Poverty Implications of Trade

Liberalization in Bangladesh: A
General Equilibrium Approach

Mustafa Mujeri

Bangladesh Institute of Development Studies

Bazlul Khondker

University of Dhaka

August 2002

This paper forms part of the 'Exploring the Links Between Globalisation and Poverty in
South Asia' study which is part of the Globalisation and Poverty Programme, funded by the
Department For International Development (DFID) UK. The Programme includes fourteen
projects on a three-year programme of research exploring the linkages between globalisation
processes and poverty.
 Contents

Page No

1. Introduction 1

2. The Bangladesh Economy and Trade Policies:

Main Features and Changes 3
2.1 Reduction of Import Barriers 4
2.2 Export Promotion Measures 6
2.3 The Exchange Rate System 7
2.4 Trade and Investment Flows 7

3. Poverty, Labour Market and Employment 10

3.1 Recent Trends in Poverty 11
3.2 Changes in Inequality 12
3.3 Developments in the Labour Market 13
3.4 Trade Liberalization and Labour Market Developments 15

4. Objectives of the General Equilibrium Framework 19

5. Outline of the Methodology 19

5.1 The Bangladesh Social Accounting Matrix for 1995/96 19
5.2 The CGE Model 21

6. Simulation Designs 25

7. Simulation Results 27

8. Concluding Observation 36

References 40

[Link] ii
 List of Tables

Page No

Table 1: Structure and Growth of the Bangladesh Economy 3

Table 2: Import Liberalization in Bangladesh 5
Table 3: Nominal and Effective Protection Rates in Bangladesh 6
Table 4: Growth and Structural Changes in Bangladesh’s Merchandise Trade 8
Table 5: Openness Ratio of the Bangladesh Economy 8
Table 6: Capital Flows into Bangladesh 9
Table 7: Major Elements of Bangladesh’s FDI Policy 10
Table 8: Changes in Incidence of Poverty in Bangladesh 11
Table 9: Poverty Reduction Rates during Pre- and Post Liberalization Period 12
Table 10: Changes in Growth and Inequality in Bangladesh 13
Table 11: Trends in Employment in Bangladesh 14
Table 12: Underemployment in Bangladesh, 1995/96 15
Table 13: Employment and Wage Rate Changes in Agriculture and
Non-Agriculture Sectors in Bangladesh 17
Table 14 Distribution and Changes in Real Per Capita Income 18
Table 15: Regional Variation in Per Capita GDP 18
Table 16: Disaggregation and Description of Factors, Institutions and Households 21
Table 17: Summary of Model Features 25
Table 18: Tariff Rates for Base Year and Simulation Experiment 26
Table 19: Selected Macro Effects of the Simulations with the Model 27
Table 20: Welfare Impacts of Simulations: Equivalent Variations for Different
Household Groups 29
Table 21: Data Layout in the 1995/96 Household Expenditure Survey 31
Table 22: Base Values of Household Poverty Profiles 34
Table 23: Income Effects Under Alternative Simulations 35
Table 24: Impacts of Policy Simulations on Poverty Profiles 36

List of Figures and Appendix

Figure 1: Structure of Production 22

Figure 2: Structure of Final Demand 23
Figure 3: Ratio of Foreign Savings to GDP 26
Appendix I: Summary of the CGE Model Specification 43
Appendix 2: Bangladesh Social Accounting Matrix for 1995-96 45

[Link] iii
 Poverty Implications of Trade Liberalization in Bangladesh:
A General Equilibrium Approach

1. Introduction

The process of globalization involves greater participation of individual countries in world

trade, foreign direct investment and the capital market. Over the last few decades, reduction
of natural barriers such as transportation and communication costs and revolution in
information technologies and removal of artificial impediments like tariff and non-tariff
barriers have fostered greater interdependence among the countries of the world through
increased trade and private capital flows. During the period, a growing number of developing
countries have also adopted outward-oriented liberalization measures in the hope that greater
global integration of their economies will lead to faster poverty reduction through
accelerating economic and productivity growth and thereby supporting the consumption gains
of the people. The contribution of globalization to growth and poverty reduction, however,
varied considerably across the countries. At the same time, concerns have been expressed
regarding the potential long-term negative effects of globalization which may outweigh the
positive impacts of economic integration. Moreover, the consensus seems to suggest that the
effects of globalization are likely to be country-specific and should be examined keeping the
country context in view.

In Bangladesh, one of the significant features of globalization has been the introduction of
measures to bring liberalization and openness in the economy. Since the 1980s, the
Government undertook reform programmes to accelerate growth through wide-ranging
policies to improve competitiveness, enhance economic efficiency, and dismantle state
interventions to create conditions for promoting export-led growth. As a part of the process,
significant reforms have been implemented in trade regime by liberalizing external trade and
foreign exchange regulations and introducing deregulatory measures to facilitate increased
participation of the private sector. Both tariff and non-tariff barriers have been reduced along
with dismantling of quantitative restrictions on imports and deregulation of import
procedures. More specifically, the measures to rationalize the tariff structure involved
reduction in average tariff rates and the number of duty slabs, lowering the gap between the
statutory nominal protection and the observed levels, narrowing down the tariff dispersion
and minimizing the control list of banned and restricted items.1 The extent of liberalization
is reflected in the fact that the mean tariff rate for all products declined to 22 per cent in 1999
from 114 per cent in 1989. Moreover, the decline was sharp for Bangladesh compared to
other South Asian countries.2

While these reforms have significantly changed the policy environment in the country, a
proper assessment of the impact of these changes, particularly their distributional
consequences, requires a comprehensive framework capable of analyzing the interactions
between different sectors along with linkages between macro-policies and various household
groups. In particular, the liberalization policies raise several issues relating to poverty in
Bangladesh: How does trade liberalization affect the welfare status of different
socioeconomic groups especially the poor? Are countervailing policies needed to make such
policies more equitable? Evidently, the issues are complex and limited availability of
empirical evidence on the nature of impact of these policy changes makes it difficult to draw
specific policy conclusions. Moreover, the poverty linkages, to a large extent, depend on
propagation channels through which the impacts of trade liberalization are transmitted to
different economic sectors and socioeconomic groups.

The present study seeks to assess the characteristics of the transmission channels of trade
liberalization policies in Bangladesh and examine their poverty implications. Although
significant interactions exist among different reform measures in practice, the study
concentrates on trade reforms alone along with welfare implications of these reforms in terms
of impact on absolute and relative poverty. The aim of the analysis is to contribute to better
understanding of the relationships between trade reforms and poverty in Bangladesh, a least
developed country, and help identify policy options that are capable of promoting
liberalization in a more equitable manner.

1 As a result, the Bangladesh economy has become more open in the 1990s compared to any period in the past.
Several indicators highlight the extent of trade liberalization achieved in the 1990s. The highest rate of customs
duty was reduced from 350 per cent in 1991/92 to 37.5 per cent in 1999/00. Similarly, four slabs of duty rates
were introduced in 1999/00 in place of 24 in the 1980s and the number of items banned and/or restricted due to
trade or non-trade reasons at the 4-digit Harmonized System (HS) code level declined from 315 in 1989/90 to
124 in 1997-2002.
2 In India, for instance, the mean tariff for all products declined to 33 per cent in 1999 from 82 per cent in 1990
and, in Sri Lanka, the decline was from 28 per cent in 1990 to 20 per cent in 1997. See World Bank 2000.

[Link] 2
2. The Bangladesh Economy and Trade Policies: Main Features and Changes

The Bangladesh economy has undergone significant sectoral changes since the 1980s (Table
1). The share of agriculture in GDP has declined to around a quarter by 2000 while nearly a
half of the GDP comes from the service sector.3 The average rate of GDP growth has been
4.3 per cent per year since the 1980s although the economy has shown a better growth
performance in the 1990s. The growth in per capita GDP also accelerated during the 1990s
both due to increased economic growth and reduction in the rate of population growth.

Table 1: Structure and Growth of the Bangladesh Economy

A. Structure
Share (per cent) in GDP at constant 1995/96 prices
1980 1990 1995 2000
Agriculture 33.2 29.5 26.0 25.6
Industry 17.1 20.8 24.3 25.7
Services 49.7 49.7 49.7 48.7
Total 100 100 100 100

B. Growth
Per cent at constant 1995/96 prices
1981-2000 1981-1990 1991-1995 1995-2000
Agriculture 2.8 2.3 1.6 4.9
Industry 6.4 5.8 7.5 6.4
Services 4.8 3.7 4.1 4.8
GDP 4.3 3.8 4.4 5.2
Per capita 2.3 1.6 2.4 3.6
GDP

Source: BBS 2000, 2001.

Since independence in 1971, three major phases of evolution in trade policies may be
identified in Bangladesh. The period covering 1972 to 1978 was characterized by the pursuit
of an import-substitution strategy through quantitative restrictions on imports, import
licensing and strict exchange control measures. The distorted incentive structure of the
period, however, led to allocative and productive inefficiencies, strained the external sector,
created anti-export bias, and consequently resulted in low growth of the economy. This
prompted the policy makers to introduce reforms towards a free market economy and export-
led industrialization although at a relatively slow rate over the 1979-1990 period.4 The third

3 Despite the declining relative share of agriculture in GDP, agriculture continues to remain the major sector in
terms of employment with about 62 per cent of total employed persons in 2000.
4 The reform programmes of the period included different measures like fiscal, financial, trade and industrial
policy reforms; public resource management and privatization; and institutional and sectoral reforms. These
economy-wide reforms and structural adjustments initiated in 1986/87 formed components of the Structural
Adjustment Facility (SAF) and the Enhanced Structural Adjustment Facility (ESAF) of the IMF and the World
Bank. For details on evolution of these policies, see Sobhan 1991, Mujeri et. al. 1993.

[Link] 3
phase (1990-2000) was characterized by greater openness of the economy through
accelerated trade liberalization, financial and fiscal reforms, and privatization.

The Bangladesh economy was highly protected and inward looking until the end of the
1970s. During 1978, for example, there were 36 different tariff rates ranging from zero to
400 per cent. Quantitative restrictions were also widespread. The reason for pursuing such a
restrictive trade policy was two-fold: to protect domestic industries and to raise revenue.
This, however, resulted in an expansion of inefficient industries and misallocation of
resources with adverse consequences on the export sector and the economy. Trade reforms,
launched in the 1980s, were aimed mainly at rationalizing and reducing tariffs and other
import taxes, and eliminating import prohibitions and quantitative restrictions. Incentives
were also introduced to boost exports and diversify the export base. In the 1990s,
Bangladesh embarked on a liberal trade and investment policy. The 1991 Industrial Policy,
for example, targeted the expansion of export-oriented industries and employment creation
through attracting foreign investment and removing all barriers to make the industrial sector
more efficient and internationally competitive.

2.1 Reduction of Import Barriers

The primary objective of reducing import barriers over the last two decades was to rationalize
and simplify the trade regime through lowering the tariff rates, phasing out the quantitative
restrictions, streamlining import procedures and introducing tax reforms. These reforms
brought significant changes in the overall tariff structure. Several features of the changes
may be summarized as follows:
(i) The number of commodities under the four-digit code subject to quantitative
restrictions declined from 550 in 1987 to 124 under the Import Policy of 1997-
2002. In 1991/92, about 12 per cent of around 10,000 tariff lines were subject
to such restrictions which declined to less than 4 per cent in 1998/99. At
present, less than 0.5 per cent of imports, mainly in the textile category, are
subject to quantitative restrictions;
(ii) Average tariff rates have been significantly reduced. The mean tariff on all
products declined from 114 per cent in 1989 to 22 per cent in 1999 and the
weighted mean tariff from 114 per cent to 19 per cent over the same period
(Table 2);

[Link] 4
Table 2: Import Liberalization in Bangladesh
A. Removal of Quantitative Restrictions (QRs) at 4-Digit HS Code-Level
Year Total QRs Trade reasons Non-trade
in placea Banned Restricted Mixed reasons
1987 550 252 151 86 61
1989 433 165 89 101 78
1990 315 135 66 52 62
1992 193 78 34 25 56
1995-1997 120 5 6 17 92
1997-2002 124 5 6 17 96
B. Changes in Tariff Barriers (Per cent)
1989 1999
A. All products
Mean tariff 114.0 22.1
Standard deviation of tariff rates 84.9 14.6
Weighted mean tariff 114.2 19.0
B. Primary products
Mean tariff 85.1 21.1
Standard deviation of tariff rates 58.7 13.1
Weighted mean tariff 76.1 21.0
C. Manufactured products
Mean tariff 123.2 22.4
Standard deviation of tariff rates 89.8 15.0
Weighted mean tariff 125.5 18.5
a
There are a total of 1,240 four digit tariff headings under the Harmonized System.
Source: Mujeri 2000, World Bank 1999, 2000.

(iii) The combination of maximum tariff rate reductions and a tariff decrease from
2.5 per cent to zero per cent on some products led to a narrowing down of
tariff bands;
(iv) Import taxes such as development surcharges, regulatory duties and sales taxes
were abolished in 19915; and
(v) Various measures were introduced with a view to simplifying import
procedures. In 1985/86, two lists were introduced to replace the ‘positive list’
(which contained all goods that could be imported into Bangladesh along with
their constituent raw and packing materials): all banned items were listed
under a ‘negative list’ and those importable under certain conditions were
registered on a ‘restricted list’.6 All other products could be imported freely.
Over the years, Import Policy Orders showed substantial reduction in the
number of banned and restricted items.

