The Trade Policies of Developing Countries by amronline

VIEWS: 166 PAGES: 51

									The Trade Policies of
Developing Countries
The Trade Policies of
Developing Countries
  Recent Reforms and
   New Challenges

         Sarath Rajapatirana

                  The AEI Press
   Publisher for the American Enterprise Institute
   W A S H I N G T O N , D.C.
Available in the United States from the AEI Press, c/o Publisher
Resources Inc., 1224 Heil Quaker Blvd., P.O. Box 7001, La Vergne,
TN 37086-7001. Distributed outside the United States by arrange-
ment with Eurospan, 3 Henrietta Street, London WC2E 8LU

ISBN 0-8447-7152-X

© 2000 by the American Enterprise Institute for Public Policy
Research, Washington, D.C. All rights reserved. No part of this pub-
lication may be used or reproduced in any manner whatsoever with-
out permission in writing from the American Enterprise Institute
except in cases of brief quotations embodied in news articles, critical
articles, or reviews. The views expressed in the publications of the
American Enterprise Institute are those of the authors and do not
necessarily reflect the views of the staff, advisory panels, officers, or
trustees of AEI.

The AEI Press
Publisher for the American Enterprise Institute
1150 17th Street, N.W., Washington, D.C. 20036

Printed in the United States of America

Acknowledgment                                       iii

Trade Trends in Developing Countries                 2

What Led to Trade Policy Reforms?                    5

The Political Economy of Trade Reforms               12

Reform Trends in the 1990s in a Twenty-Country Sample 14

Issues Arising from the International
  Trading Environment                                22

Extending the Trade Reforms of the 1990s             28

Conclusions                                          32

Notes                                                35

References                                           37

About the Author                                     41



The author is grateful to Claude Barfield, American Enterprise
Institute, for his incisive comments on an earlier version of the
paper; to Gokce Ozbilgin, Smith College, for her excellent research
assistance; and to James Morris and David Stetson for their superb
editorial advice.

                                             SARATH RAJAPATIRANA ✜ 1

        he failure of the Seattle ministerial meeting of the World Trade
        Organization (WTO) to launch a new round of multilateral
        trade negotiations brings into sharp focus the increasing divide
between developed and developing countries in their respective
agendas for reforming the international trading environment. The
developing countries need a genuine and well-defined agenda for
moving the reform process forward. While there is no unanimity
among developing countries with regard to the particular items that
belong on the agenda, there are common areas of interest that are dif-
ferent from the agenda of the developed countries. This essay
attempts to set the background for a new round of multilateral nego-
tiations, though it may take place later than was thought possible on
the eve of the Seattle meeting.
    The 1980s saw nearly all the developing countries reform their trade
regimes, improve their economic performance, and provide increased
access to exports from developed countries. Many developing coun-
tries joined the WTO, and many others are now waiting in the wings
to join. The 1990s saw a continuation of the reform trend of the 1980s,
despite a few pauses such as the 1994 payments crisis in Mexico and
the East Asian crisis that began in 1997. The implementation of the
Uruguay Round of multilateral trade negotiations since the mid-1990s
gave a definite fillip to trade liberalization in developing countries, as
the decade saw the intellectual case for free trade receive new support.
But there were also deflections from the mainstream position. New
challenges to trade liberalization in developing countries have arisen
both from the need to complete the agenda of their domestic reforms
and from developments in the international trading environment.
    Research in the 1970s and 1980s strengthened the case, both the-
oretical and empirical, for the more-open trade policies that eventu-
ally led to widespread trade liberalizations. The research gives us a
basis to judge the trade regimes that have come into being in the
1990s. Despite reforms, those regimes are still far from the ideal
regime of low levels of protection and low variance in protection,
based on tariffs and characterized by transparency. A new round of
multilateral trade negotiations is needed to address the unfinished


agenda of domestic trade reforms and to meet the new challenges to
liberal trade policies emanating from the international trading envi-
ronment. This essay deals only with the main items on that unfin-
ished agenda.
   We first discuss the trade trends of developing countries in the
1990s in a global context. We then consider the factors that led to
reforms, the political economy aspects of trade reforms, and, using a
twenty-country sample, the main trends in reform at the country
level. Next, we examine issues in the international trading environ-
ment that impinge on developing countries, the main issues that
emerge from the reforms, and the new challenges to liberal trade that
arise from the international trading environment. Before presenting
our conclusions, we identify a future agenda for reforms for the next
round of multilateral trade negotiations.

             Trade Trends in Developing Countries

During 1990–1997, the volume of world trade (export and import
volumes) grew by 6.5 percent per year, and the exports of develop-
ing countries grew by 8.7 percent per year. Although the East Asian
crisis in 1997 led to a reduction in export growth in that region, the
growth rate of exports from developing countries in the first seven
years of the 1990s was substantial. World output grew at 2.3 percent
per year over the 1990–1997 period, and developing country output
grew at 3.1 percent per year (World Bank 1999). Developing
economies also became more integrated into the world economy. The
estimates of trade openness in developing countries (as indicated by
their trade ratios) show an equally good performance (Drabek and
Laird 1998).
    World exports of commercial services grew by 8 percent per year
during 1990–1997. Starting from a low base, the share from develop-
ing countries grew faster than the world rate of growth of commercial
services (World Trade Organization 1998). In 1997, developing coun-
tries accounted for 30 percent of the world’s exports of commercial
services. The “other commercial services” category, which includes
financial services, construction, and computer services, was the
fastest-growing category of services exports. World services exports
slowed in 1997, in part because of the economic crisis that began in
East Asian countries, which were the main importers of commercial
services in the developing world. Given the further liberalization of
                                             SARATH RAJAPATIRANA ✜ 3

services trade following the Uruguay Round (1995) and the two min-
isterial meetings that took place prior to 1998, the opportunities for
faster growth in world trade in services will be even greater. Once
East Asia recovers from its crisis, developing countries will be able to
make use of the opening-up of services trade.
    The average volume in commodity and services trade during the
1990s masks many differences in the individual performances of
countries and regions. The East Asian countries led the way in the
growth of exports and permanently changed the patterns of trade
between developed and developing countries. First, manufactured-
goods exports from the developing world, and from East Asian coun-
tries in particular, were the fastest-growing component of trade. They
averaged increases of some 12–15 percent per year during the 1990s,
thereby continuing a trend that started in the late 1960s when East
Asian countries liberalized their trade regimes. Second, there was a
decline of trade in mining products from developing countries after
the decline in oil prices. The previous decade had begun with the oil
shock of 1979–1980, when oil prices rose to unprecedented levels.
Third, there was an increase in intraregional trade among developing
countries, attributable to unilateral liberalizations and to the reduc-
tion in trade barriers against one another on a preferential basis—that
is, the institution of lower tariffs and the granting of greater market
access for fellow members of regional trading agreements (RTAs).
Latin America led the developing world in rejuvenating and creating
RTAs. Africa had the slowest growth in trade in the 1990s, because
of its slower liberalization and because its commodities were con-
centrated in primary products.
    World trade slowed in the three years preceding 1998, primarily
because of trade contraction in East Asia but also because of the
decline in commodity prices, which fell sharply in 1997–1998 and
began to recover in early 1999. The price declines were so sharp that
primary product prices fell to 50 percent of their peaks for the first
time since World War II. The declines were somewhat moderated in
1999 with the recovery of oil prices. In terms of regional groups, the
most severely affected were the African and Middle Eastern
economies, because of their concentrations in primary products. The
prices of services exports fell to their lowest level in dollar value since
1983, and the prices of all the major commercial services, including
transport and travel, declined. It is worth noting, however, that the
price declines are a transitory phenomenon and that prices are sure

to recover with the revival of the East Asian economies. Indeed, evi-
dence of a recovery in the region began to emerge in 1999.
    The trade policies of developing countries in the 1990s differed
from those of the 1980s in terms of their dominant themes and the
policy experiences of different groups of countries. Writing on the
main features of the trade policies of the 1980s, Anne Krueger (1990)
concluded that “it is by no means clear that the trade policies of devel-
oping countries are any more restrictive in 1987 than they were at
the end of the 1970s!” For most developing countries, trade regimes
are more open in the late 1990s than at any time in the post–World
War II period.
    There is no better way to demonstrate the trade trends in devel-
oping countries than to contrast the themes of the 1980s (Krueger
1990) with the emerging themes of the 1990s. First, a comparison of
the trade regimes of the 1980s with those of the period 1950–1980
shows increased differentiation in trade policies among the develop-
ing countries. Many East Asian countries took the lead in liberalizing
their trade regimes, while nearly all other developing countries main-
tained restrictive regimes. In the 1990s, there is less differentiation,
because nearly all of the developing countries have begun to liberal-
ize their regimes.
    Second, a long-term trend toward less-restrictive trade regimes
was observed in the 1980s. In the 1990s, trade reform has become a
unifying goal, varying only according to the speed of liberalization.
The response to reform has differed among countries depending on
the strength and credibility of their initial reforms and the different
circumstances within individual countries.
    A third aspect of the reforms of the 1980s saw developing coun-
tries become more integrated with the world economy, in both trade
and capital flows, than in the previous three decades. This process of
integration into the world economy continued in the 1990s.
Measured by trade-intensity indexes (the ratio of the sum of exports
and imports to GDP), developing countries are more closely integrated
into world trade than ever before. They are also closely integrated into
the world capital market. In fact, the amounts of both short- and long-
term capital flowing into developing countries in the 1990s are four
times the levels of the 1980s, which was a period characterized by
debt problems. It is beyond the scope of this essay to explore the con-
sequences of integration into the world capital market, but we can
observe that manufactured exports from developing countries have
                                            SARATH RAJAPATIRANA ✜ 5

become closely linked to private foreign direct investment in the
export sectors. That became possible with the reduction of bias
against exports through trade liberalization and the creation of a more
hospitable environment for private foreign direct investment in most
of the developing countries, thanks to the reform of their regulatory
regimes in the 1990s.
    Finally, in the 1980s, real oil prices reached a peak, and mining
products were a large share of the total exports of developing coun-
tries. In the 1990s, many developing countries moved from being
exporters of primary products (agricultural and mining products) to
becoming exporters of manufactures on a wider scale than before.
That was a consequence of the changed economic environment of the
1990s and the decline in the real price of oil, which fell to its lowest
levels since the early 1970s and reduced oil exports as a component
of total exports. In the mid-to-late 1990s, the prices of agricultural
exports also fell, thereby increasing the relative value of manufactures.
The manufactured exports of developing countries significantly
increased their share of the total exports of developing countries—from
17 percent in 1980 to 24 percent in the 1990s.
    Despite the East Asian crisis of 1997, the past decade has been less
turbulent than the 1980s. Many of the countries that were in des-
perate straits in the 1980s have staged recoveries thanks to reforms
and to the return of private foreign direct investment. The irony is
that some of the East Asian countries that were setting a strong pace
for export growth, such as Malaysia, South Korea, and Thailand, fell
victim to their own success in the late 1990s. The crisis was attrib-
utable to weaknesses in their financial systems, to inadequate vigi-
lance in the conduct of macroeconomic policies (which led to the
appreciation of real exchange rates), and to the recession in Japan,
which reduced the demand for their exports.

