Draft

November 2007

Rhetoric v. Reality
in Bangladesh’s Trade Policy

Shantayanan Devarajan*
Zaidi Sattar**

Paper presented at the Golden Jubilee International Conference on Development Prospects of

Bangladesh: Emerging Challenges, 2-3 December 2007, organized by Bangladesh Institute of
Development Studies, Sheraton Hotel, Dhaka.

(*) Dr. Devarajan is Chief Economist, South Asia Region, World Bank. (**) Dr. Sattar is a staff
consultant, World Bank, Dhaka Office. Views expressed in this paper are those of the authors and
must not be attributed to the World Bank or its Board of Executive Directors.
 Rhetoric v. Reality in Bangladesh’s Trade Policy

“The ideas of economists and political philosophers, both when they are right
and when they are wrong, are more powerful than is commonly understood.
Indeed the world is ruled by little else.”--J.M. Keynes in The General Theory

One of the many paradoxes that characterize Bangladesh is that of its trade policy.
Starting in the early 1990s, Bangladesh lowered import tariffs, removed quantitative
restrictions, and reformed its industrial sector. There followed an export boom fueled by
garment exports, an acceleration of economic growth, and a reduction in poverty.
Bangladesh is living proof that trade liberalization works. Yet today the public
discussion in Bangladesh is almost uniformly opposed to further trade liberalization. We
hear every possible argument against greater trade openness, and hardly any in favor of
reforms that have contributed to the economy’s success. This gulf between reality and
rhetoric is significant because, despite the reforms of the past, Bangladesh today has the
most restrictive trade regime in South Asia [Ahmed and Sattar (2004)] and one of the
highest average tariff rates in the world. To accelerate and maintain its economic growth,
Bangladesh will have to improve its export competitiveness relative to other developing
countries—and lowering trade restrictions is essential to that objective.

In this paper, we attempt to resolve this paradox by first, in section I, describing

the evolution of trade policy and its results in Bangladesh. Next, in section II, we
summarize the main arguments made against further trade reform, and compare them
with economic logic and empirical facts. Finally, in section III, we offer a possible
solution to the puzzle. We show that the trade and industrial reforms of the 1990s were
largely driven by external policy advice. Although they have been successful, these
reforms were not initially “owned” by the broader civil society in Bangladesh.
Consequently, despite its potential to accelerate growth and reduce poverty in
Bangladesh, the idea of further trade reform is resisted because of its association with
external actors. The paper concludes by hoping that the resolution of the paradox will
lead to a more evidence-based debate in Bangladesh about the future of trade policy.

Bangladesh is not unique to popular resistance to trade liberalization. Dani Rodrik

(2002) carried out a cross-country survey of public opinion to find out why some people
and countries were more protectionist than others. He found that even in developed
countries, of the people surveyed, some 60% favored trade restrictions. Among pro-trade
people were those engaged in export and non-traded sectors, while those who worked in
industries with revealed comparative dis-advantage were anti-trade. Surprisingly, the
survey found that nationalistic and patriotic citizens were more protectionist than
cosmopolitan individuals.

2
I. The Evolution of Trade Policy in Bangladesh

At independence Bangladesh inherited from Pakistan a trade regime reminiscent

of the 1950s Prebisch-Singer [1949, 1950] approach towards import-substituting
industrialization (ISI). High tariffs, pervasive bans and quantitative restrictions on
imports along with an overvalued exchange rate characterized this regime. In the absence
of any industrial house of East Pakistani origin 1 , the principal beneficiaries of this
restrictive trade regime were the large industrial houses in West Pakistan that owned and
operated much of the manufacturing sector dominated by jute and cotton textiles.
Consumers in East Pakistan generally paid higher than world prices for domestically-
produced manufactured products, resulting in a significant transfer of monopoly rents to
West Pakistani entrepreneurs and the accumulation of capital in that part of the country.

Though a team of Harvard-trained economists located in the then Pakistan

Institute of Development Economics (PIDE) – the precursor of BIDS -- had begun to take
note of these distortions in trade policy and the consequent disparities in wealth between
the two wings of Pakistan (MacEwan 1971), their research findings had little influence in
shaping Pakistan’s trade policies, which remained largely in the grip of the bureaucratic-
industrialist nexus. Trade policy kept most competing imports out, and created space for
the emergence and prosperity of the legendary “22 families” – industrial groups – in West
Pakistan. With an overvalued exchange rate, special export premium rates were offered
to, among others, jute and jute goods exports, with rents from the premium being
siphoned off by traders and industrialists leaving jute growers with very little of the
returns.

So in 1972 Bangladesh inherited a highly restrictive, inward-looking, import-

substituting trade regime, with little debate about its relevance to the new economy
whose lifeline was actually an export sector—jute and jute goods. The inherent anti-
export bias of an ISI regime went unnoticed as Bangladesh chose a path of import
substituting industrialization with minimal impact on industrial development. By the
1980s, the shortcomings of ISI policies became all too evident globally and the strategic
importance of export-oriented policies began to be realized, particularly in developing
economies trying to enter markets for labor-intensive manufactures. Though Bangladesh
began dismantling its ISI orientation in the early 1990s, to this day, the last vestiges of
ISI continue to haunt policymakers and economists alike, with strong backing from the
emerging entrepreneurial class in Bangladesh.

Although a price taker in the world market for exports and imports, Bangladesh
had some monopoly power in the world jute market, as it was the largest single exporter
of raw jute. But that leverage was soon lost as synthetic substitutes took over the market
and Bangladesh was unable to stem the tide by way of innovating new products from
jute. Thus began the slide in the economy’s fortunes originally tied to the ‘golden fibre’.
It was a fortuitous external event -- the imposition of textile quotas under the Multi-Fibre
Arrangement (MFA) in 1974 -- that not only saved the day but laid the foundations for

1
With the lone exception of Chittagong’s A. K. Khan Group.

3
Bangladesh to exploit market opportunities on the basis of the classical notion of
comparative advantage.

Bangladesh Trade Policy in the 1970s

In 1972, there was little debate on what course trade policy should take in
Bangladesh. The ravished state of the economy, narrow manufacturing base (barely 10%
of GDP), overwhelming import dependence to meet domestic requirements of consumer
goods and industrial inputs, and exports limited to a small range of primary products
dominated by jute and jute goods—all of these factors severally and jointly set the tone
for trade policy. To keep the import bill within the limits of foreign exchange resources
generated mostly from aid and exports of jute and jute goods, a highly restrictive import
regime was formulated, the exchange rate was fixed, and export earnings and all foreign
exchange transactions were fully controlled by the central bank.