5 Despite reduction in the tariff rates, total tariffs still remain high by international standards since, in addition to
customs duty, other taxes are also levied on imports e.g. value added tax, supplementary duty, infrastructure
development surcharge and license fee. Thus, although the average implicit (expost) customs duty at present is
around 14 per cent, the total average expost ‘tariff’ is approximately 27 per cent.
6 For example, out of the 391 items which were listed under the ‘negative list’ in 1985/86, only 24 remained in
1993/94. Similarly, the number of items on the ‘restricted list’ over the same period was reduced from 351 to
91.

[Link] 5
Overall, Bangladesh’s trade policies were liberalized rapidly in the 1990s. As a result, the
economy has become significantly outward-oriented both due to quantitative changes in tariff
and removal of non-tariff barriers.7 It has been observed that Bangladesh’s ‘nominal import
protection level currently ranks among the lowest in South Asia’ and that ‘tariff reduction in
Bangladesh during the early 1990s has been one of the fastest amongst the reforming
countries’ (CPD 1997). The extent of protection of the domestic economy also declined due
to changes in the tariff structure. The effective rate of protection (ERP) declined from 76 per
cent in 1992/93 to 27 per cent in 1998/99 (Table 3). While significant liberalization has been
achieved in the 1990s, the scope for further reduction, compression and rationalization of the
tariff structure still exists with a view to reducing economic distortions and welfare losses
resulting from the trade policy.

Table 3: Nominal and Effective Protection Rates in Bangladesh

1992/93 1995/96 1998/99
Nominal protection rate:
Unweighted 55.4 27.1 27.2
Weighted 28.2 22.3 20.3
Effective protection rate: 75.7 33.0 26.8

Source: Mujeri 2000.

2.2 Export Promotion Measures

Over the years, the Government attempted to promote exports through various measures.
The policies for export promotion emphasized the need to diversify the export base, stimulate
higher value-added exports, improve the quality of exports, develop backward linkage
industries and undertake vigorous marketing efforts. Incentives are provided to the exporters
in the form of special bonded warehouses, export processing zones (EPZs), duty drawback

7 The outward-orientation or openness of an economy is, however, difficult to measure. See Pritchett 1996.
While tariff and non-tariff barriers are widely used indicators, movements in the real exchange rate can have
significant impact on an economy’s response to trade reform. Even with tariff and non-tariff barriers, the
appropriate summary measures may differ. For instance, nominal tariffs give a better indication of the distortion
in consumption while effective tariffs better represent the distortion in production. Similarly, production-
weighted effective protection rate may be more appropriate to measure the distortion in production and
consumption-weighted nominal tariff for the consumption distortion. Moreover, these measures may not reflect
the true picture since other trade policies (e.g. duty drawbacks, export processing zones, export subsidies) have
been used by Bangladesh as countervailing measures for export promotion. These are discussed in the next
section. Moreover, measures of non-tariff barriers usually reflect the coverage of particular restrictions but not
the severity of their application.

[Link] 6
and a number of other methods.8 Against the backdrop of phasing out of the Multi Fibre
Arrangement (MFA) by 2005 and the facilities that Bangladesh will have to forgo as a LDC
for the export of its textile products, the country aims at achieving self-sufficiency in fabrics
to meet the requirements of the garment industry through establishing backward linkages. It
is important, however, to encourage the promotion of backward linkages to the extent that
they do not adversely affect export competitiveness. An important concern in export
promotion in Bangladesh is to ensure enhanced coherence and consistency in export policy
through adopting a uniform export strategy that allows the private sector to respond to
predictable and stable market incentives.

2.3 The Exchange Rate System

In line with the overall framework of trade reforms, gradual liberalization of foreign
exchange restrictions has also been implemented. The Government replaced the policy of
maintaining a multiple exchange rate system by a unified exchange rate in 1992 and the
domestic currency, Taka, was pegged to a currency-weighted basket. Since then, a policy of
creeping devaluation has been followed to maintain exchange rate flexibility and export
competitiveness within a more market-determined exchange rate regime. The Taka has been
made convertible for all current account transactions along with measures to set in motion the
inter-bank foreign exchange market.

2.4 Trade and Investment Flows

The trade liberalization process since the 1980s has been associated with considerable
intensification of trade and investment flows. The growth and structural changes in
merchandise trade can be seen in Table 4. Compared with an average annual growth of about
8 per cent in the 1980s, merchandise exports in both volume and value terms increased by
around 13 per cent per year in the 1990s. In the case of imports, the rates increased
substantially to around 9 per cent for volume and 12 per cent for value in the 1990s compared
with less than 2 per cent and 4 per cent for volume and value respectively in the 1980s. The

8 The system of support to exporters that exists in the country is highly complex, fragmented and consists of
wide range of different measures applying in specific circumstances. Several measures may be noted e.g. rebate
on insurance premiums, income tax rebate, export credit guarantees, incentives for export of non-traditional
industrial products, export promotion fund, retaining foreign exchange from export earnings, VAT refunds, tax
holiday and other incentives.

[Link] 7
export trade basket, moreover, indicates an increasing concentration of a broad category of
manufactured goods consisting of an assortment of simple manufactured goods like
readymade garments, leather and leather products, fabrics, and made-up articles which
accounted for 91 per cent of total merchandise exports in 1998. This shows that
Bangladesh’s exports have increased through exporting more of the same or similar goods
and, from this perspective, have displayed little dynamism. Nevertheless, the ‘openness
ratio’, as measured by trade to GDP ratio, has increased. The share of foreign trade
(merchandise exports and imports) in GDP increased from 19 per cent in 1984/85 to nearly
35 per cent in 2000/01 (Table 5). The ratio, however, is still low in relation to the ratio
observed for all developing countries (43 per cent in 1990-94). Similarly, the pace of
integration has been slow at 0.6 per cent per year which, although better than the average of
South Asian countries, is lower than the average of all developing countries (0.7 per cent).9

Table 4: Growth and Structural Changes in Bangladesh’s Merchandise Trade

A. Growth Average annual % growth
1980-1990 1990-1997
Export: Volume 7.6 12.9
Value 7.6 13.2
Import: Volume 1.8 9.1
Value 3.7 11.8
B. Structural Change Merchandise exports Merchandise imports
1980 1998 1980 1998
Total value (US$ million) 793 5,141 2,353 6,862
% of total
Food 12 7 24 15
Agricultural raw materials 19 2 6 5
Fuels 0 0 9 7
Ores and metals 0 0 3 2
Manufactures 68 91 58 69
Source: World Bank 2000.

Table 5: Openness Ratio of the Bangladesh Economy

1984/85 1989/90 1994/95 2000/01
As per cent of GDP
Imports 13.2 13.5 17.4 20.1
Exports 5.6 6.1 10.9 14.6
Total trade 18.8 19.6 28.3 34.7
Openness ratio 18.8 19.6 28.3 34.7
Nominal level (per cent) Annual average change
1975-79 1990-94 (per cent)

Regions
Developing countries 31.8 42.8 0.7
East Asia 31.2 54.6 1.6
South Asia 17.6 25.1 0.5
Bangladesh 19.1 28.0 0.6
Source: BBS 2000, 2001 and World Bank 1997.

9 Two alternative measures, trade in goods as a share of PPP GDP and goods GDP, also indicate increased global integration
of the Bangladesh economy in the 1990s. The trade in goods as a share of PPP GDP increased to 7.0 per cent in 1998 from
4.2 per cent in 1988. Similarly, the share of trade in goods GDP increased to 56.1 per cent from 29.9 per cent over the same
period. The dynamism of the trade regime, as measured by the difference in growth in real trade and growth in real GDP,
was also high at 7.2 per cent during the period. See Mujeri 2002.

[Link] 8
The level and pace at which foreign direct investment (FDI) increases are important
indicators of the global financial integration of a country. Bangladesh’s exposure to natural
disasters and a high propensity to import, among other factors, have persistently contributed
to chronic current account deficits in the country. A key policy issue has, therefore, been to
find ways to meet the structural deficits of the current account balance. In this respect,
measures of attracting FDI have proved to be largely unsuccessful. Despite its comparative
advantage in terms of low labour costs, FDI in Bangladesh reached only US $ 125 million in
1995/96 growing from an average of less than US $ 10 million in the previous five years
(Table 6). The FDI registered with the Board of Investment (BOI), however, indicates a
rising trend which is six times higher than actual FDI flows.10 According to actual FDI
flows, FDI to GDP ratio was only 0.03 per cent in the early 1990s. The ratio is low compared
with the ratios for all developing countries (1.18 per cent) and the South Asian countries
(0.44 per cent). Similarly, the pace of financial integration is also slow compared with the
pace observed in developing countries and other South Asian countries.

Table 6: Capital Flows into Bangladesh

(million US $ unless stated otherwise)
1990 1991 1992 1993 1994 1995 1996
Total flows: 1708 1684 1440 1030 1485 906 1247
Official flows 1638 1648 1423 1023 1455 896 1203
Private flows 70 36 17 7 30 10 44
of which:
Foreign direct investment 3 1 4 14 11 2 15
Portfolio equity investment 0 0 0 0 0 0 0
Other private flows 67 35 13 -7 19 8 29
Per cent of GDP
Total private flows 0.31 0.15 0.07 0.03 0.12 0.04 0.14
Foreign direct investment 0.01 0.00 0.02 0.06 0.04 0.01 0.05
Portfolio equity investment 0 0 0 0 0 0 0
Other private flows 0.30 0.15 0.05 -0.03 0.07 0.03 0.09
Memorandum Items Nominal level Annual average
1975-79 1990-94 change
(per cent) (per cent) (per cent)
Developing countries 0.84 1.18 0.02
East Asia 0.87 2.69 0.12
South Asia 0.10 0.44 0.02
Bangladesh 0.00 0.03 0.00
Note: Other private flows consist of flows from commercial banks, bonds, and other private sources.
Source: World Bank 1997 and own estimates.

As a result, the country remained highly dependent on official aid flows from multilateral and
bilateral donors. Between 1971 and 2001, total foreign aid disbursement to Bangladesh

10 The BOI approved FDI data refer to registered amount which may not be realized due to various factors. As
a result, there exists considerable gap between BOI approved FDI figures and actual FDI inflows.

[Link] 9
amounted to US $ 37.7 billion, 48 per cent of which was in the form of grants and the rest in
loans. Although loans have assumed greater importance relative to grants in total aid flows in
recent years, Bangladesh’s debt service payment does not weigh heavily on the balance of
payments. The debt-service ratio in 2001 was about 14 per cent of the country’s merchandise
exports representing about 2 per cent of GDP which is lower than the similar ratio in other
South Asian countries (e.g. India and Pakistan) due to the concessional terms of the loans
received by the country.

In recent years, Bangladesh has significantly improved its investment and regulatory
environment which includes liberalization of industrial policy, abolition of performance
requirements, and allowing full foreign-owned joint ventures. New sectors have been opened
up for foreign investment including the telecommunications sector in 1996 (Table 7).
Nevertheless, measurable indicators of globalization such as the openness ratio, pace of
integration into the global economy, and FDI to GDP ratio indicate that the extent and pace
of integration of the Bangladesh economy into the global economy still remains low and the
process has been relatively slow in the past. The Government’s commitment is to improve
the situation for which liberal trade and investment policies have been emphasized.

Table 7: Major Elements of Bangladesh’s FDI Policy

Elements Description
Entry Barrier Closed: Defense equipment, International air transportation, Railway
transportation, Security printing, Forestry and Nuclear energy.
Full Foreign Ownership Requirement 100 per cent ownership permitted with approval.
Performance Requirement No requirement.
Transfer of Profits and Convertibility Restricted: Provision of transfer subject to control by BOI and the
Central Bank
Incentives EPZs: Interest on foreign loan is tax exempt. Tax exemption on
royalties, technical know-how, technical assistance fees and facilities
for their repatriation.
Source: World Bank 1997.

3. Poverty, Labour Market and Employment

In this section, we shall focus on changes in two areas: poverty and income distribution, and
the labour market. In a low-income country such as Bangladesh, the developments in the
labour market are important determinants of poverty. The labour market’s role in the country
largely derives from its limitations in providing productive and gainful employment
opportunities to the large majority of the labour force. With low skills and the limited scope

[Link] 10
of employment in the formal sector, the vast majority of the labour force subsists in low-
productive informal activities with limited ability to generate decent incomes required for
moving out of poverty.

