MfrRCQS~~~    ~                    [2089
i MA                         S/tb  `-AL."                 2 A  u  00
i : MErR^CUSUIR - liNt-d T A  -    C'     w       
A WO R LD B A N K PO L IC Y R E S E AR CH R EPORT
EA4qEC                e ErWR-W3A EGA  R  M A ' H R G3 - CC dOC'
cosA1
NC   M      R                                        0CU'C
ECO                                                             MEF
rA'SAARC  / SAA                                                A  .
tocm CB -3                      E5-L .f lC  0<
. ~ ~ ~. ..
LEU. .EUR                                     r    >.J4  .iA A.t
J  SAD
A1I
I (t
M A %G ; H R'. ;K '. -1
.tjv''~'    Jp_2A \_.. re LLt-






Trade Blocs
A World Bank Policy Research Report






Trade Blocs
Published for the World Bank
OXFORD UNIVERSITY PRESS



Oxford University Press
OXFORD   NEW YORK  ATHENS  AUCKLAND   BANGKOK   BOGOTA   BUENOS AIRES
CALCUTrA  CAPE TOWN   CHENNAI  DAR ES SALAAM   DELHI  FLORENCE   HONG
KONG ISTANBUL KARACHI KUALA LUMPUR MADRID MELBOURNE MEXICO CITY
MUMBAI  NAIROBI  PARIS  SAO PAULO  SINGAPORE  TAIPEI  TOKYO   TORONTO
WARSAW
and associated companies in
BERLIN IBADAN
C 2000 The International Bankfor Reconstruction
and Development / The World Bank
1818 H Street, N.W, Washington, D. C 20433, USA
Published by Oxford University Press, Inc.
198 Madison Avenue, New York, N. Y 10016
Oxford is a registered trademark of Oxford University Press.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted, in anyform or by any means, electronic, mechanical, photocopying,
recording, or otherwise, without the prior permission of Oxford University Press.
Manufactured in the United States ofAmerica
First printingAugust 2000
Thefindings, interpretations, and conclusions expressed in this study are entirely those of
the authors and should not be attributed in any manner to the World Bank, to its affiliated
organizations, or to members of its Board of Executive Directors or the countries they
represent. The boundaries, colors, denominations, and other information shown on any
map in this volume do not imply on the part of the World Bank Group any judgment on
the legal status of any territory or the endorsement or acceptance of such boundaries.
Library of Congress Cataloging-in-Publication Number: 00-028297



Contents
Foreword   ix
Acknowledgments    xiii
The Research Project and Team    xv
1.  Regional Integration Agreements    1
Introduction    1
This Report   5
2.   Politics and Policymaking    11
Introduction    1 1
2.1 Integration for Security   12
2.2 Integration for Bargaining Power   17
2.3 Project Cooperation    21
2.4 Integration for Lock-in to Reform    22
2.5 Lobbying for Integration    26
2.6 Conclusion: So ... Integration Is Political   28
3.  Economic Benefits and Costs    29
Introduction    29
3.1 Competition and Scale   30
3.2 Trade and Location: The Pattern of Trade    39
3.3 Trade and Location: Convergence or Divergence?    51
3.4 Conclusion    61
4.   Policy Choices    63
Introduction    63
4.1 Partners: With Whom?    64
4.2 External Trade Policy: How Much Preference?   73
4.3 How Deep?    79
4.4 How Wide?    85
4.5 Conclusion    90
v



CONTENTS
5.   Trade Blocs and the World Trading System      93
Introduction    93
5.1 How RIAs Set External Tariffs    94
5.2 The Dynamics of Regionalism   96
5.3 Regionalism and Multilateral Trade Negotiations    101
5.4 Regionalism and the WTO    106
5.5 The Rules for RlAs    108
5.6 Post-Seattle    115
5.7 Conclusion    117
Appendix I: WTO Provisions on Regional Integration Arrangements
(Extracts)    118
6.   Conclusion: Tell Me the Truth about Trade Blocs     123
Trade blocs are political...   123
The politically feasible alternative to a costly trade bloc is probably a
better-designed bloc...    124
So how does a trade bloc affect people, especially the poorest?    125
I'm the minister of trade. What bloc design should I choose?    128
Which countries should I take as partners?    128
How many blocs should I join?    129
How much preference should I give?    130
Should I press for a common external tariff?    130
How deep should I take liberalization?    130
How wide should I let negotiations range?    131
I went to Seattle. How can I use the WTO more effectively?    131
Bibliography     133
Boxes
1.1  Recent Regional Integration Agreements    3
3.1  The Single Market Program   35
3.2  Production Networks    60
Figures
1.1  Notifications to GATT and WTO of Regional Integration Agreements,
1949-98    2
3.1  United States Export Prices to Brazil and Rest of World    36
3.2  Republic of Korea Export Prices to Brazil and Rest of World    36
3.3  Mexican Foreign Direct Investment, Net Inflows    38
3.4  Intra-RIA Imports as Share of GDP    46
3.5  Extra-RIA Imports as Share of GDP    46
3.6  Ratio of Intra-RIA Imports over Extra-RIA Imports   46
3.7a RIA's Trade within and across Borders: Evidence of Diversion    48
3.7b RIA's Trade within and across Borders: No Evidence of
Diversion    49
vi



CONTENTS
3.8  Incomes Converge as EEC Integrates    52
3.9  CACM, Formed in 1960, Manufactures Value Added    53
3.10 CEAO, Formed in 1974, Manufacturing Value Added    54
3.11 Convergence and Divergence of Real Incomes    55
4.1  Intra-RIA Exports as a Share of the RIAs Total Exports   65
4.2  Regional Organizations in Southern and Eastern Africa    71
Tables
1.1  Membership of Selected Major Regional Integration Agreements
(RIAs) and Date of Formation    4
3.1  MERCOSUR: Net Inflows of Foreign Direct Investment    38
3.2  Customs Revenue Collected as a Percent of Total Government Revenue
in 1996 and the Implications of a Free Trade Area for SADC
Members    45
4.1  Pluses and Minuses of Hypothetical RlAs    66
vii






Foreword
T       HERE HAS BEEN A VERITABLE EXPLOSION OF REGIONAL
integration agreements (RIAs) in the last 15 years or so.
Nearly every country in the world is either a member of-or
discussing participation in-one or more regional integration
arrangements. Such agreements have been concluded among high-income
countries, among low-income countries, and, more recently, starting with
the North American Free Trade Area (NAFTA)-between high-income
and developing countries. More than half of world trade now occurs
within actual or prospective trading blocs. Why are countries so active
in their pursuit of preferential trade agreements? Do such agreements
allow countries to obtain benefits that cannot be achieved through
autonomous actions or multilateral cooperation? Or is it an instance of
"old wine in new bottles," as argued by opponents of regional
arrangements who regard them as the outcome of political economy
forces fostering discriminatory trade restrictions and undermining the
multilateral trading system?
Although regional integration has given rise to a vigorous debate-
one that shows few signs of dying down-the quality of interchange is
frequently low. There is often a lack of understanding of what determines
the economic effects of a RIA and ignorance of the available evidence on
the various potential justifications that are offered for pursuing RIAs.
Few of the more recent arguments in favor of RIAs have been satisfacto-
rily formalized or tested-for example, that regionalism stimulates in-
vestment, that it confers credibility on reform programs, or that it assists
multilateral liberalization-and no attempt has been made to weigh them
against each other in the circumstances of developing countries. At the
same time, economists in particular have not paid enough attention to
the noneconomic objectives that frequently underlie RIAs, and the role
ix



FOREWORD
of trade preferences in achieving these objectives. Understanding the po-
tential linkages between favoritism in trade and the pursuit of noneco-
nomic political and social objectives is crucial in deciding both the bot-
tom line question of whether to join a RIA, and the negotiation question
of what design is most suitable.
This report summarizes the findings of a research project to under-
stand better the political economy of regional integration, the economic
benefits and costs for developing countries, the policy choices confront-
ing governments, and the implications of regionalism for nonmembers
and the multilateral trading system.
Regional agreements frequently have political objectives and noneco-
nomic dimensions, including national security, enhancement of bargain-
ing power, and bolstering the credibility of reforms (economic and politi-
cal). The report concludes that attainment of such political objectives de-
pends primarily on agreement design, as this determines the size and dis-
tribution of net economic impacts. Economic gains depend on securing
an effective increase in competition on the regional market and minimiz-
ing incentives for consumers to start purchasing goods and services from
inefficient regional suppliers. This can be achieved by increasing openness
to trade with the outside world at the same time as intraregional trade is
being liberalized. One central finding of the report is that selecting high-
income countries as partners is often likely to be the best option for devel-
oping countries. Industrial countries tend to have lower barriers, there is
greater scope for exploitation of comparative advantage, and a higher prob-
ability of attaining political objectives. Agreements among developing coun-
tries are more likely to lead to income divergence of members, causing
potential for implementation problems and tensions.
Deeper and wider forms of integration may be required to realize sought-
after benefits of greater competition and economies of scale in produc-
tion. These forms of integration include agreements that go beyond aboli-
tion of tariffs to include other (national) policies that affect trade in goods,
such as customs procedures, product conformity assessment requirements,
abolition of antidumping and safeguard protection, as well as investment
and services policies. Such forms of integration often can and should be
applied on a nondiscriminatory basis, in which case regional agreements
will complement the multilateral trading system. But experience to date
suggests that achieving deeper and wider integration is not easy. Few ex-
tant agreements involving developing countries have gone significantly
x



FOREWORD
beyond the preferential reduction of border barriers. Deeper forms of in-
tegration frequently will require the creation of mechanisms to ensure an
equitable distribution of gains.
Making regional integration an effective element of national devel-
opment strategies is a great challenge. Regional agreements can reduce
the frictional costs of trade by harmonizing regulations and standards.
By making national policies a part of international agreements, regional
agreements can increase the credibility of reform initiatives and strengthen
security arrangements between partners. They can be vehicles for gov-
ernments to test the waters of freer trade. By reducing trade barriers on
a subset of partners, however, countries generally increase the real cost of
their imports, reduce the flow of technology from nonmember coun-
tries, and increase dependence on particular export markets. Great care
is required to ensure that such costs are compensated by other gains. If
not, membership may hinder, rather than promote, development.
Like previous Policy Research Reports, this study is the product of
the staff of the World Bank: the judgments and arguments made herein
do not necessarily reflect the views of the World Bank's Board of Direc-
tors or the governments they represent.
Josef Ritzen
Vice President, Development Policy
The World Bank
xi






Acknowledgments
T       HE PROJECT HAS BENEFITED GREATLY FROM THE WORK OF
the many scholars who prepared background papers.
Bernard Hoekman and Peter Robson contributed to early drafts
of the report. Jaime De Melo and Simon Everett provided useful
comments. Excellent research assistance was provided by Soamiely
Andriamananjara, Won Chang, Anju Gupta, Moonhui Kim, Praveen
Kumar, and Isidro Soloaga. Lawrence MacDonald and Joost Polak
provided valuable editorial services. Mary-Ann Arouna, Maria Lourdes
Kasilag, and Rebecca Martin provided superb administrative support
throughout the duration of the project.
.x.i.






The Research Project
and Team
T       HE PRINCIPAL AUTHORS OF THIS POLICY RESEARCH REPORT
were Paul Collier, Maurice Schiff, Anthony Venables, and L.
Alan Winters. The report draws on the findings of a large research
project on regional integration led by Maurice Schiff and L. Alan Winters,
who-with Bernard Hoekman-prepared the first draft of the report.
The magnitude of the research output generated by the project is too
large to do full justice in this report. A more detailed, book-length analysis
and assessment of regional integration that draws in greater depth on
the results of the research project is being prepared by Maurice Schiff, L.
Alan Winters, and Bernard Hoekman. Interested readers can download
many of the background papers from the World Bank trade website at
www.worldbank.org/trade. A number of papers were published in a
special issue of the World Bank Economic Review (May 1998).
xv






CHAPTER  1
Regional Integration
Agreements
Introduction
T       HE GROWTH OF REGIONAL TRADING BLOCS-OFTEN KNOWN
as regional integration agreements (RIAs)-is one of
the major international relations developments of recent years.
Most industrial and developing countries in the world are members of a
regional integration agreement, and many belong to more than one:
more than one-third of world trade takes place within such agreements.'
The structure of regional agreements varies hugely, but all have one thing
in common-the objective of reducing barriers to trade between member
countries. At their simplest they merely remove tariffs on intrabloc trade
in goods, but many go beyond that to cover nontariff barriers and to
extend liberalization to trade and investment. At their deepest they have
the objective of economic union, and they involve the construction of
shared executive, judicial, and legislative institutions.
During the last decade the move to regionalism has become a headlong
rush. Figure 1.1 gives the number of regional agreements notified to the
General Agreement on Trade and Tariffs and the World Trade Organiza-
tion (GATT/WTO)2 each year, and makes apparent the dramatic increase
that occurred in the 1990s. Of the 194 agreements notified at the begin-
ning of 1999, 87 were notified since 1990. Description of some of the
major agreements made in recent years is given in box 1. 1, and table 1.1
lists selected trading blocs, their memberships-and their acronyms.
The last 10 years have witnessed qualitative, as well as quantita-
tive, changes in regional integration schemes. There have been three
major developments.
The first is the recognition that effective integration requires more than
reducing tariffs and quotas. Many other barriers have the effect of segment-
ing markets and impeding the free flow of goods, services, investments, and
1



TRADE B LO CS
Figure .1 Notifications to GATr and WTO of Regional Integration Agreements, 1949-98
INotifications
25 -
20 -
15 -
10 
5 
0-
1949   1954   1959   1964   1969   1974   1979   1984   1989   1994
Sourre:)WTO data.
ideas, and wide ranging policy measures-going well beyond traditional
trade policies-are needed to remove these barriers. This so-called deep in-
tegration was first actively pursed in the Single Market Program of the Euro-
pean Union (EU), and elements of this program are now finding their way
into the debate in other regional agreements.
The second is the move from "closed regionalism" to a more open
model. Many of the trading blocs that were formed between developing
countries in the 1960s and 1970s were based on a model of import-
substituting development, and regional agreements-with high external
trade barriers-were used as a way of implementing this model.3 The
new wave of regional agreements-including resurrection of some old
agreements-have generally been more outward-looking, and more com-
mitted to boosting, rather than controlling, international commerce.
The third development is the advent of trade blocs in which both
high-income industrial countries and developing countries are equal part-
ners in agreements designed to bolster the economies of all the member
2



REG I O NAL I NT EG RAT I O N  AG R EEM E N T S
Box 1.1 Recent Regional Integration Agreements
EU HAS PLAYED A LARGE ROLE IN THE RECENT SURGE   African Cooperation sprang up where the East Af-
of activity, with the implementation of the Single Mar-  rican Community had failed.
ket Program in 1992, enlargement of its membership,  The Middle East wimessed the development of the
and numerous agreements with other countries. These   Gulf Cooperation Council (GCC), and in 1997 Arab
agreements account for two-thirds of the agreements   League members agreed to cut trade barriers over a 10-
notified to GATT/WTO since 1990 and include the   year period. In Asia, the Association of Southeast Asian
European Economic Area, the Europe Agreements   Nations (ASEAN) countries developed 25 years of po-
with the countries of Eastern Europe, the EU-Turkey   litical cooperation into a free trade area in 1992, with
customs union, and the development of a Mediterra-   the formation of the ASEAN Free Trade Area. The
nean policy potentially incorporating regional agree-   South Asian Association for Regional Cooperation
ments with most countries on the southern and east-   agreed in 1997 to transform itself into the South Asian
ern shores of the Mediterranean.               Free Trade Area becoming, in terms of the population
In Latin America, MERCOSUR was formed in   it represents, the world's largest regional agreement.
1991 and the Group ofThree in 1995. The Andean   New ground was broken in 1994 when the Canada-
Pact and Central American Common Market   U.S. Free Trade Area was extended to Mexico through
(CACM) were resurrected in 1991 and 1993, re-   NAFTA. For the first time a developing country joined
spectively.                                    industrial countries as an equal partner in a trade bloc
In Sub-Saharan Africa, the blocs in West Africa   designed to bolster economic development in all three
were reformed and reorganized. The Southern Af-   economies. Links between high-income and develop-
rican Development Community (SADC) devel-  ing countries were also being forged in the Asia Pacific
oped out of an earlier defense-based organization,   Economic Cooperation (APEC), established in 1989-
Southern African Development Coordination Con-   a looser organization, committed to trade liberaliza-
ference, anid was supplemented-for many of its  tion, on a nonpreferential basis, by 2010 for industrial
members-by the Cross-Border Initiative. The East   country members and 2020 for developing countries.
countries.4 Perhaps the most important example of this is the North
American Free Trade Area (NAFTA), formed in 1994 when the Canada-
U. S. Free Trade Agreement was extended to Mexico. EU has linked with
the transition economies of Eastern Europe through the Europe Agree-
ments, and has developed the EU-Turkey customs union and a Mediter-
ranean policy potentially incorporating agreements with nearly every
Mediterranean country. There has been discussion of replacing the EU's
trade preferences of the Lome agreements with reciprocal trade agree-
ments with these developing countries.
These developments have occurred against the backdrop of globaliza-
tion. New technologies and more liberal trading regimes have led to in-
creased trade volumes, larger investment flows, and increasingly foodoose
3



T RAED E  B LO CS
Table 1.1 Membership of Selected Major Regional Integration Agreements (RMAsO and Date of Fornation
Industrial and developing  European Union (EU): formerly European Economic Community (EEC) and European
economies              Community, 1957: Belgium, France, the Federal Republic of Germany, Italy, Luxembourg,
the Netherlands; 1973: Denmark, Ireland, United Kingdom; 1981: Greece; 1986. Portugal,
Spain; 1995: Austria, Finland, Sweden.
European EconomicArea: 1994: EU, Iceland, Liechtenstein, Norway.
Euro-Mediterranean Economic Area (Euro-Maghreb): Bilateral agreements, 1995: EU and
Tunisia; 1996.: EU and Morocco.
EU bilateral agreements with Eastern Europe: 1994: EC and Hungary, Poland; 1995:
European Community and Bulgaria, Romania, Estonia, Latvia, Lithuania, Czech Republic,
Slovak Republic, Slovenia.
Canada-US. Free TradeArea: 1988: Canada, United States.
North American Free Trade Area (NAFTA): 1994: Canada, Mexico, United States.
Asia Pacific Economic Cooperation (APEC): 1989: Australia, Brunei Darussalam, Canada,
Indonesia, Japan, Malaysia, New Zealand, Philippines, the Republic of Korea, Singapore,
Thailand, United States; 1991: China, Hong Kong (China), Taiwan (China); 1993: Mexico,
Papua New Guinea; 1994: Chile; 1998: Peru, Russia, Vietnam.
Latin America and        Andean Pact: 1969: revived in 1991, Bolivia, Colombia, Ecuador, Peru, Venezuela.
the Caribbean          CentralAmerican Common Market (CACM): 1960: revived in 1993, El Salvador, Guatemala,
Honduras, Nicaragua; 1962: Costa Rica.
Southern Cone Common Market (Mercado Comrin del Sur-MERCOSUR): 1991: Argentina,
Brazil, Paraguay, Uruguay.
Group ofThree: 1995: Colombia, Mexico, Venezuela.
Latin American Integration Association (LAIA): formerly Latin American Free Trade Area,
1960: revived 1980, Mexico, Argentina, Bolivia, Brazil, Chile, Colombia,
Ecuador, Paraguay, Peru, Uruguay, Venezuela.
Caribbean Community and Common Market (CARICOM): 1973: Antigua and Barbuda,
Barbados, Jamaica, St. Kitts and Nevis, Trinidad and Tobago; 1974: Belize, Dominica,
Grenada, Montserrat, St. Lucia, St. Vincent and the Grenadines; 1983: The Bahamas (part of
the Caribbean Community but not of the Common Market).
Sub-Saharan Africa        Cross-Border Initiative: 1992: Burundi, Comoros, Kenya, Madagascar, Malawi,
Mauritius, Namibia, Rwanda, Seychelles, Swaziland, Tanzania, Uganda, Zamsbia, Zimbabwe.
East African Cooperation: 1967: formerly East African Community, broke up in 1977 and
recently revived, Kenya, Tanzania, Uganda.
Economic and Monetary Community of Central Africa: 1994: formerly Union
Douaniere et Economique de l'Afrique Centrale, 1966: Cameroon, Central
African Republic, Chad, Congo, Gabon; 1989: Equatorial Guinea.
Economic Community of WestAfrican States (ECOWAS): 1975: Benin, Burkina Faso, Cape
Verde, C6tc d'Ivoire, Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Mauritania,
Niger, Nigeria, Senegal, Sierra Leone, Togo.
Common Marketfor Eastern and Southern Africa: 1993: Angola, Burundi,
Comoros, Djibouti, Egypt, Ethiopia, Kenya, Lesotho, Malawi, Mauritius, Mozambique,
Rwanda, Somalia, Sudan, Swaziland, Tanzania, Uganda, Zambia, Zimbabwe.
(Table continues on thefollowingpage.)
4



RE GI O NAL  I N T EG RATI O N  AG R EE M E N TS
Table 1.1 (continued)
Indian Ocean Commission: 1984: Comoros, Madagascar, Mauritius, Seychelles.
Southern African Development Community (SADC). 1980: formerly known as the Southern
African Development Coordination Conference, Angola, Botswana, Lesotho,
Malawi, Mozambique, Swaziland, Tanzania, Zambia, Zimbabwe; 1990: Namibia; 1994:
South Africa; 1995: Mauritius; 1998: Democratic Republic of the Congo, Seychelles.
Economic Community of WestAfrica: 1973: revived in 1994 as West African Economic and
Monetary Unit, Benin, Burkino Faso, C6te d'Ivoire, Mali, Mauritania, Niger, Senegal.
WestAfrican Economic and Monetary Union: 1994: Benin, Burkina
Faso, C6te d'Ivoire, Mali, Niger, Senegal, Togo, 1997: Guinea-Bissau.
Southern African Customs Union (SACU): 1910: Botswana, Lesotho, Namibia, South Africa,
Swaziland.
Economic Community of the Countries of the Great Lakes: 1976. Burundi, Rwanda,
Democratic Republic of the Congo.
Middle East and Asia      Association of SoutheastAsian Nations (ASEAN): 1967: ASEAN Free Trade Area was created
in 1992, Indonesia, Malaysia, Philippines, Singapore, Thailand; 1984: Brunei Darussalam;
1995: Vietnam; 1997: Myanmar, Lao People's Democratic Republic; 1999: Cambodia.
Gulf Cooperation Council (GCC): 1981: Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, the
United Arab Emirates.
South Asian Association for Regional Cooperation: 1985: Bangladesh, Bhutan, India,
Maldives, Nepal, Pakistan, Sri Lanka.
Source: WTO data.
production. All these considerations point to the need for a new analysis of
regional integration agreements-one that takes into account political as
well as economic effects, that assesses the opportunities for deep integra-
tion, and captures the new potentials created by North-South agreements.
This Report
N            O  COUNTRY IS IMMUNE FROM THE EFFECTS OF
regionalism as it shapes world economic and
political relationships and influences the development of the
multilateral trading system, and all countries face policy choices concerned
with regionalism. Should they enter a regional integration agreement?
With what other countries? What measures should be implemented-
simple trade liberalization or deeper harmonization of domestic policies?
There are no simple answers to these questions. Countries differ in their
5



TRA DE B LO CS
circumstances and in their political and economic objectives.
Nevertheless, the economic tradeoffs that they face in making these
choices can be identified and the process of decisionmaking made better
informed. This is the objective of this report. We use the latest research
on regionalism-research undertaken in the World Bank and elsewhere-
to evaluate the experience of existing regional integration agreements
and to draw inferences from this for the tradeoffs faced by decisionmakers.
An important question in analyzing the effects of RIAs is: Compared
to what? The choice of counterfactual-including the status quo, uni-
lateral trade liberalization, or multilateral liberalization-depends on the
objective of the analysis. Answers will vary according to the choice of
counterfactual. In fact, disagreements between policymakers or analysts
may be due as much to differences in counterfactuals they have in mind
as to differences in views. It is therefore important to be clear about the
counterfactual that is being used as the basis for the analysis.
Given the wave of new RIAs in the last decade, and the fact that
many developing countries have approached the World Bank for techni-
cal support and policy advice on whether to join or form a RIA, on the
choice of partner, on the type of RIA, and on details of implementation,
we choose the status quo as the counterfactual. However, when appro-
priate, we will also compare our results with those obtained under uni-
lateral trade liberalization.
Politics and Policymaking
Many of the arguments for membership in a regional agreement are
political, and we address these in the next chapter. There are three main
issues. The first issue is security. There may be perceived benefits from
using a regional agreement as a basis for increasing security against non-
members, and there are some examples where this has occurred. A re-
gional agreement may also enhance a country's security in its relation-
ship with other members, an important argument in the early days of
European integration. These security arguments are driven by a variety
of mechanisms. Interlocking economies can make conflict more expen-
sive; regular political contact can build trust and facilitate other forms of
cross-border cooperation. But a regional agreement can also create inter-
nal tensions, particularly if driven by economic rather than security con-
siderations and if the economics appears to bring an unfair distribution
6



REG I ONAL I NT EG RAT ION  AG RE EM EN TS
of benefits. History provides examples where this has led to the breakup
of economic unions and to conflict.
The second sort of political effect is bargaining power-the hope that
from unity comes strength. The likelihood of this occurring depends on
who the member countries are. The EU has probably been able to secure
more in some international negotiations than could its member states
acting independently. Regional agreements between small developing
countries cannot aspire to the EU's economic weight or political power,
but can nevertheless enter negotiations more effectively than separate
countries might be able to. They are more likely to "be noticed," and
thence more likely to be able to make deals. Of course, these benefits
depend on members being able to formulate a common position on
relevant issues, a goal that has often proved elusive.
The third issue is "lock-in," and relates to the effect of the regional
agreement on domestic politics. Attempts at reform are often under-
mined by expectations of reversal. A regional agreement can provide a
"commitment mechanism" for trade and other policy reform measures.
It can be a way of raising the cost, and thereby reducing the likelihood,
of policy reversal. This argument can apply to political as well as eco-
nomic reform, and there are examples where regional integration agree-
ments have reinforced democracy in member states. However, the effec-
tiveness of regional agreements as "commitment mechanisms" depends
on the interests and degree of involvement of all the countries concerned.
Domestic political pressures and the activities of lobbies will also influ-
ence the form of many regional agreements-quite possibly preventing
them from being as effective as they might otherwise have been.
Economic Costs and Benefits
The economic effects of regional agreements are studied in chapter 3,
and are of two main types. The first are "scale and competition" effects.
Removal of trade barriers is like a market enlargement, as separate na-
tional markets move toward integration in a regional market. This allows
firms to benefit from greater scale and attracts investment projects for
which market size is important, including foreign direct investment (FDI).
Removing barriers also forces firms from different member countries into
closer competition with each other, possibly inducing them to make effi-
ciency improvements. Although these are major sources of benefit, we
7



TRADE B LO CS
will see that the effects are not always achieved, and that the effects de-
pend on the design and implementation of the agreement. Some of these
benefits may be achievable with unilateral trade liberalization, although
the latter may not always be politically feasible.
The second source of economic change is "trade and location" effects.
The preferential reduction in tariffs within a regional agreement will
induce purchasers to switch demand toward supply from partner coun-
tries, at the expense of both domestic production and imports from non-
members. This is trade creation and trade diversion. Governments will
lose tariff revenue, and the overall effect on national income may be
positive or negative, depending on the costs of alternative sources of
supply and on trade policy toward nonmember countries.
Changes in trade flows induce changes in the location of production
between member countries of a regional agreement. These relocations
are determined by the comparative advantage of member countries, by
agglomeration or clustering effects, and by possible technology transfer
between countries. In some circumstances, relocations can be a force for
convergence of income levels between countries. Labor-intensive pro-
duction activities may move toward lower-wage countries, raising wages
there. In other circumstances, relocations can be a force for divergence.
Industry may be pulled toward a country with a head start or with some
natural advantage, driving up incomes ahead of other countries. We ar-
gue that divergence is more likely in "South-South"' regional agreement
schemes between economically small low-income countries. It can cre-
ate tensions that lead to failure of the agreement. "North-South" re-
gional agreements are also more likely to generate useful technology trans-
fers for Southern members, particularly if the Northern member is an
important producer of knowledge.
Policy Choices
Regionalism confronts countries with four broad areas of policy choice;
analysis of these choices is the subject of chapter 4. The first choice is
with whom, if anyone, to form an agreement. A developing country
may face options ranging from an agreement with another low-income
country, to partnership with a high-income country or bloc, or mem-
bership of a large regional grouping such as the Asia Pacific Economic
Cooperation (APEC). The political and economic effects differ widely
8



R EG I O NAL I N T EG RAT I O N  AG R E EM E NTS
across these cases, and we identify circumstances under which we think
gains or losses are more or less likely. We also investigate the costs and
benefits of belonging to more than one agreement.
The second policy choice concerns the policy stance members of a
regional agreement take toward the outside world. A regional agreement
is inherently preferential, discriminating in favor of members-how dis-
criminatory should it be? Trade policy may be set regional agreement
wide (as is the case in a customs union) or left to the discretion of indi-
vidual members (as in a free trade area). These different arrangements
each have costs and benefits, as well as creating different incentives for
setting external trade policy.
The third area of policy choice is the "depth" of integration to be
pursued. Different countries enter regional agreements with very differ-
ent objectives and constraints. For some the objectives are to secure eco-
nomic integration and to build deep political links, and there is a will-
ingness to exert political effort to meet these objectives. Thus, the EU
seeks "economic union" and the development of shared political institu-
tions. "Deep" integration of this type may bring larger benefits in many
areas of activity. For example, deep integration might involve harmoni-
zation of product standards and of parts of the fiscal system. This will
make it more likely that economic "scale and competition" gains are
realized. However, it also involves greater loss of sovereignty, greater po-
litical commitment, and far more complex policymaking than would a
looser free trade agreement.
In addition to the depth of integration, countries must also decide on
the "width" of an agreement, and this is the fourth policy area. How
wide a range of activities should be covered by an integration agree-
ment-just trade in goods, or also extension to services and factor mo-
bility? We look at some of the economic issues that arise, and some of
the practical problems that may be encountered as countries choose the
depth and width of integration.
Regionalism and the World Trading System
One of the most hotly debated issues in the area is the effect of region-
alism on the world trading system. To some, regionalism is a "stepping
stone" toward global free trade, while to others it is a "stumbling block,"
inhibiting progress in multilateral trade liberalization. We address these
9



TRA DE  B LO CS
issues in chapter 5, and find no evidence that regionalism has caused tar-
iffs on nonmembers to be higher than they otherwise would have been.
But neither do we accept the arguments that regionalism sets in train a
"liberal dynamic," or facilitates multilateral liberalization. We find some
case for strengthening the operation of WTO procedures to ensure that
regional agreements are genuinely trade liberalizing, and for increased ac-
cess to Organisation for Economic Co-Operation and Development
(OECD) markets for developing countries.
The Scope of this Report
Although we argue that analysis of regional integration agreements
should go well beyond traditional approaches of trade creation and trade
diversion, we must also circumscribe the scope of this report. Our focus
is trade, and we do not discuss either free movement of labor or mon-
etary union. These are both features of the EU, but they are rare else-
where. The EU is the only major regional agreement that gives all citi-
zens right of abode and employment in all member countries. Monetary
union occurs in francophone Africa, but it is outside the terms of refer-
ence of most regional agreements.
Notes
1. T'his figure increases to 59 percent if APEC is in-   the Economic Community of West African States,
cluded.                                             ECOWAS (1975), the Economic Community of the Coun-
tries of the Great Lakes (1976), and the East African Com-
2. General Agreement on Trade and Tariffs, which    munity (1967). In Latin America and the Caribbean, ex-
from 1994 onward has been part of the World Trade Or-   amples include the Central American Common Market,
ganization.                                         CACM (1960), Latin American Free Trade Area (1960),
the Andean Pact (1969), and the Caribbean Community
3. This model of "closed" regionalism was widely   and Common Market, CARICOM (1973).
implemented, especially in Sub-Saharan Africa and Latin
America and the Caribbean. Examples in Africa include the  4.  Unlike the preferential trade agreements implied
Economic Community of West Africa (created in 1973),   by colonial preferences.
10



C HAPT E R 2
Politics and Policymaking
Introduction
R       EGIONAL INTEGRATION IS GOOD POLITICS: IT MEETS
political  needs,  such  as  security  or  enhanced
bargaining power, and it satisfies influential lobbies. Indeed,
the purpose of integration is often political, and the economic
consequences, good or bad, are side effects of the political payoff. The
most successful regional integration to date has been the EU. Here's
how a president of the European Commission, Walter Hallstein, saw the
balance between economic and political objectives in what he was doing:
"We're not in business at all; we're in politics."
In this chapter we explore these political payoffs. When are they im-
portant, when are they illusory?
However, there is another and more worrying reason why regional
integration is good politics. This is because those economic effects that
are most apparent-larger markets and economies of scale-happen to
be favorable. As we show, other effects make the net economic impact of
integration ambiguous, but these are often overlooked in political dis-
course. So, regional integration is good politics partly because it is
"soundbite economics," based on only those effects that are easiest to
grasp. Sometimes the net effects will be highly favorable, but sometimes
"soundbite economics" is seriously misleading: the illusion of gain col-
lides with the reality of large costs and major redistributions.
The combination of political payoffs and soundbite economics is
tempting, and goes far to explain the popularity of regionalism. Regional
integration may also be good economics, but the impetus for integration
has usually not been the economics. Sometimes, good politics delivers
bad economics. The political payoff may nevertheless be worth it: If a
11



TRA DE B LO CS
country can substantially increase its security or lock in to democracy by
joining a regional integration scheme, then it may be willing to accept a
few hundred million dollars of costs as worthwhile. But sometimes the
political benefits are ephemeral and the economic costs high. We show
that the economic effects can vary enormously depending upon the char-
acteristics of the country, the countries it chooses as partners, and the
detailed design of the integration arrangement. Only by attending to
these economic choices can the political momentum behind regional
integration avoid the risk of driving an economy into costly mistakes.
2.1 Integration for Security
Intraregional Security
The politics that most concerned President Hallstein and his colleagues
in building the EU was security. The preamble to the 1951 treaty estab-
lishing the European Coal and Steel Community, out of which the EU
grew, stated its aim as follows: "To create, by establishing an economic
community, the basis for broader and deeper community among peoples
long divided by bloody conflicts." Written shortly after the third bloody
conflict between France and Germany in 70 years, it was not surprising
that security was paramount.
Security also played an important role in initiating regional integration
in the Southern Cone. The Argentine and Brazilian militaries had long
perceived each other as potential threats. Economic agreements covering
steel and automobiles were signed in the mid- 1980s as part of an attempt
to reduce tensions, and the creation of MERCOSUR in 1991 reinforced
this process. Rubens Ricupero, a former finance minister of Brazil (and
now Secretary General of the United Nations Conference on Trade and
Development) confirms the importance of MERCOSUR's security role:
Both countries were emerging from a period of military govern-
ments, during which considerable tension had characterized the
bilateral relationship... It was essential to start with agreements in
the economic areas in order to create a more positive external envi-
ronment that rendered it possible to contain the military nuclear
programs, and to replace rivalry by cooperation (Ricupero 1998).
12



P OL I TI CS AN D  PO LI CYMAKI N G
The political impetus to European and Southern Cone integration was
thus based on the belief that increasing trade would reduce the risk of
intraregional conflict. Similar motives are found in the creation of the
Association of Southeast Asian Nations (ASEAN) (to reduce tensions be-
tween Indonesia and Malaysia; see De Rosa 1995), APEC, and the Cen-
tral American Common Market (CACM), which include potential politi-
cal and militaryopponents (Page 1996). Anwar (1994) argues thatASEAN
has promoted regional peace, with intraregional conflicts among the five
founding members before ASEAN was founded, but not afterward.
The idea that increasing trade reduces the risk of conflict has a distin-
guished pedigree. It goes back to Emanuel Kant's Perpetual Peace (1795).
In the nineteenth century it was taken up by the British politician Rich-
ard Cobden, who put his belief into practice as an architect of the Anglo-
French commercial treaty of 1860 (Irwin 1993). In the twentieth cen-
tury, Cordell Hull, a U.S. secretary of state, also put belief into practice
as one of the architects of the postwar international trading order. The
idea turns out to be supported by some evidence. Polachek (1992, 1997)
used the Conflict and Peace Data Bank to investigate both the associa-
tion between trade and security and the direction of causation. He found
that a doubling of trade between two countries lowers the risk of con-
flict between them by around 17 percent.
There are various reasons why this might happen. Political scientists
suggest that negotiations between political leaders on trade issues gradu-
ally build trust, so that elites learn to form cross-national coalitions for
subsequent collaboration. The founding fathers of the European Com-
munity, Robert Schuman and Jean Monnet, went further. They consid-
ered that economic integration would make war "materially impossible,"
meaning that interlocking of steel, coal, and other strategic industries would
leave countries unable to wage war against each other (Milward 1984).
Thus, regions that are economically highly integrated may tend to
have less internal conflict. It seems an obvious step from this result to
the proposition that policies that promote trade within a region will
increase intraregional security. If so, regional integration may provide a
first-best solution and external trade barriers must fall over time and
following deep integration (Schiff and Winters 1998a,b). Sometimes
economic integration paves the way for full political integration so that
the risk of internal conflict is greatly reduced. An example is the Zollverein
of 1834, a customs union among the then-numerous German princi-
palities, which was the precursor to gradual political unification over the
next 37 years and the end of intra-German conflict. A second example is
13