By the end of the decade, one could get a sense of what the trade regime looked
like – a mere continuation of the old ISI regime. Import protection benefited the nascent
industries set up under the East Pakistan Industrial Development Corporation (EPIDC),
the nationalized enterprises of Pakistani and Bangladeshi owners under the
nationalization policy, and a few industries that were still left in private hands. By default
if not by design, agriculture received scant protection, turning the domestic terms of trade
against agriculture.

To be sure, during the first part of the decade the government was in no position
to articulate a clear policy regarding trade and its impact on economic performance.
There was much firefighting needed to meet the import demands of a devastated
economy constrained by foreign exchange shortages. It was only towards the middle of
the decade that policymakers started addressing trade policy issues albeit in fits and
starts. The institution of MFA quotas in 1974, putting a cap on textile exports from
competitive suppliers like Korea, Hong Kong, China and India, created opportunities for
Bangladesh which could only be seized by devising import schemes that addressed the
foreign exchange constraint and tariff schemes that allowed duty-free imports of inputs
for the proposed export-oriented, readymade garment (RMG) industries. Back-to-back
Letters of Credit (LC) and special bonded warehouses for duty-free imports of raw
materials were innovations borrowed from Korean garment exporters who needed to
exploit Bangladesh garment quotas for the US and European markets. Though trade
policy remained generally restrictive for the whole decade, these innovations laid the
foundations for export-oriented growth in an otherwise high-tariff import regime.

To sum up, trade policy during this first decade can be characterized by high
tariffs and extensive import bans and restrictions. Gazetted notification of Import Policy
Orders (IPO) provided a “positive list” of importable items beyond which all other
imports remained prohibited to (a) protect nascent domestic industries from import
competition, and (b) ration the use of scarce foreign exchange. The goal was to promote
industrialization within a protected economic environment while keeping the balance of
payments on an even keel. Note that during the 1972-80 period, other domestic economic

4
policies also centered on control and regulation. Manufacturing was dominated by SOEs,
the financial sector was subject to strong regulations with predominant public ownership
of banks and other financial institutions, and investment regulations were overwhelming.

This approach neither spurred industrialization nor improved the balance of

payments situation nor relaxed the foreign exchange constraint (Table 1). By 1980, the
balance of payments was characterized by massive trade and current account deficits
requiring heavy dependence on aided funds (which mostly came from the World Bank’s
IDA) to finance the import bill for critical industrial raw materials and intermediate
goods needed to enhance capacity utilization in the industrial sector.

Table 1. Trade and Production structure in the 1970s

1973-74 1979-80 1980-81
Balance of Trade (% GDP) 5.0 11.2 10.6
Current account balance (%GDP) 3.6 5.9 4.9
Share of manufacturing (%GDP) 11.7 11.2 10.5
GDP (billion Taka) 75.75 196.05 231.42
Source: Bangladesh Economic Review 1997; Economic Trends, Bangladesh Bank

Bangladesh Trade and Industrial Policy in the 1980s 2

Towards the end of the 1970s, there was a slow realization that the prevailing
approach to industrialization was not bearing fruit and some change was essential.
Although the bulk of the labor force was employed in agriculture, that sector’s capacity
to create additional employment and absorb the new entrants to the labor force was
severely limited. Industry had to be the leading sector to drive growth, generate
employment and reduce poverty. Manufacturing therefore began to be given priority
attention with incentive schemes biased in its favor. The futility of the 1972
nationalization – 92% of assets of the formal manufacturing sector were brought under
state ownership in one fell swoop – was also recognized. The limits of the public sector
in the management of manufacturing enterprises were slowly becoming apparent. The
vital role of the private sector in driving prosperity began to be realized in policy circles.
This led to the first steps being taken towards privatization of SOEs under the New
Industrial Policy (NIP) of 1982.

The NIP 1982 acknowledged the debilitating shortcomings of the prevailing

industrial and trade policy regime on export performance, and took the first important
steps to rectify the situation. Reforms thus initiated were subsequently reinforced in a
revision of NIP in 1986. The thrust of the two NIPs was to broaden and diversify the
industrial and export base, spearheaded by the private sector; limit the role of the public

2
It would not be meaningful to discuss trade and industrial policy as distinct from each other during the
decade of the 1980s since they complemented and reinforced each other. The ostensible goal of
policymakers was to get growth going and industrialization was the preferred route. To get progress on
industrialization, trade policy and consequent incentive regime had to be set right. While industrial policy
underwent changes in the first half of the decade, significant developments on the trade policy side
occurred in the latter half.

5
sector to strategic industries; and improve the efficiency of public sector industrial
enterprises.

To promote exports in an environment of high tariffs, an overvalued exchange

rate, and pervasive QRs is a major challenge. A number of instruments have to be put in
place to compensate. As promotion of export growth and diversification into non-
traditional items became a major emphasis of government policy in the 1980s, it is no
surprise that an array of policy instruments had to be employed to create sufficient
incentives for exports. These included

• Export Performance License Scheme (XPL), subsequently replaced by Export

Performance benefit (XPB) Scheme
• The Duty Drawback System
• The Bonded Warehouse Scheme
• Export financing incentives, such as Export Credit Guarantee Scheme, Export
Development Fund
• Export Processing Zone, and
• Foreign exchange rate management.

On the import side, progress on relaxation of the import regime was much more guarded
and tariff adjustments marginal

The reform proposals also included putting caps on the combined tax incidence
of tariffs and other levies to 125%, with some exceptions, such as luxury cars and
alcoholic beverages. To make scheduled QRs transparent and monitorable, the “negative
list” was introduced in FY86, replacing the former system of a “positive list”.
Subsequently, the approach was to pare down the number of items in the negative list –
particularly those included for protection purposes. The strategy of reducing QRs came
with the objective of replacing such QRs with tariffs. To back up the tariff proposals with
analytical research, for the first time, a study on effective protection analysis was
undertaken.

On the face of it, these trade policy packages, if implemented, would have moved
the economy towards significant liberalization. In practice, however, there were only
marginal changes to be seen in the reduction of the Negative List or reduction of high,
almost prohibitive, tariffs.

More significant was their overall impact on agriculture. The elimination of

import controls on minor irrigation equipment along with a reduction of tariffs
complemented the market orientation reforms in agriculture. The IPCs supported the
Government’s primary objective to increase foodgrain production and productivity by
focusing on reforms in the following major policy areas: foodgrain production,
marketing, and storage; public foodgrain distribution; fertilizer marketing, distribution
and pricing; minor irrigation equipment distribution and pricing, and so on.

