REGIONAL RULES IN THE GLOBAL TRADING SYSTEM
The proliferation of regional trade agreements (RTAs) over the past two decades has highlighted the need to look closely at the relationships between regional and WTO rules or disciplines. A major obstacle to advancing understanding of RTAs is the absence of detailed information about their contents. This has limited the debate between those who view RTAs as discriminatory instruments hostage to protectionist interests and those who see them as conducive to multilateral trade opening. This book provides detailed analysis of RTA rules in six key areas – market access, technical barriers to trade, contingent protection, investment, services and competition policy – across dozens of the main RTAs in the world. The analysis helps to provide new insights into the interplay between regional and multilateral trade rules, advances understanding of the economic effects of RTAs and contributes to the discussion on how to deal with the burgeoning number of RTAs. a n t o n i e s t e v a d e o r d a l is Manager at the Integration and Trade Sector of the Inter-American Development Bank in Washington. k a t i s u o m i n e n is Trade Economist at the Integration and Trade Sector of the Inter-American Development Bank in Washington. r o b e r t t e h is Counsellor in the Economic Research and Statistics Division of the World Trade Organization.
REGIONAL RULES IN THE GLOBAL TRADING SYSTEM Edited by ANTONI ESTEVADEORDAL, KATI SUOMINEN and ROBERT TEH
cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sa˜o Paulo, Delhi Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521759342 Ó World Trade Organization 2009 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2009 Printed in the United Kingdom at the University Press, Cambridge A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication data Regional rules in the global trading system / [edited by] Antoni Estevadeordal, Kati Suominen, Robert Teh. p. cm. Includes index. ISBN 978-0-521-76084-3 (hardback) ISBN 978-0-521-75934-2 (paperback) 1. International trade. 2. Regionalism–Economic aspects. 3. Trade blocs. 4. Foreign trade regulation. 5. Commercial treaties. I. Estevadeordal, Antoni. II. Suominen, Kati. III. Teh, Robert. IV. Title. HF1379.R4394 2009 3820 .91–dc22 2008053616 ISBN 978-0-521-76084-3 hardback ISBN 978-0-521-75934-2 paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party Internet Web Sites referred to in this publication, and does not guarantee that any content on such Web Sites is, or will remain, accurate or appropriate. The views and opinions expressed in this volume are strictly those of the authors’ and editors’ alone and do not reflect the views of the IDB, the WTO or any of their members.
From Kati Suominen, to Gen. Fred F. Woerner, Ret. and From Robert Teh, to Imelda de Leon
CONTENTS
List of figures page viii List of tables x List of Contributors xii Foreword xiii Acknowledgements xvi List of abbreviations xvii 1 Introduction 1 antoni estevadeordal, kati suominen an d r o b e r t te h 2 Big-Think Regionalism: a critical survey r i c h a r d ba l d w i n
17
3 Market access provisions in regional trade agreements antoni estevadeordal, matthew shearer and k at i s uomi nen
96
4 Trade remedy provisions in regional trade agreements robert teh, thomas j. prusa and m i c h e l e bu d e t t a
166
5 A mapping of regional rules on technical barriers to trade r o b e r t a p i e r m a r t i n i a nd mi c h el e b u d et ta 6 Services liberalization in the new generation of preferential trade agreements: how much further than the GATS? 316 m a r t i n r o y , j u a n m a r c h e t t i a n d h o e li m 7 Mapping investment provisions in regional trade agreements: towards an international investment regime? 365 b a r b a r a k o t sc h w a r 8 Competition provisions in regional trade agreements robert teh
Appendix: List of RTAs included in the survey Index 503 vii
418
492
250
FIGURES
2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 3.1 3.2a 3.2b 3.3a 3.3b 3.4a 3.4b 3.5a 3.5b
The RTA diagram’s trade pattern page 27 The RTA (preferential trade arrangement) diagram 29 Ambiguous net welfare effects 30 The Johnson diagram (small Home and Partner nations) 31 The small PTA diagram: a simple case 34 Net welfare effects, FTA to global free trade 43 Stumbling-bloc FTAs 44 Juggernaut framework 52 Juggernaut building-bloc logic 52 Imported MFN liberalization 62 An economic theory of the GATT 64 Percentage of tariff lines duty free, by selected benchmark years 112 Reciprocity of concessions: Year 5 113 Reciprocity of concessions: Year 10 113 Duty-free lines in 5 years v. TRQ incidence 114 Duty-free lines in 10 years v. TRQ incidence 114 Agricultural duty-free lines in 5 years v. TRQ incidence 115 Agricultural duty-free lines in 10 years v. TRQ incidence 115 Evolution of duty-free treatment in selected RTAs 124 Evolution of duty-free treatment in selected RTAs: North–North agreements v. agreements with a Southern party 124 3.5c Evolution of sectoral duty-free treatment in selected RTAs 125 3.6a Distribution of liberalization by RTA parties in chapters, Year 5 126 3.6b Distribution of liberalization by RTA parties in chapters, Year 10 126 3.7a Distribution of liberalization of chapters in RTA parties’ schedules, Year 5 128 3.7b Distribution of liberalization of chapters in RTA parties’ schedules, Year 10 128 3.8 Evolution of duty-free treatment as trade-weighted percentage of tariff lines 130 3.9 Evolution of duty-free treatment as percentage of imports 130 3.10a Evolution of duty-free access in Canadian market for Costa Rican goods: a comparison using various measures 131
viii
list of figures 3.10b Evolution of duty-free access in Costa Rican market for Canadian goods: a comparison using various measures 132 3.10c Evolution of duty-free access in Chilean market for Korean goods: a comparison using various measures 132 3.10d Evolution of duty-free access in Korean market for Chilean goods: a comparison using various measures 133 3.10e Evolution of duty-free access in EU market for South African goods: a comparison using various measures 133 3.10f Evolution of duty-free access in South African market for EU goods: a comparison using various measures 134 3.10g Evolution of duty-free access in Moroccan market for US goods: a comparison using various measures 134 3.10h Evolution of duty-free access in US market for Morrocan goods: a comparison using various measures 135 3.11 Coverage of selected market access disciplines in fifty RTAs 139 3.12 Coverage of selected RTAs, by NTM provision 140 3.13 Coverage of selected RTAs, by other measure provision 140 3.14 Coverage of selected RTAs, by special regime provision 141 3.15 Coverage of selected RTAs, by RoO provision 142 3.16 Coverage of selected RTAs, by customs procedure provision 143 3.17 Restrictiveness of RoO in selected RTAs 151 3.18 Facilitation of regime-wide RoO in selected RTAs 151 4.1a Frequency of anti-dumping initiations, 1980–2006 170 4.1b Frequency of anti-dumping measures, 1995–2006 170 4.2a Frequency of CVD initiations, 1995–2006 171 4.2b Frequency of CVD measures, 1995–2006 171 4.3a Frequency of safeguard initiations, 1995–2006 172 4.3b Frequency of safeguard measures, 1996–2006 172 4.4 Number of RTAs in sample coming into force, by decade 177 4.5 RTAs by level of development of members 177 4.6 Hub-and-spoke arrangement of RTAs in sample 178 6.1 Proportion of sub-sectors with new and improved commitments under mode 3, per WTO Member (when comparing the GATS offer to the GATS schedule (‘GATS’) and the PTA commitments to the GATS offer (‘PTA’)) 332 6.2 Proportion of sub-sectors with new and improved commitments under mode 1, per WTO Member (when comparing the GATS offer to the GATS schedule (‘GATS’) and the PTA commitments to the GATS offer (‘PTA’)) 361 7.1 Coverage by country 397
ix
TABLES
3.1 3.2
3.3
3.4 4.1 4.2 4.3 4.4a 4.4b 4.5 4.6 4.7a 4.7b 4.8 4.9a 4.9b 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 6.1
RTAs covered in the study page 105 Products subject to tariff rate quotas in CAFTA: US tariff quotas on products entering from Central America and the Dominican Republic 118 Products subject to tariff-rate quotas in CAFTA: Central American and Dominican Republic tariff quotas on products entering from the United States 120 Three-level matrix for mapping market-access provisions in RTAs 152 Trade remedy actions, initiations and measures, 1995–2006 169 Anti-dumping template 186 Countervailing duties template 188 Bilateral safeguards template 191 Global safeguards template 192 Anti-dumping mapping 195 Countervailing duties mapping 203 Bilateral safeguards mapping 208 Global safeguards mapping 231 Characteristics of RTAs that have disallowed trade remedies 241 Probit estimation results 244 Multinomial logit estimation results 245 The structure of the template for mapping regional rules on standards, technical regulations and conformity assessment procedures 270 An overview of TBT provisions in RTAs 272 Characteristics of RTAs by hub 279 The extent of regional TBT liberalization 283 The likelihood of provisions encouraging the harmonization of technical regulation 288 The likelihood of provisions encouraging mutual recognition of conformity assessment 288 The likelihood of provisions encouraging notification of standards and procedures 289 The likelihood of provisions establishing a dispute settlement body 289 Preferential trade agreements reviewed 322
x
list of tabl es 6.2 6.3 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 8.1 8.2 8.3
Cross-tabulation of countries and PTAs 328 Examples of commitments providing for the phasing-in of liberalization 346 Agreements used in sample and date of entry into force 376 Mapping of investment provisions: results by geographic pairing Mapping of investment provisions: results by levels of development Scope and coverage 401 Non-discrimination: most favoured nation and national treatment standards 403 Standards of treatment 406 Performance requirements and restrictions on nationality 409 Denial of benefits, transparency and dispute settlement 412 Competition template 424 Competition mapping 426 Correlation of competition provisions in selected hubs 486
xi
382 386
CONTRIBUTORS
Richard Baldwin is Professor of International Economics at the Graduate Institute of International Studies in Geneva. Michele Budetta is a graduate student at the University Cattolica del Sacro Cuore in Milan Antoni Estevadeordal is Manager in the Integration and Trade Sector at the Inter-American Development Bank. Barbara Kotschwar is Research Associate at the Peterson Institute for International Economics. Aik Hoe Lim is Counsellor at the Trade in Services Division of the World Trade Organization. Juan Marchetti is Counsellor at the Trade in Services Division of the World Trade Organization. Roberta Piermartini is Counsellor at the Economic Research and Statistics Division of the World Trade Organization. Thomas J. Prusa is Professor of Economics at Rutgers University. Martin Roy is Counsellor at the Trade in Services Division of the World Trade Organization. Matthew Shearer is Statistician and Economist at the Inter-American Development Bank. Kati Suominen is International Trade Specialist in the Integration and Trade Sector at the Inter-American Development Bank. Robert Teh is Counsellor at the Economic Research and Statistics Division of the World Trade Organization.
xii
FOREWORD
The number of regional trade agreements (RTAs) has grown enormously over the last decade. Over 200 RTAs currently in force have been notified to the World Trade Organization, with many more being currently negotiated. More and more areas, that traditionally were not part of trade agreements, are being covered by these agreements. RTAs now typically include rules on competition policy, the environment, labour, services, investments, intellectual property, trade remedies or technical barriers to trade, in addition to the usual market access provisions in merchandise trade. The proliferation of RTAs requires increased attention to be paid to the potential conflicts and complementarities between ‘regional’ and ‘global’ rules. The relationship between regionalism and multilateralism has sometimes been framed as one where RTAs are either a building bloc or a stumbling bloc to multilateralism. But, having closely witnessed integration arrangements in Europe, Latin America and the Caribbean, Africa, Asia and elsewhere around the globe, we believe that this is not as black and white. RTAs have delivered important trade gains for their participants. But, often, they have also been a source of trade diversion and have hampered movement towards greater multilateral liberalization, as is the case with certain rules of origin. In our view, the key research question is to identify those regional rules that promote complementarities with the multilateral trading system and those that conflict with it. A great amount of scholarly interest has been spawned by regionalism, with both eminent economists and political scientists making many valuable theoretical contributions. But, with a few notable exceptions like NAFTA and the EU, very little research has as yet been devoted to the actual contents of these RTAs. There is not enough understanding of the diversity in regional rules, the difference between these regional rules and multilateral rules, the feasibility of converging towards some common xiii
xiv
foreword
standard and the appropriate methods for assessing the compatibility of regional rules with multilateral rules. As part of the effort to remedy this gap in knowledge, the InterAmerican Development Bank and the World Trade Organization have carried out a joint research project to enhance our understanding of regional rules and their implications for the global trading system. This research attempts to complement the already rich vein of economic theorizing by assembling a comprehensive mapping of the trade and trade-related rules that govern RTAs around the globe. It maps various provisions or rules across dozens of the most important RTAs in six key areas: market access, trade remedies, technical barriers to trade, services, investment, and competition policy. For the first time, a framework for analysing these regional rules and a mapping of these rules have been put together in a single volume. One of the early harvests of the WTO Doha Round has been the establishment, on a provisional basis, of a new transparency mechanism for RTAs. Regional trade agreements have been the subject of some multilateral examination since the days of the GATT. This new transparency mechanism brings a higher level of examination to RTAs. It provides for early announcement of any RTA, and its notification to the WTO. It requires a factual presentation of the notified RTAs to be made to WTO Members on the basis of a report prepared by the WTO Secretariat. It mandates that any changes affecting the implementation of an RTA, or the operation of an already implemented RTA, will be notified to the WTO. Finally, at the end of the RTA’s implementation period, it calls for the parties to the RTA to submit to the WTO a written report on the realization of liberalization commitments in the RTA as originally notified. We believe that the contributions in this volume can provide relevant conceptual frameworks, methods of analysis and data which could be of help in the context of the new transparency mechanism. This volume, which addresses one of the key challenges faced by the multilateral trading system, represents another facet of the fruitful partnership between the IDB and the WTO. The IDB has a long history of research and capacity-building on trade and integration issues and, together with other regional and multilateral lending agencies, is heavily involved in supporting the WTO’s Aid for Trade initiative. We look
foreword
xv
forward to more collaborative endeavours between our two institutions that will help improve the workings of the global trading system. Pascal Lamy Director-General World Trade Organization Luis Alberto Moreno President Inter-American Development Bank
ACKNOWLEDGEMENTS
The research collaboration between the Inter-American Development Bank and the World Trade Organization Secretariat which gave rise to this publication has made the editors indebted to numerous colleagues rich in ideas and enormously interested in regional integration. We would like to express our deepest thanks to all the authors of the papers in this volume for their contribution and involvement in this joint research undertaking. We are grateful for the financial support extended by the Inter-American Development Bank without which this research would not have been possible. Along the way, several seminars and workshops were organized in Geneva and Washington, D. C. We appreciate the comments and suggestions made by seminar and workshop participants that have provided our contributors with fresh perspectives and valuable insights. Particular thanks are due to Chad Bown, Simon Evenett and Ken Vandevelde for their careful reading and thoughtful comments on several chapters of this volume. We are beholden to staff members of the IDB and WTO Secretariats for their assistance, patience and many helpful suggestions. We would like to make specific mention of Patrick Low, Clem Boonekamp, Robert Anderson, Stefania Bernabe´, Jo-Anne Crawford, Jesse Kreier, Anthony Martin, Andreas Sennekamp and Hiromi Yano of the WTO Secretariat; and Santiago Levy of the IDB, as well as Robert Devlin and Nohra Rey de Marulanda, who served at the IDB during the launch of this joint research project. Finally, we wish to thank Finola O’Sullivan of Cambridge University Press for her support in preparing this volume.
xvi
ABBREVIATIONS
ACCC ACE AD AFAS AFTA ALADI ANZCERTA ASEAN AUSFTA BIT BOP CA CACM CAFTA CARICOM CDC CEMAC CEN CENELEC CER CET CITT COMESA COMTRADE
Australian Competition and Consumer Commission Acuerdo de Complementacio´n Econo´mica Anti-dumping ASEAN Framework Agreement on Services ASEAN Free Trade Area Asociacio´n Latinoamericana de Integracio´n Australia New Zealand Closer Economic Relations Trade Agreement Association of Southeast Asian Nations Australia–US Free Trade Agreement bilateral investment treaty balance of payments Central America Central American Common Market Central America Free Trade Agreement Caribbean Community and Common Market Comite´ de Defensa de la Competencia Communaute Economique et Monetaire de l’Afrique Centrale Committee for Standardisation European Committee for Electrotechnical Standardisation Australia–New Zealand Closer Economic Relations common external tariff Canadian International Trade Tribunal Common Market for Eastern and Southern Africa United Nations Commodity Trade Statistics xvii
xviii
CRTA CU CUSFTA CVD DOC DR ECJ EEA EEC EFTA ETSI EU FDI FTA FTC FYROM GATS GATT GCC GDP GPA HKC HS IADB IAF ICSID IEC ILAC ISO ITO LAIA MAI MERCOSUR MFN MRA MTN NAFTA NATO
list of abbreviations
WTO Committee on Regional Trade Agreements customs union Canada–US Free Trade Agreement countervailing duty US Department of Commerce Dominican Republic European Court of Justice European Economic Area European Economic Community European Free Trade Association European Telecommunications Standards Institute European Union foreign direct investment free trade agreement Federal Trade Commission Former Yugoslav Republic of Macedonia General Agreement on Trade in Services General Agreement on Tariffs and Trade Gulf Cooperation Council gross domestic product Government Procurement Agreement Hong Kong, China Harmonized System Inter-American Development Bank International Accreditation Forum International Centre for Settlement of Investment Disputes International Electrotechnical Commission International Laboratory Accreditation Cooperation International Organization for Standardization International Trade Organization Latin American Integration Association Multilateral Agreement on Investment Mercado Comun del Sur most-favoured-nation mutual recognition agreement Multilateral Trade Negotiations North American Free Trade Agreement North Atlantic Treaty Organization
l ist of abbreviations
NGO NT NTM NZCC OCT OECD PTA QM RoO RoW RTA SADC SAPTA SAT SCM SEP SITC SPARTECA SPS SSG TBT TRIMs TRIPS TRQ UEMOA UN/EDIFACT UNCTAD VER WAEMU WCO WTO
xix
non-governmental organization national treatment non-tariff measure New Zealand Commerce Commission Overseas Countries and Territories Organization for Economic Co-operation and Development preferential trade agreement quantitative measures rules of origin rest of the world regional trade agreement Southern African Development Community South Asian Preferential Trade Arrangement substantially all trade subsidies and countervailing measures (Trans-Pacific) Strategic Economic Partnership Standard International Trade Classification South Pacific Regional Trade and Economic Cooperation Agreement sanitary and phytosanitary special safeguard technical barriers to trade Trade-Related Investment Measures Trade-Related Intellectual Property Rights tariff rate quota L’Union Economique et Mone´taire Ouest Africaine United Nations/Electronic Data Interchange for Administration, Commerce, and Transport United Nations Conference on Trade and Development Voluntary Export Restraint West African Economic and Monetary Union World Customs Organization World Trade Organization
1 Introduction antoni estevadeordal, kati suominen and robert teh* 1
Introduction
Regional trade agreements (RTAs) have proliferated around the world in the past decade. Some 200 RTAs currently in force have been notified to the World Trade Organization (WTO) and the number will continue to rise given the many RTAs being proposed and negotiated. It is estimated that, if one takes into account RTAs which are in force but have not been notified, signed but not yet in force, currently being negotiated, and in the proposal stage, close to 400 RTAs are scheduled to be implemented by 2010 (Fiorentino, Verdeja and Toqueboeuf, 2006). Virtually all countries are member of at least one RTA, with most countries belonging to two or more RTAs at once. The geographic reach of RTAs has also changed over time, making ‘regional’ somewhat of a misnomer. While most RTAs are still formed among countries inhabiting the same region or continent, they increasingly involve members that are not immediate neighbours and create partnerships spanning oceans. Trans-Atlantic and trans-Pacific RTAs are gaining in number through such agreements as the European Union (EU)–Mexico Economic Partnership Agreement, the EFTA–Chile free trade agreement (FTA) and the recently signed Korea–US FTA. The economic importance of regional trade agreements has continued to grow. More than half of global merchandise trade flows among countries connected by a common RTA. But RTAs are today increasingly important in areas other than merchandise trade. Indeed, the architecture of RTAs has become both more comprehensive and more complex. Besides trade in goods, many RTAs now regulate such subjects as trade in services, investments, standards, intellectual property *
The views and opinions expressed in this volume are strictly those of the authors and editors alone and do not reflect the views of the IDB, the WTO or any of their Members.
1
2
antoni estevadeordal, kati suominen , robert teh
and competition rules, as well as a host of issues not directly related to trade, such as labour and environment. The body of rules governing international trade has been extended to matters traditionally considered to be within the realm of domestic regulations (‘behind-theborder’ measures). These developments suggest that RTAs have become a major and strategic part of commercial policy for many countries. But RTAs also pose important challenges for the multilateral trading system. Indeed, the growing importance of RTAs has directed attention to the potential conflicts as well as complementarities between the rules that are adopted in RTAs (‘regional’ rules) and the multilateral rules established in such agreements as the General Agreement on Tariffs and Trade (GATT) of 1994, the General Agreement on Trade in Services (GATS) and other WTO Agreements (‘global’ rules). While the GATT had in the past conducted examinations of RTAs, scrutiny may now become more exacting. The multilateral Doha trade round launched in 2001 included a negotiating mandate aimed at clarifying and improving disciplines and procedures under the existing WTO provisions applying to RTAs – GATT Article XXIV, the Enabling Clause, and GATS Article V. These negotiations have resulted in a new transparency mechanism that was adopted by the WTO’s General Council in December 2006. The transparency mechanism obliges Members to notify the WTO of any RTA that Members enter into and to provide information about the agreement. The mechanism also mandates the WTO Secretariat to prepare a report on notified RTAs. While this report on the RTA has to be ‘factual’ and refrain from any ‘value judgment’, the increased level of scrutiny can alert the rest of the WTO membership to some of the rules and practices in RTAs that adversely affect non-RTA members. This may induce countries to adopt RTA rules that complement rather than conflict with existing WTO agreements. The growing policy attention paid to RTAs finds a parallel in the debate in the economic literature on whether RTAs are ‘building blocs’ or ‘stumbling blocs’ to multilateral trade liberalization.1 These concepts refer to the nature of the dynamics or time paths that RTA formation can generate (Bhagwati and Panagariya, 1999). RTAs are building blocs if they accelerate multilateral trade negotiations or progressively enlarge their membership so that they lead to global free trade. RTAs are stumbling blocs if they hamper the attainment of global trade liberalization. 1
Bhagwati (1991) first coined the terms ‘building bloc’ and ‘stumbling bloc’.
introduction
3
The stumbling bloc camp argues that RTAs undermine countries’ incentives to undertake further multilateral liberalization because members are unwilling to dilute the preferential access they have to the markets of RTA partners. Another argument is that RTAs can create incompatible regulatory structures and standards which lock in the members’ policies, and increase the adjustment costs associated with multilateral liberalization, thus making it less attractive. The importance of non-economic motives or interests can make RTAs a stumbling bloc to global free trade (Lima˜o, 2007). RTAs can be valuable to a large country because the preferential access to its market allows it to extract co-operation in nontrade matters from smaller partners. Multilateral tariff reductions reduce the value of preferential access to the large market and thus the surplus that can be extracted from potential RTA partners. Various political-economy models have sought to show that the establishment of an RTA weakens the motivation of the members for reciprocal liberalization with non-members. In Levy (1997), if an RTA produces disproportionately large gains and relatively small losses to the median voter so that his utility is raised above what could be achieved with a multilateral deal, multilateral liberalization will no longer be viable. Krishna (1998) argues that trade-diverting RTAs generate large rents tied to the preferences granted by the agreement for producers. Multilateral trade liberalization threatens those rents. If governments are swayed more by producer interests, then multilateral liberalization will not be pursued. Moreover, the attention that governments invest in RTA negotiations draws away scarce political and human resources from multilateral negotiations. There are strong arguments for the building bloc story as well. Baldwin (1995) has proposed a domino theory of regionalism where the establishment of an RTA increases the value for non-members of joining the agreement.2 The creation of a preferential regional arrangement will reduce the profits of the firms exporting to the region but who are located in a non-member country. They will have a reason to lobby their government to join the bloc. If the regional bloc enlarges as a consequence, the value of membership for outsiders increases since they face a cost disadvantage in an even greater number of markets. This leads to a snowballing of countries which want to join so that the RTA progressively enlarges its membership until global free trade is reached. Another building-bloc 2
However, the domino theory does not explain what incentives existing members have to accept new members.
4
antoni estevadeordal, kati suominen , robert teh
argument is that preferential trade liberalization will help enlarge the exporting sectors and diminish the import-competing sectors in RTA members. Thus a country which enters into an RTA will expand the economic and political strength of its pro-liberalization constituency (‘juggernaut’ effect), making it possible for its government to cut a multilateral deal (Baldwin, 2005). There are also those who see RTAs as a stepping stone towards a global free trade policy (Ethier, 1998). RTAs may help a government intent on carrying out economic reforms to mobilize domestic forces in support of opening up to the wider world. By initially entering into a preferential trade arrangement, the reforming country would be able to capture economic benefits, for example through FDI inflows from its RTA partners that tilt the political balance within the country in favour of economic reform and multilateral liberalization. A more recent vein of research has argued that, as more RTAs are established, the cost to producers of overlapping rules would lead them to pressure governments to harmonize or ‘multilateralize’ these rules (Baldwin, 2006). As bilateral and regional trade agreements proliferate, a ‘spaghetti bowl’ of rules of origin will emerge. This, in turn, will run against the increasing fragmentation of production as firms find it more cost-efficient to locate the manufacturing of parts and components in different countries. The requirement to comply with different rules of origin will then raise firms’ production costs. Paradoxically, RTA spaghetti bowls can become building blocs to multilateralism as offshoring becomes a force for the multilateralization of existing regional rules. Indeed, there are nascent efforts by some groups of countries in Asia as well as the Americas to examine ways to connect their common RTAs into broader trade areas so as to reduce the complexity of rules facing economic actors in the RTAs and to facilitate more trade and investment. To be sure, the multilateralization of regional rules may for many still seem to be a long-term aspiration. But what is clear today is that, given that nearly all WTO Members are RTA members and vice versa, WTO Members should have an interest in reducing conflicts between regional and multilateral rules and in ensuring compatibilities between them. Yet, despite the growing academic and policy attention to regionalism, the anatomy of RTAs remains poorly understood. Virtually all of the existing mappings that have been undertaken on RTAs focus on a single RTA discipline – rules of origin.3 The lack of a comparative look at 3
A partial list of the literature includes Estevadeordal (2000), Suominen (2004), Estevadeordal and Suominen (2005) and Cadot et al. (2006).
introduction
5
other RTA rules severely limits our understanding of the effects of RTAs and provides little foundation for recommending measures to further compatibilities between regional and global rules. We therefore take up Richard Baldwin’s recommendation in this volume to move the economic profession’s discussion from high theory to one which is more empirically grounded and policy-relevant.
2
Objectives and analytical approach
The main objective of this volume is to begin filling the gaps in our knowledge of RTA rules and their relationship to multilateral trade rules. The ultimate goal of our endeavour is to provide a firmer basis for informed policy debate and policy-making on RTAs. We seek to accomplish these objectives by developing detailed analytical mappings of regional rules in six key areas – market access, trade remedies, technical barriers to trade, services, investment, and competition policy – across dozens of the main RTAs around the world. The choice of rules to include in the mapping exercise has been dictated by a number of considerations. Our primary focus was regional rules for which there are corresponding global rules or for which a global rule is under negotiation in the Doha Round. However, we also allowed for the inclusion of regional rules that are applied in a substantial number of RTAs even though they have no multilateral counterpart (which is the case for competition policy). We feel that the choice of these six areas also strikes the right balance between the ‘traditional’ areas covered in trade agreements, such as market access and trade remedies, and the new but growing areas such as TBTs, services, investment and competition. Beyond addressing the lacuna in our knowledge of RTAs, we seek to meet three other objectives with this book. First, we hope to inspire and inform further work on disaggregating RTAs into their component parts, a task that is absolutely crucial for understanding the implications of the rising tide of regionalism on the global economic system. Secondly, by developing methodologies for dissecting RTAs, we hope to establish a rigorous starting point for further studies, as well as to provoke debate on the best prisms to examine RTAs. Thirdly, by virtue of containing detailed data on RTAs around the world, the chapters here hope to serve as inputs for fresh empirical work on the economic effects of RTAs.
6
antoni estevadeordal, kati suominen , robert teh
3
Organization and main findings
The second chapter of this volume, by Richard Baldwin, provides an analytic framework for examining the building bloc–stumbling bloc debate. The chapter discusses various economic mechanisms (Smith’s certitude, Haberler’s spillover and Viner’s ambiguity) that help determine whether preferential trade arrangements help or hinder multilateral trade liberalization. Baldwin critically reviews the long line of literature on both sides of the debate and identifies four distinct types of stumbling blocs in the literature – preference-erosion, goodies-bag, cherry-picking and the bargaining-model stumbling blocs. The building-bloc effects include the juggernaut effect, the Kemp–Wan effect, and the veto-avoidance mechanism. He then details the economic arguments that underlie these various effects. Baldwin’s main conclusion is that, while many of the models provide helpful theoretical frameworks for assessing the potential effects of RTAs on the global trading system, it is now time to move the literature’s focus from high theory to empirically grounded research which will have more policy relevance. The third chapter begins the mapping exercise. The focus of Antoni Estevadeordal, Matthew Shearer and Kati Suominen is on market access disciplines in RTAs – both tariff liberalization and a number of other disciplines that can qualify the extent of market access in RTAs, including non-tariff measures, special regimes, rules of origin, and customs procedures and trade facilitation. The study goes to the heart of GATT Article XXIV, which sets out the conditions under which the main types of RTAs – free trade agreements (FTAs) and customs unions (CUs) – are viewed as consistent with multilateral trade rules. The Article has been a source of extensive debate and interpretations. Some of the most disputed issues centre on the Article’s stipulation that RTAs are to eliminate tariffs on ‘substantially all trade’ (SAT) between the parties, and to do so within a ‘reasonable length of time’. Another key line of debate involves the meaning of the Article’s requirement that, besides tariffs, RTAs eliminate ‘other restrictive regulations of commerce’ on substantially all trade. Employing both tariff-line-level and aggregate data, the authors strive to capture the extent to which RTAs meet the Article XXIV benchmarks. There are three main findings. First, most RTAs attain a common interpretation of SAT and ‘reasonable length of time’ – liberalization of 90 per cent of tariff lines by year 10 of the agreement. Trade-weighted measures of the depth of liberalization yield similar results.
introduction
7
However, and secondly, there are a number of outlier RTA parties (in general, developing countries) and product categories (particularly sensitive sectors – agriculture, textile and apparel, and footwear) that do not attain the benchmark. Many an RTA also contains provisions that could potentially be classified as ‘other restrictive regulations of commerce’, such as tariff rate quotas, special safeguards, and demanding rules of origin. Thirdly, in terms of the aggregate provisions, many RTAs are also ‘WTO-plus’, in terms of incorporating a larger number and/or more specific provisions than are currently in force at the multilateral level. One key example is customs procedures and trade facilitation where US agreements in particular establish quite comprehensive and specific commitments. Overall, the findings are encouraging as to the extent of liberalization provided in RTAs and the potential for RTAs to serve as testing grounds for new, more comprehensive trade rules than have thus far been crafted at the multilateral level. The results could also provide insights and guidance for any future efforts to define ‘substantially all trade’ and ‘reasonable length of time’ in a more precise fashion at the multilateral level – as well as for negotiators of new RTAs to meet and go beyond the liberalization attained in past agreements. To be sure, the discriminatory potential of the RTAs remains, and must be attenuated with simultaneous unilateral and multilateral liberalization. In chapter four, Robert Teh, Thomas Prusa and Michele Budetta examine provisions on trade remedies – anti-dumping, countervailing and safeguard measures – in RTAs. Their main concern is with the possible increase in discrimination against non-members that come from regional rules on trade remedies. The elastic and selective nature of trade remedies may lead to more discrimination against non-members through greater frequency of trade remedy actions against them. The adoption of RTA-specific trade remedy rules can increase this risk of discrimination, with trade remedies against RTA members being abolished outright or being subjected to greater discipline. Given the second-best nature of preferential liberalization, any increase in intraregional trade that this brings about may simply be substituting for cheaper sources of imports from non-members. They find that about one-sixth of the RTAs surveyed have dispensed with at least one type of trade remedy. In addition, a number of RTAs have adopted RTA-specific rules that tightened discipline on the application of these remedies against RTA members. In the case of
8
antoni estevadeordal, kati suominen , robert teh
anti-dumping for example, some specific provisions have increased de minimis volume and dumping margin requirements, adopted a lesser duty rule and shortened the duration for applying anti-dumping duties. They have also pointed to the role of regional bodies with the authority to review or remand determinations made by national authorities in possibly reducing anti-dumping action against RTA partners. In the case of safeguards, they have expressed some concern about the exclusion of RTA partners in safeguard actions triggered by GATT Article XIX and the Agreement on Safeguards. This puts RTA rules on safeguards in conflict with the non-discriminatory principle that underlies multilateral rules on safeguard action and squarely raises the problem of trade diversion. Although WTO panels have ruled against such exclusions so far, it is not clear that future panels will do so consistently given the particular ground of parallelism on which previous decisions have been made. There appears to be less of a problem with countervailing duty (CVD) provisions in RTAs: no major changes have been made to CVD rules in the RTAs included in their survey. They suspect that a major reason for this is the absence of agreements in RTAs on meaningful or significant curbs on subsidies or state aid. Chapter five, by Roberta Piermartini and Michele Budetta, analyzes the variety of approaches that have been adopted at the regional level to remove technical barriers to trade (TBTs). The chapter attempts to answer whether RTA provisions have gone deeper than the Agreement on Technical Barriers to Trade (TBT Agreement) of the WTO in liberalizing TBTs and what factors determine the design of TBT rules at the regional level. The chapter develops a template that follows the structure of the WTO TBT Agreement to map rules on technical barriers to trade in regional trade agreements. The authors find that provisions on standards, technical regulations and conformity assessment procedures are widespread across RTAs. Overall, regional agreements tend to favour harmonization over mutual recognition of product standards, while equivalence and mutual recognition appear to be the preferred options to deal with TBTs arising from testing and certification. Their analysis seems to confirm theoretical predictions that similarity in the levels of development affects the likelihood of introducing provisions for mutual recognition of product standards and technical regulations, but there does not appear to be a link between the requirement to harmonize standards and the level of development. In general, regional rules on TBTs develop according to a hub-and-spoke
introduction
9
structure. In particular, the family of RTAs which involves the EU tends to promote the use of European standards. A common feature of RTAs signed by the US is the inclusion of provisions for the establishment of institutions to deal with the administration of the TBT chapter of the agreement, the resolution of disputes on TBT matters and encouragement of mutual recognition. One important question that the authors raise is whether, because of a ‘cloning’ tendency on the part of the hubs, standards become a barrier to trade between major regional groupings. To the extent that TBT provisions in RTAs succeed in locking in a country to ‘regional’ standards, RTAs act as a stumbling bloc in the process of multilateral liberalization. Trade in services makes up a substantial proportion of world trade. In 2006, commercial services exports grew by 12 per cent in nominal terms to US$2.76 trillion – accentuating the salience of both RTA and GATS services rules.4 Chapter six, by Martin Roy, Juan Marchetti and Aik Hoe Lim, analyzes services liberalization commitments in RTAs, and compares them to prevailing GATS commitments and Doha Round offers. Focusing on market access achieved in both bilateral and multilateral negotiations thus far, their main finding is that services commitments in RTAs tend to go significantly beyond GATS schedules and even Doha Round offers. These advances take the form of a high proportion of new bindings in sectors that had remained uncommitted in the GATS and improved bindings in sectors that were already committed in the GATS schedules/offers. The authors reach a number of further conclusions about the pattern of services liberalization in RTAs. First, countries that have used negative-list approaches have bound at least the existing level of openness for the large majority of sectors. This instils predictability in the bilateral relationship and spurs cross-border investment and trade. The authors also highlight that a number of agreements have led to ‘real’ liberalization on the ground. Secondly, a number of the larger developed countries tend not to go as far beyond GATS as many of the smaller developing economies. The authors argue that this might be explained by an imbalance between negotiating partners and by the fact that at least some of the larger developed countries have less room to improve their already extensive GATS schedule and offers. Thirdly, as a result, the most protected services activities in larger developed countries remain largely unaffected by RTAs, for example audiovisual for EFTA and the EC, maritime transport 4
WTO (2007).