T RADE  B LO CS
the customs union formed between Moldavia and Wallachia in 1847,
which led to the creation of Romania 34 years later (Pollard 1974;
Milward and Saul 1973).
Unfortunately, the result that trade enhances security does not allow
us to conclude that policies that promote trade within a region will nec-
essarily improve the prospects of regional peace. Indeed, they may have
precisely the opposite effect, particularly when enhancing security is not
the main concern. This is because policy-induced integration promotes
trade at a price. The tariff preferences that induce regional trade can
create powerful income transfers within the region and can lead to the
concentration of industry in a single location. The countries or regions
that lose income and industry can be sufficiently resentful that separatist
movements arise and the overall risk of conflict is increased.
A clear example of how trade preferences can trigger conflict was the
American Civil War. The Northern states produced manufactures that
they sold to the Southern states, and the Southern states produced cot-
ton that they exported to Europe. Tariffs first nearly triggered civil war
in 1828. The United States was already a customs union, but in that
year the U.S. Congress, dominated by Northern interests, sharply raised
the U.S. import duty on manufactures. This increased the price that
Northern manufacturers could charge in the South, and so generated a
massive income transfer from the South to the North. The policy was
referred to in the South as the "Tariff of Abominations." South Carolina
refused to collect it and threatened to secede unless it was rescinded. The
federal government sent in troops but Congress backed down before
fighting developed. In 1860 Northern interests tried again, because the
North had so much to gain from high tariffs. This time Congress would
not back down. This (perhaps as much as slavery) was the issue that led
the Southern states to try to quit the Union, and led to the bloodiest
conflict of the nineteenth century (Adams 1993).
Protective barriers also played a role in the conflict in which Ireland
broke away from the Union of 1807 with Britain. Although there were
many contributing factors, the single most decisive was probably the
Irish famine of the 1840s. As part of the Union, Ireland and Britain
were regionally integrated behind high tariff barriers for grain (the
"Corn Laws"). The Corn Laws enabled British grain producers to sell
to Ireland at well above the world price. This income transfer from
Ireland to Britain became most acute during the Irish potato blight
because Ireland needed to import more grain during that time (Irwin
1996). Resentment at the profits from famine forced the repeal of the
14



POLI TICS AND  PO LI CYMAKING
Corn Laws, but not before mass starvation and emigration had taken
place, sowing the seeds of future conflict.
A further example is the war during which East Pakistan broke free
from West Pakistan to become Bangladesh. Here West Pakistan was
equivalent to the Northern states during the Civil War, selling manufac-
tures to East Pakistan at prices forced well above world levels by tariff
barriers. When Pakistan (East and West) was created in 1947 the per
capita income in West Pakistan was only 17 percent higher than in the
East, but the differential grew steadily-32 percent in 1959-60,45 per-
cent in 1964-65, and 61 percent by 1969-70. Islam (1981) attributes
the growing disparity to high levels of external tariff and quota protec-
tion, the concentration of regional development and investment in
Karachi, together with the lack of labor mobility due to the distance
between East and West Pakistan. East Pakistan's desire for secession was
in part motivated by resentment at the large income transfers the tariff
barriers created and this growing income differential.
Conflict within regional integration agreements is illustrated by the
East African Common Market. In this case Kenya was the equivalent of
the Northern states during the Civil War. Tanzania and Uganda com-
plained about the income transfers that were created by the common
external tariff on manufactures. They also feared that there would be
increasing agglomeration of manufacturing in Nairobi, which had a head
start on industrialization compared with the smaller industrial centers
of Tanzania at Dar es Salaam and Uganda at Jinga. Arguments about
compensation for the income transfers contributed to the collapse of the
Common Market, the closing of borders, and the confiscation of Com-
munity assets in 1978. In turn this atmosphere of hostility contributed
to conflict between Tanzania and Uganda in 1979 (Robson 1998;
Venables 1999; and Schiff 2000).
In the CACM, Honduran dissatisfaction with the distribution of ben-
efits was one of the factors behind the 1969 military conflict with El Sal-
vador. After the war, El Salvador vetoed a proposed special development
fund to channel additional resources to Honduras, prompting Honduras
to leave the CACM (Pomfret 1997).
These examples show that trade policy can redistribute income and
produce outcomes that worsen intraregional security. The genius of the
European Community has not been simply that it promoted regional
integration but that it did so without generating the large transfers that
trigger conflict. How did the European Community avoid the problem?
Partly it was a matter of negotiating style and partly a matter of detailed
15



T RALD E B LO CS
design. The style was consensual: negotiators always looked for compro-
mise and conciliation. When a country signaled that a Community policy
would cause it major problems it was accommodated, either by being
bought off with compensation, as with the British budget rebate, or by
being granted a gradual adjustment process, as with compliance with
the rules on labor mobility by Spain and Portugal (Tsoukalis 1993). A
key design feature that eased the problem was that the Community's
external tariffs were generally low and declining. Hence, there were no
large unfair income transfers with producers in one nation able to ex-
ploit consumers in another by selling at prices well above world levels.
The one exception to this is agriculture, which has been very highly
protected and which has generated large income transfers between coun-
tries. But the reactions to this exception support the idea that the trans-
fers generated by high protection create conflict. Agriculture has been
the most contentious of the Community's policies.
Thus, it seems that regional integration can sometimes promote
intraregional security and sometimes worsen it. The net effect depends,
in part, upon the economic characteristics of the members of the region
and upon the style and design of the integration arrangement. Unless
the economics is right, the belief that regional integration enhances se-
curity can indeed be nothing better than "soundbite political science."
Extraregional Security
The political impetus for regional integration is sometimes not
intraregional security, but the need to unite to face a common exter-
nal threat. The underlying idea is that common action in the eco-
nomic sphere makes common action for security easier and more cred-
ible. The country perceived as a potential threat is typically the re-
gional hegemon.
A good example of such a response was the formation of the South-
ern African Development Coordination Conference in 1980 to provide
a united front among the small countries of the region against the apart-
heid regime in South Africa. Part of the strategy was to reduce economic
dependence on South Africa both as a trading partner and as a conduit
for the Southern African Development Coordination Conference trade
with the outside world. Hence, an objective was actually to reduce trade
with the hegemon (Foroutan 1993).
16



P OLITI CS  AN D  P O LI CYMAKING
The Gulf Cooperation Council (GCC) was created in 1981 partly in
response to the potential threat of regional powers such as Iran and Iraq
(Kechichian 1985), and ASEAN was partly motivated by a perceived
need to stem the threat of spreading Communism in the region. A ma-
jor motive of Central and Eastern European countries in applying for
membership to the EU is as protection against a perceived threat from
the Russian Federation.
Sometimes regional integration combines the objectives of intraregional
and external security. An example of how economic cooperation can be
the precursor to military cooperation is the Economic Community of West
African States (ECOWAS), originally formed in 1975. In 1986, 11 of
ECOWAS's 16 members ratified a mutual defense protocol that autho-
rized military intervention by the Community both in conflicts between
members and if conflict in a member country was instigated from outside
the membership. Clearly, 11 countries are unlikely to reach a military
agreement without some prior experience of mutual negotiation and trust
building. While the nations of West Africa have many subregional organi-
zations through which they can build trust, most do not span the
francophone-anglophone divide. ECOWAS was one of the few spanning
organizations (Pomfret 1997) and so played an important, although indi-
rect role in the establishment of the defense protocol.
While there is evidence that induced regional integration is a two-edged
sword with respect to intraregional security, there is too little evidence to
evaluate the effects on extraregional security. However, it is clear that secu-
rity considerations of one type or another have often been an impetus for
regional integration. The economic consequences of such integration are
thus side effects of political decisions. A prime purpose of this report is to
show how to assess these side effects. Sometimes the economic effects will
be favorable so that integration offers politicians both political gains, such
as enhanced security, and economic gains. Sometimes the economic ef-
fects will be unfavorable, facing decisionmakers with a tradeoff
2.2 Integration for Bargaining Power
YJOINING TOGETHER, THE WEAK CAN BECOME STRONG. JUST
as workers band together in unions to increase their
bargaining power against their employer, so small countries can
band together to increase their bargaining power in trade negotiations.
17



TR A D) E B LO CS
How important has the motive of solidarity been in regional integration,
and how effective is solidarity in trade bargaining?
The classic example of regional solidarity exerting economic power
is that of the Organization of Petroleum Exporting Countries (OPEC).
The OPEC agreement is not strictly regional and does not involve
economic integration, but it does illustrate what solidarity can achieve.
Its objective was not to increase trade among its members but rather to
coordinate a restriction in oil export volumes, equivalent to a coordi-
nated increase in export taxation. Hence, it was certainly a trade agree-
ment. This was a successful bargaining strategy in that it raised the
price of oil.
There is some evidence that one motivation for the formation of the
original European Economic Community (EEC) in 1957 was the desire
to increase bargaining power relative to the United States (Milward 1984,
1992; Whalley 1996). Some commentators argue that the formation of
the EEC influenced the U.S. negotiating position during the Dillon and
Kennedy rounds of GATT negotiations as the United States sought to
mitigate the trade-diverting effects of European integration (Lawrence
1991; Sapir 1993; WTO 1995). The members of the EEC probably
achieved two important bargaining objectives. They accelerated United
States-Europe trade liberalization in manufactures; Europe is a net ex-
porter of manufactures to America and reciprocal liberalization improved
European access to American markets, and they delayed trade liberaliza-
tion in agriculture. This raised the incomes of Europe's farmers at the
expense of the rest of its population; although not good economics, this
met the political demands of agricultural lobbies.
While this evidence for enhanced bargaining power may exaggerate
the effects of integration, it suggests that there is indeed sometimes
strength in numbers. How well might this result generalize to develop-
ing countries integration arrangements? We should expect that things
will look different. OPEC dominated oil supplies: The EU created a
grouping with a larger economy than the United States. No developing
country regional grouping can hope to be in this league in terms of sheer
economic power. In practice, developing country groupings have not
succeeded in negotiating as groups. During the Uruguay Round, despite
the existence of so many regional blocs of developing countries, most
did not negotiate as a group. The exceptions were customs unions that,
by virtue of having a common external tariff, are required to have a
single negotiating position.
18



POLI TI CS  AN D  PO LI CYMAKI NG
One reason why the regional blocs did not negotiate as single entities
is that shared location often does not mean shared economic interest.
The most successful negotiating blocs have been groups of countries
with common exports, whether or not they were in regional agreements.
OPEC is indeed an example of such a product-based alliance. During
the Uruguay and Tokyo Rounds the most important such bloc was the
Cairns Group of agricultural export countries, which had a common
interest in achieving trade liberalization in agriculture and which was
highly influential in negotiations.' It is suggested that it was only be-
cause of pressure from the Cairns Group that the United States main-
tained pressure on the EU.
Potentially, regional blocs can achieve a common negotiating posi-
tion by logrolling, even if the interests of member countries differ. Each
country agrees to support a long list of negotiating objectives-only one
or two of which it has an interest in-to get other countries to support
its own concerns. However, in practice, logrolling deals are easy for op-
posing negotiators to unravel by making offers that suit some members
enough to tempt them to defect. Only if the parties to a logrolling deal
are regularly making similar deals will they have an incentive to sacrifice
the short-run gains offered by defection to preserve the longer-run ben-
efits of cooperation.
The bargaining triumphs of OPEC and the EEC/EU are therefore
a tantalizing, but ultimately misleading, role model for developing coun-
try regional groups. However, this need not imply that there is no
scope for collective regional action. Even if a regional group cannot
aspire to this much economic power, it can still work toward raising
countries' visibility. After all, bargaining is not only about threats-it
is also about the discovery of mutually beneficial deals. Even if region-
alism cannot confer the power to be menacing, it can sometimes confer
the power to be noticed.
In trade bargaining, countries win gains from an agreement by mak-
ing concessions. When groups have the power to menace, they can use
their power to make fewer concessions. However, paradoxically, when
groups get together with the more modest objective of creating the power
to be noticed, the ultimate test of success is whether or not they end up
making more concessions. This is not because solidarity weakens their
bargaining power: Rather, it is that if small countries act individually
they may have so little bargaining power that they cannot negotiate at-
tractive deals. They find themselves on the sidelines, with their offers of
19



TRADE B LO CS
concessions seen as too unimportant to invite useful concessions in re-
turn. By combining their forces, they can cumulate potential conces-
sions to a significant scale and so cut more deals.
The clearest example of regional solidarity with the objective of get-
ting noticed is the Caribbean Community and Common Market
(CARICOM), the alliance of small Caribbean island-states. CARICOM
is not focused on achieving regional integration-rather its objective is
common action. Without solidarity it would be quixotic for any single
island to spend resources on international negotiations, and so it was
evident that there were gains from pooling the costs of negotiation.
CARICOM has been strikingly successful in getting its members no-
ticed. For example, they have taken the lead in formulating and articu-
lating the position of the African, Caribbean, and Pacific countries group
in negotiating the Lome Conventions. By pooling their support the
CARICOM countries succeeded in getting their nationals elected to
key international positions such as Commonwealth Secretary General
and the African, Caribbean, and Pacific countries Secretary General. They
also succeeded in negotiating a whole range of preferential market access
arrangements with Canada, the United States, and the EU.2
Africa is the region most fragmented into small economies. There is
thus a strong case for some cooperation in negotiations through bar-
gaining units, whether through participation in product groups such as
the Cairns Group, or through regional blocs. The national representa-
tion structure of the W'TO tends to favor small countries, somewhat
analogous to the United Nations, and so the existence of the WTO rein-
forces the "strength in numbers" argument for African collaboration.
However, realistically the rationale would be to get noticed rather than
to menace. Were Africa to use blocs to negotiate more effectively, it would
be able to grant more concessions and thus cut more beneficial deals
than it has done to date.
In one important respect regional integration can actually reduce the
bargaining power of small developing countries. Consider the negotia-
tion between a large multinational firm that is considering making an
investment and demands tax concessions from the government of a small
country. If the country is part of a regional trade bloc the firm can locate
in any country within the bloc and still get access to the entire market.
Governments within the bloc will tend to compete against each other to
offer tax concessions: To each individual government a little tax revenue
looks better than no tax revenue. By creating a regional trade agreement
they set themselves up for a "race to the bottom." Hence, multinational
20



P O L I T I CS AND  PO LI CYMAKI NG
firms can be expected to favor regional integration not just because it
offers larger markets, but because it enhances their bargaining power.
The only way for governments to prevent this effect of a regional trade
agreement is for it to be extended to control such tax concessions; this
goes well beyond the scope of most regional agreements.
A rare exception is a recent agreement on charges for cruise liners in the
Caribbean. Until recently, cruise ships could discharge waste without charge
at any island. Waste discharge inflicted cleanup costs on island govern-
ments, yet no individual island dared to levy a charge because the cruise
companies threatened simply to divert their ships to other islands. In ef-
fect, the island governments found themselves stuck "at the bottom" in
the charges race, because ships could choose freely between islands and
each offered the same attractions. Analytically, this is precisely the same as
the tax race within a regional trade bloc. Building on their previous exten-
sive experience of solidarity, the island governments were able to coordi-
nate the imposition of a cleanup charge. Despite threats of a generalized
boycott from the cruise companies, the governments won their struggle.
Hence, sometimes it is possible to counter the erosion of bargaining power
that regional integration implies.
2.3 Project Cooperation
C OUNTRIES CAN BENEFIT GREATLY FROM COOPERATION WHEN
they   share  resources-such   as  rivers,  fishing
J  grounds, hydroelectric power, or rail connections-or when they
join to overcome problems-such as pollution and transport bottlenecks.
Failure to cooperate in these areas can be very costly. For instance, the
uncoordinated exploitation of the Aral Sea by the five Central Asian
riparian countries resulted in one of the world's worst environmental
disasters, with the sea's desiccation, destruction of ecosystems, the
pollution of surface and ground water, and more.
However, finding equitable ways to share the burdens and benefits of
cooperation can be difficult. Countries sometimes find themselves unwill-
ing to cooperate because of national pride, political tensions, lack of trust,
high coordination costs among a large number of countries, or the asym-
metric distribution of costs and benefits. Regional cooperation agreements
are typically harder to achieve than national ones: they must be self-
enforcing because of absence of regional courts or authorities to enforce
them. The fact that these agreements must be self-enforcing reduces the
21



TRA DE B LO CS
set of feasible cooperative solutions and their benefits. International orga-
nizations, such as the World Bank-with high international credibility,
expertise to deal with the complex issues involved, and the ability to mo-
bilize financing-have often helped achieve an agreement where it might
not have been possible otherwise.3
Regional integration agreements can help, because the ties of collabo-
ration and frequent policy-level contact can raise the degree of trust among
the parties. Moreover, regional agreements also help by putting more
issues on the table and embedding them in a wider agreement, both of
which lower the size of compensatory transfers required and make it
easier to reach an agreement.
Regional integration was crucial in achieving project cooperation in
the support the Southern African Development Community (SADC) pro-
vided to the Southern African Power Pool. The Southern African Power
Pool provides for regional exchanges of electricity. Such pools offer large
potential gains because peak loads do not always coincide, so that by pool-
ing each country can meet peak demand while maintaining a smaller gen-
erating capacity. The gains over the period 1995-2010 are estimated at an
astonishing $785 million, a 20 percent saving. However, cooperation is
intrinsically difficult because each country has an incentive to underprovide
expensive spare capacity, hoping to free-ride upon the spare capacity of
others. Agreements can devise rules that forbid such behavior, but the
rules must somehow be enforced. It seems that the framework provided
by the SADC has provided this. The agreements between the EU and the
East European and Mediterranean countries also include numerous ini-
tiatives for joint cross-border projects, covering transport networks, en-
ergy, environment, and other infrastructure projects.
Thus, though the support from the international community is often
crucial for the success of cooperation agreements, regional integration is
likely to help by increasing trust between the parties and by embedding
the issues in wider negotiations where tradeoffs are more feasible and
agreements easier to reach.
2.4 Integration for Lock-in to Reform
T          HE ARGUMENTS SO FAR HAVE ALL HAD TO DO WITH
international relations, but integration can also
help domestically, as a government seeks to implement its
political agenda.
22



P OLITI CS AN D  P O LI CYMAKI N G
Reform plans can often be thwarted by the mere possibility that they
will be reversed. Adjusting to reform typically involves investments, but
these investments will not be made unless investors are confident that
the reform will persist. Confidence in the persistence of the reforms may
be low, particularly if the country has no track record of reform or, worse,
a history of reversing reform. If the investments are not made, then gov-
ernment is likely to face increasing pressure to reverse the reforms-a
"catch-22" situation (also known as policymakers' time-inconsistency
problem). To escape from this trap governments often need institutions
that enable them credibly to lock in to decisions. These are sometimes
referred to as commitment mechanisms.
How good are regional integration agreements as commitment mecha-
nisms? For trade liberalization, regional agreements are a well-designed
piece of commitment mechanism because they are built upon reciprocal
preferences: I reduce tariffs against you only if you reduce tariffs against
me. As a set of self-enforcing rules this is quite efficient, especially if trade
between members is large. The agreement reduces the likelihood of a coun-
try reversing its trade liberalization, because if it reneges on the trade pref-
erences it has granted, it can be sure that its partners will respond by can-
celing the preferential access they grant. MERCOSUR provides a good
example, where Brazil has had to back down on its attempts to deviate
from the agreements after coming under pressure from Argentina.
Paradoxically, although regional integration works best as a commit-
ment mechanism for trade policy, this is the area of policy where it is
least necessary. This is because trade policy is the only area for which
there is a well-established multilateral commitment mechanism, namely
the WTO. Any country that wishes to reassure investors that trade re-
form will not be reversed can do so very simply by "binding" its tariffs to
their reformed levels. This commits the country not to raise tariffs above
bound levels, and is enforced by a clear and straightforward set of penal-
ties. Indeed, a good case can be made that WTO enforcement mecha-
nisms are stronger than those found in many regional agreements
(Hoekman and Kostecki 1995).
But there is no equivalent to the WTO in other spheres of policy.
Thus, regional trade agreements can be useful as a commitment mecha-
nism if they are used as the foundation of reciprocal behavior on which
other agreements are then built.
For example, the wave of democratization that swept the world post-
1989 created a need for constitutional lock-in. Some regional trade arrange-
ments have explicidy added a commitment to democracy to their original
23



TR AD E B LO CS
design. MERCOSUR put its-then informal-democracy rule into prac-
tice in April 1996 when the commander of Paraguay's armed forces was
contemplating a military coup. The bloc's four presidents (with backing
from the United States and the Organization of American States) reportedly
quashed the rumored coup with a strong joint statement that democracy
was a condition of membership in the bloc. Two months later MERCOSUR
amended its charter to formally exclude any country that "abandons the full
exercise of republican institutions" (Presidential Declaration on the Demo-
cratic Commitment in MERCOSUR, San Luis, Argentina, June 25, 1996;
Talbott 1996; Survey on MERCOSUR, The Economist, October 12, 1996).
In forming free trade areas with MERCOSUR, Chile and Bolivia accepted
democracy as a condition for membership (Protocol of Ushuaia, July 24,
1998). However, this condition is only truly binding if the penalties for
violating it are severe and effectively enforced.
Even without such explicit commitments, membership in a regional
grouping can be understood to be restricted to democracies. This has
clearly been the case with the EU, and the granting of membership to
the new democracies of Portugal, Greece, and Spain is widely seen as
an example of the use of a regional group as a commitment mecha-
nism. We also see it with the eastern expansion of the EU, in which
the EU Articles of Agreement with accession candidates are proving
effective in promoting "full integration into the community of demo-
cratic nations" (Title 1, Article 2). For example, Latvia, which is a
candidate for accession, is reviewing its citizenship policies for its Rus-
sian minority to meet EU concerns about human rights (The Washing-
ton Post, July 14, 1998). The preambles of the EU association agree-
ments with Mediterranean countries include "the importance the Par-
ties attach to the principles of the UN Charter, particularly the obser-
vance of human rights, democratic principles and economic freedom,
that form the very basis of the Association" (Commission of the Euro-
pean Communities 1995, emphasis added).
Regional groups also sometimes use their potential as commitment
technologies to lock in to economic policies other than trade. The most
important example of this is probably NAFTA. Although NAFTA was
ostensibly about trade policy, its underlying motivation was the desire
on the part of both the Mexican and U.S. governments to lock in the
broad range of economic reforms that the Mexican government had
undertaken in the preceding years. The implicit agreement was that if
Mexico maintained its policies it would have access to the U.S. market
24



P O L I T I C S AN D  P O L I CYMAK I N G
and also would have a more general claim on U.S. assistance (Francois
1997). One motivation for the U.S. government was that it had a strong
interest in encouraging Mexican economic growth in order to curtail
Mexican emigration to the United States. By making the Mexican re-
forms more credible it hoped to raise the Mexican growth rate.
Do regional agreements work as commitment mechanisms? The en-
forceability of bloc rules depends on both the value of belonging to the
bloc and the credibility of the threat of action if rules are broken. If a
country suffers little from leaving-or being expelled from-the bloc, then
membership will not provide a credible way of committing to its rules.
This suggests that regional agreements between smaller low-income coun-
tries, which typically trade very little with each other, add little to credibil-
ity (Collier and Gunning 1995). Credibility will be achieved only if a
country that breaks the rules is likely to be penalized by other members of
the bloc.4 This will occur only if partner countries are sufficiently con-
cerned to be willing to act to enforce bloc rules-for example, by expul-
sion or other sanctions. For instance, Mexican membership in NAFTA
lent credibility to Mexico because the United States has a clear interest in
a neighboring country with a pool of potential emigrants. But when mem-
bers are more remote, engaged in less trade, or themselves less committed
to the rules of the bloc, then enforcement becomes less likely.
Have regional agreements worked as commitment mechanisms in
practice? For trade, the answer is broadly yes, although there have been
reversals. Most agreements involving high-income countries have sur-
vived, successfully locking in their policies. Among developing coun-
tries the record is less good; a number of regional agreements have col-
lapsed, or never moved far from paper to practice. When regional trade
agreements have been used as the foundation for other types of commit-
ment they have had some important successes, as shown by Mercosur's
continuing commitment to democracy and Mexico's persistence with
economic reform. Shortly after the NAFTA agreement in 1994 the
Mexican peso was subject to speculative attack. NAFTA was evidently
not sufficient to prevent this run on the currency. However, it probably
was instrumental in determining the policy response of both the Mexi-
can and the U.S. governments. Until 1994 the most common Latin
American response to a major run on the currency had been to impose
high trade restrictions and retreat into a controlled economy. This time
the Mexican government basically maintained the reforms, although it
raised some tariffs on non-NAFTA imports. The U.S. government also
25



TRA DE B LO CS
demonstrated that NAFTA meant more than just trade policy. It re-
sponded with a massive $15 billion rescue package, wholly out of line
with its previous behavior.
In developing countries NAFTA is often seen as the model for a con-
tinentalagreement. However, its true significance is not that it spanned
a continent but that it spanned the North and the South. By using a
major Northern economy that had a direct interest in its own perfor-
mance as its commitment mechanism, Mexico increased its credibility.
While there have been some attempts to create an equivalent arrange-
ment between the United States and Africa, the underlying politics are
less well suited. The United States does not have the same migration
interest in the growth of Africa as it has in Mexico, and the U.S. market
is far less important for Africa than it is for Mexico. A commitment
mechanism for Africa that could be equivalent to the United States for
Mexico is perhaps agreement with the EU, though there are some differ-
ences including the number and size of African economies and their
distance from the EU. There are signs that Africa-EU trade agreements
may in the future evolve to something more like NAFTA and away from
the nonreciprocal concessions of past Lome Agreements, although these
agreements have recently been extended.
Thus, paradoxically, regional trade agreements may sometimes be more
important for what they can do for the credibility of other policies, than
for what they can do for trade policy. If governments want to lock in to
trade policy they can most do so through the WTO.
2.5 Lobbying for Integration
G     OVERNMENTS' POLICYMAKING IS LIKELY TO BE SHAPED BY
the competing demands of different lobbies, which
is why firms invest heavily in lobbying bureaucrats and
politicians. Some interest groups will have much more influence than
others will. Paradoxically, if the benefits of a policy change are spread
very wide, no single beneficiary has an incentive to spend resources
lobbying (an example of the "free-rider" problem). By contrast, if the
benefits are concentrated, then it pays the few beneficiaries to spend
heavily. Because of this, lobbies tend to promote policies that cause
transfers from the many to the few, and this usually means that it is
26



P O L I T I CS AN D  PO LI CYMAKI NG
producer groups that dominate the lobbying scene. Usually also,
producers competing against imports are more powerful than exporters;
from a status quo with trade barriers, the import-substituting producers
are already in business, while export sectors may be relatively small and
undeveloped (Hillman 1989).
To producer groups, regional integration may appear an attractive op-
tion, particularly compared to unilateral nonpreferential liberalization. First,
it limits the increase in international competition to which home country
industries are subjected, since liberalization is only with the partner coun-
tries. Second, the agreement is reciprocal, with the home country obtain-
ing improved market access in partner countries' markets. Since these
markets also may be protected from competition from the rest of the world
quite large profits may be possible; (we will see in the next chapter that
preferential trade agreements can sometimes transfer tariff revenue from
the government to importing firms). Indeed a number of authors have
argued that politically sustainable agreements tend to be those that are
"trade diverting"-maintaining high external protection, delivering mini-
mal benefits to consumers, but raising returns to producer groups
(Hirschman 1981; Bhagwati 1993; Grossman and Helpman 1994,1995).
Political economy reasons may therefore make it easier for countries
to liberalize trade on a regional basis than unilaterally, or through the
multilateral process. Thus, for a government seeking trade reform, re-
gional integration may be seen as the only politically feasible way to
liberalize its economy, and perhaps also as the first step in a process of
further liberalization. The former minister of industry in Morocco, Hasan
Abouyoub, has mentioned (Abouyoub 1998) that trade liberalization
would have been infeasible without first entering into a free trade ar-
rangement with the EU. These arguments apply both for initial tariff
cuts, and for the commitment not to reverse policy. We saw in the pre-
ceding subsection that it is possible for a country to use the WTO as a
commitment mechanism, by setting its bound tariffs at low levels. But
in reality many developing countries-including those that have made
substantial trade liberalizations-have not used this option, but have
instead left bound tariff rates at high levels, often much higher than
actual tariffs. In a regional agreement the binding-to free internal
trade-is reciprocal, and thus more acceptable to the lobbies that influ-
ence the political process. Olarreaga and Soloaga (1998) provide an
empirical confirmation of this view for MERCOSUR.
27



T RAI) E  B LO CS
2.6 Conclusion: So...lntegration Is Political
TR       RADE POLICY GETS SHAPED BY THE POLITICAL NEEDS OF
governments and by the political pressures exerted
by well-funded lobbies. Adding a regional dimension to trade
policy, meets some political needs and changes lobbying opportunities.
Sometimes the political needs met by regionalism are highly valuable,
such as the lock-in to democracy or to policy reform. Regional integration
can also serve as an efficient tool to enhance security among member
countries, as has occurred in some of the largest regional agreements.
However, the political effects can sometimes be unexpected or opposite
of the aspirations, as when regional preferences create such unfair
redistributions that they provoke conflict.
It is not surprising if the economic impact of these policies, good or
bad, is not very influential in the decision to join a regional integration
agreement. Ordinary people will be made poorer or better off, but to
date these effects have usually been incidental to the political debate.
When regional integration makes ordinary people better off, the neglect
of economic analysis does not matter. But sometimes regionalism comes
at a cost. To know whether that cost is worth paying we need to know
how large it might be. For this, there is no alternative but to understand
the economic effects of regional trade preferences.
Notes
1. The Cairns Group consists of 15 developing and   over the waters of the Indus River. India and Pakistan were
industrial countries comprising Argentina, Australia, Bra-   unable to agree on a division of the Indus River Basin wa-
zil, Canada, Chile, Colombia, Fiji, Indonesia, Malaysia,   ters after the 1947 partition. In 1954, the World Bank pro-
New Zealand, Paraguay, the Philippines, South Africa,   posed a solution based on dividing the Indus and its five
Thailand, and Uruguay.                              tributaries.
2. See Andriamananjara and Schiff (2000) for a dis-  4. See Fernandez and Portes (1998) for a thorough
cussion of the costs of benefits for microstates of joint   analysis of the conditions under which a regional agree-
negotiation.                                        ment will enhance the credibility of policy reform.
3. The international community helped overcome
seemingly intractable problems between India and Pakistan
28



C HAPT ER 3
Economic Benefits and Costs
Introduction
M        EMBERSHIP IN A RIA HAS IMPLICATIONS FOR ALMOST ALL
parts of the economy. Some sectors will face opportunities
for expansion; others will contract. Some sources of income
will be boosted; others will decline. In this chapter we outline the main
economic mechanisms that bring these changes and the evidence on
their importance. We group these mechanisms into two main types, which
we refer to as competition and scale effects, and trade and location effects.
Competition and scale effects arise as separate national markets be-
come more integrated in a single unified market. The larger market per-
mits economies of scale to be achieved and brings producers in member
countries into closer contact-and competition-with each other. En-
trenched monopoly positions are eroded, promoting efficiency gains
within firms. Suppliers from nonmember countries will also experience
the change in market size and competition, inducing changes in the
pricing of their imports and in their attitude to foreign direct invest-
ment (FDI). We start this chapter (section 3.1) by examining these com-
petition and scale effects.
Competition and scale effects can occur even if the sectoral mix of
production in each country stays broadly unchanged. In contrast, trade
and location effects arise when the regional agreement changes the pat-
tern of trade and the location of production. The direction of trade
changes as imports from partner countries become cheaper, encourag-
ing consumers to substitute these for local production and for imports
from the rest of the world-phenomena known as trade creation and
trade diversion. These effects create real income changes for consumers
and producers, as well as changing government tariff revenues. We ana-
lyze these costs and benefits in section 3.2.
29



TRADE B LO CS
As the direction of trade changes, so too does the location of eco-
nomic activity within the integrating countries. Countries will see ex-
pansion of some activities and contraction of others. In some cases these
changes may not be evenly balanced, so some countries (or regions within
countries) will do better than others. Sometimes these changes promote
convergence of income levels, raising income levels in poorer countries
to the levels of richer partners. In other circumstances they may cause
divergence, with some countries gaining at the expense of others. In
section 3.3 we show how these location effects depend on the compara-
tive advantage of members of the agreement, on incentives for clustering
of activity, and on the potential for technology diffusion and adoption.
3.1 Competition and Scale
MAYANYCOUNTRIESARETOO SMALLTO SUPPORT, SEPARATELY,
activities that are subject to large economies of
scale. This might be because insufficient quantities of
specialized inputs are available, or because markets are too small to
generate the sales necessary to cover costs. Regional cooperation offers
one route to overcome the disadvantages of smallness, by pooling
resources or combining markets. In chapter 2 we saw some examples of
this in increased ability to undertake cross-border public sector projects.
There may be similar gains to be had in the private sector, although they
are likely to be achieved through quite different mechanisms, arising out
of a combination of scale effects and changes in the intensity of
competition. Note that the disadvantage of smallness can also be overcome
through unilateral trade liberalization.
Domestic Production
Small domestic markets make it difficult to produce profitably goods
that are subject to increasing returns to scale-declining average pro-
duction costs. Even if production is profitable, scale economies mean
that only one or a few producers can survive, typically with monopoly
power, leading to high prices, low levels of sales, and perhaps also high
costs. There is plenty of evidence of the relatively small number of firms
operating in most developing countries, and Rodrik (1988) reports that
30



ECON OM IC  BEN EF ITS AND  COSTS
measures of concentration (measures of firms' market power) in manu-
facturing sectors in large developing countries are typically between 50
percent and 100 percent higher than in industrial countries. However,
entry costs may also be relatively low in developing countries, imposing
competitive pressure on incumbents.'
What difference can a RIA make? In principle, a RIA combines mar-
kets, making it possible to reduce monopoly power as firms from differ-
ent countries are brought into more intense competition. This can yield
three types of gain. The first is the textbook gain from increased compe-
tition: firms are induced to cut prices and to expand sales, benefiting
consumers as the monopolistic distortion is reduced.2
The second source of gain arises as market enlargement allows firms to
exploit economies of scale more fully. In a market of a given size there is a
tradeoff between scale economies and competition-if firms are larger,
then there are fewer of them and the market is less competitive. Enlarging
the market shifts this tradeoff, as it becomes possible to have both larger
firms and more competition. For example, there might be an initial situa-
tion in which two economies each have two firms in a particular industry,
and these firms exploit their "duopoly' power, setting prices well above
marginal cost. After formation of the RIA this becomes four firms in one
combined RIA market. This increases the intensity of competition, and
possibly induces merger (or bankruptcy) of some firms, perhaps leaving
only the three most efficient firms. The net effect is increased competi-
tion, increased firm scale, and lower costs. "Triopoly" competition is likely
to be more intense than the original duopolies; and surviving firms are
larger and more efficient, so can better exploit economies of scale.
The third source of gains comes from possible reductions in internal
inefficiencies that firms are induced to make. If the RLA increases the
intensity of competition, it may induce firms to eliminate internal inef-
ficiencies (so called X-inefficiency) and raise productivity levels (Horn,
Lang, and Lundgren 1995). Since competition raises the probability of
bankruptcy and hence layoffs, it also generates stronger incentives for
workers to improve productivity, and increases labor turnover across firms
within sectors (Dickens and Katz 1987).
There is a good deal of evidence that general (nonpreferential) trade
liberalizations achieves many of these gains. A number of studies have
found that openness to trade reduces price-cost margins, an indicator of
competitive pressure in the industry (Roberts and Tybout 1996).2 There is
also evidence of an association between trade liberalization and increases
31



TRA DE  B LO CS
in efficiency, and between trade liberalization and a reduction in the dis-
persion of efficiency levels, as low efficiency firms adapt or are eliminated.4
Tybout (1999) concludes that most of the efficiency gains from openness
come from reductions in inefficiencies, rather than from scale effects.
When it comes to regional integration, we have less direct evidence.
The most extensively studied RIA is the EU, and here the static gains
from these effects have been estimated to range up to 5 percent of the
gross domestic product (GDP), with additional, and even more specu-
lative, dynamic growth effects on top (Baldwin 1989; Catinat and
Italianer 1989; McKibbin 1994). These estimates are based on extrapo-
lations of calculations from a handful of industries, and assume a sig-
nificant increase in competitive pressure.5 They are therefore appro-
priately characterized as somewhat "heroic" (Winters 1992). Even in
the EU, there is some evidence that general external trade liberaliza-
tion may be more important than regional integration in achieving
these gains. Estimates from the EU found that procompetitive effects
are largest not in markets where there is a high level of intra-EU trade,
but instead in markets where there is a high degree of import competi-
tion from firms outside the EU (Jacquemin and Sapir 1991).
Turning to developing countries, there are several arguments that sug-
gest that the potential gains may be larger than they are for high-income
countries. The small size and relatively closed structure of many devel-
oping countries mean that there is scope for fuller exploitation of econo-
mies of scale and for removing local monopoly power. A well-documented
example concerns duplication of plants in the tire industry of Central
America (Wilmore 1976, 1978). A plant in Guatemala had the capacity
to meet entire CACM demand, and there was another sizable plant in
Costa Rica. Rationalization did not occur, however-possibly because
the external tariff remained high, enabling the firms to impede effective
competition. The service sector too offers considerable potential gains
from opening to competition.
A number of studies calculated the potential (rather than actual)
gains that might be expected from the competition and scale effects.
Hunter, Markusen, and Rutherford (1992) construct a model of the
U.S. and Mexican automobile industries and simulate the possible ef-
fects of NAFTA; they predict large increases in output for Mexico,
increases in the scale of individual firms, and reductions in price-cost
margins. A study for MERCOSUR (Flores 1997) based on a similar
methodology suggests GDP gains of 1.8 percent, 1.1 percent, and 2.3
32