6
 Box 1: The case for and against trade protection

A study of the evolution of trade policy in Bangladesh would be incomplete without some reflection on
the case for free trade and the arguments for protection as they evolved in economic theory and literature on trade
policy. The influence of neoclassical economic thinking on policymaking – through research and teaching by the
many leading economists of the country who were trained in such economic ideas – has also to be understood and
appreciated. Also notable are the cross-currents of socialist and radical thinking on trade and other economic
policies that continue to have a measure of influence on the intensity and direction of policy reforms.

In its purest form, free trade would be the absence of any trade policy intervention. Trade policy implies
the presence of government intervention in the form of tariffs, quantitative restrictions, or other non-tariff barriers
on import and export transactions. The classical Ricardian theory of comparative advantage was founded on the
assumption of perfect markets in international trade where factor endowments guide specialization and gains from
trade for trading partners. It takes little imagination to find that the real world of today (or at any time) suffers from
all kinds of market imperfections and/or market failures. This is not limited to differences in initial conditions
between countries or differences in institutional development but a whole host of phenomenon that would negate
the existence of perfect markets in international trade. Simply put, the presence of market failure which
presumably could be corrected by government intervention therefore provides the theoretical basis for protection
by means of tariff and non-tariff instruments that hinder free flow of goods across borders.

The problem is that trade protection once introduced into the economy gathers its own momentum for
sustenance. The evidence is uncontestable: As countries embraced the argument for infant industry protection in
the 1960s, pretty soon the world became littered with geriatric infants. Discredited in the literature, the infant
industry argument soon gave way to that of learning-by-doing under which it was argued that new industries or
activities need time to learn, after which they would have been able to reduce production costs and compete with
established producers. Even this latter argument did not specify whether all or some activities were candidates for
such learning-by-doing that would eventually reduce production costs to competitive levels. Again, if this
argument for protection were to be accepted as valid, it says nothing about the periodicity of such learning-by-
doing. How long is not too long? Three, five, ten or twenty years? There is clear evidence that longer the period of
protection, greater is the inefficiency that is bred by such policy – all at the cost of the consumer who must pay the
high price and choice limitation that protection policy entails. On the basis of collected evidence around the globe,
one can argue that only time-bound protection, which carries with it a threat of elimination of support after a
reasonable period of time, can be justified on grounds of learning-by-doing.

The case of South Korea and Taiwan are examples of such time-bound protection yielding good results.
Following the Korean War (1950-53) which devastated Korean industry, the government maintained a
protectionist outlook with high trade barriers for about a decade. Thereafter, by the mid-sixties, there was a clear
policy shift towards outward orientation with a pro-export tilt. The results are well known1. Taiwan pursued the
same policy after the Chinese civil war – first, by protecting domestic industries for about a decade, then by
switching to an outward-oriented pro-export policy as it was clear to policymakers that the domestic economy was
too limited to give Taiwanese industry the market for expansion that it sorely needed. In both cases, the policy of
high industrial protection was limited to about a decade, following which they embraced import liberalization as a
way of boosting exports by making imported inputs readily available at world prices.

In sharp contrast, while South Korea opened up its economy in the 1970s, India toyed with trade
liberalization but made little progress. Between 1960-80, Korean exports grew 24% annually, imports at 18%, and
per capita income by 7%. Export share of GDP rose from 5% to 33% during the same period. India’s import share
in GDP fell from 7% in 1958 to 3% in 1976. Export growth was negligible, while per capita income grew barely
1% annually.

Thus, at the end of the decade, the trade regime remained restrictive. Despite
liberalization of the import regime undertaken since the mid-1980s through relaxation of
some bans and quantitative restrictions, actual achievements appeared modest. The

7
coverage of QRs was. The Government only guardedly addressed the issue of
rationalization and reduction of the tariff structure, so there was no appreciable change in
average tariffs or dispersion. Distortions remained in the trade regime stemming from
high tariff rates, wide rate dispersion, rate differentiation based on industry, stage of
production, end product use, and ad hoc exemptions, leading to uneven effective
protection, and absence of competitive pressure on domestic industry to improve
efficiency.

While industrial policy in Bangladesh was moving in the direction of de-

nationalization with modest relaxation in the import regime, some new thinking also
emerged in trade literature with regard to the importance of trade openness and export
orientation in the trade policies of developing countries (Box 1). The contention of the
Prebisch-Singer hypothesis was called into question. In the late 1970s, a number of
research findings on the consequences of ISI policies in developing countries showed the
harmful effects of such inward-looking trade policies on growth. Most influential
amongst these were the seminal work done by Bhagwati (1979) and Krueger (1978).
Their research revealed the adverse consequences of distortions created in the economy
through interventionist policies while presenting a strong case for outward-looking
export-oriented approach. In South Asia, Sri Lanka took the lead in quickly dismantling
high tariffs and import controls and opening up its economy by the end of the 1970s.

Like their counterparts in India at that time, Bangladesh policymakers remained

largely unconvinced by the “openness-favors-growth” argument, only toying with the
idea of trade openness and outward orientation, but, for the most part, leaving a largely
restrictive import regime untouched. A snapshot of the overall trade regime at the close
of the decade could be summarized as follows (Table 2):

Table 2. Snapshot of Trade Regime pre-1990

Policy Criteria Status
Exchange Rate Fixed, Over-valued
Payment convertibility
Current account No
Capital account No

Import restrictions
Import licensing Yes…license raj
QRs on imports Trade QRs rampant
Tariff structure
Top normal rate 1990 160
Average protective rate n/a
Tariff slabs (customs duty) numerous
Para-tariffs, levies many
Trade openness, trade-GDP ratio (%) 1990 20
Source: World Bank (2004)

8
Trade Policy Developments in the 1990s

By 1990, Bangladesh policymakers realized that inward-looking import

substituting industrialization – supported by interventionist domestic public policy-- was
not yielding growth dividends. For two decades, trade protection persisted, and
government ownership and control over the economy was pervasive. The result was
sluggish growth of the economy and lackluster export performance. With the coming of
an elected government, more definitive trade policy reforms were undertaken, starting
around June 1992.