                What Led to Trade Policy Reforms?

The 1980s were the decade of initiating reforms; the 1990s, of
continuing them. Although there was less of an urgent need to under-
take reform in the 1990s than in the 1980s, many developing countries
did so and fashioned more-liberal trade regimes. We shall now con-
sider the reasons for their actions.
   The 1980s were a decade of crises, starting with the second oil
shock of 1979–1981, a debt shock that began when Mexico was

unable to service its debt, the rise in interest rates as U.S. monetary
policy was targeted to reduce double-digit inflation, and the ensuing
worldwide recession. Those negative shocks led to crises in the
majority of developing countries (save the oil exporters). The crises
were not wholly external in origin. In many cases they were home-
grown after consumption and investment booms in the mid- to late
1970s. Whatever the origin of the crises, they led to reforms, which
are now termed “new liberalizations” (Little et al. 1993; Lal and
Snape, forthcoming). One of the principal reasons for the reforms was
the breakdown of coalitions that supported the existing protectionist
regimes. Policymakers found it easy to push through reforms when
the alternative became increasingly more costly. In my sample of
twenty countries, there is only one—Colombia in 1991—in which
reforms were undertaken without crisis. The impetus for change in
Colombia came when Cesar Gaviria assumed the office of president.
Deeply committed to change, he compressed a planned four-year
program of trade reform into eighteen months (Rajapatirana, de la
Mora, and Yatawara 1997).
    New research has strengthened the intellectual case for continu-
ing and consolidating trade reforms in the 1990s. For example, the
creation of endogenous growth models by Romer (1986), Lucas
(1988), and others permitted an analysis of policy reforms that had
not been possible within the neoclassical growth model. Again, the
empirical research associated with Dollar (1992), Sachs and Warner
(1995), and Edwards (1997), among others, built on the work done
by Little, Scitovsky, and Scott (1970), Balassa and Associates (1971),
Bhagwati (1978), and Krueger (1978).
    In the 1990s, some rejected the line of argument that liberalized
trade regimes based on neutral incentives were at least partially
responsible for the very rapid growth in exports and high GDP
growth in the East Asian countries. Amsden (1989), Wade (1990),
Lall (1996), and Rodrik (1992), in particular, have argued that the
selective promotion of specific sectors and well-targeted public
investment led to economic success. The principal agreements and
disagreements between the mainstream researchers and the revision-
ists may be stated as follows. Both sides generally agree that macro-
economic fundamentals such as low inflation, fiscal stability, and
competitive exchange rates are essential for the proper conduct of
trade policy. The revisionists also agree with the mainstream econo-
mists that education and the development of human capital, as well
                                             SARATH RAJAPATIRANA ✜ 7

as basic infrastructure, are needed for development. The revisionist
view focuses on the interpretation of the success of East Asian
economies and claims that the success can be attributed not so much
to neutral incentives as to selective government intervention. This
view emphasizes the importance both of acquiring technological
mastery as the key to successful industrialization and of building the
institutional structure needed to conduct sound industrial policy.
Revisionists consider selective intervention the means to address dis-
tortions and believe that cooperation between the government and the
private sector is essential for industrial success. They also consider the
mainstream economists’ belief in the success of neutral policies to be
an ideological commitment to laissez faire rather than a representation
of the real world. In addition to the individuals mentioned above, pro-
ponents of the revisionist view include some, such as Singer (1950),
who were ardent advocates of the import substitution strategy in the
    The revisionists gained ground in the debate about the factors
responsible for the success of East Asian economies. The main point
of contention was that, in addition to getting the fundamentals right,
the East Asian countries selectively intervened in their economies to
depart from neutral incentives. Those who espouse the revisionist
view insist that the success of the East Asian economies (prior to the
1997 crisis) was due to dynamic factors. These factors include learn-
ing by doing (associated with the adoption of new technology) and
externalities (benefits that one firm creates for others but that do not
result in compensation to the firm that creates the benefits), neither
of which can be realized by relying on market forces alone. In addi-
tion, the revisionists claim that the success of East Asian economies
is attributable not to neutral incentives but to factors that are not yet
fully understood. In the opinion of some scholars, the East Asian
Miracle study (World Bank 1993) departed from the well-established
view within the World Bank that neutral incentives were more likely
to lead to better economic performance than were selective interven-
tions and supported the view that the correct balance of laissez faire
and intervention led to the East Asian miracle. However, there is noth-
ing in the study that indicates how this correct balance is to be
achieved. In addition, the study—written before the East Asian crisis—
could not account for the fact that at least part of the crisis originated
in the attempt to promote specific industries by using the financial sys-
tem to subsidize those industries. In the event, the financial sectors

became weak and vulnerable to macroeconomic shocks. Dani Rodrik
(1994) contends that the East Asian Miracle study helped accommo-
date the view that the state and the market could cooperate to pro-
duce a superior outcome. But the debate does not rigorously define
neutral incentives and selective promotion approaches. The main
issue is to determine which type of regime provides the best oppor-
tunity for rapid development: a regime of neutral incentives that
allows comparative advantage to assert itself, or a regime that needs
selective promotion to achieve outcomes superior to those that neu-
tral incentives could provide.
    Alice Amsden (1989) and Robert Wade (1990) have been among
the most ardent recent advocates of using the East Asia arena as the
basis for asserting the superiority of the selective intervention
approach. The main argument used by Amsden is that the newly
industrializing countries do not fit the old paradigm of innovative
industrialization in developed countries—that the latter, in fact, are
special cases.
    Wade contends that “governing the market,” in which government
deliberately distorts prices to promote certain activities, is the domi-
nant explanation for the success of East Asian economies. He agrees
that the free-market and “simulating free-market” explanations can
also be used to explain that success, but that such explanations are
not as valid as those based on governing the market. Like Amsden,
Wade claims that the newcomers operate in a different paradigm from
the earlier industrializing countries, and he caricatures neoclassical
explanations, as do many of those who argue for selective interven-
tion. The caricatures portray the alleged neoclassical position as being
“get prices right, and everything will follow.” Amsden and Wade say
that the neoclassical position is not relevant to developing countries
because the concern should be not with static but with dynamic com-
parative advantage.
    Sanjaya Lall (1996), another advocate of selective intervention,
bases his position on the need to promote technology in developing
countries. Although he says that incentives are important for indus-
trialization, he advocates government intervention to induce firms in
developing countries to acquire technology, and he alleges that neo-
classical economists ignore micro-level dynamic efficiency based on
the acquisition of technology and skills.
    Many of the arguments that have been leveled against the alleged
neoclassical position of the 1970s and 1980s are based on a highly
                                             SARATH RAJAPATIRANA ✜ 9

exaggerated and distorted version of the neoclassical approach to the
use of neutral incentives. The neoclassical position does not claim
that outward-oriented strategy is the necessary and sufficient factor
for efficient industrialization, and many of its proponents discuss the
importance of property rights, infrastructure, sound macroeconomic
policies, and an institutional environment that is supportive of indus-
trialization. Where outward orientation has failed, many of those sup-
portive factors were absent.
    In addition, there has been much confusion about the use of neu-
tral trade policies and laissez faire. Many of the revisionists, including
Amsden and Wade, claim that the neoclassical position is untenable
because it is based on a very limited role for the government. All of
them assume a benevolent and omniscient government. They argue
that the role of the government should not be limited, due to volatil-
ity in the markets and to the need to support infant industries; they
also argue that coordination of the different activities of the private
sector is not just desirable, but is in fact essential for raising GDP
growth rates.
    The East Asian experience (save that of Hong Kong), they claim,
demonstrates that large and interventionist governments using selec-
tive promotion were the cause of industrial success. That position is
in sharp contrast to the literature on public choice theory, which
argues that government is an economic entity like any other, trying
to maximize its gains (Krueger 1974), and is not the benevolent en-
tity that revisionists or the neoclassical economists of the past make
it out to be. Government is not a guardian angel but, at times, a
predator. The literature on rent seeking and revenue seeking finds
that governments can be captured by different elements of the pri-
vate sector to serve their own ends, and that the opportunity to
receive selective support for different activities encourages capture by
different interest groups.
    But neither mainstream economists nor policymakers have
accepted the position of the revisionists, and some revisionists have
begun to mute their advocacy of selective intervention in the wake of
the East Asian crisis. One reason for the weakened financial systems in
East Asia was the use of financial markets to target different activities
selectively. This was a major cause of the weaknesses in the commer-
cial banks’ portfolios that precipitated the crisis: selective intervention
took the form of targeted credit subsidies to some enterprises, and
these subsidies compromised the financial viability of the banks,