ECO N O M IC  B EN EF ITS AND  CO STS
percent for Argentina, Brazil, and Uruguay, respectively (the larger
economies gaining less because they are already closer to reaping econo-
mies of scale). However, these estimates are essentially predictions of
what might be expected from regional integration, rather than mea-
sures of what was actually achieved.
The message from this section is then, that regional integration
schemes offer developing countries substantial potential from these
competition and scale effects. However, the gains are not automatic,
and making sure that they are achieved involves careful policy design-
and many of these gains can also be achieved through unilateral
(nonpreferential) trade liberalization.
Market Segmentation and Deep Integration
The competition, scale, and consequent efficiency effects we have
outlined are likely to be important sources of economic benefit, al-
though they have proved difficult to quantify in practice. However,
regional integration does not necessarily secure these gains. They arise
from firms being brought into more direct and more intense competi-
tion with each other, and lower tariffs are a necessary, but not suffi-
cient condition to achieve this.
There are some extreme examples, where RIA member governments,
under pressure from industry lobbies, have deliberately acted to stop mar-
kets from being fully opened to competition. In Europe, a well-known
example is automobile distribution, where a variety of nontariff measures
restrict the ability of European consumers to engage in intra-European
arbitrage. The measures include national product standards and licensing
requirements that make it expensive for a consumer to import an automo-
bile from another European country, and a tolerance of restrictions on
competition in car distribution (Mattoo and Mavroidis 1995). In
MERCOSUR, Argentina and Brazil have negotiated an agreement on
automobile trade, whereby each individual firn is required to balance its
Argentina-Brazil trade (Bouzas 1997). This inhibits efficient reorganiza-
tion of the industry, and favors companies with plants in both countries,
discriminating against other firms, and restricting competition.
Even if competition is not being obstructed in this way, it is likely
that remaining border "frictions" will still create substantial obstacles to
effective cross-border competition. For example, within NAFTA,
33



TRADE  B LO CS
evidence from Canadian-U.S. trade suggests exports from Canadian
provinces to other Canadian provinces are some 20 times larger than
their exports to United States states at a similar distance (McCallum
1995; Helliwell 1997); other estimates suggest that the United States-
Canada border imposes barriers to arbitrage comparable to 1,700 miles
of physical space (Engel and Rogers 1996). The significance of these
border frictions is increased by the fact that firms will typically have no
desire to compete more intensely with rivals in partner countries; they
may seek to collude, tacitly if not explicitly, agreeing not to supply each
other's markets. The implication is that markets will be left "segmented,"
rather than integrated into a single unified market. In this case the gains
we have outlined above will only be partially realized.
This is one of the main arguments for pursuing "deep integration."
Liberalization of trade between countries can involve not just removing
tariff barriers, but also removing "trade chilling" contingent protection,
and other obstacles created by frontier frictions, such as frontier red-
tape, differences in national product standards and so on. The benefit of
implementing as deep a range of measures as possible-and extending
them into areas such as service trade-is that it will force firms to com-
pete directly. It was precisely these arguments that caused the EU to
embark on the Single Market Program in 1989-a far-reaching set of
measures aimed at integrating markets (box 3.1). We return to ways of
achieving this deep integration in chapter 4.
Importers and the Terns of Trade
If regional integration makes markets more competitive, then this
should be felt not only by firms inside the agreement, but also by firms
outside that export to the RIA markets. The more intense competition
may induce them to cut prices; if so, this will be a direct source of eco-
nomic gain to purchasers in the RIA (although the gain comes at a cost
to these outside firms). The effects of RIA formation on import prices is
an under-researched area, but there is now some evidence that they have
achieved this effect.
Chang and Winters (1999) show that Brazil's membership in
MERCOSUR has been accompanied by a significant decline in the
relative prices of imports from nonmember countries. They use econo-
metric techniques to investigate changes in the prices of U.S. exports
34



ECON OM I C  B EN EF ITS AN D  CO STS
Box 3.1 The Single Market Program
THESINGLEMARKETOR'EC-1992"PROGRAMAIMED             * Progress toward deregulation of the transport
at eliminating the remaining restrictions on the ex-  sectors of EU countries, including measures to
change of goods and services in the EU. The project   reduce restrictions on hauliers from one coun-
involved adoption of almost 300 measures to elimi-    try accepting loads in another.
nate intra-EU barriers. They fall into five main types.  * The opening of public procurement in EU
* Simplification and in some cases abolition of    countries to effective competition from suppli-
intra-European Community border controls.        ers in all EU countries. Measures include the
...intra-europeanC     border p ontro bs        requirement that public projects be advertised
This involves replacing border paperwork by     i h  Uwd  ulctos
an EU wide system of administernge value         i  h  Uwd  ulctos
anded EU w   osyste of radmnisteri value      * Deregulation of service sector activities, in-
added tax on cross-border transactuons.          cuigoeigfnnilsrie  ocme
* Product standards: to remove the need for        cluding opening financal services to compe-
expensive retesting and recertification of prod-  tto   n   iigsriepoiesadpo
expensive rethEsti aond reertiiati  orecod-     fessionals the right of establishment in other
ucts in each EU country, the "mutual recog-      EU cutries
nition" principle was adopted, under which a        countnes.
product that can be legally sold in any one  Source: Pelkmans and Winters (1988); Pohl and Sorsa (1992).
EU country can be legally sold in all.
to Brazil, relative to Argentinean ones. They observe a substantial fall
in the relative price of the U.S. goods for most of the period. Formal
econometric estimates suggest that these changes in relative prices are
largely due to the reduction in tariffs on Argentine exports to Brazil
compared to those on U.S. exports.
An additional test of the hypothesis is to see what happened to U.S.
export prices in Brazil relative to U.S. export prices on sales to markets
outside MERCOSUR. Figure 3.1 shows that U.S. export prices (aver-
aged over 1,356 products) in the Brazilian market declined in absolute
as well as relative terms over the integration period. Figure 3.2 shows a
similar experience for Korean exports to MERCOSUR. These are siz-
able price reductions, and are confirmed by econometric analysis, which
were also found for Japan and EU countries. They indicate that in-
creased competition in MERCOSUR markets induced exporters to
cut prices, thereby improving the terms of trade of MERCOSUR coun-
tries and yielding them a sizable welfare gain.
Is it necessary for Argentina to export to Brazil in order for prices to fall
on exports to Brazil from the rest of the world, or is a threat of increased
competition enough? Chang and Schiff (1999) find that the prices of
35



TR AD E  B LO CS
Figure :3.1 United States Export Prices to Brazil and Rest of World
(1,356 commodities)
Average prices (percent difference from 1991)
15 -
10
O _                                                                       = World
0 - ............................
-5 --                                                   ------ Brazil
-5
-10                                   ------
-15 -                 I       l        l                     I
1991       1992      1993       1994       1995       1996
Source: Chang and Winters (1999).
Figure 3.2 Republic of Korea Export Prices to Brazil and Rest of World
(99 commodities)
Average prices (percent difference from 1990)
10
5
World
-5             
-10                - "                        , '   "            ------ Brazil
-15
-20 -                   I l        l       l
1990    1991    1992    1993    1994    1995    1996
Source: Chang and Winters (1999).
36



ECON O M I C  B EN EF I TS AN D  CO STS
exports to Brazil from the rest of the world fall even for products that Ar-
gentina does not export to Brazil, implying that the threat of increased
competition may be enough to improve the terms oftrade in MERCOSUR
Foreign Direct Investment
In addition to changing the organization of local industry, if RIAs
create large markets they may also assist in attracting FDI. Foreign firms
that want to supply their product to a particular country face a choice
between serving the market by importing or by building a local plant.
The tradeoff is between the costs of tariffs and other trade barriers in-
curred on imports, and the production costs of the local plant. If the
investment is "lumpy," requiring a certain minimum level of sales to be
viable, then the scale effect of joining markets in a RIA may well tip the
decision toward FDI. The decision will be tipped further in favor of
FDI if the RIA makes the market more competitive, favoring lower
marginal cost sources of production.
There is considerable evidence that RIAs-or at least RfAs with large
markets-have succeeded in attracting FDI. Mexico perhaps provides
the best example of this, although its position as a potential export plat-
form to the United States is clearly special. NAFTA guaranteed market
access to its Northern neighbors, and this had a profound impact on
FDI, as can be seen from figure 3.3. Flows into Mexico more than doubled
in the year following the launch of NAFTA, and Blomstrom and Kokko
(1997a) argue this increase was mainly by non-NAFTA countries' firms
taking advantage of preferential access to the larger Northern market.
For example, Japan redirected part of its FDI from the United States
and Canada toward Mexico, and many projects (in the automobile in-
dustry, for example) are intended for the NAFTA continental market.
Similar phenomena have been observed elsewhere. In Europe, a ma-
jor surge of FDI accompanied the Single Market Program; the Euro-
pean Commission (The Single Market Review 4 (1) 1998) finds that the
EU's share of worldwide inward FDI flows increased from 28 percent to
33 percent during 1982-93. In MERCOSUR too, there is evidence of
significant expansion of FDI inflows. The share of the MERCOSUR
countries in the stock of U.S. FDI increased from 3.9 percent in 1992 to
4.4 percent in 1995; the inflow to each country is given in table 3.1.
37



TRAD E  B LO CS
Figure 3,3 Mexican Foreign Direct Investment, Net Inflows
Current US$ million
12,000
10,000 -
8,000 -
6,000-
4,000-
2,000-
0        i     i
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Source: World Development Indicators (various issues); Feenstra and Hansen (1997).
Table 3.1 MERCOSUR: Net Inflows of Foreign Direct Investment
(1991 IJS$ million)
Country                 1991          1992          1993          1994         1995          1996
Argentina               2,439         3,934        2,421          2,843        3,774         3,781
Brazil                  1,103        2,005         1,224          2,847        4,387         8,728
Paraguay                  84           133           105            167          166           194
Uruguay                    0             1            97           144           142           149
Sour,e. World Development Indicators (various issues).
It is tempting to assume that inward flows of FDI are beneficial, al-
though this cannot be taken for granted. External trade barriers can pro-
vide an incentive for FDI just to avoid these barriers. If the private incen-
tive to undertake FDI is created solely by the desire not to pay tariffs to
the local government, then this "tariff jumping" FDI can reduce real
38



ECON OM I C  BENE F ITS AN D  CO S TS
income, as it is possible that local production costs exceed the costs of
imports. However, there is evidence that FDI can play a role in stimulat-
ing local production in related industries, in transferring technology, and
in raising productivity in neighboring firms. (For a survey of evidence,
see Blomstrom and Kokko 1997b, and Saggi 1999.) These benefits gen-
erally outweigh the costs associated with "tariff jumping."
Competition and Scale; Conclusions
Pulling together the strands of this section, we see that there are sig-
nificant potentialgains to be achieved, but that these depend on securing
effective competition. It can easily be obstructed, and policy is required
to prevent this from happening. We return in chapter 4 to a more de-
tailed study of the policies that are needed if effective competition is to
be achieved.
3.2 Trade and Location: The Pattern of Trade
R  <  EGIONAL INTEGRATION WILL CHANGE RELATIVE PRICES IN
member economies. Imports from  partner countries
will become cheaper due to the elimination of tariffs, and in
response demand patterns will change, causing changes in the flow of
trade and in output levels in many sectors. What do we know about
these changes, and what are their economic effects? These are intrinsically
multi-industry or "general equilibrium" issues, involving expansion of
some sectors, contraction of others, and relocation of industries from
country to country. However, it is necessary to start discussion with the
mechanics of preferential trade liberalization in a single industry-the
analysis of trade creation and diversion first put forward by Viner (1950).
The classical source of gains from trade is that global free trade allows
consumers and firms to purchase from the cheapest source of supply, hence
ensuring that production is located according to comparative advantage.
In contrast, trade barriers discriminate against some (foreign) producers
in favor of domestic suppliers. This induces domestic import-competing
producers to expand, even though their costs are higher than the cost of
imports, which in turn starves domestic export sectors of resources, raises
their costs, and causes these sectors to be smaller than they otherwise would
39



TRADE  B LO CS
be. Since a RIA liberalizes trade, reducing at least some of the barriers,
doesn't it follow that it too will generate gains from trade? Unfortunately,
as Viner pointed out, the answer is, "Not necessarily." The gains-from-
trade argument tells us what happens if all trade barriers are reduced, but
need not apply to a partial-and discriminatory-reduction in barriers,
as in a RIA. This is true because discrimination between sources of supply
is not eliminated, it is simply shifted. If partner country production dis-
places higher cost domestic production, then there will be gains-trade
creation. It is also possible that partner country production may displace
lower cost imports from the rest of the world-trade diversion.
Sourcing Imports: Trade Diversion
To understand the effects of discrimination, it is helpful to think
through a simple example. Suppose that a country can import a good
from a potential partner country at $105 per unit, and from the rest of
the world at $100, and that in both cases the country pays $10 in duty,
making the prices paid by consumers $115 and $110 respectively. In
this situation consumers obviously purchase from the rest of the world
and pay $110. If the country joins a RIA with the partner, imports
come in duty-free so the price consumers pay for imports from the
partner country falls to $105, while imports from the rest of the world
still cost consumers $1 10. Consumer choices are obvious: They switch
to the partner country, buying the $105 good and saving $5. But the
government now loses $10 per unit (the revenue it was getting on each
unit of imports from the rest of the world), so the net effect for the
country is a loss of $5-the RIA has reduced real income. Another
way of putting it is that the country (not the consumers) used to pay
$100 per imported unit, and now pays $105. This is the deleterious
welfare effect of "trade diversion."
This is just an example, and circumstances like this clearly will not
apply in all sectors-there are some where partner country costs are less
than are those in the rest of the world, and others where the country
under study is an exporter-but the example makes an important point.
Can we identify circumstances where trade diversion is more or less likely
to be a problem?
First, notice that trade diversion can occur only if the country has a
tariff on imports from the rest of the world, and that the cost of trade
40



ECO N OM IC  BEN EF ITS AND  COSTS
diversion cannot exceed the height of this external tariff. In the previous
example, if the external tariff was initially low the loss of tariff revenue
would be small, and if the external tariff were cut, the switch in supply-
source would not occur. One clear policy implication is that member
countries should lower external tariffs as much as possible.
Second, trade diversion arises only if partner country costs are out of
line with costs and prices in the rest of the world, this will not be the case
if the partner itself has low trade barriers. For example, if the partner had
a duty of just $2 per unit, then prices and costs in the partner country
could not exceed $102 (the price at which imports from the rest of the
world would be sold in the partner country). Preferential liberalization
would then cost the government $10 but save consumers $8, creating a
net loss of just $2 per unit, and mitigating the cost of trade diversion.
However, as discussed in chapter 4, the country with high trade barriers
will impose rules of origin to prevent trade deflection (where imported
goods are re-exported from the low- to the high-tariff country), and
prices could still differ by more than $2.
Third, our example is of a rather artificial and "frictionless" trade;
in reality products from different countries are not perfect substitutes,
and trade faces transport costs and other barriers apart from tariffs.
How does this change the situation? The fact that products are less
than perfect substitutes means that the change in sourcing of imports
will be less sharp than in the example, again mitigating the costs of
trade diversion. The presence of transport costs means that countries
that are close may have lower costs of supply than more distant coun-
tries. This is the "natural trading bloc" argument (Wonnacott and Lutz
1989; Summers 1991) and means forming a RIA with close countries
may be less prone to costly trade diversion than forming one with
more distant countries. This argument has not been resolved: the op-
posite view is provided in Bhagwati and Panagariya (1996), while Schiff
(1999) argues that neither view is correct.
Trade diversion is more than a theoretical possibility-the best-known
example is the EU's Common Agricultural Policy. This involves a price
structure for agricultural products designed to divert consumer purchases
toward EU farmers and away from non-European suppliers. Some of
the money is a transfer to farmers' incomes, and some is, in economic
terms, wasted, because food is produced which could be more cheaply
purchased on world markets. Messerlin (1998) estimates the cost of this
protection at, conservatively, 12 percent of total EU farm income.
41



TR A D E B LO CS
An example from NAFTA concerns clothing. Following the "Tequila"
crisis, Mexico increased tariffs on non-NAFTA imports of clothing from
20 percent to 35 percent in March 1995, just as it was reducing tariffs
on NAFTA imports. Mexican imports from the rest of the world fell by
66 percent between 1994 and 1996, while those from the United States
increased by 47 percent. Similarly in the U.S. market, imports fromn Asia
fell while imports of clothing and finished textiles from Mexico and
from Canada increased by more than 90 percent (USITC 1997).
A particularly disturbing possibility is that trade diversion may oc-
cur in capital goods or other goods used as inputs in production; this
would reduce production efficiency, and possibly slow the transfer of
technology to the country. For example, Madani (1999) examines the
effect of intermediate goods imports in three Andean Pact countries
(Bolivia, Colombia, and Ecuador) from the early 1970s to 1994. She
finds that imports of intermediate goods from the rest of the world
(primarily industrial countries) tend to raise growth while intrabloc
imports do not have this effect. She also reports that the share (al-
though not the level) of extrabloc imports in total trade has fallen,
suggesting that regional integration might have a negative effect on
growth. Her findings suggest that from the viewpoint of technology
and growth, an agreement with a large industrial country is superior to
one with a developing country.
Finally, note that a RIA between two small developing countries is
likely to only generate trade diversion and no trade creation. This can be
seen most clearly in the case of homogeneous goods. In each member
country domestic consumer prices are fixed at the world price plus the
import tariff. Since these do not change with integration, consurmption
does not change. However, production increases because each country
can now sell to the partner without tariff. Thus, each member country
substitutes cheaper imports from the rest of the world with more expen-
sive partner imports. The outcome is trade diversion and a loss for both
countries (Schiff 1997).
Exports and Trade Diversion
The focus of our discussion of trade diversion has been on imports.
What about exports? Is an importer country loss due to trade diversion
just the other side of an exporter gain, in which case the RIA as a whole
42



ECO NO MI C  BEN E F ITS AND  COSTS
would be better off? The answer to this is that there may be exporter
country gain, but it is less, per unit, than importer country loss.
Recall that in our example above, consumers switched to imports
from the partner country. Partner country export sales expand, but
how much of a gain is this for the partner country? If exports are just
selling at cost ($105 in the example), then selling more of them does
not raise income in the partner country.6 If, however, they are selling
above cost, then there will be a real income gain. But how much higher
than cost can the price go? The answer is that the price cannot go
above $110 (if it did, consumers would switch back to buying from
the rest of the world), so the exporter country gain per unit cannot
exceed the gap between the price of imports from the rest of the world
and costs (maximum $5 = $110 - $105).
This line of reasoning suggests another way of thinking about trade
diversion. Returning to our example (for the last time), the government
has given up $10 of tariff revenue per unit. We can see where this has
gone: $5 per unit goes to the higher cost of producing partner country
imports compared to the cost of imports from the rest of the world, and
the remaining $5 is divided between domestic consumers and partner
country firms, depending on whether these firms are able to raise their
prices in response to having preferential access to the domestic market.
It is often argued that an advantage of a RIA (over unilateral liberaliza-
tion) is that firms benefit from preferential access to partner markets.
This is true, but we now see that it comes only at the expense of con-
sumers and government revenue.7 The RIA acts as an inefficient way of
transferring some of the country's tariff revenue either to domestic con-
sumers or to partner country producers.
Transfers are important in North-South RIAs because developing
countries risk losing from a RIA with the North. The reason is that
developing countries typically have higher tariffs than industrial coun-
tries. Consequently, the industrial member is likely to gain more from
increased access to the partner's market than the developing member.
The latter can resolve this issue by unilaterally lowering its tariffs.
Government Revenue
For many developing countries trade taxes are an important source of
government revenue, and membership in a RIA leads to loss of tariff
43



TRADE B LO CS
revenue. This arises directly-as tariffs on intra-RIA trade are reduced-
and also indirectly, when trade diversion occurs, such as when importers
switch away from external imports subject to tariffs.
Loss of government revenue underlies the trade diversion argument, as
we saw in the preceding subsections. However, if the government is con-
strained in its alternative revenue sources, then a loss of tariff revenue can
be particularly damaging.8 Many developing countries are heavily depen-
dent on trade taxes as a source of revenue, with some African countries
raising as much as one-half of government revenues from trade taxes. In
practice, how much revenue has typically been lost by RIA formation? In
many cases we see that larger amounts of revenue have been lost in coun-
tries that are less dependent on trade taxes. This paradox arises from the
fact that intra-RIA trade volumes are typically very high in RlAs where
dependency on trade taxes has been quite low (such as the EU), while
countries with higher trade tax dependency have often formed RLAs with
countries with whom they have relatively little trade.
However, there are exceptions to this. Cambodia derived 56 per-
cent of its total tax revenues from customs duties prior to its entry into
the ASEAN free trade area, with two thirds of rhese levied on imports
from ASEAN countries (Fukase and Martin 1 999c). Entry into ASEAN
provided a powerful stimulus for the introduction of a value added tax
in early 1999. In the SADC also, where some countries are heavily
dependent on trade with South Africa, substantial amounts of revenue
are involved. Table 3.2 gives estimates of the revenue cost of going to
free internal trade, and we see that this will approximately halve cus-
toms revenue in Zambia and Zimbabwe, losing the governments 5.6
percent and 9.8 percent of government revenue respectively. These are
very substantial revenue losses, and point to the need to ensure that
alternative tax systems are in place before removing sources of trade
tax revenue.
Trade Flows: The Evidence
Trade diversion increases intrabloc trade at the expense of trade with
outside countries, while trade creation does not have this negative effect.
If we look at overall trade flows, what evidence is there of trade diversion
relative to creation? Looking at the raw numbers we typically see expan-
sions both in trade within the bloc, and in external trade, suggesting
44



ECON OM I C  B ENEF I TS  AN D  COSTS
Table 3.2 Customs Revenue Collected as a Percent of Total Govemment Revenue in 1996 and the
ImpRications of a Free Trade Area for SADC Members
Estimated change in customs duty
Customs duty as perrent        Percent custom        Percent total
Member country                              of total tax revenue             duty              tax revenue
Malawi                                            14.3                      -36.7                 -5.3
Mauritius                                        29.8                       -18.2                 -5.4
South Africa                                      3.6                          4.9                  0.2
Tanzania                                         24.0                        -8.3                 -2.0
Zambia                                            12.3                      -45.3                 -5.6
Zimbabwe                                          18.4                      -53.3                 -9.8
Note. The FTA assumes Free Trade on intra-SADC trade. The projections assume that each country's average tariff rates against SADC
members are zero. There are discrepancies between the duty revenue reported by customs departments and that reported in budget
numbers. For example Malawi reported FY96 duty revenues of 1,505.2 and 2,028.7 million kwacba against the 615 million reporred by
customs. For consistency, we have used the numbers reported by customs.
Source Staff calculations, IMF.
that there is no evidence of trade diversion. However, looking at the raw
data alone fails to distinguish the effects of regional integration from
other economic changes-including in some cases external trade liberal-
ization. To identify the effects of the RIA the researcher must try to
control for these other changes, and this can be done with varying de-
grees of sophistication. When we include these controls we find that
there is some evidence of trade diversion as well as trade creation.
The raw data on intra-RIA and extra-RIA trade for nine developing
country RIAs before (one year before implementation) and after (five
years after implementation) is given in figures 3.4-3.6. We see increases
in intra-RIA imports for all cases (figure 3.4), although perhaps the most
startling thing from the figure is how little trade there is within some of
the RIAs; Union Douaniere et Economique de l'Afrique Centrale
(UDEAC) members partner trade as a share of GDP trebled, but only
from 0.24 percent to 0.79 percent. For trade diversion, we look at the
extra-RIA trade (figure 3.5) for evidence of declines, but here too we see
increases, typically around much higher trade volumes. Looking at the
ratio of intra-RIA imports to external imports (figure 3.6) we see that
the share of intra-RIA imports expanded relative to external for seven of
the nine RIAs (the exceptions are CARICOM and the GCC). However,
since this was on the basis of rising volumes of both sorts of trade, it
provides no evidence of trade diversion.
45



TRADE  BLOCS
Figure 3.4 Intra-RIA Imports as Share of GDP
Percent
10
8
6                                                                                            U i _   1E Before
4                                             -                                               n After
MERCOSUR Andean I CACM 11 Andean 11 CARICOM   CEAO    UDEAC    AFTA           GCC
Figure 3.5 Extra-RIA Imports as Share of GDP
Percent
90
70  -                                                                                         U Before
50 -                                                                                          El After
so -                       n                            n                  11   fI
40-
30
20
10                                                     I          ]    ]   j 
0-           1
MERCOSUR Andean I CACM 11 Andean 11 CARICOM   CEAO'   UDEAC    AFfA           GCC
Figure 3.6 Ratio of Intra-RIA Imports over Extra-RIA Imports
intra-RIA over Extra-RIA Imports
0.30 -
0.25 -      _                                                                                      Before
0.20  J-                                                                                        El After
0.15 
0.10 
MERCOSUR Andean I CACM 11 Andean 11 CARICOM   CEAO'   UDEAC    AFrA           GCC
a. Economic Community of West Africa
Note. For MERCOSUR this encompasses two years 1991 and 1996; Andean Pact 1 1968 and 1974; Andean Pact 11 1990 and 1996; CACM 11 1990
and 1996; CARICOM 1972 and 1978; the Economic Community of West Africa 1965 and 1971; AFTA 1991 and 1996; and the GCC 1980 and 1986.
Source: U.N. COMTRADE data.
46



ECO N OM I C  B EN E F ITS AN D  C OS TS
The standard way to control for other effects is to build an economet-
ric model of trade, and see whether the estimated relationships change
as a consequence of implementing the RIA. The usual model for such
purposes is the gravity model, which estimates bilateral trade between
countries, generally for a sample of many countries and for several dif-
ferent dates. It explains trade between pairs of countries as a function of
their GDPs (larger economies trade more), populations, the distance
between them (as a proxy for transport costs, cultural similarity and
business contacts), and physical factors such as sharing a land border,
and being landlocked or an island. Researchers add to the list dummy
variables that capture whether or not countries are in a particular RIA. If
these show up positively for pairs of countries in a RIA, then they indi-
cate that these countries trade more than would be suggested by the
other factors. A fall in the value of a dummy for trade between a mem-
ber and nonmember is indicative of trade diversion, particularly if the
fall shows up after formation of the RIA.
Using this technique, Bayoumi and Eichengreen (1997) find that the
formation of the EEC reduced the annual growth of member trade with
other industrial countries by 1.7 percentage points, with the major at-
tenuation occurring over 1959-61, just as preferences started to bite.
Cumulating the decline in growth over 1957-73 gives lost exports to
the rest of the world of $24 billion in 1973.
A recent example of this approach is work undertaken by the World
Bank investigating nine major blocs over 1980-96 (Soloaga and Win-
ters 1999a,b). Figures 3.7a and 3.7b summarize the estimates of the
trade effects in 1980-82, 1986-88, and 1995-96. A positive value on
the vertical axis of these figures indicates that a country is trading more
than would be suggested by other factors. Looking first at figure 3.7a we
see that the EU, European Free Trade Association (EFTA), and NAFTA
had relatively high levels of extrabloc trade, but that these coefficients
fell over the period. This suggests trade diversion was occurring. Surpris-
ingly, the change in the coefficients on intrabloc trade are generally smaller,
and in some cases negative.
Figure 3.7b looks at four blocs for which the picture is rather differ-
ent. Extrabloc trade is generally lower for these blocs, but there is no
evidence of trade diversion taking place during the period. Indeed, for
ASEAN there is substantial increase in the coefficients for extrabloc trade,
accompanied by a fall in that on intrabloc trade.
What do we learn from these studies? It is extremely difficult to
control for other determinants of trade, but once we do there appears
47



T RAED E  B LO CS
Figure 3.7a RIA's Trade within and across Borders: Evidence of Diversion
(Gravity model estimates over three periods; 1980-82, 1986-88, 1995-96)
EU
1.5 -
1.0 
1.0 -             _                                                   Intra-bloc trade
0.5 -
0.0 -                                                           ------ Extra-bloc trade
-0.5  - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
-1.0 -                           I                          I
1980-82                   1986-88                    1995-96
EFTA
1.0 -
0.5 -                                                               Intra-bloc trade
0.0 -            ---- ---....                                          Extra-bloctrade
-0.5 -ll
1980482                   1986-88                    1995-96
NAFTA
1.5 -
1.0 -                               -                                  Intra-bloc trade
0.5 ---
------.                    .................  -- ----- Extra-bloc trade
0.50-
1980-82                   1986-88                    1995-96
Source: Soloaga and Winters (1999a,b).
to be weak evidence that external trade is smaller than it otherwise
might have been in at least some of the blocs that have been researched.
However, the picture is sufficiently mixed that it is not possible to
conclude that trade diversion has been a major problem. Furthermore,
we cannot infer that trade diversion has been economically damaging
without information on relative costs and tariff structures, which are
not revealed in this sort of aggregate exercise.
48



ECO N OM I C  B EN EF ITS AN D  COSTS
Figure 3.7b RIA's Trade within and across Borders: No Evidence of Diversion
(Gravity model estimates over three periods; 1980-82, 1986-88, 1995-96)
CACM
4
2                                  -                                 Intra-bloc trade
1I-  ~~~~~~~~~~~~~~~~                  ~             ~~~~~ ~ ------ Extra-bloc trade
0
1980-82                   1986-88                    1995-96
Andean
2
1                                                             _____ Intra-bloc trade
----- Extra-bloc trade
1980-82                   1986-88                    1995-96
MERCOSUR
27
Intra-bloc trade
-1                                                             ------ Extra-bloc trade
-2 - 4
1980-82                   1986-88                    1995-96
ASEAN
2
1                - -              _____ Intra-bloc trade
- - -1;-;- -r -- ------ Extra-bloc trade
-1
1980482                   1986-88                    1995-96
Source: Soloaga and Winters (1999a,b).
49



T RA D E B LO CS
Computable Equilibrium Studies
Although the gravity equations control for some of the other fac-
tors determining trade flows, they cannot control for all of them. In
addition, they do not contain the details about tariff rates and prod-
uct supplies and demands that are needed to establish whether changes
in trade flows are really beneficial or damaging. An alternative ap-
proach that enables the researcher to do this is to construct a full
computer model of the economies under study, and then simulate the
effects of the policy changes associated with the RIA. Such a model
typically contains a great deal of microeconomic detail, so it can be
used to predict changes in production in each sector and changes in
factor prices and real incomes.
Models of this type come in increasing degrees of sophistication as
researchers have refined technique.9 "First generation" models assume
that all markets are perfectly competitive, so the costs and benefits of
RIA membership arise only from trade diversion and trade creation (the
effects discussed in section 3.2). "Second generation" models include
increasing returns and imperfect competition, so incorporate some of
the scale and competition effects outlined in section 3.1. "Third genera-
tion" models contain some dynamics, allowing for capital accumula-
tion, and sometimes also technical progress.
The conclusions from these models are, broadly, that there are gains
from regional integration, but the gains are small (Francois and Shiells
1994; Harrison, Rutherford, and Tarr 1994). In the first generation mod-
els the interaction between trade diversion and trade creation brought
effects that were typically very small-a fraction of 1 percent of GDP.
Second generation models generally increased this somewhat, to around
2-3 percent of GDP. Third generation models increased the gain fur-
ther, to approximately 5 percent of GDP.
The strength of these models is that they have sufficient microeconomic
structure to enable the effects of a policy change to be traced out in detail,
and its real income effects to be calculated. They are also often used for
prediction-to estimate the likely effects of a policy change before it is
implemented. But they have the major weakness that they are not usually
fitted to data as carefully, nor are they subject to the same statistical testing
as econometric models. The cost of the microeconomic detail is a com-
plexity that makes rigorous econometric estimation impossible.
50



ECO NOM IC B ENE FITS AND  COSTS
3.3 Trade and Location: Convergence or
Divergence?
R       EGIONAL INTEGRATION WILL LEAD TO RELOCATION OF
economic activity; industries will expand in some
countries and contract in others, and as this happens demand
for labor and real income levels will change. How will this affect member
states, and which countries are likely to be gainers and which losers from
this process?
There is an empirical paradox here that needs to be explained. There
is evidence from the European experience that RIA membership is asso-
ciated with convergence in the income levels of different countries. The
overall dispersion of income levels in the EU to the mid-1980s has been
studied by Ben-David (1993), from which figure 3.8 is drawn. The ver-
tical axis of the figure measures the dispersion of income levels in Eu-
rope, and clearly shows an almost continuous convergence, from 1947
(when the BeNeLux Customs Union was created), through 1951 (the
formation of Economic Coal and Steel Community), 1957 (creation of
the EEC), 1962 (when quotas were eliminated), 1968 (when internal
tariffs were removed) to 1981. Income differences narrowed by about
two-thirds over the period, due mainly to more rapid growth of the
lower-income countries.
The most interesting features of the more recent experience are the
strong performance of Ireland, Portugal, and Spain, who have made sub-
stantial progress in closing the gap with richer members of the EU.
Whereas in the mid 1980s these countries' per capita incomes were,
respectively, 61 percent, 49 percent, and 27 percent of the income of the
large EU countries,'" by the late 1990s the numbers had risen to 91
percent, 67 percent, and 38 percent. This convergence did not take place
in Greece, although it joined the EU earlier than Portugal and Spain,
because Greece did not implement the necessary reforms after joining
the EU. This suggests that even though integrating with a large and
advanced region is potentially beneficial, economic reforms in the poorer
country are needed in order to capture these benefits.
While European experience suggests convergence, the experience of
most developing country RIAs does not. Indeed, there are several examples
of integration being blamed for divergence of economic performance, such
as the experience of the East African Community and East and West Paki-
51



T RA D E  B LO CS
Figure 3.8 Incomes Converge as EEC Integrates
Income differences
(annual standard deviations of log incomes)
0.34 - 1947: BeNeLux Union Created
1951: Creation of the Economic Coal and Steel Community
1957: Creation of EEC
0.28 -              1959:Tariff and quota reforms initiated
1962: Final quota elimination
1968: Final elimination of tauiffs
0.22 -                                     1973: Common Market expands to 9, Arab Oil Embargo
1981: EEC grows to 10
0.16-
0.10    -     I       I       I       I       I      I
1950   1955   1960   1965   1970   1975   1980
Source: Ben-David (1993).
stan, discussed in chapter 2. The concentration of manufacturing in the
old East African Community (where the Nairobi region gained at the
expense of manufacturing in Uganda and Tanzania) has been extensively
studied (Hansen 1969). Uganda andTanzania contended that all the gains
of the East African Common Market were going to Kenya, which was
steadily enhancing its position as the industrial center of the Common
Market, producing 70 percent of the manufactures and exporting a grow-
ing percentage of them to its two relatively less industrial partners. By
1958, 404 of the 474 companies registered in East Africa were located in
Kenya, and by 1960 Kenya's manufacturing sector accounted for 1]0 per-
cent of its gross national product (GNP), against 4 percent in the other
two states. The community collapsed in 1977, because it failed to satisfy
the poorer members that they were getting a fair share of the gains.
More recent examples include the concentration of industry, commerce,
and services in and around Guatemala City and San Salvador in the Cen-
tral American Common Market (due to lack of data in the early years,
52



ECO N O M I C  B EN EF I TS  AN D  CO S TS
Costa Rica is not included) and Abidjan and Dakar in the Economic
Community ofWest Africa. Figures 3.9 and 3.10 indicate how dominant
these locations have become in manufacturing in their regions.
In this section we shed light on these differing experiences, by ad-
dressing the question, how can regional integration lead to relocations
of economic activity between member countries?
Internal and External Comparative Advantage
To think about how industry will relocate within the RIA we look first
at the comparative advantage of RIA members relative to each other, and
relative to the rest of the world. It turns out that comparative advantage
alone can go a long way toward explaining the different experiences of
different RlAs, although these forces are augmented by agglomeration and
technology transfer, to which we turn in the following subsections.1"
Let us start by thinking of two developing economies that both
have a comparative disadvantage in manufactures relative to the rest of
the world, but the disadvantage is less for one of them than the other.
Figure 3.9 CACM, Formed in 1960, Manufactures Value Added
(percent of total)
1965                                              1980
Nicaragua 20%       El Salvador 33%              Nicaragua 20%      El Salvador 21%
Honduras 11%                                      Honduras 12%
Guatemala 36%                                      Guatemala 47%
1996
Nicaragua 6%
Honduras 12%
El Salvador 41%
Guatemala 41
Sourre:Venables (1999).
53