10 antoni estevadeordal, kati suominen , robert teh
and certain professional services for the US, cross-border trade in a number of financial services for a variety of countries, and education services where there has been no significant improvement for the US, EC and EFTA Member States. Preferential trade in services has its benefits and costs. On the one hand, preferential access in services may be less costly than in merchandise goods. One reason the authors provide is that barriers to trade in services are usually embedded in regulations and it may be difficult for governments to devise and enforce one set of regulations for some service suppliers and another set of regulations for a different group. But preferential access can also engender important costs: non-parties may suffer because preferences may provide lasting advantages to first movers that might be hard to reverse through subsequent extension of access to other countries; while parties to the agreement may also suffer if they are stuck with relatively less efficient suppliers. Additionally, from a political economy perspective, bilateral negotiations may have a deleterious impact on multilateral talks. For one, bilateral and regional initiatives divert important resources from multilateral negotiations. Moreover, the proliferation of RTA negotiations may well have led some WTO Members to make minimal WTO services offers in the Doha Round so as to have ‘negotiating chips’ to trade in RTA negotiations. The authors suggest two ways in which multilateral initiatives can overcome these drawbacks and help harness regionalism to reinforce multilateralism. First, countries involved in RTAs could conditionally offer a level of services commitments in the WTO closer to the one they agreed to in RTAs. Those WTO Members that have not been involved so far in RTAs could use the GATS-plus commitments in RTAs as a kind of target they could aim for in the Doha Round. Secondly, the authors propose better multilateral surveillance of the implementation of the services commitments in RTAs. They envision this surveillance to be modelled along the lines of China’s transitional review mechanism in the WTO, according to which China has an obligation to provide information on policies affecting trade in services (e.g. changes to laws and regulations, state of play of licensing applications), information which is subsequently reviewed by Members in the specialized WTO bodies overseeing the issues at hand. The same obligation could be imposed on WTO Members having signed RTAs including services obligations and commitments. A substantial body of academic literature agrees that trade policy and investment tend to influence each other: high tariff walls may encourage
introduction
11
market-seeking investment that ‘jumps’ the tariff while trade liberalization can attract investment that relies on availability of cheap intermediate goods and access to foreign markets. In chapter seven, Barbara Kotschwar analyzes investment chapters in fifty-two RTAs, finding that most cover such areas as MFN treatment, national treatment, transparency, denial of benefits and restriction of transfers, nationality of management and board of directors, performance requirements, expropriation, and investor–state dispute settlement. She finds that investment provisions in US RTAs are particularly comprehensive and often extend well beyond what can be found in multilateral instruments regulating investment such as the Agreement on Trade-Related Investment Measures (TRIMs) and the General Agreement on Trade in Services (GATS), and, less directly, in the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) and the Agreement on Subsidies and Countervailing Measures (SCM). Kotschwar discovers two main phenomena: the increasing commonalities in investment provisions at the bilateral and regional levels, and the importance of flexibility in the forging of investment provisions. Common approaches have emerged in RTAs with respect to the scope of application, treatment, transfers, expropriation and mechanisms for the settlement of disputes. At the same time, regional arrangements are allowing countries to customize their investment relations in terms of the depth of liberalization, the extent of protection, and demands by developing countries to address the development dimensions of FDI. Whereas increased commonalities could be a harbinger of an eventual multilateral investment regime, widespread exercise of flexibility may ensure that investment provisions at the bilateral and RTA level end up driving the evolution of the international investment regime so that a single, comprehensive and coordinated regime is unlikely in the near future. Nonetheless, the commonly shared elements could serve as conduits between the various RTAs – as well as between the RTAs and the multilateral system. One complication affecting investment rules that Kotschwar identifies is their multilayered nature: countries regulate investment not only through RTAs and multilateral instruments, but also through bilateral investment treaties (BITs). Ensuring complementarities between RTAs, BITs, TRIMs and the other multilateral investment rules is likely to pose one of the more salient policy challenges in the years ahead. A growing number of RTAs, many of them involving developing countries, include competition policy and competition-related provisions. While there have been a number of surveys of competition
12 antoni estevadeordal, kati suominen , robert teh
provisions in RTAs in recent years, they have tended to dwell only on the competition policy chapter of regional agreements. The principal contribution of chapter eight is to go beyond this and to examine competition-related disciplines and principles that are found in other chapters of RTAs. Robert Teh shows how extensive the scope of competition rules is in RTAs. There are competition disciplines to be found in the RTA chapters on investment, services (in telecommunications, maritime transport and financial services), government procurement and intellectual property. Between a third and a half of the RTAs that were studied in the chapter have horizontal provisions cutting across the entire agreement on transparency, procedural fairness and nondiscrimination. These findings underline the importance of examining competition-related provisions beyond those contained in the competition policy chapter. Within the competition policy chapter proper, the main obligations involve application of competition laws to curb anti-competitive behaviour and closer co-operation among competition authorities of RTA partners. The types of anti-competitive behaviour that are of concern are concerted actions, abuse of a dominant position, monopoly, state enterprises or undertakings with special or exclusive rights and state aid. For the most part, the scope of co-operation among national competition authorities is limited to exchange of information, notification and consultation. However, a small number of RTAs give a substantial role to regional bodies in carrying out surveillance, investigations and taking measures to curb anti-competitive behaviour. One of the findings of this chapter is that a significant number of RTAs carve out competition policy, either partially or completely, from dispute settlement. This further underscores the importance of the competitionrelated provisions in the sectoral chapters and the horizontal principles, since for the RTAs with carve-outs, they are the only provisions on competition that could be enforced through the RTA’s disputesettlement mechanism. The chapter goes on to argue that many elements of competition rules in RTAs are non-discriminatory in that their benefits cannot be denied to non-RTA members. This applies in great measure to the horizontal principles of transparency, procedural fairness and non-discrimination. But, beyond these principles, the requirement to apply competition law, or to set up a competition authority, or to subject state aid and public monopolies to disciplines, all improve the conditions of competition in the marketplace, and should enhance economic opportunities to firms
introduction
13
who already operate in the market, whether they are domestic, from an RTA member or from a country which is not a party to the regional agreement.
4
Conclusion
There are three policy-related lessons that arise from this volume. First, RTAs are here to stay. Countries have become comfortable in walking with two feet. Regionalism and multilateralism are being pursued simultaneously by many countries that apparently see no major conflict in the pursuit of both policies. Given their proliferation globally and with rules governing a host of policy areas, they cannot be ignored at the multilateral level. One key policy objective must be to harness RTAs as an instrument to further multilateral liberalization. Secondly, regional rules and their economic impacts are complex, and these findings are reflected in the analysis in many of the chapters in this volume. The trade remedies chapter highlights the discriminatory impact of what appears to be trade-friendly elimination of such measures. The TBT chapter points to the pitfall of harmonized product standards that may lock RTA members into a particular regulatory framework that is costly to escape from. On the other hand, the competition chapter notes that a significant number of RTA rules on competition create benefits that are non-discriminatory and that extend beyond firms from RTA members. An almost similar finding is made in the services chapter. Since many services restrictions are embedded in regulatory regimes, governments will often not put in place different regulatory regimes for different supplier countries. Thus once some restrictions on trade in services are removed for RTA members, they will not continue to be applied to other countries. But the services chapter also warns about the possibility of lasting advantages to first-movers that might be hard to reverse through subsequent extension of access to other countries. The third lesson is that RTAs can have far-reaching implications for domestic politics – they create coalitions and vested interests as any policy of trade liberalization does – as well for national policies. RTAs prescribe new rules, regulations and institutions, such as domestic bodies to implement RTA provisions. RTAs thus both regulate trade and create the means to regulate. The key lesson for any signatory is that the negotiation of RTAs is only the beginning of a long process of implementing and investing in the implementation of numerous RTA provisions while at the same time seeking to ensure compatibilities
14 antoni estevadeordal, kati suominen , robert teh
between a country’s old and new RTAs as well as between RTAs and multilateral commitments. It is our hope that this volume will inspire further research on regionalism. It is the task of empirical analysts to harness and hone the data assembled in this volume to better understand the full range of RTA effects. There are four key areas of inquiry where greater knowledge is needed. The first is the political economy of deep v. shallow integration or comprehensive v. limited integration. What determines the choice between these options in RTAs? Why are countries increasingly negotiating and signing RTAs containing behind-the-border disciplines such as competition policy or technical barriers to trade? Is the purpose to limit the liberalization provided by the market access chapter, to expand it or to guard against nullification of market access through other measures? A broader question that this refers to is the complementarities between the various provisions within RTAs. Do tariff and investment provisions foster each other or work at cross-purposes? Are services trade provisions complementary to provisions on competition policy? Secondly, as pointed out by Baldwin, why have developing nations agreed to disciplines in RTAs that they have tended to resist at the multilateral level, such as intellectual property rights, tariff bindings, TBTs, services, government procurement and competition policy? Is this simply a matter of sequencing? Or do political-economy considerations make it easier to advance on these issues with a handful of partners than with the entire WTO membership? To what extent does this result hinge on the uneven negotiating positions of developing and developed county members in an RTA? Or is this a matter of modalities – should multilateral talks be structured in a different fashion to provide momentum for negotiations of these disciplines? Thirdly, do any of the differences identified in the mappings between RTAs matter for economic decisions, and do they have real economic consequences? Does the RTA spaghetti bowl impose costs on producers and traders? And, if so, how could these differences between RTAs be bridged so as to facilitate trade and investment between them? And what would connecting RTAs mean to the global trading system? Fourthly, how could the interplay between multilateralism and regionalism be made more constructive? What should be done at the national level to build synergies between a country’s RTA and WTO commitments? How could multilateral disciplines regulating RTAs be improved? How can multilateral reporting and oversight of RTAs be
introduction
15
strengthened? How might best practices and lessons learned from RTAs be better transposed to the multilateral level? At the regional level, what can be done by parties embarking on these arrangements to ensure compatibility with global trade rules? How should groups of countries interested in connecting their various RTAs devise such broader integration areas so as to best facilitate economic exchange, both among each other and with the rest of the world? And, finally, how best to employ RTAs as a means to arrive at global trade liberalization? References Baldwin, Richard E., 1995. ‘A Domino Theory of Regionalism’, in Richard E. Baldwin, P. Haaparanta and J. Kiander (eds.), Expanding Membership of the EU, Cambridge: Cambridge University Press, pp. 25–48. 2005. ‘Stepping Stones or Building Blocs? Regional and Multilateral integration’, in Julie McKay, Maria O. Armengol and Georges Pineau (eds.), Regional Economic Integration in a Global Framework, Frankfurt: European Central Bank, pp. 113–34. 2006. ‘Multilateralising Regionalism: Spaghetti Bowls as Building Blocs on the Path to Global Free Trade’, The World Economy 29: 1451–518. Bhagwati, Jagdish, 1991. The World Trading System at Risk, Princeton University and Harvester Wheatsheaf. Bhagwati, Jagdish, and Panagariya, Arvind, 1999. ‘Preferential Trading Areas and Multilateralism’, in Jagdish Bhagwati, Pravin Krishna and Arvind Panagariya (eds.), Trading Blocs, MIT Press, pp. 33–100. Cadot, Olivier, Estevadeordal, Antoni, Suwa-Eisenmann, Akiko, and Verdier, Thierry (eds.), 2006. The Origin of Goods: Rules of Origin in Regional Trade Agreements, Oxford: Oxford University Press. Ethier, Wilfred J., 1998. ‘Regionalism in a Multilateral World’, Journal of Political Economy 106: 1214–45. Estevadeordal, Antoni, 2000. ‘Negotiating Market Access in the Americas: The Case of NAFTA’, Journal of World Trade 34: 141–66. Estevadeordal, Antoni, and Suominen, Kati, 2005. ‘Rules of Origin in Preferential Trading Agreements: Is All Well with the FTA Spaghetti Bowl in the Americas?’, Economı´a 5: 63–103. Fiorentino, Roberto V., Verdeja, Luis, and Toqueboeuf, Christelle, 2006. ‘The Changing Landscape of Regional Trade Agreements: 2006 Update’, World Trade Organization Discussion Paper No. 12. World Trade Organization. Krishna, Pravin, 1998. ‘Regionalism and Multilateralism: A Political Economy Approach’, Quarterly Journal of Economics 113: 227–51. Levy, Philip I., 1997. ‘A Political Economic Analysis of Free-Trade Agreements’, American Economic Review 87: 506–19.
16 antoni estevadeordal, kati suominen , robert teh Lima˜o, Nuno, 2007. ‘Are Preferential Trade Agreements with Non-Trade Objectives a Stumbling Block for Multilateral Liberalization?’, Review of Economic Studies 74: 821–55. Suominen, Kati, 2004. ‘Rules of Origin in Global Commerce’, PhD Dissertation, University of California, San Diego. World Trade Organization, 2007. International Trade Statistics 2007, World Trade Organization.
2 Big-Think Regionalism: a critical survey richard baldwin* 1
Introduction
In the late 1940s and 1950s, the profession’s best and brightest minds were focused on regionalism: Jacob Viner, James Meade, Richard Lipsey, Harry Johnson and Max Cordon, inter alia. The reason was simple. Europe’s post-war architecture was one of the world’s greatest problems and a free trade area was to be part of it, economists were muddled over the issue. The thinking of the 1950s straightened out the economics and established the intellectual paradigm that dominated the regionalism literature right up to 1991.1 The paradigm was framed around the Vinerian question: ‘Would a nation gain from joining various preferential trade configurations?’, This literature – what could be called Small-Think Regionalism – ignored systemic implications since the only large preferential arrangement – the EEC – was viewed as sui generis. All this changed in the late 1980s when large-scale regionalism was reignited in North America. Regionalism seemed to sweep the world trading system like wildfire while the multilateral GATT talks proceed at a glacial pace. Then GATT negotiations slipped into a four-year coma in December 1990 with the acrimonious failure of the summit that was supposed to have wrapped up the Uruguay Round. Regionalism seemed to threaten the multilateral trade system. Once again, regionalism attracted attention from the profession’s leading lights. In 1991, Paul Krugman, Larry Summers and Jagdish Bhagwati laid out lines of analysis – what might be called Big-Think Regionalism – that continue to guide the profession’s thinking on regionalism even
*
1
I would like to thank the following scholars for comments and suggestions: Caroline Freund, Antoni Estevadeordal, Kati Suominen, Nuno Lima˜o, Richard Pomfret, Alan Winters, Patrick Low and all the economists in the WTO Research Department. Particularly, Viner (1950), Meade (1955a, 1995b), Lipsey (1957), Johnson (1957, 1958a, 1958b) and Corden (1957).
17
18
richard baldwin
today.2 While Small-Think Regionalism focused on the welfare of the individual nation, Big-Think Regionalism focuses on the systemic implications. It comprises two principal lines of inquiry: 1. Does spreading regionalism harm world welfare? Krugman (1991a) provided a line-sketch model that crystallized the profession’s thinking around a simple comparative static exercise: ‘Will an exogenous variation in the number of regional trade blocs raise or lower world welfare?’ He admitted, however, that this exercise missed much: In a fundamental sense, the issue of the desirability of free trade areas is a question of political economy rather than of economics proper . . . The real objection is . . . the fear that regional deals will undermine the delicate balance of interests that supports the GATT.3
This lead to the second line of inquiry: 2. Does regionalism help or hinder multilateralism? Krugman (1991b, 1993) sketches a bargaining model where regionalism can help or hinder multilateralism, but it was Jagdish Bhagwati’s bon mot that organized the profession’s thinking: do regional trade blocs ‘more readily serve as building blocks of, rather than stumbling blocks to, GATT-wide free trade’ (Bhagwati 1991 p. 77)? Specifically, the second line of inquiry crystallized around the question of whether an exogenous variation in regional trade blocs made multilateral tariff cooperation more or less likely. Summers’ (1991) contribution was to stake out one extreme of the debate on both lines of inquiry. He argued that ‘plausible’ regional arrangements were natural trade blocs and thus would raise world welfare. He also asserted that reasonable regional arrangements were as likely to accelerate the general liberalization process as to slow it down. Hence, his famous assertion that all the ‘-isms’ are good: unilateralism, bilateralism, plurilateralism and multilateralism. Summers rejected the notation that regionalism and multilateralism were enemies as Krugman, Bhagwati and many suspected; regionalism and multilateralism were the two legs on which the world was walking towards global free trade. The two main lines of inquiry embraced the notion that one could reasonably view changes in the number of blocs as exogenous. 2
3
See, particularly, Krugman (1991a, 1991b), Bhagwati (1991 Chapter 5), Summers (1991), but also Krugman (1993) and Bhagwati (1993). Krugman (1991b) p. 23.
big-thin k regionalism: a critical survey
19
A third line of inquiry extended the issue by endogenizing regionalism’s spread. The focus was on cause and extent of spreading regionalism; it turned on the positive political-economy question of which trade blocs would actually emerge (Baldwin 1993b, 1995, Grossman and Helpman 1995, Yi 1996, Freund 2000a, 2000b, Aghion, Antras and Helpman 2004). In 2008, the world finds itself again in a situation like the 1990s when regionalism is booming and multilateralism seems to be ailing. But this time, the nature of regionalism is quite different. The trade deals being signed are much broader and are often not even regional, as the other chapters in this volume demonstrate.
1.1
Plan of the paper
My goal is to summarize and evaluate the Big-Think Regionalism literature. My focus is almost exclusively on theoretical work to limit the length of this already-too-long paper. Before turning to the literature, section 2 presents the elemental economic effects that concern Big-Think Regionalism in a way that helps to fix ideas and terms. Section 3 considers the stumbling/building bloc issue and section 4 looks at the ‘Is bilateralism bad?’ literature. The third line of inquiry is covered in section 5. The summary and concluding remarks are in section 6.
2
Basic economic effects
Before launching into a review of Big-Think Regionalism, it proves useful to conduct a quick review of the basic economics of preferential trade liberalization as far as the political economy interaction between regionalism and multilateralism is concerned. There is nothing new in this review, and it ignores many elements that are important economically, for example scale effects, growth effects and location effects. It is necessary since the literature is marked by a conceptual ‘spaghetti bowl’ – a tangle of conflicting, overlapping and competing terminologies for basic effects. My sole aim here is to establish a common set of labels and notation for the key concepts underpinning thinking in the Big-Think Regionalism literature. There are only three core effects. All have been known at least since 1950. To avoid creating yet another set of terms, I label them according to their intellectual fathers.
20
richard baldwin
2.1 Smith’s certitude, Haberler’s spillover and Viner’s ambiguity The theory of preferential trade did not begin in 1950 with Jacob Viner.4 Early contributions include Smith (1776), Taussig (1892), and Torrens (1844). Adam Smith’s contribution highlights one of the most robust finding in the field – what might be called ‘Smith’s certitude’: When a nation ‘exempt[s] the good of one country from duties to which it subjects those of all other . . . the merchants and manufacturers of the country whose commerce is so favoured must necessary derive great advantage’.5
Much later, Gottfried Haberler asserted that all members of a preferential trade agreement must gain while third nations must lose.6 The first part of the assertion is wrong, but what might be called ‘Haberler’s spillover’ – the part about third nations losing – turns out to be almost as robust as Smith’s certitude. Haberler’s spillover and Smith’s certitude are the linchpins of the Big-Think Regionalism discussion. The only basic element added in the post-war period came with Jacob Viner’s famous 1950 book, The Customs Union Issue. Viner’s key finding is that discriminatory tariff liberalization has ambiguous welfare effects (‘Viner’s ambiguity’).7 Viner’s ambiguity is quite general, but one is hard pressed to see this from the analysis in his book. An RTA is nothing more than a special case of non-uniform commodity taxation, but Viner did not have the benefit of modern economic tools for tax analysis.
4
5 6
7
See the excellent survey, Pomfret (1997 Chapter 8.1), on pre-Vinerian contributions; also Pomfret (1986) and O’Brien (1976). Smith (1776) as quoted in Pomfret (1997). His discussion runs over several pages but here he asserts it fairly directly: ‘There is no difference in kind, but only one of degree, between the grant of lower preferential duties upon imports from certain country and a general reduction in tariffs. A partial reduction is better than none at all (although, of course, a general reduction would be still better, from an economic standpoint).’ Haberler (1936 p. 384). Haberler’s spillover was certainly understood by scholars before Haberler (e.g. Bismarck used this aspect of customs union to force/cajole many German-speaking states to join his unified Germany), but I assign it to Haberler since Haberler’s 1936 book shows that mainstream trade economists were confused about the theory of the second best. This illustrates why Viner’s 1950 book was viewed as such a landmark. Viner’s consideration of other effects of customs union formation (its impact on the terms of trade, economies of scale, cartels, administrative efficiency, the pressure to harmonize excise taxes, and the necessity to go beyond tariff removals in order to remove trade barriers) made much less lasting impact on the literature, but remains a fascinating and highly accessible read.
big-thin k regionalism: a critical survey
21
Rather, he conducted the analysis using the enduring but imprecise concepts of ‘trade diversion’ and ‘trade creation’. The analysis will be directed towards finding answers to the following questions: in so far as the establishment of the customs union results in change in the national locus of production of goods purchased, is the net change one of diversion of purchase to lower or higher money-cost sources of supply, abstracting from duty-elements in money costs . . . If the customs union is a movement in the direction of free trade, it must be predominantly a movement in the direction of goods being supplied from lower money-cost sources than before. If the customs union has the effect of diverting purchases to higher money-cost sources, it is then a device for making tariff protection more effective. None of these questions can be answered a priori, and the correct answers will depend on just how the customs union operates in practice.8
‘Trade diversion’ and ‘trade creation’ are misleading since they suggest trade volumes are the key even though his words clearly indicate that cost/price changes are what matter. Moreover, they fail to cover all the effects generated by discriminatory tariff liberalization – even in a simple Walrasian setting. Given these shortcomings and the decadeslong debate on ‘what did Viner really mean’ (a debate in which Viner himself participated without notable effect), it is curious that the terms have enjoyed such enduring success.9 The generality of Viner’s ambiguity is glaringly obvious to readers schooled in the ‘theory of the second best’ (preferential liberalization induces new distortions while removing others) but Viner’s book was a landmark. The ‘theory of the second best’ was unknown in 1950, and many of Viner’s contemporaries – Haberler, for example – were muddled over the key differences between general and preferential liberalization.
8
9
Viner (1950) p. 44. Viner (1950) is worth reading in the original. His informal reasoning is full of insights and it anticipates much of the economic and political-economy theory as well as the political-economy debates that have surrounded economic integration in the subsequent six decades. The key passages are reproduced verbatim in Box 2.1. The basic problem was that the profession found the simple trade creation/diversion paradigm to be effective in communicating the crucial welfare-ambiguity result but the words did not fully capture all the basic economic effects. Arvind Panagariya, for example, suggests that the terms persist since they are ‘highly effective tools of focusing policy-makers’ attention on the ambiguous welfare effects of RTAs’ (Panagariya 2000).
22
richard bald win
box 2.1 viner verbatim The relevant passages from Viner’s famous book, The Customs Union Issue, are relatively short and worth reading. (If nothing else, it gives students an idea of what economics was like before diagrams and maths became de rigueur.) The analysis will be directed towards finding answers to the following questions: in so far as the establishment of the customs union results in change in the national locus of production of goods purchased, is the net change one of diversion of purchase to lower or higher money-cost sources of supply, abstracting from duty-elements in money costs: (a) for each of the customs union country taken separately; (b) for the two combined; (c) for the outside world; (d) for the world as a whole? If the customs union is a movement in the direction of free trade, it must be predominantly a movement in the direction of goods being supplied from lower money-cost sources than before. If the customs union has the effect of diverting purchases to higher money-cost sources, it is then a device for making tariff protection more effective. None of these questions can be answered a priori, and the correct answers will depend on just how the customs union operates in practice. All that a priori analysis can do, is to demonstrate, within limits, how the customs union must operate if it is to have specific types of consequence. The removal of ‘nominal’ duties, or duties which are ineffective as barriers to trade, can be disregarded, and attention can be confined to the consequences of the removal, as the result of customs union, of duties which previously operated effectively as a barrier, partial or complete, to import. There will be commodities, however, which one of the members of the customs union will now newly import from the other but which it formerly did not import at all because the price of the protected Home product was lower than the price at any foreign source plus the duty. This shift in the locus of production as between the two countries is a shift from a high-cost to a lower-cost point, a shift which the free trader can properly approve, as at least a step in the right direction, even if universal free trade would divert production to a source with still lower costs. There will be other commodities which one of the members of the customs union will now newly import for the other whereas before the customs union it imported them from a third country, because that was the cheapest possible source of supply even after payment of duty. The shift in the locus of production is now not as between the two member countries but as between a low-cost third country and the other, high-cost, member country. This is a shift of the type which the protectionist approves, but it is not one which the free trader who understands the logic of his own doctrine can properly approve. Simplified as this exposition is, it appears to cover most of the basic economic issues involved. The primary purpose of a customs union, and its major consequence for good or bad, is to shift sources of supply and the shift can be either to lower- or to higher-cost sources, depending upon the circumstances.
big-thin k regionalism: a critical survey
23
box 2.1. (cont.) From the free trade point of view, whether a particular customs union is a move in the right or in the wrong direction depends, therefore, so far as the argument has as yet been carried, on which of the two types of consequences ensue for that customs union. Where the trade-creating force is predominant, one of the members at least must benefit, both may benefit, the two combined must have a net benefit, and the world at large benefits; but the outside world loses, in the short-run at least, and can gain in the long-run only as the results of the general diffusion of the increased prosperity of the customs union area. Where the trade-diverting effect is predominant, one at least of the member countries is bound to be injured, both may be injured, the combination will suffer a net injury, and there will be injury to the outside world and to the world at large. The question as to what presumptions can reasonably be held to prevail with respect to the relative importance in practice of the two types of effects will be examined subsequently.10 There is quite a bit of discontent among modern academic economists with the trade-creation-trade-diversion terminology, for example Kowalczyk 1990.
2.1.1 Kemp–Wan logic A fourth elemental effect in the regionalism literature concerns the interaction between preferential and multilateral tariff cutting. It is not really a basic economic effect but rather a specific combination of effects motivated by the fact that the most important regional liberalization over the last sixty years (Europe and North America) has been accompanied by multilateral liberalization. When thinking about this teaming of multilateral and regionalism liberalization, the guiding light is the Kemp–Wan logic. Meade (1955a) introduced analysis that produced one of the few general statements that can be made about RTAs – the Kemp–Wan theorem.11 Kemp–Wan (1976) demonstrate that RTAs could be 10 11
Viner (1950) pp. 42–4. Meade identified the basic result in 1955 when he argued that, were it not for external trade considerations (the PTA’s trade with non-members), duty-free internal trade would be optimal: ‘if all trade barriers take the form of fixed and unchanged quantitative restrictions, then a customs union must increase economic welfare’ (Meade 1955b p. 98). As often happened in ‘customs union theory’, the result was reinvented repeatedly (Vanek 1965, Ohyama 1972, and Kemp and Wan 1976). The profession knows it as the Kemp–Wan theorem. Recently, Krishna and Panagariya (2000) follow-up Meade’s
24
richard baldwin
designed to be Pareto improving for every member of the RTA and the world at large. The logic is elegant. Assume two nations sign an RTA and alter their external tariffs to freeze their external trade flows; the external trade flows can then be treated as part of the bloc’s endowment. Removal of all intra-RTA barriers thus shifts the twonation bloc from a second-best situation to a first-best situation (i.e. laissez-faire in goods and factors given tastes, technology and endowments). The first welfare theorem of Walrasian economics guarantees an increase in economic efficiency and lump-sum transfers within the RTA ensure welfare gains for all. Third nations are unaffected since their trade vectors are unaffected. Dixit and Norman (1980) generalize the analysis, showing that the Kemp–Wan improvement can be obtained without lump-sum transfers; intra-RTA commodity taxes and subsidies are sufficient. Of course, real-world RTAs do not adjust external tariffs in a Kemp– Wan manner, nor do they have access to large lump-sum transfers. Nevertheless, the theorem is important from a policy perspective. It proves that RTAs are not necessarily bad for world welfare. Moreover, it helps us think about why the duo of multilateral and preferential tariff cutting – a duo that has been in operation since the 1950s – may have been critical to explaining why post-WWII regionalism has had a relatively benign impact on the world trade system to date – certainly much more benign that the European regionalism between the wars.
2.2
Modern treatment of Viner’s ambiguity
The first economists to apply modern economic analysis to Viner’s question were Meade (1955a, 1955b), Lipsey (1957) and Gehrels (1956–7). Modern tax analysis shows that the welfare impact of any tax change is captured by two terms in the Walrasian setting – one related to the change in consumption over the tax wedge, the other related to the level of consumption times the change in the actual price paid. James Meade’s pioneering analysis applied this to import taxes (tariffs) where the two terms may be called the trade volume and trade price effects (the trade price effect is often called the terms-of-trade effect). For the nation imposing the tariff, the net welfare effects is related to the initial tariff insight that the key is to freeze external tariff vectors to show that Kemp–Wan holds for FTAs as well as for customs unions, if the FTA is free to choose all external tariffs.
big-thin k regionalism: a critical survey
25
wedges on bilateral trade, changes in bilateral imports, and the changes in bilateral border prices according to: Net Home welfare effect ¼ ðp p ÞdM ðMÞdp
ð1Þ
where p and p* are the vectors of internal and border prices, M is the vector of bilateral imports (exports are negative imports), and dM and dp* are the vector of changes in bilateral trade volumes and border prices, respectively. See Box 2.2 for a derivation of this expression in a simple linear case, and the Annex to this chapter for a more general demonstration that allows for a variety of other effects (pro-competitive effects, scale effects, location effects and accumulation effects) in a more general economic setting.
box 2.2 walrasian net welfare effects: linear example The Annex to this chapter works through the general derivation of the modern analysis of a marginal tariff liberalization. This box works through a linear example that makes the reasoning more transparent and directly connects it to the standard undergraduate analysis involving consumer surplus, producer surplus and tariff revenue. It also specifies the translation from marginal tax analysis into discrete tariff changes. We work in a three-nation Walrasian setting with tariffs as the only distortion. We assume Home imports from both Partner and RoW nations and initially apply the same specific tariff T to imports from both sources. Assuming a linear demand function, p ¼ 1 – C, where C is Home consumption and p is the Home’s internal price, and a linear supply function, p ¼ b þ Q, where Q is Home production, the sum of consumer surplus, producer surplus and tariff revenue is: W ¼ ða pÞ=2 þ ðp bÞðQ=2Þ þ ðp p ÞðC QÞ where p* is the common border price and the three terms are, respectively, consumer surplus, producer surplus and tariff revenue (C – Q equals imports). After the tariff liberalization, welfare can be written as: W' ¼ ða p'Þ=2 þ p'ðQ'=2Þ þ ðp' p 'ÞðC' Q'Þ where the ‘primes’ indicate post-liberalization quantities and prices. Subtracting the post-liberalization welfare, W’, from the pre-liberalization W, and rearranging: W ¼ ðp' pÞðM' MÞ=2 ðp ' p ÞM where M ¼ C – Q, and M' ¼ C' – Q' . Intuitively, the traditional focus on consumer surplus, producer surplus and tariff revenue is unnecessarily cumbersome since the
26
richard baldwin
box 2.2. (cont.) incidence of the tariff can always be decomposed into a part that falls on domestic agents (consumers) and foreign agents (foreign exporters). That is, the incidence of the tariff on consumers is an internal transfer; only the incidence on foreigners (via change in the border price) matters for net welfare.