which were not encouraged to evaluate the financial risks when mak-
ing loans but were instead pressured by governments to make loans
to certain enterprises.
    In addition to the revisionists’ failure to influence policymakers—
which led to strong import liberalization (a departure from the revi-
sionists’ alleged East Asia model)—there was a strong negative lesson
from the collapse of the Soviet model that had been emulated by
countries like India. The demise of that planned, state-led, and highly
inefficient closed-economy model may have had a salutary effect on
countries that were not previously prepared to commit to liberalized
trade in one form or another.
    The negotiations that preceded the signing of the Uruguay Round
were another important factor in facilitating trade reforms in the
1990s. Preparation for the round moved trade issues to center stage,
brought an external force to bear on the domestic policies of devel-
oping countries, and committed the countries to reducing trade bar-
riers and binding their tariffs. Though domestic crises created the
impetus for trade reform, the Uruguay Round provided a mechanism
to reduce barriers and to link reform to an external guard against slip-
page. To be sure, many aspects of the round could have served a lib-
eral trade environment better, but it did create the atmosphere for
trade reform in the 1990s.
    Prior to the 1990s, developing countries were reluctant to partic-
ipate in previous rounds and preferred to “free ride” when developed
countries reduced barriers on a most-favored-nation basis. They
invoked “special and differential” treatment as the reason not to enter
into reciprocal negotiations with developed countries. That stance
changed in the 1990s, for several reasons. First, given the increas-
ingly strong intellectual argument that liberalized trade was good for
developing countries, multilateral trade negotiations based on reci-
procity became attractive. Second, developing countries realized that
they would lose out by not participating and by forgoing an oppor-
tunity to influence the agenda for negotiations. Third, they realized
that special and differential treatment (in which developing countries
did not have to reciprocate when they received access to markets in
developed countries) was not beneficial. In fact, such treatment had
prevented them from using external pressures as a rationale for trade
policy changes; consequently, many developing countries had con-
tinued their highly restricted trade regimes, much to their own detri-
ment. In the same vein, developing countries had received
                                          SARATH RAJAPATIRANA ✜ 11

concessions through the Generalized System of Preferences, under
which these countries could have zero or low tariffs for their prod-
ucts in industrial country markets. This access was not bound, which
meant that the developed countries could withdraw the concessions
at will. Fourth, because many developing countries had become
important players in world trade, particularly in manufactures, devel-
oped countries would no longer allow them to “free ride.”
Consequently, developing countries joined GATT, and later the WTO,
in larger numbers than ever before. By 1998, the WTO had some 132
members, and some 32 countries were in the process of negotiating
entry. All of the potential new members are developing countries and
transition countries. In short, the 1990s saw an unprecedented
increase in the participation of developing countries in the multilat-
eral trading system; they accepted its laws and processes, and they
attempted to secure greater access to the world market on a bound
    Another notable stimulus to trade liberalization in the 1990s was
the large-scale adoption of flexible exchange rates by developing
countries. This move allowed countries to remain competitive inter-
nationally. Changes in domestic prices came to be reflected in exchange
rates and thus removed a constraint on trade liberalization that had
existed in the 1960s and 1970s. The period since 1984 can be
described as a time of flexible exchange rates. In fact, very few trade
liberalizations were associated with fixed exchange rates.
    Finally, the international financial institutions (the World Bank
and the International Monetary Fund), along with GATT and the
WTO, continued their support for trade reform. The World Bank had
led the way in the 1980s, but its support declined in the 1990s with
the decline in structural adjustment lending. In contrast, the IMF has
increasingly included trade policy conditions in its standby programs
(IMF 1998). These programs allow countries with balance of pay-
ments difficulties to undertake a smoother adjustment by financing
a part of their deficits using IMF resources. Access to these resources
is conditioned on a program of actions designed to address the bal-
ance of payments disequilibrium. The creation of the WTO, the
inclusion of agriculture and services within multilateral trade disci-
pline, the establishment of the trade review mechanism, and the
newly reformed dispute-settlement mechanism have all contributed
to a more favorable environment for liberalizing trade further on a
most-favored-nation basis.

             The Political Economy of Trade Reforms

To understand the timing of trade reforms in the late 1980s and the
early 1990s one must consider the reforms in the context of political
economy. A new literature emerged in the 1980s that formalized the
view that trade policymaking was subject to political influences. This
literature recognized that the existence of trade protection in the
1960s and 1970s reflected the political equilibrium of that time:
powerful and influential groups within developing countries favored
protectionist trade policies. Those groups differed from country to
country and over time in different country settings. They appropri-
ated the power to make trade policy and passed on the costs of pro-
tection to the society, largely to the agricultural and rural sectors. The
prevailing ideologies of the time gave intellectual support to protec-
tion as the way that developing countries could industrialize.
    The political economy framework must be set within the context
of the demand for, and the supply of, protection. The supply side of
trade policymaking involves the trade policymakers’ preferences
(infused by ideology) and the institutional structure of government
(democratic or authoritarian, multiparty versus biparty). The demand
side involves individual preferences and organizations that lobby for
protection. From the research of Mancur Olson, it is well known that
the costs of protection are dispersed across a large part of the popu-
lation, while the benefits of protection accrue to a much smaller
group of individuals and firms; those who benefit from protection
have the ability to organize and lobby against the removal of protec-
tion (Olson 1982). In the 1960s and 1970s the protectionist lobbies
held sway; they demanded and received protection. The equilibrium
that existed was somewhat stable and was dominated by different
groups in different countries, such as the military and labor in
Argentina, the industrialists and bureaucrats in India, and the oli-
garchs and industrial interests in Mexico.
    In the late 1980s and the 1990s, the political economy equilibrium
changed in the developing countries, for several reasons. First, there
were large macroeconomic shocks, including oil and debt shocks,
that upset the balance of power among the different groups. Second,
in some parts of the world, particularly in Latin America, there was
a return to democracy, which led to the exercise of majority choices
and the emergence of dynamic leaders—both within the existing
milieus, as in Colombia and India, and, more dramatically, within
                                            SARATH RAJAPATIRANA ✜ 13

changed political milieus, as in Argentina, Brazil, and Peru. Third, as
noted earlier, market-favoring ideologies took hold because of the
failure of command economies and the visible success of market-
based economies.
    The large economic shocks that turned into crises and unsettled
the existing political equilibrium were the common factor in all of the
trade liberalizations (with the possible exception of the reform in
Colombia, which liberalized under stable macroeconomic conditions
in 1991 after the election of President Gaviria).
    An examination of the trade liberalization experiences of these
countries reveals that many populist (that is, democratically elected)
governments were prepared to undertake nonpopulist reforms—
such as trade reforms—because the cost of populist policies had
begun to exceed the cost of not adjusting to the macroeconomic
shocks. The crisis factor seems to be ubiquitous, and the research of
Bates and Krueger (1993), Haggard and Webb (1994), and Little,
Cooper, Corden, and Rajapatirana (1993) confirms that crisis was
indeed a prerequisite for reform.
    The theoretical literature on political economy supports the crisis
hypothesis. Alesina and Drazen (1991) have developed a model in
which fiscal adjustment, the primary element in overcoming a crisis,
is delayed because of the uncertain distribution of the burden of
adjustment. A crisis allows one group to dominate and to carry
through the reform. More often than not, trade reforms are tacked on
to fiscal reforms, because the costs of not adjusting to macroeconomic
reforms are higher. By contrast, the absence of a macroeconomic crisis
delays trade reforms. There might exist a seeping or a quiet crisis, as in
India, but the macroeconomic crisis in the country has to grow before
the introduction of trade reforms.
    Other models, such as that of Fernandez and Rodrik (1991),
explore a system that reinforces inaction to maintain the status quo.
Given the narrow time horizons of politicians and the uncertain dis-
tribution of the benefits of reform, the equilibrium of the status quo
can be broken only when a crisis becomes so severe that reform is
the only recourse. Small crises generally preserve the status quo.
Severe crises can be cathartic. All theoretical models postulate the
presence of a distributive constraint—the cumulative pressures and
costs of nonadjustment leading to the emergence of a group that can
undertake the reforms. In Latin America, for example, severe crisis
and strong leadership have been the principal factors behind trade

reform (Rajapatirana, de la Mora, and Yatawara 1997).
   Thus, both the empirical evidence and the theory argue that the
political economy of trade reform supports crisis-driven reforms.

                 Reform Trends in the 1990s in a
                     Twenty-Country Sample

In this section we identify the main trends in trade policy reforms in
the 1990s within twenty individual countries (table 1). The countries
in the sample are from four continents, and the sample is suffi-
ciently diverse to represent developing countries in general. It
includes the relatively rich developing countries (“middle-income,”
as defined by the World Bank), such as Argentina, Chile, South
Korea, and Venezuela, and poor countries, such as India, Kenya,
Nigeria, Sri Lanka, and Pakistan. It also includes countries with large
populations, such as India, Indonesia, and Brazil, and less populous
countries, such as Bolivia and Costa Rica. There are countries that are
relatively more industrialized, such as Brazil, Mexico, and South
Korea, and countries that are mostly agricultural, such as Côte
d’Ivoire, Kenya, and Thailand. Three countries of the twenty are oil
exporters: Cameroon, Indonesia, and Mexico.
   The time period chosen to identify the trends in developing coun-
try trade policies is 1985–1996, because only by extending the period
back to the mid-1980s can we clarify the trends of the 1990s. As we
have seen, many countries began trade reform in the 1980s in response
to crises. Many others had begun even earlier, in the 1970s, and some,
such as South Korea, in the 1960s. Thus, to have limited the time
period for the sample to the 1990s would have made it impossible to
judge accurately the main trends of the decade.
   Many countries in the sample undertook reforms in the 1990s that
continued reforms begun in the 1980s (the reform years are indicated
in parentheses): Argentina (1991–1993), Brazil (1990–1992),
Cameroon (1990–1994), Colombia (1990–1991), Costa Rica (1992),
Côte d’Ivoire (1994), India (1991–1992), Kenya (1989–1995), Nigeria
(1995), Pakistan (1994), Thailand (1990), and Venezuela (1996).
   Korea was the earliest of the strong reformers. Five other countries
undertook reforms in the 1970s and 1980s that qualify those countries
not only as early reformers but also as strong reformers, in comparison
with the mild and weak reformers identified in the table. Early reform-
ers in the strong reformer group are Bolivia (1985), Chile (1974), Korea
                                                        SARATH RAJAPATIRANA ✜ 15