It was fortuitous that India embarked on trade liberalization at about the same
time under the leadership of Finance Minister, Dr. Manmohan Singh, who was on
friendly terms with his Bangladeshi counterpart. While the Indian budget would be
announced at end February, the Bangladesh budget, which spelled out the tariff
adjustments and trade policy stance in detail, was submitted in June. That gave sufficient
time for the Bangladesh Finance Minister to review the steps towards unilateral
liberalization that India was pursuing at the time and chart out its own course. It turned
out that throughout the four budgets – FY93-96 – which saw sharp reduction in tariffs,
customs revenues continued to rise, growing at an average annual rate of 15%. Two
factors were responsible for this outcome: (a) simultaneous removal of QRs, which
eliminated pervasive bans and restrictions on imports, and (b) the response of imports
coming from elimination of high and often prohibitive tariffs. The two simultaneous
events stimulated imports while bringing their domestic prices closer to world prices.

Table 4: Evolution of import restrictions 1991-2005

IPO IPO IPO
IPO IPO 2003-06 2003-06, 2006-
IPO IPO IPO
1989- 1997-02 Jul 03 Jul05 09,
1991-93 1993-95 1995-97
91 Jun07
Number of 320 193 111 120 122 67 25(chk) 26
items in the (25.8%) (15.6%) (9.0%) (9.7%) (9.8%) (7.3%) (2.0%) (2.1%)
control list at
the HS 4-
digit level
Number of n.a. 79 19 27 28 19 5 2
trade-related (6.4%) (1.5%) (2.2%) (2.2%) (1.9%) (0.3%) (0.2%)
items in the
control list at
the HS 4-
digit level
Source: IPOs various years; Ministry of Commerce

At the end of 1986, roughly 50% of importables remained under some form of
bans or restrictions. The first significant steps towards removal of QRs were taken in
1988-89, and in the IPOs of 1989-91 and 1991-93 (Table 4). However, the declining
trend was halted during much of 1993 through 2002. At the end of the millennium,
Bangladesh remained the only country to still use traditional QRs for industrial
protection. It seems that policymakers took heed of this fact and resumed dismantling of
QRs in 2002. Progress over the next four years was phenomenal as the last of the trade-

9
related QRs (Figure 1), particularly on the textile categories, were removed, leaving only
a ban on eggs and some restriction on salt import 3 . Thus the latest IPO 2006-09, wiped
the slate clean of protective QRs, As of June 2007, tariffs remain the only formal
instrument of protection.

Figure 1. Evolution of Import Restrictions (1991-2007)

(Number of items in the control list at the HS-4 digit level)
250

193
200

150
120 122
111

100
79

50 27
28 26
19 25
3 2
0
1991-93 1993-95 1995-97 1997-02 2003-06 (Jul 05) 2006-09 (Jun 07)

Overall QRs Trade-related QRs

While there was progress in removing QRs, tariff reduction in the 1990s appeared
to be a more contentious and hotly debated topic of trade policy with strong resistance
coming from producer groups engaged in production for the domestic market. As
described earlier, substantial dismantling of QRs began during 1989-91, finally ending
with the dissolution of the import licensing system in 1992.

Exchange liberalization. Liberalization of the exchange rate also began around

this time, with the unification of the exchange rate in 1992, and the move towards a more
flexible exchange rate system.

Tariff rationalization and reduction. During the 1990s, Bangladesh significantly

reduced its tariff rates and rationalized the structure, progressively moving towards the
goal of simplicity and transparency of the customs tariff. The top custom duty (CD) rate
came down from 160 percent in 1990 to 25 percent today (Table 5). The average
(unweighted) customs duty declined to 16% in FY08 as compared with 57% in FY92,
and 100% in 1985. Considerable rationalization of the tariff structure occurred and
progress was made towards achieving a degree of uniformity and removing some tariff
anomalies that existed due to higher tariffs on intermediates than final products

3
MOC had received WTO waiver for ban on salt imports under Article XVIII until end 2007. IPO 2006-
09 replace that ban with some restriction.

10
 Table 5: Progress in tariff rationalization
FY92 FY96 FY97 FY02 FY05 FY08
Tariffs
Average CD (unweighted) 57 22 22 17 16.3 15.9
Average protective tax 61 74 26 27 26.5 21.9
Top CD rate 160 50 45 37.5 27.5 25.0
Tariff slabs 17 6 6 5 4 3

Source: NBR and staff estimates

Figure 2. Bangladesh 1991/92-2007/08

Unweighted Average Protective tariffs

80
60
Avg rate

40
20
0
1991/92

1992/93

1993/94

1994/95

1995/96

1996/97

1997/98

1999/99

1999/2000

2000/01

2001/02

2002/03

2003/04

2004/05

2005/06

2006/07

2007/08
Custom s Duty Para-tariff
Source: NBR Tariff database

Two features stand out in the evolution of protective tariffs: (a) the sharp
reductions which began in FY92 halted around FY96 (see Figure 5 and Annex-I); (b) the
emergence of para-tariffs, additional levies such as supplementary duties, regulatory
duties, and infrastructure development surcharge. Trade liberalization slowed down by
the mid-1990s, and was even somewhat reversed during the last few years of the decade,
on the popular notion that earlier reforms had been “too much too fast”. Resistance to
liberalization from business and industry became strong enough at this time to forestall
further reduction in protection. One reason could be that productivity improvements and
efficiency adjustments were not fast enough for those industries that had gotten used to
high protection. At the end, these groups exerted enough pressure to undermine political
will to carry forward the liberalization agenda. Thus although customs duties kept
coming down, para-tariffs emerged and soon assumed importance as protective
instruments, so much so that by FY04, nearly a third of nominal protection was being
derived from these add-ons. It was then that trade reform had to address the para-tariff
issue head-on with some results in June 2007.

While Bangladesh reduced tariffs, -- rapidly at first, and gradually later -- other
countries in the South Asia region and across the globe moved much faster.
Consequently, despite a decade and a half of liberalization, judged by the level of average
protective tariffs, Bangladesh today has the most restrictive trade regime in South Asia, if
not in the world (Table 6, and Annex-II).