2.2.1
Viner’s special case: adding up created and diverted trade An antiquated but enduring rule-of-thumb for evaluating RTAs turns on the volume of trade created and diverted. We can use Meadean analysis to show exactly what is being assumed away. Assuming tariffs are the only barriers (no export taxes or subsidies so p – p* equals the vector of bilateral tariffs, T) and ignoring changes in all border prices (so Mdp* is zero), the welfare effect boils down to T times dM. Further assuming that tariffs are identical on all imports, the net welfare effect is proportional to the sum of changes in imports. In other words, this says an RTA member gains if the RTA creates more trade than it diverts. A slightly more general test – one that allows for different bilateral tariffs – is to check the change in tariff revenue collected. Of course, this test ignores Smith’s certitude and Haberler’s spillover, but it was the best economists could do in the 1950s and 1960s without computers. 2.2.2 Meade’s primary, secondary and tertiary effects Meade (1955b) described the trade volume and border price effects as the primary effects, but he listed two other categories of effects. Meade’s secondary effects are the substitution and income effects of a tariff change on other markets. His tertiary effect concerns general equilibrium adjustments necessary to ensure the balance of payment. 2.3
Illustration of basic economic effects
Smith’s certitude, Haberler’s spillover and Viner’s ambiguity capture most of the basic economics of RTAs, and, together with the Kemp– Wan logic, most of the political-economy reasoning in the Big-Think Regionalism literature.12 It is possible to deal with these mathematically, 12
Of course, when considering the full economic impact, one must consider scale economies, pro-competitive effects, variety effects, location effects and growth effects
big-thin k regionalism: a critical survey
27
Good 1 Good 2
Home
Partner
Good 2
Good 1
Good 3
Good 3 RoW
Figure 2.1
The RTA diagram’s trade pattern
however, to demonstrate the basic interactions among the elemental effects, and, to facilitate the exposition of the logic of the Big-Think Regionalism literature in the sequel, it is useful to have an amenable and flexible analytic framework, especially one that lends itself to graphical analysis. The simplest framework that meets these requirements is a Walrasian three-nation model (Home, Partner and RoW) with three goods (goods 1, 2 and 3); each nation exports two goods and imports the other good. Since each nation has two sources of imports, tariff discrimination can be a real issue in all markets. To rule out Meade’s secondary effects, tastes are assumed to be identical across nations and additively separable in all goods.13 For simplicity’s sake, the three nations are symmetric in terms of size and the MFN tariff they initially impose.14 The two trading equilibria (regionalism versus multilateral free trade) in a typical market (good 1) can be worked out with the help of the ‘RTA Diagram’. The analyses for imports of goods 2 (into Partner) and 3 (into RoW) are isomorphic due to the strong symmetry. Figure 2.2 shows the export supply curves (marked XS with the appropriate superscript to indicate the origin nation) for Home’s two
13
14
(Baldwin and Venables 1995). However, most of these are not critical in the Big Think Regionalism literature. The reader can mentally insert a fourth untaxed good that enters the utility function linearly to formally eliminate Meade’s tertiary effects. The RTA diagram can be thought of as a modification of the Blackhurst (1972) diagram.
28
richard baldwin
potential suppliers (the two left-most panels). The horizontal sum of the XS curves is shown in the right-most panel (as MSFT) along with Home’s import demand curve, MD. Under global free trade, the domestic and border price is PFT as shown in all nations for all goods. Assuming all nations impose a specific tariff T on an MFN basis, the internal price in Home is driven up to P while the border price is driven down to P – T for both suppliers. Home imports drop, with the reduction divided equally among the two suppliers.
2.3.1 From MFN tariffs to FTA If an FTA or customs union is formed between Home and Partner, the total import supply curve becomes the kinked MSFTA curve. The resulting internal price falls to P' , but there are now two border prices. The FTA raises the price facing Partner exporters from P – T to P’, while it lowers the RoW border price from P – T to P' – T. Partner exports expand while RoW exports contract. Identical things happen in the market for good 2, but here Home is the exporter and Partner the importer. Nothing happens in the market for good 3 (where RoW is the importer) since RoW maintains its MFN tariff and the strong separability assumptions rule out Meade’s secondary and tertiary effects. We see Smith’s certitude and Haberler’s spillover immediately in Figure 2.2. Smith’s certitude shows up as Partner gains (the areas a þ b) that result from the higher exports and higher border price. Since the FTA is reciprocal and nations are symmetric, Home gains the same from a higher border price and greater exports to Partner in good 2. Haberler’s spillover shows up as the RoW lose (the area ‘e’) from the drop in the border price it faces (from P – T to P' – T) and the reduction of its exports to Home. 2.3.2 The preference rent A critical observation, as far as the regionalism–multilateralism debate is concerned, touches on a decomposition of Smith’s certitude, namely, how FTA exporters gain from two distinct features of their improved market access. First, the removal of the intra-FTA tariff boosts their market access directly. Secondly, FTA-based exporters benefit from the reduction in RoW exports induced by the tariff discrimination. The second part of the gain – area ‘a’ in Figure 2.2 – could be called the ‘preference rent’ since, if the tariff-cutting were multilateral instead of preferential, FTA partners would gain only ‘b’, not ‘a’ þ ‘b’. This
big-thin k regionalism: a critical survey Border price
Internal price
Border price XS R
MSMFN
XSP
MSFTA
P P⬘ PFT P–T P⬘–T
c
a
d
b
29
MSFT
P⬘ PFT P–T
e
MD T
X R⬘
Figure 2.2
R oW E xports
XP⬘
Partner E xports
M
Home imports
The RTA (preferential trade arrangement) diagram
preference rent ‘a’ is vulnerable to so-called preference erosion and, as such, it plays a leading role in the stumbling-bloc logic. On the import side (Figure 2.3), Home gains a trade-volume effect (equal to area A) from expanding its imports, i.e. replacing highcost domestic production with lower-cost imports. Home also gains from a border-price effect, i.e. the terms-of-trade improvement against RoW (area B) while losing from the terms-of-trade loss against Partner (area C1 þ C2). Home’s terms-of-trade gain on the export side partly offsets the terms-of-trade loss on the import side (D1 ¼ C1), so Home’s net welfare change is A þ B þ D2 – C2.15 As drawn, it looks like Home and Partner gain, but this depends upon elasticities and the initial MFN tariff; in general Viner’s ambiguity holds in this framework.16 The net welfare impact on RoW is unambiguously negative (Haberler’s spillover). RoW experiences no change on the import side, but twice loses area ‘e’ (right-most panel in Figure 2.2) – once on its exports of good 1 to Home and once on its exports of good 2 to Partner. The ‘Haberler spillover’ is an externality as far as the global trade system
15
16
Area C2 might be called the ‘trade diversion’ effect, while D2 and A might be call the ‘trade creation’ effect, but, as usual, the trade creation/diversion dichotomy is incomplete; here it leaves out the third nation terms-of-trade gain, B. As we shall see in the mathematical version of these diagrams, the FTA lowers welfare when the MFN tariff is sufficiently high. See Box 2.3.
30
richard baldwin Border price
Domestic price XS P
D2
A P
P⬘ D1
C1
P–T
B P⬘–T
C2
MD
NB:C1= D1
X
X⬘
Figure 2.3
FTA Exports
X R⬘ X R
M
Home imports
Ambiguous net welfare effects
is concerned, and as such it plays a central role in the Big-Think Regionalism literature.
2.4
Influential diagrams that ignore some of the three-elemental effect
Up to the 1990s, the literature’s main points were presented using diagrammatic analysis – two diagrams were particularly pivotal. The fact that these diagrams ignored some of the three basic effects distorted the direction of the literature and with it academic trade economists’ perceptions of RTAs. Since these older, incomplete diagrams occasionally enter today’s regionalism debate, it is worth presenting them briefly and highlighting their shortcomings. The first is the Johnson diagram that is still used in most undergraduate textbooks.
2.4.1 The Johnson (1960) diagram and the JCM proposition Meade’s analysis was not integrated into mainstream trade theory in part because it was marginal and trade economists were interested in studying the discrete liberalization implied by RTAs. Viner provided no diagrams, so ‘Customs union theory’, as it was known at the time, was a distinctly wordy subject until Johnson (1960a) introduced his famous diagram that illustrated Viner’s ambiguity in a manner that was immediately transparent to all economists.
big-thin k regionalism: a critical survey euros
euros
D
31
S
PB+T
XSB+T
PA+T
XSA+T
PB PA
B+D E
A
B
C
D
F
G
E
H
XSB XSA
MD M
Figure 2.4
M⬘
imports
quantity
The Johnson diagram (small Home and Partner nations)
In the diagram (see Figure 2.4), Home imports can come from partner nation A or B. Home’s demand is an infinitely small share of world demand, so it faces perfectly elastic export supply curves from both sources (labelled XSA and XSB). We start with the Home imposing an MFN-specific tariff of T, so all imports come from the low-cost supplier, nation A. The domestic price is PA þ T, while the border price is PA. Home can form a customs union with nation A or B, so we consider both. The customs union with B would remove the tariff only on imports from B (the high-cost supplier), and this produces supply switching. Home switches from importing everything from A to importing everything from B. Home’s domestic price falls from PA þ T to PB. Assuming a utilitarian metric, the net welfare effects are (B þ D) – E, which may be negative or positive depending upon elasticities and the height of the initial tariff; this is Viner’s ambiguity.17 The customs union with nation B was called ‘purely trade diverting’, yet, if the initial tariff is high and the PB – PA border-price gap was small, it can be welfare-improving for Home. This result – a welfare-improving but purely trade-diverting customs union – seemed to contradict Viner’s reasoning and it produced the first of what was to be a long series of ivory-tower debates over terminology; this one pitted Meade (1955b) against Johnson (1960a) and Corden (1965).18 17
18
The left panel translated the effects into Meade’s two-part framework: B þ D is the trade-volume effect (related to the change in the volume of imports), and E is the tradeprice effect (related to the change in the border price). The 1950s, 1960s and 1970s saw a rather extended and fruitless discussion of what Viner really meant. It featured contributions from the greatest trade economists of the time, including Meade (1955a), Johnson (1960a), Corden (1965), Bhagwati (1971, 1973),
32
richard baldwin
If Home chooses to form a customs union with A, the ambiguity disappears. Such a customs union is unambiguously welfare-improving since its positive effects are identical to MFN free trade (both before and after, all imports would come from A). Home’s domestic price falls from PA þ T to PA, and the net welfare gain is B þ G þ D þ H.19
2.4.2 Omitted elemental effects Readers will immediately note that Smith’s certitude and Haberler’s spillover are missing. Third nations are entirely unaffected by the trade policy of an infinitely small nation like Home. In Johnson’s diagram, the partner nations care no more about Home’s trade policy than a perfectly competitive firm does about gaining or losing one atomistic buyer. This was an attractive feature in the Small-Think Regionalism literature where national welfare was the pivotal issue, but it renders the diagram useless for the Big-Think Regionalism debate. Quite simply, the diagram assumes that Home’s decision to form an FTA has no systemic effects at all. Also missing from the diagram is an analysis of the preferential access that Home’s exports win in its partner’s market. For two decades, the Johnson diagram dominated economic analyses of RTAs to such an extent that Smith’s certitude and Haberler’s spillover came to be largely forgotten by academic trade economists. This went so far that many mainstream trade theorists came to view RTAs as economically irrational – a view encapsulated in the Johnson–Cooper– Massell (JCM) proposition which states that a small nation should always prefer unilateral MFN liberalization to any RTA. The point is easily illustrated in Figure 2.4; cutting T to zero on imports from A and B would always yield net welfare gains that are at least as high as any customs union.
19
Kirman (1973) and Johnson (1974). Viner himself participated in the exchange without fully clarifying matters (Viner 1965). See Kowalczyk (1992) and Pomfret (1986) for summaries of the ‘what Viner really meant’ literature. The quandary was thickened by the fact that the deep economics of taxation in a Walrasian world really do only need two effects, as Meade (1955a) demonstrated, but only one of the two deals with trade volumes. One reaction to this cognitive dissonance was to expand the terminology (e.g. external trade creation, gross trade creation, etc., as in Balasaa 1967); another was to stretch the meaning of the terms to cover the trade-price effects (which is possible since bilateral border prices and imports are related by the export supply curves). This contrast is the source of the rule of thumb that an FTA with your main trading partners is more likely to be welfare-improving since you are giving preferences to the partners that have demonstrated themselves to be the low-cost supplier by winning the largest market share in an even competition with other suppliers.
big-thin k regionalism: a critical survey
33
From the modern perspective, Johnson’s analysis seems impossibly simplistic and the disconnect between academic and real-world thinking is truly astounding. For instance, when the United Kingdom put in its first application for EU membership in 1961, better market access for UK exporters was the key concern, but academic economists were working with the Johnson diagram that assumed this away. Moreover, the main RTA in existence at the time – the EEC – counted for a substantial fraction of world imports and the key nations – the United Kingdom, France and Germany – were far from atomistic. As Pomfret (1997) points out, a number of frameworks were developed at the time that would have allowed the necessary extension, including Johnson (1957, 1958), Humprey and Ferguson (1960), and Blackhurst (1972), but the Johnson diagram’s hold on the literature was so firm that the early efforts were obliged to stick with his small-country fiction.20
2.4.3 The ‘Small FTA’ diagram An important analytical extension of the Johnson diagram came with the ‘small FTA’ analysis (Shibata 1967). It allows for Smith’s certitude even though it continues to assume away Haberler’s spillover. The diagram continues to be used even today (e.g. Grossman and Helpman 1995), so it is worth presenting it briefly. The small FTA diagram looks somewhat different under various assumptions on the pattern of comparative advantage and the smallness of the two partners. The various combinations of assumptions yield a range of results that have been covered by three decades of literature (see Panagariya 2000 for a comprehensive survey of papers using the small FTA diagram in recent decades). Here we study a fairly standard case and illustrate the diagram’s properties by demonstrating two classic results in the regionalism literature. The diagram (see Figure 2.5) presumes that the two FTA partners, Home and Partner, import the same good from RoW. Home and Partner are ‘small’ with respect to RoW, and so face a perfectly elastic RoW export supply curve, XSRoW. This sets the initial border price to PR in both nations. Home has a higher MFN tariff than Partner to start with – TH as opposed to TP – so the pre-FTA price is higher in Home. 20
The early 1980s saw a number of widely read studies that sought to reverse the JCM proposition while staying in the small-country framework. These efforts, e.g. Wonnacott and Wonnacott (1981) and Berglas (1983), strike the modern reader as awkward due to the small-nation assumption and the intricate diagrammatic analysis.
34
richard baldwin euros Trade price gain
euros
Partner D
Home Trade volume gain
S
Trade price loss
PR+TH
PR+TP
XSRoW
PR
S
D
M⬘H
Figure 2.5
quantity
M⬘H
quantity
The small PTA diagram: a simple case
When Home and Partner form their FTA, Partner-based firms initially see a higher price in Home and so begin exporting to Home. In equilibrium, all post-FTA Home imports, M' H, are supplied by Partner firms. Partner’s internal price remains at PR þ TP, so its consumption and production are unchanged, which means that the new exports to Home are replaced – one for one – by new Partner imports from RoW. In the case illustrated in the diagram, Partner is large enough relative to Home to ensure that Home’s entire demand can be satisfied by Partner producers at PR þ TP. In terms of welfare, the FTA results in a positive trade volume effect for Home but a negative border price effect (Home pays PR þ TP for its imports instead of PR). Partner expands its imports across the tariff wedge, and this results in a positive trade volume effect equal to TR times the expanded imports, i.e. M' H.
2.4.4 Irrelevance of rules of origin Although it seems an odd objective from today’s perspective, Shibata’s goal was to illustrate the irrelevance of rules of origin (RoO). His point is that RoO only prevent blatant trade deflection. Because goods from Partner and RoW are fungible, the equilibrium is the same with and without RoO as long as Partner’s supply is sufficient. If Partner’s supply were not large enough to supply all of Home’s imports at PR þ TP, the FTA with RoO would have somewhat higher prices than one without. The use of this exercise in the Big-Think Regionalism literature comes in the form of ‘imported liberalization/protection’ (see section 3.3.4 below), and Grossman and Helpman (1995).
big-thin k regionalism: a critical survey
35
2.4.5 Unsustainability of the FTAs Another application that found popularity in the academic literature but seems odd today is the proposition that FTAs will always break down. Using a diagram like Figure 2.5, Vousden (1990 p. 234) argues that Home would be tempted to lower its MFN tariff to just under that of Partner in order to recapture the tariff revenue, and Partner would have an incentive to reply. The resulting race-to-the-bottom tariffcutting was viewed as making FTAs ‘unsustainable’. Vousden’s (1990) diagram did not attract much attention until Richardson (1995) extended and popularized it. These two results (irrelevance of rules of origin and unsustainability of FTAs) are classic examples of how academic thinking on regionalism has often followed literature-driven paths that have little relevance to real-world policy concerns. 3
Stumbling and building-bloc logic
From 1960 to the late 1980s, regionalism was a simple matter. It consisted of: (1) the EEC, which encompassed one-third of world trade in a highly effective customs union; and (2) a slew of RTAs among developing nations that covered a trivial fraction of world trade and in any case never operated effectively. Regionalism’s systemic implications were simply not an issue. Regionalism got complicated in the late 1980s when Canada and Mexico changed their minds on regionalism (Krugman 1991b p. 7).21 The US had long been interested in regional preferential trade, but Mexico and Canada resisted, fearing domination by their giant neighbour. Canada proposed an FTA with the US in 1985 that entered into force in 1989. Mexico proposed an FTA with the US in 1990, and this evolved into NAFTA at Canada’s insistence (to safeguard its Auto Pact preferences). 21
Bhagwati (1991 p. 71) ascribes the shift to the US’s conversion to regionalism, but this contradicts the judgments of trade policy scholars who were engaged in the details of policy at the time (Smith 1988 p. 41, Wonnacott 1987 p. 17, Schott 1988 p. 29, Hufbauer, Schott and Clark 1994 p. 100, Whalley 1993). It also contradicts the facts. The US’s long-standing interest in regionalism is testified by a long string of deals that were struck, or almost struck, in 1854, 1874 and 1911. In March 1948, they concluded a secret draft protocol eliminating most tariffs and quotas bilaterally, but this was ultimately rejected by the Canadians. In 1958, US government procurement was preferentially liberalized in Canada’s favour. The US–Canada Auto Pact came into force in 1965. The 1974 Trade Act authorized the US President to negotiate an FTA with Canada, and the 1979 Trade Agreements Act required the President to study an FTA in North America.
36
richard baldwin
The US–Mexico initiative triggered a wave of Latin American requests for bilateral FTAs with the US and gave greater urgency to arrangements among Latin Americans, most notably MERCOSUR.22 The rise of North American regionalism coincided with two other major developments in the world trade system. First, GATT negotiations were lurching from crisis to crisis in the late 1980s and then, as mentioned in the introduction, seemed to die with the acrimonious collapse of the Uruguay Round’s ‘final’ summit in December 1990. Secondly, European regionalism was reignited by the Single European Act and the collapse of the Soviet Union. Many respected thinkers looked at this temporal correlation and saw causality. They feared that regionalism’s spread might kill the proverbial gold-laying goose – the GATT-centred world trade system. The fears are easy to understand. Two-thirds of world imports went to North America and Europe; 40 per cent of this was intra-bloc trade and soon to be covered by discriminatory liberalization schemes. Still more worrisome, North Americans and Europeans were the stalwarts of the GATT system. If regionalism weakened their support for multilateralism, the goose was indeed in deep trouble. Spreading regionalism had become much more than a Small-Think ‘should I join?’ question. These fears promoted regionalism to a status on the world’s policy agenda that it had not enjoyed since the 1950s. This naturally attracted paradigm-setting efforts from the profession’s leading international economists.
3.1
Framing the new regionalism debate
Krugman (1991b) is clearest in rejecting the relevance of the 1950s Small-Think approach and delineating the outlines of a new line of inquiry – what I call Big-Think Regionalism: In a fundamental sense, the issue of the desirability of free trade areas is a question of political economy rather than of economics proper. While one could argue against the formation of free trade areas purely on the grounds that they might produce trade diversion . . . [t]he real objection is a political judgment: fear that regional deals will undermine the delicate balance of interests that supports the GATT.23
22 23
See Baldwin (1997) or Serra et al. (1997) for an account of this domino effect. Krugman (1991a) p. 23.
big-thin k regionalism: a critical survey
37
The Krugman (1991b) framing of what he identified as the key issue – the impact of regionalism on support for the GATT system – did not catch on.24 His 1991 papers, however, did reframe the 1950s national welfare question into a global-level question. Krugman (1991a) introduced a new approach by asking whether spreading regionalism raises or lowers world welfare. This spawned a decade-long literature and continues to influence research even today. This ‘Is bilateralism bad?’ literature – also known as the multilateralism versus regionalism literature – looks distinctly odd from today’s perspective. It tries to use simple theory to answer what is intrinsically a complicated empirical question. At the time, however, it was the best they could do. Economists had limited access to the necessary data and lacked the panel econometric techniques to exploit them. Moreover, spreading regionalism was at the time more of a threat than a reality, so there was little to test empirically. In my mind, this literature is now mainly of interest to historians of thought, but many participants in today’s regionalism debate cite it, so I review it in section 3.5 below. The focus of this section is on what I consider to be the central theoretical question: Does regionalism help or hinder multilateralism? Ultimately this also is an empirical question, but, given the relatively little experience the world has had of the regionalism–multilateralism interface (only one MTN has been completed since 1991), convincing empirics is not yet feasible – although some tantalizing results are beginning to emerge. Moreover, given the complexity of the
24
He argued that the multilateral process had run aground with the December 1990 failure and was unlikely to get afloat anytime soon as the system was plagued by deep-seated problems. ‘[W]hile some kind of face-saving document will probably be produced, in reality the Uruguay Round has clearly failed either to significantly liberalize trade or to generate good will that would help sustain further rounds of negotiation.’ Regionalism, however, was not one of those deep-seated problems. ‘But while the move to free trade areas has surely done the multilateral process some harm, it is almost surely more a symptom than a cause of the decline of the GATT . . . [T]he problems of the GATT are so deep-seated that it is unlikely that a world without regional free trade agreements would do much better.’ He closes his essay with a prediction that history falsified a faute de mieux view on regionalism. ‘The world may well be breaking up into three trading blocs; trade within those blocs will be quite free, while trade between the blocs will at best be no freer than it is now and may well be considerably less free. This is not what we might have hoped for. But the situation would not be better, and could easily have been worse, had the great free trade agreements of recent years never happened.’
38
richard baldwin
interlinkages, a clear theoretical understanding is a necessary condition for well-structured empirical work.
3.1.1 Are regionalism and multilateralism friends or foes? Bhagwati’s book, The World Trading System at Risk, does not focus on regionalism. Indeed, his Part One, ‘The GATT Architecture under Threat’, listed four main threats; regionalism was number four. Nevertheless, his writing helped establish Big-Think Regionalism as the new paradigm. In the lead paragraph to his chapter on regionalism, he writes: ‘These regional alignments have led to fears of fragmentation of the world economic into trading blocs in antithesis to GATT-wide multilateral free trade. Does such regionalism truly constitute a threat to multilateralism?’ Although he does not set out an analytic framework for answering the question, his writing influenced the intellectual paradigm for more than a decade. 3.1.1.1 Framing the issue three ways Theory requires explicit questions. Asking whether regionalism and multilateralism are friends or foes is not sufficient. Pure logic identifies three mutually compatible ways that regionalism and multilateralism could interact: regionalism could affect multilateralism; multilateralism could affect regionalism; and both multilateralism and regionalism could be driven by third factors.
The literature has looked at all of these, but the first has dominated since Krugman (1991b, 1993) presented a simple analytic framework for posing the question. His explicit question was: how does an exogenous variation in regionalism (specifically, the formation of a new RTA) affect nations’ incentives to cut tariffs multilaterally? 3.1.1.2 Modelling choices and branches of the literature Answering this question requires an economic model that links tariff choices to equilibrium outcomes and a political-economy model that endogenizes the MFN tariff choice (the RTA tariff levels are taken as exogenous in this literature). The choice of economic models is quite open, but the literature naturally gravitated to the simplest possible models that yielded the elemental economics effects – Smith’s certitude and Haberler’s spillover – that are pivotal to regionalism’s systemic implications (Viner’s ambiguity is more a Small-Think Regionalism issue).
big-thin k regionalism: a critical survey
39
The choice of political-economy modelling is facilitated by two key institutional features of the global trade system: the nature of RTA tariffcutting and MFN tariff-cutting. Specifically: Real-world RTAs involve bargaining on very few tariff lines; they cut
tariffs to zero on most goods.25 This institutional fact made modelling of the preferential liberalization easy – most authors assumed that forming an RTA meant zero tariffs on all goods traded among members. Real-world multilateral tariff-cutting talks also involve bargaining on very few tariff lines. Since the 1963 Kennedy Round began, multilateral tariff ‘bargaining’ was simple: GATT members pre-agreed tariff-cutting rules with exceptions for specific products (the usual ‘sensitive’ products such as clothing, footwear, etc.). This real-world feature is easily understood. Nations’ tariff schedules typically list about ten thousand individual lines and at least a dozen nations participate actively in MTN tariffcutting negotiations. (The number has increased greatly in the ongoing Doha Round, but only the richest OECD nations participated substantially in the Kennedy, Tokyo and Uruguay Round tariff cuts.) If each of the dozen nations bargained for one minute over each of its tariff lines with each of the other nations, the talks would take twenty-five years of 24/7 discussion. Trade diplomats avoid this by agreeing the basic tariff-cutting before the talks start – i.e. it is set in the agenda. Specifically, the Kennedy and Tokyo Rounds used formulas that cut tariffs by about one-third with some sensitive industrial goods excluded (Winters 1991 p. 171). The Uruguay Round agreed, at its 1988 Ministerial Midterm Review in Montreal, to cut tariffs by at least as much as in the Tokyo Round, i.e. by about onethird and again sensitive products were predictably excluded (Croome 1995 p. 183). This institutional fact made modelling of the multilateral liberalization easy (or at least should have). Many authors modelled multilateralism as saying yes-or-no to an exogenously defined MFN tariff cut. One branch of the literature, however, adopted the alternative tack of twisting reality to the theory. These authors ignored the institutional
25
RTAs involving a developed WTO member must respect this; those involving only developing nations do not, but de facto they have, as chapter 3 in this volume shows.
40
richard baldwin
features of MTNs in order to use game theory’s simple bargaining models – models that had been developed to deal with much simpler bargaining situations. While I do not believe that such models provide useful insights into the key Big-Think Regionalism questions, they are often referred to in the academic literature, so I review them in section 3.4 below. While most of the literature has followed Krugman’s lead in asking how exogenous variations in regionalism affect multilateralism, two other branches consider (1) how RTAs can affect MTNs (competitive liberalization), and (2) how RTAs and MTNs are driven by deeper forces (Summers’ notion that all the ‘-isms’ are good). I review thinking on these in section 3.5 below. With this structuring of the literature out of the way, I turn to review what I consider the most relevant theory on the Big-Think Regionalism literature.
3.2
Stumbling-bloc logic
In its cleanest form, the stumbling-bloc logic asserts that, if the stumbling bloc RTAs were forbidden, global free trade would be obtained, but, since they are allowed, global free trade becomes impossible. Weaker forms are put forth in the help-or-hinder literature – for example, RTAs may slow the achievement of global free trade – but clarity has led the profession to focus on the strong form. In my opinion, only three distinct stumbling-bloc logics are relevant to real-world policy analysis: (1) the preference-erosion/exploitation stumbling bloc; (2) the goodies-bag stumbling bloc; and (3) the cherrypicking stumbling bloc, although I am sure many more shall be illuminated in coming years. For simplicity’s sake, these stumblingbloc logics are demonstrated under the naı¨ve but transparent assumption that national governments choose tariffs to maximize national welfare.
3.2.1 Preference-erosion/exploitation stumbling bloc The logic of the preference-erosion/exploitation stumbling bloc is easy and quite general. Starting from a world where all nations have MFN tariffs, the question is: Can some group of nations raise their collective welfare above the free trade level by forming a trade bloc and thus exploiting other nations? If the answer is ‘yes’, then that bloc is a
big-thin k regionalism: a critical survey
41
stumbling bloc on the road to multilateral free trade because the bloc members would veto global free trade as undermining their exploitation of third nations. In the Walrasian setting, and in many other trade models, the answer is almost always ‘yes’, but the answer may depend upon the level of MFN tariffs when the bloc is formed. Given Smith’s certitude and Haberler’s spillover, some combination of nations is bound to be able to better exploit third nations by acting as a bloc. This is almost trivially true if the bloc can violate its WTO tariff bindings by raising external tariffs. After all, the bloc as a whole has more buying power than its constituents do individually, so the bloc can better exploit foreigners. Less obviously, but equally true, stumbling blocs can be found even when external tariffs are maintained (as has been the case for all of the major post-war RTAs). The preference-erosion/exploitation stumbling-bloc logic can be illustrated more concretely with the RTA diagram. 3.2.1.1 A Walrasian illustration Start by noting that MFN free trade would be approved by all governments in the simple model laid out above (see Figure 2.2 and Box 2.3). The first welfare theorem tells us that global free trade is efficiency-enhancing (a move to the first-best), and symmetry ensures that each nation would get an equal slice of the gains. This conclusion, however, can be reversed when we start from the situation where Home–Partner have formed an FTA. Taking the FTA as the base case, a move to global free trade eliminates the preference margin Home exporters enjoyed in Partner and this would lead to a terms-of-trade loss of area C1' and a trade-volume loss of C2' (see Figure 2.6). On the import side, global free trade would win Home an additional trade-volume gain of area A' , a terms-of-trade improvement with respect to Partner exporters of area C1' , and a terms-of-trade loss on imports from RoW, area B' . Global free trade would also improve Home exporters’ market access to RoW, and this would boost Home welfare by area D' . Overall, the net welfare change of moving from the FTA to global free trade is – C2' – B' þ D' þ A' ; the sign of this is ambiguous. As Box 2.3 shows, however, it is always negative when the initial MFN tariff is low enough. Working out magnitudes in the diagrams for the whole range of possible T’s is tedious, so we lighten the analysis by introducing the
42
richard baldwin
box 2.3 the walrasian rta diagram model in maths The three goods (1, 2 and 3) are produced according to rising marginal costs. For good 1, we assume Partner and RoW have comparative advantage, while Home does not. The assumed linear supply curves that yield this are: p ¼ b þ X1H ;
p ¼ X1P ;
p ¼ X1R ;
0
ð2Þ
where the superscript indicates the producing nation and the subscript indicates the good. Demand for good 1 is identical in all nations and is given by: j
j
p i ¼ a C1 ;
j ¼ H; P; R;
0
ð3Þ
The supply and demand curves for goods 2 and 3 are similar, but Partner has the comparative disadvantage in good 2 and RoW in good 3. The demand curves are identical for all goods. As is well known, the utility function that generates this demand structure involves quasi-linear preferences (utility is separable and quadratic in consumption of each of the three goods). The model has a fourth untaxed good that is only introduced to formally eliminate Meade’s tertiary effects (it is produced under constant returns using only labour and identical technology worldwide, thus ensuring trade balance as long as the parameters are such that all three nations produce some in all equilibria, an assumption we maintain throughout). We take labour as nume´raire and choose units so that one unit of labour is needed to produce one unit of the fourth good. The supply curves are thus the marginal cost curves where marginal cost is, for example, the wage ‘w’ times total output X for nations with a comparative advantage in the good in question. As usual, rising supply curves generate producer surplus which we take to be the reward to the implicit, scarce, specific factor that is generating the diminishing returns. As is well known, consumer surplus plus producer surplus plus tariff revenue is an exact measure of utility, assuming tariff revenue is returned lump sum to the workers/owners.
equations behind the diagrams (see Box 2.3). Solving for equilibrium prices under the FTA and global free trade situations and plugging these into the welfare functions, we can plot the welfare of a typical FTA nation (Home) against the MFN tariff. The results are shown in Figure 2.7. For initial MFN tariff levels that are sufficiently low, we see that Home’s welfare is higher with the FTA than it is with global free trade – and this despite the fact that Home would have agreed to global free trade starting from the initial situation without the FTA. (The line marked MFN is
big-thin k regionalism: a critical survey Internal price
Border price C2⬘ P⬘ PFT
XS
P
A⬘
C1⬘
C1⬘
D’
B⬘
P–T P’–T
Exports
Figure 2.6
43
MD
XR⬘
Imports
Net welfare effects, FTA to global free trade
everywhere below the line marked global free trade, but the dashed FTA line is above the global free trade line for sufficiently low tariffs.) Intuitively, the FTA allows Home to exploit RoW both on the import side (by pushing down the price it pays RoW good 1 exporters) and on the export side (by raising the price in Partner at the expense of RoW good 2 exports). The move to global free trade undoes these two forms of exploitation, but in exchange it provides better access to the RoW market and more liberalization in Home’s import market. When the initial T is low, the market-access and home-liberalization gains are modest so the net is negative. In other words, the basic logic of the stumbling-bloc result turns on the way that an FTA allows the FTA partners to exploit excluded nations. The Walrasian model presented here is very special, but the heartand-soul of this stumbling-bloc effect – the exploitation of excluded nations – is quite a general result; one that is surely an important consideration in the real world. The opposition of small developing nations to agricultural liberalization in the DDA – especially those who benefit from the EU’s unilateral preferences – is a classic example of the preference-erosion stumbling bloc. Had the EU not unilaterally granted these nations preferences, these nations would probably have been pushing for EU market opening in sugar and other goods.