    Table 1 Trade Reform in Twenty Countries, 1985–1996
                                  Reform                                Overall
                                   Years                               Outcome
                                                               Average          GDP
                                                                 GDP           growth
                  Before                         Type of       growth           post-
Countries         1985         1985–1996         reforms      1985–96         reforma
Bolivia             ---         1985–87             S           3.26              3.70
Chile            1974–79        1992–93             S           7.03              6.70
Colombia         1984–87        1991–92             S           4.23              4.70
Korea            1966–71           ---              S           8.44              8.60
Mexico              ---         1985–90             S           2.03              2.50
Sri Lanka        1977–80        1989–90             S           4.40              5.10
Turkey             1980         1985–86             S           4.73              4.40
  Mean                                                          4.87              5.10
  SD                                                            2.19              2.01
  (Mean/SD)b                                                    2.23              2.54
Argentina          1976         1991–93             M           2.17              3.30
Brazil             1968         1990–92             M           2.94              3.20
Indonesia           ---         1986–89             M           7.19              8.00
Thailand            ---         1988–91             M           8.72              8.50
  Mean                                                          5.26              5.75
  SD                                                            3.20              2.89
  (Mean/SD)                                                     1.64              1.99
Cameroon            ---         1989–93             W          -0.61              -0.70
Costa Rica          ---         1986–89             W           3.85               3.90
Côte d’Ivoire       ---         1986–89             W          2.15               2.00
India            1980–85        1991–92             W           5.91               7.00
Kenya               ---         1991–92             W          3.72               2.90
Morocco          1984–85           ---              W           3.96               3.70
Nigeria          1973–74        1986–90             W           4.59               4.80
Pakistan         1972–73        1987–88             W           5.41               4.70
Venezuela           ---         1989–90             W           2.58               2.80
  Mean                                                         3.51               3.46
  SD                                                            1.96              2.13
  (Mean/SD)                                                    1.79               1.62
Note: S = strong, M = mild, W = weak

a. Estimated with one-year lag after initiation of reforms.
b. Mean/variance = coefficient of variation.
Source: World Bank 1997 database.

(1966), Mexico (1985–1990), Sri Lanka (1977), and Turkey (1980).
Bolivia and Chile enjoy the distinction of having introduced reforms
that amounted to complete regime changes and that led to low pro-
tection and very low variance in protection. Bolivia has a two-tier tar-
iff system of 5 percent and 10 percent, and Chile has a 10 percent
uniform tariff. In both countries quantitative restrictions (QRs) apply
to less than 1 percent of their imports. The six countries in this group
of strong and early reformers have low protection—in the 11 percent to
13.5 percent range—and low variance in tariffs (measured by tariff dis-
persion); QRs apply to less than 10 percent of their tariff lines. All six
countries entered the 1990s with liberalized trade regimes. Colombia
undertook reforms in 1991–1992 and joined the group of strong
reformers. (It should be noted that the coefficient of variation for GDP
growth is higher for strong reformers than for the other two groups.)
    The mild reformers defined in the sample are Argentina, Brazil,
Indonesia, and Thailand. With the exception of Indonesia, all of the
countries in the group undertook their trade reforms in the early 1990s.
Indonesia attempted to cope with the sharp decline in oil prices by
means of a package of reforms that began in 1986. Thailand began the
reform process in 1988, but the more substantial reforms came in the
early 1990s. For the group as a whole, tariffs are between 13.5 percent
and 33 percent. Tariff dispersion ranges from zero to 80 percent. QR
coverage is between 11 percent and 30 percent of their imports.
    The third group, made up of what we characterize as weak reform-
ers, comprises the largest number in the sample: Cameroon, Costa
Rica, Côte d’Ivoire, India, Kenya, Morocco, Nigeria, Pakistan, and
Venezuela. Like the other two groups, this one includes countries that
began the reform process in the mid-1980s. The reforms were limited
in scope and were implemented slowly by comparison with the
action of the strong reformers. The levels of protection in these coun-
tries, as measured by nominal tariffs, range from 19 percent to 70
percent, and tariff dispersion from 19 percent to 134 percent. QRs
still cover a large part of the tariff lines. QR coverage is as high as 30
percent for India and 50 percent for Pakistan.
    Table 2 shows the main features of the reforms during the period
1985–1996. The reforms are noteworthy in several respects. First, the
most common reform was to reduce barriers against exports as a first
step. This type of reform included the reduction and elimination of
export duties and the provision of duty-free status to imported inputs
used in the production of exports, either through duty exemptions
                                                                     Table 2 Content of Trade Reform in Twenty Countries, 1985–1996
                                                   Trade Regime before Reformsa                                                Trade Regime after Reformsb
                               Country              AT         TD          CQR             Reform Content                      AT          TD          CQR                Comments
                           Argentina (1986)       42 (p)     0–115          60    Reduced barriers on exports and              --        0–40           43      Liberalized its trade regime in 1991.
                                                                                  streamlined QR coverage.
                           Bolivia (1985)         12 (m)      5–10          <2    Streamlined import and export               7.5        0–10          <2       Liberalized its trade regime in 1985.
                                                                                  procedures and reformed customs.
                           Brazil (1987)           51        0–100          39    Reforms mostly on exports.                  14.3       0–35           20      Liberalized its trade regime during
                           Cameroon (1988)          --          --          --    Reduced export taxes.                       18.8         --           43      Reforms during 1990–94.
                           Chile (1985)            35           0           <2    Promoted nontraditional exports              11          0            <2      Undertook a strong liberalization in
                                                                                  and improved export procedures.            (1991)                             1974–79.
                           Colombia (1984)         61        0–220          83    Provided imported inputs for exports        43         0–100          63      Undertook a strong reform in
                                                                                  and automatic access to foreign                                               1990–91 and reduced AT to 12.5%.
                           Costa Rica (1984)      53 (p)     0–1400         --    Replaced QRs with tariffs and reduced       11.2       5–20          <10      Exchange controls rescinded in 1994.
                           Côte d'Ivoire (1985)    26        25–350         --    Reduced export taxes and QRs on             15         5–30          <10      Stronger liberalization following
                                                                                  imports and abolished reference                                               devaluation of CFA franc in 1994.
                           India (1989)            128       0–300          70    Reformed export incentives; removed         71           --           32      Reforms in 1991–92. Consumption
                                                                                  import restrictions on imported                                               goods mports remain restricted.
                           Indonesia (1986)        35        0–225          20    Removed restrictions on exports; reduced    25         0–70           13      Main reforms 1990–93.

                           Kenya (1985)            53        0–100          40    Import licenses reduced. Export              45          --           60      Reforms 1989–95.
                                                                                  compensation system replaced               (1989)                   (1989)
                                                                                  by duty drawback.

                                                                                                                                                               Table continues on next page.
                                                                                        Table 2 Content of Trade Reform in Twenty Countries, 1985–1996 (cont’d)
                                                                          Trade Regime before Reformsa                                                                Trade Regime after Reformsb
                                                  Country                  AT             TD            CQR                   Reform Content                         AT             TD            CQR                      Comments
                                              Korea (1988)               18.1              --           15.5        Korea's second reform package                    8.3             --            <2          First liberalized in 1965–67; second
                                                                                                                    1977–83.                                                                                   episode in 1977–83. By 1994.
                                                                                                                                                                                                               AT=7.9% and CQR<2%.
                                              Mexico (1985)              24 (p)         0–100            92         Reduced tariffs and QRs; eliminated             12.5             --           17.3         Greater liberalization after joining
                                                                                                                    reference prices.                                                                          NAFTA (1991).
                                              Morocco (1985)               60           0–150            90         Removed restrictions on exports and on           45           0–100             --         By 1995 AT=23%. Liberalization
                                                                                                                    imports used for exports. Reformed                                                         possible after abandoning fixed
                                                                                                                    temporary admission regime.                                                                exchange rate in 1994.

                                              Nigeria (1986)               50           0–100             --        Abolished import surcharge. Export              34.3           0–60             --         1993 AT=32.8%. By 1995 CQR=88%.
                                                                                                                    procedures simplified. Improved duty                           (est)
                                                                                                                    drawback system.
                                              Pakistan (1986)              69           0–140            80         Increased export subsidies, bonded               65              --           14.5         AT=35% in 1994. Remains protected
                                                                                                        (est)       warehouses, and studies.                                                      (est)        country.
                                              Sri Lanka (1990)             20          10–114             --        Reduced import duties and tariff                 12            0–50            <5          Main liberalization in 1977.
                                                                                                                    dispersion.                                    (1985)          (est)

                                              Thailand (1986)              23           0–100            30         Main reforms 1988–91.                           11.4             --            5.5         AT=27% (1993), 17% (1994).
                                                                                         (est)                                                                     (1990)
                                              Turkey (1985)              31.4              --             --        Liberalized agriculture imports and               9              --            2.7         Liberalized in 1980.
                                                                                                                    improved procedures.                           (1993)
                                              Venezuela (1988)             37           0–135            40         Reduced import tariffs and CQR.                  19            0–50            10          Adopted the Andean Group common
                                                                                                                                                                   (1991)                                      external tariff 5, 10, 15, 20 in 1995.
                                              Note: AT = average nominal tariff (average unweighted ad valorem rate unless otherwise noted), p = production weighted, m = import weighted, TD = tariff dispersion,
                                              CQR = coverage of quantitative restrictions (percentage of import items), -- = not available.

                                              a. AT, TD, and CQR are shown for the year before trade reforms.
                                              b. AT, TD, and CQR are shown for the year following trade reforms.