11
 Table 6. Average MFN tariffs in South Asia

Average tariff (%) Year

Bangladesh 21.9 2007
India 15.0 2007
Nepal 14.7 2005
Pakistan 14.6 2007
Sri Lanka 11.3 2005
Source: WDI, World Bank staff estimates

Only a couple of years ago, India would have claimed that distinction. But after
the removal of all QRs in April 2001, India has brought down its tariff levels
substantially – the top general level of non-agricultural tariffs in India as of February,
2007, is at 10 percent, though agricultural tariffs are still higher. India’s average
protective tariff now is around 15% compared to Bangladesh’s average of 22%. Though
Bangladesh’s top customs duty rate has been scaled down to 25% in FY05, it gave a
misleading impression of protection levels since a good number of consumer goods (e.g.
dairy and bakery products, beverages, sugar, dry cell batteries, textile fabrics, toys and
furniture) are subject to extremely high protective tariffs, ranging from 50% to 100%,
made possible with the imposition of supplementary duties. To rationalize the tariff
structure and bring transparency, the budget of June 2007 announced the withdrawal of
one para-tariff – IDSC – thus leaving only supplementary duties as the only additional
complement to tariffs.

Table 7 gives a snapshot of the current trade regime in Bangladesh. The most
significant changes from the pre-1990 snapshot presented in Table 2 are the absence of
trade QRs, a floating (albeit managed) exchange rate system, only one para-tariff (SD),
few tariff slabs (4), and a doubling of the trade-GDP ratio.

Table 7: Snapshot of Bangladesh Trade Regime 2007

Policy Criteria Status
Exchange Rate Unified, Managed Float
Payment convertibility
Current account Yes, some limits
Capital account No
Import restrictions
Import licensing No
QRs on imports Trade QRs gone
Tariff structure
Top CD rate 25
Average protective rate 21.9
Tariff slabs (customs duty) 5, 10, 15,25
Para-tariffs SD*
Trade openness, trade-GDP ratio (%) 2007 40

What has been the impact of one and a half decade of trade liberalization in
Bangladesh? To run causality tests within some econometric formulation would be
problematic and run into the usual degrees of freedom problem, not to mention the

12
difficulty of getting a good handle on data on trade openness. Instead, we will ask the
question: did the predicted (by critics) harmful impacts – such as de-industrialization,
declining growth rates or economic stagnation, etc. – actually materialize? Did industries
one after another close down with massive job losses? Did the economy slide backward
with greater openness? Answers to these questions can easily be seen in the numbers
(Table 8).

Without going into rigorous attribution analysis, it is possible to ask: what

differentiated the dominant policy environment of the 1990s from that of the 1980s and
before? Trade openness, deregulation, privatization and market orientation as a set of
policy package began to be emphasized as the cornerstone of public policy, beginning in
the early 1990s. This represented a significant departure from past approaches in favor of
protective import substitution, market interventions for controlling prices of many
outputs and inputs, and pervasive public sector ownership and operation of industrial
enterprises.

To be sure, with regard to market orientation and integration with world markets,
Bangladesh adopted a gradualist approach as opposed to the big bang. Yet, two key
statistics stand out: the economy is much more open today than it was pre-1990, with
trade-GDP ratio at 40% in 2007, compared to 20% in 1990; annual manufacturing growth
averaged 6.9% during 1991-00, and 8% since 2001, compared to 5% during 1981-90 –
the pre-reform period.

Table 8. Growth performance and poverty reduction, 1974-2007

Avg Average Poverty Trade-GDP Manufacturing
Avg GDP
Period Per Capita Inflation Headcount % Ratio growth
Growth
GDP Growth Rate end of period End period

1972-1975 2.75 -0.10 47.14 11.2 na

88.14*
1976-1990 3.89 1.58 9.12 58.8** 20.0 5.0

1991-2000 4.80 3.09 5.64 49.8*** 30.1 6.9

2001-2007 5.76 4.26 5.1 40.0**** 39.7

8.0
*refers to 1973-1974, ** refers to 1991-1992, *** refers to 1999-2000, average for 1972-75,
**** refers to 2005
Source: Updated from Ahmed and Sattar (2004)

Though in relative terms Bangladesh still remained behind its comparators in

South Asia in terms of trade openness, it appears to have reaped benefits from the
liberalizing policies begun in 1991 in the form of a superior growth performance. Per
capita GDP grew 3% in the 1990s and over 4% since 2000. Annual growth in industry
and services sectors averaged 6-7% since 1991. Though direct causality cannot be
established between trade liberalization and growth performance the empirical evidence
suggests that economic performance certainly benefited from a more open trade regime.

13
 Figure 3. Results on the Ground

Average Growth Rate of Industry and Services (FY81-FY06)

9.0
7.5
6.0
4.5
3.0
1.5
0.0
F Y81

F Y83

F Y85

F Y87

F Y89

F Y91

F Y93

F Y95

F Y97

F Y99

F Y01

F Y03

F Y05
The post-1990 era also saw Bangladesh experience the highest rate of export
growth, averaging 13% in the 1990s, led by the surge in RMG exports. That exemplary
performance has continued into the next decade and is sustained beyond 2005 even after
the phase out of the MFA. What remains a policy failure perhaps is the inability of the
economy to reduce its predominant reliance on RMG exports which leaves the economy
vulnerable to external shocks arising from any future decline in export markets for
garments. Despite repeated emphasis on an export strategy whose main thrust lay in
diversifying Bangladesh’s export basket, no progress is discernible over the past ten years
or so, during which RMG exports continued to be the most dynamic sector, comprising
roughly 75% of the export basket.

II. The Current Debate on Trade Reform in Bangladesh

Despite the abject failure of trade protection in Bangladesh, and the apparent
success of trade liberalization—in stimulating exports, generating growth and reducing
poverty—the current debate about trade policy is heavily biased against trade reforms.
The debate is more than just an academic exercise because, as noted above, Bangladesh
remains one of the most protected countries in the world. A case can be made therefore
that the country needs further trade liberalization in order to maintain its competitiveness
vis-à-vis other developing countries. In this section, we examine some of the arguments
against trade liberalization in Bangladesh with a view towards understanding their
internal logic as well as their relationship to Bangladesh’s history with trade reform.

14
 Domestic market v. exports. One argument is that Bangladesh has a large (in
population terms) domestic market, so policy should be geared towards supplying that
market rather than the foreign market. Proponents of this view 4 acknowledge that the
domestic market is small in economic terms because Bangladesh is a poor country. But
this speaks to the fallacy of the argument. Bangladesh can help the domestic market
grow by selling its goods to the (huge) foreign market and use the income earned from
this to increase the size of the domestic market. This is precisely how it has grown in the
past fifteen years. And the experience of the previous two decades show that producing
for the domestic market behind protective trade barriers has neither stimulated growth nor
saved the country from foreign exchange problems.