44
richard baldwin Home Welfare Global free trade
0.6
0.59
0.58
0.57 Stumbling bloc T’s
MFN
0.56
FTA 0
0.05
0.1
0.15
0.2
0.25 MFN T
T
Figure 2.7
Stumbling-bloc FTAs
3.2.1.2 Who did what, when? The logic of the preference-erosion/ exploitation stumbling bloc was demonstrated in a Walrasian setting by Reizman (1985) and Kennan and Reizman (1990), and in a Brander– Krugman setting by Krishna (1998) and Freund (2000a, 2000b).26 The core algebra for the preference-erosion/exploitation stumbling bloc in the Brander–Krugman setting is laid out in Box 2.4 where the main Freund and Krishna results are also demonstrated.
box 2.4 preferential liberalization in the brander–krugman model One set of stumbling/building bloc papers employs the Brander–Krugman model, i.e. Cournot oligopoly with segmented markets. As is well known, reciprocal dumping trade among all nations arise in this model for a wide range of parameters. This set up is useful since every nation sells to every other nation. Moreover, the Cournot oligopoly and market segmentation immediately give us Smith’s certitude and Haberler’s spillover. Interestingly, Viner’s ambiguity does not arise in the simple symmetric version since the pro-competitive effects of
26
Also see Goto and Hamada (1997, 1999), Nordstrom (1995) and section 3 of Bond and Syropoulos (1996a) for examples of preference erosion/exploitation stumbling blocs.
big-thin k regionalism: a critical survey
45
box 2.4. (cont.) preferential liberalization are always strong enough to outweigh any terms-oftrade losses for the integrating nations. One important drawback in this approach is the inability to distinguish between import-competing firms and export firms (they are the same firms). On the demand side, the model is quite similar to the Walrasian model in Box 2.3. It has three nations (Home, Partner and RoW) and uses the same quasi-linear preferences with an untaxed nume´raire to neutralize Meade’s secondary and tertiary effects. The demand curves are identical to (3). Marginal costs are flat (set to zero without loss of generality, given the linear demands), there is one firm per nation, and firms play Cournot in all markets (assumed to be segmented). Solving the Nash first-order conditions, profits of a Home firm depend upon tariffs according to: 1 ð4Þ pH ¼ ða þ tPH þ tRH Þ2 þ ða þ tRP 3tHP Þ2 þ ða þ tPR 3tHR Þ2 16 where the subscripts on the directional specific tariffs (the t’s) are ‘from, to’ (e.g. tRP is the tariff on the good shipped from RoW to Partner). The expressions for Partner and RoW profits are isomorphic. Equilibrium welfare of a typical nation – defined as the sum of consumer and producer surplus plus tariff revenue – depends upon the matrix of tariffs according to: 1 WH ¼ f15a2 þ tHP ð18tHP 12ða þ tRP ÞÞ þ tHR ð18tHR 12ða þ tPR ÞÞ 32 ð5Þ þ tHR ð18tHR 12ða þ tPR ÞÞ þ tRP ð4a þ 2tRP Þ þ tPR ð4a þ 2tPR Þ þ tRH ð6a 21tRH Þg Krishna (1998) uses profit at the government’s objective function, while Freund 2000a uses welfare. The value of Home profits under the three regimes – symmetric MFN tariffs ‘MFN’, FTA with Partner ‘FTA’, and hub-and-spoke FTA with Partner and RoW ‘HS’ – are: pMFN ¼ pFT H H þ
3t 2 at ; 4
FT pFTA H ¼ pH þ
3t 2 ; 8
FT pHS H ¼ pH þ
t 2 þ 2at 8
ð6Þ
2
3a where pFT H ¼ 16 . The corresponding expressions for Partner are the same for the MFN and FTA cases (by symmetry of nations), and RoW’s expression is also identical for the MFN case. The profits for the three remaining cases are:
5t 2 2at ; 8 2 5t 2at þ 8
FT pHS P ¼ pH þ FT pHS R ¼ pH
¼ pFT pFTA R H þ
11t 2 4at ; 8
ð7Þ
46
richard baldwin
3.2.2 Goodies-bag stumbling bloc Another stumbling-bloc logic – one which follows closely the fundamental economic logic of the preference-erosion stumbling bloc – might be called the ‘goodies bag’ stumbling bloc.27 In a nutshell, the rents corresponding to ‘Smith’s certitude’ can be thought of as a ‘bag of goodies’. These goodies can be used by one or both RTA parties to buy non-economic benefits from its partners. Since the size of non-economic benefits that can be ‘purchased’ is linked to the richness of the ‘goodies bag’, i.e. the margin of preferences, RTA members have an extra incentive to maintain high margins of preference by avoiding multilateral liberalization. The goodies-bag logic, however, extends to a far greater range of issues than the tariffs that are the focus of the preferenceerosion stumbling bloc. In the case of RTA between very large and very small nations – a case that is extremely common in the new century (e.g. US and Costa Rica or Japan and Singapore) – the large country’s interest in the RTA can hardly be the preferential market access. The EU, for example, grants extensive preferences to its members’ former colonies, justifying these on the basis of international solidarity. In other words, the economic gains to the EU’s partners count as a plus inside the EU since they advance one of the EU’s non-economic objectives – fostering development. Similarly, but more explicitly, the US justifies many of its FTAs with small, poor nations on the basis of noneconomic objectives (typically anti-drug and/or anti-terror policies). The previous section illustrated how the desire to guard rents created by an RTA could make a nation reject global free trade when MFN free trade would have been embraced without the RTA. The goodies-bag stumbling-bloc logic amplifies this mechanism by making both nations interested in each other’s export rents – the area corresponding to C1' , in Figure 2.6, with the link coming through the pursuit of noneconomic objective (non-economic in the narrow sense). 27
Here is an example of the terminology in the context of US unilateral concessions from The Hindu newspaper, 24 September 2001. Referring to the mood in Washington, the correspondent notes: ‘There is tremendous support and sympathy for Pakistan in the light of the decisions taken by its President, Gen. Pervez Musharraf, to support the US wholeheartedly in its war against terrorism. As for Pakistan, the question is not merely the lifting of US sanctions. Rather, it is in the larger “goodies bag”‘ that most certainly is coming up. As it is, there is disappointment in the official Pakistan establishment here that Mr Bush has not gone the whole hog.’ See www.hinduonnet.com/2001/09/25/ stories/03250007.htm.
big-thin k regionalism: a critical survey
47
3.2.2.1 Who did what, when? The theoretical notion was formalized by Lima˜o (2007) and tested in Karacaovali and Lima˜o (2005) for the EU, and Lima˜o (2006) for the US. Both papers suggest that these trade powerhouses cut their MFN tariffs less on those items where they granted important preference margins. In short, the goodies-bag did seem to act as a ‘slowing bloc’ even if it did not act as a stumbling bloc. The evidence in Lima˜o (2006) is widely misrepresented as showing that the US raised tariffs in the Uruguay Round for items on which it granted FTA preferences. Of course, this cannot be right since MTN market access talks only involve tariff bindings and the US did not violate any of its bindings in the Uruguay Round. Indeed, the data shows US tariffs decreasing for all but twelve of the thousands of tariff lines (defined at the HS-8 products level in the WTO’s database).28 Formally, Lima˜o estimates an econometric model of US tariff cuts in the Uruguay Round. His famous stumbling bloc results that he finds is that the US cut tariffs by less than his econometric model predicted they should have on items for which the US had granted FTA preferences before the Uruguay Round. In short, he shows that the US preferences acted as a ‘slowing bloc’ not a stumbling bloc.
3.2.3 Cherry-picking stumbling bloc An entirely distinct mechanism is at work in the third type of stumbling bloc – the cherry-picking stumbling bloc. Moving to global free trade will typically involve some pluses and some minuses from the national perspective. The question is: Can one find a group of nations whose integration would provide many of the pluses with few of the minuses? If the answer is ‘yes’, the trade bloc is likely to be a stumbling bloc. Starting from the bloc situation, a move to global free trade involves many minuses and fewer pluses for bloc members, so bloc members may veto global free trade even when they would have embraced it without the trade bloc. Here is an example of a setting that could yield a cherry-picking stumbling bloc. Assume the trading environment is marked by Helpman–Krugman trade; there is both intra-industry trade in differentiated products and inter-industry trade in Walrasian sectors. In this world, trade liberalization will produce gains from trade due to the Krugman variety effect. It will also produce gains from trade from comparative advantage effects, but the comparative advantage gains 28
As per a bilateral communication with the author.
48
richard baldwin
come bundled with politically difficult Stolper–Samuelson effects on domestic factor prices. Now suppose two large nations have quite similar factor endowments. If they form a trade bloc, they will win a large share of the variety gains that would come with global free trade, but will experience little of the Stolper–Samuelson pain. Taking the trade bloc as the base case, the bloc members may find a move to global free trade unattractive. It would entail a good deal of Stolper–Samuelson pain and only a modest amount of additional variety or comparative advantage gains. Depending upon parameters, especially the political power of Stolper–Samuelson sufferers, the gains may not be sufficient to make global free trade attractive to the bloc members. 3.2.3.1 Who did what, when? Levy (1997) illustrates a cherry-picking stumbling bloc in a highly stylized set-up, but his main result is surely more general than his model.
3.3
Building-bloc logic
While many trade policy scholars – such as Paul Krugman (1991b) and Jagdish Bhagwati (1991) – worried that regionalism was a stumbling bloc to global free trade, others – such as Larry Summers (1991) and Fred Bergsten (1991) – viewed regionalism as a largely benign or even constructive force in the world trade system. This sub-section considers the economic logic of the assertion that RTAs could foster multilateral liberalization, i.e. how RTAs may be building blocs on the road to global duty-free trade. There are four main logics in the literature. We start with the one that permeates the justifications used by nations which simultaneously pursue regional and multilateral liberalization – the notion that preferential liberalization creates a political-economy momentum that makes multilateralization liberalization easier (and vice versa).
3.3.1 Juggernaut building-bloc logic Liberalization begets liberalization, according to the juggernaut building-bloc logic. The logic comes in two parts that are most easily explained in the context of multilateral liberalization. When the GATT started in 1947, import duties were high worldwide, since they had been set without international coordination during the Smoot–Hawley tariff wars. The tariffs balanced the supply and demand for protection in the ‘political market’ of each nation separately. The main demanders of
big-thin k regionalism: a critical survey
49
import protection were import-competing firms and the workers they employed. Governments were the supplier of protection, but concern for the general economic well-being of the nation meant that the supply of protection was not perfectly elastic. Starting from this situation of uncoordinated tariff-setting, announcement of an MTN based on the principle of reciprocity alters the array of political forces inside every participating nation. The central point is reciprocity; it converts each nation’s exporters from bystanders in the tariff debate to anti-protectionists. For exporters, lobbying against domestic tariffs becomes a means of lowering foreign tariffs. Because the MTN rearranges the political economy forces inside every nation, all governments find it politically optimal to choose tariff levels that are lower than the unilaterally optimal tariffs.29 This is the first part of the juggernaut theory.30 The logic is not new. Informed observers have long known that the GATT’s reciprocal MTNs were mostly about helping nations internalize a political economy externality inside their own polities, i.e. about making it easier for national politicians to put together a national coalition in support of freer trade. Writers such as Robert Baldwin (1970, 1985) and Mac Destler (1992) are explicit on this point, but historical accounts of the Cobden–Chevalier Treaties show that using external trade deals to realign domestic political forces was very much in the minds of nineteenth-century thinkers (Irwin 1993 p. 96). Even Krugman (1991b) writes: ‘[T]he process of multilateral negotiation . . . sets each country’s exporting interests as a counterweight to import-competing interests; as trade negotiators bargain for access to each others’ markets, they move
29
30
More formally, without MTNs governments maximize a politically weighted objective function that includes things affected by the nation’s own tariffs – profits of importcompeting sectors, consumer surplus and tariff revenue. During the MTN, a nation’s tariff affects all these things, but foreign tariff levels are linked to domestic tariffs via reciprocity; the objective function now includes the impact of foreign tariffs on exporter’s profits. Since this new impact is negative (higher domestic tariffs lower exporters’ profits via reciprocity), announcement of the MTN leads the government to find it politically optimal to choose a tariff that is lower than the politically optimal tariff before the MTN. The word ‘juggernaut’ – defined as ‘any massive inexorable force that advances crushing whatever is in the path’ – stems from a British mispronunciation of the Hindu deity of the Puri shrine, Jagannath. A festival is held in Puri involving the ‘chariot of Jagannath’, an enormous and unwieldy construction that requires thousands of people to get it rolling. Once started, however, it rolls over anything in its path.
50
richard baldwin
toward free trade despite their disregard for the gains from trade as economists understand them.’ The second part of the juggernaut logic concerns the impact of the tariff cuts on openness. The tariff cuts make all nations’ more open export sectors expand with the foreign tariff cuts, and importcompeting sectors contract with domestic tariff cuts. Assuming political influence is linked to industry size, this economic re-landscaping strengthens pro-liberalization forces and weakens anti-liberalization forces in all nations – although of course such industrial restructuring takes years. In other words, the initial reciprocal tariff cuts start a liberalization juggernaut rolling. Due to the economic re-landscaping that occurs during the phase-in of the initial tariff cuts, all governments find that their politically optimal tariff in the next MTN is below the levels that they found politically optimal during the previous MTN. These fresh tariff cuts continue the re-landscaping and the juggernaut continues to roll forward. Thus, once the liberalization juggernaut starts rolling, it crushes all tariffs in its path.31 To the extent that regionalism can start the juggernaut rolling, RTAs can be building blocs – hence the name ‘juggernaut building-bloc logic’.32 The precise mechanism is a simple extension of the juggernaut logic. RTAs re-landscape members’ economies – making export sectors 31
32
The juggernaut logic is from Baldwin (1994 chapter 2.5) and is elaborated in Baldwin (2006) and formalized in part by Freund (2000b). The first part of the juggernaut mechanism – reciprocity’s realignment of domestic special interests – has long been recognized in histories of trade liberalization, e.g. Destler (1992 p. 17) and Bergsten (1996) under the name ‘bicycle theory and export politics’; the point was also made by many others including Baldwin (1984), and Hoekman, Bernard and Kostecki (2001). The basic idea dates much further back, as Irwin (1996) points out. Even more recently, the first-half of the juggernaut logic has been studied formally by Grossman and Helpman (2001), and Bagwell and Staiger (2002). Juggernaut-like mechanisms were discussed independently by Hufbauer, Erb and Starr (1980), Hufbauer, Schott and Clark (1994 p. 164), and Richardson (1993). Bergsten (1998) mentions an alternative source of political-economy momentum (‘modest liberalization begets broader liberalization by demonstrating its payoff and familiarizing domestic politics with the issue’). Staiger (1996, section 5.4) uses a repeated game setting with workers moving slowly out of the import-competing sector to generate gradualism, but MTNs and GATT reciprocity play no role. Milner (1997) and Oye (1993), working independently in the International Political Economy context, discuss mechanisms by which FTAs can create a pro-liberalization political economy momentum. More recently, Hathaway (1998) presents a similar logic in her positive-feedback model. But, likewise, multilateral tariff-cutting may lower tariffs to the level where PTAs become feasible, when previously they were not (this may have been the case with the US–Canada FTA).
big-thin k regionalism: a critical survey
51
larger and import-competing sectors smaller. Thus the RTA can alter the member governments’ stance in MTNs, making it politically optimal to cut MFN tariffs to levels that would not have been politically optimal without the RTA. Of course, if an RTA results in higher external tariffs (as in the case of the EU’s agriculture tariffs), then an RTA can start the juggernaut rolling backwards. The basic idea can be presented in a diagram, Figure 2.8. The two curves, FE and GFOC, show how the size of the import competing sector depends upon the tariff (free entry) and how that tariff depended upon the size of the import-competing sector via politics. Here, ‘n’ (the number of firms) measures the size of the import-competing sector. The politically optimal tariff choice – taking as given the size of the import-competing sector – is plotted as GFOCunil (solution to the government’s first-order condition without MTN). The politically optimal tariff rises with n, since the larger the import-competing sector, the higher is the political benefit from a marginal increase in the tariff. The free entry curve, FE, relates the equilibrium number of firms to the tariff. As the tariff rises, more firms find it optimal to enter the market. These two relationships assume that the government and firms are short-sighted and the government chooses T taking n as given, while firms choose n taking T as given. Note that this only captures the size of anti-trade forces; the size of the pro-trade export sector is suppressed to avoid a three dimension diagram. To see the two steps of the juggernaut effect, note that announcement of the MTN shifts the GFOC curve to GFOCMTN. This is lower, since the government finds it politically optimal to set a lower tariff for any given level of n when domestic protection is linked to foreign protection via reciprocity. As drawn, the new long run equilibrium, Efinal, entails free trade, but, since entry and exit occur slowly, the tariff and state of the import-competing industry do not jump to Efinal. The diagram illustrates one possible adjustment path. Each MTN results in an instantaneous drop in the tariff, but slow entry/exit means the horizontal arrows reflect slow movement of the state variable. The juggernaut effect acts as a building bloc if the FTA reduces the importance of the import-competing industry in governments’ objective functions. As we saw in the previous section, an FTA between Home and Partner does reduce the size of the Home import-competing sector, so, in that model, FTAs are building blocs on the road to global free trade. Figure 2.9 illustrates the juggernaut building-bloc logic diagramatically (the maths of the juggernaut stumbling bloc are illustrated in Box 2.7 in the Annex to this chapter). The FE curve in Figure 2.9 was
52
richard bald win Tariff,T FE Eo
To
GFOC (unil)
GFOC (MTN)
Size of import competing sector, n
Efinal
Figure 2.8
Juggernaut framework
Tariff,T FE (with FTA) Eo
To
E1
FE GFOCunil
GFOCMTN
E2
Size of import competing sector, n
Figure 2.9
Juggernaut building-bloc logic
drawn for symmetric MFN tariffs. When the FTA is signed, the FE curve rotates inwards as shown since the additional competition from Partner producers lowers that Home price facing import-competing firms so some of them exit. Consequently, the GFOC under reciprocal trades will yield a lower MFN tariff after the FTA (point E2) than before (point E1). Of course, if the FTA somehow raised protection of the Home
big-thin k regionalism: a critical survey
53
import-competing sector, the effect is reversed and the FTA acts as a stumbling bloc.33 3.3.1.1 The Wei and Frankel momentum Frankel and Wei (1995) illustrate another juggernaut-like mechanism. In their model, imperfect information makes workers uncertain as to whether they will win or lose from global free trade. Since an FTA is an intermediate form of liberalization, they show that an FTA could be politically feasible even when global free trade would not be. After the FTA is signed, the nation’s true comparative advantage is revealed, and workers now know whether they will win or lose from free trade. If the parameters are chosen carefully, the certainty resolution may mean that global free trade is politically feasible only after the FTA. Thus the FTA is a building bloc, and, since it operates by altering the political-economy landscape, it can be thought of as a momentum-generating mechanism. The juggernaut logic exploits the fact that nations do have preferences over trade arrangements in the way that individual consumers have preferences over consumption bundles. The nature of the trade deal proposed can affect a nation’s ranking over choices (unilateral versus reciprocal, for example). Moreover, nations’ rankings over choices – unlike the rankings of a standard Walrasian consumer – are pathdependent since historical liberalization can affect the current political strength of various pro- and anti-trade special-interest groups. There are, however, a number of building-bloc logics that assume nations – like Walrasian consumers – have exogenous preferences over outcomes. The next section considers the easiest of these, assuming a representative consumer and assumes that the government acts to maximize this individual’s well-being.
33
Agriculture in Europe is a good example. The customs union formation realigned special interests in the Community in a way that fostered higher agricultural tariffs. Since EEC tariffs on agricultural goods were not bound until the 1990s, the EEC was free to raise its agricultural tariffs without appealing to the Article 24 exception. Since EEC tariffs were bound for manufacturing goods, EEC nations need the Article 24 exception to establish the common external tariff (CET), and this led them to respect the Article’s requirement that the CET not be higher on average; roughly speaking, France lowered its tariffs, Germany raised its, and the Benelux nations did little since their tariffs were initially between the French and German levels. In this way, the EEC customs union formation probably reduced the overall size of the EEC import-competing sector in manufacturing but raised it in agriculture.
54
richard baldwin
3.3.2 Kemp–Wan building-bloc logic The assertion that trade blocs may be building blocs in a static Walrasian world is as easy and as general as the assertion that they may be a stumbling bloc. Starting from a world where all nations have MFN tariffs, the question is: Can a group of nations always raise their collective welfare by forming a trade bloc? If the answer is ‘yes’, then a piecemeal enlargement of the bloc will raise bloc members’ well-being monotonically. Bloc members attain the highest welfare when all nations are part of the bloc. In this world, the formation of a single bloc should trigger a domino effect that leads to worldwide free trade. The Kemp–Wan theorem tells us that the answer to the question above is always ‘yes’ in a Walrasian setting when nations have access to international lump-sum transfers (Kemp–Wan) or a complete set of commodity taxes and subsidies (Dixit–Norman). Kemp and Wan (1976) – which is probably the first formal contribution to the building/ stumbling bloc discussion – make exactly this point. See Aghion, Antras and Helpman (2004) for an elaboration of the Kemp–Wan argument that uses modern co-operative game theory concepts.34 While this Kemp–Wan building-bloc logic is flawless, it falls down on the very real-world problem that nations do not have access to massive lump-sum transfers. Indeed, assuming that such international transfers are a realistic possibility basically assumes away most of the core difficulties facing the international trade system (and international relations more broadly). Without international transfers, the logic of preferenceerosion stumbling blocs and cherry-picking stumbling blocs suggests that, in many blocs, some bloc members would eventually veto some enlargements. 3.3.3 Veto-avoidance building-bloc logic The exploitation/preference-erosion stumbling-bloc logic turned on the fact that bloc members could veto the move to global free trade. The veto-avoidance building-bloc logic points out that, although the bloc
34
The key concepts are ‘coalition externalities’ (Haberler’s spillover) and ‘grand-coalition superadditivity’ (global free trade is first best). The authors assume one nation is the undisputed agenda-setter and that unlimited international transfers (transferable utility in game theory parlance) are possible.
big-thin k regionalism: a critical survey
55
members could veto a multilateral trade liberalization, they cannot veto further RTAs that may eventually eliminate all tariffs globally. The explosion in the number of RTAs among small nations that has been witnessed in the new century may very well be due to a combination of the juggernaut effect and the veto-avoidance logic. We start by considering the development of an extremely common form of regionalism, namely, hub-and-spoke FTAs – where one nation (e.g. the US, the EU or India) has a network of ‘radial’ bilateral FTAs with some of its trading partners while these trading partners do not have FTAs with each other. Simplicity dictates our continued use of the symmetric Figure 2.2 framework, so we arbitrarily bestow hub status on Home. 3.3.3.1 Incentives for hub-and-spoke FTAs Roughly speaking, Home found the bilateral FTA with Partner attractive since the improved market access for Home exporters in Partner’s market more than outweighed the potential welfare losses from trade diversion in Home’s import market. This suggests that Home might also find a second bilateral FTA with RoW to be welfare-enhancing, especially given all the separability that rules Meade’s secondary and tertiary effects. As it turns out, in the RTA diagram framework, Home always gains from signing a second FTA with RoW. Intuitively, the point is that Home gains the same preferential market access as it did from the first FTA and it undoes the potentially harmful trade diversion by fully liberalizing its import market. To see this in more detail, we reinterpret Figure 2.6. On the export side, Home’s second FTA wins it preferential access to RoW’s market without giving up its preferences in Partner; this has a net welfare value of areas þ D' þ C1' þ C2' , in the left panel.35 On the import side, the second FTA brings the price in Home’s market for good 1 to the global free trade level, PFT. The welfare impact of this is the positive trade-volume effect area A' plus the conflicting terms-of-trade effects, areas –B' and þC1' . The formulas in Box 2.5 show that the overall welfare change for Home is always positive.
35
Given the separability of the markets, the second FTA with RoW would yield a price for good 3 equal to P' .
56
richard baldwin
box 2.5 the walrasian rta diagram model: equilibrium prices and welfare The model described in Box 2.1 is a system of linear equations. The equilibrating variable is the internal price in the nation that imports the good (1, 2 or 3; the price of good 4 is always unity). Solving for typical good 1 when T ¼ 0, the equilibrium price for the typical good and the equilibrium welfare of a typical nation (Home) are: 3a þ b 9a2 6ab þ 5b2 ð8Þ ; WHFT ¼ 12 6 The equilibrium price is identical for goods 2 and 3; internal and border prices are equal. When nations impose a symmetric MFN tariff, T, we get the MFN equilibrium price, p1MFN, and, when Home and Partner sign a bilateral FTA, we get the FTA equilibrium price p1FTA. Direct calculation shows these to be: p1FT ¼
2 1 ð9Þ p1MFN ¼ p1FT þ T; p1FTA ¼ p1FT þ T 3 3 The MFN price applies to all three goods, but the FTA price concerns only the liberalized goods (good 1, which Home imports from Partner, and good 2, which Partner imports from Home). Under the bilateral FTA, the price of good 3 remains at the MFN level (additive separability and free trade in the fourth good eliminate all cross-good effects). The prohibitive MFN tariff is T ¼ 1/2b; we assume throughout that T < 1/2b. The so-called optimal tariff (Nash tariff) is T ¼ 1/8b. The welfare of a typical FTA partner, Home, under MFN tariffs and bilateral FTA, are given by the first two expressions in: 2 b WHMFN ¼ WHFT T ; WHFTA ¼ WHFT þ T T 2 ; 3 9 ð10Þ 5b 2 FTA FT WR ¼ WR T 9 The third expression gives the welfare of the excluded nation under the FTA. The welfare levels of Home and RoW under the hub-and-spoke FTA arrangement are: 2b 2 b 7 ð11Þ T þ T 2 ; WRH&S ¼ WRFT T T 2 9 9 9 9 Comparing RoW welfare under free trade and H&S, we see the gain from moving to free trade is bT/9 þ 7T2/9 > 0. WHH&S ¼ WHFT þ
Would RoW accept Home’s offer of a second FTA? As it turns out, RoW gains from such an FTA as long as T is not too high (see Box 2.5 for details). That is, the hub-and-spoke situation is better for RoW than
big-thin k regionalism: a critical survey
57
the initially disadvantaged position when it was excluded from the Home–Partner FTA. On the export side, an FTA with Home would improve RoW’s market access a great deal (its export price would rise from the depressed level of P' – T up to the free trade price PFT) and the liberalization on the import side would have the usual positive trade-volume effect and conflicting terms-of-trade effects (identical to those experienced by Home in its first FTA). The threshold tariff for RoW to gain from the hub-and-spoke arrangement is calculated in Box 2.5. Plainly, Partner will be harmed by the formation of the hub-and-spoke system around Home. Its preferences in Home are eroded and it receives nothing in compensation. Partner would thus like to veto Home’s second FTA, but, except in extraordinary circumstances, third nations cannot veto FTAs; the main exception is that of customs unions. Functioning customs unions, however, are quite rare in the modern world. Customs unions require supranational decisionmaking capacity to keep all external tariffs in line despite changes in anti-dumping duties, special unilateral preferences to third nations (GSP, etc.) and tariff changes in multilateral trade talks. In fact, the groups of nations that manage such coordination fall into exactly two types: the EU and nations involved in super-hegemon relations (France and Monaco, Switzerland and Liechtenstein, and the South African Customs Union, etc.). In fact, the real world is covered with hub-and-spoke trade arrangements, so we assume henceforth that Partner has no veto over Home’s FTA policy and the hub-and-spoke system gets set up. 3.3.3.2 Incentives for spoke-spoke FTAs The story, however, is not finished. As it turns out, the two spokes may find a spoke-spoke FTA to be advantageous and this would achieve global duty-free trade (although trade would not necessarily be free, due to the exclusion of various ‘sensitive sectors’, rules of origin and cumulation). To see this, note that the hub-and-spoke FTA puts Home in an enviable position – giving it the benefits of free trade as far as its imports go and preferential market access for all of its exporters. In this sense, hub-and-spoke bilateralism might be thought of as another example of the preference-erosion/exploitation stumbling-bloc logic; Home would veto WTO talks aimed at achieving global free trade. However, this simple world can reach global duty-free trade without multilateral talks. An FTA between Partner and RoW would do the job. So, would Partner and RoW be interested in an FTA?
58
richard baldwin
Taking hub-and-spoke bilateralism as the point of departure, it is clear that the spokes – Partner and RoW in the example at hand – have a very different view of global free trade than does the hub. Taking Partner as an example, a move to global free trade would do nothing to erode Partner’s preferences in Home, since those were already eroded by Home’s second FTA. For Partner, the shift to the global free trade regime would involve a standard exchange of market access with RoW; Partner would see its export price to RoW rise from P' – T to PFT for good 3, and RoW would see a symmetric border price rise for its exports of good 2 to Partner (see Figure 2.2). The attendant liberalization of the two nation’s import markets would have the usual conflicting tradevolume and terms-of-trade effects, but overall the two nations could find the exchange to be welfare-enhancing. In fact, Partner and RoW would always prefer global free trade to the hub-and-spoke situation, as the formulas in Box 2.5 confirm. This is certainly not to be taken as a general result. It does, however, illustrate how regionalism could be a building bloc in a world where overall free trade would be in the interest of all nations, but achieving the goal is blocked by nations fearing erosion of their preferences. See Lloyd (2002) for a clear development of the veto-avoidance logic. This line of thinking is one strand in the widely discussed ‘competitive liberalization’ logic of Bergsten (1996a).
3.3.4 Related logics: induced liberalization and protection Before ending this review of the helps-or-hinders literature, it is useful to cover two economic mechanisms that link RTAs and MFN tariffs without formally making the connection with multilateral trade talks. Both mechanisms consider the impact of RTAs on a nation’s MFN stance in the absence of a new MTN. The first links RTAs to unilateral MFN liberalization. The second looks at how an RTA can lower or raise a nation’s effective MFN tariff rate. 3.3.4.1 RTAs and unilateral liberalization: are RTA and MFN tariffs complements or substitutes? The building-bloc logics examined above directly address the issue of whether RTAs help or hinder the attainment of global free trade. Here, we look at a related, but logically distinct, question: What is the impact of an RTA on the tariffs a nation would find unilaterally optimal to impose on third nations? Intuitively, the question is whether preferential tariffs are complements or substitutes for MFN tariffs.
big-thin k regionalism: a critical survey
59
The easiest way to organize the various mechanisms is to start from Meade’s formula for the welfare impact of any trade-policy change in a Walrasian economy, namely, TdM – Mdp*. A nation choosing its bilateral tariffs optimally would view this as a first-order condition and set it to zero to find its optimal tariff. The optimal bilateral tariffs are:36 dp Tod ¼ Mod ð12Þ dM od where ‘o’ indicates the origin nation and ‘d’ the destination nation (i.e. the nation choosing the tariffs). In general, anything can happen to Tod when the nation signs a free trade agreement since – according to the Slutsky equation – the direct and cross-good income and substitution effects of the FTA-induced price changes could raise or lower the righthand side of (12). Attempts to resolve the inherent ambiguity have led to several economic mechanisms being stressed in the literature. 3.3.4.2 Preferential and MFN tariffs as complements If the RTAinduced price changes have little impact on the equilibrium slopes of the third-nation import-supply curves, then Tod is likely to fall since RTAs typically reduce an RTA member’s trade with third nations (Harberler’s spillover), i.e. Mod is likely to fall. For example, in the simple RTA diagram model presented in section 2.1 above, the import-supply curves are linear, so dp*/dM does not change, but Haberler’s spillover lowers third-nation trade, and so Home’s optimal tariff on RoW exports fall. Another mechanism that yields complementarity turns on the general principle that taxes become more distortionary when the cross-product variance of rates increases (the so-called uniform tax rate principle). This is a feature of many economic models, especially when administrative and enforcement considerations are taken into account. (This is why most nations impose fairly even indirect tax rates across products.) Since the RTA automatically makes the import tax structure more uneven, there is some presumption that the RTA makes the third-nation tariffs more distortionary. In models where this is true, nations are likely to lower third-nation tariffs when it re-optimizes its trade tax structure, i.e. RTAs encourage nations to lower applied MFN tariffs.