                                              Sources: World Bank and GATT-WTO trade policy reviews.
                                           SARATH RAJAPATIRANA ✜ 19

or duty drawbacks. A second feature of the reforms that came early
on was the replacement of QRs with tariffs. Of course, some coun-
tries—India and Pakistan in the 1990s, and Argentina, Brazil, and
Colombia until 1992—retained QRs for a large part of their imports.
A third feature, which emerged in the 1990s, was the reduction of
import tariffs. Cameroon and Côte d’Ivoire, the two CFA countries1
in the sample, were able to undertake reforms following their deval-
uation of the CFA franc in 1994; Morocco, although not a CFA coun-
try, followed suit. Finally, there were institutional reforms in trade
policy making and in the administration of customs and of export
and import procedures.
    We can make several generalizations, based on the sample, about
the trade reforms and trade regimes of the 1990s—and, in so doing,
we can begin to shape an agenda for future reforms.
    First, although the 1990s saw most countries make further
progress with reforms begun in the 1980s, the trade regimes that have
evolved still have different rates of protection for different goods,
leading to a wide variance in protection. It is true that many coun-
tries have liberalized their trade regimes and that protection is lower
than at any time in the last five decades. But all of the countries in
the mild and weak reform groups need to undertake a substantial
agenda of additional reforms if they are to move toward the ideal lib-
eralized trade regime. If Chile is the yardstick, because it has low pro-
tection, low variance, and hardly any QRs (and thus is close to the
ideal trade regime identified in research), the other nineteen coun-
tries have a long way to go. To be competitive, countries must reduce
their levels of protection as compared with those of other countries—
they must not be satisfied with reductions that diminish protection
relative to their own previous levels of protection. Those in the strong
reformers group have less far to go to equal Chile than those in the
mild and weak categories. Chile resolved to reduce its uniform tar-
iffs to 7.5 percent in 1996 and then postponed action on the deci-
sion; it has now resolved to adopt a 6 percent uniform tariff by 2003.
Similarly, Sri Lanka had decided to move toward a uniform 15 per-
cent tariff in 1998, but it has now postponed the change indefinitely.
The reason these countries have deferred further liberalization is
unclear. But the delay indicates a slowing down of the reform effort,
even among the leaders.
    Second, variance in protection may have increased in some coun-
tries in the 1990s, for several reasons. Many countries have continued

to protect agriculture, even after they, like the developed countries,
have reduced overall protection. A high level of protection for agri-
culture is a key feature of current trade regimes in the developing
countries; Nigeria, for example, has tariffs of over 100 percent.
Others have bound tariffs on agriculture at much higher rates than
those for manufactures. Brazil has bound agricultural tariffs in the
80–230 percent range, while India has bound its agriculture tariffs at
a 0–300 percent range. Many of the rates cited have been reduced
over time, and the applied tariff rates for agriculture are currently
much lower than those bound rates. Yet the potential to raise rates in
the future still exists. In addition, various nontariff barriers are more
prevalent in agriculture than in industry. For some agricultural prod-
ucts, Colombia uses a domestic procurement scheme that treats
imports as a residual, after domestic production is bought up. The
Andean Group as a whole uses a price band system for some agri-
cultural products that imposes a variable levy. That levy can rise to
inordinately high levels, above the bound rates, when the interna-
tional prices of agricultural goods decline—as in fact happened in the
late 1990s. Another reason for the increasing variance in tariffs is the
RTAs. Their tariff preferences have led to zero or low tariffs among
the member countries, thereby widening the variance between both
high and low tariffs and between member- and nonmember-country
import tariffs.
    Third, as described above, most of the reforms in the 1990s were
confined to export rather than import liberalization. A typical
sequence has been to reduce barriers on exports, such as export
duties; to provide duty drawbacks and credit subsidies; and to com-
pensate for tariffs on imported inputs that are used to produce
exports. These measures compensate for the bias against exports
found in many developing countries. Some countries have gone
beyond the point of eliminating the bias against exports by provid-
ing direct subsidies based on performance criteria. Thus in many
instances they do more than provide “free trade status” for exports.
These countries place less emphasis on reducing import protection
and instead attempt to promote exports through subsidies. Brazil,
India, Indonesia, Nigeria, Thailand, Turkey, and Venezuela have followed
this sequence. Venezuela has gone so far as to identify “sector leaders”
and target them for special concessions, such as energy subsidies. A
number of the countries extend credit at concessional interest rates
to small and medium-sized exporters. Under the Uruguay Round
                                           SARATH RAJAPATIRANA ✜ 21

agreement, many of these subsidies granted by countries to their
domestic producers must be phased out. Developing countries have
an eight-year time horizon for phasing out subsidies; developed coun-
tries have a five-year time horizon to do the same. Irrespective of the
country in which they occur, such subsidies distort trade regimes and
hence prevent the best allocation of resources worldwide.
    A bias against exports arises from the prevailing import protection,
and it cannot be corrected through export subsidies without incur-
ring additional costs. Many countries do not appreciate the presence
of the bias (Clements and Sjaastad 1984). Part of the problem can be
traced to the attempt by countries to follow the “Japanese-Korean
model” espoused by the revisionists, which delayed import liberal-
ization for a few years until export success was achieved. This view
presupposes that import tariffs do not harm exports, that a period of
import protection is not only harmless but necessary, and that there
are clever bureaucrats who can implement a system of targeted export
subsidies efficiently and can be immune to corruption in the bargain.2
Recent research and observable experience in East Asia confirm that
a sequence that neglects import liberalization has led to lower total
productivity growth in Japan and Korea (Beason and Weinstein 1996;
Lawrence and Weinstein 1999).
    Fourth, although many developing countries have reduced tariff
barriers and QRs, they have begun to use antidumping on a wide
scale. Of the countries in the twenty-country sample, Argentina,
Brazil, and Colombia have increased their antidumping initiations.
Interestingly, antidumping duties have been imposed within RTAs as
other barriers to trade have been reduced. In their increased use of
antidumping, developing countries are beginning to imitate devel-
oped countries. Australia, the European Union, and the United States
have led the way in the use of antidumping, which has become the
weapon of choice for protection. The Uruguay Round left antidump-
ing open to abuse, even though the round did tighten the circum-
stances in which it could be used. One irony is that some countries
that did not have antidumping, countervailing-duty, or safeguards
laws have introduced them in the 1990s following the Uruguay
Round and have begun to use them as protectionist devices.
Colombia, which had strict antidumping and safeguard laws, made
them less stringent after the Marrakech Agreement of June 1, 1994.
In general, it has become easier for countries to initiate antidumping
and safeguards actions (Rajapatirana 1998).

    Finally, the increasing number of RTAs is an important feature of
trade regimes in the 1990s. Their number has quadrupled since the
1970s. Just over one hundred regional trading agreements have been
notified to the WTO as required by GATT and now WTO rules. Some
agreements are not formally notified to the WTO; they are established
outside the standard rules for regional integration, under the enabling
clause of 1979 (part 4 of GATT), which confers a special concession
to regional agreements formed among developing countries. The
character of RTAs has changed since the 1960s and 1970s when they
were conceived as protectionist devices. In the 1990s, they have
become devices that can restrict trade—not deliberately, but because
of the sheer difficulty of implementing them. The main problem
arises from the rules of origin usually found in free-trade agreements.
To give one example, Bolivia has signed twenty-eight regional agree-
ments with a variety of rules of origin; they give customs officials wide
discretion in classifying imports, and they provide wide opportuni-
ties for corruption and rent seeking.3 All RTAs have the potential for
trade diversion—trade that is generated due to cost differences aris-
ing from the concessions exchanged, rather than from genuine dif-
ferences in the costs of production and transport. Such trade
diversion is welfare-reducing.4

    Issues Arising from the International Trading Environment

The international trading environment has become more important
to developing countries than it was in the past. Reductions in trans-
port costs, the increased participation of developing countries in
global trade, the opening of developed-country markets (as well as
those of developing countries), and increased private foreign direct
investments in the countries have led to the phenomenon now uni-
versally acknowledged as globalization. It has coincided with, and
contributed to, the liberalization of trade policies in developing coun-
tries. Moreover, the increase in private foreign direct investment has
created many investment and production activities that can, and do,
move across borders with much greater ease than before. The com-
parative advantage that determines the investments has a thin edge.
Since technology diffuses rapidly, and capital markets are closely inte-
grated, the variance in costs across countries tends to decline. This
decline increases the sensitivity of domestic trade and investment
regimes to the international trade and investment environment.
                                          SARATH RAJAPATIRANA ✜ 23