3.1 Recent Trends in Poverty

The inter-temporal estimates of poverty in Bangladesh show substantial variations due to

differences in underlying assumptions and methodologies.11 Nevertheless, some trends can
be discerned with the available data (Table 8). It shows that the incidence of poverty, as
measured by the head-count index, declined to 50 per cent in 2000 from 59 per cent in
1983/84. Both urban and rural poverty have declined although the incidence of rural poverty
remains higher than that of urban poverty. Over the entire period since the early 1980s,
poverty incidence has declined at a slow rate with substantial variations over different sub-
periods and between rural and urban areas. Between 1984 and 2000, rural poverty declined
by only 7 percentage points. Given the fact that 80 per cent of the poor live in rural areas,
total poverty incidence declined by only 9 percentage points although the decline in incidence
of urban poverty has been higher. Moreover, the absolute number of the poor increased to 63
million from 56 million between 1984 and 2000 – an increase of 7 million over a period of 16
years when total population increased by about 30 million.

Table 8: Changes in Incidence of Poverty in Bangladesh

Year Head count ratio (per cent) Number of poor (million)

Rural Urban Total Rural Urban Total
1983/84 59.6 50.2 58.5 50.3 5.6 55.9
1988/89 59.2 43.9 57.1 53.7 5.7 59.4
1991/92 61.2 44.9 58.8 57.5 6.4 63.9
1995/96 55.2 29.4 51.0 53.6 5.7 59.3
2000 53.0 36.6 49.8 53.5 9.2 62.7

Note: The figures are based on the Household Expenditure Surveys of the BBS. The poor have been estimated using the
cost of basic needs (CBN) method and are taken as those living below the poverty line which corresponds to an
intake of 2,122 kcal/person/day and a nonfood allowance corresponding to nonfood expenditure among household
whose food expenditure equals the food poverty line. The number of the poor has been derived by the authors
using estimated population and its rural-urban distribution implicit in respective surveys.

Source: World Bank 1998, BBS 2001.

If we take the period of the 1980s as the pre-trade liberalization period, variations in the rate
of poverty reduction during the period and afterwards may be noted (Table 9). During the

[Link] 11
1980s, the incidence of poverty marginally increased due to increasing rural poverty despite
the decline in urban poverty at a rate of 1.4 per cent per year. In contrast, the 1990s
witnessed a decline in the incidence of total poverty at an annual rate of 1.9 per cent when
both rural and urban poverty declined at the rates of 1.7 per cent and 2.3 per cent per year
respectively. This shows that Bangladesh’s performance in reducing absolute poverty has
been better in the 1990s compared with earlier periods.

Table 9: Poverty Reduction Rates during Pre- and Post Liberalization Period
(Per cent per year)
1984-1992 1992-2000
Rural 0.33 -1.67
Urban -1.38 -2.31
Total 0.07 -1.91

Note: Calculated from figures in Table 6.

3.2 Changes in Inequality

The nature of impact of economic growth and other macroeconomic changes on poverty is
influenced by changes in the distribution of income and consumption. The favourable impact
of economic growth on income poverty is likely to be reduced if growth leads to increased
income inequality. In Bangladesh, inequality increased rather sharply during the early 1990s
which coincided with the period of rapid trade liberalization (Table 10). The Gini index of
consumption expenditure in both rural and urban areas remained largely unchanged till
1992. A similar trend may also be noted for income distribution in both rural and urban
areas. The urban Gini index for consumption expenditure, however, rose sharply to nearly 38
per cent in 1996 (from 32 per cent in 1992) in urban areas and marginally declined to 37 per
cent in 2000. In rural areas, inequality in consumption expenditure also increased. In the
case of income inequality, the trends were similar with a sharp increase in Gini index during
the mid-1990s. Moreover, income inequality is much higher than consumption inequality in
both rural and urban areas. One may also note that urban inequality increased more than
rural inequality along with widened disparity between rural and urban areas.

11 For an analysis of the implications of different methodologies on poverty estimates, see Ravallion 1990, Ravallion and
Sen 1996. The alternative poverty estimates highlight important issues of measurement of poverty, aggregation of numbers,
choice of calorie norms, and other dimensions. For a review of available estimates, see Hossain and Sen 1992, Mujeri 1999.

[Link] 12
Table 10: Changes in Growth and Inequality in Bangladesh

Poverty line Mean consumption Mean Gini index (%)

(Tk/person/month) (Tk/person/month) consumption/
Poverty line (%)
Consump- Income
tion
Urban
1983/84 301.72 396.53 131 29.8 37.0
1988/89 453.65 695.19 153 32.6 38.1
1991/92 534.99 817.12 153 31.9 39.8
1995/96 650.45 1,372.47 211 37.5 44.4
2000 724.56 1,291.53 178 36.6 45.2
Rural
1983/84 268.92 284.84 106 24.6 35.0
1988/89 379.08 435.39 115 26.5 36.8
1991/92 469.13 509.67 109 25.5 36.4
1995/96 541.77 661.47 122 27.5 38.4
2000 634.48 820.20 129 29.7 36.6
Source: BIDS 2001, BBS 1998, 2001.

3.3 Developments in the Labour Market

A significant aspect of the demographic developments in Bangladesh is the rapid growth of

the labour force. The labour force grew at a much faster rate than the growth in population
and the demand for labour.12 The trends in employment since the mid-1980s can be seen in
Table 11. The figures indicate that, between 1985/86 and 1995/96, total employment in the
country increased by around 10 million with an employment level between 40.3 million and
54.6 million in 1995/96 depending on the alternative statistical definition of employment.
During the period, the non-agricultural sectors were the main engine of job creation
contributing nearly 60 per cent of the additional employment. Moreover, the bulk of the
employment generation between 1989 and 1995/96 took place in the informal sector and
currently nearly 60 per cent of the urban employment and about two-thirds of the rural
employment outside agriculture are estimated to be in the informal sector.

12 During 1961 to 1991, total population of the country increased from 50.8 million to 111.5 million (that is, by
nearly 120 per cent) while the labour force grew from 16.9 million to 51.2 million (an increase of 203 per cent).
Similarly, between 1989 and 1995/96, the population increased by about 17 million and the net entrants into the
labour force was over 8 million.

[Link] 13
Table 11: Trends in Employment in Bangladesh
(in million)
1985/86 1989 1990/91 1995/96
A. Usual definition
Employed population 30.5 32.7 34.9 40.3
Male 27.4 29.4 30.4 33.2
Female 3.1 3.3 4.5 7.1
B. Extended definition
Employed population ... 50.1 50.2 54.6
Male ... 29.4 30.5 33.8
Female ... 20.7 19.7 20.8
Note: The usual definition refers to any person aged 10 and over employed (worked at least one hour in a week)
with/without pay or profit during the reference period excluding own household economic activities. The extended
definition uses a similar concept but includes some household economic activities e.g. care of poultry and
livestock; threshing, boiling, drying, processing and preservation of food, and similar other activities.

Source: BBS 1998.

Despite the overall increases in the level of employment, significant imbalances in the labour
market exist. The total civilian labour force increased from 50.7 million in 1989 to 56.0
million in 1995/96 (that is, by more than 10 per cent) while the number of employed persons
during the period rose by about 9 per cent – from 50.1 million to 54.6 million. This indicates
that the unemployment rate more than doubled over the period from 1.2 per cent to 2.5 per
cent.13 While ‘open’ unemployment is relatively low due to the dominance of the informal
activities in the labour market and the compulsion of the vast majority of the poor households
to earn subsistence for their survival, the problems of the labour market are manifested in the
high rate of underemployment (Table 12). The problem of under employment reflects the
fact that more than one-third of the employed persons work less than 35 hours a week, a low
level for a developing country such as Bangladesh. Moreover, the situation seems to have
deteriorated over time. During 1989, 43 per cent of the employed labour worked for less than
40 hours a week and the share increased to more than 49 per cent in 1995/96. This, combined
with the relatively low female participation rate in the labour force (in 1995/96, the female
participation rate for persons aged 10 and over as per the ‘usual’ definition of the labour force
was only 18.1 per cent compared with 77 per cent for the males), indicates that the labour
market in Bangladesh is characterized by the existence of significant ‘surplus labour’.14

13 The definition of unemployed persons used in the surveys is, however, somewhat unrealistic in the context of
Bangladesh since only persons in the civilian workforce not doing any work at all (even an hour) and engaged as
unpaid family helpers and working less than 15 hours during the reference week are treated as unemployed.
14 The labour market has several other disquieting features as well e.g. disproportionately high unemployment
rates for the youth, labour market discrimination against women, existence of child labour, and low education
and skill level of the labour force. During 1995/96, around 80 per cent of the female labour force were unpaid
family workers compared with 20 per cent for males, almost 20 per cent of the children (aged 5 to 9) were in the
labour force and two-thirds of the working children were engaged in agriculture, forestry and fishery activities
mostly as unpaid family helpers or day/casual workers, and 51 per cent of the labour force (aged 15 and over)
had no education and another 25 per cent had education only at the primary level. See BBS 1998.

[Link] 14
Table 12 Underemployment in Bangladesh, 1995/96
(Persons aged 15 and over)
Category Bangladesh Urban Rural
Absolute unemployed persons (‘000) 1,266 401 865
Unemployed persons
(Unpaid workers < 15 hrs/week; ‘000) 1,802 163 1,639
Underemployed persons (<35 hrs/week; ‘000) 18,903 1,942 16,961
Total unemployed and underemployed (‘000) 21,971 2,506 19,465
Underemployment rate (% of total labor force) 38.5 22.1 42.1
Male 13.7 10.9 14.5
Female 79.0 54.2 82.4
Unemployment and underemployment rate 39.2 24.6 42.5
(% of total labor force)
Source: BBS 1998.

Wage developments during the period indicate that real wages have grown at a moderate rate
of 2 per cent per year since the early 1990s. This has been lower than the productivity gains,
estimated at around 2.5 to 3 per cent since the beginning of the trade liberalization. This
tends to indicate that the share of wages in total national income may have declined over time
in Bangladesh although no firm data on distribution of income by factors of production are
available.

3.4 Trade Liberalization and Labour Market Developments

As we have indicated earlier, the growth performance of the Bangladesh economy indicates
some acceleration in GDP growth in the 1990s which coincided with the post-liberalization
period. Between 1973 and 1980, GDP grew by 2.8 per cent per year on an average which
increased to 3.8 per cent in the 1980s. By contrast, GDP increased by an average of 4.8 per
cent per year in the 1990s. The growth in real per capita income also accelerated during the
1990s. The per capita GDP in real terms increased by only 16 per cent between 1980 and
1990 but the increase was 34 per cent between 1990 and 2000. Another significant feature of
the 1990s is the rapid increase in export earnings. While export earnings represented around
one-third of total imports in 1979/80 and 40 per cent in 1989/90, the share was nearly 75 per
cent in 1998.

Although it is difficult to assess the productivity trends in the economy due to lack of
comparable data over the period, the available information indicates that the productivity
growth has been insignificant during the first half of the 1980s. During the period, GDP grew
by 3.1 per cent per year while employment grew by 3.3 per cent. During 1989 to 1995/96,

[Link] 15
GDP increased by over 4 per cent per year compared with the employment growth of about 3
per cent indicating a small productivity gain. It seems likely, therefore, that the period of
trade liberalization has witnessed some modest productivity gains.

The impact of trade liberalization policies on the labour market is difficult to infer since the
outcomes are not straight forward in a country such as Bangladesh. Although the standard
analysis in a labour abundant country like Bangladesh suggests that the demand for labour,
especially in the formal sector, should increase as the trade restrictions are removed, the
actual outcomes depend on the structure of the labour market.15 Moreover, the skill
characteristics and segmentation in the labour market are important elements in determining
who benefits from the changes in the labour market. For example, a necessary pre-requisite
for the poor labourers to benefit is that the production of tradable goods uses the unskilled
workers as the most intensively used factor so that a positive impact on unskilled wages is
created. Otherwise, wages of skilled (or semi-skilled) workers will increase with trade
liberalization with those of unskilled workers remaining unchanged. In Bangladesh, the
changes in agricultural wages are significant since unskilled workers in the rural areas form
the largest majority of the poor in the country.

If we assume that the technology and other factors have a fairly constant impact on changes
in employment and wages over the years, then an analysis of the trends in employment and
wages can reveal some impact of trade liberalization in the labour market. We have
compared the changes for two periods – late-1980s (1986-1990) and early-1990s (1991-1996)
using available data and the results are given in Table 13. It shows that the period of the
early-1990s, which was associated with rapid trade liberalization, was characterized by
significant deceleration in the rates of employment creation in both agriculture and
manufacturing sectors. The rate of increase in real wages also decelerated. During the
period, agricultural growth declined to 2.2 per cent per year (compared with 2.8 per cent
during the late-1980s) while manufacturing GDP grew by an annual rate of 9.5 per cent. The
above results suggest that the period of the early-1990s witnessed significant adjustments in
the labour market, particularly in its formal segment, whereby growth in employment and
real wages slowed down. In terms of growth, the urban economy seems to have gained from
trade liberalization through higher growth in manufacturing output.

[Link] 16
Table 13: Employment and Wage Rate Changes in Agriculture and Non-Agriculture Sectors in Bangladesh
(Average annual growth rate in per cent)
1986-1991 1991-1996
GDP Employment Real wage GDP Employment Real wage
Agriculture 2.8 2.2 3.2 2.2 0.8 1.9
Manufacturing 6.5 19.2 2.4 9.5 -6.2 1.6
Total 3.7 2.8 2.5 4.4 3.1 1.3
Source: BBS 1999, MOF 2001.