36
Divide both sides by pod* to get the optimal ad valorem tariff in terms of the importsupply elasticity.
60
richard baldwin
3.3.4.3 Preferential and MFN tariffs as substitutes The most obvious mechanism that suggests the substitutes result (i.e. nations find it optimal to raise third-nation tariffs after having signed an RTA) concerns the market power of the new bloc. If the RTA allows RTA members to better coordinate their third-country tariffs, they are likely to raise external tariffs since they will have more purchasing power than before. This, of course, presumes three fairly unrealistic things: they can coordinate external tariffs; the governments share sufficiently similar objective functions; and their external tariffs are not subject to WTO bindings (or they are willing to violate their WTO commitments). Since most effective RTAs are among developed nations whose tariffs are almost universally bound at near-zero levels (apart from a few lowvolume items) and such nations rarely violate their WTO bindings, this mechanism is probably of little real-world relevance except in a few commodities (agriculture before the Uruguay Round), and a few lowtrade-volume RTAs among developing nations. 3.3.4.4 Who did what, when? Contributions to the literature that have looked at the complementarity-versus-substitutes effects include, inter alia, Reizman (1985), Kennan and Reizman (1990), Krugman (1991a, 1993), Bond and Syropolous (1996a), and Ornelas (2005, 2008). 3.3.4.5 Imported MFN liberalization and protection A closely related line of reasoning considers the automatic impact of FTAs on the external protection of FTA members, when members impose different tariffs on third nations. Under some circumstances, the FTA effectively lowers the higher MFN tariff (imported MFN liberalization); in other circumstances, the FTA effectively raises the lower MFN tariff (imported MFN protection). A good example of imported MFN liberalization can be found in North–South FTAs. The concept can be explained intuitively with reference to Mexico’s experience. Mexico signed FTAs with the US and Canada in 1994 (NAFTA) which phased in tariff cuts over ten to fifteen years. Mexican MFN tariffs were (and still are) much higher than US and Canadian MFN tariffs, but, as NAFTA brought Mexican prices down to the US internal level, domestic prices in Mexico came to resemble those that Mexico would have observed if it had lowered its MFN tariffs to US levels. To put it differently, the high Mexican MFN tariffs became irrelevant since the same goods could be purchased from the US duty-free and the US internal price was linked to the world price
big-thin k regionalism: a critical survey
61
via its low MFN tariff (leaving aside the small sectors still protected by high US MFN tariffs, such as clothing, textiles and footwear). In this sense, Mexico ended up ‘importing’ the US’s low MFN tariffs.37 The argument can be made more precise with Figure 2.10. The two left panels show import demand of the US (left-most) and Mexico (middle); the right panel shows the world market for the good under consideration. The US total supply curve is shown in Mexico’s panel for reasons that will become clear. The US initially imposes a zero tariff on imports from RoW while Mexico imposes a tariff of TMX on imports from both the US and RoW. When Mexico eliminates duties on US imports, US–made goods can enter Mexico duty-free. Since the Mexican internal price is initially above the US internal price, US firms sell to the Mexican market and in doing so drive down Mexico’s internal price to the US internal price – which of course is just the world price. US production entirely displaces Mexican imports from RoW and the Mexican MFN tariff becomes irrelevant. The feasibility of this outcome is established by noting that the US supply at Po is more than sufficient to cover the entire Mexican import demand (point 2 is to the right of point 1). Note that there would be a secondary effect on world prices as the US expands its imports. In the diagram, the new world price would be at the intersection of the dotted MD curve and the XSRoW curve. For simplicity’s sake, this second-order impact is not shown in the two left-most panels. If one combines this imported MFN liberalization with the juggernaut logic, the FTA can eliminate all the firms in Mexico that would have otherwise opposed MFN liberalization. That is, Mexican industry has no interest in lobbying for the maintenance of high Mexican MFN tariffs since those tariffs provide no protection to Mexican industry. In the case at hand, the Mexican government signed a vast array of FTAs to exchange its now-politically useless MFN tariffs against preferential access for its exporters. 37
Extension of this analysis led to the ‘unsustainability of the FTAs’ proposition. Vousden (1990 p. 234) argues that Home would be tempted to lower its MFN tariff to just under that of Partner in order to recapture the tariff revenue and Partner would have an incentive to reply, with the resulting race-to-the-bottom tariff making FTAs ‘unsustainable’. Richardson (1995) extended and popularized the result. The main results in Shibata (1967), Vousden (1990) and Richardson (1995) – the irrelevance of rules of origin and the unsustainability of FTAs – are of little relevance to real-world policy concerns (rules of origin are at the heart of many current policy debates and FTAs, not customs unions, are by far the most prevalent form of PTAs).
62
richard bald win euros Partner
euros Home
XSRoW
Po +TMX
SUS
1
Po
RoW border price
2
Po
MDUS
MDUS+MX
MDMX
Imports
Figure 2.10
MoMX
Imports
Imports
Imported MFN liberalization
Since developed nations (the North) tend to have much lower MFN tariffs on most manufactured goods than developing nations (the South), the mechanism suggests that an important implication of North–South FTAs for the world trading system is the way they lower the South’s resistance to further liberalization. Given that most of the South does not participate in MTN tariff-cutting exercise on the basis of reciprocity, the North–South FTAs are one of the few ways of triggering juggernaut effects in developing nations. 3.3.4.6 Rules of origin and imported MFN protection The opposite result – an RTA importing MFN protection to a nation with low MFN tariffs – can occur when highly restrictive rules of origin are imposed. The argument can be illustrated with reference to NAFTA. Since the US’s first foray into regionalism – the 1965 US–Canada Auto Pact – US and Canadian rules of origin on autos have been highly restrictive. One of Canada’s motives in pushing for the trilateralization of the US– Mexico free trade agreement was to extend its restrictive rules of origin to Mexico and thereby avoid the undermining of the Auto Pact. The rules of origin forced Mexican-based car producers to import parts and components from the US or Canada instead of from third nations. As before, NAFTA equalized US and Mexican internal prices, but this meant that the Mexican prices were linked to the world prices via the higher MFN tariffs in the US and Canada. In this way, ‘imported MFN protection’ occurred. NAFTA, with its rules of origin, had effects that mimicked a rise in the Mexican MFN tariffs to the US and Canadian
big-thin k regionalism: a critical survey
63
levels. The imported-MFN-protection reasoning is most often associated with Krueger (1993), although it has played a role in the literature discussed above.38 Although the distortionary impact of rules of origin is limited by the level of the MFN tariff, the all-or-nothing feature of rules of origin for final goods can lead to large ‘effective rates of protection’. For example, if a US$20,000 NAFTA-origin car pays zero tariff while the same nonNAFTA car would pay 5 per cent, a rule of origin that stipulates that a particular component must be made inside NAFTA could make it economic to pay up to US$1,000 more for the local versus imported component. While the distortion in the final good market is limited to 5 per cent, the distortion in the component market can be much larger (this is the traditional effective rate of protection logic). 3.3.4.7 Who did what, when? Shibata (1967), Vousden (1990), Richardson (1993, 1994, 1995), and Grossman and Helpman (1995) are all important contributors to or users of this line of analysis.
3.4
Bargaining-model stumbling/building-bloc logics
The stumbling/building bloc mechanisms discussed above resonate strongly with real-world considerations, in my judgment, since they took advantage of the simple institutional features of real-world tariffcutting in RTAs and MTNs (see section 3.1 above). This, however, is not how the Big-Think Regionalism started thinking about the issue (Krugman 1991b, 1993). As Krugman (1993 p. 58) puts it: ‘In this realm of foggy discussion it is natural for economists to grab hold of any analytic tool they can find, even if they are ill-adapted to the work at hand.’ I do not believe that the literature that followed up this lead by Krugman has contributed to policy insights since it is built on foundations of sand. Nevertheless, a large literature has developed around it, and many academic participants in the Big-Think Regionalism debate refer to the literature’s insights, so I will review the basic logic here. When Krugman wondered how regionalism would affect the GATT, the tool he grabbed was simple bargaining game theory with two nations that are considering setting tariffs co-operatively (under GATT), or non-co-operatively (Nash tariffs). As Figure 2.11 shows, both nations 38
See note 37 above.
64
richard baldwin Nation 2’s tariff, t2
Nash tariffs
t *2
GATT tariffs Efficient bargaining curve tariffs
t *1
Figure 2.11
Nation 1’s tariff, t1
An economic theory of the GATT
prefer the co-operating outcome. He notes: ‘Trade bargaining . . . is characterized by a Prisoners’ Dilemma. This Dilemma arises in part from a terms-of-trade effect of conventional optimal tariff analysis, but also (and presumably in practice mostly) from the effect of each country’s tariff on the other country’s producer interests’ Krugman (1993 p. 72). He goes on to invoke all the usual theorems of repeated games to think about the building/stumbling-bloc issue and concludes: ‘Trade liberalization must be supported by the belief of countries that if they cheat they will lose from the subsequent collapse of the cooperative outcome.’39 The crux of his analysis is to examine the impact that an exogenously formed RTA has on the costs and benefits of cheating.
3.4.1 Terms-of-trade approach in a nutshell Much of Krugman’s reasoning is informal, so it is worth spelling it out explicitly. The whole analysis turns on three equilibrium welfare levels: W
GATT
is the level of a nation’s welfare with global co-operation (GATT tariffs);
39
This is not a new point. It is very clear in the discussion of Johnson (1953), but probably dates much further back. Indeed, the notion that a quid pro quo would be mutually advantageous was probably well understood by trade diplomats as far back as Roman times.
big-thin k regionalism: a critical survey
65
Nash
W is national welfare under non-co-operative tariffs; and cheat W reflects the nation’s welfare when its government ‘cheats’, i.e.
chooses a tariff to maximize its own welfare when the foreign government embraces its GATT tariffs. There are two logical steps in the approach. The first step consists of the obvious point that tariffs are worse than a zero-sum game from the global perspective so some form of co-operation could be Paretoimproving, but nations have an incentive to cheat. Formally, this is a prisoners’ dilemma and it arises when Wcheat > WGATT > WNash. The second step involves a dynamic game that models when co-operation is sustained. As Krugman notes, co-operation is self-enforcing when the gains from cheating are more than offset by the losses from the (infinite) punishment. Taking d as the discount factor, the present value of cooperating forever in symbols is WGATT / (1 – d). If co-operation is to be sustained, this must exceed the one-period gain from cheating, Wcheat, plus the present value of the infinite sequence of the Nash outcomes that kick in next period after the foreigners realize cheating has occurred, dWNash / (1 – d). Clearing the (1 – d) terms, the condition for selfsustaining global free trade is: W GATT >ð1 dÞW cheat þ dW Nash In words, each nation compares the value of welfare under co-operation to a weighted average of the cheating outcome and the Nash outcome. So far this is a trivial relabelling of an undergraduate lecture on repeated games. The contribution of this approach comes in considering how an RTA changes the three levels, Wcheat, WFT and WNash. Krugman (1993) asks whether the formation of a trade bloc among nations makes those nations more or less able or willing to co-operate. His answer is: It can cut either way. Krugman’s core insight – an insight that has been followed up in a dozen articles since – is that RTAs typically reduce the RTA members’ trade with the rest of the world, and so reduce both the costs and the benefits of cheating. Since these work in opposite directions, some bargain-approach papers find that RTAs are building blocs (thus making cooperation more likely) while others find that RTAs are stumbling blocs. 3.4.1.1 Who did what, when? This approach came to be known as the terms-of-trade approach after Kyle Bagwell and Bob Staiger,
66
richard baldwin
and students formally modelled the issues, starting with Bagwell and Staiger (1996a, 1996b). For examples of this sort of application to the regionalism questions, see Bond and Syropoulos (1996b), Campa and Sorenson (1996), Bond, Syropoulos and Winters (1993), Bagwell and Staiger (1996a), Conconi and Perroni (2000), Conconi (2000), Yi (1996), and Ornelas (2005, 2008).
3.4.2 Critique of the self-enforcing approach to tariff-liberalization The approach set out by Krugman seems entirely natural to readers schooled in game theory that was developed to explain strategic interactions among private agents. This sort of model resonates strongly when thinking about how OPEC sustains its cartel, or chemical companies agree to fix the price of vitamin C. When it comes to Big-Think Regionalism, however, it is vitiated by two factually incorrect assumptions: (1) that the cheating period is long enough to make it tempting; and (2) that punishment can be usefully modelled as consisting only of tariff changes. First, real-world tariffs are part of a nation’s tax code and are thus entirely transparent – at least in the advanced industrialized nations that account for more than 80 per cent of world trade. The domestic laws of the US, the EU and Japan, for example, require their governments to publicly announce the tariff schedule. Consequently, tariff-cheating is detected by foreigners at exactly the same time the change is implemented (if not before). See Box 2.6 for a real-world example of a surprise tariff cut being detected immediately. This means that the cheating period is zero, i.e. d is unity, so the sustainability condition becomes WGATT > WNash. In other words, as in the stumbling-bloc logics discussed above, nations co-operate if and only if it is in each case in their own interest to do so. The whole ‘selfenforcing’ module is irrelevant. This is a lethal problem for the approach since its only contribution was to examine how an RTA affected the costs and benefits of cheating for one period. The second critical flaw is even more damaging to the relevance of its insights. When Krugman set out the basic approach, he took the repeated Prisoners’ Dilemma model straight off the self – including its assumption that the punishment can only come in the form of a change in the variable of interest. That is, he assumed that nations can only punish via tariffs. This restriction of the strategy space makes perfect
big-thin k regionalism: a critical survey
67
box 2.6 nixon’s tariff ‘cheating’ A quite extreme example can be found in the surprise announcement of a 10 per cent rise in all US tariffs on 15 August 1971. The announcement was made on national television by President Nixon, acting under the emergency authority in the Trade Expansion Act (authority that was inserted to allow the US to respond instantly to changes in the international environment, including sudden tariff hikes by foreigners). The nightly news coverage of the announcement included interviews with European and Japanese policy-makers discussing the policy change even before it was implemented the next morning. The only other legal option the US had was an Act of Congress, which would have been equally public. Interestingly, the 10 per cent surcharge was to take immediate effect, but bureaucratic delays in US Customs and confusions over the exact coverage meant that implementation was not immediate. The reaction by foreigners, however, was immediate and the ‘import surcharge’ was formally removed after just four months. US tariff revenue as a percentage of US imports was lower in 1971 than in 1970.40 While this episode demonstrates that nations have and might use tariffs as a stick to get their way (in this case, the problem was the European and Japanese refusal to revalue their currencies despite the growing US trade deficit), it also clearly illustrates that there is no period in which the cheater could profit before the others work out what has been done.
sense in the analysis of collusion among firms. It does not make sense when it comes to the co-operation in the global trade system. In the real world, co-operation in the GATT/WTO system is supported by the threat of a breakdown in co-operation far broader than tariffs. The system’s members can and do use a wide range of ‘carrots and sticks’ to induce trade co-operation: foreign development aid, military aid, political support in the international arena, participation in NATO, US troops stationed in Germany and Japan, etc. Moreover, if international trade co-operation did breakdown, the costs would far exceed lost markets for exporters. It would halt or seriously hinder international co-operation on many pressing issues – climate change, international terrorism, money laundering, organized crime, humanitarian crises and the control of illegal drugs, to mention just a few. Formally, we would 40
See www.state.gov/r/pa/ho/frus/nixon/iii/8698.htm. You can watch Nixon announce it at http://themessthatgreenspanmade.blogspot.com/2008/03/nixon-ends-gold-conver tibility.html.
68
richard baldwin
include this in the repeated game by giving the players the ability to impose large costs (unrelated to tariffs) on any player that deviated from co-operation. In summary, the basic idea that trade co-operation should be modelled explicitly is not useless. Trade co-operation is almost surely dependent upon nations’ perceptions of the costs and benefits of deviating. Unfortunately, the terms-of-trade approach is focusing on the trees and missing the forest. Its insights as to whether an RTA makes cheating on a tariff more or less attractive misses the point altogether, since the main carrots and sticks supporting co-operation have nothing to do with tariffs. This means that RTAs do not have a first-order impact on the underpinnings of international trade co-operation – at least not via its impact on the costs and benefits of deviating from co-operation. The mainstream stumbling/building-bloc ideas discussed in sections 3.2 and 3.3 above ignored the sustainability issue. They just assumed that nations would keep their word if they did agree to go to free trade. This strikes me as a fairly reasonable assumption even though threats are explicitly or implicitly part of all forms of co-operation. There is little to gain from explicitly modelling the threats since the most important forms of retaliation are entirely outside the model.
3.5
Other links from RTAs to MTNs
There are a number of points made in the literature that do not fit neatly into the stumbling/building-bloc framework as I have delineated it. One line of reasoning views a nation’s MFN tariffs, or yes–no stance on multilateral co-operation, as dependent upon the strength of various domestic special-interest groups. The Big-Think Regionalism question here is whether an RTA weakens or strengthens the pro-trade or antitrade interest groups. At one end, Winters (1993) argues that regionalism (especially the EU on agriculture) strengthened the hand of protectionists since it worsened Olsen’s Asymmetry (i.e. winners from protectionism are few in number and easy to organize, while the losers are dispersed, numerous and difficult to organize politically). He calls this the ‘restaurant bill’ problem. Just as diners at a table where the bill will be split equally tend to order too much, the EU tended to grant too much protection to farmers. At the other end, De Melo et al. (1993), Richardson (1994) and Panagariya and Findlay (1996) argue that an RTA tends to dilute the influence of special-interest groups via various mechanisms.
big-thin k regionalism: a critical survey
69
Another important line of thinking asserts that the formation of RTAs creates forces that induce nations to begin and/or finish multilateral trade talks. For example, Lawrence (1991), Sapir (1993) and WTO (1995) all argue that the threat of regionalism was a critical element in inducing GATT members to initiate the Uruguay Round and to accept the final Uruguay Round agreement. Windham (1986) makes the same argument for the Tokyo Round. Bergsten (1996a) dubs this ‘competitive liberalization’ and asserts that regionalism fosters multilateralism and vice versa. A somewhat related line of thinking, which has not been formalized, is that RTAs are a testing grounds for the GATT/WTO (Lawrence 1996, Bergsten 1996b). The prime example here is the EU, which dealt with deeper-than-tariff-cutting liberalization for decades before the issues arrived on the GATT agenda in the Tokyo and Uruguay Rounds. See Ludema (1996) for a partial formalization of the idea. Another line of thinking suggests that RTAs can provide commitments that boost the credibility of a nation’s policy reforms (Fernandez and Portes 1998). This was explicitly mentioned by Mexico in its request to the US for an FTA.
4
Multilateralism and regionalism
The analytic contributions discussed above help frame our thinking on whether regionalism is likely to help or hinder multilateral trade liberalization. The bulk of the theoretical literature over the past two decades, however, has been concerned with an analytically distinct question: Would more regionalism be good or bad for world welfare? The question of whether RTAs raise or lower welfare is empirical, so one might wonder why so many theoretical articles were published on the topic. The answer lies in the history-of-thought. When leading thinkers such as Larry Summers, Paul Krugman and Jagdish Bhagwati set the intellectual ‘terms of reference’ for the regionalism literature in the early 1990s, the EU was the only effective regional arrangement. Since it was widely viewed as unrepresentative, the analysis was conducted by speculating on two scenarios – what the world would look like with, and without, more regionalism. The comparison of these two scenarios was taken as providing insight on whether spreading regionalism would raise or lower world welfare. From the perspective of the new century, this literature looks distinctly dated. The world has seen a great deal of regionalism develop,
70
richard baldwin
and it does not look anything like the speculated scenarios of the theoretic models from the 1990s. Since my goal is to review the main theoretical literature, even that which is no longer very useful, I cover the ‘Is bilateralism bad?’ papers, extracting the literature’s main insights without focusing too much on ‘Who did what, when?’ issues.
4.1
Is bilateralism bad?
This literature was kicked off by Paul Krugman’s 1991 article entitle ‘Is Bilateralism Bad?’ (Krugman 1991a). In typical Krugman style, it laid out a seductively simple model, asked it a seductively simple question, and found an answer that was provocative and seductively simple – world welfare would be lowest with three symmetric blocs.41 This provocation produced an impressive string of contributions from authors who felt that Krugman’s speculated scenarios model left out critical elements of reality. To understand the basic logic behind the three-is-worst result, it is important to note that the simplicity of Krugman’s model rules out most of the standard effects of RTAs. On the production side, he assumes each nation produces a single good, so no production distortions are possible. Since nations and blocs are all symmetric, no nation gains on net from taxing imports. While a tariff tends to lower the cost of imports via the usual terms-of-trade effect, foreign tariffs exactly offset this since they are the same size and applied to the same amount of trade (each bloc’s imports equal its exports). In short, all welfare effects are channelled through the consumer distortion.
4.1.1 More regionalism with fixed tariffs. To fix ideas, we first hold the tariff constant while varying the size of the symmetric blocs. At one extreme is the case of only one bloc, in which case there is no tariff and thus no consumer distortion; welfare is maximized. At the other extreme there are zero blocs in the sense that all goods – including domestically produced goods – are taxed at the same rate. Since no production distortion is possible and there is no net terms-of-trade effect, this extreme also attains the first-best outcome. After all, consumer distortion is driven by distorted relative prices. 41
Note that the exercise was anticipated by Riezman (1985) and Kennan and Riezman (1990), but these did not catch on.
big-thin k regionalism: a critical survey
71
Moving away from either extreme lowers welfare. This is the key intuition for the U-shaped relationship between the number of blocs and welfare. The fact that the minimum is at three is not a result that is robust to parameter changes in Krugman’s own model, and it was shown repeatedly to fall down in less simplified models. Digging slightly deeper into the intuition, consider what happens when moving from one bloc (global free trade) to two, and then three blocs. When the world forms into two blocs, the price of imported goods rises. Since consumers do not see the true price their nation pays for goods, this leads to socially inefficient expenditure switching; localgood expenditure shares rise and imported-good expenditure shares fall. When the world reconfigures into three blocs, two things happen to the consumer distortion. Some goods that were previously untaxed become taxed; this increases the distortion. The distortion between a typical domestic good and a typical imported good, however, falls since more varieties are now taxed. The tension between these two effects is what generates the U-shaped relationship.
4.1.2 Impact of regionalism on tariffs Krugman assumes the blocs are customs unions and that they set the tariff at the naı¨ve optimal tariff level, i.e. the Nash tariff that is equal to the inverse of the bloc’s import elasticity. Since the bloc gets more market power as it enlarges, Krugman has tariffs rising as the number of blocs declines, although not very much according to his numerical simulations. 4.1.3 Extensions and modifications A series of papers considered more complicated models, with the two main complications being the introduction of national comparative advantages and trade costs. When authors introduced comparative advantage, it is easy to make the three-is-worst result go away, but the U-shaped relation remains in all the models that do not allow for production distortions. Bond and Syropoulos (1996a), for example, find that the worst number of blocs may be two, three or more.42 More interestingly, considerations of comparative advantage can reverse Krugman’s result that bigger blocs would like to charge higher 42
Srinivasan (1993) and Deardorff and Stern (1994) provide other examples.
72
richard baldwin
tariffs. The intuition can be had by linking the optimal tariff to market power and market power to comparative advantage. When nations produce entirely distinct goods, as in Krugman’s model, grouping nations into, say, two trade blocs does nothing to make the blocs’ production structures more similar, and thus market power is merely a matter of size. When nations’ production structures differ in a richer way, formations of blocs can make the blocs’ output-mix more similar and this will reduce each bloc’s market power. For example, if there are four nations, two labour-poor and two labour-rich, the formation of two blocs with one nation of each type would produce a world where blocs had equal factor endowments. In a Heckscher–Ohlin world, such a shift would extinguish the market between the two blocs and with it the incentive to charge a tariff. Of course, that is not the only possibility, but it shows that the monotonic link between bloc size and optional tariffs can be broken. Sinclair and Vines (1995) undertake a more thorough study of the bloc-size-tariff issue, examining, for instance, the case of FTAs as well as customs unions. They make the intuitive point that increasing the size of an FTA does not have a first-order impact on each FTA member’s market power and so the tendency for higher tariffs is absent. The idea that trade costs (i.e. frictional trade barriers) would affect the comparison between regionalism and multilateralism is obvious and was pointed out by Krugman (1991a) in a simple extension of his basic model. He assumes that there are continents as well as countries – where continents are defined as groups of nations that have low trade costs among themselves but high trade costs with other continents. If the inter-continental trade costs are high enough, continental free trade comes very close to approximating global first trade (since the remaining tariffs fall on very little trade). In the extreme of infinite intercontinental trade costs, the first-best outcome can be obtained with continental trade blocs. The basic point that ‘natural trade blocs’ can liberalize most of the world’s trade had strong resonance with the real-world pattern of RTAs. It also attracted a large number of papers that elaborated on Krugman’s basic insight by considering more general sets of trade costs.43 Altogether, the ‘Is bilateralism bad?’ literature is best thought of as a string of parables from which we might glean some insight into 43
See Frankel, Stein and Wei (1995, 1996), Frankel (1996), Nitsch (1996a, 1996b), Schiff (2001), and Spilembergo and Stein (1996).
big-thin k regionalism: a critical survey
73
real-world events. The two main insights, however, are not particularly difficult to grasp. The first is that spreading regionalism might or might not harm global welfare. The second is that the more regional free trade comes closer to mimicking global free trade, the more regionalized is the world trade pattern (due to transport costs or comparative advantage).
5
Endogenous bloc formation
In the early 1990s, regional liberalization seemed to be the easy route to integration. A central issue in the Big-Think Regionalism literature was therefore why countries were eager to open markets regionally but reluctant to do so multilaterally. From the new-century perspective, this question looks odd. In both North America and Europe, regionalism ran into severe political problems (NAFTA, failure to renew fast-track, Maastricht, Nice and Constitutional Treaty rejection, Turkish enlargement, etc.), while the Uruguay Round passed without much discussion once the final compromises had been completed. For example, NAFTA and the Uruguay Round came into effect in the same year. Moreover, the multilateral trade system continues to enjoy a high level of background support in all major trading nations. Yet, the slowness of the Doha Round, together with booming regionalism worldwide, has brought discussion of the causes of regionalism back to the forefront of trade policy discussions. Paul Krugman, Jagdish Bhagwati and other intellectual leaders of Big-Think Regionalism put forward many accounts of the causes of regionalism.44 The main explanations were: (1) that members were frustrated with the GATT’s slow progress, especially since the issues had become so much more complex than tariffs and the number of members had grown; (2) that regionalism negotiations were easier; (3) the US conversion from devoted multilateralist to ardent regionalist removed one of the key restraints on regionalism; and (4) the US’s aggressive unilateralism scared Western Hemisphere nations into seeking a ‘safe harbour’ against the US policies and/or a breakdown in the GATT system. In a 1993 paper, I proposed a very different explanation with the domino theory of regionalism (Baldwin, 1993b, 1995, 1997). The theory 44
See Krugman (1991a, 1991b, 1993), Bhagwati (1991, 1993), Anderson and Blackhurst (1993), Whalley (1996), Lawrence (1996), Bergsten (1996), Panagariya (1996), inter alia.
74
richard bald win
views a nation’s decision to join an RTA as endogenous and notes that Haberler’s spillover provides a de novo political economy force that might make nations change their minds after some of their trade partners form or deepen a preferential trade agreement. This section reviews the domino theory, its intellectual antecedents and its subsequent contribution to the theory of endogenous bloc formation.
5.1
Domino theory
All the contributions in this area require two basic elements – a model of the economy that connects tariff choices to economic outcomes, and a political-economy model that connects economic fundamentals to policy choices. The economic model is not very important as long as Haberler’s spillover arises, i.e. as long as the preferential policies of a nation’s partners can create export discrimination. The political-economy model only requires that exporters tend to support membership in an RTA and import-competitors tend to oppose it. The logic proceeds in two steps. First is the immediate impact of an idiosyncratic deepening of integration in an RTA. Given an initial political equilibrium where the nation in question has chosen to be outside the RTA, the idiosyncratic deepening or widening of the RTA generates new political economy forces. Specifically, non-member exporters now have a greater stake in membership – they face more discrimination if their nation stays out and greater market access if it joins. Anti-membership forces may also be strengthened in non-member nations, but, if the industrial output of export sectors is systematically larger than the output of importcompeting sectors (as is usually the case since the export sector produces for both domestic and foreign consumers) and the sectors’ political power is linked to their size, the shock raises the pro-membership forces more than the anti-membership forces. For outsiders that were previously close to indifferent to membership (politically), these changes shift the domestic political-economy equilibrium to the projoiners’ camp. The second stage starts if one non-member actually does decide to join. The PTA enlargement implies that discrimination facing the remaining non-members expands and this again heightens the promembership political-economy forces in outsiders, potentially producing a membership application from an outsider that previously found
big-thin k regionalism: a critical survey
75
it politically optimal to stay out. The cycle repeats itself until a new political equilibrium membership in the PTA obtains. If the world was marked by perfect information and synchronized periodicity in political decision-making, PTA membership bids would be perfectly coordinated and bloc enlargement would happen in a steplike fashion. Uncertainty, imperfect information and mismatches of decision timings suggest that the new political economy equilibrium may be reached only gradually. During the transition, it might look like regionalism was spreading like wildfire.
5.2 The supply side The domino theory ignores the ‘supply side’ membership – i.e. thinking about whether the incumbents would allow the applicants to join. This was not an omission but rather a strategic choice reflecting a judgment that most ‘applications’ to join or form RTAs were driven by economics for the demandeurs (often small nations) and by politics for the re´pondeurs. The issue of the economic impact on the re´pondeurs was thus ignored for parsimony’s sake. Given the rather promiscuous approach that nations are taking in the new century to RTAs, the assumption seems to me as continuing to be relevant. Much of the subsequent literature on endogenizing the membership of RTAs has focused on putting the supply side into a model with domino-like features. These studies follow the lead of Riezman (1985) in using co-operative game theory to model this ‘club formation’ issue.45 Yi (1996) was one of the first to formalize the domino logic using cooperative game theory with and without considering the supply constraint. In his model, the domino effect leads to global free trade, if membership is open (i.e. the supply side is ignored), but not if the joiners require the assent of the incumbents and the incumbents care only about how tariffs affect their national welfare. Haveman (1992), Syropoulos (1999), Abrego, Riezman and Whalley (2006), and Melatos and Woodland (2007) consider a number of extensions including asymmetric nations. Aghion, Antras and Helpman (2004) is perhaps the most complete study of what happens when the supply side is fully incorporated. To make headway and reduce the proliferation of cases, they allow for transferable utility among nations and assume there is a single nation 45
Also see Kennan and Riezman (1990), Riezman (1999) and Kose and Riezman (2000).
76
richard baldwin
proposing various configurations. While these assumptions are rather strong and rather at odds with the main outlines of international trade relations in today’s world, they do allow the authors to engage the powerful tools of game theory that were developed for simpler problems. Their main conclusion is that almost anything could happen, but one possible is that – as Yi (1996) found – the domino effect could lead to every nation joining a single RTA.
5.2.1 Asymmetric lobbying The political-economy forces driving the domino effect are strengthened by the peculiar tendency of special-interest groups to fight harder to avoid losses than to secure gains. Joining allows excluded firms to avoid damage as well as to win new commercial opportunities, so trade diversion may play a particularly important role in generating new, promembership political-economy activity. Many explanations for this ‘loser’s paradox’ are possible, but one simple economic interpretation that is relevant to the domino theory is based on unrecoverable investments, i.e. sunk costs. Entry into most industries and markets involves large unrecoverable investments in product development, training, brand name advertisement and production capacity. In such situations, established firms can earn positive profits without attracting new firms, but only insofar as these profits constitute a fair return on the entry investments, i.e. sunk costs create quasi-rents, not pure rents. Given that firms in an industry will already have incurred the sunk costs, deepening of an existing bloc or formation of a new one will destroy quasi-rents, and thus generate strong, de novo political forces pushing the government to redress the new discrimination. The most direct way would be to join the bloc, but other modalities are possible. Governments of excluded nations may seek to restore quasi-rents by calling for a multilateral trade round, or forming a new trade bloc among excluded nations.46 6
Directions for future research and concluding remarks
In 2008, a Big-Think Regionalism review might seem like an intellectual indulgence from two distinct angles.