TRADO E  B LO CS
Figure 3.10 CEAO, Fonned in 1974, Manufacturing Value Added
(percent of total)
1972                                               1980
Benin 9%                                            Benin 5%
Cote d'lvoire 41 %          Burkina Faso 12%        Cote d'lIvoire 57 %
Burkina Faso 19%                                            Niger 4%
Mauritania 3%
Mali 5%
Niger 8%
Mauritania 2%                                       Senegal 14%
Mali 7% Senegal 14%
1997
Benin 5%
Burkina Faso 13%        Cote d'lvoire 52 %
Niger 3%
Mauritania 3%
Mali 5%
Senegal 19%
Sourc: Venables (1999).
Kenya and Uganda can serve as examples. Their comparative disad-
vantage in manufactures could come from many alternative sources-
technological, geographical or institutional differences-but let us sup-
pose that it is because of different endowments of capital: Kenya has
little capital per worker relative to the world average, and Uganda has
even less. The initial position is one in which both Kenya and Uganda
have some manufacturing, serving local consumers and surviving be-
cause of relatively high tariff protection.
What happens if these two countries form a RIA? Since Kenya has a
comparative advantage in manufacturing (relative to Uganda, but not
relative to the rest of the world), it will draw manufacturing production
out of Uganda, so consumers in both countries will be supplied with
manufactures from Kenya. This moves Kenya's production structure fur-
ther away from its true comparative advantage, while moving Uganda's
closer. What are the effects of this on real income? Surprisingly, Kenya
will gain from the relocation, and Uganda may lose (and will certainly
do less well than Kenya). The reason is that Uganda is suffering trade
diversion-some manufactures that were previously imported from the
rest of the world are now imported from Kenya. But for Kenya there are
54



ECO N O M I C  B EN E F ITS  AN D  C OSTS
gains from being able to supply manufactures in the Ugandan market,
protected from competition with the rest of the world.
There is a general argument here, which is that countries with com-
parative advantage closer to the world average do better in a RIA than do
countries with more extreme comparative advantage. Interposing the "in-
termediate" country between the "extreme" one and the rest of the world
distorts the extreme country's trade, causing it to switch import supplier.
But the intermediate country does not experience this switch in supply; its
trade with the "extreme" country and with the rest of the world are less
close substitutes, and therefore less vulnerable to trade diversion.
A further implication follows. A RIA between two poor countries
will tend to cause their income levels to diverge, but a RIA between
two rich ones will tend to cause convergence. The logic can be seen
from figure 3.11. The vertical line measures each country's endow-
ment of capital per worker; countries higher up the line have a greater
comparative advantage in manufacturing, and also higher initial per
capita income. For two countries below the line, we see the extreme
country losing, and the intermediate country gaining-Uganda and
Kenya, as marked. Similarly, for countries above the line, the extreme
Figure 3.11 Convergence and Divergence of Real Incomes
France         |        Direction of real
income change
Spain         ¶
Capital to labor              Wol average
endowment
-  Kenya
- Uganda
Source: Venables (1999).
55



T RA El E B LO CS
country loses and the intermediate gains (labeled France and Spain on
the figure). However, in the former case the extreme country is the
poorer of the two, and in the latter it is the richer. The same basic
forces therefore mean that regional integration between rich countries
causes their incomes to converge, whereas integration between poor
ones causes divergence.
The two cases analyzed above were for a pair of low-income coun-
tries and a pair of higher-income countries. What about regional agree-
ments that link high- and low-income countries? The mechanism driv-
ing the changes is simply relocation of industry in response to differ-
ences in factor endowments, and associated differences in factor prices.
The changes might be particularly large, and particularly beneficial, for
a lower-wage economy in a RIA with an industrialized and high-wage
economy. We have already seen how Mexico acts as a platform for FDI
to serve the U.S. market, and there is substantial evidence of the reloca-
tion of manufacturing production from the United States to Mexico
(Feenstra and Hanson 1997).
There is also evidence of a similar process underway in Europe, both
within the EU and also in its relationship with some of the transition
economies, with relatively labor intensive activities moving to lower-
wage economies, and promoting convergence in wage rates. This may
also become more important over time, as new technologies make it
easier to fragment production processes-or "slice up the value chain"-
moving labor-intensive elements of the process to lower-wage econo-
mies. Often this occurs through FDI, and within production networks,
closely linked networks involving intrafirm trade, or trade between firms
and established suppliers.
Agglomeration
Comparative advantage is not the only force that drives relocation of
activity in a RIA. As economic centers start to develop, so "cumulative
causation" mechanisms come into effect, leading to clustering (or ag-
glomeration) of economic activity, and extending the advantage of loca-
tions that have a head start.12
Spatial clustering of economic activities is all-pervasive. Cities exist
because businesses, workers, and consumers benefit by being in close
proximity. Particular types of activity are frequently clustered, the most
56



ECON O M I C  B EN EF I TS AN D  C O STS
spectacular examples being the electronics industries of Silicon Valley,
cinema in Hollywood, and the concentration of banking activities in
the world's financial districts. Clustering also occurs in many manufac-
turing industries-for example U.S. automobile manufacturing in the
Detroit area, or industries such as medical equipment, printing machin-
ery, and others studied by Porter (1998).
Clustering or agglomeration typically arises from the interaction be-
tween "centripetal" forces, encouraging firms to locate close to each other,
and "centrifugal" forces, encouraging them to spread out. The centrip-
etal forces are usually classified in three groups (Marshall 1920). The
first are knowledge spillovers, or other beneficial technological exter-
nalities that make it attractive for firms to locate close to each other-in
Marshall's phrase, "The mysteries of the trade become no mysteries, but
are, as it were, in the air." The second are various labor market pooling
effects, which encourage firms to locate where they can benefit from
readily available labor skills-perhaps by attracting skilled labor away
from existing firms.
The third centripetal force arises from "linkages" between buyers and
sellers. Firms will, other things being equal, want to locate where their
customers are, and customers will want to locate close to their suppliers.
These linkages are simply the "backward" (demand) and "forward" (sup-
ply) linkages of Hirschman (1958). They create a positive interdependence
between the location decisions of different firms, and this can give rise to a
process of cumulative causation, creating agglomerations of activity. '3
These centripetal or agglomeration forces can operate at quite an ag-
gregate level, or can be much more narrowly focused. For example, ag-
gregate demand creates a backward linkage, drawing firms from all sec-
tors into locations with large markets. Other forces affect broad classes
of business activity-providing basic industrial labor skills, or access to
business services such as finance and telecommunications. In contrast,
knowledge spillovers affecting particular technologies, or the availability
of highly specialized inputs might operate at an industry level. In this
case the forces work for clustering of the narrowly defined sector, rather
than for clustering of manufacturing as a whole.
Pulling in the opposite direction are "centrifugal forces," encourag-
ing the dispersion of activity. These include congestion, pollution, or
other negative externalities that might be associated with concentra-
tions of economic activity. Competition for immobile factors will de-
ter agglomeration, as the price of land and perhaps also labor is bid up
57



TRAD E  B LO CS
in centers of activity. Also, there are demands to be met from consum-
ers who are not located in the centers of activity; dispersed consumers
will encourage dispersion of producers, particularly if trade barriers or
transport costs are high.
How might the balance between centripetal and centrifugal forces be
upset by membership of a RIA? Can membership cause, or amplify, the
clustering of economic activity, and if so might it widen income differ-
entials between partner countries?
By reducing trade barriers, membership in a RIA makes it easier to
supply consumers (or customers more generally) from a few locations.
This suggests that the balance of forces may be tipped in favor of ag-
glomeration, although the ensuing relocation of industry could develop
in several different ways.
One possibility is that particular sectors become more spatially con-
centrated, and this is likely if the centripetal forces act at a quite narrow,
sectoral level. For example, industries in the United States are much
more spatially concentrated than in Europe (even controlling for the
distribution of population and manufacturing as a whole), suggesting
that regional integration in Europe could cause agglomeration at the
sectoral level (for example, Germany gets engineering, the UK financial
services, and so on). The possibility that this might happen is generating
some concern in Europe, although evidence for it is so far rather weak
(see Midelfart-Knarvik and others 1999). If it does happen it will create
considerable adjustment costs-as the industrial structure of different
locations changes-but aggregate gains, as there are real efficiency gains
from spatial concentration. This sectoral agglomeration need not be as-
sociated with increases in cross-RIA inequalities; each country or region
may attract activity in some sectors.
An alternative possibility is that, instead of relatively small sectors
each clustering in different locations, manufacturing as a whole comes
to cluster in a few locations, de-industrializing the less-favored regions.
Under what circumstances might this be the outcome? It will be rela-
tively more likely to occur if manufacturing as a whole is a small share of
the economy. This is because fitting the whole of manufacturing in one
(or a few) locations is then less likely to press up against factor supply
constraints and to lead to rising prices of immobile factors (such as land).
It will be relatively more likely if linkages are broad, across many sectors,
rather than narrowly sector specific. This in turn is more likely in early
stages of development, where a country's basic industrial infrastructure-
58



ECON OM IC  BEN E FITS AND  COSTS
transport, telecommunications, or access to financial markets and other
business services-is thinly developed and unevenly spread.
This suggests that there is a real possibility that RIA membership
could lead to agglomeration and growing divergence between member
countries, as we saw in the examples earlier in this section. The agglom-
eration forces we have outlined here will interact with comparative ad-
vantage and may well reinforce each other. It seems likely that both com-
parative advantage and agglomeration are at work in some South-South
RIAs. As Nairobi, Abidjan, and Dakar have attracted manufacturing, so
they have started to develop business networks and the linkages that
tend to lock manufacturing in to the location. The process might be
further accelerated by the propensity of foreign direct investment to cluster
in relatively few locations. Agglomeration is then accentuating the forces
for divergence that we outlined in the preceding subsection.
In other circumstances agglomeration forces may pull against com-
parative advantage. For example, firms choosing location in Europe
may want the agglomeration benefits of locating in Germany, but fac-
tor price differences create an incentive for them to locate in Hungary.
An important final point is that agglomeration forces will be strongest
at "intermediate" levels of trade barriers (or transport costs). When
barriers are very high, each country will have its own industry to sup-
ply local consumers. When they are very low firms go where labor
costs are cheapest, because they can bring in their inputs and ship their
output at very low cost-as with the production networks described
in box 3.2. But at "intermediate" barriers firms are reluctant to move
away from suppliers and other agglomeration benefits, yet are able to
supply foreign markets through exports.
Knowledge Flows
Both the comparative advantage and agglomeration mechanisms sug-
gest that integration may cause the performance of members of develop-
ing country RIAs to diverge. One further factor needs to be added to the
convergence or divergence calculus. RIAs may promote knowledge flows
between member countries.
An influential-although not universally accepted-body of work
argues that trade flows provide a powerful mechanism for the transfer
of technology between countries. A good example of this are the works
59



T RAID E  B LO CS
Box 3.2 Production Networks
LOW TRADE BARRIERS AND NEW INFORMATION    rest ofEurope. By 1993 foreign owned plants produced
technologies make it possible to split the production    60 percent of gross output and accounted for nearly
process in many goods, relocating labor-intensive parts   45 percent of manufacturing employment (Barry and
of lhe process to lower-wage economies. This has led    Bradley 1997). They are concentrated in high-technol-
to growth of trade in components, which now ac-   ogy sectors, and import two-thirds of their inputs and
counts for a substantial part of some country's im-   export about 86 percent of their output.
poits (WDR 1999). Such tightly integrated interna-  Similar developments are occurring in some of the
tional production chains are sometimes referred to as   East European countries that have regional agreements
production networks.                             with the EU, in particular Hungary and Estonia. Pro-
Ireland provides a good example of production net-   duction networks are particularly important in the
works in Europe. Since joining the European Com-   automotive, telecommunications, and office machin-
munity in 1973 Ireland has attracted multinationals   ery sectors, and have accounted for a rising share of
which use Ireland as an export platform to supply the   the trade of these countries.
of Coe and Helpman (1995) and Coe, Helpman, and Hoffmaister
(1997), which seek to explain the rate of increase in total factor pro-
ductivity across OECD and developing countries. They construct an
index of total knowledge capital (measured by accumulated invest-
ment in research and development) in each industrial country, and as-
sume that trading partners get access to a country's stock of knowledge
in proportion to their imports from that country.
These authors find that access to foreign knowledge is a statistically
significant determinant of the rate of growth of total factor productivity.
For developing countries, Coe, Helpman, and Hoffmaister find that pro-
ductivity growth is related to the interaction between the openness of the
economy (imports relative to GDP) and access to foreign knowledge. Thus,
an economy benefits from foreign knowledge, first, according to how open
it is, and, second, according to whether it is open to those countries that
have the largest knowledge stocks. These results are intuitively very attrac-
tive and suggest, again, that trade is a major conduit for spillovers between
countries.'4 Of course, these results may be due to other factors highly
correlated with trade, such as foreign direct investment.
Although the research was not undertaken explicitly for RIAs, it has
clear implications. Increasing trade with high-income countries by form-
ing a North-South RIA may lead to beneficial transfers of technology,
60



ECO N OM IC  BEN EF ITS  AND  COSTS
and consequent convergence of incomes. Developing country RlAs do
not offer such good prospects-particularly if they are relatively closed
to external trade and cause trade diversion.
Convergence or Divergence: The Balance
We have seen how there are conflicting forces at work, some tending
to lead to convergence of member countries, others pushing divergence.
How do we think these forces balance out? In regional agreements that
include high-wage countries with industrial centers and lower-wage coun-
tries, our judgment is that the convergence is probably dominant. But
for South-South RIAs-particularly between the lowest income coun-
tries where manufacturing is small and business infrastructure thin-
the analytical arguments suggest that there is a real danger of divergence.
The analytical arguments are supported by the empirics, suggesting that
RIAs containing high-wage countries have promoted income conver-
gence, in a way that South-South RIAs have not.
3.4 Conclusion
A S POINTED OUT IN THE INTRODUCTION TO THIS CHAPTER,
regional integration  will affect many  aspects  of
economic life. Whether or not trade diversion dominates trade
creation depends on the specific circumstances. There is potential for
gains from "competition and scale" effects in industrial sectors of the
economy, but achieving these might require "deep integration" policies,
and these gains might also be achievable through unilateral trade
liberalization. The government will lose revenue, and some of this may
be dissipated through trade diversion. Industries are likely to relocate,
benefiting some countries and possibly harming others.
The ambivalence of these conclusions does not reflect ignorance but
rather the wide range of country circumstances and policy options that
exist. There are thus no fast and easy conclusions, but a number of rea-
sonably firm policy conclusions are provided at the end of chapter 4.
61



T RA D E  B LO CS
Notes
1.  Tybout (1999) fails to find evidence that price-   8. Any additional cost of losing government revenue
cost margins are systematically higher in developing than    applies only if the shadow price on government revenue is
in industriol countries, although reports a number of ex-    greater than unity.
amples where this is so.
9. This classification is from Baldwin and Venables
2.  Djankov and Hoekman (1998) report the posi-   (1997), who survey some of these studies in greater detail.
tive effects of trade reform on competition in Slovakia.
10. We use the average of France, Germany, Italy, and
3.  See Levinsohn (1993); Harrison (1994); Foroutan    the United Kingdom.
(1996); and Krishna and Mitra (1997).
11. This section is based on Venables (1999).
4. See Nishimuzi and Page (1982); Tybout and oth-
ers (1991); Haddad (1993); Haddad and Harrison            12. This section is based on Fujita, Krugman, and
(1993); Tybout and Westbrook (1995); Harrison    Venables (1999).
(1996).
13. This argument only works if there are increasing
5.  The original industry studies in this area were un-    returns to scale in production. (If not, firms can put small
dertaken by Smith and Venables (1988).                 plants in many different locations.) For formal analysis
see Fujita, Krugman, and Venables (1999).
6. Because some other sector must contract to release
resource s for the expansion.                             14. The conclusion has been challenged because the
paper assumes, rather than tests, that imports from indus-
7.  Partner country consumers and government for   trial countries provide the correct weights with which to
exports; own consumers and government for imports   combine stocks of foreign knowledge. Keller (1998) has
from partners.                                         suggested that the results are little better than would be
obtained from relating total factor productivity to a ran-
dom weighting of foreign knowledge stocks.
62



C HAPT ER 4
Policy Choices
Introduction
W         E HAVE NOW REVIEWED MUCH OF THE THEORY AND
evidence on the political and economic effects
)VVV,of regional integration schemes. How is this discussion
brought to bear on a particular country faced with decisions on RIA
membership? What choices will the country face, and how should the
theory and evidence inform these choices?
We group the choices under four headings. The first is who-if any-
one-a country's RIA partners should be. The political and economic
forces described in the previous chapters will have different effects de-
pending on the characteristics of the countries concerned. Here we bring
these together, showing how the effects might work for hypothetical
pairings of countries. Related to the choice of partner are issues to do
with the size of the RIA, and with multiple membership: should a coun-
try focus on membership of a single RIA or opt for many?
The second set of policy choices is to do with the external policy of
countries in a RIA, which in turn depend importantly on the type of
RIA that is adopted: a free trade area or a customs union. There are
considerable benefits from merging external trade policy through mem-
bership of a customs union, since it allows freer circulation of goods
within the RIA. However, it also involves greater loss of sovereignty and
more political commitment, as well as possibly leading to the formation
of new trade policy lobbies.
We then turn to the third and fourth policy choices, the "depth"
and "width" of integration. The question is whether to deepen an agree-
ment to cover domestic policies that affect the ability of consumers
63



TRADE B LO CS
and producers to engage in intra-RIA trade in goods; and whetlher to
widen the coverage to extend beyond merchandise trade. The tradeoff
running through all these cases is between the potential for greater
benefit from making the integration deep and wide, versus the consid-
erable practical problems and potential political difficulties that may
be involved. We look at a number of issues, including contingent pro-
tection, product standards, trade in services, and investment rules to
see the benefits and costs involved.
4.1 Partners: With Whom?
S     OME RIASAREJUST BETWEEN HIGH-INCOME COUNTRIES (SUCH
as EU or the European Economic Area), some only middle-income
(such as MERCOSUR or the ASEAN Free Trade Area), and some
only low-income (such as the West African Economic and Monetary
Union (UEMOA)). Others contain a mixture, often with a dominant
high-income partner (such as NAFTA or the EU Association agreements
with Central European and Mediterranean countries). RIAs come in all
sizes, both in terms of number of members, and in terms of income; the
EU has 15 members and GDP of US$8.4 trillion, while UEMOA's eight
members have a combined GDP of US$24 billion; many RIAs are just
bilateral agreements between a pair of countries.
In some RIAs most trade is between members, while in others very
little is. We see from figure 4.1 that the 1996 share of member countries'
exports that goes to other member countries is 62 percent for EU- 12 and
47 percent for NAFTA. In middle-income RIAs, the share is smaller, at 23
percent for MERCOSUR, dropping still further for low-income RIAs, at
9 percent for ECOWAS and UEMOA, and only 1.9 percent for the Eco-
nomic and Monetary Community of Central Africa.
While no single country will face a menu of choices containing all
these possibilities, we want to show how the effects discussed in previ-
ous chapters apply to these widely different options for regional inte-
gration. We do this by considering some hypothetical country pair-
ings, which are given in table 4.1. Pairs of country types are given in
the columns, and our main political and economic arguments are in
the rows. The country pairings are not to be taken literally, and the
signs of the effects, which we give in the body of the table, are not
drawn from research on the particular named economies. Instead, the
country pairs are representative of types of RIAs, and the body of the
64



PO LI CY CHO ICES
Figure 4.1 Intra-RIA Exports as a Share of the RIA's Total Exports
RIA
APEC
European Union
NAFTA
ASEAN
MERCOSUR
CACM
Andean Group
UEMOA
SADC
GCC
Union du Maghreb Arabe
Economic and Monetary
Community of Central Africa
ECOWAS
0           20          40           60          80
1990                I  1996
Source: U.N. COMTRADE data.
table summarizes our judgment about the sign and strength of the
forces we have discussed in preceding chapters. The four country group-
ings we have chosen are therefore intended to represent a middle-
income country and a high-income country-we label them "Poland"
and the "EU"; two middle-income countries-"Brazil" and "Argen-
tina"; a pair of small low-income countries-"Burkina Faso" and "C6te
d'Ivoire"; and a small low-income country and a large high-income
country or bloc-"Kenya" and the "EU."
65



T RA DE  B LO CS
Table 4.1 Pluses and Minuses of Hypothetical RiAs
European                       Burkina     Cote              European
Category         Poland       Union      Brazil   Argentina    Faso    d 'Ivoire   Kenya    Union
Political
Secur.ty           +            0          +         +          ?         ?        +?         0
Bargaining         0            0          +         +          0         0         0         0
Being noticed      0            0          +         +          +         +         0         0
Policy lock-in     +            0          +         +         O?         O?       +?         0
Coop,ration        +            +          +         +          +         +        +?         0
Economic
Scale and
competition      +            +          +?        +?        O?         +?       +?         0
Trade diversion    -?           0               -               -         -?       -?         0
Fiscal             -            0         -?         -?         -         +?        -         0
Trade and location    +         +          +?        +?         -         +         +?        0
Technology transfer   +          0         +?        +?         0         0         +         0
Source: Authors.
Poland-EU
We start with the case of a RIA between a middle-income develop-
ing or transition economy and a large high-income country or bloc-
such as with East European economies and the EU, or Mexico and
the United States. On the political side, the most important element
of likely gain comes from the policy lock-in argument. We saw in
chapter 2 that two conditions need to be met for this argument to
work; the country seeking lock-in must care about sanctions that the
partner might impose, and the partner must have the incentive to
impose sanctions, rather than let policy reversals go unremarked. Both
these conditions seem likely to be met in this case. Suppose that
"Poland" reneges on part of the agreement-perhaps imposing trade
barriers, or violating political conditions. The partner (the "EU") is
so large, and likely to account for such a high proportion of the
country's trade that its actions will certainly impact on "Poland." At
least when the countries are geographically close, the "EU" is likely to
want to see economic and political stability in the region, and be pre-
pared to act to enforce it. We have also scored "cooperation" positive;
there is great potential for schemes ranging from environnmental
projects to technical assistance programs.
66



P OLI CY  CH OI C ES
Turning to the economic arguments, there is considerable scope for
gains from the "scale and competition" effects outlined in chapter 3. The
middle-income country might typically have a broad range of industrial
firms, many of them inefficient, perhaps because of lack of competition,
or because they operate at small scale. Competition in a much larger
market provides the opportunity for solving these problems.
What about the changes in trade flows and location? It seems likely
that factor price differences should induce movement of relatively labor-
intensive industrial activities to the middle-income country, in line
with the experiences of Mexico, Hungary, and other parts of the Euro-
pean periphery we saw in chapter 3. Much of this might be driven by
FDI, bringing with it new technology. A counter argument to this is
that the established agglomeration of activity in the EU might draw
activity out of the middle-income country, particularly since the EU is
a "hub," benefiting from numerous bilateral "spoke" agreements.
Whether or not this happens will depend on how closely the two re-
gions are integrated. If barriers-including transport costs-are suffi-
ciently low then the middle-income country is likely to be drawn into
Europe-wide production networks. But if obstacles to trade or to FDI
remain, the outcome may be less positive; the middle-income country
might find its industry threatened by imports, yet not be a sufficiently
attractive place for FDI inflows.
Finally, if the middle-income country retains high external tariffs,
there is scope for it to suffer from trade diversion, although this is likely
to be small insofar as affected sectors have low protection in the high-
income country (with obvious exceptions, such as agriculture in the EU).
Tariff revenue loss may be significant, since large volumes of trade are
likely to be covered by an agreement of this sort.
Brazil-Argentina
Now consider a RIA between two middle-income countries, both
perhaps of considerable size. We have marked in possible benefits under
all the political headings.
On the economic side, a pair of countries at this stage of develop-
ment offers perhaps the greatest potential for scale and competition ef-
fects. This might increase efficiency levels in domestic firms, attract FDI,
and also lead to terms-of-trade improvements, as foreign suppliers react
67



TRA DE  B LO CS
to the more competitive market. FDI might bring with it benefits of
technology transfer. However, the usual question mark remains over these
effects. Securing effective competition has often been obstructed, and
achieving it requires investing in "deep integration"; policies to achieve
this are discussed later in this chapter.
What about the relocation of industries between countries? If there
are differences in comparative advantage or in market size and access,
then integration will create forces for relocation. However, to the extent
that the economies are similar, and already have established manufac-
turing sectors and the infrastructure that goes with it, it seems unlikely
that this would be a one-way traffic. Some industries will expand in one
country, others in the other, so there may be clustering of particular
sectors rather than of activity as a whole, bringing gains from specializa-
tion rather than from the costs of divergence.
These are the benefits, but there are also likely costs. There is a danger
of substantial trade diversion in such a RIA. The countries involved may
have developed their industry behind protective barriers, meaning that
production costs are well above world minima. If tariff preferences in-
duce importers to switch the source of supply from the rest of the world
to inefficient production in the partner country, this will reduce income.
There will also be loss of government tariff revenue, its significance de-
pending on initial tariff rates, on initial trade volumes, and on the ease
with which governments are able to activate alternative fiscal instrunments.
Burkina Faso-Cote d'lIvoire
Regional integration between two small low-income countries may
offer some real opportunities for benefits from increased cooperation on
economic projects-such as water management, or development of in-
frastructure, particularly between coastal and landlocked countries. As
far as bargaining power goes, the main potential benefit is that of "being
noticed," providing that the member countries are willing and able to
take a concerted position on world issues. "Lock-in" effects are unlikely
to be strong; they require both that the partner is itself committed to
reform, and that it has political capital to invest in securing reform in
the partner country.
Turning to economics, some sectors may benefit from scale and com-
petition effects. Rationalization and removal of inefficient duplication
68



P OLI CY  CH O I C ES
of plants is a possible outcome, bringing with it efficiency gains. Market
enlargement may also bring an FDI inflow. As usual, these potential
gains can easily be frustrated, and it is also possible that even the com-
bined market is too small for scale and competition effects to operate.
Against these gains are the costs. If external tariffs remain high then
trade diversion is likely. Related to this, inward-FDI may be "tariffjump-
ing," in which case it is not necessarily beneficial. Both these effects will
be associated with loss of tariff revenue, likely to be a major source of
government revenue.
It is also in relatively closed South-South RIAs that we think that the
scope for uneven internal development is greatest, with production con-
centrating in a few locations. If one region has a head start in manufac-
turing-due perhaps to its location, endowment of factors of produc-
tion, or simply due to history-then this region may well expand at the
expense of other regions ("Cote d'Ivoire" in table 4.1). Linkages are likely
to be strong, because of the paucity of the business infrastructure, and
manufacturing as a whole is sufficiently small that it will not run up
against the "centrifugal" forces outlined in chapter 3. In this case then,
the effects of the RIA will be very different across the members; benefi-
cial for some, but possibly adverse for others.
Kenya-EU
Kenya-EU represents a RIA between a small low-income country and
a large industrial country or bloc. In table 4.1 we have set effects on the
EU at zero-simply reflecting the relative magnitudes of the countries.
What of the effects on the low-income partner? On the political side,
we have scored the bargaining and being noticed rows at zero. In the do-
mains of security, cooperation and policy lock-in there are potential gains.
Once again, for a regional agreement to work as a commitment mecha-
nism we look for two conditions to be satisfied. The first is that the coun-
try seeking lock-in must care about sanctions that the partner might im-
pose. This condition seems likely to be met:If the partner is large, consti-
tuting a significant market for exports, and perhaps also a considerable
source of aid and technical assistance the threat of sanctions will be power-
ful. The second condition is that the partner must have the incentive to
impose sanctions, rather than let policy reversals go without response. This
is more questionable. The "EU" is unlikely to be direcdy affected by policy
69



T RA DE B LO CS
reversal or even instability in a small remote developing country, but there
may nevertheless be reasons for it to enforce an agreement. One might be
a history of involvement with the region. Another is the reputation of the
"EU" itself; if it has entered many such RIAs, then the effectiveness of all
of them can be damaged by the failure of one.
Evidently, large question marks surround the willingness of our hy-
pothetical industrial country to constitute an effective commitment
mechanism. This is an argument for designing the agreement in a way
that makes explicit the commitments to reform, and perhaps also the
sanctions that are to be followed in the event of policy reversal.
Turning to the economics, there are three dominant issues. The first
is trade diversion: Is a RIA with the "EU" likely to cause the source of
imports to switch from other lower-cost suppliers to relatively high-cost
"EU" production? This effect is likely to be small, insofar as the "EU"
itself has low protection, so production costs are close to world minima.
Transport costs could however be important if our hypothetical low-
income country were much farther away from the "EU" than from other
sources of supply of manufactures. The second issue is government rev-
enue. There may be significant loss of tariff revenue, and the low-
income country should look to the agreement to make this up.
Third, and perhaps most importantly, are the trade and location is-
sues. Will such an agreement enable a low-income country to develop
effective export activities, supplying the partner country? Is there a real-
istic likelihood that the low-income country can be drawn into a "pro-
duction network," undertaking labor-intensive stages of production ac-
tivity? Low-wage costs suggest so, but working against this are transport
costs, quality of infrastructure, and security of market access. The po-
tential advantage of a regional agreement is that it can make progress on
reducing these obstacles. This suggests a need for such an agreement to
be relatively deep, so it can overcome trade frictions and secure guaran-
teed market access, for example, by removing contingent protection.
Size and Multiple Membership: How Many?
Many countries are members of more than one RLA. Probably the most
extreme example of this is the EU, in which each member is also a inem-
ber of EU agreements with EFTA countries (the European Economic Area
agreements), most Mediterranean countries (the Euro-Med agreements),
70



PO LI CY  CHO IC ES
and with East European countries (the Europe Agreements). A structure
of this type has been termed "hub and spoke regionalism"; it places Eu-
rope at the hub of agreements with other countries, most of which are not
linked to each other through RIAs.
Complex patterns of multiple membership appear elsewhere. In ad-
dition to the large Latin American RIAs, many Latin American coun-
tries also have bilateral agreements with other countries or groupings.
For example, Chile is party to 11 trade agreements (APEC and the Latin
American Integration Association, plus bilateral links with Argentina,
Bolivia, Canada, Colombia, Ecuador, MERCOSUR, Mexico, Peru, and
Venezuela). Panama is a member of nine trade agreements; Mexico of
eight; Bolivia, Costa Rica, and Nicaragua of five; and El Salvador, Gua-
temala, and Honduras four each. The picture is equally complex for
Eastern Europe, with the Slovak Republic belonging to nine RIAs; the
Czech Republic and Slovenia belonging to eight; Estonia to six; and
Hungary, Latvia, Poland, and Romania to five. The pattern of RIA mem-
berships in East and southern Africa is given in figure 4.2, which reveals
a pattern of overlapping memberships.
Figure 4.2 Regional Organizations in Southern and Eastern Africa
Cross-Border Initiative
Common Market for Eastern                                            Burundi
and Southern Africa                                                  Rwanda
Djibouti ---------------------.
Egypt      SADC                              -       ----       ----       
Eritrea   I                    Malawi          East African
Ethiopia  I Angola             Zambia          Cooperation
Sudan      Congo               Zimbabwe                              Kenya
Tanzania            I Uganda
-----------------------------___________  -     -     -     -     -     -     -     -     -     -     -     -   -     -     -     -     -  _ -
I  : Southern African                Indian Ocean
I  , Customs Union                    Commission
I  l                  Namibia
Lesotho          Swaziland      Mauritius            I Comoros
Seychelles          I Madagascar
South Africa                   Mozambique
I  | Botswana
Source. Authors.
71



TRA DE B LO CS
In some cases membership in multiple RIAs creates obligations made
in one that contradict those made in others. Customs union mernbers
must all set the same external tariffs, yet Bolivia, Colombia, Ecudor,
the Republica Boliviarana de Venezuela, and Peru form the Andean
pact (a customs union), while Colombia and the Republica Boliviarana
de Venezuela are also in the Group of 3, a free trade area with Mexico.
Similarly (see figure 4.2), Lesotho, Namibia, and Swaziland belong to
the Common Market for Eastern and Southern Africa while also be-
longing to Southern African Customs Union (SACU), a customs union
with Botswana and South Africa. Tanzania belongs to SADC while
being a member of the East African Cooperation, a customs union
with Kenya and Uganda. These obligations are contradictory, so it may
be unclear which will prevail in practice, and special conditions and
exclusions may have to be formulated.
What are the benefits and costs of being in multiple RIAs? First, there
are circumstances where gains can be derived, by some countries at least,
from being in more than one RIA. The best example of this is being the
hub of "hub and spoke" regionalism. If one country (or group of coun-
tries) has RIAs with a number of countries that maintain barriers be-
tween each other, this hub country becomes the preferred location for
investment-firms can reach more markets tariff-free than they can from
any of the other locations-and this will tend to bid up factor prices and
raise real income in the hub.
Second, membership in multiple RIAs may lead to complexity, which
can hinder private sector decisionmaking. The extent to which this hap-
pens depends on how complex the individual RIAs are. Membership in
a number of relatively loose free trade areas may be straightforward, but
even here there is a possibility for conflicting rules-for example, rules
of origin requirements-that are nontransparent and complicate the busi-
ness environment. At worst, there are inconsistencies (as when some
customs union members are also in a free trade area), which can only
create uncertainty about how the inconsistency is to be resolved, and is
thus likely to dampen investment.
Third, securing the full gains from a RIA may require considerable
government commitment-for example, tackling the difficult issues of
deep and wide integration-and membership of many RIAs may be a
diversion from this.
These points are relevant considerations not only for a government
considering membership of multiple RIAs, but also for incumbent
72



POLICY  CHO IC ES
members deciding on accession of additional countries. There is, in
most cases, a tradeoff between the depth of integration that can be
achieved and the size of the RIA. (This tradeoff has been evident in
the different positions taken by EU member states in discussion of
enlargement of the EU). The tradeoff can be shifted by requiring that
new members accept the entire package of policies implemented by
existing members (the acquis communitaire in EU parlance). While
this may make it easier for incumbents to accept new members, it may
reduce the desirability of membership for these new countries. In gen-
eral then, we expect early members to have more say in setting the
rules-and therefore receive larger benefits-than later members.
4.2 External Trade Policy: How Much Preference?
A       FUNDAMENTAL ISSUE CONFRONTING ANY GOVERNMENT
considering a regional integration agreement is the
external trade policy that will prevail. There are two dimensions
to this issue: first, the level of external protection that will apply after the
agreement is made, and second, whether to set this external trade policy
in a concerted manner.
Openness to the Outside World
Some of the costs and benefits of RIA membership depend directly on
the external trade policy stance. This creates strong arguments for pursu-
ing a policy of external openness in conjunction with regional integration.
First, trade diversion is both more likely-and more costly should it
occur-the higher the external trade barriers. It is more likely, since the
relative price differences created by preferential liberalization will be greater
with a higher external tariff, inducing trade diversion in more sectors. It is
also more cosdy, since a higher external tariff will provide greater incen-
tives for inefficient sectors to expand. Producers are able to charge high
prices (because the tariff protects them from world competition) and cap-
ture what was previously tariff revenue on internal trade. Fundamentally,
the gains from competition with low-cost suppliers-gains to consumers,
gains in developing an efficient industrial structure, and competition-
induced efficiency gains at the firm level-may be forgone if competition
73



TRA DE B LO CS
from the lowest cost suppliers is inhibited. These are arguments both for
low tariffs on average, and for tariff schedules that are relatively uniform,
avoiding peaks. Very high rates in particular sectors are almost certain to
produce diversion-such as in EU agriculture.
A further argument for low external tariffs relates to the likelihood of
agglomeration occurring, and the RIA consequently causing divergence
of economic structure. We argued in chapter 3 that one of the forces
driving agglomeration is linkages between firms-the dependence of
firms on other local firms for supplies, and on local firms and markets
for their sales. The more closed is the RIA, the more inward-oriented
will be its firms, increasing the strength of local linkages and making it
more likely that the RIA will develop a monocentric economic struc-
ture. Research by Ades and Glaeser (1995) examined 85 countries, and
showed that the population of the largest city is greater the higher are
tariff barriers and the lower the share of imports in GNP. Krugman and
Hanson (1993) have documented how Mexico's opening to trade (largely
with the United States) brought a deconcentration of manufacturing
from the congested Mexico City region.'
A counter argument to external openness might be that the RIA lib-
eralization brings with it adjustment costs, and simultaneous external
liberalization magnifies these to an unacceptable level. The problem with
this argument is that adjustment costs are only worth paying, for an
adjustment in the right direction. High-tariff peaks might induce costly
economic changes that are moves away from economic efficiency. Simu-
lation studies of regional integration involving developing countries and
large, high-income nations or blocs (such as the EU) suggest that the
adjustment costs associated with RIA implementation are as high as those
that would arise if trade liberalization were implemented on a nondis-
criminatory basis (Rutherford, Rutstr6m, and Tarr 1999).
Customs Union or Free Trade Area?
In a free trade area (FTA), countries are free to set their own external
trade policy, whereas in a customs union (CU) the RIA as a whole sets a
common external policy. Of the 162 RlAs notified to the GATT/WTO
by August 1998, 143 were FTAs and 19 were CUs.2 A CU typically re-
quires greater political commitment, because countries have to agree to a
common external policy and set up budgetary mechanisms to distribute
74