An important issue, moreover, is to see how the gains (in terms of income) have been shared.
While the issue is the focus of analysis of the present study and will be further examined later
on using a specifically adopted analytical framework, we present here some empirical
evidence on how the incomes have been shared both spatially and by broad income groups.
The data, presented in Table 14, show that the average per capita income in real terms
increased by more than 40 per cent between 1986 and 2001.16 However, the per capita
income of the poorest 20 per cent of the households increased by only 24 per cent while the
per capita income of the richest 20 per cent increased by nearly 60 per cent during the period.
The percentage increase during the 1990s, moreover, is much higher for the 20 per cent
richest households (49 per cent) compared with only 21 per cent for the 20 per cent poorest
households. Similar increases during the earlier period (1986-1992) were 7 per cent for the
richest 20 per cent households and less than 3 per cent for the poorest 20 per cent households.
This indicates that the rich households gained relatively more during the period of
liberalization relative to the poor households. The period also witnessed widening
inequalities in the regional income distribution. Urban pockets of development e.g. Dhaka,
Chittagong and Khulna enjoyed relatively better living standards (as measured by per capita
GDP) during the 1990s compared with the rest of the country (Table 15). On the other hand,
growth in several rural regions stagnated. Regional disparities, as measured by the
coefficient of variation of regional per capita income, also increased.

15 For example, if the elasticity of labour supply is zero, wages will increase but not employment whereas, if the
elasticity is infinite, employment will increase but not wages.
16 The estimates are based on Household Expenditure Survey (HES) data on income distribution and income
data of the national accounts. For example, the income of the poorest 20 per cent of the households is estimated
as the product of national income times the share of these households as given in the HES income distribution.
The per capita income is then derived by using the one-fifth of the total number of households and average
household size. One limitation of the methodology is the neglect of the movements of the households across the
income categories over different periods. Despite this caveat, the results show important aspects of income
inequality.

[Link] 17
Table 14: Distribution and Changes in Real Per Capita Income
(Taka at 1995/96 constant prices)
Year Per capita income Ratio of the poorest 20% and
Average 20 per cent richest 20 per cent poorest the richest 20%
households households
1985/86 11,199 25,780 3,914 0.15
1988/89 11,474 26,504 3,809 0.14
1991/92 12,286 27,564 4,005 0.15
1995/96 13,788 34,109 3,889 0.11
2000/01 15,788 41,104 4,855 0.12
Source: Author’s calculations based on Household Expenditure Survey and national accounts data.

Table 15: Regional Variation in Per Capita GDP

(Country average in 1989/90 = 100 at constant 1984/85 prices)

Region 1989/90 1998/99 Total variation
Dhaka 101.3 143.1 41.4
Mymensingh 84.0 107.1 23.1
Jamalpur 95.6 120.5 24.9
Tangail 91.2 114.5 23.3
Faridpur 98.2 122.1 23.9
Chittagong 151.9 202.4 50.5
Chittagong Hill Tracts 297.1 398.6 101.5
Noakhali 84.4 101.4 17.0
Comilla 93.6 119.1 25.5
Sylhet 98.6 127.3 28.7
Rajshahi 83.7 103.0 19.3
Dinajpur 96.4 115.2 18.8
Rangpur 84.4 103.1 18.7
Bogra 99.2 118.6 19.4
Pabna 73.0 93.3 20.3
Khulna 111.3 144.7 33.4
Barisal 105.2 120.0 14.8
Patuakhali 110.8 135.4 24.6
Jessore 97.3 125.2 27.9
Kushtia 98.0 120.3 22.9
Country average 100.0 131.0 31.0
Coefficient of Variation (per cent) 47.1 48.1 -
Source: BBS 1999.

The empirical evidence presented in this section suggests that, despite some positive
developments, the gains associated with trade liberalization have been unevenly shared, both
among various income groups and over different geographical regions, of the country. The
critical policy issue for Bangladesh, therefore, is to enhance the overall gains from trade
liberalization along with improving the access of all social groups to the benefits of
globalization and growth. In the following sections, we shall examine the issues in terms of
the general equilibrium framework developed under the study.

[Link] 18
4. Objectives of the General Equilibrium Framework

The major objective of the framework is to examine and analyze the poverty and distribution
impacts of the measures adopted by Bangladesh to integrate its economy to the rest of the
world. More specifically, the framework intends to analyze the consequences of
“globalization” measures on household poverty and income distribution. Understanding the
impact of globalization measures is important for Bangladesh since it would help the
policymakers in formulating and implementing countervailing measures that would offset, or
at least reduce, the deleterious impact of the globalization measures on the poor households.
Using the framework, the consequences of several measures (e.g. trade liberalization and the
inflow of capital) on allocation of resources, distribution of income, and the poverty situation
of different household groups have been examined. For the purpose, simulation exercises
were conducted using the multi-sectoral, multi-factor and multi-households computable
general equilibrium (CGE) model calibrated to the 1995/96 social accounting matrix (SAM)
database of the Bangladesh economy.

5. Outline of the Methodology

The general methodology uses a framework of analysis which allows to examine the
consequences of policy changes both at sectoral and macro levels and to estimate their
poverty and distribution impacts at the household level. For the purpose, a computable
general equilibrium (CGE) model has been employed which allows to examine the
consequences of policy reforms within a constrained optimization framework. A Social
Accounting Matrix (SAM) for the year 1995/96 has been developed to serve as the consistent
and comprehensive database for the above-mentioned exercise.

5.1 The Bangladesh Social Accounting Matrix for 1995/96

As mentioned above, the CGE model has been numerically calibrated to a 1995/96 Social
Accounting Matrix (SAM) for which the main sources of information are: (a) 1993/94 Input-
Output Table prepared by the Bangladesh Institute of Development Studies (BIDS 1998); (b)
Household Expenditure Survey (HES) 1995/96 by the Bangladesh Bureau of Statistics (BBS
1998); (c) 1995/96 Labour Force Survey (LFS) by the Bangladesh Bureau of Statistics (BBS

[Link] 19
1998); and (d) National Income Estimates by the Bangladesh Bureau of Statistics (BBS).
The major features of the SAM may be summarized as follows:

Accounts
The 1995/96 SAM identifies the economic relations in the economy through four types of
accounts: (i) production activity accounts for 26 sectors; (ii) 7 factors of production with 6
different types of labour and one type of capital; (iii) current account transactions between 3
main institutional agents: households and unincorporated capital, the government, and the
rest of the world; and (iv) one consolidated capital accounts to capture the flows of savings
and investment by institutions and sectors respectively.17

Activity
The activity account has been represented by 26 producing activities. These are derived from
the 79 sectors of the 1993/94 input-output table. Due to lack of adequate information, no
distinction has been made between activity and commodity and hence they are synonymous
in the SAM.

Households
An important feature of the SAM is the decomposition of the households into 7 groups. The
household groups differ with respect to employment status, income levels and expenditure
patterns. Pyatt and Thorbecke (1976) have suggested location, sociological and wealth
criteria to classify the household groups18. In our case, location (rural-urban), land
ownership, occupational status, and the level of education information, contained in the
1995/96 HES, has been used for household classification.

Labour Factor
The 1995/96 SAM also accounts for decomposition of the labour factor into 6 groups based
on gender and skill level of the workers. The labour factor classification is important to

17 The details of the SAM are provided at Appendix 2.

18 For instance, the location criterion which distinguishes a household as urban or rural is useful since it
captures many aspects of duality. Depending on this distinction, individuals with otherwise similar
characteristics are likely to be paid different wages, have different job opportunities and employment
expectations and generally be subject to different sets of parameters in their socio-economic behaviour (Pyatt et
al, 1984).

[Link] 20
examine the consequences of policy measures on “factorial” income distribution. The
information on the level of education and gender, contained in the 1995/96 LFS, has been
used for labour factor classification.

The disaggregation of factors, households, activities and institutions in the SAM and the CGE
model is given in Table 16.

Table 16: Disaggregation and Description of Factors, Institutions and Households

Set Description of Elements
Factors of Production
• Female: 3 categories according to skill levels (low, medium and high)
Labour (6) Low: grades 0-5; Medium: grades 6-10; High: grades 11 and above
• Male: 3 categories according to skill levels (low, medium and high)
Low: grades 0-5; Medium: grades 6-10; High: grades 11 and above
Capital (1) • 1 type only
Institutions
• Rural Agriculture: 3 categories according to land ownership
Households (7) Labourer household: 0-0.49 hectares; Small Farmers: 0.5-2.49 hectares,
Large Farmers: >2.5 hectares.
• Rural Non-Farm: 1 category according to occupation
• Urban: 3 categories according to the level of education of the household’s head
Low Skilled: grades 0-5; Medium Skilled: grades 6-10;
and Professional: grades 11 and above

Others (2) • Government

• Rest of the World
Activities
• Crops Non-traded: Rice (Aman and Boro)
Agriculture (7) • Crops Traded: Other Grains and Commercial Crops
• Non-crops Non-traded: Forestry
• Non-crops Traded: Livestock and Fish

• Food Processing Traded: Rice Milling, Atta and Flour, Other Food and Tobacco
Industries (12) • Textiles Traded: Clothing, Ready Made Garments and Leather.
• Others Traded: Chemicals, Fertilizer, Petroleum Products, Machinery and Miscellaneous
Industries

Services (6) • Non-Traded: Construction, Gas, Trade Services, Social Services, Public Administration,
Financial Services and Other Services

5.2 The CGE Model

The computable general equilibrium (CGE) models capture the detailed accounts of the
circular flows of receipts and outlays in an economy. It satisfies the general equilibrium
conditions in the markets simultaneously. Given the framework, such models are useful to
analyze the associations among various agents of the economy.

[Link] 21
In line with most of the CGE models, the present model has been solved in the comparative
static mode which provides an instrument for controlled policy simulations and experiments.
The solution of each simulation presents the complete sets of socio-economic, meso and
macro level indicators such as activity/commodity prices, household incomes and
expenditures, factor demand and supplies, gross domestic products, exports and imports, and
household poverty situation. To begin with, the model was calibrated to the 1995/96 SAM to
exactly reproduce the base year values19.

The structure of the model and its main features are discussed below. The schematic
presentation of the production structure and the structure of demand are shown in Figure 1
and Figure 2 respectively.

The Production Structure

The nested production structure in each sector is presented in Figure 1. At the top level, the
real value added and the intermediate inputs are combined via a Constant Elasticity of
Substitution (CES) production function to produce the gross output. At the bottom level,
there are two CES functions: one for labour and capital factors to produce the real value
added and one for imported and domestic intermediates to generate composite intermediate
inputs.

Figure 1: Structure of Production

Gross Output

Value Added CES Intermediate Input

CES CES

Capital (1) Labour (6) Domestic Imported

CES

Male (Skilled, Semi-skilled, Unskilled) Female (Skilled, Semi-skilled, Unskilled)

19 In the calibration procedure, most of the model parameters are estimated endogenously keeping the various
elasticity values fixed.

[Link] 22
The Demand Structure
The structure of demand is presented in Figure 2. It shows the demand for private and public
consumption expenditures, investment demand and exports demand. The private
consumption demand is specified by a Cobb-Douglas function which is combined with a
nested CES function of composite products. The distribution of investment by sector is
modeled using a fixed-coefficient specification. The Leontief specification applies to both
domestically produced and imported investments. The formulation of investment function is
static: there is no link between increased savings at the current period and additional
investment in a subsequent time period. In a dynamic model, a policy which has a negative
impact on welfare in the current period, may yield substantial welfare gains in the long run.
These inter-temporal features, however, have not been explicitly considered in the present
structure of the model. The total government expenditure has been taken to be exogenous.
The distribution of government expenditure by sectors has been modeled using a fixed-
coefficient specification. The export demand is specified by a downward sloping world
demand for exports.

Figure 2: Structure of Final Demand

Structure of Final Demand

Final Demand

Consumption Investment Government Export

CD
Leontief Leontief Downward
Sloping

26 Composite 26 Composite 26 Composite

3 Products
Products Products Products

CES
CES

(1) As in (1) Domestic

Imported Domestic

[Link] 23
System Constraints and Equilibrium Conditions
There exist four constraints in the system of the specified model. The real constraint refers to
domestic commodity and factor market; whereas the nominal constraint represents two macro
balances: the current account balance of the rest of the world and the savings-investment
balance.

The sectoral supply in the model is a composite of imports and outputs sold in the domestic
market. The composite demand, on the other hand, includes final demands (i.e. private and
public consumption expenditure and investment) and intermediate input demands. The
variations in the sectoral prices assure equilibrium between sectoral supply and demand in the
model.

In the case of the factor market, it is generally assumed that total quantities of factor supply
are fixed and hence variations in factor returns (i.e. wages and rents) ensure the equilibrium
between the factor demand and the fixed supply. This specification implies the existence of
full mobility of the factors across the producing activities specified in the model. However,
given the comparative static and short-run nature of our analysis, the full mobility
specification is adopted for the six types of labour factors where variations in their wages
assure the equilibrium in the labour market. However, capital is not treated as mobile rather it
is taken as sector specific and hence the capital market equilibrates through explicit
parameters that allow for differential rents for different sectors.