    While developing countries were engaged in reform in the period
1985–1996, several changes occurred in the international arena that
are of great importance to the countries’ future role in the interna-
tional trading environment and to their trading relationship with the
developed countries. The first is a growing concern in developed
countries that the greater penetration of exports from developing
countries would reduce real wages in the industrial countries; this
has led to a clamor that labor standards be imposed on developing
countries. Second is the movement in developed countries to impose
environmental standards on developing countries—to harmonize the
standards and to prevent an alleged “race to the bottom” in setting
and observing them. Third is the concern in developing countries
about the fulfillment of undertakings that developed countries gave
in the Uruguay Round, in particular with respect to textiles, clothing,
and agriculture. And fourth is the developing countries’ concern
about the increased use of antidumping and countervailing duties
against their exports.
    The United States and the European Union have espoused labor
standards, some of which are already recognized—such as the
International Labor Organization (ILO) rule against the use of prison
labor for economic activities. In 1998, the board of the World Bank
resolved to include labor standards as a condition in approving loans,
and labor unions have strongly encouraged developed countries to
take an interest in the standards. In the United States there is a fear
that rising imports from developing countries will drive down the
wages of unskilled labor. In Europe the fear is that exports from
developing countries will increase unemployment levels, which are
already high. Some groups in developed countries call for labor stan-
dards on moral grounds: they argue that the absence of standards
leads to child labor and paltry wages for unskilled labor. Thus, the
interests of those who favor labor standards on moral grounds to sup-
port high wages in developing countries converge with the interests
of those who want to protect their real wages by keeping out labor-
intensive products.
    Studies in the United States show that imports from developing
countries have had very little impact on the wages of unskilled labor.
The reason the wages of unskilled labor fell in the 1980s and have
been stagnant in the 1990s is technological change (Krugman and
Lawrence 1993). In fact, Bhagwati (1998) believes that free trade has
had a moderating influence on the decline in real wages of unskilled

labor, because trade leads to the more efficient use of unskilled labor
in the United States. The introduction of labor standards in develop-
ing countries will create a bias against the exports of developing
countries and will prevent both developed and developing countries
from realizing the full gains from trade. Moreover, if labor standards
were formally introduced into international trade rules, they could
easily deteriorate into a protectionist device, with trade unions lob-
bying to raise the standards over time. And yet, ignoring the issue of
labor standards may lead to a worse outcome. It is true that free trade
might be harmful to the real wages of unskilled labor, but protection
is no solution: it will make everyone, including unskilled labor, worse
off. In addition, it will undermine United States leadership, which
has been an important factor in the liberalization of trade worldwide.
Thus all could lose as a result of protections for unskilled labor in the
United States.
    This is not to argue that the welfare of unskilled labor is not impor-
tant, but merely to point out that trade measures are not the appro-
priate means to assist unskilled labor. The proper forum to address
labor issues is the ILO, rather than the WTO. But the better solution
would be for each country to determine its own labor standard
through its political process. In this context, the failure of the Seattle
ministerial to launch a new round of trade negotiations was particu-
larly disappointing, for it was at least partly due to the present United
States administration, which wanted at the last minute to appease
trade unions in an election year by insisting that labor standards
become an important issue for a new round. In the previous two min-
isterials and in the meeting of trade ministers of Latin America, devel-
oping countries have argued against the inclusion of labor standards
as an agenda item for a new round of multilateral trade negotiations.
    Like labor standards, environmental standards have become a
salient issue on the agenda of developed countries. In the United
States, for example, lobbies led by the Sierra Club, among others,
have begun to influence policymaking. Environmental standards
become an international trade issue when economic activities have
cross-border environmental effects; if they are confined within the
borders of a country, they can be addressed through domestic regu-
lation rather than through international trading rules. Domestically,
the issue is one of identifying both positive and negative effects of the
actions of firms on the environment—the positive and negative exter-
nalities as commonly defined by economists—and devising a system
                                           SARATH RAJAPATIRANA ✜ 25

of subsidies and taxes to address the positive and negative external-
ities respectively. With such a subsidy and tax scheme in place, along
with a free-trade regime, countries have two instruments that they
can deploy to achieve two targets. The deployment of these two dif-
ferent instruments to achieve the two goals sought—free trade and a
clean environment—leads to much better progress toward both goals
than use of the trade instrument alone. Trade policy will not serve
either goal well, but it will prevent developing countries from spe-
cializing in labor-intensive products and thus will harm developing
countries, which are much less responsible for global environmental
problems than are the developed countries. Additionally, global trade
issues are unsuited to deal with domestic environmental problems;
in most instances, proper domestic policies (such as tax and subsidy
policies) will also minimize global environmental problems.
    The environmental lobbies join those demanding the harmoniza-
tion of environmental standards across countries. Such harmonization
is conceptually flawed to begin with—and, in addition, environmen-
tal standards, like labor standards, could be captured by protectionist
interests. Harmonization attempts to equate costs across countries. But
resource endowments differ among countries. For example, the costs
of reducing carcinogens through a capital-intensive technology will be
high in a developing country, while a labor-intensive technology for
the same purpose would cost less there. Additionally, tastes can differ
among countries. One country may rank reducing water pollution as
of greater importance than reducing air pollution. Developed coun-
tries are greater polluters than developing countries, given their larger
use of natural resources, so they may end up having to pay more than
developing countries to reduce pollution. Given the different supply
(endowments) and demand (tastes) conditions, the costs of reducing
pollution differ across countries. Harmonization that attempts to
equate the costs does not work. And if it should work (by accident),
the basis for trade is removed. Hence, policies that try to harmonize
environmental standards deny countries their gains from trade, irre-
spective of their level of development.
    Empirical research on Latin America shows that liberalized
economies have less pollution than those with restricted trade
regimes (Birdsall and Wheeler 1993). One reason is that the machin-
ery used for production in liberalized economies is more up-to-date
and pollution-reducing than older machines that do not incorporate
the new pollution-minimizing technology. In addition, a liberalized

trade regime is apt to use more price-oriented approaches to dealing with
a problem like pollution and to employ the “polluter-pays” principle.
    Developing countries worry that obligations entered into by devel-
oped countries, related to the Uruguay Round, may not be fully met.
Laws relating to these obligations may be passed in developed countries,
but the actual implementation may not take place. The developing coun-
tries’ concern in this regard centers on implementation of the Textiles
and Clothing Agreement, on protection for agriculture, and on the
increased use of antidumping and countervailing-duty actions by devel-
oped countries.
    The Textiles and Clothing Agreement brought the Multifiber
Arrangement under GATT-WTO discipline and was intended to end
a system of bilaterally “agreed” quotas with developing countries.
Bringing textiles and clothing under WTO scrutiny is a major
achievement, but the benefits that developing countries will reap
from this change will be delayed. The liberalization of trade is to take
place in four stages, and only in the last stage, in the year 2005, will
trade be completely liberalized. Meanwhile, the items that are being
liberalized by developed countries are those that are least competi-
tive (i.e., very low value added textile and clothing items), and the
gains they may bring to developing countries are therefore limited.
In addition, developed countries are using the transitional safeguard
measures provided under the agreement to keep developing coun-
tries from the competitive export of textiles and clothing. The Textile
Monitoring Body created to supervise the implementation of the
agreement suffers from the old GATT problem of acting only in con-
sensus, which means that the implementation of the agreement is
slow and countries are not subject to any sanctions if obligations are
not met.5
    The Uruguay Round also caused agriculture to be brought under
international trade discipline. To begin with, domestic subsidies and
quantitative restrictions were “tariffied” at high rates, which led to an
increase in the protection of agriculture after the Uruguay Round. It
is true that agriculture was liberalized to the extent that prices were
substituted for nonprice restrictions such as QRs. However, the trade
gains from that liberalization are far from being realized. Agricultural
liberalization will lead to gains for exporters from developing coun-
tries and losses (through terms-of-trade effects) for the importers of
agricultural goods in developing countries. But all will gain in the
long run, as the world’s resources are better allocated.
                                           SARATH RAJAPATIRANA ✜ 27

    Meanwhile, developed countries have been using the special pro-
vision of tariff quotas to continue protecting agriculture in areas
where there could be further liberalization. The Uruguay Round
allowed the gradual reduction of protection in agriculture through
the use of tariff quotas. Countries are allowed to apply temporary
quotas to a part of their imports within a given commodity group,
and the quotas are reduced over time to allow tariffs to determine the
level and composition of imports. Since many developed countries
carry over unfulfilled quotas across commodity groups within agri-
culture and from one year to another, the liberalization of imports of
agriculture is delayed further. While this is not WTO-illegal, it is
against the spirit of liberalizing agriculture trade. Moreover, agricul-
ture subsidies persist within the “green box” category (those allowed
under environmental grounds) and lead to the persistence of agri-
culture protection. Developing countries are concerned that these
measures amount to trade protection rather than environmental pro-
tection. Developing countries also protect agriculture, as we have
seen above. There is no reason for them to have a special concession,
apart from the longer period of time they have been accorded to end
all export subsidies, including those for agriculture. Though they
seem not to use the “green box” category, developing countries retain
high levels of protection for agriculture.
    A final issue of concern to developing countries is the increased
use of antidumping and countervailing duties against their exports in
developed countries. Goods produced in developing countries tend
to be labor-intensive; marginal costs are therefore a large proportion
of their average costs. Productivity enhancements and shifts away
from labor-intensive production processes can yield reductions in
marginal costs for developing countries, but antidumping measures
are often invoked to challenge such reductions. Developing coun-
tries are therefore particularly vulnerable to accusations of dumping;
this is especially true because of current WTO law and procedures
for calculating “constructed costs,” under which it is relatively easy
to allege that there has been dumping even when dumping has not
occurred. In addition, incentives exist for a country that is bent on
imposing antidumping duties to construct costs that will support its
case and that will lead to a finding that dumping has indeed occurred.
The initiation of antidumping actions of itself has a protective effect,
because dumping duties can be imposed while a case is pending.
Moreover, developing countries lack the human and financial

resources that are required to challenge antidumping actions. They
may agree to reduce exports in lieu of a continuation of antidump-
ing actions. A similar situation holds for countervailing duties, even
though it may be difficult to show that a public subsidy has been
involved in the production and export of a good. Developing coun-
tries themselves have begun to imitate developed countries in the use
of antidumping, and the practice has now become more widespread
than ever (Guasch and Rajapatirana 1998).