P OLI CY CH O I CES
the tariff revenue between member countries. A central issue for countries
planning to integrate their trade is whether to choose an FTA or CU.
The great advantage of a CU is that, because members have a com-
mon external tariff, it is possible to have much simpler internal border
formalities-and possibly none at all. In contrast, an FTA leaves exter-
nal trade policy to individual member governments, and faces a prob-
lem known as trade deflection; the redirection of imports from outside
countries through the FTA member with the lowest external tariff, to
exploit the tariff differential. If unconstrained, this reduces the effective
tariff of every member to that of the lowest plus the transportation cost
involved in indirect importing (which is a wasted real resource cost).
The usual solution is rules oforigin-the apparently reasonable require-
ment that goods qualifying for tariff-free trade should be produced in a
member country, rather than just passing through it.
In practice, the costs of implementing rules of origin are high. They
mean that controls on goods crossing internal frontiers have to be retained
to ensure compliance and to collect customs duties that are due. Some
years ago these costs were estimated at 3 to 5 percent of fo.b. prices for
EFTA-European Community trade (Herin 1986). They also allow cus-
toms authorities-and individual customs officers-a good deal of discre-
tion, and the attendant danger that such discretion might be abused.
Rules of origin are particularly complex because they have to take
into account tariffs on imported intermediate goods used in products
manufactured within the FTA. The principle behind a rule of origin is
that imports from outside the FTA should pay the tariff of the country
of final sale, but additional value added in FTA members should be
tariff-free. For example, if one million dollars worth of shirts are manu-
factured in FTA member A and exported to member B, and these shirts
use $1 00k of cotton imported from outside the FTA on which the country
B tariff is 20 percent, then $20k of duty is payable to B; furthermore,
the exporting firm should be rebated any tariff it paid to country A on
the cotton. In practice calculations are not made on such an exact basis
but instead according to more or less arbitrary rules, typically stating
that exports have to derive a certain proportion of their value from local
content or undergo certain production processes within the FTA to ob-
tain duty-free treatment.
The rules are complex and hard to negotiate (the EU's agreement
with Poland has 81 pages of small print in its rules of origin section, and
NAFTA some 200 [Krueger 1997]). Since they do not match the exact
75



TRA DE B LO CS
inputs in each commodity, they introduce further biases and sources of
distortion. For example, NAFTA rules of origin in some sectors have
serious protective effects that shift trade and investment patterns from
lower- to higher-cost sources. Most clothing produced in Mexico gains
tariff-free access to the North American U.S. and Canadian markets
only if its inputs are virtually 100 percent sourced in North America
(WTO 1995). In the automobile industry, the origin requirement of
62.5 percent local content has induced Japanese automobile manufac-
turers with plants in Canada to produce components in the United States
rather than import cheaper ones from Japan. NAFTA rules of origin
require color television tubes to be of North American origin, causing
five television tube factories to be planned or established in North America
by Japanese or the Republic of Korean firms, probably at the expense of
expansion in Southeast Asia.
It is also worth noting that even with complex rules of origin in place,
the problem of imports to the FTA entering through the country with
the lowest external tariff is not entirely solved. A low-tariff partner can
meet its own requirements for a product from the rest of the world, and
export a corresponding amount (or all) of its own production to its part-
ners. This is indirect trade deflection (Robson 1998).
Thus, there are substantial benefits to going to a CU, yet, as we saw
above, only a small minority of RIAs notified to the GATT/WTO are in
fact CUs. What are the costs?
First, harmonization of external trade policy means a loss of national
autonomy. Second, we saw in chapter 2 the potential for politically divi-
sive redistributions due to a common external tariff-for example, in
the antebellum United States. Political institutions need to be put in
place to ensure that these tariffs are set in a consensual way. Fuirther-
more, tariff revenues generated by the common external tariff have to be
distributed between member countries, and this too can be divisive. In
the EU these revenues are part of the central budget and are spent on
agreed programs, yet the level of each member's net contribution or re-
ceipt from the budget remains contentious. In many developing coun-
try customs unions, difficulties of agreeing on a common external tariff
and distribution of revenues have proven to be great. Thus, the GCC
has to date not been able to achieve consensus on its common external
tariff. Similarly, the implementation of a common external tariff by
CARICOM, originally scheduled for 1981, was delayed until the 1990s;
the original scheme, adopted in 1991, was subsequently revised, and
76



PO LI CY CHO ICES
CARICOM members are in the process of implementing a new tariff
structure (IDB 1998). In the case of the CACM, the common external
tariff was rendered largely ineffective because of exemptions granted by
some members for "necessary" imports (De la Torre and Kelley 1992).
The third problem with a common external trade policy is the addi-
tional adjustment costs-and lobby opposition-that may be encoun-
tered in moving to the common schedule. A good example is the diffi-
culty that the EU had in harmonizing nontariff barriers. Despite being a
customs union, for its first 30 years the EU allowed members to main-
tain their own quotas on certain third country imports (for instance,
clothing, footwear, and steel) and prevented those goods from crossing
internal borders (Winters 1992, 1993).
These three costs of forming a CU are minimized of member coun-
tries are more similar to each other, as in South-South RIAs rather than
North-South RIAs. Schiff (2000) shows that the ratio of FTAs to CUs is
6 to 20 times higher for North-South RIAs then for South-South ones.
CUs, FTAs, and the External Tariff
There is a potentially important interaction between our first point-
the desirability of open external trade-and the choice between an FTA
and a CU, because FTAs and CUs create different incentives for setting
external tariffs.
Several arguments suggest that an FTA may create downward pres-
sure on external tariffs. First, trade diversion may become apparent; if a
country sees itself importing a good from a partner country at a higher
cost than similar goods from nonmembers, this may induce it to cut
tariffs on these external imports. Second, if there is trade deflection,
high-tariff countries lose tariff revenue, as imports come in through low
tariff countries. This creates an incentive to cut tariffs to just below the
level of their partners, and thereby capture the tariff revenue. Third, if
duties on inputs used to make exports to other members cannot be re-
bated, high import tariffs render final goods uncompetitive to exporters.
This concern was apparently behind Canada's decision to reduce 1,500
tariffs on inputs in 1995, shortly after NAFTA started.
The situation in CUs is quite different. Creating a customs union
provides an (unavoidable!) opportunity to review the tariff structures
and create new institutions for determining trade policy. National tariffs
77



TR ALD E  B LO CS
must be harmonized at some agreed level, and in the process, interna-
tional obligations-notably those to the WTO-must be respected. Un-
fortunately, most of the arguments about the incentives for tariff setting
in a CU do not lean toward external openness.
The first argument is that, by coordinating their trade policies, CU
members may have a market power that individual countries lack. Import
tariffs can improve a country's terms of trade, since by cutting the volume
imported they reduce the world price of the product. This effect is going
to be larger, the larger the country or bloc imposing the tariff. Further-
more, countries in a RIA may be able to increase their negotiating power
against the rest of the world. If they are able to negotiate effectively as a
bloc (which does not always happen) this additional power will change
the outcome of trade negotiations. A related argument is that, in trade
negotiations, a CU may be able to gain disproportionate power over cer-
tain issues by letting a particular member "lead" negotiations. If a more
aggressive member leads negotiations with the rest of the world on an
issue, the union may be able to extract a more favorable deal because threats
to retaliate (with the whole of the union's resources) will be more credible.'
Whether this leads to lower protection overall depends on whether a more
aggressive union can achieve a more liberal outcome by virtue of its readi-
ness to retaliate, or whether it actually needs to use its retaliatory rnuscle.
The internal process of decisionmaking within the CU may also place
an upward pressure on tariffs. Consider a situation in which each mem-
ber country has an interest in raising protection in one sector and reduc-
ing it in all others, and each country has a veto over reductions in pro-
tection. Unless there is effective intra-CU negotiation, this creates the
possibility of a classic "prisoners' dilemma" outcome with high protec-
tion in all sectors, even though each country would be better off with
low protection in all sectors. Although it might be hoped that negotia-
tion within a CU could avoid prisoners' dilemmas, there are enough
examples from the EU to not engender optimism. The EU allows coun-
tries disproportionate influence over policy-up to (and including) veto
power-in areas in which they claim "vital interests." Winters (1994)
shows that the prisoners' dilemma outcome is quite likely to arise in
small groups of decisionmakers, and experience suggests the same out-
come. Indeed, the unanimity with which trade policy decisions pass (in
1995 the EU Council of Ministers passed 92 of its 94 common trade
policy decisions unanimously) suggests that countries were willing to
acquiesce to perceived "vital interests" of particular members (Bilal 1998).
78



P OLI CY CH OI CES
The CU will also change the power of lobbies. It is possible that lobby-
ing pressure within a CU may be diluted, compared with national lobby-
ing for protection within an FTA. There may be more opposition to over-
come (Panagariya and Findlay 1994; de Melo, Panagariya, and Rodrik
1993) or more representatives to influence (Richardson 1994). Other ar-
guments cut in the other direction. For example, each country might find
that initially its industry and agriculture more or less cancelled each other
out, but if integration lets agriculture lobbies cooperate (because they pro-
duce the same things) while the industry lobbies compete (because they
produce different things) the CU may end up with agricultural protec-
tion. It is worth noting that there is a massive lobbying industry in Brus-
sels, where the number of lobbying organizations grew from 300 in 1970
to some 3,000 in 1990. Expenditure on lobbying was estimated at $150
million in 1990, and increased rapidly. By 1998, there were 13,000 pro-
fessional lobbyists in Brussels, approaching one for every European Com-
mission staff member (The Economist, August 14, 1998).
4.3 How Deep?
T * 8 HERE IS A LARGE MENU OF CHOICES CONCERNING THE DEPTH
of integration that can be sought in a RIA. We argued
in chapter 3 that the gains from competition and scale effects
might not be achieved unless other policies causing segmentation of
markets were removed, and firms in different countries were induced to
compete head-on. Simple removal of tariffs while other obstacles remain
in place may not be sufficient to achieve this. However, removing these
obstacles is not without cost. For example, a CU is "deeper" than an
FTA (it avoids the border costs associated with enforcement of rules of
origin), but it brings with it other complexities related to choosing the
common external tariff and distributing the revenue. In this section we
look in detail at other policy measures that can be adopted to try and
secure greater market integration.
Contingent Protection
Many RIAs retain contingent protection-restrictions on intrabloc
trade that are applied in a more or less well-defined set of circumstances.
79



T RA D E B LO CS
These include antidumping, countervailing duties (in response to for-
eign subsidies), and "emergency protection" to address balance of pay-
ments problems or to protect an industry from surges in imports. While
contingent protection has been abolished within the EU, it remains ap-
plicable in all the EU's agreements with other countries except the Eu-
ropean Economic Area. Within some RIAs contingent protection is
widely used; in MERCOSUR, Argentina initiated 33 antidumping cases
on imports from Brazil between 1992-96 (Tavares and Tineo 1998).
Where these contingent protection measures have been left in place,
countries recognize that they provide obstacles to trade, but have weighed
the political and economic costs of their removal greater than the ben-
efits. What are the benefits and costs of removing these measures, as has
been done in the EU, the Canada-Chile FTA, and the Australia-New
Zealand agreement?
Contingent protection provides a major barrier to trade, not only
when actually applied to trade flows, but also by its mere existence, which
has a "chilling" effect on trade. Just the surveillance of trade flows has
been shown to have a negative impact on the volume of trade (Winters
1994). The threat of initiation of antidumping actions can lead to an
immediate loss of markets for exporters, as importers seek to avoid the
costs of posting the bonds required by customs authorities while the
investigation is ongoing. Empirical estimates of the "chilling" effect on
trade find that threatened exporters often agree to raise prices and main-
tain historical market shares, so contingent protection becomes an in-
strument facilitating collusion between domestic and foreign firms
(Messerlin 1990; Staiger and Wolak 1989).
These are powerful arguments for abolition of contingent protection,
but they encounter several counter-arguments. The first is that dump-
ing can be predatory, when a foreign firm (or cartel) seeks to force do-
mestic competitors out of the market by pricing below cost-with the
intention of raising prices once the competition is gone. However, re-
search suggests that predation is very much the exception, not the rule
in antidumping cases, and that in over 90 percent of actual antidump-
ing actions an antitrust authority would not have intervened on compe-
tition, let alone predation, grounds (Messerlin 1997; Schone 1996). The
consensus among most economists is that antidumping as it is practiced
today has nothing to do with predation. Furthermore, if there is preda-
tion, national antitrust should be able to address the problem; and this
can be done unilaterally-there is no need for international agreement
or harmonizing competition regimes.
80



POLI CY  CH OI CES
The second counter-argument is that if countervailing duties are to
be suspended, then measures also need to be taken to restrain or coor-
dinate industrial subsidies that may have a negative impact on domes-
tic firms. When such subsidies are being used there are valid theoreti-
cal economic justifications for the use of countervailing duties (Dixit
1988). However, in general, contingent protection is an inefficient in-
strument to deal with the effects of foreign subsidies or industrial poli-
cies, because it imposes additional costs on domestic consumers with-
out greatly increasing the incentives of the foreign government to change
its policies. Small countries in particular are unlikely to have much
success by pursuing retaliatory policies-all they will end up doing is
adding an additional distortion to consumption. The appropriate policy
is to draft rules that restrict the ability of RIA members to use indus-
trial policies in ways that are detrimental to the welfare of other mem-
ber countries. In practice, this may be difficult to achieve; only a lim-
ited number of RIAs have done much to discipline the ability of mem-
bers to provide subsidies.
Finally, there may be strident opposition from lobby groups to the
abolition of these measures. Antidumping cases are frequently initiated
by producer lobbies, and these may be virulent opponents of their with-
drawal. In practice the political power of these groups goes a long way
toward explaining why contingent protection frequently remains in place
in RIAs, despite the overall commitment to liberalize intrabloc trade
flows. Continued access to such instruments may be required in order to
"sell" the more general liberalization reform package.
Borders, Product Standards, and Red Tape
Trading across international borders encounters many real costs: de-
lays, form-filling, recertification of products, and so on. Even if there are
no duties, border formalities themselves create barriers and can be quite
wasteful. For example, customs procedures can be duplicative or redun-
dant, as when tax authorities in an exporting country require data simi-
lar to that demanded by the importer's customs officials-but in a dif-
ferent format. It has been estimated that border formalities on intra-EU
trade in the early 1 990s were equivalent to over 1.2 percent of the gross
value of internally traded goods. The EU had already implemented pro-
cedures to cut these costs, and in many other RIAs the costs of border
formalities are many times larger.
81



T RA DE B LO CS
Additional barriers to trade are created by variations in national prod-
uct standards. Estimates from Egypt in the early 1990s, showed just how
significant standards-related "red tape" in customs were; redundant test-
ing and idiosyncratic standards alone imposed taxes equivalent to between
5 and 90 percent of the value of shipments (Hoekman and Konan 1]999).
Barriers of this type prevent effective international competition. They
provide opportunities for corruption. They are usually real resource
costs-unlike tariffs, where revenue gets transferred to the government,
the barriers use up administrative and technical time and effort. What
can be done, within a RIA, to simplify them?
Border Formalities
Border formalities can be reduced by good customs administration and
the use of standard practice on procedures. Much of this is possible inde-
pendently of RIA membership, by adopting best practices in the area of
customs administration and implementing the relevant international con-
ventions aimed at trade facilitation. The World Customs Organization is
the primary organization promoting the standardization and simplifica-
tion of customs procedures around the world, in particular the Kyoto
Convention of 1973, currently undergoing a major revision. Adopting
the revised procedures has been argued to be the single most comprehen-
sive prospect for true international trade facilitation (Staples 1998). Some
of the needed measures are easier to implement within a RIA. One reason
is simply the inherent simplicity of not collecting duties on internal trade.
Another is that it may be easier to develop common approaches and insti-
tutions, such as the adoption of standard forms.
Product Standards
The issues raised by product standards are more complex, since coun-
tries can genuinely differ on what they regard as acceptable levels of
standards. Countries seeking to reduce the barriers created by differing
standards can follow two alternative routes-harmonization and mu-
tual recognition.
The baseline principle governing treatment of imports in most trade
agreements is "national treatment," which requires that governments
82



POLICY  CHO ICES
treat foreign products or producers that enter their territory in the
same way as domestic counterparts, in terms of internal taxes, health
and safety standards, competition rules, and so on. The principle has
always been a basic building block of international trade treaties, and
ensures that liberalization commitments cannot be circumvented by
discriminatory application of domestic policies-such as an excise tax
that is higher for foreign than for domestic products. While national
treatment is a powerful source of discipline on RIA members, prob-
lems arise if national product standards differ, requiring the sort of
expensive retesting we saw above.5
These issues are nothing new. In the past they have often been re-
solved by harmonization around international standards. Between 1860
and 1914 more than 30 intergovernmental organizations emerged to
harmonize product standards, particularly on infrastructure: for example,
mail (1863), marine signaling (1864), technical railwaystandards (1883),
ocean telegraphy (1897), and aerial navigation (1910) (Murphy 1994).
International interconnection norms, agreed under auspices of the In-
ternational Telecommunications Union, eliminated the need for tele-
grams to be printed at each border post, walked across, and retyped. The
Radiotelegraph Union aimed to prevent a global radio monopoly by
requiring interconnection across different technologies. Intergovernmen-
tal organizations proliferated after World War II, including such
standards-setting bodies as the International Maritime Organization, the
World Customs Organization, and the Bank for International Settle-
ments. However, these bodies do not cover the great bulk of products
that make up most of world trade.6
An alternative is to harmonize by unilaterally adopting the standards
of another country or group of countries. In 1992 Canada adopted U.S.
auto emission standards to ensure that its automakers could realize econo-
mies of scale by avoiding separate production lines for their home and
U.S. markets. Switzerland, similarly, adopted the EU regime on techni-
cal regulations and industrial standards so Swiss goods can enter and
circulate in the EU on the same basis as EU-produced goods (Messerlin
1998). Many developing countries use legal regimes developed in Eu-
rope or the United States, usually by maintaining systems inherited from
a colonial past or military occupation. Others have deliberately adopted
foreign norms. South Korea imported many West German and U.S.
product standards in the 1950s as part of a strategy to upgrade the qual-
ity of industrial production and foster exports.
83



TRA DE  B LO CS
Despite these possibilities, most countries still maintain their own
distinct product standards. Attempts to harmonize within a RIA require
developing a set of RIA-specific common standards, set either by inter-
governmental cooperation or by the cession of sovereignty to common
or supranational institutions. In the EU, the European Commission has
been delegated the power of proposing directives and regulations and
the Court of Justice given the task of enforcement. EU experience sug-
gests that this process is extremely slow and painful. Early efforts toward
harmonization centered on food standards-the first "harmonization
directive" issued in 1962 dealt with food coloring-and progress was
very slow, in part because adoption of a Community-wide norm required
unanimity. It took over a decade to reach agreement on the composition
of fruit jams and mineral water, and only nine directives on foodstuffs
were adopted between 1962 and 1979. Differences in national norms,
reflecting national tastes, history, legal regimes-and producer lobbies
seeking to restrict competition from imports-made it difficult to achieve
the required consensus. For example, Germans set great store by their
Reinheitsgebot, a standard established in 1516 specifying that beer may
have only four ingredients (malted barley, hops, yeast, water). Other
countries include preservatives or additives.
If harmonization is not pursued, the alternative route to settling stan-
dards issues is "mutual recognition," under which member countries
simply accept that goods that are legally introduced into circulation in
one member state cannot be barred from entering and being sold in
another. This principle was incorporated into the Single Market pro-
gram and has proved to be a powerful tool for increasing cross-border
competition in European markets.7
However, mutual recognition raises two quite difficult issues. First,
countries have to be able to agree on the minimum norms that should
be met to safeguard public health and safety or maintain the integrity of
public networks. Second, there has to be mutual trust in the compe-
tence and ability of the national institutions responsible for enforcing
the relevant mandatory standards. Thus, even if all members accept the
levels at which standards are set, it may require significant institutional
strengthening for one country to accept that another's testing and certi-
fication procedures are adequate. One way of addressing this constraint
is to rely on third party conformity assessment of goods and services,
and seek agreement that certification by such specialized entities will be
84



PO LI CY CHO ICES
accepted by all members of a RIA. To date, however, this has been an
option that has not been pursued vigorously in RIAs.
Discrimination in Public Procurement
If RIA membership is to secure effective competition, then competi-
tion should extend to government procurement-an area that often ac-
counts for as much as 10 percent of GDP. Yet, in practice, governments
frequently permit or require public entities to discriminate in favor of
domestic firms when procuring goods and services. This can take the
form of price preferences, local content rules, or residency requirements.
Progress in eliminating discrimination in public procurement can, in
principle, be made independently of RIA membership, by consistent ap-
plication of the national treatment principle and procedures outlined in
the WTO Agreement on Government Procurement. In practice, progress
at the multilateral level has been slow-virtually no developing countries
have signed the voluntary WTO agreement (Hoekman and Mavroidis
1997). RIAs offer a potential avenue to make more rapid progress. How-
ever, only a limited number of RlAs have included liberalization of public
procurement as an objective or made progress in forcing government enti-
ties to abide by the national treatment principle (EU, NAFTA). Progress
in this area has proven to be difficult to achieve. A recent evaluation of the
pattern of purchasing by European government entities found that public
sector import penetration increased from an estimated 6 percent in 1987
to some 10 percent in 1994 (Gordon, Rimmer, and Arrowsmith 1998).
Thus, on average, 90 percent of all purchases in the EU continue to be
sourced from national firms, despite vigorous efforts by the EC Commis-
sion to eliminate discriminatory procurement practices.
4.4 How Wide?
P     OLICYMAKERS ALSO FACE CHOICES OVER THE RANGE OF
activities to integrate. The discussion so far has
focused on merchandise trade and integration of goods markets,
but there are other aspects of cross-border economic interaction that
may remain tightly regulated. The main examples concern trade in factors
85



T RA D E B LO CS
of production and trade in services. What are the costs and benefits of
extending liberalization to these other cross-border interactions?
Investment Flows
At various points we have referred to the likely effects of FDI flows
in a RIA. These may be flows from outside, or from partners within
the RIA, as plants move to exploit the comparative advantage of dif-
ferent locations and to compete more directly with host country firms.
In practice, there can be substantial barriers to such investment. The
barriers take many forms, including: absolute barriers to establishment
in some sectors or activities; requirements that foreign equity not ex-
ceed a certain percentage of the total; domestic content requirements
or export requirements, requirements that FDI projects meet targets
not imposed on national firms; and obstacles to repatriation of profits.
Little progress has been made on liberalizing these restrictions on a
multilateral basis. The OECD's proposed Multilateral Agreement on
Investment sought to overcome some of these barriers, but agreement
could not be reached. A WTO working group is studying the desirabil-
ity of multilateral investment rules, but there are strong differences of
opinion regarding the costs and benefits of a multilateral set of disci-
plines in this area. Within RIAs too, progress has often been slow, al-
though individual member states of many RIAs have been unilaterally
liberalizing FDI regimes. Argentina and Brazil still restrict FDI flows in
a number of sectors, independent of whether or not the potential inves-
tor originates in MERCOSUR; CARICOM has only liberalized invest-
ment in banking; the CACM countries do not have a common invest-
ment regime. But many RIAs do include provisions to liberalize invest-
ment flows: such as NAFTA, the Group of Three, GCC, and SACU.
The case for liberalizing investment is strong. As we have seen at
many points above, FDI is an important route through which devel-
oping countries can benefit from RIA membership. Agreeing to apply
national treatment and the right of establishment for investors helps
ensure that production choices are not distorted, and is important for
governments seeking to use RIAs as "lock-in" or commitment devices.
Leaving FDI off the table sends a strong signal to the international
financial community that a government may wish to continue to re-
strict international transactions.
86



P OL ICY  CHO IC ES
Services Liberalization
Service trade is inherently more complex than goods trade, for two
reasons. First, in many service activities problems of asymmetric infor-
mation are particularly acute; the purchaser does not know the quality
of a professional service being purchased until after it has been paid for
and consumed. And second, service trade frequently requires consumers
and providers to be at the same place at the same time.
The first of these complexities creates a proper need for regulation of
such service activities. Service suppliers must obtain certification or li-
censing in such fields as financial services, law, accountancy, and medi-
cine. However, standards are often set by professional bodies that have
an interest not only in creating a reputation for ensuring quality, but
also in restricting entry and limiting competition. The second complex-
ity-that services providers typically have to be established in the coun-
try they are supplying-is an inherent obstacle to international trade.
The combination of these two considerations has made services trade
notoriously prone to trade restrictions, and hard to liberalize. We see
existing services trade restrictions taking many forms. Many countries
restrict the access of foreign services and service suppliers to domestic
markets, and sometimes trade in services is simply prohibited. Rights of
supply may be restricted to domestic firms (such as in domestic trans-
portation and basic telecommunication services), or to domestic resi-
dents (such as in legal, insurance, educational, surveying, or investment
advisory services). Even if there are no formal prohibitions, there are
often major barriers to entry. Professional standards are set in a way that
requires foreigners to engage in costly recertification. Rights of access to
telecommunications networks are often restricted: for example, where a
dominant telecommunication carrier-public or private-imposes re-
strictions on the ability of new service providers to link to the network,
or forces them to build infrastructure to reach interconnection points.
Discrimination in ancillary services-not being listed in computer res-
ervation systems-can substantially reduce the competitiveness of an
airline. Limitations on advertising are a common way to limit the ability
of foreign insurance firms to compete, and distribution arrangements
can effectively bar market access for branded products.
This restrictive starting point, and the fact that most of the barriers are
essentially quantity restrictions (prohibitions and regulations rather than
tariffs) means that gains from opening up services trade to international
87



TRADE  B LO CS
competition are likely to be particularly large. Indeed, from a starting
point of prohibitive barriers, a preferential liberalization cannot be "trade
diverting," so a major source of ambiguity about gains from preferential
liberalization is absent. Furthermore, there will be no loss of tariff revenue
for government, since rents created by the regulatory obstacles are typi-
cally collected by protected firms and workers. This does, of course, raise
its own political economy difficulties in seeing through a liberalization.
In addition, the fact that the service sector is an input to so many
other activities in the economy-production, commerce, trade, and edu-
cation-makes it particularly important that the sector function effi-
ciently. For example, the fact that services are an input to production
means that failure to liberalize them faces local producers with higher
costs than necessary, at a time when the RIA is possibly increasing com-
petition and reducing the price of their output. This can give rise to the
phenomenon of negative effective protection; output is not receiving
tariff protection, but inputs are (implicitly, in the form of the barriers to
service trade), so creating negative incentives for the development of
activities that use these services. Estimates for Egypt suggest that trade
barriers in services reduced effective rates of protection for manufactur-
ing activities by some 30 percentage points (Djankov and Hoekman
1997). Since services are an input to trade, service sector inefficiencies
can be damaging for many sectors of the economy. Agricultural output
can be lost due to poor transportation and storage facilities, and sub-
standard communication networks can raise the costs of doing business.
Recent studies of Egypt drew attention to the fact maritime shipping
was a monopoly-and fees were 25 percent higher than those in neigh-
boring countries for equivalent routes. Fees charged by the public com-
panies providing port services for handling and storage of goods were 30
percent higher (Mohieldin 1997). It has been estimated that a service
liberalization that reduces average services prices by 15 percent could
lead to estimated welfare gains for Egypt of some 5 to 10 percent of
GDP (Hoekman and Konan 1999).
Most RIAs have not gone much beyond what was achieved in the
General Agreement on Trade in Services (GATS). The main exceptions
are the EU and NAFTA, although in all cases effective liberalization of
services was not initiated until the 1 990s. NAFTA has made substantial
progress using a negative list system, so all service sectors are covered
unless specifically exempted. Other RIAs vary in their coverage of ser-
vices. The Group of Three is similar to NAFTA, although sectoral cov-
erage is narrower. In MERCOSUR, free circulation of services is a
88



PO LI CY  CHO IC ES
long-term objective to be achieved by 2007; progress has been slow,
with members still negotiating a framework agreement. ASEAN mem-
bers have agreed to full liberalization (on a preferential basis) in most
services by 2020. The FTAs between the EU and Mediterranean coun-
tries do not include services, while those with the Central and Eastern
European countries do. ANDEAN, CACM, and SADC have so far
made little progress (Hoekman and Sauve 1994; Page 1997).
If liberalization is to be pursued, it can proceed through various
channels. GATS was negotiated at the WTO as part of the Uruguay
Round, and WTO members have agreed to begin a new round of ne-
gotiations on services in the year 2000. This provides an opportunity
to improve the structure of the rules and to achieve a higher level of
liberalization commitments.
There are several reasons to think that RIAs may, for many countries,
offer a more effective route to service liberalization. First, service liberal-
ization may involve labor mobility-service providers need to become
established locally, and this typically involves temporary if not permanent
residence. Politically, this may be easier to achieve in a RIA than on a
nondiscriminatory basis. Second, procedures are needed to make sure that
quality standards are met. These issues are conceptually similar to those
we saw with product standards-agreement on harmonization or, if mu-
tually agreeable standards are in place, on mutual recognition. Achieving
these within a RIA may be easier than establishing them for all comers.
However, service liberalization within a RIA is not something that is
likely to be achieved easily. The very fact that these sectors are highly
protected by nontariff instruments means that there are substantial vested
interests, and considerable political effort will be needed to overcome
them. Many RIAs have found it difficult to make much progress, and
only now is the EU effectively securing market opening in these areas.
Further Areas
Investment and services are the two most important areas, beyond
merchandise trade, where there are gains from widening the scope of inte-
gration. Of course, the scope can be set wider still-to include labor mo-
bility, fiscal harmonization, and monetary union. We end this chapter by
cautioning that extending the scope too broadly may detract from the
economic benefits of integration. Some of the benefits from trade that
arise out of differences between countries and the attempts to iron these
89



TRA DE B LO CS
out-by, for example, RIA-wide minimum wage legislation-will inhibit
the very mechanisms by which countries can gain from integration.
4.5 Conclusion
O NE  REASONABLY FIRM CONCLUSION IS THAT FOR MOST
developing  countries,  and  especially  for  the
poorer ones, a North-South RIA with a large industrial country
is likely to be superior to a South-South RIA with a developing country.
The reasons are that:
* South-South RIAs are more likely to generate divergence, with the
less developed member losing relative to the more developed one.
* South-South RIAs are more likely to generate trade diversion.
* North-South RIAs are more likely to generate useful transfers of
technology.
* North-South RIAs are more likely to provide lock-in mechanisms
in the area of politics (such as democracy) and economics (in terms
of policy credibility).
* Given the industrial partner's superior institutions, a North-South
RIA may provide more benefits from "deep integration" than a
South-South RIA.
i Given a larger endowment difference between member countries
in a North-South RIA than in a South-South one, a developing
country may be able to better exploit its comparative advantage in
a North-South RIA.
* The developing country partner is unlikely to capture most of the
above-mentioned benefits of North-South integration unless it un-
dertakes economic reforms.
* Though this chapter advocates forming or joining a RLA with a
industrial country or region, whether and under what circumstances
the latter will allow the developing country to join is unclear. This
is examined in chapter 5.
Other conclusions include:
* For any RIA, lowering external trade barriers (up to the optimum
level) is beneficial: it will increase the RIA's gains or reduce its losses.
90



PO LI CY  CHO IC ES
* Though there are substantial benefits from a CU compared with
an FTA in terms of simplicity of intrabloc trade regulations there
are also costs in terms of loss of sovereignty, difficulty in setting the
common external tariff and sharing the tariff revenues. Thus, most
recent RIAs have been FTAs, and mostly North-South ones where
these costs of forming a CU are likely to be high.
* For middle-income countries-such as the members of
MERCOSUR or ASEAN-there may be sufficient gains from
scale and competition effects to justify a RIA between them, but
fully capturing these effects will require "deep integration" mea-
sures, and expansion and contraction of sectors and firms de-
cided by the market.
* However, if a middle-income country is located close to a large
industrial RIA-such as Mexico, and Eastern European and Medi-
terranean countries-then it is likely to best capture the benefits
of enhanced scale and competition by joining that RIA.
Notes
1.  For development of this argument see Fujita,     6.  Neither do they usually include enforcement
Krugman, and Venables (1999) and Venables (1999).    mechanisms. Resolving interstate conflicts has always fig-
ured on the agenda of international conferences and orga-
2.  Source: WTO. Of course, many RIAs contain    nizations, but has often been ineffective. Early in this cen-
other elements as well; this classification refers only to their   tury a number of proposals were made to develop interna-
policies on trade in goods.                           tional instruments to extend binding arbitration to intel-
lectual property, the principle of equality of foreigners in
3.  Gatsios and Karp (1991, 1995).                 taxation (national treatment), civil and commercial pro-
cedure, customs tariffs, the right of foreigners to hold prop-
4.  In 1988 the EU had already adopted a Single Ad-   erty, the regulation of companies, and claims for damages.
ministrative Document and many members had simpli-   These were opposed by countries that sought to retain their
fied procedures to reduce customs burdens for large trad-   national sovereignty (Murphy 1994). Most of the bodies
ers. The average customs clearance transaction in devel-   mentioned earlier (except for the WTO) do not have bind-
oping countries involves 25 to 30 different parties, 40 docu-   ing dispute settlement mechanisms.
ments, 200 data elements, some 30 of which are requested
at least 30 times, and 60 to 70 percent of which must be  7.  It originates from a landmark 1979 case, in which
re-keyed at least once (Roy 1998).                   the European Court of Justice found that a German ban
on the sale of a French Cassis de Dijon used to prepare an
5. Article 36 of the Treaty of Rome permits EU    aperitif, kir, could not be justified on the basis of public
members to maintain domestic policies that restrict trade   safety or health. At the same time, procedures for harmo-
if this is needed to protect national health, security, mor-   nization were simplified by shifting to a majority voting
als, or the environment. Virtually all RIAs have similar   rule for standards harmonization (Neven 1996). During
provisions.                                           1986-97 some 250 harmonization directives were adopted
91



TRA D E  B LO CS
in the EU tinder majority voting rules, compared to a to-  11. For example, a NAFTA Professional Services An-
tal of only 225 between 1958 and 1985, when the una-    nex sets out procedures for developing standards of pro-
nimity principle applied (Messerlin 1998).             fessional practice, requires the abolition of citizenship and
permanent residency requirements in licensing and certi-
8.  'We look only at long-term investment, not at the    fying professional service providers, and establishes work
issue of;-egulation of short-run capital flows.        programs to liberalize licensing for foreign legal consult-
ants and engineers.
9.  What follows draws on Hoekman and Braga
(1997).
10. rypically a doubling of protective barriers more
than doubles the real income cost of the barrier.
92



C HAP TER 5
Trade Blocs and the World
Trading System
Introduction
D        OES REGIONAL INTEGRATION ENCOURAGE EVOLUTION
toward globally free trade, or does it place
obstacles in its way, and perhaps even increase the likelihood
of trade wars between competing blocs? The stakes in the bet as to
whether RIAs are stepping stones to multilateral trade liberalization or
millstones around its neck are truly huge.' Opening trade and increasing
competition have been behind virtually every sustained economic
growth experience, and the unprecedented postwar growth of world
output and income has clearly been allied to the growth of world trade
and trade liberalization. Progress here affects everyone, and is
particularly important for the small and medium economies that depend
heavily on international trade, and are the principal beneficiaries of an
orderly and nondiscriminatory trading regime.
This chapter takes up these issues. First, we investigate the external
tariffs of RIAs, and ask whether there are reasons to believe that a world
of relatively few large RIAs will have lower or higher tariffs than a world
composed of a large number of separate countries. We then turn to the
dynamics of the world trading system. Does the presence of RIAs create
incentives for excluded countries to join existing RIAs, to form new
RIAs of their own, or to change their external trade policy in other ways?
What are the prospects for "open regionalism" as proposed by some APEC
countries? This leads us into the effects of RIAs on multilateral trade
negotiations-the rounds of GATT/WTO talks. Have RIAs prompted
countries to initiate and become involved in these negotiations, and do
they facilitate or impede successful outcomes of the talks? Finally, we
turn to the rules of the WTO itself, and ask whether the WTO should
treat RIAs differently from its present lax stance.
93