The inflows (transfers to and from domestic institutions) are specified as fixed but imports
and exports are determined endogenously in the model. The foreign savings is also taken as
fixed in this model and nominal exchange rate is allowed to vary to clear the foreign
exchange market. In this case, the equilibrating variable is the nominal exchange rate. Under
the specification, fixing of foreign savings is equivalent to keeping the trade deficit fixed.

Finally, for the savings-investment equilibrium, the model treats the investment decision as
given and hence savings adjust to ensure its equality to the fixed value of the investment. The

[Link] 24
basic approach is to allow the savings propensity of one of the domestic institution to vary.
The main features of the model are summarized in Table 17.20

Table 17: Summary of Model Features

• Labour factor is mobile across producing activities.

• Capital is immobile and sector specific.
• Primary factor supplies are exogenous and fixed.

• The world prices of imports and exports are exogenous invoking the small country assumption.

• Current account balance or deficit is fixed.

• Imports and domestically produced goods are imperfect substitutes.
• Output produced for domestic and export markets reflects differences in quality.

• Savings of domestic institution adjust to equate to given investment.

• General price index acts as the numeraire.
• Excess demand conditions are satisfied.

6. Simulation Designs

For policy analysis, two simulations have been conducted to examine the impacts of
globalization measures on poverty and income distribution of the seven representative
household groups. These simulation designs are done in line with the measures of
globalization adopted in Bangladesh and discussed in section 2 of the study.

Simulation 1 (Sim1): In the first simulation, the base values of the tariff rate are set equal to
zero to encourage the volume to trade to expand such that the “openness” indicator of
globalization is enhanced. Consequently, the base values of all other parameters are retained.
The base and simulation values of the tariff rates are presented in Table 18.

20 For a summary of the equations of the model, see Appendix 1.

[Link] 25
Table 18: Tariff Rates for Base Year and Simulation Experiment
Base Year Simulation 1
Import values Tariff Revenues Tariff Rates Tariff Rates
Grains 4.21 0.69 16.51 0.00
Commercial Crops 7.49 0.54 7.18 0.00
Livestock 2.34 0.00 0.00 0.00
Fish 0.00 0.00 0.00 0.00
Forestry 0.00 0.00 0.00 0.00
Rice Milling 0.57 0.02 3.43 0.00
Ata and Flour Mill 0.04 0.00 12.16 0.00
Other Food 8.39 3.14 37.42 0.00
Leather 0.28 0.00 0.00 0.00
Cloth 22.90 5.59 24.42 0.00
Ready Made Garment 1.37 0.04 2.96 0.00
Tobacco 0.11 0.002 2.04 0.00
Chemical 24.01 5.04 20.98 0.00
Fertilizer 2.84 0.00 0.08 0.00
Petroleum Products 9.78 4.61 47.15 0.00
Machinery 70.98 12.69 17.88 0.00
Miscellaneous Industry 99.01 6.62 6.69 0.00
Average 254.33 39.00 15.33 0.00

Simulation 2 (Sim2): In the second simulation, the base value of foreign savings is
augmented to reflect the pattern of foreign investment inflow into Bangladesh during the last
few years. The inflows of foreign investment are concentrated mainly in the “gas” sector
depicting the rise in “foreign investment to GDP ratio” (e.g. investment indicator of
globalization) as well as to assess impacts of such investment. As a result of such inflow of
capital there is usually a primary resource boom in the country where the resource is
effectively in an enclave. The direct outcome of this is the repatriation of export earnings
leading to rise of domestic prices relative to world prices and the contraction of tradable
sector relative to non-tradable sector. The base values of all other parameters are retained.
The base and simulation values of the foreign savings are presented in Figure 3.

Figure 3: Ratio of Foreign Savings to GDP

Foreign Savings

6.0
5.0
4.0
3.0
2.0
1.0
0.0
Base Sim2

[Link] 26
7. Simulation Results

The major results of the simulations are discussed in this section. For this, the simulation
outcomes have been presented under three main headings: impacts on selected macro
variables; welfare effects; and poverty implications.

Impacts on Macro Indicators:

The impacts on selected macro indicators are reported in Table 19. The real GDP growth in
simulation one is 2.43 percent compared with the base case. The complete elimination of the
tariffs (which were mainly concentrated on a few manufacturing sectors) led the resources to
move from the protected (i.e. manufacturing) to the unprotected sectors (i.e. agriculture and
services). Such a reallocation of resources to the unprotected sectors resulted in higher
growth of agriculture (0.73 per cent) and service (0.84 per cent). The experience of the
manufacturing sector is mixed with respect to growth and allocation of resources. Resources
have moved from some manufacturing sectors and these are the major losers of the tariff
elimination experiment. Such sectors belong to the “import-substituting” type of
manufacturing activities. On the other hand, the elimination of tariff rates depreciated the
nominal exchange rate , which helped to increase Bangladesh’s exports. Thus, resources also
moved to the export oriented manufacturing sectors and led to their growth. The net effect on
the manufacturing activity is the growth of the sector by 0.86 percent compared with the base
case.

Table 19: Selected Macro Effects of the Simulations with the Model
Shares (%) Growth Rates (%)
Base Case Simulation 1 Simulation 2
Real GDP 2.43 0.21
Agriculture 0.22 0.73 -0.004
Manufacturing 0.22 0.86 -0.93
Service 0.56 0.84 1.14

Traded 0.33 1.40 -1.13

Non-traded 0.67 1.03 1.33
Consumption 1.38 0.95

Imports 4.23 12.89

Exports 14.77 -11.19
Trade Openness 29.54 28.62

[Link] 27
The observed pattern of manufacturing sector growth is reflected in high growth of the export
sector by nearly 15 percent compared to the base case. The growth of imports by 4 per cent is
moderate considering the full elimination of tariffs. Substantial depreciation of nominal
exchange rates countered the large fall in the domestic import prices. These two opposing
impacts on the domestic import price resulted in the moderate increase in import by about 4
per cent.

The patterns of growth effects under the second simulation are different from the first
simulation. In the second simulation the resources moved from both agriculture and
manufacturing sectors to generate growth in the service sector. As expected the pattern of
resource reallocation resulted in the growth of non-traded sectors (1.33 per cent) at the
expense of the traded sectors (-1.13 per cent).

The growth of imports has been relatively high (13 per cent) in the second simulation. The
decline of domestic manufacturing and agriculture activities manifested in higher prices of
domestic products relative to the import price of their import substitutes. This led to the
substantial growth of imports in the simulation. Similarly, higher prices of domestic supplies
compared to the export prices manifested in sharp decline of exports (11 per cent) in this case
compared to the base case.

Welfare Effects:
The concept of efficiency or welfare is the starting point of any policy analysis. Unlike in a
theoretical approach where an ordinal measure of alternative states is examined, some
measures of welfare are employed in applied policy analysis to compare the movement from
one state to another.

In applied policy analysis, this is done through using some monetary representation of the
individual utility functions which is defined as the amount of money required to attain a level
of utility at a reference price vector. This is termed as the money metric, and its value is
derived from the expenditure function. The expenditure function, which is the inverse of the
indirect utility function, is a vital tool for such an welfare analysis which allows the
“measurement of utility”. Since the value of the expenditure function depends on the set of
prices used, there are different money metrics which one can use. The most widely used ones
are, however, the compensating variation (CV) and the equivalent variation (EV). These are

[Link] 28
commonly used due to their easy interpretation in terms of the compensated demand curves.
In the EV approach, the idea is to measure in money terms, how much income needs to be
given to the consumer at the “pre-policy change” level of prices ( P0 ) in order to enable him
to enjoy the utility level which arises after the policy change is effected (“post-policy change
level of utility”). The CV, on the other hand, uses the post-policy prices ( P1 ). It thus
measures the income change necessary to compensate the consumer for the changes in
prices21. In the present exercise, the Equivalent Variation (EV) has been used as a measure
of welfare to examine welfare impacts of the simulations. The results are given in Table 20.

Table 20: Welfare Impacts of Simulations: Equivalent Variations for Different Household Groups
Base value Simulation 1 Simulation 2
Consumption Consumption Consumption
Household Groups (Billion Tk) Growth (%) EV Growth (%) EV
Agricultural Labourers 95.59 1.25 1.19 0.75 0.71
Small Farmers 176.25 1.36 2.54 1.06 1.97
Large Farmers 188.63 1.35 2.93 1.45 3.13
Non-farms 268.77 1.33 3.9 0.91 2.65
Worker-Low Skilled 168.94 1.41 2.47 0.89 1.56
Worker-Medium Skilled 151.75 1.46 3.03 0.89 1.84
Professionals 329.07 1.35 5.57 0.76 3.10
Total 1379.00 1.36 ... 0.95 ...

It can be observed that, in both simulations, Equivalent Variations (EVs) are positive for all
household groups. The positive EV values are the manifestation of positive real GDP growth
and consumption growth. Except for the non-farm household group, the observed EV is
larger for the relatively high-income household groups (e.g. professionals, medium skilled
workers, and large farmers) compared with the low-income household groups (agricultural
labourers, low-skilled workers and small farmers). This suggests that the welfare gains
emanating from the “globalization” measures accrued more to the well-off household groups
compared to their less well-off counterparts. Among the less well-off household groups, only
the non-farm group is observed to benefit due to their higher participation in non-traded and
service activities which exhibit high growth under the simulations.

21 In a many consumer economy, the use of aggregate EV or CV as a measure of welfare changes, although avoids any
explicit Social Welfare Function (SWF), has an implicit SWF because of the adding up approach. Boadway and Bruce
(1984) show that there are some well-known problems in interpreting the aggregate EVs or CVs and one needs to be careful
in interpreting the result of such measures. The social ordering requires more data and judgment than do household ordering
and it may not be possible to measure changes in welfare simply on the basis of household orderings of social status drawn
from their market behaviour. When EV is used as a measure of welfare, it is implicitly assumed that aggregate market
behaviour is generated by a single household whose preferences coincide with the social ordering.

[Link] 29
Poverty Implications:
In the present exercise, Foster-Greer-Thorbecke (FGT) measure of poverty has been used to
evaluate the policy effects on poverty profiles of the representative household groups. The
measurement of poverty profiles has been done following the method adopted by Decaluwe
et al (1999). Specifically, the methodology requires: (a) explicit proposition of income
distribution formulation corresponding to each household group’s characteristics and (b)
postulation of an unique and constant basket of basic needs based poverty line whose
monetary value is altered by endogenously determined commodity prices. The derivation of
poverty profiles of the representative household groups in the present study follows the above
methodology for which the major steps are summarized below:

1. The income distribution formulation under the approach depends on the “minimum” and
the “maximum” incomes and on the skewness of the distribution. The “Beta” distribution
function (equation 1) has been used to represent these characteristics of the household
groups. The implementation of “Beta” distribution requires minimum (mny) and
maximum (mxy) incomes within each of the seven groups and values of shape and
skewness parameters (i.e. p and q) of the distribution..
h h
−1 −1
h h h h 1 ( y h − mny h ) p ⋅ (mxy h − y h ) q
I (y , p ,q ) = ⋅
Bh ( ph, qh )
h
+ q h −1
(mxy h − mny h ) p

h h
−1 −1
h h h mx h ( y h − mny h ) p ⋅ (mxy h − y h ) q
B (p ,q ) = ∫
h
dy (1)
mn h + q h −1
(mxy h − mny h ) p

2. The 1995/96 Household Expenditure Survey (HES) of the Bangladesh Bureau of

Statistics (BBS), which is a representative survey for the country as a whole, was used to
derive the household classification scheme and subsequent derivation of the values (p,q)
and the values of the minimum and maximum incomes for each household group. In the
survey, these data were recorded in nine sets that are referred to as cards. The layout is
given in Table 21.

[Link] 30
Table 21: Data Layout in the 1995/96Household Expenditure Survey
Attribute Card Sample of 7,420 households with 39,051 members

Household 1 Sex, relationship, age, marital status, work status, occupation, activity, activity code, industry
Structure code;
2 Land property, housing, sanitation, electricity, water supply, occupational status;

Expenditure 3 Permanent and temporary expenses (fuel & light, gas, washing & cleaning, communication
& travel, miscellaneous items);
4 Other monthly expenses (house rent, educational and medical expenses and other
miscellaneous expenses);
9 Daily expenses on food items

Income 5 Agricultural and related activities

7 Other sources of income

Community 8 Demographic characteristics, agricultural practices, quality of social and physical

infrastructure, availability of schools and health facilities, access to various amenities and
development programmes and activities by the Government and NGOs

Excluding the missing records, the total sample size was 39,044 of which 26,446 belonged to
the rural location and 12,598 to the urban location.22 For computing the above parameters,
the data records in the HES were grouped. The ‘betafit’ function was applied in the
‘MATLAB’ environment to derive p and q values for the household groups. The steps
followed are outlined below:

Sorting each household group in ascending order of income.

Derive maximum and minimum income for each household group.

Derive mean income for each decile group within each household category.23

The following formula was then applied to each decile group of a particular
household category:
(meani-min) / (max-min)
Where, meani = mean income of the ith decile of the household group.
min = minimum income of the household group.
max = maximum income of the household group.
Thereby, we got a vector (x) consisting of 10 values corresponding to each decile
group of a particular household.