Backward linkages. Another argument is that Bangladesh’s major export,

garments, has very little “backward linkages”. At least some of the protected domestic
industries, such as textiles, have strong backward linkages with yarn manufacturers, etc.
While this may be true in the abstract, the numbers don’t bear this out. The garment
export sector employs 2.2 million workers 5 ; the backward linkage textile sector employs
some 150,000 workers. Even if the multiplier (thanks to backward linkages) were 4 or 5,
the overall employment effect would not come close to the level of garment exports.
Furthermore, developing industries with backward linkages behind protective tariff
barriers, so the economy loses money on every unit produced, does not benefit
Bangladesh, as the experience of the jute mills showed.

Since garments already enjoy duty-free inputs, there is no need for further trade
liberalization. Import tariffs and quotas reduce the competitiveness of all exports. While
garments enjoy duty-free inputs under the bonded warehouse scheme, a reduction in
tariffs and quotas would enable garments and other industries to export more. Why?
Reductions in all tariffs and QRs will permit other exporters to compete on an equal
footing with their counterparts in other countries as they can import raw materials and
intermediate inputs cheaper and without hassle. This is particularly important as
Bangladesh seeks to diversify its exports.

Bangladesh should not lower its tariffs until the Europeans and Americans reduce
their tariffs and agricultural subsidies. The OECD countries have tariffs and subsidies
for domestic political reasons. The only way they will lower them is if there is a change
in their domestic politics. They are unlikely to lower them as part of a bargain with
Bangladesh to lower its tariffs. Given that it cannot affect tariffs in OECD countries,
Bangladesh’s best strategy is still to lower its tariffs. This would enable Bangladesh to
export more to these countries for a fixed level of protection in the OECD.

Opening up risks running large trade deficits. The evidence is the opposite.
When Bangladesh was a relatively closed economy, trade deficits were higher than today.
Bangladesh suffered a serious balance of payments crisis at the end of 1980s. Subsequent

4
For a recent expression of this view, see Salim Rashid’s comments on the World Bank report,
“Bangladesh: A Strategy for Sustained Growth”. http://www.thedailystar.net/story.php?nid=10804
5
Indirectly, RMG linkage activities like packaging, courier, banking services, and the like employ another
million or so workers.

15
exchange rate liberalization and opening up of the economy actually helped improve the
BOP situation with current account deficits within reasonable limits ever since (under 1%
of GDP). Foreign exchange reserves also accumulated reaching comfortable levels today
(nearly $6 billion and growing).

Trade liberalization will put a lot of people out of work, creating hardship.
Undoubtedly, industries that were protected will contract with liberalization, putting
people who work in them out of work. But as the experience with trade reform in the
1990s showed, the exporting industries will create jobs at a faster pace than jobs are
destroyed (the protected industries were not very labor intensive to begin with). The
challenge is to help the displaced workers find these newly-created jobs, or other jobs.
This is not an argument against trade liberalization; it is an argument for manpower
training and employment matching services. Trade openness heightens competition and
creates winners and losers. Typically, number of winners far outweighs losers. Some
mechanism to compensate losers is a good strategy. Finally, the long run beneficial
impact of competition on efficiency and productivity must not be ignored.

Agriculture will suffer. On the contrary, agriculture benefited from the trade
reforms of the 1990s because cheaper inputs were available. Farmers produced those
goods that were competitive in world markets, and stopped producing those that weren’t.
The economy imported agricultural products that were cheaper to import.

In short, most of the arguments against trade reform in Bangladesh are either
logically inconsistent or at variance with the facts about the experience with trade policy
in Bangladesh. Yet, these are the only arguments heard in the debate about trade. The
puzzle is why no one has come forward with a public position in support of further trade
liberalization, especially with so much empirical evidence to support it. The next section
attempts to solve this puzzle.

III. Bridging the Gap between Evidence and Rhetoric

The reality is that Bangladesh economy today is much more open than it was in
the 1970s or 1980s. The economy as a whole and the manufacturing and services sector,
in particular, have reaped the benefits of such openness, if one goes by growth rates or
employment creation. One might ask where the impetus for these policy changes came
from. In reviewing the history of trade reform in Bangladesh, it is difficult to ignore the
fact that the earlier strategy of import protection led to a series of foreign exchange crises
that required drastic policy intervention, particularly in early 1991. Trade policy reforms
in Bangladesh benefited from analytical research and diagnostic work done at the Bank
on Bangladesh’s trade regime with particular focus on ensuring export competitiveness.

Between 1975 and 1985 the major thrust in industrial and trade policy reform was
supported by thirteen World Bank adjustment credits (Import Program Credits) worth
nearly a billion dollars. Apart from ensuring a minimum level of essential imports and
alleviating balance of payments difficulties, these credits eventually supported some
macroeconomic and institutional reforms, along with further government initiatives in

16
market orientation of agriculture. This resulted in modest relaxation of the import regime
and the beginning of private sector dominance in industry with a shrinking role of the
public sector.

Next, the refinements in industrial policies along with another round of relaxation
in the import regime came in 1987 with broad-based trade policy reforms that covered
tariffs and QRs. Government policy at the time favored removal of QRs and replace them
with tariffs, as the principal instrument of protection. To that end, the first step was to
replace the “positive list” with a “negative list” (which later became the Control List in
IPO 1989-91) before the latter could be pared down. In addition, this adjustment loan
supported, for the first time, a number of broad-based reforms in the direction of trade
liberalization, such as:

• Rationalization of the tariff structure by reducing tariff variation and reducing the
range of effective protection among different products;
• Phasing out quantitative restrictions and streamlining import procedures; and
• Complimenting tariff rationalization with tax reforms to generate alternative
revenue sources.

The second, and perhaps, more intensive phase of import liberalization and export
orientation was introduced under the Second Industrial Sector Credit (ISAC-2) in 1992.
This intensive phase of trade liberalization was preceded by a package of programs that
put in place basic measures to achieve macroeconomic stability prior to undertaking trade
policy reforms.

ISAC-2 brought in substantial import liberalization, tariff reduction, as well as

deregulation of the investment regime during 1992-95. In fact, the entire package was
rather ambitious to the point of being unrealistic in so far as implementation was
concerned. It involved too many ministries and government agencies that were required
to fulfill stringent conditions. In consequence, the program had to be aborted midway.

Between 1996 and 2002, government policy was geared to sustain the trade
liberalization process, albeit at very modest levels. Nevertheless, as can be seen in Figure
2, protectionist pressures remained strong. Soon, para-tariffs emerged to compensate for
the reduction of protection via custom duties. A World Bank study [World Bank (2004)]
showed that Bangladesh had the most protected economy in South Asia.