46
See Baldwin and Robert-Nicoud (2002, 2005) for a formal treatment of the ideas and Baldwin (1993b) for an early formal model.
big-thin k regionalism: a critical survey
77
1 Regionalism is here to stay. Regionalism has been raging for two decades and it shows no signs of abating. Since the existing RTAs are not going to be disbanded and more will surely be signed, discussion of whether these RTAs help or hinder multilateral liberalization is something of a luxury. The real-world issue facing policy-makers is how to increase the likelihood that these RTAs help the world trade system and reduce the likelihood that they hurt it. While there may be some room for highly abstract reasoning on this issue, the devil is in the detail. Thinking up ways of making regionalism fit in better with multilateralism will require highly detailed knowledge of matters such as rules of origin, and RTAs’ treatment of non-tariff barriers. The chapters in this volume provide the most comprehensive effort to gather such information and should therefore provide an excellent springboard for future empirical and theoretical work. 2 Deep, multilateral integration. For the GATT’s first thirty years, multilateralism meant shallow integration (mainly tariff-cutting). Deep integration (liberalization of tariffs and many behind-theborder measures) was the purview of regionalism.47 This began to change with the 1979 Tokyo Round, although it was not until the Uruguay Round’s Single Undertaking that deep integration became an integral part of multilateralism. In 2008, the WTO is dealing with, or talking about dealing with, a wide range of areas where liberalization requires discipline of behind-the-border measures – trade in services, trade-related intellectual property rights, traderelated investment measures, multilateral investment disciplines more generally, technical barriers to trade, government procurement, competition policy, and trade facilitation. A critical issue now facing policy-makers is whether the deep integration initiatives in RTAs are throwing up impediments to deep multilateral liberalization. From this perspective, the exclusive focus on border measures in the helpor-hinder literature seems somewhat amiss.
47
Until the 1980s, regionalism principally referred to European PTAs that were implemented (EFTA and the EEC) and a handful of developing-nation schemes that were never effectively implemented, the main exception being the ties between Australia and New Zealand. The EEC involved deep integration from its inception, but the trend accelerated with the 1986 Single European Act and its extension to EFTA via the European Economic Area (EEA) negotiations. The ‘Down Under’ pair embraced deep integration with their 1984 Closer Economic Relationship (CER).
78
richard bald win
But does it matter that the stumbling/building-bloc theory deals almost exclusively with tariff liberalization? Can one not model deep integration as the removal of barriers that – like tariffs – drive a wedge between internal and external prices? Answering these questions requires detailed knowledge of exactly what sort of deep integration is going on in the RTAs. The chapters in this book provide an excellent starting point. Importantly, the findings suggest that deep integration in RTAs may indeed be creating incompatibilities that will impede deep integration at the multilateral level – at least in some areas. Several of the chapters provide evidence that at least two ‘families’ of RTAs are emerging as far as deep integration schemes are concerned – one based on the NAFTA model and another based on the EU model. For example, the chapter on technical barriers to trade (TBTs) suggests that the EU family is fostering a convergence to the EU’s product standards and regulations, while the NAFTA family is fostering convergence to international norms. Although the EU and international norms are not always in conflict, it is easy to think of situations where this trend in RTAs will hinder rather than help multilateral TBT liberalization.
6.1
Future research topics
One of the astounding contrasts that emerge from the detailed mapping of RTAs is the extent to which developing nations accept disciplines in RTAs that they resist at the WTO level. In areas across the board – intellectual property rights, tariff bindings, TBT liberalization, access commitments in services, investment agreements, government procurement, subsidy disciplines, trade facilitation, and competition policy – the RTAs reviewed in the chapters show that developing nations have accepted things that they refuse even to discuss at the multilateral level. Does this indicate that something is wrong with the WTO’s negotiating procedures? An important topic for future theoretical and empirical research is the identification of the determinants of deep integration in RTAs and the failure of such initiatives at the multilateral level. A second topic of research concerns the deep difference between the political economy of shallow and deep integration. For example, while discrimination is relatively simple when it comes to tariffs, some behind-the-border measures do not lend themselves to preferences. For example, many TBTs and services trade restrictions are justified on ‘good governance’ grounds – protection of consumers, etc. – that do not logically admit preferences. If the restrictions are necessary to protect
big-thin k regionalism: a critical survey
79
consumers from fraud from one nation, then it seems natural that they should be applied to imports from all partners. Moreover, many behind-the-border measures, especially in services, act to shield incumbents from competition from other domestic firms as well as from foreign firms. In such a situation, any liberalization with respect to foreigners is likely to bring on competition from domestic rivals as well. Such considerations suggest that the trade political-economy models of tariff-liberalization need to be modified when thinking about behindthe-border measures. Theoretical advances in the political economy of various forms of deep integration should be greatly assisted by the mappings contained in the chapters in this volume. The near future should see many empirical studies trying to determine why some RTAs include various deepintegration initiatives while others exclude them. The answers will surely involve some general considerations – for example, rich nations inevitably place greater trust in the legal and governance structures of other rich nations and such trust is necessary for many of the deeper forms of integration. Other answers, however, are likely to be highly specific to particular areas. For example, services trade liberalization that goes beyond GATS may be systematically easier between nations that share similar legal and/or educational environments (as is often the case between former colonizing nations and their former colonies). A third topic – one that would involve a high level of abstraction – would be to think about the design of a WTO negotiating framework that would result in successful deep integration in the same way that the GATT’s framework fostered successful tariff-cutting.
6.2
Concluding remarks
The bulk of this paper presented a survey regarding the analytics of the ‘classic’ debate on regionalism versus multilateralism. The intent of the paper, however, was to suggest that there is a need to move the literature’s focus from the high theory of shallow integration to a more policy-relevant issue, the theory and empirics of deep integration in the regional versus multilateral contexts. There is also a need to advance the profession’s thinking on how the liberalization in RTAs can be made more supportive of multilateral liberalization, i.e. on how one can promote convergence/harmonization of RTAs. The vast datasets that are contained in the other chapters in this volume provide a rich stimulus to future research that goes beyond effect.
80
richard bald win
Annex Derivation of the welfare-organizing framework Consider an economy with an infinitely lived representative agent with G sectors. Some sectors consist of differentiated products, and for these sectors we assume that all varieties produced in a single country are symmetric; the total number of varieties available in sector i is denoted as ni, where ni ¼ 1 for homogenous good sectors. The consumer’s intertemporally separable preferences are: R 1 qt V ½p; n; Edt ð13Þ t¼0 e where V[p,n,E] is the instantaneous indirect utility function (i.e. indirect utility function for a moment in time) where p and n are the vectors of consumer prices and the number of varieties in each sector, and E is total consumption expenditure. Expenditure is home income less home savings, S, where income is the sum of home factor income, revenue from tariffs and other import barriers (assumed to be returned lump-sum to consumers) and pure profits, i.e.: E ¼ wL þ rK þ Tm þ — S
ð14Þ
where w and r, and L and K, are the prices and stocks of home’s labour force and capital stocks, m is the vector of sectoral trade (imports enter with a positive sign, exports with a negative sign), T is the vector of the specific-tariff equivalent of home-imposed DCR import barriers so Tm is trade rent term that includes tariff revenue (we ignore export taxes and subsidies), — is pure profit and S is savings. Note that T /p – p*, where p* is the border price. The border price is what the country actually pays for imports and what it actually gets paid for exports. This price need not equal the world price in the case of frictional trade barriers, or quantitative restrictions where foreigners get the trade rents. The stock of labour, L, is assumed to be fixed, but the capital stock, K, is endogenous. Pure profits are related to the vector of local prices, p, the vector of sectoral outputs, Q, and the vector of average costs, a, according to: — ¼ ðp aÞQ;
ai ½w; r; xi
ci ½w; r; xi ; xi
Qi ni xi
ð15Þ
where Qi is a typical element of the sectoral output vector Q, ai is a typical element in the vector of sectoral average costs a, defined as
big-thin k regionalism: a critical survey
81
the total cost of a typical firm (recall that all domestic firms within a sector are assumed to be identical), ci[w,r,xi], divided by typical firm output, xi; as usual, total cost is a function of w, r and firm output x. Notice that pure profits are decomposed here. That is, — is defined as if all goods in a particular sector were sold at the domestic (i.e. local) price, p. For perfect competition sectors and for sectors marked by Dixit–Stiglitz monopolistic competition, this does not matter since, in both cases, firms charge the same producer price in all markets (perfectly competitive firms never price discriminately across markets, and non-discrimination, i.e. mill pricing, is optimal under Dixit–Stiglitz competition). However, for oligopolistic sectors, producer prices can vary across markets. For such sectors, the profitability of such price discrimination shows up in the (p – p*)m term in the E definition. For example, suppose markets are segmented and there are costs to exporting, so a profit-maximizing oligopolist would typically charge a lower producer price for exports (as in the Brander–Krugman reciprocal dumping model). Then the total profit of the firm would be paqa þ pbqb – a[w,r,qa þ qb](qa þ qb), where the ‘a’ indicates the producer price and quantity for local sales and ‘b’ indicates the producer price and quantity for exports, but in notation of (15), the profit is pa(qa þ qb) – a[w,r,qa þ qb](qa þ qb) þ (pb – pa)qb with the last term showing up in the trade rents term.
Calculations Totally differentiating V and converting utils (which is the unit of measure in dV) into dollars by dividing through by the scalar VE yields (the units of VE are dollars per util): Vp dV Vn Vn ¼ ð Þdp þ dn þ dE ¼ Cdp þ dn þ dE VE VE VE VE
ð16Þ
where Vp and Vn are vectors of partial derivatives. The second expression follows from the first using Roy’s identity. To calculate dE, we totally differentiate the definition of E, (14), and use T ¼ p – p* to get: dE ¼ Ldw þ Kdr þ rdK þ ðp pÞdm þ mdp mdp þ ðp aÞdQ þ Qðdp daÞ dS
ð17Þ
82
richard baldwin
Where da ¼ {(cwdw þ crdr)/x} þ {cr/x – c/x2}dx, and {@a/@x} ¼ {(@c/@x)/x} – {c/x2}, so: X @c i dw @c i dr @ai Q þ Qi þ Qi dxi Qda ¼ i i @w x @r xi @x i i ð18Þ X @ai ¼ Ldw þ Kdr þ Q dxi i i @x i The second expression follows from the first by Shepard’s lemma (e.g. (@ci/@w) equals the optimal labour demand for a typical firm and in equilibrium, the total demand for all firms in the economy must equal labour supply L). Plugging (6) and (5) into (4) and using –C – m – Q, and defining @a/@x ¼ ax, the expression for home welfare changes simplifies to: dV =VE ¼ ðp pÞdm mdp þðp aÞdQ þ Qðax Þdx þ ðVn =VE Þdn þ rdK dS
ð19Þ
This expression is our organizing framework.
Discussion The first row shows effects that occur even when one ignores scale economies and imperfect competition; these are referred to as NICNIR (no imperfect competition, no increasing returns), or Walrasian effects. As in the public finance literature, welfare effects in a Walrasian world stem from quantity changes times initial price wedges and price changes times net trades. We refer to the first as the trade volume effect and to the second as the trade price effect (in certain special cases, the border price effect is equivalent to the terms-of-trade effects). The second row shows the ICIR (imperfect competition and increasing returns) effects. That is, the effects that appear when we assume the economy is marked by scale economies and imperfect competition. The first effect is the production-rent effect (i.e. if there are pure profits in a sector, increased production in that sector tends to improve national welfare). In a common special case, it corresponds to the profit-shifting effect that is well known in the strategic trade policy literature. The second term is the scale effect (–ax is positive under scale economies) average cost falls as output increases. The third term includes two distinct effects, the well-known variety effect (i.e. an
big-thin k regionalism: a critical survey
83
increase in the total number of varieties available tends to improve domestic welfare) and a location effect. The location effect concerns implications for whether a particular variety is produced locally or is imported. For example, when trade is costly, domestic welfare is higher when a variety is produced domestically instead of abroad since this allows avoidance of the trade cost. The third row constitutes the accumulation effects, i.e. the welfare implications of induced capital formation and the induced savings that is required. (Recall that, with balanced trade and no trade in factors, savings must equal investment.) Induced capital formation has two counteracting welfare effects. A higher capital stock raises income and thereby expenditure. This is reflected in the first term, and it is the positive aspect of induced capital formation. However, induced capital formation means that consumption must be foregone. This negative influence on the current flow of utility is captured by the second rowthree term. Accumulation effects concern the way in which trade policy can affect a nation’s stock of productive resources. Typically, we focus only on the impact on a nation’s human capital stock, physical capital stock and knowledge capital stock (i.e. level of technology), assuming away Malthusian effects. Since the accumulation of human, physical and knowledge growth is the source of all per capita output growth, many authors refer to accumulation effects as growth effects, or dynamic effects. Interested readers can easily extend this framework to capture additional effects. Considerations, such as adjustment costs that drive a wedge between income and expenditure, enter as did dS. Factors that affect wealth – environmental damage or changes in the value of vintage capital – could be included by the full intertemporal indirect utility functions that include current wealth and/or putting environment variables – such as emissions – in the cost functions. Considerations of income distribution could be incorporated by using an indirect social welfare function that included a Gini coefficient. An X-inefficiency term could be introduced into the cost functions. All sorts of price wedges could be inserted, and external economies could be introduced via the average cost function.
84
richard baldwin
box 2.7 juggernaut effect maths Using the basic Walrasian model from Box 2.3, we must add two elements: the political-economy model of tariff choice, and free entry. Now the number of firms per nation and per industry can vary. The national supply curves, taking good 1 as an example, are: p ¼ b þ X1H =nH ;
p ¼ X1P =nP ;
p ¼ X1R =nR ;
0
ð20Þ
where the ni’s are the number of symmetric firms in nation-i, and the prices refer to the prices actually faced by the various firms (with tariffs, these will differ). The supply curves for goods 2 and 3 are isomorphic, except Partner has the comparative disadvantage in good 2 and RoW in good 3. Firm-level operating profit is (p – b)2 / 2 for Home firms and (p)2 / 2 for Partner and RoW firms. Assuming entry costs are subject to congestion, the fixed entry cost in ‘f’ time ni for nation-i, where f > 0 is a parameter, the free entry conditions under the MFN regime are: ðp bÞ2 ðp T Þ2 ¼ n2H f ; ¼ n2i f ; i ¼ P; R 2 2 and the free-entry costs under the FTA regime are similar. The equilibrium prices under the MFN and FTA regimes are: pMFN ¼
3a þ T ð2 þ nP þ nR Þ þ bnH ; 3 þ nP þ nR þ nH
pFTA ¼
3a þ T ð1 þ nR Þ þ bnH 3 þ nP þ nR þ nH
ð21Þ
ð22Þ
Plugging (22) into the appropriate free-entry condition, we can solve for the FE curve, i.e. the n’s as a function of the T. The MFN solutions are: bT nMFN ¼ nMFN ¼ nMFN ; ; nMFN H R P R f pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 289 þ 400T 800T 2 10T nMFN ¼ þ R 3 3 6
ð23Þ
where the last expression takes numerical values for a, b and f to reduce the clutter (there is an analytic solution but it is too cumbersome to be revealing); specifically, this takes a ¼ 1, b ¼ 1/4 and f ¼ 1/10. For the FTA regime, the n’s as functions of the T’s are: bT T FTA ; nFTA nFTA ¼ nFTA þ ; H ¼ nR P R f f pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 3 10T 921 þ 1200T 3200T nFTA ¼ þ R 4 3 12
ð24Þ
Notice that, since nR is lower in the FTA than in the MFN regime (RoW firms face a lower border price under the FTA), the size of the Home import-competing sector is smaller with the FTA.
big-thin k regionalism: a critical survey
85
box 2.7. (cont.) The political-economy model of tariff choice is based on a politically realistic objective function where consumer surplus, the various producer surpluses (import-competing and exporting sectors’ profit net of fixed costs) and tariff revenue are weighted by coefficients that are exogenously determined by the nation’s political system. To keep things simple, we take the weight to be unity on everything except the import-competing sector’s producer surplus, so the Home government’s objective function is: G ¼ apM þ CS þ TR þ pX where the p’s are producer surplus in the import-competing sector (M) and export sectors (X) respectively, CS is consumer surplus and TR is tariff revenue. The first three terms always depend upon T, while the fourth term is a function of T only under reciprocity. Thus: *
*
*
*
Gunil ¼ apM ½T ; n þ CS½T ; n þ TR½T ; n þ pX ½n *
*
*
*
GMFN ¼ apM ½T ; n þ CS½T ; n þ TR½T ; n þ pX ½T ; n
ð25Þ
Here we impose a very simple version of reciprocity in MTN talks; the foreign tariffs, TP and TR, equal the Home tariff, T. Under the FTA regime, we set the Home– Partner T’s to zero exogenously. The building/stumbling-bloc issues then turn on whether the Home government finds it optimal to choose a higher or lower T – under reciprocity – after the FTA has been implemented. The first-order conditions of these two objective functions are the GFOCunil and GFOCMFN in the diagram; * these show how the politically optimal tariff depends upon the vector of n’s, n. Using the solutions for the n’s from the FE conditions in the GFOC conditions, we can see what T would be chosen under the three scenarios: unilateral tariff setting, MFN tariff setting with reciprocity without an FTA between Home and Partner, and MFN tariff setting with reciprocity but with the FTA. An analytic solution for the T is available, but, since it is the solution to a system of quadratic equations, it is too unwieldy for building understanding. If we use the numerical values for a, b and f as above, and take the political overweight of import-competing profits, a, to be two, we get: Tunil ¼ 0:09;
TMFNðrecipÞ ¼ 0:06;
TFTAðrecipÞ ¼ 0:005
This demonstrates the juggernaut building bloc formally. Intuitively, the FTA downsizes the overweighted import-competing profits, and this in turn makes the Home government find it politically optimal to set a lower tariff in reciprocal trade talks.
86
richard bald win References
Abrego, Lisandro, Riezman, Raymond, and Whalley, John, 2006. ‘How Often Are Propositions on the Effects of Customs Unions Theoretical Curiosa?’, Journal of International Economics 68(1): 59–78. Aghion, P., Antra`s, P., and Helpman, Elanan, 2004. ‘Negotiating Free Trade’, NBER Working Paper No. 10721. Anderson, Kym, and Blackhurst, Richard, 1993. ‘Introduction and Summary’, in Kym Anderson and Richard Blackhurst (eds.), Regional Integration and the Global Trading System, London: Harvester Wheatsheaf. Bagwell, Kyle, and Staiger, Robert W., 1993. ‘Multilateral Tariff Co-operation During the Formation of Regional Free Trade Areas’, NBER Working Paper No. 4364. 1996a. ‘Regionalism and Multilateralism Tariff Co-operation’, Paper prepared for International Economics Association Round-Table Conference, International Trade Policy and the Pacific Rim, 15–17 July 1996. 1996b. ‘Preferential Agreements and the Multilateral Trading System’, mimeo. 2002. The Economics of the World Trading System, Cambridge, MA: MIT Press. Baier, S. L., and Bergstrand, J. H., 2003. ‘Economic Determinants of Free Trade Agreements’, Journal of International Economics 64: 29–63. Balassa, Bela, 1967. ‘Trade Creation and Trade Diversion in the European Common Market’, Economic Journal 77. Baldwin, Robert, 1970. Non-Tariff Distortions of International Trade, Washington DC: Brookings Institution. 1984. ‘The Changing Nature of US Trade Policy Since World War II’, in R. E. Baldwin and A. O. Krueger (eds.), The Structure and Evolution of Recent US Trade Policy, Chicago: University of Chicago Press. 1985. The Political Economy of US Import Policy, Cambridge, MA: MIT Press. 1993a. ‘Asymmetric Lobbying: Why Governments Pick Losers’, GIIS mimeo, available by email. 1993b. ‘A Domino Theory of Regionalism’, CEPR Discussion Paper No. 857; NBER Working Paper No. 4465. 1994. Towards an Integrated Europe, London: CEPR. 1995. ‘A Domino Theory of Regionalism’, in Richard Baldwin, P. Haaparanta and J. Kiander (eds.), Expanding Membership in the European Union, Cambridge: Cambridge University Press, pp. 25–48; republished in Jagdish Bhagwati, Pravin Krishna and Arvind Panagariya (eds.), Trading Blocs: Alternative Approaches to Analyzing Preferential Trade Agreements, Cambridge, MA: MIT Press, 1999, pp. 479–502. 1997. ‘The Causes of Regionalism’, The World Economy 20: 865–88. 2006. ‘Multilateralising Regionalism: Spaghetti Bowls as Building Blocs on the Path to Global Free Trade’, World Economy 29(11): 1451–518.
big-thin k regionalism: a critical survey
87
Baldwin, Richard, Haaparanta, P., and Kiander, J. (eds.), 1995. Expanding Membership in the European Union, Cambridge: Cambridge University Press. Baldwin, Richard, and Robert-Nicoud, F., 2002. ‘Entry and Asymmetric Lobbying: Why Governments Pick Losers’, NBER Working Paper No. 8756. 2005. ‘Juggernaut Model – The Lego Version’, Geneva Graduate Institute of International Studies, mimeo. Baldwin, Richard, and Venables, Anthony, 1995. ‘Regional Economic Integration’, in Gene Grossman and Kenneth Rogoff (eds.), Handbook of International Economics, vol. III, Amsterdam: North Holland, pp. 1597–644. Baldwin, Richard, and Wyplosz, Charles, 2006. The Economics of European Integration, 2nd edn, McGraw-Hill. Berglas, Eitan, 1983. ‘The Case for Unilateral Tariff Reductions: Foreign Tariffs Rediscovered’, American Economic Review 73(5): 1141–2. Bergsten, C. Fred, 1991. ‘Commentary: The Move Toward Free Trade Zones’, Federal Reserve Bank of Kansas City Economic Review 76: 27–35. 1996a. ‘Competitive Liberalization and Global Free Trade: A Vision for the Early 21st Century’, Institute for International Economics Working Paper 96–15. 1996b. ‘Globalizing Free Trade: The Ascent of Regionalism’, Foreign Affairs May/June 1996. 1998. ‘Fifty Years of the GATT/WTO: Lessons from the Past for Strategies for the Future’, Institute for International Economics Working Paper 98–3. Bergstrand, J., and Baier, S., ‘Economic Determinants of Free Trade Agreements’, Journal of International Economics 64: 29–63. Bhagwati, Jagdish, 1971. ‘Trade-Diverting Customs Unions and WelfareImprovement: A Clarification’, Economic Journal 81: 580–7. 1973. ‘Reply to Professor Kirman’, Economic Journal 83: 895–7. 1991. The World Trading System at Risk, Hemel Hempstead: Harvester Wheatsheaf. 1993. ‘Regionalism and Multilateralism: An Overview’, in Jaime de Melo and Arvind Panagariya (eds.), New Dimensions in Regional Integration, Cambridge: Cambridge University Press. Bhagwati, J. N., and Panagariya, A., 1996. ‘Preferential Trading Areas and Multilateralism: Strangers, Friends or Foes?’, in Jagdish Bhagwati and Arvind Panagariya (eds.), Free Trade Areas or Free Trade? The Economics of Preferential Trading Agreements, Washington DC: AEI Press. Blackhurst, David, 1972. ‘General versus Preferential Tariff Reduction for LDC Exports: An Analysis of the Welfare Effects’, Southern Economic Journal 38: 350–62. Bond, Eric W., and Syropolous, Constantinos, 1996a. ‘The Size of Trading Blocs: Market Power and World Welfare Effects’, Journal of International Economics 40: 411–37.
88
richard bald win
1996b. ‘Trading Blocs and the Sustainability of Inter-Regional Cooperation’, in M. Canzoneri, W. Ethier and V. Grilli (eds.), The New Transatlantic Economy, Cambridge: Cambridge University Press. Bond, E. W., Syropoulos, C., and Winters, Alan, 1996. ‘Deepening of Regional Integration and External Trade Relations’, CEPR Discussion Paper No. 1317. Cadot, Olivier, De Melo, Jaime, and Olarreaga, Marcelo, 1996. ‘Regional Integration and Lobbying for Tariffs Against Non-Members’, mimeo, University of Geneva. Campa, J. M., and Sorenson, T. L., 1996. ‘Are Trade Blocs Conducive to Free Trade?’, Scandinavian Journal of Economics 98(2): 263–73. Cheng, I., and Wall, H., 2005. ‘Controlling for Heterogeneity in Gravity Models of Trade and Integration’, Federal Reserve Bank of St Louis Review 87: 49–63. Collie, D. R., 1997. ‘Bilateralism Is Good: Trade Blocs and Strategic Export Subsidies’, Oxford Economic Papers 49: 504–20. Conconi, P., 2000. ‘Trade Bloc Formation under Imperfect Competition’, The Warwick Economics Research Paper Series (TWERPS) 571, University of Warwick, Department of Economics. Conconi, P., and Perroni, C., 2000. ‘Issue Linkage and Issue Tie-in in International Negotiations’, Fondazione Eni Enrico Mattei, Note di Lavoro: 57/2000. Cooper, C., and Massell, B., 1965. ‘A New Look at Customs Union Theory’, Economic Journal 75: 742–7. Corden, W. Max, 1957. ‘The Calculation of the Cost of Protection’, Economic Record 33: 29–51. 1965. ‘Recent Developments in the Theory of International Trade’, Special Papers in International Economics, No. 7, Princeton University, International Finance Section. Croome, John, 1995. Reshaping the World Trading System: A History of the Uruguay Round, New York: DIANE Publishing. De Melo, Jaime, and Panagariya, Arvind (eds.), New Dimensions in Regional Integration, Cambridge: Cambridge University Press. Deardorff, Alan V., and Stern, Robert M., 1994. ‘Multilateral Trade Negotiations and Preferential Trading Arrangements’, in Alan V. Deardorff and Robert M. Stern (eds.), Analytical and Negotiating Issues in the Global Trading System, Ann Arbor, MI: University of Michigan Press. Destler, I. M., 1992. American Trade Politics, Washington DC: Institute for International Economics. Dixit, Avinash, and Norman, Victor, 1980. Theory of International Trade, Cambridge: Cambridge University Press. Fernandez, Raquel, and Portes, Jonathan, 1998. ‘Returns to Regionalism: An Analysis of Non-Traditional Gains from Regional Trade Agreements’, World Bank Economic Review 12: 197–220.
big-thin k regionalism: a critical survey
89
Finger, Michael, Ulrich, Reincke, and Castro, Adriana, 1999. ‘Market Access Bargaining in the Uruguay Round – Rigid or Relaxed Reciprocity?’, Policy Research Working Paper Series 2258, The World Bank. Frankel, Jeffrey A., 1996. Regional Trading Blocs, Washington DC: Institute for International Economics. Frankel, Jeffrey A., and Wei, Shang-Jin, 1995. ‘Trading Blocs and the Americas: The Natural, the Unnatural, and the Super-Natural’, Journal of Development Economics 47: 61–95. 1996a. ‘Regionalization of World Trade and Currencies: Economics and Politics’, in Jeffrey A. Frankel (ed.), The Regionalization of the World Economy, Chicago: Chicago University Press. 1996b. ‘Continental Trade Blocs: Are They Natural and Super-Natural?’, in Jeffrey A. Frankel (ed.), The Regionalism of the World Economy, Chicago: Chicago University Press. Freund, C., 2000a. ‘Different Paths to Free Trade: The Gains from Regionalism’, Quarterly Journal of Economics 115: 1317–41. 2000b. ‘Multilateralism and the Endogenous Formation of Preferential Trade Agreements’, Journal of International Economics 52: 359–76. Gatsios, K., and Karp, L., 1995. ‘Delegation in a General Equilibrium Model of Customs Unions’, European Economic Review 39: 319–33. Gehrels, Franz, 1956–7. ‘Customs Unions from a Single-Country Viewpoint’, Review of Economic Studies 24: 61–4. Goto, J., and Hamada, K., 1994. ‘Economic Integration and the Welfare of Those Who Are Left Behind: An Asian Perspective’, RIEB Kobe University, Discussion Paper No. 47. 1995. ‘EU, NAFTA and Asian Responses: A Perspective from the Calculus of Participation’, NBER Working Paper No. 5325. 1997. ‘EU, NAFTA, and Asian Responses: A Perspective from the Calculus of Participation’, in Takatoshi Ito and Anne O. Krueger (eds.), Regionalism versus Multilateral Trade Agreements, Chicago: University of Chicago Press, pp. 91–118. 1999. ‘Regional Economic Integration and Article XXIV of the GATT’, Review of International Economics 7(4): 555–70. Grossman, Gene, and Helpman, Elhanan, 1994. ‘Protection for Sale’, American Economic Review 84: 833–50. 1995. ‘The Politics of Free Trade Agreements’, American Economic Review 86: 667–90. 2001. Special Interest Politics, Cambridge, MA: MIT Press. Haberler, Gottfried, 1936. The Theory of International Trade with Its Applications to Commercial Policy, New York: Macmillan. Hathaway, Oona, 1998. ‘Positive Feedback: The Impact of Trade Liberalization on Industry Demands for Protection’, International Organization 52: 575–612.
90
richard bald win
Haveman, Jon, 1996. ‘Some Welfare Effects of Sequential Customs Union Formation’, Canadian Journal of Economics 29(4): 941–58. Helpman, Elhanan, 1995. ‘Politics and Trade Policy’, CEPR Discussion Paper No. 1269. Hoekman, Bernard, and Kostecki, Michel, 2001. The Political Economy of the World Trading System, Oxford: Oxford University Press. Holmes, Tammy, 2005. ‘What Drives Regional Trade Agreements That Work?’, HEI Working Paper No. 07/2005, Graduate Institute of International Studies, Geneva. Hufbauer, Gary, 1989. Background Paper for the Free Trade Debate: Reports of the 20th Century Fund Task Force on the Future of American Trade Policy, New York: Priority Press. 1990. Europe 1992: An American Perspective, Washington DC: Brookings Institution. Hufbauer, Gary C., Erb, J. S., and Starr, H. P., 1980. ‘The GATT Codes and the Unconditional Most-Favoured Nation Principle’, Law and Policy in International Business, No. 1. Hufbauer, Gary C., Schott, Jeffrey, and Clark, D., 1994. Western Hemisphere Economic Integration, Washington DC: Institute for International Economics. Humphrey, D. C., and Ferguson, C. E., 1960. ‘The Domestic and World Benefits of a Customs Union’, Economia Internazionale 13: 197–213. Irwin, Douglas, 1993. ‘Multilateral and Bilateral Trade Policies in the World Trading System: A Historical Perspective’, in Jaime De Melo and Arvind Panagariya (eds.), New Dimensions in Regional Integration, Cambridge: Cambridge University Press, pp. 90–119. 1996. Against the Tide: An Intellectual History of Free Trade, Princeton, NJ: Princeton University Press. Jackson, John H., 1991. ‘The European Community and World Trade: The Commercial Policy Dimension’, Discussion Paper No. 298, Institute for Public Policy Studies, University of Michigan. Jacquemin, A., and Sapir, A., 1988. ‘European Integration or World Integration?’, Weltwirtschaftliches Archiv 124: 127–39. Johnson, Harry, 1953. ‘Optimum Tariffs and Retaliation’, Review of Economic Studies 21: 142–53. 1957. ‘Discriminatory Tariff Reduction: A Marshallian Analysis’, Indian Journal of Economics 5: 39–47. 1958. ‘Marshallian Analysis of Discriminatory Tariff Reductions: An Extension’, Indian Journal of Economics 6: 177–82. 1960a. ‘The Economic Theory of Customs Unions’, Pakistan Economic Review 10: 14–30. 1960b. ‘The Cost of Protection and the Scientific Tariff’, Journal of Political Economy 68: 327–45.
big-thin k regionalism: a critical survey
91
1965. ‘An Economic Theory of Protectionism, Tariff Bargaining and the Formation of Customs Unions’, Journal of Political Economy 73: 256–83. 1974. ‘Trade Diverting Customs Unions’, Economic Journal 84: 618–21. Karacaovali, Baybars, and Lima˜o, Nuno, 2005. ‘The Clash of Liberalizations: Preferential versus Multilateral Trade Liberalization in the European Union’, CEPR Discussion Paper No. 4973. Katz, E., 2001. ‘Bias in Conditional and Unconditional Fixed Effects Logit Estimation’, Political Analysis 9: 379–84. Kemp, Murray, and Wan, Henry Y., 1976. ‘An Elementary Proposition Concerning the Formation of Customs Unions’, Journal of International Economics 6: 95–8. Kennan, John, and Riezman, Raymond, 1990. ‘Optimal Tariff Equilibria with Customs Unions’, Canadian Journal of Economics 23: 70–83. Keohane, R. O., 1984. After Hegemony: Cooperation and Discord in the World Political Economy, Princeton, NJ: Princeton University Press. Kirman, Alan P., 1973. ‘Trade Diverting Customs Unions and Welfare Improvement: A Comment’, Economic Journal 83(331): 890–4. Kose, M. Ayhan, and Riezman, Raymond, 2000. ‘Understanding the Welfare Implications of Preferential Trade Agreements’, Review of International Economics 8(4): 619–33. Kowalczyk, C., 1990. ‘Welfare and Customs Unions’, NBER Working Paper No. 3476. Kowalczyk, C., and Sjostrom, T., 1992. ‘Bringing GATT into the Core’, Economica 61: 301–17. Krishna, Pravin, 1998. ‘Regionalism and Multilateralism: A Political Economy Approach’, Quarterly Journal of Economics 113: 227–51. Krishna, Pravin, and Panagariya, Arvind, 2000. ‘A Unification of Second-Best Results in International Trade’, Journal of International Economics 52: 235–57. Krueger, Anne O., 1993. ‘Free Trade Agreements versus Customs Unions’, Journal of Development Economics 54: 169–87. Krugman Paul, 1991a. ‘Is Bilateralism Bad?’, in Elhanan Helpman and Assaf Razin (eds.), International Trade and Trade Policy, Cambridge, MA: MIT Press. 1991b. ‘The Move Toward Free Trade Zones’, Federal Reserve Bank of Kansas City Economic Review 76: 5–24. 1993. ‘Regionalism versus Multilateralism: Analytical Notes’, in Jaime de Melo and Arvind Panagariya (eds.), New Dimensions in Regional Integration, Cambridge: Cambridge University Press, pp. 58–79. Lawrence, R. Z., 1991. ‘Emerging Regional Arrangements: Building Blocs or Stumbling Blocks?’, in Richard O’Brien (ed.), Finance and the International Economy 5: The AMEX Bank Review Prize Essays, New York: Oxford University Press, pp. 23–35.