TRADE B LO CS
5.1 How RIAs Set External Tariffs
A        RE THERE ANY REASONS TO BELIEVE THAT A WORLD OF
trading blocs is likely to be more prone to high
tariffs between blocs than a world of separate nations? Some
commentators have foreseen doomsday scenarios in which "Fortress
Europe" and other major trade blocs engage in trade wars, damaging to
themselves and above all to excluded developing regions. Is there any
basis for such views?
A famous insight on the possible effects of trade blocs on tariffs comes
from Paul Krugman (1991a and 1991b)i He noted that trade barriers
would be lowest-and consequently world income greatest-in two
opposite circumstances. One is when there is a single world trade bloc
containing all countries, that is, global free trade. The other is when
trade policy is set by many small independent countries, each so small as
to have no market power and no reason to deviate from free trade. How-
ever, between these extremes each trading bloc has an incentive to use
external tariffs to try and improve its terms of trade (reducing trade vol-
umes to drive up the price of exports and reduce the price of imports).
This reduces world real income, which reaches a minimum for some
intermediate number of trade blocs.3 Extending the insight in a simple
(and not very robust) example, Krugman suggested that the worst num-
ber of similar sized RIAs for world welfare was three. Each sets a tariff to
try and turn the terms of trade in its own favor, but this can only be at
the expense of other blocs. The "prisoners' dilemma" in tariffs is worst
with three prisoners!
This paper sparked a large literature, and many counter-examples.
Perhaps the most pertinent criticism of the analysis surrounds the fact
that tariff setting and trade negotiations involve repeated interactions
between the same countries or blocs, so that the simple logic of the
prisoner's dilemma need not apply. Countries or blocs may be able to
cooperate, particularly if they perceive that the cost of breaking an agree-
ment on tariffs is a future trade war, with all the costs this imposes.
Analysis of this situation (for example, Bond and Syropoulos 1996a and
1996b) suggests that as bloc sizes get larger the returns to cheating on a
trade agreement grow, but so too does the loss from the resulting trade
war. In some cases at least, the former effect dominates, with the result
that it is more difficult to maintain free trade in a world of large trading
blocs, suggesting that regionalism increases the likelihood of protection.
Considerable caution needs to be employed in interpreting these results;
94



TRADE B LO CS AND  THE WORLD  TRAD ING  SYSTEM
they are not very robust, and no consensus has emerged about the mag-
nitude nor even the net direction of general effects.
Turning from the analytical arguments, what is the evidence on the
external trade policy of countries in RIAs? We start by noting that WTO
rules governing RIAs expressly forbid increases in trade barriers (see sec-
tion 6 of this chapter). However, in practice, the rules can be circum-
vented, potentially allowing RIA formation to be associated with tariff
increases. For many developing countries there is a wide gulf between
their actual (applied) tariffs and the maxima committed to in their for-
mal bindings in the GATT. For example, when Mexico nearly doubled
tariffs on 503 imported items from non-NAFTA sources in 1995, it did
so without violating any bindings. WTO rules are ambiguous and poorly
enforced, and a determined government can make trade policy more
restrictive in ways more or less immune to WTO disciplines-say,
through antidumping actions or health regulations. Also, in a world edg-
ing toward general trade liberalization, we need to ask not only whether
RIAs raised tariffs, but whether they caused them to fall less than they
otherwise might have.
We can do this by comparing the external trade policy of RIAs with
that of countries not members of effective RIAs. Foroutan (1998) un-
dertook such a study across developing countries. She divides countries
according to whether or not they are in "effective" RIAs, defined as hav-
ing a material effect on the share of intrabloc trade in total trade, and
compares aspects of their external trade policies.4 Her main findings are:
Average applied tariffi and nontariff barrier coverage. The Latin
American RIAs now have among the lowest average tariffs and
nontariff barrier coverage among developing countries and have
achieved the greatest liberalization since the mid-1980s. Except
for Chile, the small non-RIA group has made much less progress.
Until 1994 neither RIA nor non-RIA countries in Africa had dis-
played much tariff liberalization, and the mean average tariff is
almost the same between the two groups. The most liberal coun-
tries in the sample are the members of North-South RIAs-Israel,
Mexico, and Turkey.
* Uruguay Round Concessions. Here the only feasible comparison is
between the Latin America RIA group and all non-RIA countries.
The former group cut its bound tariff by more and bound more of
its tariffs in the Round than did the latter, although levels started,
and remain, relatively high.
95



TRADE B LO CS
This work is useful for refuting the simple hypothesis that RIAs lead
directly to protectionism and are consistent with the idea that regional-
ism helps to lock in previous liberalization. However, the jury is still out
on the issue; evidently many countries were seeking to liberalize trade by
regional, multilateral, and possibly also unilateral routes, and we are far
from knowing whether incentives for the multilateral and unilateral are
sharpened or blunted by following the regional route.
5.2 The Dynamics of Regionalism
IN CHAPTER 1 WE SAW THE EXPLOSIVE GROWTH OF RIAS IN RECENT
years. Is this, in part at least, because of an underlying
dynamic by which forming a RIA increases the incentive for outside
countries to become members, and so on, until the world is completely
divided into RIAs? The process has been termed "domino regionalism"
by Baldwin (1995), who analyzes how, after three decades of resistance,
three Scandinavian countries decided to seek EU membership in the
late 1 980s. Although they were still uncomfortable with the EU politically,
the economic pressures from the Single Market Program and from EU
expansion were overwhelming. Arguably, the same process occurred when
Canada sought access to the U.S.-Mexican talks that eventually created
NAFTA, and motivated several Latin American and Caribbean countries
to seek accession afterward. The same happened with Chile and Bolivia
seeking association with MERCOSUR, with Mediterranean countries
racing to get Euro-Med Agreements, and even perhaps with a number of
late entrants seeking membership in the Cross-Border Initiative in Africa.5
In part, the drive toward regional agreement is driven by positive
example; countries perceive benefits of membership and act to join or
set up RIAs. But in part, the force comes from perceived adverse effects
of nonmembership.
The Costs of Nonmembership
Why might the existence of some RIAs create or increase incentives
for other countries to join? One obvious reason is simply that countries
perceive benefits of membership, and become increasingly unwilling to
96



TRADE B LO CS AND  THE WO RLD  TRAD ING  SYSTEM
forgo them. But another, more malign reason, is that countries suffer
from being left out, and it is this that creates the rush to join. So how
does the existence of RIAs affect nonmember countries?
The first and most direct way in which nonmembers are affected is
through the change in trade flows caused by a RIA, causing both the
exports and imports of nonmember countries to be smaller than they
otherwise would have been. Such changes do not necessarily have an
adverse effect, although there are several circumstances when they will.
One is when they lead to a terms-of-trade deterioration for excluded
economies. That is, the fall in demand for their exports (and reduction
in supply of imports) may reduce their export price (and raise their
import price). We saw in chapter 3 that member countries have gained
from this, and that excluded country loss is simply the other side of
the coin. Another set of circumstances under which excluded coun-
tries will be harmed by the relative decline in their trade is if their
trade flows are already too small, that is, if they are running at a level at
which marginal benefits exceed marginal costs. This will happen if the
excluded economies have their own restrictive trade policies in place.
It will also happen if firms are operating in imperfectly competitive
markets and are making a positive price-cost margin on each unit of
sales. Losing sales in a RIA market might be particularly damaging to
such firms, causing them to lose profits, and perhaps causing them to
be unable to cover their fixed costs.
Probably more importantly, countries fear that firms may relocate in
search of the benefits of a larger market. This is consistent with what
actually happened to FDI in Europe. Following announcement of the
Single Market Program in the late 1980s, FDI fell in every single EFTA
country. In order to restore their share of FDI, the governments of EFTA
had little choice but to accept the Single Market Program. All except
Switzerland did this, either by applying for EU membership or by join-
ing the European Economic Area. Only once they had announced these
moves did FDI recover (Baldwin, Forslid, and Haaland 1996).
Another source of loss from nonmembership of RIAs is the risk at-
tached to being isolated if a trade war occurs. Of course, all countries-
inside or outside RIAs-will usually lose from a trade war, but countries
in RIAs have the insurance of knowing that they will still have free trade
with partner countries. Whalley (1996) undertakes some numerical simu-
lations of the costs of being outside a bloc during a trade war, showing
97



T RA DE B LO CS
that they can be very sizable. He argues that this creates a strong insur-
ance motive for being inside a large RIA.
The argument is not all one-sided, however. Against these losses, there
are a number of sources of gains that the rest of the world might expect
from the formation of a RIA. We saw in chapter 3 that a RIA might
improve the supply side of the integrating economies-perhaps by im-
proving policy, increasing firms' efficiency, and thereby raising income
levels. Such supply-side improvements will tend to reduce the prices of
products exported by the RIA. This will be beneficial to countries who
import these goods, although damaging to competing exporters.
Generally all these third country effects are likely to be small, particu-
larly when the RIA is between relatively small economies, but there are
exceptions. The important role NAFTA seems to have played in miti-
gating Mexico's 1994 peso crisis clearly benefited the rest of the world.
Bloc Fonnation
If we accept that the existence of RlAs creates demand for additional
RIA membership, the demand can be met either by formation of new
RIAs or by expansion in the membership in existing ones. Expansion
requires the permission of existing members, and we discuss it in the
next subsection (on open regionalism). Formation of new blocs has been
an important alternative, as is evident from the figures on new RIA no-
tifications. The process of bloc formation has been analyzed by Frankel
(1997). He studies a hypothetical world of many countries and four
continents, with zero trading costs between countries within the same
continent and positive costs between continents. Starting from a situa-
tion in which each continent has a nondiscriminatory trade policy, any
one continent could then improve its welfare by forming an FTA and
imposing preferential tariffs. This would harm overseas producers since
they would have to lower their prices to mitigate their loss of competi-
tiveness, and would then suffer from the terms-of-trade decline. From
here a second continent benefits by creating a RIA, switching a loss of
welfare into gain, and thence a third continent, converting larger losses
into smaller ones. Even the fourth continent gains by creating a RIA,
although by then all continents are worse off than under nondiscrimina-
tory policies. World welfare falls at every stage, but no continent has the
incentive to undo the regionalism unilaterally.
98



TRAD E  B LO C S A N D  T H E WO R L D  TRAD I N G  S YS TE M
Open Regionalism?
Open regionalism originated from APEC whose leaders envisioned
a community based on openness and free exchange of goods, services,
and investment.
The term open regionalism has been used in different ways, and with
two quite distinct meanings. One is unconditional nondiscriminatory liber-
alization (or concerted unilateralism). The idea is that as member states
liberalize trade within the bloc, so they simultaneously cut trade barriers
on imports from external countries as well. This was the definition of the
early APEC advocates, who saw the coalition as a means of encouraging
countries to liberalize together and so provide for each other some of the
terms of trade and political economy benefits of a full GATT/WTO round.
Relative to forming a RIA, the policy brings the additional gains of liber-
alizing external trade and thereby removing the source of trade diversion.
Relative to a liberalization by a single country, the fact that it would be
concerted, with all members liberalizing together, brings additional ben-
efits as access for imports is eased, so firms get improved access to partner
markets; this reduces the likelihood of liberalization leading to a terms-of-
trade loss. However, despite these attractions, the idea of concerted
unilateralism does not now seem likely to make headway. Within APEC,
the United States is implacably opposed to the idea. More generally, the
WTO system is so firmly based on the notion of trade liberalization being
a concession (to be granted in return for some concession by trading part-
ners) that the idea is unlikely to catch on.
A second, and quite different concept is open access whereby the RIA
announces that any country willing to abide by its rules may join. In
terms of the economics, such an arrangement is still preferential, giving
discriminatory benefits to members. Its main importance would be as a
stepping stone to multilateralism: Could an open access RIA attract an
increasing number of members, to the point where almost all countries
became members?
Analytical treatments of this issue are not optimistic. As we saw in
chapter 4, there is generally an optimal size for a RIA, from the point
of view of existing members, so it is not clear why members should
want unrestricted access. In practical terms, the feasibility of open ac-
cess depends crucially on the depth of the scheme. Where a RIA in-
volves few conditions, then open access can be quite easily envisaged.
Perhaps the best example is the Cross-Border Initiative in East and
99



TRA DE  B LO CS
southern Africa, in which neither internal preferences nor external tar-
iff harmonization are rigorously enforced. But for deeper agreements,
open access is harder to envisage.
NAFTA, where the principal rules are completely free trade in goods
and open investment, might seem to be a candidate, but has in fact re-
jected many overtures. This seems more to avoid adjustment in the United
States than because other issues such as quotas for professional migration
or dispute settlement require negotiation. MERCOSUR has expressed will-
ingness to accept new members, implying a degree of open access. How-
ever, given that fairly deep integration is planned, detailed negotiation is
required. Association-the status preferred by Chile- is easier, but does
not offer full integration and still requires several years of talks.
The EU stands ready to sign association agreements with many neigh-
boring countries, and ostensibly with the African, Caribbean, and Pa-
cific countries, but only on its own terms covering issues such as rules of
origin, excluded sectors, the use of antidumping duties, and so on. Full
membership in the EU is anything but open access. The United King-
dom had to ask three times before it was allowed to join, and Turkey was
rejected until recently. Negotiations are tortuous once accession is agreed
in principle. Each of the five current Eastern European candidates faces
a formidable list of demands and requirements prior to membership. In
several cases they are required to adopt policies from which some exist-
ing members have negotiated exemptions (such as the Social Chapter).
Still, EU membership has increased from 6 to 15 countries and is ex-
pected to increase to 25 countries over the next decade or so.
In practice, what we are seeing here is essentially that any country is free
to apply; but the price of entry is set separately for each entrant. This can
lead to asymmetric agreements in which benefits to developing country
candidates are reduced and possibly appropriated by existing members
through side conditions on issues such as the environment, labor regula-
tions, and rules of origin. Since the more complex aspects of RlAs-espe-
cially those with budgetary implications-have to be negotiated, access
can never be automatic and unconditional. In practice, it will not be easy
for the WTO to write or enforce general rules for open access, and it is
unlikely that this route will lead to ever expanding membership.
However, even though full and unconditional open access to RIAs is
unlikely to be feasible, increased access by developing countries to the
markets of the trade blocs in the industrial North is more likely through
100



TRADE B LO CS AND THE WORLD  TRAD ING  SYSTEM
a successful WTO or through association agreements with the EU, the
United States, or Japan. The latter requires the WTO to modify its rules
regarding RIAs and create a presumptive right of association. As with
the most favored nation (MFN) clause, if association is granted to one
country, there should be a presumption that it should be available to
others. For instance, if Iceland is offered reciprocal freedom from anti-
dumping suits by the EU, the same option should be available to Ghana.
In practice, association is complex and each accession needs to be nego-
tiated, but the poor should not be automatically denied association rights
provided to middle-income countries by the EU and NAFTA. This is-
sue is examined in more detail in section 5.6.
What about APEC itself, where the idea originated? It has yet really
to decide between these alternatives, because there has not yet been any
APEC liberalization. Members have certainly not yet introduced any
discriminatory trade policies (with the minor exception of the APEC
business visa), but neither have they yet moved beyond implementing
their Uruguay obligations and, for developing members, their own uni-
lateral reforms. The manifesto still contains a commitment to global
liberalization, but it is worth recalling that, although more positive in its
approach and timetable, this manifesto is not very different from the
EEC's 1957 statement that "by establishing a customs union between
themselves, member states aim to contribute, in the common interest,
to the harmonious development of world trade, the progressive aboli-
tion of restrictions on international trade and the lowering of customs
barriers" (Article 1 10.1, Treaty of Rome).
5.3 Regionalism and Multilateral Trade
Negotiations
O  +  NE OF THE MAIN VEHICLES FORTRADE LIBERALIZATION HAS
been the series of GATT/WTO rounds of trade negotiations.
From the late 1940s through the present, these rounds have
reduced average tariff rates on manufactures from more than 40 percent
to less than 5 percent. How do RIAs impact on multilateral trade
negotiations? Do they promote the initiation and conclusion of these
rounds, or provide obstacles to their progress? Three sorts of arguments
have been made.
101



T R A D E B L O C S
Multilateralism as a Response to Regionalism
The first argument is that countries outside RIAs may react to their
exclusion by attempting to accelerate multilateral liberalization. Many
commentators suggest that the creation of the EEC in 1957 had this ef-
fect. This, they suggest, led directly to the Dillon and Kennedy Rounds of
GATT negotiations as the United States sought to mitigate the EEC's
potential for diverting trade (Lawrence 1991; Sapir 1993; WTO 1995).
Although perfectly conceivable, this is not a straightforward argument.
First, it seems unlikely that multilateral negotiations would have
stopped completely had the EEC not been created, especially given the
global reach of the United States during the 1960s. Thus, at most, the
EEC affected the timing, not the occurrence, of the Rounds. Second,
agriculture played an important role in the formation of the EEC, and
the EEC was probably more successful in resisting that sector's liberal-
ization in the multilateral trade negotiations than its individual mem-
bers would have been. This has probably made future liberalization more,
not less, difficult. Third, suppose that the hypothesis were true, that the
creation of the EEC had led to negotiations. The logic of the argument
is essentially coercive: EEC members did something that their trading
partners considered harmful, and then offered to mitigate it in return
for concessions. Coercion may be warranted and the outcome may have
been beneficial, but this is a dangerous game. It depends critically on the
willingness of the partners to fold-(by negotiating) rather than fight
(by raising tariffs) and to respond multilaterally rather than regionally.
It has also been argued that regionalism was behind the Tokyo
Round. Winham (1986) reports both the first EEC enlargement (in-
cluding free trade with EFTA) and the restrictiveness of Europe's Com-
mon Agricultural Policy as factors in the United States desire for a
Round . The former observation seems no more compelling than those
surrounding the creation of the EEC, while the latter is distinctly two-
edged: it is based on regionalism having increased trade restrictions in
agriculture (Common Agricultural Policy), and a response to this be-
ing negotiation. For the net effect of this on multilateral progress to be
beneficial requires a negotiating structure in which might and
countervailing power are the critical elements of liberalization.
Finally, consider the Uruguay Round, of which the WTO (1995)
says, "There is little doubt that ...the spread of regionalism [was a] major
factor in eliciting the concessions needed to conclude" the Round. There
102



TRADE B LO CS AND THE WO RLD TRAD ING  SYSTEM
was, indeed, a perception that the failure of the Round would lead to
regional fragmentation. This almost certainly encouraged the spread of
defensive regionalism, but whether this pressured the two major parties
to agree is not clear, for they were the prime "regionalists," and they
would certainly not have been the principal casualties of fragmentation.
Some senior EU negotiators have said that the 1993 Seattle APEC Sum-
mit induced the EU finally to concede on agriculture and conclude the
Uruguay Round (Bergsten 1997). Again, this may be true, but there are
counter-arguments. For example, APEC was not advertised as a discrimi-
natory RIA, and any discrimination would, anyway, have been far in the
future. Also, the principal necessary condition for the EU to complete
the round was agricultural reform, which was initiated in 1990 and com-
pleted in 1992 (Hathaway and Ingco 1996).
Does Regionalism Facilitate Negotiation?
If regional integration agreements made trade negotiations easier,
perhaps they would help the world evolve toward freer trade. Coordi-
nated coalitions may have greater negotiating power than their mem-
bers do individually and such coalitions may facilitate progress just by
reducing the number of players represented in a negotiation (Kahler
1995). Whether this really occurs is a moot point. For example, a nego-
tiation comprising one dominant partner and a competitive fringe of
small countries might be easier and proceed further than if the fringe
coalesced into a significant counterforce.
This line of reasoning prejudges the issue of whether blocs are genu-
inely unified in their approach to trade negotiations. This is not usually
so, meaning that any gains from having fewer players in the last stage of
a negotiation are offset by the complexity of agreeing joint positions in
the first phase. The difficulties of achieving a European position on ag-
riculture and cultural protection in the Uruguay Round are well known,
and formulating EEC positions in the Tokyo Round proved complex
(Winham 1986). Moreover, two-stage negotiations need not be more
liberal than single-stage ones (Basevi, Delbono, and Mariotti 1995). To
be sure, Germany and the United Kingdom pressured France to agree to
the agricultural deal in the Uruguay Round, but the liberalizers had to
make potentially trade-restricting concessions on "commercial defense
instruments" to clinch the deal. Wang and Winters (1998) argue that
103



T RA DE B LO CS
the benefits to African countries of enhancing their bargaining power by
cooperation are likely to be outweighed by the costs of combining their
different interests into a single negotiating position.
While the EU has internal procedures for arriving at a common posi-
tion, many of the CUs that attend the next round of global trade talks
have not yet developed procedures for determining their negotiating po-
sitions. SACU's previous practice of delegating all responsibility to South
Africa begins to look less tenable as divisions emerge between members,
and MERCOSUR has yet to devise really robust internal decisionmaking
capacity. Thus, at least into the foreseeable future, RIAs do not seem
likely to facilitate even a traditional trade negotiation. Moreover, as WTO
has extended its reach, it has embraced subjects in which most central
CU authorities have no mandate to negotiate. Mixing national and CU
responsibilities seems unlikely to simplify matters, and it is not realistic
to expect member countries to surrender sovereignty on sensitive issues
to regional bodies just because trade negotiations are in process.
Successful trade negotiations also require political will and adminis-
trative effort. Reserves of administrative skill, political capital, or imagi-
nation are finite; if they are devoted to a RIA they are not available for
multilateral objectives. These arguments were advanced to explain both
EU and U.S. behavior during the Uruguay Round, but they must be
several times more important for developing countries. Negotiating a RIA,
especially with a major power that has its own objectives, will absorb a
huge proportion of the scarce policymaking skills of a developing coun-
try. Perhaps one of the opportunity-costs of RIAs is that less negotiating
capacity and political capital are available for multilateral negotiations.
One alarming possibility is that regionalism might undermine U.S.
7
or EU willingness to participate actively in the multilateral system. Over
the last three decades they have been major players, monitoring both
smaller countries' policies and each other's. A loss of interest by either
would reduce WTO's overall effectiveness and could upset the current
fine balance.
Regionalism and the Frontiers of Liberalization
One of the strengths that is frequently claimed for the regional ap-
proach to liberalization is that it makes it easier to handle the tough
issues (Kahler 1995); there are areas in which regional liberalization or
104



TRAD E B LOC S AN D  TH E WO RLD TRAD I NG  SYSTEM
harmonization between similar or like-minded countries is feasible when
multilateral progress is not. As we have seen (chapter 4) the EU has
addressed a wide range of "deep integration" issues; NAFTA has tackled
investment; Brazil has free trade in information technology goods within
MERCOSUR; Chile and Canada have eschewed antidumping actions
on mutual trade.
But until recently even RIAs among industrial countries, let alone
those among developing ones, had not advanced much further with lib-
eralization than had the multilateral system (Hoekman and Leidy 1993).
Thus, for example, agriculture frequently remained restricted, transport,
culture, and other sensitive services were excluded; and government pro-
curement was ignored de facto if not de jure. The EU, especially in the
Single Market Program, has now advanced further, but this took 30
years to initiate and is, to date, unique. Overall, RIAs have not led on
the tough issues to the extent that is sometimes supposed.
The benefit of having RIAs tackle these "tough issues" depends heavily
on whether they are liberalizing and whether they are otherwise well-
suited to developing country needs and capacities. The EU has used the
Europe Agreements to obtain action on environmental and labor condi-
tions and on intellectual property, and the United States has used NAFTA
as a tool for enforcing Mexican labor and environmental standards. It is
quite possible that practices developed by the major RIAs will not suit
developing countries well, and may also be less open and liberal.
An extension of the "tough issues" argument is that RIAs help develop
blueprints for subsequent multilateral negotiations (Bergsten 1996). For
example, the EU pioneered "bulk" mutual recognition for industrial stan-
dards and services harmonization, and NAFTA's investment chapter may
inform a multilateral investment negotiation (if there is one). On the other
hand, the EEC also suggested the Common Agricultural Policy as a model
for agriculture in the Kennedy Round. Once again then, the blueprints
might not suit developing countries well. There is the further danger that
coming to negotiations with a fully developed blueprint can appear to be
a pressure tactic that actually makes negotiation more difficult; develop-
ing countries' de facto rejections of the OECD draft Multilateral Agree-
ment on Investment in 1998 contains at least an element of this.
Overall then, we find the arguments that regionalism has promoted or
facilitated multilateral trade negotiations to be rather weak. There is little
evidence that RlAs have either prompted or facilitated negotiations, and real
dangers that they might dilute countries' involvement in such negotiations.
105



T RAD E B LO CS
5.4 Regionalism and the wTO
T       HE PREVIOUS ANALYSIS SUGGESTS THAT INTERNATIONAL
policy toward regionalism should aim to:
* Encourage RIAs to achieve trade creation and avoid trade diver-
sion, both for the sake of members and to minimize harm to ex-
cluded countries, for instance, by setting low external tariffs
* Permit deep integration, including nation building, between
members
* Preserve the effects of previous liberalization and provide credibil-
ity for any liberalization that form part of the RIA
* Support a liberalizing dynamic within member countries and in
the world trading system as a whole.
The instrument we have for international policy on regionalism is the
WTO, and this section explores how it manages regionalism and whether
its rules could be reformed to help it do better. RIAs are officially sanc-
tioned-but conditional-exceptions to the GATT's rules on nondis-
crimination. The conditions imposed on RIA formation doubtless con-
strain and mold the pattern of regionalism in the world, but they are
neither adequate, nor adequately enforced, to ensure that regionalism is
economically beneficial for either its members or excluded countries.
The responsibility for good outcomes falls on governments themselves;
they cannot tie their own or each other's hands sufficiently tightly in the
WTO to preclude the possibilities of signing harmful RIAs.
The world trade system works-pragmatically and consensually. The
GATT was created in 1947 as a temporary body to assist countries in
trade liberalization. Its role was to codify and record a series of tariff reduc-
tions that its members wished to make, and provide a structure to give
credibility to those reductions. It discouraged the reversal or nullification
of tariff cuts by restricting policies that impose duties on an ad hoc basis,
such as antidumping duties and emergency protection, and equivalent
policies such as internal taxes on imports. It also defined important me-
chanics of trade, such as the valuation of trade for customs purposes. A
key concept of the GATT, indeed the cornerstone of the present world
trading system, is nondiscrimination between different sources of the same
imported good, which is achieved by requiring members to give each other
MFN treatment, except in specified circumstances. With an assurance of
106



TRADE B LO CS AND  THE WO RLD  TRAD ING  SYSTEM
nondiscrimination, when A negotiates a reduction in one of B's tariffs, it
knows that the commercial value of its effort will not be undermined by B
then offering C an even lower tariff This, in turn, makes A more willing
to "buy" the concession by reducing one of its own tariffs on B, and so
encourages trade liberalization.
Over 50 years of operation the GATT continued in this low-key,
member-driven, fashion. It was essentially consensual in approach and
pragmatic in operation. The GATT did not adjudicate trade disputes,
but had a dispute settlement process, less concerned with law than with
solving disputes in a way that preserved consensus and allowed the liber-
alizing bandwagon to continue to roll. The WTO, which was created in
1995 to oversee the GATT and certain other agreements, is more legal-
istic, but still focuses heavily on pragmatic and mutually acceptable so-
lutions to problems. The WTO/GATT administers a set of rules for
behavior, not a set of outcomes-it is concerned with meeting agreed
obligations and rights rather than with economic outcomes per se. The
WTO/GATT has undoubtedly been a force for economic good, but its
role has not been defined in those terms.
The GATT traditionally did not intrude into domestic politics. It
had no ability to force member countries to liberalize if they did not
wish to, and was extremely light-handed in its requirements about the
shape of domestic legislation. The WTO is rather more far-reaching;
through its greater breadth and its "single undertaking," under which
members must subscribe to virtually all its rules, rather than, as previ-
ously, treat some as optional extras, it has constrained governments more
tightly. Nevertheless, the WTO can still be effective only if it proceeds
more or less by consensus.
Given this background, the WTO can enhance the economic well-
being of developing countries in four ways. First, if sufficient members
wish, it can organize periodic rounds of tariff negotiations that offer
opportunities and incentives to members to reduce their barriers to trade.
Second, it provides guidelines for domestic policy-directly in some cases,
but more often indirectly by shaping the terms of the debate. Govern-
ments resisting pressures to protect particular lobbying groups are im-
measurably strengthened if they can point to prohibitions in the WTO
agreements. Third, the WTO can protect the rights of members against
certain rules violations by other members. It cannot necessarily, how-
ever, protect members against harm.10 Fourth, it provides a forum and
mechanism for governments to manage the spillovers from members'
107



TRADE B LO CS
trade policies onto their partners. These four links provide the frame-
work for assessing the WTO's current rules about RIAs and exploring
whether they can be improved.
5.5 The Rules for RIAs
A        RTICLE XXIV OF THE GATT SETS LIMITS ON THE RIGHTS OF RIA
member countries to violate the MFN principle. It imposes three
principal restrictions (appendix 1). A RIA must:
* Not "on the whole" raise protection against excluded countries
* Reduce internal tariffs to zero and remove "other restrictive regula-
tions of commerce" other than those justified by other GATT ar-
ticles
* Cover "substantially all trade."
These conditions ensure that RIAs do not undermine the access of
other countries to the RIA market. The first preserves the sanctity of
tariff bindings by ensuring that forming a RIA does not result in a whole-
sale dissolution of previous bindings. It is supplemented by a rule that
compensation is due to individual partners for tariff increases induced
by the RIA if other reductions to keep the average constant do not main-
tain a fair balance of concessions. Together with the 1994 Uruguay Round
Understanding on the Interpretation of Article XXIV on how to mea-
sure tariff barriers for RIAs, these provisions offer reasonable assurances
that the barriers facing nonmembers will not be raised.
The second condition helps defend the MFN principle by making it
subject to an "all-or-nothing" exception. If countries were free to nego-
tiate different levels of preference with each trading partner, binding
and nondiscrimination would be fatally undermined: no member could
be sure that it would receive the benefits it expected from negotiating
and reciprocating for a partner's tariff reduction. Also, if a customs union
is a first step toward nation building, it is inappropriate for an interna-
tional trade treaty to stand in the way of such progress. Thus, internal
free trade, such as one (usually) achieves within a single country, would
seem to be an acceptable derogation of MFN, whereas preferences would
not. The third condition reinforces this by requiring a serious degree of
commitment to a RIA in terms of sectors.
The second and third conditions-no internal tariffs and substantial
coverage-are important in heading off pressure to use tariffs to fine-tune
10o8



TRAD E B LOCS AND  TH E WO RLD  TRAD I NG  SYSTEM
political favoritism toward either domestic industries or partner countries;
they help to prevent governments that restrict RIAs from swapping trade-
diverting concessions and, thus, from avoiding politically more painful
trade creation. These conditions essentially require a serious commitment
to integrating member markets as a condition for proceeding.
Article XXIV is generally an aid to better RIAs, but it is certainly not
sufficient for good economic policy. Even if the conditions were applied
without exception they would not preclude harmful RIAs: Wholly GATT-
compatible RIAs can be predominantly trade diverting, excluded coun-
tries can suffer terms-of-trade declines, and institutions can arise that
make liberal policies less likely.
There are major difficulties in interpreting the conditions of Article
XXIV Even following the Uruguay Round Understanding there is no agree-
ment about what "substantially all trade" means, nor even whether it re-
fers to the proportion of actual trade covered or the inclusion of all major
sectors of the economy. Similarly the treatment of nontariff barriers in
assessing the overall level of trade restriction is not defined, nor is that of
rules of origin. The requirement that "other restrictive regulations of com-
merce" be removed between members is ambiguously worded: several ex-
ceptions to this requirement are identified explicitly but other barriers,
including antidumping duties and emergency protection, are not. Com-
plete integration between members of a RIA would abolish these barriers
and so their continuation-in NAFTA or the Euro-Med agreements-
suggests an unwillingness to proceed too far in that direction.
Perhaps because of its ambiguities, Article XXIV has been notori-
ously weakly enforced. RIAs have to be notified to the GATT and until
1996, each was then reviewed by an ad hoc working party to see if it was
in conformity with the Article. WTO (1995) reports that of 69 working
parties reporting up to and including 1994, only 6 were able to agree
that a RIA met the requirements of Article XXIV, of which only
CARICOM and Czech-Slovak CU remain operative. However, the re-
mainder did not conclude that agreements were not in conformity-
they merely left the matter undetermined.
This agnosticism is essentially the product of the GATT's consensual
nature. The first major test of the article was the Treaty of Rome estab-
lishing the EEC. The political pressure to permit it was enormous: EEC
countries would almost certainly have put the EEC before the GATT in
the event of conflict and the United States strongly supported the treaty.
The treaty, however, clearly violated Article XXIV, and so the only fea-
sible solution was not to push the review to conclusion. Given a start
109



T RA DE B LO C S
like this, the EEC's willingness to support more or less any RIA in the
GATT, the need for working parties to reach consensus, and the GATT's
inability to make an adverse determination without the acquiescence of
the party at fault," it is hardly surprising that future reviews proved little
more demanding. Nor have matters improved with the establishment in
1996 of a single Committee on Regional Trading Agreements to conduct
the reviews. The inability to rule on whether RIAs conform to article
XXIV does not mean that the rules have had no effect, for we do not
know the extent to which they have influenced the structure of RIAs that
have come forward, nor which potential arrangements they have discour-
aged.1 It is not an encouraging record, however, either from the point of
view of enforcing current rules or from that of rewriting the rules to
increase their ability to distinguish good from bad RIAs.
Finally, Articles XXIV.10 and XXV of the GATT can be used to
grant waivers to make otherwise inadmissible policies GATT-legal. This
was done for the European Coal and Steel Community (1952) and the
U.S.-Canada Auto Pact (1965). Under WTO, waivers are still feasible
but are time-limited.
Article XXIV of the GATT refers to trade in goods. The equivalent
for services is Article V of the GATS, which is modeled closely on it.
The requirement not to raise barriers to third countries is rather tighter:
it is applied sector by sector rather than "on the whole," and third country
suppliers already engaged in "substantive business" in a RIA territory
before the RIA is concluded must receive RIA treatment. The "substan-
tially all trade" ambiguity is only slightly abated, with an explicit note
that the word "substantially" be "understood in terms of number of
sectors, volume of trade, and modes of supply." For covered sectors
"substantially all discrimination" is to be removed, but since this is de-
fined as comprising elimination of barriers or prohibition on new or
more discriminatory barriers, or both, it need amount to very little.
Developing countries receive "flexibility" on "substantially all" discrimi-
nation and exemption from the need to give RIA treatment to least
third country firms with "substantial business" in member countries.
The Rules for Developing Countries
If all this were not enough, a further complication for developing
countries is an "Enabling Clause"' introduced in 1979 that significantly
110



TRADE B LO CS AND  THE WO RLD  TRAD ING  SYSTEM
relaxes the conditions for creating RIAs that include only developing
countries. It drops the conditions on the coverage of trade, and allows
developing countries to reduce tariffs on mutual trade in any way they
wish, and nontariff measures "in accordance with criteria which may be
prescribed" by the WTO members. It then supplements the first condi-
tion with the nonoperational requirement that the RIA not constitute a
barrier to MFN tariff reductions or cause "undue difficulties" for other
contracting parties.
In practice, developing countries have had virtual carte blanche. Twelve
preferential arrangements have been notified under the Enabling Clause,
including the Latin American Integration Association, ASEAN, and the
GCC. Internal preferences of 25 percent and 50 percent figured in
ASEAN's trading plans and also in many of the arrangements concluded
under the Latin American Integration Association and in the GCC. There
is little sign that internal preferences have undermined MFN agreements
with other trading partners but then, until recently, these countries did
not make many MFN agreements. Indeed, until the late 1 980s, the Latin
American and African countries' frequent use of regional arrangements
and weak participation in the multilateral rounds might suggest a sub-
stitution of one form of liberalization for the other. More worrying were
the sectoral agreements that abounded in Latin America.
The Enabling Clause dilutes the weak discipline that Article XXIV
imposes. Even if Article XXIV does not actually stop many harmful prac-
tices, it does at least avoid automatically giving them the respectability
of legal cover. Thus, while the GATT knowingly and willingly permit-
ted Latin American Free Trade Area (1960) and the initial notification
of ASEAN (1977) to violate Article XXIV (Finger 1993), at least it re-
quired continuing consultation with partners and left open the possibil-
ity of challenge in the dispute settlement process. The Enabling Clause
offers more cover in various areas, and thus erodes even this discipline.
Refonn of the Rules?
The WTO rules on RIAs are not exactly broken, but they are creaky,
and it is worth asking what might be done about them. We focus here
on their economic content, and on the feasibility of reform. Feasibility
seems to be a more binding constraint than devising economically sen-
sible rules. Indeed, major political backing for tightening looks im-
111



TRA DE  B LO CS
probable, as few countries within RIAs appear to seek tighter disci-
pline, as the EU continues its Mediterranean agreements and consid-
ers replacing the trade provisions of the Lome Convention with an
FTA, and as the United States contemplates the Free Trade Area of the
Americas.'7 However, see section 5.6 for a proposal on changing rules
for both industrial and developing countries.
A RIA that does not reduce external barriers may cause trade diver-
sion. One discipline on this would be to require RIA members to liber-
alize, both to reduce diversion and to induce external trade creation with
nonmembers. Finger (1993) views these reductions as a price to be ne-
gotiated to persuade nonmembers to forgo their MFN rights. How far
the parties are prepared to go in a negotiation, however, is determined
by the prevailing rules and enforcement mechanisms that define the
outcome if negotiations fail; unfortunately, these currently leave non-
members almost no negotiating power. Hence other authors have made
more concrete proposals.
Bhagwati (1993) suggests requiring that for each tariff heading a CU's
common external tariff be bound at the minimum tariff for that head-
ing among all members. This does not guarantee the elimination of trade
diversion-suppose the tariffs of three members were 98 percent, 99
percent, and 100 percent-but it will clearly reduce it. It would impose
a high (mercantilist) price on RIA formation, so only "serious" integra-
tors would pay it, and would, overall, be quite trade-liberalizing. As a
reform it is admirably clear, and if feasible, it would be desirable eco-
nomically. Its demanding nature, however, makes it very unlikely to suc-
ceed in the present circumstances.
Related is a proposal that members of FTAs be required to bind
their tariffs at actual applied rates on the eve of the RIA. Apart from
what this might do to pre-FTA applied rates, this suggestion is ran-
dom in its liberalizing effect, which reduces its moral force. Bhagwati
would just ban FTAs. This is also consistent with seeking to restrict
RIAs to those that are committed to far-reaching integration, but again
faces severe feasibility constraints, especially since some FTAs proceed
quite far in other directions.
Tied up with the FTA question is that of rules of origin. Some sug-
gest a requirement that they be no more restrictive than before the
RIA, but this is difficult to determine, ad hoc, manipulable in nature,
and potentially very complex in the face of technological changes. Bet-
ter would be a requirement precluding the manipulation of such rules
112