It was not until 2003 that the Government re-visited the status of the trade policy
agenda and decided to move forward. Addressing para-tariffs and removing the
remaining trade-related QRs were the main thrust of policy in the trade area at this time.
The resumption of a modest liberalization agenda was not without the usual skepticism
and downright resistance, which essentially came from the following groups:

• Tax Department, which is worried about revenue loss

• The domestic industrial lobby which would like trade protection to protect their
profit from competition

17
 • The labor union which is worried about job loss in the short-term
• The economist profession, of whom the more vocal group tends to be left-leaning.

The net result is that, after prolonged protection to many industries, a wide range
of import substitute products (e.g. biscuit and bakery products, prepared foods, paper
products, batteries, plastics, and so on) still retain very high protection. That Bangladeshi
consumers continue to be taxed by this protectionist stance is rarely heard in popular
discourse.

Before concluding this section, it would be useful to review the salient features of
a typical trade policy reform package supported by the Bank. It usually has four
components:

• Import-related: Policies such as eliminating or reducing quantitative restrictions and

other non-tariff barriers, reducing import tariff levels and dispersion, making tariff
regimes more uniform and transparent, reducing import surcharges, improving import
procedures, and eliminating official reference prices.
• Export-related: Policies related to making imports available for exporting; reducing
export bans, taxes, and licenses; reducing the anti-export bias; export credit and
financing; and other export incentives.
• Exchange rate and foreign exchange management: Moving toward market-
determined exchange rates, exchange rate devaluation, or step adjustment, and away
from the administrative allocation of foreign exchange.
• Industrial and other supporting policies: Pricing reform, investment promotion,
competition policy, marketing, privatization, labor markets, and safety nets.

Beginning from the mid-1980s through the mid-1990s, this was essentially the
package of trade and industrial policy measures that were pursued by Bangladesh
policymakers, either in whole or in part. Though these policies were part of what became
known as the Washington Consensus, they were in line with mainstream thinking of
economists on trade policy. They were also consistent with the literature on trade and
development which argued for policies to back up trade reforms (Box 1). As Nicholas
Stern (2002) once asserted, there was nothing wrong with the principles underlying the
Washington Consensus; what was more important were issues that were left out. For
instance, it said nothing about governance and institutions, the role of empowerment and
democratic representation, the importance of country ownership, or the social costs and
the pace of transformation.

Was Bank advice to Bangladesh in line with the guidance of mainstream

economic policy research? A quick review of the reform agenda underlying several IDA
projects between 1985 and 1995 would show that trade reforms were not initiated in
isolation but accompanied several complementary reforms that were identified in the
literature.

First, research suggested the importance of macroeconomic stabilization to

complement trade reforms as a prior requirement. Trade reforms in the presence of

18
serious macroeconomic instability could have disastrous consequences. In 1990-91, the
Government initiated steps to restore macroeconomic stability in the Bangladesh
economy, which had run into fiscal and balance of payments problems by the end of the
1980s, thanks to the political upheaval that brought an end to a dictatorial regime coupled
by the disastrous floods of 1988.

Second, it is crucial that complementary measures such as removing marketing

and price distortions as well as competition policy, strengthening financial
intermediation, reducing labor market rigidities, and improving the regulatory
environment accompany trade reforms. In principle, this was achieved through the series
of reforms envisaged in the NIPs and adjustment loans for financial sector reforms which
opened up the industrial and investment regime to both domestic and foreign investors.
Deregulation, privatization and removal of wage rigidities were envisaged through
ISAC2 compliance requirements. Full compliance with the entire package of substantial
changes, however, appeared to be a pipe dream, underscoring the political and other
difficulties associated with their implementation.

Finally, what gave rise to the most potent critique of trade liberalization policies
in Bangladesh was its inadequate attention to the poverty and distributional outcomes,
including labor market dynamics. The development community has since recognized the
crucial importance of compensating losers, of which there will be many – enterprises that
fail to adjust, raise productivity and become competitive; and workers in such firms.
Failing to explicitly take care of the (possibly temporary) pain of those who lose out
turned out to be a major weakness in project design.

IV. Concluding Remarks

This paper has attempted to answer a puzzle: Why is trade reform, which has
generated so many benefits to Bangladesh, still so unpopular? After reviewing the
history of trade policy and the current state of debate in the country, we proposed a
reason: most, if not all, trade reforms were driven by external actors, especially the
World Bank. That these reforms were externally introduced makes people suspicious,
even if the evidence that they generated benefits is compelling.

The gap between rhetoric and reality on trade policy in Bangladesh is more than a
curious puzzle. Inasmuch as Bangladesh has the most restrictive trade regime in South
Asia, if not the world, there are important policy issues to be resolved. There may be
good reasons for Bangladesh to approach trade reform cautiously, but none of them has
been proffered in the current debate. By identifying the source of the puzzle, we hope we
can shift from rhetoric to a more evidence-based debate on trade policy in Bangladesh.

19
References

Ahmed, S. and Zaidi Sattar (2004). “Trade Liberalization, Growth and Poverty
Reduction: The Case of Bangladesh,” Chapter 4 in Annual Bank Conference on
Development Economics 2003, joint publication of The World Bank and National
Council of Applied Economic Research.

Bhagwati, J. N.(1978). Foreign Trade Regimes and Economic Development: The

Anatomy of Exchange Control. Cambridge, Mass.: NBER/Ballinger.

Bhagwati, J. N.(1979). Anatomy and Consequences of Trade Control Regimes. National

Bureau of Economic Research, New York.

Bhagwati, J. N. and T.N. Srinivasan (1975). Foreign trade regimes and economic
development: India, NBER, Columbia University Press, New York and London.

Bhuiyan, A.R. and Rashid, M.A. (1993), Trade Regimes and Industrial Growth: A Case
Study of Bangladesh, Bureau of Economic Research, Dhaka University and
International Center for Economic Growth, San Fransisco, USA.

Greenaway, D. and Chris Milner (2003), “Effective Protection, Policy Appraisal and
Trade Policy Reform”, The World Economy, Vol. 26 (4), pp. 441-456.

Government of Bangladesh (1982). The New Industrial Policy. Ministry of Industries.

Government of Bangladesh (1986). The New Industrial Policy. Ministry of Industries.

IDA (1987). Development Credit Agreement. Industrial Sector Project. Credit No. 1816
BD.