92
richard bald win
1996. Regionalism, Multilateralism and Deeper Integration, Washington DC: Brookings Institution. Levy, Philip I., 1994. ‘A Political Economy Analysis of Free Trade Agreements’, Economic Growth Centre Discussion Paper No. 718, Yale University. 1996a. ‘Lobbying and International Cooperation in Tariff Setting’, mimeo, Yale University. 1996b. ‘Comment on ‘Regionalization of World Trade and Currencies: Economics and Politics’, in Jeffrey Frankel (ed.), The Regionalization of the World Economy, Chicago: Chicago University Press, pp. 223–6. 1997. ‘A Political Economic Analysis of Free-Trade Agreements’, American Economic Review 87: 506–19. Lima˜o, Nuno, 2006. ‘Preferential Trade Agreements as Stumbling Blocks for Multilateral Trade Liberalization: Evidence for the United States’, American Economic Review 96: 896–914. 2007, ‘Are Preferential Trade Agreements with Non-trade Objectives a Stumbling Block for Multilateral Liberalization?’, Review of Economic Studies. Lipsey, Richard, 1957. ‘The Theory of Customs Unions: Trade Diversion and Welfare’, Economica 24: 40–6. Lloyd, Peter J., 2002. ‘New Bilateralism in the Asia–Pacific’, The World Economy 25: 1279–96. Ludema, Rodney D., 1996. ‘On the Value of Preferential Trade Agreements in Multilateral Negotiations’, Georgetown University Working Paper No. 97–22. Mansfield, E. D., and Reinhardt, E., 2003. ‘Multilateral Determinants of Regionalism: The Effects of GATT/WTO on the Formation of Preferential Trade Agreements’, International Organization 57: 829–62. Maskus, Keith, and Wilson, J. S. (eds.), 2001. Quantifying the Impact of Technical Barriers to Trade: Can It Be Done?, Ann Arbor, MI: Michigan University Press. Mclaren, J., 2002. ‘A Theory of Insidious Regionalism’, Quarterly Journal of Economics 117: 571–608. Meade, James E., 1955a. The Theory of Customs Unions, Amsterdam: NorthHolland. 1955b. The Theory of International Economic Policy, vol. II, Trade and Welfare, Oxford: Oxford University Press. Melatos, Mark, and Woodland, Alan, 2007. ‘Endogenous Trade Bloc Formation in an Asymmetric World’, European Economic Review 51: 901–24. Messerlin, P. A., 1983. ‘Bureaucracies and the Political Economy of Protection: Reflections of a Continental European’, Weltwirtschaftliches Archiv 117: 468–96. Milner, Helen V., 1997. Interests, Institutions and Information: Domestic Politics and International Relations, Princeton: Princeton University Press.
big-thin k regionalism: a critical survey
93
Nitsch, V., 1996a. ‘Do Three Trade Blocs Minimize World Welfare?’, Review of International Economics 4: 355–63. 1996b. ‘Natural Trading Blocs: A Closer Look’, prepared for the European Economic Association Meeting, Istanbul, August 1996. Nogues, J., and Quintanilla, R., 1993. ‘Latin America’s Integration and the Multilateral Trading System’, in Jaime de Melo and Arvind Panagariya (eds.), New Dimensions in Regional Integration, Cambridge: Cambridge University Press, pp. 278–313. Nordstrom. H., 1995. ‘Customs Unions, Regional Trading Blocs and Welfare’, in Richard Baldwin, P. Haaparanta and J. Kiander (eds.), Expanding Membership in the European Union, Cambridge: Cambridge University Press, pp. 54–78. O’Brien, D. P., 1976. ‘Customs Unions: Trade Creation and Trade Diversion in Historical Perspective’, History of Political Economy 8: 540–63. Ohyama, M., 1972. ‘Trade and Welfare in General Equilibrium’, Keio Economic Studies 9: 37–73. Ornelas, Emanuel, 2005. ‘Trade Creating Free Trade Areas and the Undermining of Multilateralism’, European Economic Review 49: 1717–35. 2008. ‘Feasible Multilateralism and the Effects of Regionalism’, Journal of International Economics 74(1): 202–24. Oye, Kenneth A., 1992. Economic Discrimination and Political Exchange: World Political Economy in the 1930s and 1980s, Princeton: Princeton University Press. 1993. Economic Discrimination and Political Exchange: World Political Economy in the 1930s and 1980s, Princeton: Princeton University Press. Panagariya, Arvind, 1996. ‘APEC and the United States’, CIES Discussion Paper, University of Adelaide. 2000. ‘Preferential Trade Liberalization: The Traditional Theory and New Developments’, Journal of Economic Literature 38(2): 287–331. Panagariya, Arvind, and Findlay, R., 1994. ‘A Political Economy Analysis of Free Trade Areas and Customs Unions’, Policy Research Working Paper No. 1261, World Bank, Washington DC. Perroni, C., and Whalley, John, 1994. ‘The New Regionalism: Trade Liberalization or Insurance?’, NBER Working Paper No. 4626. Pomfret, Richard, 1986. ‘The Trade-Diverting Bias of Preferential Trading Arrangements’, Journal of Common Market Studies 25: 109–17. 1997. The Economics of Regional Trading Arrangements, Oxford: Clarendon Prees. Richardson, M., 1993. ‘Endogenous Protection and Trade Diversion’, Journal of International Economics 34: 309–24. 1994. ‘Why a Free Trade Area? The Tariff Also Rises’, Economics and Politics 6: 79–96.
94
richard bald win
1995. ‘Tariff Revenue Competition in a Free Trade Area’, European Economic Review 39: 1429–37. Richardson, M., and Desruelle, D., 1994. ‘Fortress Europe: Jericho or Chateau d’If?’, mimeo. Rieder, R., 2006. ‘Playing Dominoes in Europe: An Empirical Analysis of the Domino Theory for the EU, 1962–2004’, HEI Working Paper No. 11/2006, Graduate Institute of International Studies, Geneva. Riezman, Raymond, 1985. ‘Customs Unions and the Core’, Journal of International Economics 19: 355–66. 1999. ‘Can Bilateral Trade Agreements Help Induce Free Trade’, Canadian Journal of Economics 32: 751–66. Sapir, A., 1993. ‘Discussion of Chapter 7’, in Jaime de Melo and Arvind Panagariya (eds.), New Dimensions in Regional Integration, Cambridge: Cambridge University Press, pp. 1491–506. 1997. ‘Domino Effects in Western European Trade, 1960–1992’, CEPR Discussion Paper. Published as ‘Domino Effects in Western European Regional Trade, 1960–92’, (2001) European Journal of Political Economy 17: 386. Schattschneider, E. E., 1935. Politics, Pressures and the Tariff, New York: PrenticeHall. Schiff, Maurice, 2001. ‘Will the Real “Natural Trading Partner” Please Stand Up?’, Journal of Economic Integration 16(2): 245. Schott, Jeffrey, 1988. ‘The Free Trade Agreement: A US Assessment’, in J. Schott and M. Smith (eds.), The Canada–US Free Trade Agreement: The Global Impact, Washington DC: Institute for International Economics. Serra, Jaime, et al., 1997. Reflections on Regionalism: Report of the Study Group on International Trade, New York: Carnegie Endowment for International Peace. Shepsle, K. A., and Weingast, B. R., 1981. ‘Political Preferences for the Pork Barrel: A Generalization’, American Journal of Political Science 25: 96–111. Shibata, H., 1967. ‘The Theory of Economic Unions: A Comparative Analysis of Customs Unions, Free Trade Areas, and Tax Unions’, in C. S. Shoup (ed.), Fiscal Harmonization in Common Markets, vol. I, Theory, New York: Columbia University Press, pp. 145–264. Sinclair, P., and Vines, D., 1995. ‘Bigger Trade Blocs Need Not Entail More Protection’, mimeo, University of Birmingham. Smith, M., 1988. ‘The Free Trade Agreement in Context: A Canadian Perspective’, in J. Schott and M. Smith (eds.), The Canada–United States Free Trade Agreement: The Global Impact, Washington DC: Institute for International Economics. Spilembergo, Antonio, and Stein, Ernesto, 1996. ‘The Welfare Implications of Trading Blocs among Countries with Different Endowments’, NBER Working Paper No. 5472.
big-thin k regionalism: a critical survey
95
Srinivasan, T. N., 1993. ‘Regionalism v. Multilateralism: Analytical Notes’ Comment’, in Jaime de Melo and Arvind Panagariya (eds.), New Dimensions in Regional Integration, Cambridge: Cambridge University Press, pp. 84–9. Summers, L., 1991. ‘Regionalism and the World Trading System’, Federal Reserve Bank of Kansas City Economic Review 76: 295–302. Syropoulos, Constantinos, 1999. ‘Customs Unions and Comparative Advantage’, Oxford Economic Papers 51(2): 239–66. Vanek, Jaroslav, 1965. General Equilibrium of International Discrimination: The Case of Customs Unions, Cambridge, MA: Harvard University Press. Viner, Jacob, 1950. The Customs Union Issue, New York: Carnegie Endowment for International Peace. 1965. ‘Introduction Guide to John Rae’s Life of Adam Smith’, in John Rae, The Life of Adam Smith, New York: Augustus M. Kelley. Vousden, Neil, 1990. The Economics of Trade Protection, Cambridge: Cambridge University Press. Whalley, J., 1993. ‘Regional Trade Arrangements in North America: CUSTA and NAFTA’, in Jaime de Melo and Arvind Panagariya (eds.), New Dimensions in Regional Integration, Cambridge: Cambridge University Press. 1996. ‘Why Do Countries Seek Regional Trade Agreements?’, NBER Working Paper No. 5552. Winham, G. R., 1986. International Trade and the Tokyo Round Negotiation, Princeton, NJ: Princeton University Press. Winters, Alan L., 1991. International Economics, 4th edn, London: Routledge. 1993. ‘The European Community: A Case of Successful Integration’, CEPR Discussion Paper No. 755. 1994. ‘The EC and Protection: The Political Economy’, European Economic Review 38: 596–603. 1995. ‘Who Should Run Eastern European Trade Policy and How?’, in Alan L. Winters (ed.), Foundations of an Open Economy: Trade Laws and Institutions for Eastern Europe, London: CEPR, pp. 19–39. 1996. ‘Regionalism versus Multilateralism’, World Bank Policy Research Working Paper 1687, Washington DC: World Bank. Wonnacott, Paul, 1987. The United States and Canada: The Quest for Free Trade, Washington DC: Institute for International Economics. Wonnacott, Paul, and Wonnacott, Ronald J., 1981. ‘Is Unilateral Tariff Reduction Preferable to a Customs Union? The Curious Case of the Missing Foreign Tariffs’, American Economic Review 71: 704–14. World Trade Organization, 1995. Regionalism and the World Trading System, Geneva: WTO. Yi, S., 1996. ‘Endogenous Formation of Customs Unions under Imperfect Competition: Open Regionalism Is Good’, Journal of International Economics 41: 151–75.
3 Market access provisions in regional trade agreements antoni estevadeordal, matthew shearer and kati suominen* Regional trade agreements (RTAs) have proliferated over the past decade around the world to cover nearly half of global trade. The number of RTAs notified to the WTO is approaching 200, while the total number of RTAs around the world exceeds 300. The global RTA spree has forged a veritable spaghetti bowl of multiple and often overlapping agreements. The various rules included in each RTA entangle the bowl further. Besides market access of goods, many RTAs today include provisions in such trade disciplines as services, investment, standards, intellectual property, and competition rules, as well as a host of issues not directly related to trade, such as the environment. While RTAs can generate important economic benefits, their spread has also produced a number of concerns. First, the manifold disciplines embedded in RTAs can introduce policy frictions that increase the costs of trading. Each new RTA rule represents a new policy for firms to consider in their export, outsourcing and investment decisions. Each also has legal, administrative and economic implications to the RTA parties. Secondly, differences in rules across RTAs can translate into transaction costs to countries operating on two or more RTA fronts simultaneously. This is a growing consideration today given that nearly all WTO members are party to at least one RTA; in some regions such as
*
The authors are, respectively, Manager, Economist/Statistician, and Trade Economist at the Integration and Trade Sector of the Inter-American Development Bank (IDB). The authors wish to thank Jaime Granados, Jeremy Harris, Jessica Luna, Fazia Pustela and Carolyn Robert of the IDB, and Robert Scollay of the APEC Study Centre of the University of Auckland for their extensive inputs and excellent comments. Maria Jose Casanovas, Santiago Florez-Gomez and Carlos M. Gutierrez Jr provided outstanding research assistance. The opinions expressed are those of the author(s) and do not necessarily reflect the views of the IDB, the WTO, or their Member countries.
96
market access provisions in rtas
97
Latin America, the average number of RTA memberships per country rises to more than half a dozen. And, thirdly, at the systemic level, the burgeoning universe of RTAs is feared to set back the implementation of the multilateral trade rules of the General Agreement on Tariffs and Trade (GATT) and the WTO Agreements, and undercut states’ incentives for multilateral trade liberalization. The concerns about RTAs have focused policy attention on GATT Article XXIV, which sets out the conditions under which the main types of RTAs – free trade agreements (FTAs) and customs unions (CUs) – are viewed as consistent with the multilateral trade rules. The Article is, however, ambiguous, and thus a source of extensive debate and a host of interpretations. Some of the most disputed issues centre on the Article’s stipulation that RTAs are to eliminate tariffs on ‘substantially all trade’ between the parties, and to do so within a ‘reasonable length of time’. Another key line of debate involves the meaning of the Article’s requirement that, besides tariffs, RTAs eliminate ‘other restrictive regulations of commerce’ on ‘substantially all trade’. Further discussions have focused on the extent to which RTAs are ‘WTOþ’, or cover a broader range of disciplines than the multilateral agreements, and, as such, regulate the behaviour of their members more – and potentially also involve deeper liberalization – than multilateral rules do. Alongside these debates has developed an important body of academic literature aimed at establishing whether RTA formation is ultimately trade-creating and conducive to multilateral trade liberalization; however, even on this front, a clear consensus has yet to emerge. Notwithstanding the forceful emergence of RTAs and the contentious debates on them, there are few rigorous efforts to disaggregate RTAs in order to analyze the operation and effects of their component provisions. This implies that assessments of RTAs’ economic outcomes have yet to disentangle the various RTA rules’ respective causal effects from each other, let alone from the effects of variables beyond RTAs – which, in turn, limits the usefulness of the arguments of both those who view RTAs as discriminatory instruments that work to obstruct global trade liberalization, and those who regard RTAs as compatible with multilateral trade rules and conducive to global free trade. Yet, an improved understanding of the regional-multilateral nexus is crucial in the face of the growing importance of RTAs in the global trading system. Particularly timely are efforts (1) to establish appropriate methods for assessing and measuring the compatibility of the RTA rules with the
98
estevadeordal, shearer, suominen
multilateral trade rules; (2) to evaluate the legal and economic implications of the system of RTAs on the multilateral trade regime; and (3) to explore the utility and feasibility of harmonizing the RTA rules towards a common standard. The purpose of this chapter is to start shedding light on the extent of compatibilities among RTAs as well as between RTAs and the multilateral trade rules in the area of market access. We strive to accomplish this by presenting a detailed mapping of six market-access provisions – tariffs, non-tariff measures (NTMs), so-called other measures, special regimes, rules of origin (RoO), and customs procedures – in fifty RTAs around the world. The focus here is on both the main texts of RTAs, and RTA annexes containing the member states’ tariff liberalization programmes. While the findings are preliminary, the chapter discusses their potential broader implications for market access issues in the multilateral trade regime, including their usefulness for sharpening the contested GATT terminology related to RTAs. The first section elaborates on some of the main debates surrounding the meaning of Article XXIV, and discusses the potential contribution of this paper in light of the existing literature on market access in RTAs. The second section presents the mappings. Section three focuses on the potential implications of the findings, while section four charts avenues for future research. Section five concludes.
1
Market access in RTAs: key debates and analytical approaches 1.1
Main debates surrounding RTAs
Preferential market access is the most fundamental characteristic of any RTA. The margin of preference that an RTA creates depends on the RTA’s market-access provisions – on whether the barriers to trade between the RTA parties are fully removed or only partly removed, and in the latter case also on the extent to which they are removed. The preferential margin also depends on the level of the most-favourednation (MFN) barriers of the RTA parties, as well as on whether the MFN barriers are increased, lowered or left unchanged when the RTA is formed. As such, the measurement of the extent of market access granted by an RTA – and any assessment of the RTA’s economic effects – has more than one dimension. While product coverage is important, the pace and extent of tariff reductions also matter, as do the parties’ trade policies towards third countries. Furthermore, the relevance of an RTA
market access provisions in rtas
99
to its members depends on whether the products of greatest export interest to them are covered by the RTA. These issues permeate the legal debates on RTAs. From a legal perspective, the key questions surrounding Article XXIV are three-fold: (1) the extent of product coverage by an RTA; (2) the length of the transition period to the RTA; and (3) RTA instruments that are deemed key arbitrators of market access. The first two issues are subject to the GATT Article XXIV stipulation that the elimination of tariffs on ‘substantially all trade’ (SAT) between the parties must occur within a ‘reasonable length of time’. Importantly, the 1979 Enabling Clause imposes less stringent requirements on RTAs involving only developing countries; while such RTAs are expected to carry out some lowering of barriers between the parties, they can stop short of complete tariff elimination. The third key question on RTAs pertains to the Article XXIV provision that, besides tariffs, RTAs eliminate ‘other restrictive regulations of commerce’ on ‘substantially all trade’. There is no clear consensus as to which of the various trade policy instruments should be regarded as ‘other restrictive regulations of commerce’. The following three parts elaborate on the various interpretations of these key GATT terms.
1.1.1 Substantially all trade GATT Article XXIV stipulates that the elimination of tariffs on ‘substantially all trade’ between the parties must occur within a ‘reasonable length of time’. Efforts to define ‘substantially all trade’ have followed two main approaches: A quantitative approach that would use a statistical benchmark, such
as a percentage of trade between the parties. The commonly suggested percentages are 90, 85 and 80 per cent. A main objection to this approach is that it would not preclude the exclusion of entire sectors from liberalization. A qualitative approach, whereby no sector (or at least no major sector) would be kept from liberalization. One difficulty of this approach is defining ‘sector’; a further question is whether the inclusion of a minor segment of a major sector would satisfy the definition. Four further approaches have been suggested as possible ways to resolve the above ambiguities: A definition of product coverage ‘in terms of a certain percentage of
tariff lines’. For instance, Australia has suggested using a threshold
100
estevadeordal, shearer, suominen
of 95 per cent of all Harmonised System (HS) tariff lines at the six-digit level. A definition based on the calculation of the percentage of trade between the RTA parties that is carried out under the preferential rules of origin applying to the RTA. A definition stating that all sectors should be included. A definition along the lines suggested by footnote 1 of GATS Article V, which precludes an a priori exclusion of any sector from an agreement. The economic rationale for the SAT requirement is not altogether clear. Laird (1999) notes that exclusion of products in which at least one of the parties is internationally competitive will indeed reduce the scope for trade creation and limit the potential welfare gains from the agreement. Meanwhile, exclusion of products in which both parties are internationally uncompetitive will lower the potential for trade diversion, and could therefore be welfare enhancing. However, it is not difficult to conceive of situations where inclusion of sensitive products could be trade diverting, even if one party is an internationally competitive producer. If the protected market is shared between several internationally competitive suppliers, the supplier that secures preferential access through an RTA may be able to displace the other competitive suppliers from the market. If the preferred partner is unable to supply the entire market, the price in the importing partner market may not fall; Panagariya (1999) points out that the result may be some trade diversion, combined with a welfare loss for the importing partner that is not offset by the gain to the exporting partner.
1.1.2 Reasonable length of time Article XXIV, paragraph 5(c), requires that ‘any interim agreement’ leading to the formation of an FTA or a CU ‘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’.1 While ‘reasonable length of time’ is generally accepted to apply only after the arrangement has been fully implemented, there are various ambiguities, such as what exactly constitutes a ‘reasonable length of time’, and whether the elimination of barriers scheduled for beyond such time should be counted towards the fulfilment of the SAT requirement. The 1
Interestingly, even though most FTAs and CUs have been, at least in part, implemented by stages, very few have expressly been notified as ‘interim agreements’ to the WTO. See WTO (2002c).
market access provisions in rtas
101
multilateral 1994 Understanding on the Interpretation of Article XXIV states that the length of time in question should exceed ten years only in ‘exceptional cases’. However, there is no guidance as to what constitutes an ‘exceptional case’.
1.1.3 Other restrictive regulations of commerce Article XXIV requires that, besides tariffs, RTAs eliminate ‘other restrictive regulations of commerce’ on ‘substantially all trade’. RTAs carry several disciplines that can impose qualifications to the extent of market access provided by tariff liberalization, such as tariff rate quotas (TRQs), special safeguards (SSGs), NTMs and RoO. However, whether these or any other RTA instruments should be subject to the ‘other restrictive regulations of commerce’ phrase is unclear. TRQs can limit the extent of market access by restricting the traded quantities of goods that are provided preferential access. While they can be interpreted as a means to smooth the importing partner’s adjustment to the effects of liberalization, such a cushioning role could be considered relatively more important in cases where TRQs are applied only during the transition period. It is in cases where TRQs continue being applied beyond the end of the transition that they can be viewed as geared to limiting market access – and thus potentially subject to the phrase ‘other restrictive regulations of commerce’. SSG provisions allow RTA members to impose additional duties (usually with the existing MFN rate at the time as a ceiling) in the event that their market is disrupted by imports from the partner. In some cases, SSGs can be invoked automatically on the basis of a price or volume trigger. Bilateral emergency actions (BEAs) generally require some formal investigation prior to their imposition. Such requirements apply sometimes also to SSGs. Preferential rules of origin, a market access instrument that has yet to be subjected to any meaningful multilateral rules,2 have long been viewed as a protectionist instrument and a particularly appropriate subject to the ‘other restrictive regulation of commerce’ phrase. Recent 2
The only multilateral effort to deal with preferential RoO is the Common Declaration with Regard to Preferential Rules of Origin annexed to the Uruguay Round’s Agreement on Rules of Origin. The declaration provides disciplines for preferential rules of origin; in particular, Article 3(c) requires that laws and regulations relating to them be published ‘as if they were subject to, and in accordance with, the provisions of Article X of GATT 1994’ that regulates the publication and administration of trade regulations. However, the declaration has yet to be acted upon.
102
estevadeordal, shearer, suominen
debates have paid attention also to the role of customs procedures in determining the actual market access in RTAs; indeed, much like in the case of RoO, the potential uses of customs procedures as non-tariff instruments have gained growing attention at the multilateral level.3
1.2
State of the literature on market access in RTAs
Despite the lively debate on the meaning of Article XXIV, there are as yet only a handful of detailed surveys of RTAs’ market-access provisions and the extent to which they comply with the various interpretations of the Article. The World Trade Organization (2002a) carries out an extensive inventory of the coverage and liberalization of tariff concessions in 47 RTAs of a total of 107 parties. The data cover tariff treatment of imports into parties to selected RTAs, tariff line treatment as obtained from individual countries’ tariff schedules, and tariff dispersion for a number of countries. Scollay (2005) performs a similarly rigorous analysis of tariff concessions in a sample of eighteen RTAs. The IADB (2002) presents an exhaustive survey of market access commitments of RTAs in the Americas, while the World Bank (2005) carries out a more general mapping of the various disciplines in RTAs around the world. Besides tariff concessions, there are surveys on RoO. WTO (2002b) breaks ground, mapping the general rules of origin in ninety-one RTAs around the world. Estevadeordal and Suominen (2006) and Suominen (2004) cover a larger set of RTAs, and include both general and productspecific RoO in their analyses. Most empirical analyses have tended to abstract from the contractual details of RTAs, employing a simple dummy variable to represent RTAs and generally assuming that RTAs liberalize all trade immediately for all partners. While RTAs are in general found to boost trade among the partners, academic literature remains divided on whether RTAs are ultimately trade-creating or trade-diverting – and whether RTAs are a stepping stone or a stumbling bloc to global free trade.4 Among some 3
4
The WTO has raised RoO to a systemic issue in the Doha Development Round, while the World Customs Organization recently highlighted customs procedures as having ‘significant influence on the economic competitiveness of nations and the growth of international trade and the development of the global marketplace’. See World Customs Organization, www.wcoomd.org. For early works on the welfare effects of RTAs and customs unions, in particular, see Viner (1950), Meade (1955a), Lipsey (1960), Johnson (1965), Mundell (1964), Corden (1972), and Kemp and Wan (1976). Richardson (1994) and Panagariya and Findlay
market access provisions in rtas
103
examples, Kemp and Wan (1976), Deardorff and Stern (1992), Baldwin (1993), Wei and Frankel (1995), Bergsten (1996), Frankel, Stein, and Weil (1997), Ethier (1998) and, on the political science side, Oye (1992) and Kahler (1995), provide grounds for believing that RTAs can be ever-expanding and propel strategic interactions conducive to global free trade. In contrast, Bhagwati (1993) argues that reduced protection between RTA members will be accompanied by increased protection vis-a`-vis outsiders, with RTAs ultimately undermining multilateral liberalization.5 However, the conclusiveness of these analyses remains hampered by the crude operationalization of RTAs. Some recent empirical studies do draw finer distinctions between RTAs. Li (2000) and Adams et al. (2003) introduce a measure on the ‘depth’ of RTAs in a gravity model. Lima˜o (2006), examining tariff concessions, finds that the United States and the EU have limited their multilateral tariff liberalization in goods traded with the RTA partners. However, Estevadeordal and Robertson (2004) and Estevadeordal, Freund and Ornelas (2005), operationalizing tariff liberalization in a number of Western Hemisphere RTAs, find that RTAs in the Americas have not only been liberalizing and conducive to trade in the region, but also helped further multilateral liberalization. Estevadeordal and Suominen (2005b) and Suominen (2004) operationalize the restrictiveness of rules of origin in some 100 RTAs, finding that, while RTAs foster trade, restrictive product-specific RoO dampen it.6 This paper aims to make four contributions to the nascent pool of mappings of RTA provisions, and, as such, help advance the empirical literature on the effects on RTAs. First, we pioneer in integrating six major market access disciplines into one single study. Secondly, unlike most existing mappings, we focus both on the main RTA texts, and on
5
6
(1996) extend the political economy analysis of RTA formation to looking at welfare implications of endogenously determined RTAs. For further works on the effects of RTAs, see Haveman (1992), Bagwell and Staiger (1993), Bond and Syropoulos (1995), Krueger (1997), Krishna (1998), Lawrence (1996), Bond, Syropoulos and Winters (2001), Melitz (2001), Coe et al. (2002), Frankel and Rose (2002), Rose (2002) and World Bank (2005). However, RoO likely also defy the traditional welfare analysis based on the unidimensional friction, the tariff, because any given RoO regime entails a layer of frictions – restrictiveness, complexity and various regime-wide components – that can work in different directions. This renders efforts to capture the welfare effects of RoO a hefty challenge at best. Furthermore, the global trade effects of the differences across RoO regimes have thus far escaped both theoretical and empirical scrutiny, as have the verification costs of RoO.
104
estevadeordal, shearer, suominen
the sectoral tariff concessions in RTA annexes and protocols. Thirdly, this paper presents three alternative measurements to examine RTA tariff schedules’ compliance with SAT and ‘reasonable length of time’. And, fourthly, this paper goes beyond the geographical coverage of many existing analyses by incorporating RTAs formed throughout the world. We also include a number of very recently concluded RTAs.
2
Descriptive mapping
This section consists of a descriptive mapping of six market access disciplines – tariffs, NTMs, other measures, special regimes, rules of origin, and customs procedures – in fifty RTAs. The mapping seeks to capture two dimensions of RTAs: coverage (or ‘comprehensiveness’) and liberalization (or ‘depth’). We seek to measure the two dimensions through two distinct approaches. The first approach involves mapping out the obligations included in the annexes and protocols of the RTAs’ market access chapters, and centres primarily on different measurements of RTA parties’ tariff liberalization schedules. The second approach focuses on the provisions spelled out in the main texts of RTAs. While the first set of indicators is both country- and product-specific and largely based on quantitative information, the second set draws on inherently regime-specific and largely qualitative information. The data in the first set of indicators are based on thirty-eight RTAs, and in the second set on a sample of fifty RTAs (see Table 3.1).
2.1 RTA coverage and liberalization: an assessment at the tariff-line level This section examines the coverage and liberalization at the tariff-line level in the thirty-eight RTAs outlined in Table 3.1. The first part consists of a survey of the overall approach of the tariff liberalization regimes – divided here into basket, sectoral and preferential tariff approaches.7 While most RTAs fall into one single category, some 7
Various prior studies characterize tariff elimination as carried out on the basis of a positive or a negative list, or as based on a certain formula. This study strives to abstract from these characteristics and classify liberalization programmes by their categorization of goods into distinct paths of liberalization. To be sure, some of the categories are more aligned with a positive-list approach, while others lend themselves to a negative-list approach.
Table 3.1 RTAs covered in the study Agreement
Year of entry into effect
Chile–China Panama–Singapore Australia–Thailand CAFTA
1/10/06 24/7/06 1/1/05 17/12/04 (SV), 03/3/05 (HO), 10/3/05 (GU), 11/10/05 (NI), 27/7/05 (US)a 5/7/97 1/11/02 1/4/04 1/8/99 1/1/04 1/1/00 1/1/95 1/3/00 1/7/01 30/11/02 1/4/05 1/1/01 28/7/03 1/1/05
Canada–Chile Canada–Costa Rica Chile–Korea Chile–Mexico China–Hong Kong, China EU–South Africa EU–Lithuania EU–Morocco EFTA–Mexico Japan–Singapore Mexico–Japan New Zealand–Singapore Singapore–Australia United States–Australia
Tariff line schedules
Trade-weighted chapters
Import-linked tariff lines
Aggregate provisions
Table 3.1 (cont.)
Agreement
Year of entry into effect
United States–Chile United States–Jordan United States–Morocco United States–Singapore Chile-central America Chile–New Zealand–Singapore– Brunei EU–Chile Mexico–Bolivia Mexico–Costa Rica Mexico–Nicaragua Mexico–Northern Triangle
1/1/04 17/12/01 1/1/06 1/1/04 15/2/02 (CR), 3/6/02 (SV) 8/11/06 (CHL); 6/06 (NZL, SGP, BRN) 1/2/03 1/1/95 1/1/05 1/7/98 15/3/01 (SV, GU), 1/6/01 (HO), 14/3/01 (MEX) 15/7/04 1/4/94 N/A 1998 (original) N/A 1/10/96 28/2/97 27/6//05
Mexico–Uruguay NAFTA United States–Peru Chile–Peru United States–Colombia MERCOSUR–Chile MERCOSUR–Bolivia ACE 58
Tariff line schedules
Trade-weighted chapters
Import-linked tariff lines
Aggregate provisions
ACE 59 Australia–New Zealand Canada–Israel Central America–Dominican Republic COMESA EU–Mexico EU–Romania EFTA–Singapore Mexico–Colombia–Venezuela Mexico–Israel New Zealand–Thailand United States–Bahrain United States–Israel Total agreements a
Refers to ratification dates.
27/6/05 27/3/83 1/1/97 7/3/02 (CR), 4/10/01 (SV), 3/10/01 (GU), 19/12/01 (HO) 8/12/94 1/7/01 1/2/95 1/1/03 1/1/95 1/7/00 1/7/05 13/12/05a 15/8/85
38
27
23
50
108
estevadeordal, shearer, suominen
employ a combination of two of the approaches. Other agreements, such as some of the more recent RTAs in the Asia–Pacific region liberalize trade immediately on all products. The second part analyzes tariff-line data emerging from the RTA parties’ tariff liberalization schedules. The third part examines alternative measurements – share of trade-weighted tariff lines that are liberalized and share of trade that is liberalized from the RTA partner in a given year – in sub-samples of twenty-seven and twenty-three RTAs, respectively.