The vector ‘x’ was created in MATLAB environment.

22 We used the exhaustive method. For the rural sample, 7 out of 26,453 values were found to be missing while
classifying them as per our classification criteria. So, we excluded them from the sample. The number, however,
is too small to affect our estimates.

23 In this case we used 10 decile groups for each representative household group.

[Link] 31

The command “phat=betafit (x)” was applied to generate corresponding ‘p’ and ‘q’
values for each household group.

The ‘p’ and ‘q’ values were then used in the integral formula of poverty estimation in
the ‘Mathematica’ environment for poverty estimates.

The reported minimum and maximum incomes and the estimated values of p and q
parameters are reported in Table 22.

3. The derived distribution has been employed to assess the poverty implication within each
of the household groups. For our purpose, it has been assumed that, following a policy
change, intra-group distributions shift proportionally due to mean income change
implying the constancy of intra-group distributions. That is, if the mean income changes
by a factor k, the income of each group within each household groups is also altered by
the same factor. Analogously, the minimum and maximum incomes of each household
group will also alter. The income effects of the simulations are provided in Table 23.

4. In order to derive the poverty profiles, the per capita income of each household group has
been contrasted with the poverty line. For the purpose, two poverty lines applicable for
rural and urban locations have been defined to capture price and other characteristics. The
poverty lines (z in equation 3) have been determined endogenously within the CGE
model by a basket of quantities of commodities reflecting the basic needs (BN).
Although, the basket ( ω il ) remains invariant under different simulations, the commodity
price ( Pi ) changes to alter the monetary values of the poverty lines. A rise in commodity
prices shifts the poverty line to the right (compared to the base case) and vice versa.

l l l
Monetary Poverty Line: z = ∑ ω i ⋅ Pi (2)
i

5. The above estimates (Beta distributions and poverty lines) have been used in the FGT
poverty measure to derive pre- and post-simulations poverty incidence for the 7
representative household groups. The FGT class of measures satisfies the desirable
axioms and allows us to measure poverty incidence for different groups that adds up to
the total.24 The FGT index ( Pα ) also allows us to generate three measures of poverty:

24 Any poverty measure is expected to satisfy the following three desirable axioms. (1) Focus axiom, which
requires poverty measures to be insensitive to increase in income of a non-poor person. (2) Monotonocity axiom

[Link] 32
 Head Count Index (when α = 0 ); Poverty Gap Index (when α = 1 ) and Squared Poverty
Gap Index (when α = 2 ). The simplest measure of the incidence of poverty, headcount
ratio, is the proportion of population with a per capita income below the poverty line. The
depth of poverty is measured by the poverty gap index, which estimates the average
distance separating the income of the poor from the poverty line as a proportion of the
income indicated by the poverty line. The severity of poverty, as measured by the squared
poverty gap index, quantifies the aversion of the society towards poverty. This implies an
increase in “our measured poverty due to a fall in the standard of living will be greater the
poorer you are” (Ravallion, 1994). These three measures for rural and urban household
groups can be computed using the following formula:
zl  zl − yh  h h h h
Pαh = ∫   ⋅ I ( y , p , q )dy h (3)
 zl
mny h 


where,
l ∈ {rural, urban} refers to location;
h∈ {1,2, …, 7} refers to the 7 households groups;

Pαh is the FGT index by household groups;

Base Year Poverty Profiles:

The base year poverty profiles of the household groups and in rural and urban locations are
provided in Table 22. The main features may be summarized as follows:
(i) Almost 54 percent of the rural population are poor while, in urban areas, it is
around 29 percent. This suggests that the incidence of poverty in rural areas is
much higher than that in urban areas. Moreover, the depth and severity of
poverty, as measured by the poverty gap and the squared poverty gap indexes,
are observed to be worse in rural areas.
(ii) Among the rural households, the agricultural labourers form the most deprived
group. More than 78 per cent of them are poor. In terms of poverty gap and
squared poverty gap, they also constitute the most vulnerable group. The
group is closely followed by the small farmers and the non-farm households in
terms of incidence of poverty.

which refers to the condition where a reduction in a poor person’s income should increase the value of the
poverty measure; and (3) Transfer axiom, which demands that, ceteris paribus, a transfer of income from a poor
person to a richer person should raise the value of the poverty index. For details see Subramanian (1997).

[Link] 33
 (iii) A significant share of the population of the large farmers group lives below
the poverty line as indicated by the head count ratio of the group (0.293).
However, the values of the poverty gap (0.097) and the squared poverty gap
(0.047) suggest that most of the poor in the group are marginally poor and a
relatively small increase in their income would graduate a significant portion
of them out of poverty.
(iv) As expected, the incidence of urban poverty is concentrated mostly among the
low skilled workers. More than 37 percent of the low skilled workers have
income less than the urban poverty line. Moreover, relatively high values of
the poverty gap and the squared poverty gap (0.14 and 0.07 respectively) of
the group indicate higher vulnerability of these workers compared with other
urban groups. The incidence of poverty is low for other two urban household
groups.

Table 22: Base Values of Household Poverty Profiles

Income (Tk per capita per month) Population Beta FGT Poverty Measure
p q Squared
Poverty Share Head Poverty Poverty
Household Minimum Maximum Mean line (%) Count Gap Gap
Rural 18 9140 697 650 78.65 2.9 37 0.535 0.197 0.099
Agricultural Labourers 73 4245 507 650 29.63 2.9 26 0.781 0.305 0.153
Small Farmers 152 6369 694 650 21.65 2.3 24 0.523 0.164 0.070
Large Farmers 18 9140 981 650 11.32 2.7 22 0.293 0.097 0.047
Non-farms 91 6935 721 650 37.41 2.3 22 0.486 0.168 0.079
Urban 73 26533 1359 725 21.35 1.7 33 0.287 0.109 0.057
Workers-Low Skilled 73 16376 987 725 58.60 2.3 38 0.377 0.136 0.066
Workers-Medium
441 14833 1884 725 21.92 1.3 11 0.107 0.019 0.005
Skilled
Professionals 358 26533 2927 725 19.48 1.4 12 0.062 0.013 0.004
National 30 12854 838 665 100 2.0 56 0.482 0.178 0.089

Post Simulation Poverty Profiles:

Among others, the incomes of the representative household groups and the commodity prices
are altered as a result of the policy changes under the simulations. The changes in incomes
and prices also change the minimum and maximum incomes within each household group
and the monetary values of the rural and urban poverty lines. The estimated post simulation
values of the minimum and maximum incomes and the poverty lines are given in Table 23.

[Link] 34
Table 23: Income Effects Under Alternative Simulations
Simulation 1 Simulation 2
Household Groups Minimum Maximum Mean Poverty line Minimum Maximum Mean Poverty line
Rural 23 11,614 886 806 20 10,242 780 728
Agricultural Labourers 93 5,404 646 806 82 4,770 570 728
Small Farmers 193 8,100 882 806 170 7,141 778 728
Large Farmers 23 11,614 1,247 806 20 10,242 1,099 728
Non-farms 116 8,831 918 806 102 7,763 807 728
Urban 93 33,826 1,734 899 82 29,595 1,516 812
Workers-Low Skilled 93 20,837 1,256 899 82 18,360 1,107 812
Workers-Medium Skilled 565 18,990 2,412 899 490 16,474 2,092 812
Professionals 456 33,826 3,731 899 399 29,595 3,265 812
National 38 16372 1067 826 33 14374 937 746

These estimated values of income and the new prices generated under the simulations have
been used in the FGT index (equation 3) to derive the post simulation poverty profiles. The
poverty profiles under the two simulations are presented in Table 24 from which the
following major impacts of the changes may be identified:
(i) Due to relatively high growth of income in the first simulation, poverty status
of all household groups has improved. The gain, however, is marginally higher
for the urban households compared with the households who reside in the rural
location. The highest gains, in terms of reduction of poverty, are observed for
the relatively well-off household groups. The highest reduction in poverty is
observed for the medium skilled households (10 per cent), followed by the
professionals (7 per cent) and the large farmers (4 per cent) groups. One
reason for the relatively higher gains in terms of poverty reduction for these
groups is the fact that the depth and the severity of poverty were not intense, to
begin with, for these relatively well-off household groups. Therefore, a small
increase in real income has been able to move a significant portion of the poor
population of these household groups out of poverty as compared with the less
well-off households whose depth and severity of poverty are more intense.
(ii) In the second simulation, the poverty situation in the rural location improved
in contrast to the generally worsening poverty situation in the urban location.
The head count index of poverty declined by 0.11 percent in the rural location
which increased by 0.42 percent in the urban location. One important
observation, however, is that the poverty situation worsened for all relatively
well off household groups except the large farmers. The rise in the incidence
of poverty can be specifically noticed for the medium skilled workers (2.5 per
cent), and the professionals (1.2 per cent) households. The relatively large
decline in the manufacturing income in the simulation led to a reduction in the

[Link] 35
 real incomes of these two household groups. The fall has been manifested in
widening of the poverty gap, deepening of the severity of poverty and
worsening of the head count index.

Table24: Impacts of Policy Simulations on Poverty Profiles

Simulation 1 Simulation 2
Head Poverty Severity Head Poverty Gap Severity
Household Groups Count P0 Gap P1 P2 Count P0 P1 P2
Rural 0.514 0.187 0.094 0.534 0.196 0.099
Agricultural Labourers 0.762 0.293 0.145 0.778 0.304 0.152
Small Farmers 0.503 0.155 0.065 0.522 0.164 0.065
Large Farmers 0.280 0.092 0.044 0.293 0.097 0.046
Non-farms 0.468 0.160 0.075 0.486 0.168 0.079
Urban 0.276 0.104 0.054 0.288 0.110 0.058
Workers-Low Skilled 0.363 0.124 0.059 0.377 0.130 0.063
Workers-Medium Skilled 0.097 0.016 0.004 0.110 0.020 0.005
Professionals 0.058 0.012 0.004 0.063 0.014 0.004
National 0.463 0.169 0.855 0.481 0.177 0.902
Percentage Change from the Base Run
Head Count Poverty Severity Head Poverty Severity
P0 Gap P1 P2 Count P0 Gap P1 P2
Rural -3.82 -4.93 -5.57 -0.11 -0.30 -0.53
Agricultural Labourers -2.32 -4.10 -5.18 -0.27 -0.49 -0.61
Small Farmers -3.73 -5.34 -6.53 -0.08 -0.06 -6.53
Large Farmers -4.21 -4.86 -5.57 -0.02 0.00 -0.28
Non-farms -3.75 -5.00 -5.88 0.052 0.052 0.045
Urban -3.84 -4.52 -5.01 0.42 0.80 0.95
Workers-Low Skilled -3.74 -9.10 -9.71 0.00 -4.33 -4.37
Workers-Medium Skilled -9.64 -13.52 -19.06 2.49 3.82 5.14
Professionals -6.92 -9.52 -11.99 1.17 1.68 2.18
National -3.90 -4.89 -3.98 -0.11 -0.20 1.40

8. Concluding Observation

In the present study, two simulations have been conducted to examine the impacts of
globalization measures on poverty and income distribution of seven representative household
groups in Bangladesh. In the first simulation, the base values of the tariff rate have been set
equal to zero to encourage trade expansion such that the “openness” indicator of globalization
is enhanced. The base values of all other parameters are retained. In the second simulation,
the base value of foreign savings has been augmented to reflect the recent pattern of foreign
investment inflow into Bangladesh. The inflows of foreign investment, however, are
concentrated mainly in the “gas” sector depicting a rise in “foreign investment to GDP ratio”
(e.g. investment indicator of globalization) as well as invoking deleterious impacts of such

[Link] 36
investment. The base values of all other parameters have been retained. The simulation
results highlight several macro and poverty impacts.

The real GDP growth in simulation one is 2.43 percent compared with the base case.
Moreover, the reallocation of resources to the unprotected sectors manifests in higher growth
of agriculture and service sectors. The resources are also observed to move to export oriented
manufacturing sectors resulting in their higher growth. The pattern of manufacturing growth
led by export-oriented industries results in high growth of the export sector. The increase in
imports is observed to be moderate despite the full elimination of tariffs due to substantial
depreciation of the nominal exchange rate which countered a large fall in the domestic prices
of imports.

The growth effects under the second simulation are somewhat different from similar effects
of the first simulation. In the second simulation, the resources move from both agriculture
and manufacturing sectors to generate growth in the service sector. The growth of imports is
relatively high due to the decline in domestic manufacturing and agriculture activities
resulting in higher prices of the domestically produced products relative to the import prices
of their import substitutes. Similarly, higher prices of domestic supplies result in sharp fall in
exports in the simulation.

In the case of welfare measures, Equivalent Variations (EV) are positive for all household
groups. The EV values reflect growth in real GDP and consumption. Except for the non-farm
household group, the observed EV is larger for the relatively high-income household groups
(e.g. the professionals, the medium skilled workers, and the large farmers) compared with the
low-income households (agricultural labourers, semi-skilled workers and the small farmers).
This suggests that the welfare gains emanating from “globalization” measures accrue more to
the well-off household groups compared with their less well-off counterparts. Among the less
well-off household groups, only the non-farm groups are observed to benefit due to their
wider participation in non-traded and service activities.