Keynes, J.M. (1936). The General Theory of Employment, Interest and Money. London:
Macmillan.

Khondker, B.H. and S. Raihan (2007). “Poverty Impacts of Remittance and RMG in
Bangladesh. A Computable General Equilibrium Analysis”. Draft paper prepared
for the World Bank study of Poverty Assessment based on HIES 2005.

Krueger, A.O. (1978). Liberalization: Attempts and Consequences. National Bureau of

Economic Research, New York.

Krugman, P. (1986). Strategic Trade Theory and the new International Economics. MIT
Press, Cambridge, Massachusetts.

MacEwan, Arthur (1971), “Contradictions in Capitalist Development: the Case of

Pakistan”, Review of Radical Political Economics, Vol. 3, No. 1, 40-57.

20
Mahmud, W. (2001). Adjustment and Beyond: The Reform Experience of South Asia,
Basingstroke, UK, and New York: Palgrave-MacMillan in association with
International Economic Association.

Ministry of Finance (1997). Bangladesh Economic Review 1997. Economic Adviser’s

Wing, Finance Division.

Prebisch, Raul (1950), The Economic Development of Latin America.

Rahman, S.H. (1992). “Structural Adjustment and Macroeconomic Performance in

Bangladesh in the 1980’s,” Bangladesh Development Studies, Special Issue,
Volume XX, No. 2&3.

Rahman, S.H. (1994). “Trade Policy and Industrialization in Bangladesh: An

Assessment”, in G.K. Helleiner (ed.) Trade Policy and Industrialization in
Turbulent Times: Routledge.

Rodrik, D. and A.M. Mayda (2002). “Why are some people (and countries) more
protectionist than others?” Mimeo, Kennedy School of Government, Harvard
University, Cambridge, Massachusetts.

Singer, Hans (1949), Post-War Relations between Under-developed and Industrialized

Countries.

Stern, Nicholas (2002). “Dynamic Development: Innovation and Inclusion”. Paper

presented at Munich Lectures in Economics, Center for Economic Studies,
Ludwig Maximilian University, Munich.

Rashid, S. (2007). http://www.thedailystar.net/story.php?nid=10804. Comments on the

World Bank report, “Bangladesh: A Strategy for Sustained Growth”.

Taslim, M.A. (2004). “Trade Policy of Bangladesh: 1990-2003”, published in Souvenir

of the Dhaka University Economics Department Alumni Association (DUEDAA)
Annual Meeting.

World Bank (1990). Bangladesh Import program Credits IX-XIII and Overview of Policy
Reforms in the 1990s. Program Performance Audit Report. Operations Evaluation
Department. Report No. 8871.

World Bank (1993). The East Asian Miracle: Economic Growth and Public Policy.
Washington DC: World Bank.

21
World Bank (1996). Second Industrial Sector Adjustment Credit. Implementation
Completion Report. Report No. 15866. South Asia Region, Private Sector
Development and Finance Division.

World Bank (2004). Trade Policies in South Asia. An Overview. Report No. 29949,
Poverty Reduction and Economic Management, South Asia Region.

Yilmaz, K. and S. Varma (1994). “Trade Policy Reforms in Bangladesh” ISS Working
Paper #28, Industry and Finance Division, South Asia Region, World Bank.

22
 Annex -I

Table 6. Bangladesh 1991/92-2007/08: Unweighted Average Protective Import Duty Rates

All tariff lines Industrial tariff lines Agriculture tariff lines
Customs Para- Total Customs Para- Total Customs Para- Total
Duty tariff protection Duty tariff protection Duty tariff protection
rate rate rate
1991/92 70.64 2.98 73.62 69.72 3.44 73.16 76.64 -0.01 76.63
1992/93 57.93 2.59 60.52 57.34 2.99 60.33 61.83 -0.03 61.80
1993/94 43.47 2.43 45.90 43.13 2.84 45.97 45.58 -0.17 45.41
1994/95 34.24 3.30 37.55 33.52 3.54 37.06 37.49 2.23 39.72
1995/96 28.70 3.26 31.96 28.40 3.47 31.87 30.07 2.28 32.36
1996/97 28.24 3.38 31.61 27.79 3.58 31.37 30.25 2.48 32.73
1997/98 27.27 5.88 33.15 26.80 5.98 32.78 29.42 5.42 34.83
1999/99 26.59 5.82 32.41 26.23 5.92 32.15 28.19 5.37 33.56
1999/2000 22.40 6.99 29.39 21.86 7.33 29.19 24.87 5.41 30.28
2000/01 21.10 7.43 28.54 20.39 7.84 28.23 24.53 5.46 30.00
2001/02 21.02 8.41 29.43 20.28 8.47 28.75 24.60 8.15 32.74
2002/03 19.91 6.51 26.42 19.08 6.74 25.82 23.85 5.44 29.29
2003/04 18.82 10.29 29.11 18.02 8.81 26.82 22.56 17.22 39.77
2004/05 16.31 10.19 26.50 15.63 9.72 25.35 19.64 12.46 32.10
2005/06 15.50 10.99 26.49 14.70 10.58 25.28 19.39 13.01 32.40
2006/07 14.85 9.42 24.27 13.99 8.87 22.86 19.03 12.11 31.15
2007/08 15.91 6.03 21.93 15.11 5.46 20.58 20.14 9.04 29.18
Source: NBR import and tariff database; staff estimates

23
 Annex-II

Average MFN tariffs in South Asia compared with average tariffs in some
other large and medium size developing countries
Average tariff (%) Rank Date Year
Slovak Republic 22.1 1 2002
Bangladesh 21.9 2 2007
Morocco 19.4 7 2005
Egypt 18.9 9 2005
Iran 18.7 11 2004
India 15.0 16 2007
Nepal 14.7 22 2005
Pakistan 14.6 25 2007
Tunisia 13.4 32 2005
Ghana 13.2 33 2004
Vietnam 13.2 34 2005
Brazil 12.3 48 2005
Colombia 11.9 51 2005
Nigeria 11.6 53 2005
Sri Lanka 11.3 57 2005
Argentina 10.6 60 2005
Thailand 10.6 61 2005
China 9.2 70 2005
Mexico 9.2 71 2005
Korea, Rep. 9.0 76 2004
South Africa 8.5 79 2005
Malaysia 7.5 87 2005
Indonesia 6.5 96 2005
Philippines 5.4 104 2005
Chile 4.9 110 2005
Turkey 2.4 132 2005
Mean 12.2
Source: WDI, World Bank staff estimates

24