            Extending the Trade Reforms of the 1990s

The foregoing concerns help to shape agendas, both domestic and
international, that will affect how developing countries carry forward
the trade liberalization of the 1990s.
The Domestic Agenda. Let us begin by citing the actions to which
developing countries should assign priority on their domestic agendas.
    1. They must reduce protection toward the level of Chile. A 10 per-
cent uniform tariff would be ideal and would be within reach of the
majority of the developing countries. Reducing variance in protection
is as important as reducing the level of protection, for wide variances
interfere with resource allocation. There is now evidence that the
reduction of tariffs, when combined with a devaluation and the replace-
ment of QRs with tariffs, does not lead to a reduction in tariff revenues
in most cases (the loss of revenue is a common reason given by devel-
oping countries for not reducing tariffs). As a first step in reducing tar-
iffs, countries need to consider reducing the bound rates to applied
tariff levels and replacing QRs with tariffs. There is also little justifica-
tion on balance-of-payments grounds to impose QRs (GATT article
19), for the simple reason that they do not work. Instead, they lead to
inefficiencies and rent seeking. Only a regime based on tariffs can
ensure that variance in protection is kept within known limits. In both
these moves, namely replacing QRs with tariffs and reducing the vari-
ance in protection, the average level of protection could remain the
same; such moves are easy to accomplish and send a signal to eco-
nomic agents that a country is on the path of reform and that its pro-
tection levels have become predictable. It is noteworthy that, despite
the East Asian crisis of 1997, Korea and Indonesia have not intro-
duced new trade restrictions. Instead, they have resolved to reduce
protection in the near future. Some countries (Argentina, Brazil,
                                           SARATH RAJAPATIRANA ✜ 29

India, and Thailand) did increase protection when they confronted
negative shocks in 1994 and 1997. The increases have been relatively
small and have not changed the character of their trade regimes.
    2. New evidence confirms that reducing barriers on the export side
while maintaining import protection does not realize the full benefits
of a trade liberalization (Lawrence and Weinstein 1999). Imports pro-
vide an efficient and significant channel for learning and for acquir-
ing technological knowledge. Before the Uruguay Round, there were
no strong strictures against export subsidies. That allowed countries
to liberalize on the export side while maintaining import restrictions.
This alternative may not be available in the future, given the chal-
lenges to export subsidies through countervailing duties and the
dispute-settlement mechanism. While developing countries have
been given eight years (as compared to five years for developed coun-
tries) to eliminate export subsidies, the longer the subsidies remain
in place, the longer the misallocation resulting from such subsidies
will persist.
    3. As noted earlier, antidumping has increased in the 1990s, and
developing countries have begun to use it more than ever before
(although the main users of the measure are still developed countries,
led by Australia, the European Union, and the United States).
Argentina initiated fifteen antidumping cases in 1997, Brazil eleven,
India thirteen, Korea fifteen, and Malaysia eight, as reported to the
WTO (WTO 1998). Except in the case of predation—and there is lit-
tle or no opportunity for predation today—there is hardly a sound
economic argument to be made in favor of antidumping. As a tem-
porary measure, safeguards are certainly preferable, given that they
are imposed on a most-favored-nation basis. Safeguards are time-
bound and subject to a stricter procedure for initiation. Developing
countries could adopt safeguards in place of antidumping.
    4. The best way to minimize trade diversion within RTAs is to
lower tariffs on nonmember countries, to keep those arrangements
open to new members, and to move toward common markets rather
than free-trade areas. In theory, common markets do not require rules
of origin, because they rely on a common external tariff. However,
many common markets have used differential rules of origin as tem-
porary measures until they reached agreement on common external
tariffs. To that extent, the adverse effects of rules of origin are found
in common markets too. There is also the danger that the need to
agree on a common external tariff could lead to higher tariff rates than

an individual country would be willing to adopt on its own. In addi-
tion, further liberalization becomes more difficult if the stronger
members of the common market resist it. A short-term palliative for
regional arrangements is to adhere to article 24 and to avoid use of
the “escape clause” under which developing countries are exempt
from article 24 provisions.
The International Agenda. If the full benefits of unilateral trade lib-
eralization are to be realized, four issues in the international trading
environment, as it affects the trade policies of developing countries,
need attention. The issues are difficult because they are at the heart
of the political economy processes at work in developed countries.
    1. It would indeed be a step backward for the liberalization
process to introduce labor standards into international trade rules.
Labor standards could be handled through the ILO, which has the
responsibility to ensure that the more egregious practices, such as
forced and prison labor, are not used. Prohibiting child labor may
mean greater deprivation, and even starvation, for children. The bet-
ter way to help poor children is to supply poor countries with the
resources they need to ensure at least minimum levels of education
and health for the children. That approach serves both free trade and
the children’s welfare.
    2. The optimal policy for dealing with the environment is to
employ two instruments to achieve two goals. Free trade can be com-
patible with a clean environment through the use of domestic regu-
lations that make polluters pay. Where there are cross-border
pollution issues, the best solution is market-based and requires pol-
luters to buy permits or pollution “rights.” Those rights can be bought
and sold to create a market—an international analogue of the domes-
tic “polluter-pays” principle. The real problem is how best to allocate
pollution permits. The permits must be given to those countries that
have had low pollution and have not consequently “used up” their
“quotas” of pollution, as developed countries have. Developing coun-
tries could become sellers of the permits, and developed countries
the buyers.
    Developing countries argue for allocation on the basis of popula-
tion; developed countries, on the basis of income level. Such differ-
ences should be resolved through negotiation. Witness the allocation
of special drawing rights (SDRs) in the 1970s, when the principle of
combining population, income level, and trade openness was used to
                                           SARATH RAJAPATIRANA ✜ 31

distribute SDRs among the countries. A similar approach may be used
to construct “pollution balance sheets” on a consistent set of princi-
ples: countries that have already used up a part of their quota will get
less than those whose use has been limited. This approach would
require a round of negotiations and compromise. The main point
here is that the exercise must not be related to trade rules. The United
Nations Environmental Program, not the WTO, is the appropriate
forum to negotiate and to implement allocation principles for pollu-
tion permits.
    3. Three ministerial meetings on the implementation of the
Uruguay Round have already taken place. The third ministerial meet-
ing, which took place in Seattle in late November 1999, turned out
to be a debacle. This was largely due to the Clinton administration’s
last-minute sop to labor unions and environmental lobbies. That
action helped protectionist forces to gather strength and to resist the
call for a new round of multilateral trade negotiations to carry for-
ward the agenda identified at the end of the Uruguay Round and in
the earlier two ministerial meetings. These two ministerials had cre-
ated a process to analyze and exchange information on implement-
ing the Uruguay Round and to identify items for a new round. That
process would be the proper forum in which to discuss the Textiles
and Clothing Agreement and to handle some of the problems that
have arisen with the implementation of the agreement (such as the
transitional safeguard provisions and the inappropriate use of the tar-
iff quota system that was intended to ease implementation). The dif-
ferences between the developed and the developing countries can be
reduced by undertakings to be made by both sides. After all, both
sides protect their textile and clothing industries.
    A similar approach will benefit agriculture. The Uruguay Round
left agriculture protection even higher than it had been. The analysis
and information-exchange process has already led to some discussion
of how best to proceed. The goal is clear: to reduce agriculture pro-
tection to a level similar to that of manufactures, sooner rather than
later, so as to reap sooner the benefits of reduced protection. To be
sure, views will differ, even among the developing countries, because
of the different interests of the exporters and the importers of agri-
cultural goods. Nevertheless, a negotiation would identify those differ-
ences and would build coalitions across countries and interest groups
to reduce protection. The Cairns Group (a group that met in Cairns,
Australia, comprised of both developed and developing countries,

including Australia, the United States, and Argentina) was singularly
successful in getting agriculture onto the Uruguay Round agenda. The
same group could repeat its success with an expanded membership.
    4. The increased use of antidumping and countervailing duties
demands attention. The WTO rules have to be tightened with respect
to calculating the costs of production and closing the current loopholes.
But the main improvement would lie in replacing antidumping with
safeguards and in tightening the rules for the measurement of public
subsidies to limit countervailing actions. Here again, the ministerial
meetings could be used to highlight the problems, and the analysis and
information-exchange process could bring the issue before the next
round of multilateral trade negotiations.

    To resolve many, if not all, of the foregoing agenda issues (both
domestic and international) would require a new round of multilateral
trade negotiations, notwithstanding the failure of the Seattle minister-
ial meeting to launch a new round. Such a negotiation would bring
trade issues to the fore again and would provide opportunities to
address, for example, lowering of the levels and variance in protection;
eliminating export subsidies earlier than promulgated by the Uruguay
Round; reforming antidumping and countervailing-duty rules; chang-
ing the law with respect to the “escape clause” in RTAs; evaluating labor
and environmental standards; and completing the Uruguay Round
issues that affect the continued protection of textiles, clothing, and agri-
culture. In addition, a new round of negotiations should discuss top-
ics that go beyond the “built-in agenda” of the Uruguay Round and are
of particular interest to developing countries because they are matters
in which the developing countries enjoy a comparative advantage.
    Some leading developing countries, such as India, have argued that
a new round is not warranted until the Uruguay Round agreements are
fully implemented (Hedge 1998). They are the very countries that were
initially less than enthusiastic about the Uruguay Round. But to delay
multilateral trade negotiations is to delay the achievement of higher
growth and living standards through further trade liberalization.


Trade regimes in developing countries were more liberal in the late
1990s than at any time since World War II. Yet theoretical and empir-
ical work on trade in the 1990s reveals that the benefits would be
                                           SARATH RAJAPATIRANA ✜ 33

greater still if developing countries maintained more-liberalized trade
regimes than they now do.
    Developing countries would do well to reduce protection and its
variance, to rely on tariffs, to avoid imitating the protectionist stance
of developed countries in their use of antidumping and the contin-
ued protection of agriculture, and to embrace most-favored-
nation–based trade over regional arrangements.
    Beginning a new round of multilateral trade negotiations sooner
rather than later would best serve the agenda for reform. The nego-
tiations would facilitate further liberalization and would allow devel-
oping countries to realize their objectives in many of the areas of
interest to them. In the 1980s and 1990s, the macroeconomic crises
that many countries experienced were the principal stimulus to
reform, and both theoretical and empirical aspects of the political
economy of reforms support a crisis-driven hypothesis. But crises are
not an efficient way to generate reforms; hence the need for an exter-
nal impetus, such as multilateral trade negotiations. Country-to-
country or multilateral negotiations do not generate optimal
outcomes, for they may be confined to limited sectors. Even multi-
lateral trade negotiations limited to a few sectors would not be opti-
mal, because they would advance the trade interests of only the
more-powerful countries, which would take the opportunity to
“cherry pick” from an agenda the items that might not be as benefi-
cial to developing countries. As soon as developed countries accom-
plished their goals through this process, they would be unwilling to
negotiate measures to reduce barriers in areas of interest to develop-
ing countries.
    Of course, developing countries could liberalize their trade
regimes unilaterally—and some have already done so. (Chile is the
prime example.) That strategy remains the best option for devel-
oping countries. But multilateral trade negotiations, such as the
proposed Development Round or Millennium Round (whatever
the politicians choose to call it), would allow developing countries
to make gains over and above those of unilateral trade liberaliza-
tion, and would allow them greater latitude as well in addressing
items of interest to them. Further, such negotiations would help to
check domestic protectionist interests, which may resist further
    Multilateral trade negotiations are a more efficient mechanism
than crises to induce reforms, because the give-and-take of the nego-

tiations, if properly conducted, allows all parties to benefit, and to a
greater measure than any one of them could achieve unilaterally. The
failure at Seattle should be seen as only a temporary setback, and it
should not be allowed to spoil the case for a new round.
                                          SARATH RAJAPATIRANA ✜ 35


