-poor (e.g. primary stipend
program benefits large share of non-poor)
21
Regional patterns in human development
• Human dev outcomes in poorer divisions are often better
– E.g. child mortality is lowest in Barisal and secondary enrollment is
highest in Khulna; Sylhet is lagging behind in most HD indicators
(e.g. girls secondary enrolment, infant mortality rate)
• The contradiction between economic and social outcomes
– do differences in social norms/conservatism play a role?
Male
Source: Al Samarrai
Female
0
Male
Female
Sylhet
20
Rajshahi
Sylhet
Rajshahi
Khulna
Dhaka
Chittagong
0
40
Khulna
20
60
Dhaka
40
80
Chittagong
60
100
Barisal
80
Gross secondary enrolment by division(2005)
Gross sec enrollment (%)
100
Barisal
Gross prim enrollment (%)
Gross primary enrolment by division (2005)
22
How do households cope with shocks?
• Rapid national survey of 2,000 households to
assess the impact of rice price shock (July, 2008)
• The way households cope may have longer-term
adverse consequences for productive assets and
human capital
– 76% of households reduced quantity of food intake;
88% switched to lower quality food
– 8% took children out of school; 39% reduced education
expenses
– 45% drew upon savings or pawned belongings
23
Therefore the need for effective safety nets
• Government has raised safety net expenditures
steadily since mid-1990s
• But public resource allocation across divisions is
puzzling
– Not consistent with poverty incidence; e.g. 22% of
households in Sylhet safety net beneficiaries while
national average is 13%
• Low coverage and transfer values
– Even among bottom 10%, less than one-fourth receive
any safety net benefit
• Little coverage of urban poor except for OMS
24
Looking ahead: accelerating the pace of
poverty reduction
(A) Raising growth beyond 7%
– Stable macro and political environment, removing infrastructure
bottlenecks, export diversification, greater female participation in labor
force and sustaining remittance growth
– Investments to raise agricultural productivity (closing the “yield gap” with
other developing countries) –will reduce regional disparity since lagging
areas rely disproportionately on agriculture
(B) Spreading growth to lagging regions
– Investment in physical infrastructure (roads, Padma bridge) to improve
access to markets and existing growth poles
– Investment in urban infrastructure and services in lagging regions
– Investment in human capital/skills – conditional transfer programs can
serve dual (safety net and human devlopment) objectives
– Improving access to remittances by reducing barriers to migration
– Spatially targeted incentives and complementary investments to spur the
creation of regional growth poles
25
Looking ahead (cont.)
(C) Moderating population growth rates, tackling malnutrition and
infant and maternal mortality
- Measures to increase use of contraceptives, scaling up demand side
interventions (e.g. maternal care voucher pilot) and provision of
fortified food
(D) Reducing inequities in education and strengthening quality
- Raising public education spending in line with regional benchmarks
and focus on access by poor, preventing drop-outs and
strengthening links with labor market
(E) Expanded role for safety nets
- Current fiscal allocation for safety nets can be used to expand
benefits to most needy by improving targeting and consolidating
programs under an umbrella body
- Regional allocation of safety net resources need to be aligned better
with poverty incidence
26