TRAD E  B LOCS AN D  TH E WO RL D  TRAD I NG  SYSTEM
for protectionist purposes, such as that countries should adhere to a
single set of rules of origin agreed internationally, or that a country's
preferential rules should be the same as its nonpreferential ones.
Wonnacott (1996) suggests a number of milder reforms in this direc-
tion: for example, that rules of origin be banned where tariffs differ
between members by less than, say, 2 percentage points, or that for
each commodity they be banned for the FTA member with the lowest
tariff. These might be acceptable, but would only scratch the surface.
One proposal has been made to adopt ex post reviews to determine
whether nonmember exports have fallen since a RIA was created and de-
mand changes in policies if they have (McMillan 1993). Although fre-
quently taken seriously (for example, Frankel 1997), the proposal is wrong
in virtually every respect. Exports are the wrong criterion, quantitative
targets are the wrong way to formulate trade policy, the internal costs of
trade diversion are ignored, economic modeling is still too imprecise to
identify causes with any credibility, and expostadjustment after five years
is no basis for the policy predictability sought by investors.
There are three major proposals for creating a "liberal dynamic."
Srinivasan (1998) proposes that RIAs be permitted only temporarily by
requiring all RIA concessions to be extended to all countries within, say,
five years. This is effectively a ban on RIAs, and certainly foregoes any
gains that they might offer in terms of deep integration or nation build-
ing. It is not a serious contender.
Second, stretching back at least to the United States submission to
the Preparatory Committee of London Monetary and Economic Con-
ference of 1933, scholars and policymakers have argued that requiring
RIAs to admit any country willing to accept their rules both reduces
their adverse effects on excluded countries and establishes a liberal dy-
namic (Viner 1950). While this may be true if admission can be guaran-
teed, virtually every RIA extant has geographical restrictions on mem-
bership and has features that require negotiation. The latter vitiate the
promise of "open access." However, though unconditional open access
seems unfeasible, section 5.6 suggests a way to improve developing coun-
tries' access to the large blocs in the North.
A more feasible approach than unconditional open access is to define
and enforce current rules more rigorously. A precise definition and en-
forcement of "substantially all trade" would be a useful innovation. A
quantitative indicator would be clear, but it would need to be high given
that the kinds of trade restrictions countries wish to maintain typically
113



T RAD E B LO CS
constrain existing trade quite fiercely. The frequently cited 80 percent,
which dates from consideration of the Treaty of Rome is not adequate.
Even 90 percent, which seems to inform current EU-MERCOSUR talks,
is not indicative of serious intent to integrate. We would advocate 95
percent after 10 years and 98 percent after 15 years. Similarly, a more
constraining view of "other restrictive regulations of commerce" would
be useful-ensuring that they include the effects of rules of origin on
excluded countries, and that obvious barriers such as safeguards actions
and antidumping duties are abolished internally. The latter requirement
would increase the degree of trade creation, since these policies are ex-
plicitly aimed at preserving domestic output levels. Thus they would
raise the bar for "serious" regionalism.
However, even these changes might encounter fierce opposition and
will require major political commitment by many WTO members to be
implemented. To be acceptable to the major powers, they will certainly
need to be accompanied by a grandfathering clause to assure current
RIAs that they will not be undermined by new interpretations.
A vehicle to take forward reform measures is the Committee on Re-
gional Trading Agreements (CRTA), which reports to the WTO's Gen-
eral Council, and was established in 1996 to increase the transparency,
efficiency, and consistency of the WTO's treatment of RIAs. It was seen
as a means of ensuring more rigorous review of new RIAs because a
single group would review all of them using the same criteria and with
more searching notification and information requirements. It would also
undertake periodic review of existing RIAs, and could resolve some of
the systemic issues that remained after the Uruguay Round. The more
thorough review was seen as a route to better compliance with WTO
requirements, while the consideration of conceptual issues was a step
toward refining and codifying the rules more precisely.
Unfortunately the CRTA has not yet reached its stride. Its assessments
of particular cases have been stymied by the lack of clear systemic rules,
and the discussion of rules stalemated on exactly the same "substantially
all" and "other regulations" issues as the previous Uruguay Round discus-
sions. By December 1997 the Committee had initiated consideration of
59 RIAs (including 32 inherited from previous working parties). It had
completed factual analysis of 30 of these, and was "elaborating conclu-
sions" on 26. To date, no analyses have been released or conclusions reached.
The future development of the CRTA could take several routes. Re-
view of existing RIAs is not likely to be productive. Although RIAs are
open to dispute if third party countries feel aggrieved (at least until the
114



TRADE B LO CS AND  THE WO RLD  TRAD ING  SYSTEM
CRTA has formally certified their WTO-conformity), the rules seeking
serious intent to integrate are not really susceptible to this process. Non-
member countries are unlikely to press RIAs to include more sectors
when they expect this to increase trade diversion. Similarly, why would a
nonmember country seek to free RIA members from the threat of each
other's antidumping legislation? Members do not normally bring inter-
nal disputes to the WTO.
When it comes to new RlAs, one possibility is to have a detailed case-
by-case study of the likely effects to the RIA. But the CRTA faces a serious
timing problem. Unless agreements are submitted to the WTO early in
the process of negotiation-in which case they will be very provisional-
reviews will generally be too late to influence their initial form. Otherwise,
reviews will be too late to affect public debate and will, if they call for
changes, upset carefully negotiated compromises. For that reason they will
be resisted and resented by members, which is bad news for a consensual
organization. Thus considerable political courage will be called for to en-
force CRTA findings until their requirements are sufficiently understood
and respected by members to be met ab initio. This process of review and
response would have to be aided by detailed economic studies of RlAs
stretching well beyond the legalities of Articles XXIV and V.
The better way forward is for review of new RIAs to be restricted to
ensuring compliance with the tighter Article XXIV rules on liberaliza-
tion of "substantially all trade" and "other restrictive regulations of com-
merce" which we proposed above. As we have argued, these rules would
raise the bar for "serious regionalism," while leaving it clear that WTO
approval does not validate the economic benefits of a RIA for its mem-
bers. The responsibility for good RlAs lies with governments themselves.
5.6 Post-Seattle
T       HE FAILURE OF WTO MEMBERS TO AGREE TO A NEW ROUND
of negotiations at the Ministerial in Seattle in
December 1999 has dealt a blow to the WTO. One of the
dangers is likely to be that developing country members that were
thinking of fuller participation in the WTO and were considering further
unilateral trade liberalization and binding their lower tariffs at the WTO
may have become further disappointed by the multilateral system after
Seattle and may be looking more seriously at regional options. As
examined in the report, RIAs among developing countries are unlikely
115



TRADE B LO CS
to provide the benefits available in North-South RIAs or in the
multilateral system. Developing countries have much to gain from
multilateral liberalization as enforced by the MFN clause: it strengthens
weaker countries by limiting the ability of stronger ones to make deals
with each other that exclude the weaker ones.
In order to more fully integrate the developing countries in the mul-
tilateral system and make it work for them, simply extending the disci-
plines of Articles XXIV and V to their RIAs will not work. We believe a
quid pro quo is necessary where industrial countries offer something in
return for the developing countries' acceptance of these disciplines.
Mike Moore, secretary-general of the WTO, argued at the January
2000 United Nations Conference on Trade and Development-X Con-
ference in Bangkok that richer nations need to bring down trade barriers
to exports from developing countries. In his presentation he stated that,
"It makes no sense to spend extra billions on enhanced debt relief if, at
the same time, the ability of poorer countries to achieve debt sustain-
ability is impeded by a lack of access for their exports." James Wolfensohn,
president of the World Bank, provided a similar message.
Industrial countries are trying to help developing countries through
foreign aid, technical assistance, and debt forgiveness, especially to the
heavily indebted poor countries. What is the logic in providing foreign
aid, debt forgiveness, and other assistance, but not opening up markets in
order to help developing countries expand their exports and get out from
under these large debt burdens? Moreover, the least-developed countries
export mostly products that do not compete with OECD industries, and
the GNP of all these countries put together does not even amount to that
of a mid-sized European country. Thus, the cost of opening markets to
these countries would be small for the OECD and would help ensure that
these countries continue with their own unilateral liberalization efforts by
enabling them to obtain more benefits from liberalization.
The poor need secure access to the North and they can get this in
only two ways: through a successful WTO pursuing multilateral
nonpreferential liberalization as enforced by the MFN clause, or
through association agreements with the EU, NAFTA, or Japan. These
are, in fact, the two different uses of the term "open regionalism": con-
certed multilateralism, and open access to membership of the North-
ern clubs. The poor can both support the WTO against the menace of
Northern protectionist lobbies and at the same time pressure for the
right of access to the clubs.
116



TRADE B LO CS AND THE WO RLD TRAD ING  SYSTEM
We have proposed that the WTO should modify its rules concerning
trade blocs to create a presumptive right of association. Analogous to the
MFN clause, if association is granted to one country, there should be a
presumption that similar terms should be available to others: if Iceland
is offered reciprocal freedom from antidumping suits by the EU, then
the same option should be available to Ghana. Naturally, association is
complex, and so in practice each accession must be negotiated; regard-
less, the poor should not be denied the association rights already con-
ferred on several middle-income countries.
We have proposed a package negotiating offer by the South to the
North concerning the WTO rules governing trade blocs. The South
would offer to extend the existing rules concerning North-North trade
blocs to South-South blocs. Although this is a concession, it would
strengthen the MFN principle that is very much in the interest of the
South. In return, the South would demand an open access rule, in which
the right to equal treatment of applications for association in all trade
blocs would be enshrined.
5.7 Conclusion
T       HE CONCLUSION FIRST DEALS WTH CHANGES IN WTO RULES
to benefit developing countries, and second, with
improvement and enforcement of rules. We first recommend that:
* Industrial countries should fully open their markets to developing
country exports, particularly those from least-developing countries.
i The WTO should modify its rules concerning trade blocs to cre-
ate a presumptive right of association.
e In return, developing countries should accept the disciplines of
Articles XXIV and V for their RIAs.
Second, we recommend that:
* The WTO enforce the disciplines of Articles XXIV and V rigor-
ously in the CRTA, especially those on coverage and depth of lib-
eralization.
* The WTO define the rules more rigorously: On "substantially all
trade," we advocate 95 percent of trade after 10 years and 98 per-
cent after 15; on "other restrictive regulations of commerce," we
117



T RA E E  B LO CS
advocate inclusion of the abolition of internal barriers, such as safe-
guard actions and antidumping duties.
The WTO use the dispute settlement procedure to enforce the rights
of third countries not to face increases in protection either directly
or indirectly through the use of tools such as rules of origin.
Appendix 1: WTO Provisions on Regional
Integration Arrangements (Extracts)
Article XXIV of GAiT
4.   The contracting parties... .also recognize that the purpose of a cus-
toms union or of a free trade area should be to facilitate trade be-
tween the constituent territories and not to raise barriers to trade.
5(a). With respect to a customs union.. .the duties and other regula-
tions of commerce imposed at the institution...shall not on the
whole be higher or more restrictive than the general incidence of
the duties and regulations of commerce applicable in the constitu-
ent territories prior to the formation of such union...
(b). With respect to a free trade area... .the duties and other regulations
of commerce maintained in each of the constituent territories and
applicable at the formation of such free-trade area...shall not be
higher or more restrictive than the corresponding duties and other
regulations of commerce existing in the same constituent territo-
ries prior to the formation of the free-trade area...
(c). Any interim agreement... .shall include a plan and schedule for the
formation of such a customs union or of such a free-trade area
within a reasonable length of time.
7(a). Any contracting party deciding to enter into a customs union or a
free-trade area, shall promptly notify the CONTRACTING PARTIES and
shall make available to them such information...
8(a). A customs union shall be understood to mean the substitution of
a single customs territory for two or more customs territories, so
that: (i) duties and other restrictive regulations of commerce (ex-
cept, where necessary, those permitted under Article XI, XII, XIII,
XIV, XV and XX) are eliminated with respect to... .substantially all
the trade in products originating in such territories...
8(b). A free trade area shall be understood to mean a group of two or
more customs territories in which the duties and other restrictive
118



TRADE B LO CS AND  THE WO RLD  TRAD ING  SYSTEM
regulations of commerce (except, where necessary, those permit-
ted under Articles XI, XII, XIII, XIV, XV and XX) are eliminated
on substantially all the trade between the constituent territories in
products originating in such territories.
The Enabling Clause
1.  Notwithstanding the provisions of Article I... .contracting parties
may accord differential and more favorable treatment to develop-
ing countries, without according such treatment to other contract-
ing parties.
2(c). The provisions of paragraph 1 apply to the.. .regional or global
arrangements entered into amongst less-developed contracting
parties for the mutual reduction or elimination of tariffs and, in
accordance with criteria or conditions which may be prescribed by
the CONTRACTING PARTIES, for the mutual reduction or elimination
of nontariff measures, on products imported from one other;
The Uruguay Round Understanding on the Interpretation of Article
XXIV
2.  The evaluation.. .of the duties and other regulations of
commerce... shall.. .be based upon an overall assessment of
weighted average tariff rates and of customs duties collected... .For
this purpose, the duties and charges to be taken into consideration
shall be the applied rates of duty. It is recognized that for the pur-
pose of the overall assessment of the incidence of other regulations
of commerce for which quantification and aggregation are diffi-
cult, the examination of individual measures, regulations, prod-
ucts covered and trade flows affected may be required.
3.  The "reasonable length of time" referred to in Article XXIV 5(c)
should exceed ten years only in exceptional cases.
GATS Article V
1.  This Agreement shall not prevent any of its Members from being a
party to or entering into an agreement liberalizing trade in services
119



TRADE  B LO CS
between or among the parties to such an agreement, provided that
such an agreement:                  9
(a).  has substantial sectoral coverage  and
(b). provides for the absence or elimination of substantially all
discrimination, in the sense of Article XVII, between or
among the parties, in the sectors covered under subparagraph
(a)....
3 (a). Where developing countries are parties to an agreement of the type
referred to in paragraph 1, flexibility shall be provided for regard-
ing the conditions set out in paragraph 1, in particular to subpara-
graph (b), in accordance with the level of development of the coun-
tries concerned, both overall and in individual sectors and
subsectors...
4.   Any agreement referred to in paragraph 1 shall be designed to facili-
tate trade between the parties to the agreement and shall not in re-
spect of any Member outside the agreement raise the overall level of
barriers to trade in services within the respective sectors or subsectors
compared to the level applicable prior to such an agreement.
Notes;
1.  Multilateral trade liberalization means nearly all  5.  The interwar period offers further examples. Be-
countries re(lucing barriers on imports from nearly all part-   tween the 1920s and the mid-1930s the world trading
ners.                                                 system switched rapidly from being relatively even-handed
to a pattern of regional preferences. First Britain and France
2.  ')e have already discussed some of the different   introduced colonial preferences. In response, Germany
incentives for external tariff setting in CUs as compared    built its own system of preferences, starting with a pro-
to FTAs (chapter 4), where we argued that CUs might be    posed customs union with Austria in 1931. In 1934 the
inclined to higher tariffs than FTAs. Krugman's analysis    United States responded with the Reciprocal 'rrade Agree-
concentiates on CUs.                                  ments Act.
3.  In addition, the welfare loss due to trade diversion  6.  Frankel's model is not at all robust, but it does
for a given level of tariffs is greatest when trade is fairly   demonstrate formally the interaction between RIAs.
evenly divided between partner and nonpartner countries.
7. Levy (1995) shows theoretically how acquiring
4.  Effective RlAs among developing countries (up to   fringes of FTA partners could weaken EU and U.S. inter-
1995) are defined as including: CACM (1960-75; since   est in mutual negotiations.
1990); Andean Pact (since 1990), MERCOSUR,
UEMOA, and SACU. Individual countries affected by        8.  Oye (1992) argues that the 1930s also fit this de-
their RIA memberships include: Cameroon, Israel, Kenya,   scription. He argues that regional arrangements, such as
Mexico, and Zimbabwe.                                 U.S. bilateral arrangements, under the Reciprocal Trade
120



T RAD E  B LO CS A N D  T H E WO RL D  TRAD I N G  S YS T EM
Agreements Act were politically feasible, because they al-  the United States refused to waive their MFN rights, pre-
most guaranteed export expansion in partner markets in   venting the implementation of the Ouchy Convention, a
return for import liberalization. In this way they started to   forerunner of Benelux (Viner 1950). Similarly, negotia-
relax restrictions that were immune to multilateral efforts.   tors of the draft Multilateral Agreement on Investment
found no way of preventing some concessions on services
9. The rules of international commerce are embod-   among member from also applying to nonmembers via
ied in three main agreements: GATT, GATS, and the   the GATS MFN clause. Hence, they held back such con-
Agreement on Trade-Related Aspects of Intellectual Prop-   cessions. GATS Article V permits regional arrangements,
erty Rights. They are administered by the WTO, two of   but the Multilateral Agreement on Investment was far too
whose major tools are the Trade Policy Review Mecha-   narrowly defined to qualify.
nism and the Understanding on Dispute Settlement. The
WTO has 132 members; the major nonmember econo-        14. Together these requirements seem to impose no
mies, all of which are seeking accession, are China, Russia,   discipline on the sectors that are excluded from the RIA,
Saudi Arabia, and Taiwan (China); the other major group    although they may still be covered by the members' GATS
of current candidates are the countries of Eastern Europe   obligations.
and Central Asia.
15. The Decision on Differential and More Favorable
10. For example, if a country is harmed by another's   Treatment, Reciprocity and Fuller Participation of Devel-
breaking into its export markets, there is "properly" no    oping Countries (1979).
redress under the GATT.
16. There is also an unresolved dispute about whether
11. For example, in reviewing the treaty, the GATT    Article XXIV can be applied to an arrangement notified
executive secretary expressed the view, with which he   under the Enabling Clause, as the United States demands
thought there was no disagreement, that the incidence of   for MERCOSUR.
the common tariff was higher than that of the rates actu-
ally applied by the member states at the time of entry into  17. The EU has explicitly decided that it will not pro-
force of the Treaty of Rome (GATT Document C/M18    pose any changes to Articles XXIV and V in the next round
p.6; cited in GATT 1994, p. 750).                   of trade talks.
12. Under GATT procedure, finding a party in viola-  18. The United States and its partners still use the
tion of its obligations required unanimity. This is not true   WTO dispute settlement, but in recent years there has
of WTO.                                             been no WTO dispute between the EU and any country
with which it has a formal RIA.
13. Within the GATT there was a feeling that the ar-
ticle had influenced the structure of U.S.-Canadian and  19. This condition is understood in terms of number
U.S.-Israeli agreements (private communication). We can    of sectors, volume of trade affected, and modes of supply.
also identify cases where WTO rules, or their equivalent,   In order to meet this condition, agreements should not
have prevented RIAs. For example, in 1932 Britain and    provide for the apriori exclusion of any mode of supply.
121






CHAPT ER 6
Conclusion: Tell Me the Truth
about Trade Blocs
Trade blocs are political...
D        URING THE 1990S TRADE BLOCS PROLIFERATED. BY 1999 MORE
regional agreements had been notified to the WTO
than it had countries as members. Evidently, there were
powerful forces driving this process. These forces were political: trade
blocs have economic effects, but that is not why they are established.
The main political objective has probably been enhanced security.
International trade reduces the risks of military conflict between countries
and so there might appear to be a reasonable case for preferentially
promoting trade relations between neighbors. Unfortunately, whereas
international trade is normally mutually beneficial, preferentially induced
trade can sometimes create powerful transfers so that one partner gains
at the expense of another. There are numerous historical examples of
such redistributions causing conflict because they are seen as unfair.
Hence, even when the objective of a regional arrangement is purely
political, the economic consequences need to be understood.
Another motivation has been to enhance bargaining power. OPEC
demonstrated that it was possible under some conditions to improve the
terms of trade by collective action. The scope for OPEC-type trade agree-
ments proved very limited. However, small developing countries may
still find that by negotiating collectively with industrial countries on
trade issues they would gain, not by increasing their power, but by en-
hancing their ability to get noticed in bargaining rounds, enabling them
to conclude more reciprocal deals.
Regional cooperation on trade issues may help countries to cooperate
on other issues. Small neighboring countries have plenty of scope for
cooperation. Some infrastructure, such as power, telecoms, and railways,
123



TRAID E B LO CS
may be better provided regionally than nationally. If the tax treatment of
multinationals is harmonized, countries can increase their bargaining
power and avoid a race to the bottom. Thus, even if regional coopera-
tion starts with trade issues, it should not stop there. The main benefit
from cooperation on trade issues may be the development of a habit of
trust and cooperation between neighboring governments that can then
be extended to issues on which there is more scope for mutual gain.
Many developing and transition economies are in the process of re-
forming their economic policies and their governance systems. Regional
cooperation has sometimes proved useful as a commitment rnechanism,
locking in the change. The most spectacular examples of this have prob-
ably been the North-South cooperation arrangements: Mexico gaining
credibility through NAFTA, Eastern Europe through accession agree-
ments with the EU, and North Africa through association agreements
with the EU. Some South-South agreements have also acted as commit-
ment mechanisms, notably MERCOSUR.
Security, bargaining power, cooperation, and lock-in are probably the
main political motors for regional integration. Sometimes these motives
receive a veneer of economic rationalization. Frequently there is an ap-
peal to the benefits flowing from scale economies. Such "soundbites" of
economic analysis are not usually wrong, but they are so incomplete and
lopsided as to be seriously misleading: weapons casually hurled by advo-
cates who have already decided their position, rather than serious at-
tempts to understand the economic consequences of choices.
The politically feasible alternative to a costly
trade bloc is probably a better-designed bloc...
I N THIS REPORT WE HAVE TREATED THE POLITICAL FORCES FOR
the  creation  of  regional  arrangements  as  largely
unstoppable and have focused on choices of design. For example,
only if regional schemes adopt common external tariffs, such as in the
EU, can they bargain collectively in world trade rounds. Yet a common
tariff precludes unilateral liberalization, and also prevents individual
developing countries from joining their appropriate product-based
groupings in global negotiations. Thus, schemes that do not adopt a
common external tariff probably have a lower opportunity cost in terms
of other trade policy options. It is therefore more reasonable to treat the
effects of such schemes as being additional to whatever unilateral and
124



C O N C LU S I O N : TELL M E TH E TRUTH AB O UT TRA D E B LO C S
global liberalization might be underway. This has been our approach in
analyzing the effects of trade blocs.
So how does a trade bloc affect people,
especially the poorest?
IN ORDER TO MAKE SENSE OF DESIGN CHOICES, THE BASIC
economic effects of trade blocs need to be understood.
The overall policy message is that the effects are very sensitive to
the choices of design. It matters enormously who else is in the bloc and
how preferences are implemented. We distinguish between two broad
types of effect: the first is competition and scale, and the second is trade
and location.
Competition and Scale Effects
The simple "soundbite" image of the benefits of a trade bloc is per-
haps the scale benefits of having a single big factory serving the regional
market. This will almost never be a good idea. Such a factory would be
a protected monopoly, and such monopolies are usually inefficient and
exploitative. Competition is a vital discipline on private behavior, yet
there is obviously a tradeoff between the number of competitors in a
market and the average size of factory. More competition means smaller
factories, and so within any given market there is a tradeoff between
competition and scale. Regional integration enlarges the market and so
enables both more competition and a larger average scale. Instead of
having two national markets, each with three firms producing 100 units,
there can be four firms in the regional market, each of which produces
150 units. Both the increase in competition and the increase in scale
will lower prices, as we have shown happened in MERCOSUR; so re-
gional integration will have been beneficial, but in order to reap these
gains, two of the six firms will have closed. Hence, in order for regional
integration to secure the gains of competition and scale, the least effi-
cient firms must be allowed to exit. A successful regional integration is
an omelet that cannot be made without breaking eggs. Furthermore,
removing tariffs is likely to be insufficient to achieve these gains that
depend upon national markets becoming integrated into a single re-
gional market. This will require many other supporting policies of "deep
125



TRADE BLOCS
integration" to harmonize product standards and ensure that all firms
have real penetration in all the nations within the region.
A trade bloc that succeeds in reaping these competition and scale ef-
fects will not only lower the prices of manufactures produced within the
region. As a result, importers will be forced to lower their prices so the bloc
will improve its terms of trade. If developing countries want to use trade
blocs to improve their terms of trade, the most pertinent model is not
OPEC but MERCOSUR. The policy instrument is thus not a collective
increase in trade taxes, but a collective increase in competition.
Trade and Location Effects
Competition and scale effects accrue to the region as a whole, but trade
and location effects are predominandy about transfers between one part
of the region and another. The key trade effect is that money, which prior
to the trade bloc accrued to the government as tariff revenue, will now
accrue to firms in the partner country. The government loses tariff rev-
enue and the country as a whole loses income. This effect is known as
trade diversion. We have looked to see how substantial this effect is in
seven recent regional arrangements by modeling the effect on trade be-
tween the countries in the bloc and the rest of the world. In four of the
seven there was no problem, but in three the problem was large enough to
be visible. Hence, diversion is neither so common as to be general, nor so
unusual as to be dismissable. The analysis has to be done bloc by bloc. In
some circumstances the loss of revenue will be serious, notably where tar-
iffs are high and tariff revenue is a substantial share of total government
revenue. For example, in a small, poor country such as Burkina Faso, re-
gional integration will involve a large diversion from government revenue
to manufacturing firms in C6te d'Ivoire and Senegal. A price that Burkina
Faso might have to pay for the political drive to regionalism might thus be
fewer children in primary education. Recent research suggests that there
niay be a further hidden cost to diversion. One by-product of trade is
knowledge: firms learn from their trading partners. Evidence shows that
trade has more knowledge benefits the larger is the stock of knowledge of
the trading partner, with the stock of knowledge measured by the accu-
mulated investment in research and development. Hence, if a poor South-
ern country diverts its trade from a Northern country with a large knowl-
edge stock, to another Southern country with a much smaller knowledge
stock, it will reduce its learning. Since within a South-South trade bloc it
126



CONCLUS IO N : TELL M E TH E TRUTH  ABO UT TRAD E BLOCS
is the poorest countries that experience the most diversion, they are the
ones liable to suffer the largest reduction in knowledge transfer.
The formation of a trade bloc will cause economic activities to shift
location. Potentially, this can create convergence or divergence between
the members of the bloc. One contribution of this report has been the
discovery that the conventional forces of comparative advantage have a
disturbing implication for South-South trade blocs. We show that com-
parative advantage produces convergence in North-North blocs, but di-
vergence in South-South blocs. We show that this is not just theory. In
the EU the poorer Northern countries, such as Ireland, Portugal, and
Spain, have caught up with the richer countries: there are dramatic signs
of convergence. By contrast, in CACM and the Economic Community
of West Africa there are symptoms of divergence: the richer Southern
countries have substantially gained market share at the expense of the
poorer. Comparative advantage works in this way in these trade blocs by
advantaging the middle-income countries. A North-North bloc discrimi-
nates against the South and so helps those countries within the bloc that
are the closest competitors with the South, namely the lowest-income
countries in the bloc. A South-South bloc discriminates against the North
and so helps those countries within the bloc that are the closest competi-
tors with the North, namely the highest-income countries in the bloc.
Thus, the same force produces convergence in Northern blocs and di-
vergence in Southern blocs.
A further force for divergence within Southern blocs is industrial ag-
glomeration. Firms within an industry gain from clustering together,
and when freed from trade barriers will choose to do so. Trade blocs will
thus always increase agglomeration within each industry: if they fail to
do so it is because they have failed to remove the real barriers to trade.
Such forces may or may not cause overall industrial agglomeration. For
example, in the United States, although each industry is highly agglom-
erated, there is little overall industrial agglomeration because different
industries cluster in different cities. The key issue is whether the big
gains from agglomeration are specific to each industry or accrue to in-
dustry in general. These processes have not yet been very thoroughly
researched empirically, but what seems probable is that at an early stage
of industrialization the main benefits of clustering accrue for industry as
a whole-for example, the provision of good infrastructure-whereas at
an advanced stage the main benefits are industry-specific-for example,
a skilled work force. Unfortunately, as with the forces of comparative
advantage, this will also tend to produce convergence within Northern
127



TRADE BLOCS
blocs and divergence within Southern blocs. These forces for divergence
within Southern trade blocs have evident and serious implications for
poverty. They may also make the blocs politically unviable and even
become a cause of conflict.
I'm the minister of trade. What bloc design
should I choose?
T  t AKING THE POLITICAL IMPETUS FOR THE FORMATION OF
trade blocs as a given, it is evident that the poorest
countries may find membership of the conventional South-
South blocs quite problematic. While the soundbite regional economics
of scale economies might seem to offer most to the poorest, smallest
economies, a more serious analysis reveals much scope for loss: revenue
diversion, reduced knowledge transfer, and divergence from richer
partners. We now review some of the design choices that can determine
whether a trade bloc is economically advantageous to all of its members,
or is liable to be a source of contention.
Which countries should I take as partners?
T       ABLE 4.1 SUMMARIZES OUR ASSESSMENT OF PARTNER
suitability.  For  example,  a  Central  European
transition economy may look to association with the EU primarily
for the political benefits of security and policy lock-in. These political
benefits are likely to be so large that the country would probably want to
join the trade bloc even if the economic effects were on balance highly
negative. In fact, they are likely to be positive. There will probably be some
losses from revenue diversion, but the scale and competition, and trade
and location effects are likely to be positive. Thus, the political impetus
also happens to make economic sense. A trade bloc between two large,
middle-income countries can also have very substantial political benefits.
Regional security may be enhanced, there may be policy lock-in, the
increased trust may facilitate other types of regional cooperation, and
negotiating power may increase. The economics are less clear. The increased
competition is likely to improve the terms of trade, and this can be a large
benefit. If manufacturing is already well established, although there will
be powerful and beneficial agglomeration effects within each industry, the
128



C O N CLU S I O N: T ELL M E T H E T RUTH AB O UT TRAD E B LO C S
bloc may avoid significant overall industrial agglomeration. Offsetting this,
there may be some revenue diversion. Thus, the strong political impetus
may not deliver commensurate economic benefits, but there is perhaps
little danger of large net costs.
The most problematic blocs are evidently those between two small,
poor countries, one of which is significantly poorer than the other. There
may still be some political gains: the bloc may find it easier to get no-
ticed than the countries individually, and if the experience of trade co-
operation builds up trust, it may facilitate cooperation on other issues.
However, there may also be political costs as the unintended economic
transfers generate frictions. The economic effects look worrying: as dis-
cussed above, the poorer country stands to lose through several distinct
processes. What else might such poor countries do? One option is for a
South-South bloc to negotiate an associate agreement with a Northern
bloc. Such a negotiating opportunity is currently available from the EU
through the new Lome agreement, and might also become available from
the United States. Politically, membership of a North-South bloc may
bring benefits of policy lock-in, as in Central Europe. Economically, it
offers enhanced knowledge gains, and should at least mitigate the prob-
lem of divergence within the Southern bloc. While the forces of indus-
trial agglomeration would still favor the more developed Southern part-
ner, the forces of comparative advantage would favor the poorest most.
Hence, small, poor countries should probably aim to rechannel the po-
litical impetus for trade blocs from expanding South-South blocs into
South-North blocs. The less-poor members of South-South blocs might
also benefit from such a change of strategy. Although they would lose
protection in the tiny markets of the poorest countries, they would be
the most likely beneficiaries of investment in industries exporting to the
new Northern market.
How many blocs should I join?
M        ORE IS FINE AS LONG AS THEY ARE NOT INCOMPATIBLE,
although the resources spent in negotiating and
administering them may be better used on other issues. At
present some countries have signed multiple agreements that are not legally
compatible. This is not merely bad law, it gives rise to investor uncertainty:
it is simply unclear which tariffs will actually end up being applied. And
even when compatible, a proliferation of agreements may leave investors
129



TRADE B LO CS
confused. The existence of incompatible agreements is a classic example of
political dreams colliding with practical decisions. In such cases the political
impetus to regional agreements needs to be rechanneled.
How much preference should I give?
B      IG PREFERENCES CAUSE INDUSTRY TO AGGLOMERATE IN A
single location within a South-South trade bloc. This
implies strong transfers within the bloc with the poorer members
losing out. It is therefore in the interest of the poorer members of a
South-South bloc to set their external tariff at a moderate level, and if
the bloc has a common tariff, to insist that it be fairly low. Another
reason for low external tariffs is that big preferences increase revenue
diversion. Also, since protection in poorer members is typically higher
than in richer ones, the poorer ones will lose more from opening up to
the rich ones than they gain from free access to them. This can be resolved
by a reduction in the poor member's external tariff.
Should I press for a common external tariff?
C 1 OMMON EXTERNALTARIFFS HAVE ONE BIG ADVANTAGE: THEY
avoid  the  need  for  "rules  of  origin,"  the
enforcement of which creates a large amount of bureaucracy
and scope for fraud. However, they also have serious disadvantages.
Neighboring countries differ as to their need for tariff revenue, and hence
as to the height of tariff that is appropriate. They also differ in their
chosen pace of trade liberalization and in their preferences and
opportunities for tariff bargaining. Finally, the common pool of revenue
has to be divided on some basis, and this may strain political cooperation.
In practice, governments usually opt out of a common external tariff
through exemptions, even if they sign up in principle.
How deep should I take liberalization?
T       HE BIG GAINS FROM TRADE BLOCS COME FROM INTEGRATING
markets.  Removing  tariffs  but  leaving   other
impediments will inflict all the costs of revenue diversion without
130



C O N CLU S I O N: TELL M E TH E TRUTH AB O UT TRAD E B LO C S
any of the compensating benefits of competition and scale. Thus deeper
is better. Potentially, agreements can preclude the use of antidumping
suits, which is part of the agreement between members of the EU. Since
the EU has already extended this benefit to Iceland, there would seem to
be in principle no obstacle to its being included in prospective South-
North blocs involving the EU. Agreements can also cover border
procedures where there is often large scope for illicit protection that
undermines the bloc. Finally, as in the EU, they can cover product
standards. Rather than agree on common standards, which is slow and
may be costly for the poorer members, the most practical step may be
mutual recognition: If a product can be sold in one country, it can be
sold anywhere in the bloc.
How wide should I let negotiations range?
T g : HE FOCUS OF SOUTHERN TRADE BLOCS IS PRIMARILY ON
trade in manufactures. However, many of the big
gains to liberalizing trade are to be found in services. Services
are often less exposed to competition, and a high-cost service sector can
handicap all the other sectors of the economy for which it supplies inputs.
It is also important to extend cooperation beyond trade. For example, in
South-South blocs that fail to harmonize the taxation of foreign
investment, the creation of the trade bloc weakens the bargaining power
of each government relative to the investor. The investor can now serve
the entire regional market by locating in that country that offers the
lowest taxation, and so the trade bloc encourages a tax race to the bottom.
I went to Seattle. How can I use the wrO more
effectively?
F     INALLY, THE REPORT HAS CONSIDERED TRADE BLOCS IN THE
context of the WTO. Developing countries have much
to gain from continued multilateral nonpreferential liberalization
as enforced by the MFN clause. MFN strengthens the weak by limiting
the power of the strong to cut deals with each other that exclude the
weak. The biggest exception to MFN that the GATT and WTO have
permitted has been the EU, which has fully liberalized trade among its
members without extending the same opportunities to other nations.
131



TRADE B LO C S
Developing countries have an interest in protecting the MFN principle
from further erosion, but they also have an interest in gaining access to
the trade blocs that the North has already constructed. The strategy of
imitating the North by constructing South-South blocs is unlikely to be
beneficial for the poorest. South-South blocs cannot do for the South
what North-North blocs did for the North. This is not because of a lack
of political will, it is because the same economic forces will produce
radically different outcomes. South-South blocs offer little to the poorest
countries and may even harm them.
The poor need secure access to the North, and they can get this in
only two ways: through a successful WTO, or through association agree-
ments with the EU, Japan, or the United States. These are, in fact, the
two different uses of the term "open regionalism": concerted
multilateralism, and open access to membership of the Northern clubs.
Fortunately, these are not alternatives. The poor can support the WTO
against the menace of Northern protectionist lobbies at the same time
that they pressure for the right of access to the clubs. Since this report is
about trade blocs, we have focused upon the latter. We have proposed
that the WTO modify its rules concerning trade blocs to create a pre-
sumptive right of association. Analogous to the MFN clause, if associa-
tion is granted to one country, there should be a presumption that simi-
lar terms should be available to others. If Iceland is offered reciprocal
freedom from antidumping suits by the EU, then the same option should
be available to Ghana. Naturally, association is complex, and so, in prac-
tice, each accession must be negotiated. But the poor should not be
denied the association rights already conferred by both the United States
and the EU on several middle-income countries. The voice of the poor
is not loud in global trade forums and is easily hijacked by Northern
special interests. We have proposed a package-negotiating offer by the
South to the North concerning the WTO rules governing trade blocs.
The South would offer to extend the existing rules concerning North-
North trade blocs to South-South blocs. Although this is a concession, it
would strengthen the MFN principle, which is very much in the inter-
ests of the South. In return, the South would demand an open access
rule, in which the right to equal treatment of applications for associa-
tion in all trade blocs would be enshrined.
132