In terms of the impact on poverty, the poverty situation of all household groups is observed to
improve in the first simulation due to relatively high growth of income that results from the
policy change. The urban households gain more than the rural households and the gains in
terms of poverty reduction accrue more to the relatively well-off households. The highest

[Link] 37
reduction in the incidence of poverty is observed for the medium-skilled workers followed by
professionals and large farmer households.

In the second simulation, poverty situation in the rural location improves while it somewhat
worsens in the urban location indicating a contrasting pattern for rural-urban locations.
Moreover, the poverty indicators under the simulation indicate worsening of the poverty
situation for the relatively well-off household groups except for the large farmers. In
particular, the medium-skilled workers and the professionals experience higher incidence of
poverty due to resource reallocations from agriculture and manufacturing sectors to the
service sector.

What policy conclusions can we draw from the above results? The underlying premise for
promoting globalization in Bangladesh, as elsewhere in the world, rests on standard
arguments: wider involvement and greater integration of the domestic economy with the
global economy will benefit the country through improved external competitiveness,
increased exports, and higher economic growth. The process also brings with it the potential
to relax several constraints that Bangladesh faces such as a small domestic market, low
savings, and limited access to technology and finance. The inflow of foreign direct
investment can benefit the country by facilitating technology transfer, improving the
managerial and technical skills, and accessing the global marketing networks. Although such
‘growth’ arguments are important considerations, the equity and poverty implications of
globalization have significant policy relevance in Bangladesh in view of the central place of
poverty reduction in its development priorities. The analysis in the present study indicates
that, while the globalization efforts in Bangladesh are generally pro-poor, the gains are
relatively small and these differ across various household groups in the presence of structural
bottlenecks and other constraints. In particular, the gains accrue more to the relatively well-
off households while the extreme poor households benefit less. This indicates that the full
potential of globalization is not readily translated into poverty reduction in Bangladesh. In
order to make the liberalization policies sufficiently pro-poor, the process needs
complementary measures aiming at strengthening the institutional capabilities, addressing the
structural bottlenecks, and improving the anti-poverty policy regimes in the country. In
addition to ensuring the consistency of the macroeconomic policy regime with the
globalization efforts, this would require that the process be made sensitive to the social costs
and shaped by domestic policy regimes, be consistent with appropriately targeted social

[Link] 38
safety nets measures for the affected poor and guided by the institutional capacity to manage
the transition process.

[Link] 39
 References

BBS 2001, Preliminary Report of Household Income and Expenditure Survey 2000,
Bangladesh Bureau of Statistics, Government of the People’s Republic of
Bangladesh, Dhaka.
BBS 2001, National Accounts Statistics, National Accounting Wing, Bangladesh Bureau of
Statistics, Government of the People’s Republic of Bangladesh, Dhaka.
BBS 2000, Preliminary Estimates of Gross Domestic Product 1999-2000 and Final
Estimates of Gross Domestic Product 1998-99, National Accounting Wing,
Bangladesh Bureau of Statistics, Government of the People’s Republic of
Bangladesh, Dhaka.
BBS 1999, Statistical Year Book of Bangladesh, Bangladesh Bureau of Statistics,
Government of the People’s Republic of Bangladesh, Dhaka.
BBS 1998, 1995/96 Labour Force Survey, Bangladesh Bureau of Statistics, Government of
the People’s Republic of Bangladesh, Dhaka.
BBS 1998, 1995/96 Household Expenditure Survey, Bangladesh Bureau of Statistics,
Government of the People’s Republic of Bangladesh, Dhaka.
BIDS 2001, Fighting Human Poverty: Bangladesh Human Development Report 2000,
Bangladesh Institute of Development Studies/Bangladesh Planning
Commission, Dhaka.
BIDS 1998, An Input-Output Table for Bangladesh Economy 1993-94, Bangladesh Institute
of Development Studies/Bangladesh Planning Commission, Dhaka.
Boadway, R.W. and N. Bruce 1984, Welfare Economics, Basil Blackwell, New York.
CPD 1997, Growth or Stagnation? A Review of Bangladesh’s Development 1996, Centre for
Policy Dialogue/University Press Limited, Dhaka.
Decaluwe, B., A. Patry, L, Savard and E. Thorbecke 1999, Poverty Analysis Within A
General Equilibrium Framework, (mimeo), Laval University, Quebec City
and Cornell University, Ithaca, New York.
Foster, J.E., J. Greer and E. Thorbecke 1984, ‘A Class of Decomposable Poverty Measures’,
Econometrica, 52.
Hossain, M. and B. Sen 1992, ‘Rural Poverty in Bangladesh: Trends and Determinants’,
Asian Development Review, 10.
MOF 2001, Bangladesh Economic Review 2001 (in Bengali), Ministry of Finance,
Government of the People’s Republic of Bangladesh, Dhaka.

[Link] 40
Mujeri, M.K. 2002, ‘Globalization-Poverty Links in Bangladesh: Some Broad Observations’,
in A Review of Bangladesh’s Development 2001, Centre for Policy
Dialogue/University Press Limited, Dhaka.
Mujeri, M.K. 2001, ‘Macroeconomic Developments in the 1990s’ in A. Abdullah (ed)
Bangladesh Economy 2000: Selected Issues, Bangladesh Institute of
Development Studies, Dhaka.
Mujeri, M.K. 2000, Position Paper on Agriculture Sector, Background Paper for The Sixth
Five Year Plan 2002-2007, General Economics Division, Planning
Commission, Government of the People’s Republic of Bangladesh, Dhaka.
Mujeri, M.K. 1999, Poverty Alleviation in Bangladesh: Role of Economic Growth and
Special Programmes, International Labour Office, Dhaka.
Mujeri, M.K., Q. Shahabuddin and S. Ahmed 1993, ‘Macroeconomic Performance,
Structural Adjustments and Equity: A Framework for Analysis of Macro-
Micro Transmission Mechanisms in Bangladesh’ in Monitoring Adjustment
and Poverty in Bangladesh Report on the Framework Project, Centre on
Integrated Rural Development for Asia and the Pacific, Dhaka.
Pritchett, L. 1996, ‘Measuring Outward Orientation in LDCs: Can It be Done?’ Journal of
Development Economics, 49.
Pyatt, G., J. I. Round, and J. Denes 1984, ‘Improving the Macroeconomic Data Base: a SAM
for Malaysia 1970’ World Bank Staff Working Papers, No. 646.
Pyatt, G. and E. Thorbecke 1976, Planning Technique for a Better Future, International
Labour Office, Geneva.
Ravallion, M. 1994, Poverty Comparisons, Harwood Academic Publisher, New York.
Ravallion, M. 1990, ‘The Challenging Arithmetic of Poverty in Bangladesh’, Bangladesh
Development Studies, 18.
Ravallion, M. and B. Sen 1996, ‘When Method Matters: Monitoring Poverty in Bangladesh’,
Economic Development and Cultural Change, 44.
Sobhan, R. 1991, (ed) Structural Adjustment Policies in the Third World: Design and
Experience, University Press Limited, Dhaka.
Subramanian, S. 1997, (ed) Measurement of Inequality and Poverty, Oxford in India
Readings, Oxford University Press, New Delhi.
World Bank 2000, World Development Indicators 2000, World Bank, Washington D.C.
World Bank 1999, World Development Indicators 1999, World Bank, Washington D.C.

[Link] 41
World Bank 1998, Bangladesh: From Counting the Poor to Making the Poor Count, Poverty
Reduction and Economic Management Network, South Asia Region, World
Bank, Washington D.C.
World Bank 1997, World Development Indicators 1997, World Bank, Washington D.C.

[Link] 42
Appendix I: Summary of the CGE Model Specification

Equation Description
Price Block
PM i = PWM i ⋅ ER ⋅ (1 + tmi + tvi ) Import price

PM i = PWEi ⋅ ER Export price

Pi ⋅ Qi = PDi ⋅ Di + PM i ⋅ M i Composite price

PX i ⋅ X i = PDi ⋅ (1 − td i − tvi ) ⋅ Di + PEi ⋅ Ei Activity price

PN i = ∑ τ ji ⋅ Pj Input price
j

PVi ⋅ Vi = PX i ⋅ X i − PN i ⋅ INTi Value added price

PK i = ∑ κ ij ⋅ P j Capital price
j

PINDEX = GDPVA Numeraire price

RGDP
Production and Supply Block

−
1 Value added function
−µ µi
Vi = AVi ⋅ [ ∑ α if ⋅ FDif i ]
f

1 Factor demand
 α ⋅ PV  1+ µi
if i
FDif = Vi ⋅  
µ
 AV i ⋅ W ⋅ϖ 
 i f if 

Qi = AQi ⋅ [δ i ⋅ M i
− ρi
+ (1 − δ i ) ⋅ Di
− ρ i −1 / ρ i
] Composite supply
(Armington Function)
PDi ⋅ δ i Import-domestic demand ratio
M i = Di ⋅ [ ]σ i
PM i ⋅ (1 − δ i )
Qi = M i + Di Composite commodity aggregation for
perfect substitutes
Qi = Di Composite supply for non-imported
commodities
Qi = M i Composite supply for non-produced imports

X i = ATi ⋅ [γ i ⋅ E i
−φi
+ (1 − γ i ) ⋅ Di
−φi 1 / φi
] Composite supply function

PE i ⋅ (1 − γ i ) ϕ i Export supply
E i = Di ⋅ [ ]
PDi ⋅ (1 − td i )
ηi Export demand
 PWE i 
Ei = E i0 ⋅ 
 PWSE i 
Institutional Income
[ ]
Yh = YFh + RM h ⋅ (1 − thh − s h ) Household income

YG = ∑ thh ⋅ Yh + ∑ tmi ⋅ PWM i ⋅ M i ⋅ ER + ∑ td i ⋅ X i ⋅ PDi Government income

h i i

CDih ⋅ Pi = β ih ⋅ Yh Consumption demand

[Link] 43
 Equation Description
GDi = β ig ⋅ GTOT Government demand

PK i ⋅ DK i = ξ i ⋅ I Investment by destination

IDi = ∑ κ ij ⋅ DK j Investment by origin

j

INTi = ∑ τ ij ⋅ N j Intermediate demand

j

Equilibrium Condition
S = ∑ SH h + SG + SF Total savings by institutions
h

Qi = INTi + ∑ CDih + GDi + IDi Product market balance: supply equals

h demand
∑ PWM i ⋅ M i − ∑ PWE i ⋅ E i − ∑ RM h − SF = 0 Current account balance: receipts equal to
i i h outlays
I = S = ∑ SH h + SG + SF Macro balance: investment equals savings
h

GDPVA = ∑ PVi ⋅ Vi + INDTAX + TARIFF GDP value added

i

RGDP = ∑ CDi + IDi + GDi + E i − (1 − ∂ i ) * M i Real GDP

i

[Link] 44
Appendix 2: Bangladesh Social Accounting Matrix for 1995-96 (in Billion Taka)
Accounts Activity
Aman Boro Grains Commer Livestock Fish Forestry Rice Mill Ata

Aman 9.81 0.00 0.00 0.00 3.62 0.00 0.00 87.50 0.00

Boro 0.00 10.33 0.00 0.00 10.96 0.00 0.00 105.82 0.00

Grains 0.00 0.00 1.24 0.00 1.53 0.00 0.00 0.00 12.54

Commercial crop 0.00 0.00 0.00 17.18 3.03 0.00 0.00 0.00 0.00

Livestock 11.70 10.68 1.53 5.45 1.65 0.25 0.00 0.00 0.00

Fish 0.00 0.00 0.00 0.00 0.11 3.32 0.00 0.00 0.00

Forestry 0.00 0.00 0.00 0.77 0.00 0.10 0.00 0.75 0.00

Rice Mill 0.00 0.00 0.00 0.00 4.26 0.47 0.00 0.00 0.00

Ata 0.00 0.00 0.00 0.00 3.47 0.13 0.00 0.00 0.00

Other Food 0.00 0.00 0.00 0.09 5.11 0.42 0.00 0.00 0.00

Leather 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Cloth 0.00 0.01 0.00 0.00 0.11 0.27 0.00 1.40 0.12
Activity

RMG 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Tobacco 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Chemical 2.29 3.16 0.12 0.92 1.60 0.26 0.00 0.00 0.00