    1. Exchange rates in CFA (francophone) countries are fixed to the
French franc. The countries have access to foreign exchange
resources from the Bank of France to meet balance-of-payments
deficits and are subject to rigid rules of monetary and fiscal conduct
in exchange for this access.
    2. This view is attributed to Amsden, Wade, Lall, and others, as
well as to the World Bank’s East Asian Miracle study (1993).
    3. The extent of the choices can be gauged by the fact that, with
some 7,200 tariff lines, twenty-eight different agreements with dif-
ferent rules of origin, and three phases of implementation, the theo-
retical choice set exceeds 600,000! While a yard of cloth cannot be
classified as a live animal, the set remains large enough to give great
discretion to customs officials.
    4. Yeats (1997) demonstrated that trade diversion was likely in the
automotive sectors in the Mercosur agreement.
    5. Of course, developing countries cannot be fully absolved of
their conduct. Textiles and clothing are the among the most pro-
tected goods in developing countries, however much they have
liberalized their trade regimes.

                                                REFERENCES ✜ 37


Alesina, Alberto, and Allan Drazen. 1991. “Why Are Stabilizations
   Delayed?” American Economic Review 81(5): 1170–88.
Amsden, Alice. 1989. Asia’s Next Giant: South Korea and Late
   Industrialization. New York: Oxford University Press.
Balassa, Bela, and Associates. 1971. The Structure of Production in
   Developing Countries. Baltimore, Md.: Johns Hopkins University
Bates, Robert, and Anne Krueger, eds. 1993. Political and Economic
   Interactions in Economic Policy Reform: Evidence from Eight
   Countries. Oxford: Oxford University Press.
Beason, Richard, and David E. Weinstein. 1996. “Growth,
   Economies of Scale, and Targeting in Japan, 1955–1990.” Review
   of Economics and Statistics 78: 286–95.
Bhagwati, Jagdish N. 1978. Foreign Trade Regimes and Economic
   Development: Anatomy and Consequences of Exchange Control
   Regimes. Cambridge, Mass.: Ballinger Press.
———. 1998. “Free Trade: What Now?” Keynote address delivered
   to the International Management Symposium, May 25, University
   of St. Gallen, Switzerland.
Birdsall, Nancy, and David Wheeler. 1993. “Trade Policy and
   Pollution in Latin America: Where Are the Pollution Heavens?”
   Journal of Environment and Development 2(1): 137–49.
Bruno, Michael. 1996. Deep Crisis and Reform: What Have We
   Learned? Washington, D.C.: World Bank.
Clements, Kenneth, and Larry A. Sjaastad. 1984. “How Protection
   Taxes Exporters.” Thames Essay 39. London: Trade Policy
   Research Centre.
Dollar, David. 1992. “Outward-Oriented Developing Economies
   Really Grow More Rapidly: Evidence from 95 LDCs, 1976–85.”
   Economic Development and Cultural Change 40(4): 523–44.


Drabek, Zdenek, and Sam Laird. 1998. “The New Liberalism: Trade
   Policy Developments in Emerging Markets.” Journal of World Trade
   32(5): 241–69.
Edwards, Sebastian. 1997. “Trade Liberalization Reforms and the
   World Bank.” American Economic Review 87(2): 43–48.
Fernandez, Raquel, and Dani Rodrik. 1991. “Resistance to Reform
   Status Quo Bias in the Presence of Individual Specific
   Uncertainty.” American Economic Review 81(5): 146–55.
Fishlow, Albert, Catherine Gwin, Stephan Haggard, Dani Rodrik,
   and Robert Wade. 1994. Miracle or Design? Lessons from the East
   Asian Experience. Washington, D.C.: Overseas Development
Guasch, J. Luis, and Sarath Rajapatirana. 1998. “Antidumping and
   Competition Policies in Latin America and the Caribbean: Total
   Strangers or Soul Mates?” World Bank Policy Research Working
   Paper 1958. Washington, D.C.: World Bank.
Haggard, Stephan, and Steven Webb, eds. 1994. Voting for Reform:
   Democracy, Political Liberalization, and Economic Adjustment. New
   York: Oxford University Press.
Hedge, Ramkrishna. 1998. “India and the WTO.” Message at Second
   Ministerial Conference of the WTO, May 18, Geneva.
International Monetary Fund. 1998. Trade Liberalization in IMF-
   Supported Programs. Washington, D.C.: World Economic and
   Financial Surveys.
Krueger, Anne O. 1974. “The Political Economy of the Rent-Seeking
   Society.” American Economic Review 64(7): 291–303.
———. 1978. Foreign Trade Regimes and Economic Development:
   Liberalization Attempts and Consequences. Cambridge, Mass.:
   Ballinger Press.
———. 1990. “Trends in Trade Policies of Developing Countries.” In
   The Direction of Trade Policy: Papers in Honor of Isaiah Frank, ed.
   Charles S. Pearson and James Riedel. Cambridge, Mass.:
   Blackwell, 87–107.
Krugman, Paul R., and Robert Z. Lawrence. 1993. “Trade, Jobs, and
   Wages.” Working Paper 4478. Washington, D.C.: National Bureau
   of Economic Research.
Lal, Deepak, and Richard H. Snape. Forthcoming. Trade,
   Development, and Political Economy: Essays in Honor of Anne O.
   Krueger. New York: Macmillan.
Lall, Sanjaya. 1996. “Paradigms of Development: The East Asian
                                                   REFERENCES ✜ 39

   Debate.” Oxford Development Studies 45(2): 147–74.
Lawrence, Robert Z., and David E. Weinstein. 1999. “Trade and
   Growth: Import-Led or Export-Led? Evidence from Japan and
   Korea.” World Bank Workshop on Rethinking the East Asian
   Miracle, February, San Francisco. Mimeo.
Little, I. M. D., Richard N. Cooper, W. Max Corden, and Sarath
   Rajapatirana. 1993. Boom, Crisis and Adjustment: The
   Macroeconomic Experience of Developing Countries. Oxford: Oxford
   University Press.
Little, I. M. D., Tibor Scitovsky, and Maurice F Scott. 1970. Industry
   and Trade in Some Developing Countries: A Comparative Study.
   London: Oxford University Press.
Lucas, Robert. 1988. “On the Mechanics of Economic
   Development.” Journal of Monetary Economics 22(1): 3–42.
Olson, Mancur. 1982. The Rise and Decline of Nations: Economic
   Growth, Stagnation, and Social Rigidities. New Haven, Conn.: Yale
   University Press.
Rajapatirana, Sarath. 1997. “Evaluating Integration Choices: The
   Case of Bolivia.” Journal of Economic Integration 12(3): 298–324.
———. 1998. “Colombian Trade Policies and the 1996 WTO Trade
   Policy Review.” World Economy 21(4): 515–27.
Rajapatirana, Sarath, Luz Maria de la Mora Sanchez, and Ravindra A.
   Yatawara. 1997. “Political Economy of Trade Reforms 1965–94:
   Latin American Style.” World Economy 20(3): 307–38.
Rodrik, Dani. 1992. “The Limits of Trade Policy Reform in
   Developing Countries.” Journal of Economic Perspectives 6 (win-
   ter): 87–105.
———. 1994. “King Kong Meets Godzilla: The World Bank and the
   East Asia Miracle.” In Miracle or Design? Lessons from the East Asia
   Experience, ed. Albert Fishlow, Catherine Gwin, Stephan Haggard,
   Dani Rodrik, and Robert Wade. Washington, D.C.: Overseas
   Development Council.
Romer, Paul M. 1986. “Increasing Returns and Long-Run Growth.”
   Journal of Political Economy 94(5): 1002–37.
Sachs, Jeffrey D., and Andrew Warner. 1995. “Economic Reform and
   the Process of Global Integration.” Brookings Papers on Economic
   Activity, no. 1. Washington, D.C.: Brookings Institution.
Singer, Hans W. 1950. “The Distribution of Gains between Investing
   and Borrowing Countries.” American Economic Review 40(2):

Wade, Robert. 1990. Governing the Market: Economic Theory and the
   Role of the Government in East Asian Industrialization. Princeton, N.
   J.: Princeton University Press.
World Bank. 1986. World Development Report. Washington, D.C.:
   World Bank.
———. 1993. The East Asian Miracle. New York: Oxford University
———. 1999. Global Economic Prospects and the Developing Countries,
   1998/99. Washington, D.C.: World Bank.
World Trade Organization. 1998. Annual Report. Geneva: World
   Trade Organization.
Yeats, Alexander J. 1997. “Does Mercosur’s Trade Performance Raise
   Concerns about the Effects of Regional Trade Arrangements?”
   World Bank Policy Research Working Paper 1729. Washington,
   D.C.: World Bank.
                   About the Author

SARATH RAJAPATIRANA is a visiting scholar at the American Enterprise
Institute. He joined AEI after a long and distinguished career at the
World Bank. His latest book (his sixth) is Liberalization and Industrial
Transformation in Sri Lanka (written with Premachandra Athukorala),
published by Oxford University Press in 2000.


To top