Bibliography
Abouyoub, Hasan. 1998. Personal communication        . 1995. "A Domino Theory of Regionalism."
at the World Bank conference on "Regionalism   In Richard E. Baldwin, P. Haaparanta, and J.
and Development." Geneva (May).                Kiander, eds. Expanding Membership in the Euro-
pean Union, Cambridge, United Kingdom: Cam-
Adams, Charles. 1993. For Good and Evil: The Im-  bridge University Press.
pact of Taxes on the Course of Civilization. Lanham,
Maryland: Madison Books.                     Baldwin, Richard E., and Anthony Venables. 1997.
"International Economic Integration." In G.
Ades, Alberto E and Edward L. Glaeser. 1995. "Trade  Grossman and K. Rogoff, eds., Handbook ofInter-
and Circuses: Explaining Urban Giants." Quar-   nationalEconomics, volume 3. Amsterdam: North
terlyjournalofEconomics 11O(February): 195-227.  Holland.
Andriamananjara, Soamiely, and Maurice Schiff.   Baldwin, Richard E., R. Forslid, and J. Haaland.
Forthcoming. "Regional Groupings among          1996. "Investment Creation and Investment Di-
Microstates." Review of International Economics.  version: A Simulation Study of the EU's Single
Market Programme." World Economy 19(6): 635-
Anwar, Dewi E 1994. Indonesia in ASEA,N: Foreign  59.
Policy and Regionalism. New York: St. Martin's
Press.                                       Barry, Frank, and John Bradley. 1997. "FDI and
Trade: The Irish Host-Country Experience." Eco-
Bagwell, Kyle, and R.W Staiger. 1998. "Will Prefer-  nomic Journal 107(November): 1798-1811.
ential Agreements Undermine the Multilateral
Trading System?" Economic Journal 108(July):   Basevi, G., E Delbono, and M. Mariotti. 1995. "Bar-
1162-82.                                       gaining with a Composite Player: An Application
to the Uruguay Round of GATT Negotiations."
Baldwin, Richard E. 1989. "The Growth Effects of  Journal of International Comparative Economics
1992." Economic Policy 9: 247-81.              3:161-74.
133



B I B L I 0 G RA P H Y
Bayoumi, Tamim, and Barry Eichengreen. 1997. "Is   Blomstrom, Magnus, and Ari Kokko. 1997a. "Re-
Regionalism Simply a Diversion: Evidence from   gional Integration and Foreign Direct Investment:
the Evfolution of the EC and EFTA." In T Ito and  A Conceptual Framework andThree Cases." Policy
Anne 0. Krueger, eds., Regionalism Versus Multi-  Research Working Paper no. 1750. World Bank,
lateral Trade Arrangements. National Bureau of  Washington, D.C.
Economic Research. East Asia Seminar on Eco-
nomics, vol. 6, Chicago: University of Chicago     . 1997b. "How Foreign Investment Affects
Press.                                          Host Countries." Policy Research Working Paper
no. 1745. World Bank, Washington, D.C.
Ben-David, Dan. 1993. "Equalizing Exchange: Trade
Liberalization and Income Convergence." Quar-   Bond, Eric W, and Constantinos Syropoulos. 1996a.
terlyjournal of Economics 108(3): 653-79.       "The Size of Trading Blocs: Market IPower and
World Welfare Effects." Journal of International
Bergsten, C. Fred. 1996. "Globalizing Free Trade."  Economics 40(May): 411-37.
In Jeffrey J. Schott, ed. The World Trading Sys-
tem: C(hallenges Ahead. Washington, D.C.: Insti-   . 1996b. "Trading Blocs and the Sustainabil-
tute for International Economics.               ity of Inter-regional Cooperation." In M.
Canzoneri, W. Ethier, and V. Grilli, eds., The New
. 1997. "APEC in 1997: Prospects and Pos-    Transatlantic Economy, Cambridge, United King-
sible Strategies." Special report 9, pp. 3-17. Wash-  dom: Cambridge University Press.
ingtorn, D.C. : Institute for International Econom-
ics.                                          Bouzas, Roberto. 1997. "MERCOSUR's .Economic
Agenda: Short and Medium-term Policy Chal-
Bhagwati, Jagdish. 1993. "Regionalism  and         lenges." Integration and Trade 10:64-5.
Multilateralism: An Overview." In Jaime de Melo
and A. Panagariya, eds., New Dimensions in Re-   Catinat, M., and A. Italianer. 1998. "Completing the
gional Integration. London: Centre for Economic  Internal Market: Primary Microeconomic Effects
Policy Research.                                and their Implementation in Macroeconometric
Models." Commission of the European Commu-
Bhagwati, Jagdish, and Arvind Panagariya. 1996.    nities.
"Preferential Trading Areas and Multilateralism-
Strangers, Friends, or Foes?" In J. Bhagwati and   Chang, Won, and Maurice Schiff. 1999. "Regional
A. Pariagariya, eds., The Economics of Preferential  Integration and the Price Effects of Contestability."
Trade Agreements. Washington, D.C.: American    World Bank Development Economics Research
Enterprise Institute Press.                     Group, International Trade Unit, Washington,
D.C. Processed.
Bilal, Sanousi. 1998. "Why Regionalism May Increase
the Demand forTrade Protection."JournalofEco-   Chang, Won, and L. Alan Winters. 1999. "How Re-
nomic Integration 13(1): 30-61.                 gional Blocs Affect Excluded Countries: The Price
134



B I B L I O G RA P H Y
Effects of MERCOSUR." Discussion Paper Se-   Dixit, Avinash. 1988. "Antidumping  and
ries no. 2179. Centre for Economic Policy Re-  Countervailing Duties under Oligopoly." European
search, London.                               Economic Review 32:55-68.
Coe, David T, and Elhanan Helpman. 1995. "Inter-   Djankov, Simeon, and Bernard Hoekman. 1997. "Ef-
national R&D Spillovers." European Economic    fective Protection and Investment Incentives in
Review 39(5): 859-87.                         Egypt andJordan: Implications of Free Trade with
Europe." World Development 25:281-91.
Coe, David T., Elhanan Helpman, and Alexander W.
Hoffmaister. 1997. "North-South  R&D             . 1998. "Avenues ofTechnologyTransfer: For-
Spillovers." Economic Journal 107: 134-49.    eign Investment and Productivity Change in the
Czech Republic." Working Paper 16/98.
Collier, Paul, and J.W Gunning. 1995. "Trade Policy  Fondazione Eni Enrico Mattei Note di Lavoro.
and Regional Integration: Implications for the
Relations between Europe and Africa." World   Engel, Charles, and John H. Rogers. 1996. "How
Economy 18:387-410.                           Wide is the Border?" American Economic Review
86(5): 1112-25.
Commission of the European Communities. 1995.
"Proposal for a Decision of the Council and the   Feenstra, Robert, and Gordon Hanson. 1997. "For-
Commission on the Conclusion of a Euro-Medi-  eign Direct Investment and Relative Wages: Evi-
terranean Agreement Establishing an Association  dence from Mexico's Maquiladoras." Journal of
between the European Communities and Their    International Economics 42: 371-93.
Member States and the Republic of Tunisia."
COM(95)235 final (May 31).                  Fernandez, Raquel, and Jonathan Portes. 1998. "Re-
turns to Regionalism: An Analysis on Nontradi-
De laTorre, Augusto, and M. Kelley. 1992. Regional    tional Gains from Regional Trade Agreements."
Trade Arrangements. Washington D.C.: Interna-  World Bank Economic Review 12(2):197-220.
tional Monetary Fund.
Finger, J. Michael. 1993. Antidumping: How it Works
De Rosa, Dean A. 1995. "Regional Trading Arrange-  and Who Gets Hurt. Ann Arbor, Michigan: Uni-
ments among Developing Countries: The ASEAN   versity of Michigan Press.
Example." Research Report no. 103. International
Food Policy Research Institute, Washington D.C.   Flores, R. 1997. "The Gains from MERCOSUL: A
General Equilibrium, Imperfect Competition
Dickens, William, and Lawrence Katz. 1987. "Inter-  Evaluation." Journal of Policy Modeling 19: 1-18.
Industry Wage Differences and Industry Charac-
teristics." In Kevin Lang and Jonathan Leonard,   Foroutan, Faezeh. 1993. "Regional Integration in
eds., Unemployment and the Structure of Labor  Sub-Saharan Africa: Past Experience and Future
Markets. New York: Blackwell.                  Prospects." In Jaime de Melo and Arvind
135



B I B L I O G RA P H Y
Pana,ariya, eds., New Dimensions in Regional In-   Fukase, Emiko, and William Martin. 1999. "Evalu-
tegrainion. London: Center for Economic Policy  ating the Implications of Cambodia's Accession to
Research.                                       the ASEAN Free Trade Area: A General Equilib-
rium Model (CGE) Approach." World Bank,
. 1996. "Turkey, 1976-85: Foreign Trade, In-  Washington, D.C. Processed.
dustrial Productivity, and Competition." In Mark
J. Roberts and James R. Tybout, eds., Industrial   Gatsios, Konstantine, and Larry Karp. 1991. "Del-
Evolution in Developing Countries: Micro Patters of  egation Games in Customs Unions." Review of
Turnover, Productivity, and Market Structure. Ox-  Economic Studies 58(2): 391-97.
ford, United Kingdom: Oxford University Press.
-____ 1995. "Delegation in a General Equilibrium
1. 1998. "Does Membership of a Regional Pref-  Model of Customs Unions." European Economic
erential Trade Arrangement Make a Country More  Review 39(2): 319-33.
or Less Protectionist?" The World Economy 21:
305-36.                                       GATT (General Agreement on Tariffs and Trade).
1994. "Report on Regional Integration." Secre-
Francois, Joseph. 1997. "External Bindings and the  tariat, Geneva.
Credibility of Reform." In A. Galal and B.
Hoekman, eds., Regional Partners in Global Mar-   Gordon, H., S. Rimmer, and S. Arrowsmith. 1998.
kets. London: Center for Economic Policy Research.  "The Economic Impact of the EU Regime on
Public Procurement." World Economy 21:159-88.
Francois, J., and C. Shiells. 1994. Modeling Trade
Policy: Applied General Equilibrium Assessments of  Grossman, G., and E. Helpman. 1994. "Protection
North American Free Trade. Cambridge, United    for Sale." American Economic Review 84(4): 833-
Kingdom: Cambridge University Press.             50.
Frankel, Jeffrey. 1997. Regional Trading Blocs in the  .1995. "The Politics of Free-Trade Agreements."
Work/Economic System. Washington, D.C.: Insti-  American Economic Review 85(4): 667-90.
tute for ]International Economics.
Gupta, Anju, and Maurice Schiff. 1997. "Outsiders
Fujita, Masahisa, Paul Krugman, and Anthony        and Regional Trade Agreements among Small
Venables. 1999. The Spatial Economy: Cities, Re-  Countries: The Case of Regional Markets." Policy
gions, and International Trade. Cambridge, Mas-  Research Working Paper no. 1847. World Bank,
sachusett:s: MIT Press.                         Washington, D.C.
Fukase, Emiko, and L. Alan Winters. 1999. "Pos-   Haddad, Mona. 1993. "The Effect of Trade Liberal-
sible Dynamic Benefits of ASEAN/AFTA Acces-     ization on Multi-factor Productivity: The Case of
sion for the New Member Countries." World       Morocco." Discussion Paper Series no.4, World
Bank, Washington, D.C. Processed.               Bank, Washington, D.C.
136



B I B L I O G RA P H Y
Haddad, Mona, and Ann Harrison. 1993. "Are There  Round." In William Martin and L. Alan Winters,
Positive Spillovers from Direct Foreign Invest-  eds., The Uruguay Roundand the Developing Coun-
ment? Evidence from Panel Data for Morocco."   tries. Cambridge, United Kingdom: Cambridge
Journal ofDevelopment Economics 42: 51-74.     University Press.
Haddad, Mona, Jaime de Melo, and Brendan   Helliwell, J. 1997. "How Much Do National Bor-
Horton. 1996. "Morocco, 1984-89: Trade Lib-    ders Matter?" Series on Integrating National
eralization, Exports and Industrial Performance."  Economies. The Brookings Institution, Washing-
In Mark J. Roberts, and James R. Tybout, eds.,    ton, D.C.
Industrial Evolution in Developing Countries: Mi-
cro Patters of Turnover, Productivity, and Market   Herin, Jan. 1986. "Rules of Origin and Differences
Structure. Oxford, United Kingdom: Oxford      between Tariff Levels in EFTA and the EC." Oc-
University Press.                              casional Paper no. 13. European Free Trade Asso-
ciation, Secretariat, Geneva.
Hansen, Lyle M. 1969. "The Economy of Burundi."
Report AE-1 a. International Bank for Reconstruc-   Hillman, Arye. 1989. "The Political Economy of Pro-
tion and Development, Washington, D.C.         tection." Fundamentals of Pure and Applied Eco-
nomics. Switzerland, London, New York: Harmead
Hanson, Gordon. 1996. "Economic Integration,     Academic.
Intraindustry Trade and Frontier Regions." Euro-
pean Economic Review 40: 941-49.             Hirschman, Albert 0. 1958. The Strategy ofEconomic
Development. New Haven, Connecticut: Yale Uni-
Harrison, Ann. 1994. "Productivity, Imperfect Com-  versity Press.
petition, and Trade Reform." Journal of Interna-
tional Economics 36: 5 3-73.                      . 1981. Essays in Trespassing: Economics to Poli-
tics and Beyond. Cambridge, United Kingdom:
. 1996. "Openness and Growth: A Time-Se-    Cambridge University Press.
ries, Cross-Country Analysis for Developing
Countries." Journal of Development Economics   Hoekman, Bernard. 1998. "PreferentialTradeAgree-
48(2): 419-47.                                 ments." In R. Lawrence, ed., Brookings Trade Fo-
rum 1998. Washington, D.C.: The Brookings In-
Harrison, Glenn, Thomas Rutherford, and David    stitution.
Tarr. 1994. "Product Standards, Imperfect Com-
petition and the Completion of the Market in the  . 1999. "Rents, Red Tape, and Regionalism:
EC." Policy Research Working Paper no. 1293.   Economic Effects of Deeper Integration." In B.
World Bank, Washington, D.C.                   Hoekman, and J. Zarrouk, eds., Catching up with
the Competition: Trade Opportunities and Challenges
Hathaway, Dale E., and Merlinda D. Ingco. 1996.  forArab Countries. Ann Arbor, Michigan: Univer-
"Agricultural Liberalization and the Uruguay   sity of Michigan Press.
137



B I B L [0  G RA P H Y
Hoekman, Bernard, and Michael Leidy. 1993. "Holes   IDB (Inter-American Development Bank). 1998.
and Loopholes in Regional Integration Agree-   "Integration and Trade in the Americas." Periodic
ments." In Kym  Anderson and Richard           Note (August).
Blackhurst, eds., Regional Integration. London:
Harvester-Wheatsheaf.                       Irwin, Douglas. 1993. "Multilateral and Bilateral
Trade Policies in the World Trading System: An
Hoekman, Bernard, and Carlos Primo Braga. 1997.  Historical Perspective." in Jaime de Melo and
"Protection and Trade in Services: A Survey." Open  Arvind Panagariya, eds., New Dimensions in Re-
Economies Review 8: 285-308.                   gional Integration. London: Centre for Economic
Policy Research.
Hoekman, Bernard, and Denise Konan. 1999. "Deep
Integration, Nondiscrimination, and Euro-Medi-    . 1996. Against the Tide: An Intellectual His-
terranean Free Trade." Discussion Paper no. 2095.  tory ofFree Trade. Princeton, New Jersey: Princeton
Centre for Economic Policy Research, London.   University Press.
Hoekman, Bernard, and Michel Kostecki. 1995.   Islam, Nurul. 1981. ForeignTrade andEconomic Con-
"The Political Economy of the World Trading    trols in Development: The Case of United Paki-
System: From GATT to WTO." Oxford and New      stan. New Haven and London: Yale University Press.
York: Oxford University Press.
Jacquemin, A., and Andre Sapir. 1991. "Competi-
Hoekman, Bernard, and Petros C. Mavroidis, eds.   tion and Imports in the European Market." In L.
1997. Law and Policy in Public Purchasing: The  Alan Winters and Anthony J. Venables, eds., Eu-
WTOAgreementon GovernmentProcurement. Ann      ropean Integration: Trade and Industry. Cambridge,
Arbor, Michigan: University of Michigan Press.  New York, Melbourne: Cambridge University
Press.
Hoekman, Bernard, and Pierre Sauve. 1994. Regional
and Multilateral Liberalization ofTrade in Services:   Kahler, Miles. 1995. International Institutions and the
Complements or Substitutes?" Journalof Common  PoliticalEconomy ofIntegration. Washington, D.C.:
Market Studies 32: 283-317.                    The Brookings Institution.
Horn, Hernrik, Harold Lang, and Stefan Lundgren.   Kant, Emanuel. 1795. Perpetual Peace: A Philosophi-
1995. "Managerial Effort Incentives, X-Ineffl-  cal Essay. Bristol, United Kingdom: 'rhoemmes
ciency and International Trade." European Eco-  Press.
nomic Review 39: 117-38.
Kechichian, Joseph. 1985. "The Gulf Cooperation
Hunter, Linda, James R. Markusen, and Thomas E    Council: The Search for Security." Third World
Rutherford. 1992. "U.S.-Mexico Free Trade and  Quarterly 7: 853-81.
the North American Auto Industry: Effects on the
Spatial Organization of Production of Finished   Keller, Wolfgang. 1998. "Are International R&D
Autos." World Economy 15(1): 65-81.            Spillovers Trade-Related? Analyzing Spillovers
138



B I B L I O G RA P H Y
among Randomly Matched Trade Partners." Eu-         . 1996. Regionalism, Multilateralism, and
ropean Economic Review42(8): 1469-81.           Deeper Integration. Washington, D.C.: The
Brookings Institution.
Krishna, Pravin, and Devashish Mitra. 1997. "Trade
Liberalization, Market Discipline, and Productiv-   Levinsohn, James. 1993. "Testing the Imports-as-
ity Growth: New Evidence from India." Brown     Market-Discipline Hypothesis." Journal of Inter-
University, Providence, Rhode Island. Processed.  national Economics 35: 1-22.
Krueger, Anne 0. 1997. "Free Trade Agreements   Levy, Philip. 1995. "Free Trade Agreements and In-
Versus Customs Union." Journal of Development   ter-Bloc Tariffs." Yale University, New Haven,
Economics 54: 169-87.                           Connecticut. Processed.
Krugman, Paul. 1991a. "Is Bilateralism Bad?" In E.   Madani, Dorsati. 1999. "South-South Regional In-
Helpman and A. Razin, eds., International Trade  tegration and Industrial Growth: The Case of the
and Trade Policy. Cambridge, Massachusetts: MIT  Andean Pact." World Bank, Washington, D.C.
Press.                                          Processed.
___. 1991 b. "The Move towards Free Trade   Magee, Stephen P, and Hak-Loh Lee. 1997. "Tariff Cre-
Zones." In Policy Implications of Trade and Cur-  ation and Tariff Diversion in a Customs Union: The
rency Zones. A symposium sponsored by the Fed-  Endogenous External Tariff of the EEC 1968-83."
eral Reserve Bank of Kansas City, Jackson Hole,  Fondazione Eni Enrico Mattei Note di Lavoro38(97).
Wyoming.
Marshall, Alfred. 1920. Principles ofEconomics. Lon-
. 1993. "Regionalism Versus Multilateralism:  don: MacMillan.
Analytical Notes." In Jaime de Melo and Arvind
Panagariya, eds., New Dimensions in Regional In-   Mattoo, Aaditya, and Petros Mavroidis. 1995. "The
tegration. Cambridge, New York, Melbourne:      EC-Japan Consensus on Cars: Interactions be-
Cambridge University Press.                     tween Trade and Competition Policy." World
Economy 18(3): 345-65.
Krugman, Paul, and Gordon Hanson. 1993.
"Mexico-U.S. Free Trade and the Location of Pro-   Matusz, Steven J., and David Tarr. 1999. "Adjusting
duction." In Peter Garber, ed., The Mexico-U.S.  to Trade Policy Reform." Policy Research Working
Free TradeAgreement. Cambridge, Massachusetts;  Paper no. 2142. World Bank, Washington, D.C.
London: MIT Press.
McCallum, J. 1995. "National Borders Matter:
Lawrence, Robert Z. 1991. "Emerging Regional Ar-   Canada-US Regional Trade Patterns." American
rangements: Building Blocs or Stumbling Blocks?"  Economic Review 85(3): 615-23.
In O'Brien, ed., Finance and the International
Economy 5: TheAMEXBank. Review Prize Essays.   McKibbin, Warwick J. 1994. "Dynamic Adjustment
New York: Oxford University Press.              to Regional Integration: Europe 1992 and
139



B I B L I O G RA P H Y
NAFTA." Journal of Japanese and International   Midelfart-Knarvik, H.G., S. Redding Overman, and
Economics 8(4): 422-53.                         Anthony J. Venables. 1999. "The Location of In-
dustry in Europe." Centre for Economic Policy
McMillan, John. 1993. "Does Regional Integration   Research, London.
Foster Open Trade? Economic Theory and GATT's
Article 24." In Kym Anderson and Richard   Milward, Alan. 1984. The Reconstruction of Europe,
Blackhurst, eds., Regional Integration and the Glo-  1945-51. Berkeley: University of California Press.
bal TirailngSystem. London: HarvesterWheatsheaf.
.1992. TheEuropeanRescueofthe ANtionState.
Melo, Jaime de, Arvind Panagariya, and Dani Rodrik.  Berkeley: University of California Press.
1993. "The New Regionalism: A Country Perspec-
tive." In Jaime de Melo and Arvind Panagariya,   Milward, Alan, and S.B. Saul. 1973. The Economic
eds., New Dimensions in RegionalIntegration. New  Development of Continental Europe, 1780-1870.
York and Melbourne: Cambridge University Press.  Totowa, New Jersey: Rowman and Littlefield.
Messerlin, Patrick. 1990. "Antidumping Regulations   Mohieldin, Mahmoud. 1997. "The Egypt-EU Part-
or Pro-Cartel Law? The EC Chemical Cases."      nership Agreement and Liberalization of Trade in
Workl Economy 13: 465-92.                       Services." In A. Galal and B. Hoekman, eds., Re-
gional Partners in Global Markets: Limits and Pos-
. 1996. "Competition Policy and Antidump-    sibilities ofthe Euro-MedAgreements. London: Cen-
ing Reform." Presented at the conference on The  tre for Economic Policy Research.
World Trading System: Challenges Ahead, June
24-25, Institute for International Economics,   Murphy, Craig. 1994. International Organization and
Washington D.C.                                  Industrial Change: Global Governance since 1850.
New York: Oxford University Press.
. 1997. "Reforming the Rules of Antidump-
ing Policies." In Horst Siebert, ed., Towards a New    Neven, Damien. 1996. "Regulatory Reform and the
Global Framework for High-Technology Competi-   Internal Market." Processed.
tion. I nstitut fur Weltwirtschaft an der Universitat
Kiel Symposia and Conference Proceedings.   Nishimuzi, Mieko, and John M. Page. 1982. "Total
Tubingen: Mohr (Siebeck).                       Factor Productivity Growth, Technological
Progress, and Technical Efficiency Clhange: Di-
. 11998. "Technical Regulations and Industry  mensions of Productivity Change in Yugoslavia."
Standards in the EU." World Bank, Washington,   Economic Journal92: 920-36.
D.C. Processed.
Nogues, Julio J., and Rosalinda Quintanilla. 1993.
Messerlin, Patrick, and PK.M. Tharakan. 1999. "The  "Latin America's Integration and the Multilateral
Question of Contingent Protection." World        Trading System." In Jaime de Melo and Arvind
Economy 22(9): 1251-70.                         Panagariya, eds., New Dimensions in Regional
140



B I B L I O G RA P H Y
Integration. London: Centre for Economic Policy  Economics ofArms Reduction and the Peace Process.
Research.                                      Amsterdam: North-Holland.
Olarreaga, Marcelo, and Isidro Soloaga. 1998. "Ex-   _ . 1997. "Why Democracies Cooperate More
plaining MERCOSUR's Tariff Structure: A Politi-  and Fight Less: The Relationship between Inter-
cal Economy Approach." World Bank Economic     national Trade and Cooperation." Review oflnter-
Review 12(2): 297-320.                         national Economics 5(1): 295-309.
Oye, Kenneth. 1992. Economic Discrimination and   Pollard, Sidney. 1974. European Economic Integration,
Political Exchange: World Political Economy in the  1815-1970. London: Thames and Hudson.
1930s and 1980s. Princeton, New Jersey: Princeton
University Press.                           Pomfret, Richard. 1997. The Economics of Regional
TradingArrangements. New York: Oxford Univer-
Page, Sheila. 1996. "Intensity Measures for Regional  sity Press.
Groups." Paper prepared for the European Asso-
ciation of Development Research Institutes. Ninth   Porter, Michael. 1998. On Competition. Cambridge,
General Conference, Vienna, September.         Massachusetts: Harvard Business School Press.
-   1997. "Regions and Developing Countries."   Richardson, Martin. 1994. "Customs Unions and
Overseas Development Adminstration, London.    Domestic Taxes." Canadian Journal of Economics
27(3): 79-96.
Panagariya, Arvind, and Ronald Findlay. 1994. "A
Political Economy Analysis of Free Trade Areas and   Ricupero, Rubens. 1998. "What Policy Makers
Customs Unions." Policy Research Working Pa-   Should Know about Regionalism." Keynote Ad-
per no. 1261. World Bank, Washington, D.C.     dress presented at the World Bank Conference on
What Policy Makers Should Know about Region-
Pelkmans, Jacques, L. Alan Winters, and Helen    alism, Geneva, May.
Wallace. 1998. Europes'DomesticMarket. Chatham
House Papers Series no. 43. New York and Lon-   Roberts, Mark, and James Tybout, eds. 1996. Indus-
don: Royal Institute of International Affairs.  trial Evolution in Developing Countries: Micro Pat-
ters of Turnover, Productivity, and Market Structure.
Pohl, Gerhard, and Piritta Sorsa. 1992. European In-  Oxford, United Kingdon: Oxford University Press.
tegration and Trade with the Developing World.
Policy and Research Series no. 21. Washington,   Robson, Peter. 1998. The Economics ofInternational
D.C.: World Bank.                              Integration. London: Routledge.
Polachek, Solomon W  1992. "Conflict and Trade:   Rodrik, Dani. 1988. "Imperfect Competition, Scale
An Economics Approach to Political Interactions."  Economies, and Trade Policy in Developing Coun-
In Walter Isard and Charles H. Anderton, eds.,    tries." In R.E. Baldwin, ed., Trade Policy Issues and
141



B I B LI C) G RAP HY
Emp)iricalAnalysis. Chicago: University of Chicago  Working Paper Series no. 2501. World Bank,
and National Bureau of Economic Research.       Washington, D.C.
Roy,Jayanta. 1998. "TradeFaciitation:TheWorldBank   Schiff, Maurice, and L. Alan Winters. 1998a. "Re-
Experience." Paper presented at a World Trade Or-  gional Integration as Diplomacy." World Bank Eco-
ganization Trade Facilitation Symposium, Geneva,  nomic Review 12(2): 271-96.
Switzerland, March.
_ .___ 1998b. "Regional Integration, Security, and
Rutherford, Thomas, Elisabet E. Rutstr6m, and      Welfare." In Regionalism and Development. Report
David Tarr. 1999. "The Free-Trade Agreement     of the June 1997 European Commission and World
between the European Union and a Representa-    Bank Seminar. European Commission Studies Se-
tive Arab Mediterranean Country: A Quantitative  ries no. 1. Washington, D.C.: World Bank.
Assessment." In B. Hoekman and J. Zaarouk, eds.,
Catching up with the Competition: Trade Policy   Schmidt, Klaus. 1994. "Managerial Incentives and
Challenges in the Middle East and North Africa. Ann  Product Market Competition." Discussion Paper
Arbor, Michigan: University of Michigan Press.   no. 430. University of Bonn, Bonn.
Saggi, Kamal. 1999. "Trade, Foreign Direct Invest-   Schbne, Rainer. 1996. Alternatives to Antidumping
ment, and International Technology Transfer."   from an Antitrust Perspective. Ph.D. dissertation,
Southern Methodist University, Dallas, Texas. Pro-  University of St. Gallen. St. Gallen, Switzerland.
cessed.
Smith, Alasdair, andAnthonyVenables. 1988. "Com-
Sapir, Andre. 1993. "The European Community: A     pleting the Internal Market in the European Com-
Case of Successful Integration? A Comment." In  munity: Some Industry Simulations." European
Jaime tie Melo and Arvind Panagariya, eds., New  Economic Review 32(7): 1501-25.
Dimensions in Regional Integration. London: Cen-
tre for Economic Policy Research.             Soloaga, Isidro, and L. Alan Winters. 1999a. "How
Has Regionalism in the 1990s Affected Trade?"
Schiff, Maurice. 1997. "Small Is Beautiful: Preferen-  Policy Research Working Paper Series no. 2156.
tial Trade Agreements and the Impact of Country  World Bank, Washington, D.C.
Size, Market Share, and Smuggling." Journal of
Economnic Integration 12: 359-87.                  . 1999b. "Regionalism in the Nineties: What
Effect on Trade?" Discussion Paper Series no. 2183.
. 1999. "Will the Real 'Natural Trading Part-  Center for Economic Policy Research. Cambridge,
ner' Please Stand Up?" Policy Research Working  Massachusetts.
Paper no. 2161. World Bank, Washington, D.C.
Srinivasan, T.N. 1998. "Regionalism and the WTO:
_ .____ 2000. "Multilateral Trade Liberalization, Po-  Is Nondiscrimination Passe?" In Anne 0. Krueger,
litical Disintegration, and the Choice of Free Trade  ed., The WTO as an International Organization.
Areas Versus Customs Unions." Policy Research    Chicago: University of Chicago Press.
142



B I B L I 0 G RA P H Y
Staiger, Robert, and Frank Wolak. 1989. "Strategic  Technical Efficiency.' "Journal ofInternational Eco-
Use of Antidumping to Enforce Tacit Interna-    nomics 31: 231-50.
tional Collusion." Working Paper no. 3016. Na-
tional Bureau of Economic Research. Cambridge,   USITC (United States International Trade Commis-
Massachusetts.                                  sion). 1997. "Study on the Operation and Effects
of the NAFTA." Washington, D.C.
Staples, Brian. 1998. "Trade Facilitation." World
Bank, Washington, D.C. Processed. Available on   Venables, AnthonyJ. 1999. "Integration Agreements:
http://www.worldbank.org/wbiep/trade/.          A Force for Convergence or Divergence?" Proceed-
ings of World BankABCDE Conference. Policy Re-
Summers, L. 1991. "Regionalism and the WorldTrad-  search Working Paper Series no. 2260. World
ing System." In Policy Implications of Trade and  Bank, Washington, D.C.
Currency Zones. A symposium sponsored by the
Federal Reserve Bank of Kansas City, Jackson Hole,   Viner, Jacob. 1950. The Customs Union Issue. Wash-
Wyoming.                                        ington, D.C.: Carnegie Endowment for Interna-
tional Peace.
Talbott, Strobe. 1996. "Democracy and the National
Interest." ForeignAffairs75: 47-63.           Wang, Zhen Kun, and L. Alan Winters. 1998.
"Africa's Role in Multilateral Trade Negotiations:
Tavares, Jose, and Luis Tineo. 1998. "Harmoniza-   Past and Future." Journal of African Economies
tion of Competition Policies among MERCOSUR     Supplement 1 (June): 1-33.
Countries." Antitrust-Bulletin 43(1): 45-70.
Whalley, John. 1996. "Why Do Countries Seek Re-
Tsoukalis, Loukas. 1993. The New European          gional Trade Agreements?" In J. Frankel, ed., The
Economy: The Politics and Economics of European  Regionalization of the World Economy. Chicago:
Integration. Oxford, United Kingdom: Oxford     Chicago University Press.
University Press.
Wilmore, Larry. 1976. "Trade Creation, Trade Di-
Tybout, James. 1999. "Manufacturing Firms in De-   version, and Effective Protection in the Central
veloping Countries: How Well Do They Do, and    American Common Market." Journal of Develop-
Why?" Journal ofEconomic Literature 37(3).      ment Studies 12: 396-414.
Tybout, James, and M. Daniel Westbrook. 1995.   _     . 1978. "The Industrial Economics of Intra-
"Trade Liberalization and the Dimensions of Ef-  IndustryTrade and Specialization." In H. Giersch,
ficiency Change in Mexican Manufacturing In-    ed., On the Economics of Intra-Industry Trade.
dustries." Journal of International Economics 39:  Tubingen, Germany: J.C.B. Mohr.
53-78.
Winham, G.R. 1986. International Trade and the
Tybout, James, Jaime de Melo, and Vittorio Corbo.  Tokyo Round Negotiation. Princeton, New Jersey:
1991. "The Effects of Trade Reforms on Scale and  Princeton University Press.
143



B I B LI CO G RA P H Y
Winters, L. Alan. 1992. "The Welfare and Policy Im-   Winters, L. Alan, and Won Chang. Forthcoming.
plications of the International Trade Consequences  "Regional Integration and the Prices of Imports:
of'1992."' American Economic Review 82(2): 104-  An Empirical Investigation." Journal of Interna-
108.                                           tional Economics.
-____ 1993. "The European Community: A Case   Wonnacott, R.I. 1996. "Canadian Trade Policy: The
of Successful Integration." Discussion Paper no.  GATT's 1995 Review." The World Economy
775. Centre for Economic Policy Research, Lon-  Supplement: 67-80.
don.
Wonnacott, R.I., and Mark Lutz. 1989. "Is there a
.1994. "The EC and Protection: The Political  Case for Free Trade Areas." In Jeffrey Schott, ed.,
Economy." European Economic Review 38: 596-    Free Trade Areas and US Trade Policy. Washington,
603.                                           D.C.: Institute for International Economics.
. 1997. "Lebanon's Euro-Mediterranean Agree-   World Bank. 1999. World Development Report 1999.
ment: IPossible Dynamic Benefits." In W  Shahin  Washington, D.C.
and Shehadi K., eds., Pathways to Integration: Leba-
non and the Euro-Mediterranean Partnership. Bonn:   WTO (World Trade Organization). 1995. "Region-
Konrad Adenauer Foundation.                    alism and the World Trading System." Geneva:
Secretariat.
. 1999. "Regionalism Versus Multilateralism."
In Richard Baldwin, Daniel Cohen, Andre Sapir,   Yeats, A. 1998. "Does MERCOSUR's Trade Perfor-
and Anthony Venables, eds., Market Integration,  mance Raise Concerns about the Effects of Re-
Regionalism and the Global Economy. Cambridge,  gional Trade Arrangements?" The World Bank Eco-
United Kingdom: Centre for Economic Policy     nomic Review 12(1): 1-28.
Research.
144






THE WORLD BANK
The number of regional trade blocs has grown rapidly in recent decades. More than one-third
of world trade now takes place within such blocs. New forms of bloc, encompassing both rich
and poor countries, attempt to secure "deep integration" of economic activities. This book goes
beyond existing studies to address the implications of these new forms of trading arrangement
and to analyze the policy options that countries now face.
Trade Blocs analyzes both the political and the economic benefits of regional trade blocs. It
argues that the benefits can sometimes be illusory and that careful economic choices have to be
made if the schemes are to bring benefits. The trade-offs faced in these choices are analyzed,
and the design of successful blocs is studied. The focus of the book is on policy choices for
developing countries, with wider issues about the effect of trade blocs on the world trading
system also addressed.
The contents: 1. Regional Integration Agreements. 2. Politics and Policymaking. 3. Economic
Benefits and Costs. 4. Policy Choices. 5. Trade Blocs and the World Trading System. 6.
Conclusion: Tell Me the Truth about Trade Blocs.
THE WORLD BANK                                               OXFORD UNIVERSITY PRESS
1818 H Street N.W
Washington, D.C. 20433 USA
Telephone: 202-477-1234
Facsimile: 202-477-6391
Telex: MCI 64145 WORLDBANK
MICI 248423 WORLDBANK 
Internet: wwxv.worldbank.org
E-mail: books@worldbank.org