Fertilizer 3.41 5.65 1.47 7.28 0.02 0.17 0.00 0.00 0.00

POL 0.15 2.36 0.13 0.16 0.03 1.26 0.00 0.45 0.02

Machinery 0.31 0.41 0.04 0.13 0.24 0.66 0.44 0.66 0.31

Miscellaneous Industry 0.00 0.00 0.00 0.13 1.60 1.87 0.17 0.92 0.05

Construction 0.15 0.25 0.02 0.02 0.00 0.00 0.01 4.39 0.18

Utility 0.04 0.65 0.01 0.03 0.19 0.00 0.00 7.28 0.38

Trade 12.69 24.71 1.71 26.47 9.74 32.68 38.86 18.41 2.26

Social 0.00 0.00 0.00 0.00 0.35 0.00 0.00 0.00 0.00

Public Administration 0.03 0.05 0.01 0.12 0.03 0.02 0.72 0.01 0.01

Financial Ser 0.00 0.00 0.04 0.19 0.06 0.13 0.00 0.37 0.03

Other Service 0.13 0.24 0.01 0.04 0.96 0.18 0.00 0.03 0.00

Male-Low Skill 24.29 25.34 2.44 13.32 16.08 0.66 8.35 1.46 0.17

Male-Med Skill 5.45 5.68 0.55 2.99 3.61 0.88 1.46 0.84 0.10

Male-High Skill 2.43 2.53 0.25 1.34 1.61 0.83 0.76 0.64 0.08
Factors

Female-Low Skill 4.16 4.28 0.19 4.26 12.85 0.45 0.04 1.03 0.00

Female-Med Skill 0.70 0.71 0.03 0.72 2.18 0.06 0.00 0.05 0.00

Female-High Skill 0.15 0.16 0.01 0.16 0.46 0.04 0.00 0.04 0.00

Capital 23.04 29.93 2.33 43.26 32.40 30.49 30.78 89.63 7.35

Labourer

Small Farmers
Households

Large Farmers

Non-Farm

W-Low Skilled

W-Skilled

Professional

Government 0.00 0.00 0.69 0.63 0.00 0.15 0.00 0.02 0.00

Rest of the World 0.00 0.00 4.21 7.49 2.34 0.00 0.00 0.57 0.04

Consolidated Capital

Total Supply 100.93 127.12 17.04 133.15 120.20 75.73 81.60 322.29 23.66

[Link] 45
 Accounts Activity
Oth Food Leather Cloth RMG Tobacco Chemical Fertilizer POL Machinery

Aman 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Boro 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Grains 0.32 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Commercial crop 25.45 0.01 20.26 0.05 3.46 0.02 0.00 0.00 0.00

Livestock 0.41 12.97 0.00 0.00 0.00 1.00 0.00 0.00 0.00

Fish 10.07 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Forestry 0.27 0.00 0.07 0.00 0.00 0.78 0.00 0.00 0.03

Rice Mill 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Ata 2.58 0.00 2.16 0.00 0.00 0.01 0.00 0.00 0.00

Other Food 11.00 0.12 0.02 0.06 0.01 0.38 0.01 0.00 0.03

Leather 0.00 1.33 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Cloth 0.34 0.18 33.07 41.08 0.00 0.02 0.01 0.00 0.04
Activity

RMG 0.00 0.00 0.00 0.53 0.00 0.00 0.00 0.00 0.00

Tobacco 0.00 0.00 0.00 0.00 0.06 0.00 0.00 0.00 0.00

Chemical 0.98 3.27 5.04 0.29 0.16 14.67 3.86 0.06 2.92

Fertilizer 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

POL 0.99 0.03 0.29 0.11 0.01 0.12 0.03 3.10 0.31

Machinery 4.58 0.65 3.80 1.37 0.40 2.22 2.67 3.72 38.32

Miscellaneous Industry 1.92 0.13 0.90 1.47 1.82 2.79 0.23 0.05 1.22

Construction 0.99 0.19 0.77 0.84 0.04 0.55 0.45 0.45 0.90

Utility 1.14 0.39 3.53 0.36 0.04 0.56 1.60 0.15 3.25

Trade 6.13 1.65 11.07 5.12 4.26 4.06 1.38 1.32 3.21

Social 0.02 0.02 0.00 0.00 0.00 0.00 0.01 0.00 0.14

Public Administration 0.25 0.09 0.19 0.05 0.01 0.20 0.04 0.12 0.25

Financial Ser 1.78 0.79 1.61 0.34 0.04 0.65 1.36 4.04 2.35

Other Service 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Male-Low Skill 1.95 0.91 14.17 2.23 1.02 1.56 0.62 0.92 2.50

Male-Med Skill 1.17 0.46 7.15 0.86 0.59 1.05 0.40 0.22 2.01

Male-High Skill 0.87 0.51 7.93 1.46 0.46 2.53 0.99 0.20 1.59
Factors

Female-Low Skill 0.11 0.02 1.39 11.05 0.47 0.15 0.02 0.01 0.08

Female-Med Skill 0.01 0.00 0.40 2.96 0.02 0.02 0.00 0.00 0.00

Female-High Skill 0.00 0.00 0.13 2.06 0.02 0.45 0.07 0.00 0.00

Capital 21.69 7.76 28.86 25.65 7.26 3.50 2.50 12.43 18.48

Labourer

Small Farmers
Households

Large Farmers

Non-Farm

W-Low Skilled

W-Skilled

Professional

Government 5.65 0.19 8.39 0.15 8.70 9.39 0.00 8.46 20.66

Rest of the World 8.39 0.28 22.90 1.37 0.11 24.01 2.84 9.78 70.98

Consolidated Capital

Total Supply 109.08 31.94 174.09 99.48 28.94 70.69 19.12 45.03 169.27

[Link] 46
 Accounts Activity
[Link] Construction Utility Trade Social Pub Adm Fin Ser Other Serv

Aman 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Boro 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Grains 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Commercial crop 0.03 0.49 0.00 0.00 0.03 0.00 0.40 3.57

Livestock 0.02 0.00 0.00 0.00 0.13 0.00 0.00 5.76

Fish 0.00 0.00 0.00 0.00 0.02 0.00 0.00 2.38

Forestry 3.54 37.84 0.00 0.00 0.00 0.00 0.00 0.00

Rice Mill 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.98

Ata 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.44

Other Food 0.28 0.00 0.00 12.62 0.02 0.06 2.21 4.09

Leather 0.01 0.00 0.00 0.06 0.00 0.00 0.00 0.00

Cloth 0.29 0.06 0.01 1.84 0.28 0.16 0.44 1.10

Activity

RMG 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Tobacco 0.00 0.00 0.00 0.00 0.00 0.00 0.04 0.25

Chemical 3.64 4.28 0.02 4.87 2.12 0.55 2.20 0.00

Fertilizer 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

POL 1.39 1.26 2.09 20.74 0.54 0.65 0.31 0.00

Machinery 2.59 30.76 0.85 9.34 0.92 0.87 0.82 2.17

Miscellaneous Industry 14.66 29.02 0.17 16.73 1.38 2.71 6.48 1.00

Construction 0.98 0.12 0.15 0.63 0.54 5.08 0.20 10.62

Utility 5.22 6.58 3.51 4.92 0.91 0.61 1.39 1.26

Trade 6.61 35.09 4.08 79.47 12.52 1.54 43.31 31.83

Social 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.00

Public Administration 0.07 0.74 0.09 12.13 0.94 1.50 0.25 0.27

Financial Ser 0.54 0.95 0.13 20.35 0.78 3.98 3.62 3.61

Other Service 0.00 0.00 0.00 33.24 1.05 0.25 11.05 1.05

Male-Low Skill 6.68 17.46 1.24 152.32 1.42 2.95 2.70 27.21

Male-Med Skill 3.00 3.74 0.98 62.24 2.72 5.12 2.28 12.25

Male-High Skill 2.98 5.41 5.30 57.91 20.15 22.63 17.77 13.79
Factors

Female-Low Skill 1.62 0.63 0.02 4.44 0.43 0.23 0.09 10.54

Female-Med Skill 0.23 0.00 0.00 0.51 0.42 0.42 0.12 1.49

Female-High Skill 0.03 0.00 0.41 0.46 5.99 2.26 1.12 1.13

Capital 22.24 59.66 35.52 76.27 29.96 7.91 69.86 110.25

Labourer

Small Farmers
Households

Large Farmers

Non-Farm

W-Low Skilled

W-Skilled

Professional

Government 10.41 0.00 5.83 0.27 0.02 0.10 1.23 0.23

Rest of the World 99.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Consolidated Capital

Total Supply 186.07 234.07 60.39 571.41 83.30 59.57 167.91 248.28

[Link] 47
 Accounts Factors Households
M-Low Sk M-Med Sk M-High Sk F-Low Sk F-Fed Sk F-High Sk Capital Labourer Small

Aman 0.00 0.00

Boro 0.00 0.00

Grains 0.13 0.29

Commercial crop 4.26 10.06

Livestock 2.47 5.70

Fish 3.46 8.10

Forestry 4.20 7.81

Rice Mill 45.37 53.09

Ata 1.74 2.51

Other Food 3.52 10.09

Leather 0.30 1.31

Cloth 4.60 10.17

Activity

RMG 0.35 0.85

Tobacco 2.75 5.06

Chemical 0.97 1.85

Fertilizer 0.00 0.00

POL 0.30 0.57

Machinery 0.42 1.02

Miscellaneous Industry 2.82 12.94

Construction 0.00 0.00

Utility 0.42 1.03

Trade 2.51 9.24

Social 1.10 3.65

Public Administration 0.04 0.08

Financial Ser 5.91 12.50

Other Service 7.95 18.31

Male-Low Skill

Male-Med Skill

Male-High Skill
Factors

Female-Low Skill

Female-Med Skill

Female-High Skill

Capital

Labourer 74.90 7.06 1.24 12.24 0.51 0.23 0.00

Small Farmers 63.96 18.14 9.98 12.35 1.70 1.25 79.57

Households

Large Farmers 28.67 17.78 14.70 6.36 1.95 1.34 146.00

Non-Farm 78.54 28.62 36.53 14.75 3.16 2.87 123.88

W-Low Skilled 80.66 10.30 5.62 9.39 0.95 0.67 66.16

W-Skilled 1.91 42.54 6.82 1.37 1.55 0.69 133.77

Professional 1.35 3.33 98.04 2.11 1.21 8.10 279.65

Government

Rest of the World

Consolidated Capital 0.59 12.25

Total Supply 329.98 127.79 172.94 58.58 11.03 15.13 829.03 96.18 188.51

[Link] 48
 Accounts Households Other Institutions Capital
Consolidated
Large NFHH W-Low Skilled W-Med. killed Professional Government Rest of World Capital Total Demand

Aman 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.93

Boro 0.00 0.00 0.00 0.00 0.00 0.00 0.00 127.12

Grains 0.29 0.30 0.16 0.11 0.13 0.00 0.00 17.04

Commercial crop 11.02 9.77 5.03 4.11 14.45 0.00 0.50 133.15

Livestock 9.56 9.80 8.56 10.13 22.31 0.00 0.14 120.20

Fish 8.56 10.34 5.67 4.85 7.06 0.00 11.78 75.73

Forestry 6.53 9.91 4.71 2.51 1.78 0.00 0.00 81.60

Rice Mill 28.28 89.40 47.97 28.67 23.77 0.00 0.01 322.29

Ata 1.44 3.68 1.79 1.19 1.51 0.00 0.00 23.66

Other Food 12.66 11.28 6.83 7.00 13.74 0.00 7.43 109.08

Leather 1.78 1.91 1.16 1.70 5.54 0.00 16.84 31.94

Cloth 11.49 14.55 9.19 8.60 16.05 0.00 18.59 174.09

Activity

RMG 0.98 1.17 0.71 0.73 1.42 0.00 92.73 99.48

Tobacco 4.40 6.33 3.59 3.01 3.44 0.00 0.00 28.94

Chemical 1.90 2.96 2.12 1.52 1.76 0.00 0.30 70.69

Fertilizer 0.00 0.00 0.00 0.00 0.00 0.00 1.11 19.12

POL 0.48 1.13 1.78 1.67 2.07 0.00 0.49 45.03

Machinery 1.64 2.05 1.38 1.22 1.51 0.00 0.75 50.02 169.27

Miscellaneous Industry 24.86 14.51 8.17 7.94 16.74 0.00 1.74 8.93 186.07

Construction 0.00 0.00 0.00 0.00 0.00 0.00 0.00 205.57 234.07

Utility 0.92 1.79 2.65 3.19 6.39 0.00 0.00 60.39

Trade 14.08 21.00 13.87 18.43 72.11 0.00 0.00 571.41

Social 6.94 6.44 3.86 5.69 11.57 43.46 0.00 83.30

Public Administration 0.10 0.13 0.09 0.10 0.24 40.58 0.00 59.57

Financial Ser 16.08 19.76 13.38 14.85 37.68 0.00 0.00 167.91

Other Service 24.63 30.57 26.27 24.51 67.81 0.00 0.00 248.28

Male-Low Skill 329.98

Male-Med Skill 127.79

Male-High Skill 172.94

Factors

Female-Low Skill 58.58

Female-Med Skill 11.03

Female-High Skill 15.13

Capital 829.03

Labourer 0.00 96.18

Small Farmers 1.56 188.51

Households

Large Farmers 2.76 219.56

Non-Farm 7.50 295.85

W-Low Skilled 3.38 177.13

W-Skilled 21.57 210.22

Professional 23.51 417.31

Government 2.00 5.00 8.00 96.16

Rest of the World 254.33

Consolidated Capital 30.93 25.08 8.19 53.47 80.24 12.11 41.64 264.51

Total Supply 219.56 295.85 177.13 210.22 417.31 96.16 254.33 264.508 7124.59

[Link] 49