2.1.1 Empirical survey: tariff liberalization regime models 2.1.1.1 Basket approach In a basket approach, the tariff elimination programme assigns all products into a set of distinct categories. The categories provide a timeframe and trajectory towards complete elimination of tariffs (as opposed to providing only an end-point preferential tariff, such as a reduction of tariffs from 35 per cent to 10 per cent). The baskets include TRQs, typically with a reference to an appendix with the quantities, as well as exceptions to preferential treatment (that are typically entered into a basket of continued MFN treatment).8 The United States tends to follow the basket approach, generally subjecting nearly the entire tariff universe to eventual complete tariff elimination. Some of the less visible ‘action’ in the US agreements can be found in the annexes on TRQs, where tariff liberalization generally takes place over longer time horizons and is accompanied by increasing in-quota quantities. Even sugar, a sensitive product from the US perspective, typically receives an increasing in-quota quantity (albeit from a small starting point), such that, even when sugar receives continued MFN treatment, as is the case in the Central American Free Trade Agreement (CAFTA), at a theoretical future point an infinite quantity of sugar will enter the RTA partner free of duty.9 8
9
Thailand–Australia and Thailand–New Zealand FTAs defy categorization, as they do not use any clearly defined baskets, but, rather, implement staging simply by cross-tabbed reduced tariff rates. This lends itself mostly to the basket approach, due to the use of comprehensive schedules. However, there are a large number of case-by-case trajectories, which suggests a preferential tariff approach, as well. It should be noted that the in-quota quantities (and even the existence of in-quota treatment) in these agreements differ greatly within CAFTA. Although the United States has given the same schedule with the same baskets to the other countries, the treatment within these baskets differs greatly between countries. So, although the statistics will reflect identical treatment of all Central American countries, this will not be the case, especially when considering that a number of the products subject to TRQs are those where Central America will have a strong comparative advantage (such as in sugar).
market access provisions in rtas
109
2.1.1.2 Sector approach The sector approach, typically favoured by the EU, subjects all industrial products to a general tariff elimination schedule. A separate list for exceptions and separate annexes or protocols govern the treatment of agricultural and/or fish and/or processed agricultural products. The protocols tend to be quite complex, featuring different types of regimes, such as end-point preference margins or residual preferential tariffs, TRQs, reference quantities, and a phased reduction of tariffs to a final level (which is often non-zero). The sections referring to the scope of the agreement and definitions of certain product categories are as important to understanding the process of tariff reduction as is the section on the tariff reduction programme. The EU–Chile FTA that entered into effect in 2003 diverges from the EU’s standard practice of dividing tariff elimination into separate venues. Instead, the agreement establishes a single schedule for each party that contains all products. In its category column, the schedule includes various measures that will be maintained, such as TRQs, elimination of only the ad valorem component of a mixed duty (including in cases where the non ad valorem component is linked to an entry price),10 products subjected to a tariff concession of 50 per cent of the basic customs duty, and cases where no liberalization takes place, for instance due to ‘protected denominations’. The agreements negotiated by EFTA follow the EU model in carrying a general tariff elimination programme and separate schedules for fish and agriculture. In these sectors, rather than having a single tariff liberalization schedule, EFTA’s agreements include country-specific schedules.11 2.1.1.3 Preferential tariff approach In addition to the basket and sector approaches, some agreements, such as those forged under the umbrella of the Latin American Integration Association (LAIA), place a greater emphasis on the end-point preferential tariff. The Bangkok Agreement also focuses on the end-point preferences, with additional concessions provided to less developed RTA members. Both the LAIA and Bangkok models take a positive-list approach to the concessions, whereby the schedules contain the products to which the market-access 10
11
These two cases, as in TRQs, are additional examples of where the basket incidence statistics do not adequately present what is occurring within the ‘black box’. In this paper, the data on tariff elimination in the EFTA–Mexico FTA are based on Switzerland’s tariff schedules.
110
estevadeordal, shearer, suominen
provisions of the RTA apply (as opposed to the negative-list approach, which catalogues the products to which the market-access provisions do not apply).
2.1.2 Tariff liberalization statistics This part turns to analyzing tariff-line data developed on the basis of the tariff liberalization schedules of seventy-six parties in thirty-eight RTAs.12 There are two main sets of indicators. The first set contains basic statistics that indicate the percentage of the tariff universe subject to tariff reductions, immediate tariff elimination, eventual tariff elimination, exclusions (or products subject to continued MFN treatment), and continued MFN treatment but with increasing TRQ access. The second set of indicators strives to capture the share of each individual RTA party’s tariff lines that are accorded some tariff reductions, and the share of lines that are duty-free by certain benchmark years (generally 1, 5, 10 and 15) since the launching of the RTA.13 Year 1 12
13
The tariff liberalization schedules were obtained from the Foreign Trade Information System at www.sice.oas.org, and some national sources, including websites. Some tariff data were obtained from TRAINS. Dummies are assigned according to when tariff reductions start (whether or not in year 1, 5, 10 or 15), as well as when a product becomes duty-free, and when a product that was not previously duty-free becomes duty-free. The dummies are subsequently multiplied by the number of lines with that treatment, and then divided by the total number of lines to obtain the percentage incidence. The total number of lines does not include previously duty-free lines for the incidence of reductions and non-duty-free lines that are now duty-free, but does include all lines for the total duty-free number. A line that has split treatment (i.e. part has a reduction one year and part does not) is treated as having the most generous reductions, but the least generous duty-free treatment (as a reduction on any part of a line is a reduction, but duty-free should cover a product in its entirety). Note that, in CAFTA, indicators for the Dominican Republic and each of the five Central American countries were calculated individually and then averaged together to create a single, indicative partner to the United States. The percentages for these three basic indicators include lines subject to TRQs, based on when out-of-quota tariff rates are reduced or phased out. For example, where tariff reductions or elimination are made on in-quota tariff rates, the product in question is treated as not receiving tariff reduction. Products subject to entry prices, when relevant, are counted as receiving tariff reduction, but not as having tariffs eliminated. Safeguards are not taken into account here (i.e. as interfering with tariff elimination). Split products or products partially covered by an agreement, as a general rule are accorded a reduction at the first date at which any of the various baskets accord a reduction, but are treated as having tariffs eliminated at the last date at which any of the various baskets accorded elimination. Other side-notes are dealt with on an ad hoc basis. Any TRQ, regardless of whether reductions occur on the in-quota or out-of-quota tariff rate, are counted in the TRQ incidence measure.
market access provisions in rtas
111
refers here to the year of entry into force. We focus on the absolute and relative number of duty-free lines, as well as on the speed at which concessions are made.14 Also calculated is the number of staging categories in an agreement, and some simple statistics indicating the share of tariff lines that carry TRQs. Figure 3.1 provides a view of the share of tariff lines liberalized by the partners in the thirty-eight RTAs. It maps out the shares of national tariff lines that become subject to liberalization in year 1, years 2–5, years 6–10, years 11–20, and more than 20 years into the RTA. The three-letter ISO code of each country giving the concession (i.e. the importing country) precedes the arrow, while the code of the partner country follows the arrow. Agreements formed in the Americas and particularly those signed by the NAFTA members generally liberalize trade relatively fast, with about three-quarters or more of lines freed in the first year of the agreement. The figure reveals the share of tariff lines subject to backloaded liberalization – particularly marked in Morocco and South Africa’s schedules in FTAs with the EU, largely due to the persistent protection in the agricultural protocols. Asia–Pacific RTAs stand out for being particularly frontloaded: they liberalize the bulk of the tariff universe in the first year of the RTA. This is in good part due to Singapore’s according duty-free treatment to all products upon the entry into force of its agreements. Figures 3.2a and 3.2b assess the extent to which tariff elimination is reciprocal among the RTA members by the benchmark years 5 and 10. They are sorted in a descending fashion from the least to the most reciprocal. While the parties’ respective product coverages often diverge markedly in year 5, with some partners (such as Korea) liberalizing up to twice as many lines as their partners (such as Chile), the differences shrink considerably by year 10 with the exception of EU’s agreements with Morocco and South Africa. Figures 3.3a and 3.3b display the relationship between tariff elimination and the incidence of TRQs in the 5- and 10-year benchmarks. Tariff elimination (y-axis) uses a reverse scale, so that values closer to the origin denote greater liberalization. While the diagrams measure the progress of tariff elimination at different points in time, TRQ incidence is here treated as time invariant (so that, if a TRQ is completely phased out by year 9, it will appear in the diagrams for both year 5 and year 10). 14
An argument for the third indicator is that it measures actual concessions, while the second one measures the persistence of residual tariffs.
CHL->MEX MEX->CHL CHL->CA CA->CHL CHL->PER PER->CHL MEX->BOL BOL->MEX MEX->TN TN->MEX MEX->NIC NIC->MEX MEX->CRI CRI->MEX MEX-.URY URY->MEX CAN->CHL CHL->CAN CAN->CRI CRI->CAN MERC->CHL CHL->MERC MERC->BOL BOL->MERC MERC->AND3 AND3->MERC MERC->PER PER->MERC CANUSA->MEX MEX->CANUSA USA->CHL CHL->USA USA->CAFTA CAFTA->USA USA->COL COL->USA USA->PER PER->USA USA->SGP SGP->USA USA->MAR MAR->USA USA->JOR JOR->USA USA->AUS AUS->USA EU->CHL CHL->EU EU->ZAF ZAF->EU EU->LTU LTU->EU EU->MAR MAR->EU EFTA->MEX MEX->EFTA KOR->CHL CHL->KOR JPN->MEX MEX->JPN CHL->CHN CHN->CHL PAN->SGP SGP->PAN CHL->P3 P3->CHL JPN->SGP SGP->JPN AUS->SGP SGP->AUS NZL->SGP SGP->NZL AUS->THA THA->AUS HKG->CHN CHN->HKG
National Tariff Lines (%) 100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Agreement / Concession
Year 1 (EIF)
Figure 3.1 Years 2–5 – Years 11–20 >20 Years or Never
Percentage of tariff lines duty free, by selected benchmark years
market access provisions in rtas
113
120 Party 1 Party 2 100
% of lines duty-free
80
60
40
20
0 Descending Percentage Point Difference Between Parties (Ordinal Scale)
Figure 3.2a
Reciprocity of concessions: Year 5
120 Party 1 Party 2 100
% of lines duty-free
80
60
40
20
0 Descending Percentage Point Difference Between Parties (Ordinal Scale)
Figure 3.2b
Reciprocity of concessions: Year 10
To be sure, many RTAs do not have TRQs; in such cases, the points are concentrated along the y-axis. Some agreements entail very rapid tariff elimination, with a handful of sensitive products being subject to TRQs. These appear lower on the y-axis, yet feature a significant TRQ incidence.
114
estevadeordal, shearer, suominen 0
% of Tariff Lines Duty-Free in 5 Years
10 20 30 40 50 60 70 80 90 100 0
0.5
Figure 3.3a
1
1.5 2 TRQ Incidence
2.5
3
3.5
3
3.5
Duty-free lines in 5 years v. TRQ incidence
0
% of Tariff Lines Duty-Free in 10 Years
10 20 30 40 50 60 70 80 90 100 0
0.5
Figure 3.3b
1
1.5
2 TRQ Incidence
2.5
Duty-free lines in 10 years v. TRQ incidence
Note that TRQs in RTAs are usually additional to TRQ entitlements under the WTO Agreement on Agriculture, so that the RTA parties’ existing entitlements are not affected. Nevertheless, in quota-controlled markets where the Agreement on Agriculture allocates quotas to several supplying countries, the expansion of the quota of one supplying RTA
market access provisions in rtas
115
0
% of Tariff Lines Duty-Free in 5 Years
10 20 30 40 50 60 70 80 90 100 0
2
4
6
8
10
12
14
TRQ Incidence
Figure 3.4a
Agricultural duty-free lines in 5 years v. TRQ incidence
0
% of Tariff Lines Duty-Free in 10 Years
10 20 30 40 50 60 70 80 90 100 0
Figure 3.4b
2
4
6 8 TRQ Incidence
10
12
14
Agricultural duty-free lines in 10 years v. TRQ incidence
partner will put downward pressure on prices, causing some erosion in the quota rents available to all quota-holders, while only the RTA partner is compensated by increased market access. Given the possible negative impact on other quota-holders, it is not clear that TRQs in RTAs are consistent with the WTO rules on quotas. It is also unclear
116
estevadeordal, shearer, suominen
whether Article XXIV provides a dispensation from those rules – or from GATT Article I.15 The aggregate tariff reduction statistics return a narrow range of averages for all of the agreements, but disguise what could be expected to be important variation in the speed of liberalization across product categories.16 Figures 3.4a and 3.4b take the first cut at the sectoral data, examining the relationship between tariff elimination and the incidence of TRQs for agricultural products. The diagrams indicate a stronger positive relationship between the two indicators than that emerging from the aggregate analysis. As such, they are suggestive of the sensitivity of agricultural products for many RTA partners. Box 3.1 and Tables 3.2 and 3.3 detail the operation of TRQs in CAFTA.
box 3.1 tariff rate quotas in cafta The United States presented a single schedule of tariff concessions to the Central American countries and the Dominican Republic in CAFTA. However, there are some differences in the actual concessions to each Latin American party. The differences in treatment arise from the granting of immediate elimination of duties for finite quantities of some goods by means of a tariff rate quota. While some of the parties receive duty-free access under a quota, others do not, and, while the products subject to quotas are similar across the parties, the quantities vary widely among them (see Table 3.2).17 The differences can have substantial implications, as the products in question are among the most sensitive, and as the tariff reduction takes a long time and may be subject to grace periods before actual reductions begin. 15
16
17
GATT Article I establishes disciplines on general most favoured-nation treatment and for preferential margins in arrangements that are mentioned in the Article. The Appellate Body in the Turkey – Restrictions on Imports of Textile and Clothing Products dispute found that a dispensation could be available in cases where it could be shown that the proposed measure is essential to the formation of the PTA, but did not set the criteria by which this condition could be fulfilled in practice. Viewing the percentages of lines that are duty-free by a certain benchmark year (e.g. year 10) by two-digit HS chapters may be ideal given that the level of disaggregation is detailed enough to provide distinct product categories. A four-digit approach may be useful as well, but can be excessively complex and disguise the more general trends. The best method could be to identify some two-digit chapters that have the least comprehensive tariff elimination, and then use these as priors to conduct four- or six-digit analysis within these chapters. These tables are summary versions of those used in the Comparative Guide to the Chile– United States Free Trade Agreement and the Dominican Republic-central America– United States Free Trade Agreement, a joint project of the Tripartite Committee (IDB, OAS and ECLAC). The categories in the US table are in order of appearance in the US General Notes, while those for the Central America/Dominican Republic table are an alphabetized common set.
market access provisions in rtas
117
box 3.1. (cont.) Each of the Central American parties and the Dominican Republic have their individual schedules on products entering from the United States. The concessions are rather similar for the various product categories among these countries. Table 3.3 displays the TRQs by the Central American countries and the Dominican Republic on the United States.18 Indeed, while there are some differences in the tariff elimination treatment within Central America for individual products and for the in-quota quantities, the products on which the Central American parties open TRQs tend to be very similar. The Dominican Republic has a slightly different list of products than the Central American parties do; however, the differences can in part be explained by the aggregation of the TRQ in terms of product coverage.
Figures 3.5a to 3.5c display in light gray the year-to-year evolution of liberalization by the seventy-six RTA parties over a period of twenty years.19 The analysis is most relevant for RTAs that take the basket approach, and, in particular, for RTAs with a wide range of baskets. The percentage of duty-free national tariff lines in the overall total is calculated for each schedule. The bold line shows the average evolution of the various schedules. The diagrams also explore the parties’ attainment of the hypothetical interpretation of Article XXIV by placing the bar for SAT at product coverage of 90 per cent (vertical axis), while using year 10 (horizontal axis) as the benchmark time period for ‘reasonable length of time’.20 Figure 3.5a shows that, while some countries employ a ‘stair-step’ approach (stemming from the use of various gradual baskets) to tariff liberalization, others have constant percentage coverage of tariff lines in what could be characterized as a ‘now-or-never’ approach. Still others start from a low coverage, proceeding through one or two jumps to a near-100 per cent coverage. The overall average lies slightly above 90 per cent in year 10. However, there is wide variation across the parties, with a number of outliers lagging well behind the benchmark. 18
19
20
TRQs between the Dominican Republic and Costa Rica and Nicaragua are also part of the Agreement, but are not shown in these tables. The entry-into-force dates differ between agreements, thus the actual years of a given benchmark (e.g. agreement year 10) differ as well. Product coverage in terms of a benchmark based on trade, as opposed to tariff lines, may be more indicative of tariff liberalization, but is beyond the scope of the current analysis, due to the challenges of linking trade with tariff data at the product level. However, trade in a given product is also endogenous to the level of tariff protection on that product, although there may also be a tendency for tariff lines to be more highly concentrated in some sectors as opposed to others; this concentration may differ across countries based on their varying criteria for creating tariff openings.
Table 3.2 Products subject to tariff rate quotas in CAFTA: US tariff quotas on products entering from Central America and the Dominican Republic
Product category Beef Sugarc Sugar (organic)d Peanuts
Peanut butter Cheese Milk powder Butter Other dairy products Ice cream Fluid fresh milk and cream, and sour cream Ethyl alcohol (Central America originating) Ethyl alcohol (non-Central America originating) a
Out-of-quota tariff elimination treatmenta
Initial quantityb
15 year Continued MFN Continued MFN 15 year, nonlinear, 6 year grace period 15 year 20 year, 10 year grace period 20 year, 10 year grace period 20 year, 10 year grace period 20 year, 10 year grace period 20 year, 10 year grace period 20 year, 10 year grace period
10,536 11,000 2,000b *
1,320 10,000 * *
105 24,000 * 500
* 32,000 * *
525 8,000 * *
10,500 22,000 * 10,000
* 300
* 413
* 450
* 500
* 350
280 625 (250e)
Metric tonnes Metric tonnes
50
*
*
*
*
*
Metric tonnes
50
*
60
*
100
*
Metric tonnes
150
110 (220f)
120
250
*
100
Metric tonnes
97,087
160,194
77,670
194,174
48,544
266,989
Litres
407,461
*
366,715
305,596
560,259
254,663
Litres
Unlimited
Unlimited
Unlimited
Unlimited
Unlimited
Unlimited
Gallons
31,000,000b
*
6,604,322g
*
*
*
Gallons
Immediate
Most-FavouredNation
CRI
DOM
SLV
GTM
NIC
HND
Unit Metric Metric Metric Metric
tonnes tonnes tonnes tonnes
In-quota imports shall be free of duty as of entry into force of the Agreement. With the exceptions of imports of ‘Sugar (organic)’ and ‘Ethyl alcohol (non-Central America originating)’ from Costa Rica, which remain fixed, access quantities will be subject to growth over time. c TRQ access based on trade surplus condition. d A fixed 2,000 metric tonne TRQ was allocated by the US to Costa Rica for organic sugar under the US specialty sugar TRQ, and applies to tariff lines AG17011110, AG17011210, AG17019110, AG17019910, AG17029010 and AG21069044. e In the case of Nicaragua, an additional initial quantity of 250 metric tonnes applies to five tariff lines of the 52 total tariff lines making up the entire cheese TRQ. f In the case of the Dominican Republic, an additional initial quantity of 220 metric tonnes applies to four tariff lines of the 46 total tariff lines making up the entire ‘Other dairy products’ TRQ. g Or 10 per cent of the base quantity of dehydrated alcohol and mixtures established under Section 423, whichever is the lesser. * No TRQ. Source: Adapted from: Tripartite Committee, Comparative Guide to the Chile–United States Free Trade Agreement and the Dominican Republic–Central America–United States Free Trade Agreement, based on TRQ Annexes to CAFTA Agreement. b
Table 3.3 Products subject to tariff-rate quotas in CAFTA: Central American and Dominican Republic tariff quotas on products entering from the United States Out-of-quota tariff elimination treatmenta
Initial quantity in metric tonnesb
Product category
CRI
DOM
SLV
GTM
HND
NIC
CRI
DOM SLV
GTM
HND
NIC
Bacon Beans Beef
* * *
10 year 15 year *
* * 15 year, NL, specialc * * 20 year, 10yr GP 20 year, 10yr GP 20 year, 10yr GP * *
* * 10 year
* * *
* * *
* * *
220 * 8,560 * * 105
* * 1,060
* * *
* * *
* * 20 year, 10yr GP *
* * 20 year, 10yr GP *
* * 20 year, 10yr GP *
* * 150
1,100 * 220 * 220 100
* * 100
* * 100
* * 150
*
*
10
*
*
*
20 year, 10yr GP * *
20 year, 10yr GP * *
20 year, 10yr GP * *
410
*
410
450
410
575
* *
138 138
* *
* *
* *
* *
* *
* *
* *
* *
* *
138 440
* *
* *
* *
* *
Beef, prime and choice * Beef, trimmings * Butter 20 year, 10yr GP Buttermilk, curdled * cream and yoghurt Cheese 20 year, 10yr GP Cheese, cheddar * Cheese, mozzarella *
Cheeses, other Chicken meat, mechanically deboned
* *
15 year 15 year 10 year * * 15 year 20 year, NL, 10yr GP 10 year 10 year
Chicken leg quarters
Corn, white
17 year, NL, 10yr GP *
20 year, NL, 10yr GP *
Corn, yellow
*
*
Fresh onions
Con’t MFN Con’t MFN 5 year * 20 year, 10yr GP *
Fresh potatoes Frozen french fries Glucose Ice cream Liquid dairy Liquid milk Milk powder Other dairy products Pig fat Pork
* 20 year, 10yr GP 20 year, 10yr GP * 15 year, 6yr GP
18 year, NL, 10yr GP Con’t MFN 15 year, NL, 6yr GP *
330
550
0
21,810b
0
0
*
*
35,700
20,400
23,460
5,100
*
*
367,500
525,000
190,509
68,250
*
18 year, NL, 10yr GP Con’t MFN 15 year, NL, 6yr GP *
300
*
*
*
*
*
*
*
*
*
300
*
*
*
*
*
* * 20 year, 10yr GP 20 year, 10yr GP * 20 year, 10yr GP 20 year, 10yr GP * 15 year, NL, 6yr GPd
* * 20 year, 10yr GP *
* * 20 year, 10yr GP *
* * 20 year, 10yr GP *
2,631 * 150
* * 1,320 * 165 120
* * 160
* * 100
*
*
*
*
* * 72, 815f *
* 20 year, 10yr GP 10 year
* 20 year, 10yr GP 20 year, 10yr GP * 15 year, NL, 6yr GP
* 20 year, 10yr GP 20 year, 10yr GP * 15 year
* 200
220 * 2,970 300
* 400
* 300
* 650
140
*
120
182
140
50
* 1,100
550 *
* 1,650
* 4,148
* 2,150
* 1,100
18 year, NL, 10yr GP Con’t MFN 10 year
*
18 year, NL, 10yr GP Con’t MFNd 15 year, NL, 6yr GPd *
* * 12 year 12 year * 10 year 20 year, 10yr GP * 12 year *
* 15 year
10
Table 3.3 (cont.) Out-of-quota tariff elimination treatmenta
Initial quantity in metric tonnesb
Product category
CRI
DOM
SLV
GTM
HND
NIC
CRI
DOM SLV
GTM
HND
NIC
Pork cuts
*
*
*
*
*
*
3,465 *
*
*
*
Rice, brown
*
*
*
*
*
*
2,140 *
*
*
*
Rice, milled
20 year, NL, 10yr GP 20 year, NL, 10yr GPd * * *
15 year, NL, 6yr GP 20 year, NL, 10yr GP 20 year, NL, 10yr GP *
18 year, NL, 10yr GP 18 year, NL, 10yr GPd,e 15 year * *
18 year, NL, 10yr GP 18 year, NL, 10yr GPd * * *
18 year, NL, 10yr GP 18 year, NL, 10yr GPd * * *
18 year, NL, 10yr GP 18 year, NL, 10yr GPd * * *
5,250
8,560 5,625
10,500
8,925
13,650
51,000
*
54,600
91,800
92,700
* * *
* 263 3,850 * 110 *
* * *
* * *
* * *
Rice, rough
Sorghum Turkey meat Yogurt
a b
* 12 year 20 year, 10yr GP
62,220
With the exception of ‘Milk powder’ in the Dominican Republic, in-quota imports shall be free of duty as of entry into force of the Agreement. With the exception of imports of ‘Chicken leg quarters’ by Guatemala from the United States, where there are reductions in the duty-free quantity in several years, followed by unlimited access in year 18, access quantities will be subject to growth over time.
c
Duties in this category shall be reduced to 15 per cent in year 1. May be subject to performance requirements. e The aggregate quantity of goods entered into El Salvador from the United States under SAC provision 1006 shall be free of duty in any calendar year specified, ‘and shall not exceed 3,000 MT for “parboiled rough” rice or its equivalent “parboiled milled” rice quantity in any such year. Parboiled milled equivalency shall be calculated according to a 0.7 conversion factor, where 1 MT of parboiled rough rice is equivalent to 0.7 MT of parboiled milled rice’. f Quantities are measured in litres for the Nicaragua ice cream TRQ. * No TRQ. Source: Adapted from: Tripartite Committee, Comparative Guide to the Chile–United States Free Trade Agreement and the Dominican Republic-central America–United States Free Trade Agreement, based on TRQ Annexes to CAFTA Agreement. GP ¼ grace period; NL ¼ non-linear. d
124
estevadeordal, shearer, suominen 110 100 90
% of Lines Duty-Free
80 70 60 50 40 30 20 Individual Agreement Average
10 0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Agreement Year
Figure 3.5a
Evolution of duty-free treatment in selected RTAs
110 100 90
% of L ines Duty-F ree
80 70 60 50 40 30 North-North Avg.
20
North-South Avg. 10
South-South Avg.
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
A greement Y ear
Figure 3.5b
Evolution of duty-free treatment in selected RTAs by signatory party
market access provisions in rtas
125
110 100 90
% of L ines Duty-F ree
80 70 60 50 40 30 20 Overall Average Agriculture Avg. Industrial Avg.
10 0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
A greement Y ear
Figure 3.5c
Evolution of sectoral duty-free treatment in selected RTAs
An interesting result is that, while most concessions are made before year 10, there are much smaller movements in the subsequent years, and almost none after year 15. Overall, the findings echo those of WTO (2002a) and Scollay (2005), which note that RTAs in general do attain the 90 per cent mark within ten years. However, it should also be noted that a small number of agreements contain phase-outs even after year 20 – although the number of products subject to prolonged phase-outs is quite small. Figure 3.5b shows the evolution of duty-free treatment, distinguishing between North–North (solid lines), North–South (broken lines) and South–South (dotted lines) agreements.21 The averages for the three sets of lines are marked in bold. The North–North average is around 95 per cent of tariff lines covered at year 10, while the North–South and South– South averages are slightly above 90 per cent. This is a somewhat intuitive result given the sensitivity of agricultural products and the relative advantages of the developing nations in agriculture, as the overall average for all products in the diagram is pulled upwards due to the larger number of industrial products than agricultural products in the data. The 21
A somewhat arbitrary definition of North versus South is used here. North consists of the EU, EFTA, the United States, Canada, and the Asia-Pacific economies except for China and Thailand.
126
estevadeordal, shearer, suominen 0.5 0.45
StDev of Agreements
0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Average of Agreements
Figure 3.6a
Distribution of liberalization by RTA parties in chapters, Year 5
0.5 0.45
StDev of Agreements
0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
Average of Agreements
Figure 3.6b
Distribution of liberalization by RTA parties in chapters, Year 10
North–North average is also pulled upward by the treatment of Singapore as a Northern country. Moreover, only five of the agreements are classified as North–North, and four of these have Singapore as a party. Figures 3.6a and 3.6b provide further nuance by measuring the average liberalization (x-axis) and dispersion of liberalization (y-axis) across sixty-four RTA partners’ (in a total of thirty-two RTAs) liberalization
market access provisions in rtas
127
schedules in the ninety-seven Harmonised System chapters. The black dots indicate chapters consisting entirely of agricultural products, while squares refer to chapters consisting of industrial products.22 The chapters in the southeast corner are those in which all RTAs analyzed here feature deep liberalization, with negligible dispersion values resulting.23 Chapters in the northwest corner indicate limited liberalization across RTAs and particularly shallow liberalization in some RTAs, with high dispersion resulting. The pattern is clear: agricultural chapters in RTAs feature the least liberalization and also the highest dispersion of liberalization across RTAs, indicating that these chapters are particularly protected in some RTA parties’ schedules. The diagrams also show the relatively slow pace of liberalization: on average, RTA parties liberalize well below 50 per cent of tariff lines in the most sensitive chapters – dairy (chapter 04) and sugars (chapter 17) by the fifth year of the agreement, and less than 55 per cent in several others, including meat, cocoa, prepared cereals and baked goods, tobacco, and footwear (chapters 02, 18, 19, 24 and 64, respectively), while sugar and dairy still remain below 60 per cent at year 10. Encouragingly, however, RTA parties on average liberalize more than 75 per cent of tariff lines in the bulk of chapters by year 5 and more than 90 per cent of tariff lines in most chapters by year 10. The fastest and deepest liberalization takes place in such non-sensitive products as ores (chapter 26), fertilizers (chapter 31), pulp of wood (chapter 47), and some base metals (chapter 81); perhaps one of the reasons is that these are inputs into other products. The results indicate that most ‘action’ in the extent of liberalization occurs in agricultural and textile and apparel products. To be sure, there is significant movement in the textile chapters between the 5- and 10-year benchmarks, while dairy and sugar show little additional liberalization. The differences across parties dissipate by year 10; the persistent variation in agriculture owes largely to the EU’s agreements where liberalization is postponed – at times in perpetuity, as is the case,
22
23
For ease of presentation, in these diagrams chapters 1–24 (excluding chapter 3) are highlighted as agriculture. However, in the analyses of tariff liberalization statistics, agricultural and industrial products are defined at the six-digit HS level. An average liberalization level of one (100 per cent coverage) will necessarily be accompanied by a standard deviation of zero, but there are cases where one chapter may exhibit a higher standard deviation (dispersion among agreements) than another for a given level of average liberalization, or vice versa.
128
estevadeordal, shearer, suominen 0.5 0.45
StDev of Chapters
0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 0
0.2
0.4
0.6
0.8
1
1.2
Average of Chapters
Figure 3.7a
Distribution of liberalization of chapters in RTA parties’ schedules, Year 5
0.45 0.4
StDev of Chapters
0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 0
Figure 3.7b
0.2
0.4
0.6 Average of Chapters
0.8
1
1.2
Distribution of liberalization of chapters in RTA parties’ schedules, Year 10
for example, for certain live animals, meat, dairy and sugar products originating in South Africa in the EU–South Africa RTA. Figures 3.7a and 3.7b turn the analysis around, examining the degree and dispersion of liberalization in the ninety-seven Harmonised System chapters within the liberalization schedules of sixty-four RTA parties. Parties in the southeast corner feature deep liberalization
market access provisions in rtas
129
across the board. Meanwhile, those in the northwest corner are marked by limited liberalization as well as by particularly limited liberalization in some chapters, which entails high dispersion of liberalization across chapters.
2.1.3
New approaches to measuring SAT: trade-weighted tariff liberalization The measurement of SAT simply as a share of tariff lines fails to fully account for the exclusion of sensitive products from RTAs, if the products are covered in a very small number of tariff lines. Does RTAs’ conformity with SAT vary, then, with the measurement that is employed? We strive to shed light on this question by combining the data on liberalization as a share of tariff lines with data on trade flows. In particular, we introduce two alternative methods of exploring the depth and speed of liberalization in RTAs: liberalization statistics examined above as weighted by trade, and the percentage of total trade (imports) from the RTA partner that is completely liberalized.24 By seeking to capture the economic significance of RTA concessions, the approach helps overcome the deficiencies of the measurements based on simple counts of tariff lines, and, as such, to further the debates on the appropriate definition for SAT. Figure 3.8 takes the first cut, examining the evolution of the duty-free share of tariff lines trade-weighted by chapter. The similarity with the unweighted data in Figure 3.5a is striking. This is hardly surprising: most trade occurs in sectors that are opened up rapidly, while sectors with backloaded liberalization have a low share of the overall trade given that they also tend to feature the highest tariff and non-tariff barriers. To be sure, while the bolded averages in the two diagrams are also similar, they are not immediately comparable due to different numbers of observations – twenty-seven v. thirty-eight RTAs. Figure 3.9 measures the evolution of duty-free treatment as a share of the imports from the partner that are liberalized. By this measure, RTA partners on average reach the 90 per cent mark right at year 10.25 The result could be seen as encouraging: RTA partners to liberalize a larger 24 25
The calculations are based on COMTRADE data. Ideally, imports were averaged over a three-year period immediately prior to the entry into force of the agreement. However, due to data-availability constraints as well as to ensure consistency between versions of the Harmonised System, the number of years taken as well as the years themselves varied somewhat from party to party.
130
estevadeordal, shearer, suominen
% of Tariff Lines Duty Free (Trade-Weighted by Chapter)
120
100
80
60
40
20 Individual Agreement Average (27) 0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Agreement Year
Figure 3.8 tariff lines
Evolution of duty-free treatment as trade-weighted percentage of
120
% of Imports Duty Free
100
80
60
40
20 Individual Agreement Average (23) 0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Agreement Year
Figure 3.9
Evolution of duty-free treatment as percentage of imports
20
market access provisions in rtas
131
share of trade than they do of tariff lines. However, again care should be taken not to compare Figure 3.9 directly with Figure 3.5a and 3.8, given that it contains an even less comprehensive dataset than that displayed in Figure 3.8 (twenty-three RTAs). Moreover, and importantly, the figure does not capture the potential trade among the RTA partners. Even if the share of actual trade excluded from an RTA were very small, the potential trade could be very significant in the absence of policy barriers. Figures 3.10a to 3.10h take a closer comparative look at liberalization under the three alternative measurements. They chart the evolution of the liberalization as an unweighted share of liberalized tariff lines, tradeweighted share of liberalized tariff lines, and share of liberalized imports by eight RTA parties in four RTAs – Canada–Costa Rica, Chile–Korea, EU–South Africa, and US–Morocco FTAs. While the paths of the three measurements tend to mirror each other, liberalization as a share of trade-weighted lines and as a share of actual imports tends to be deeper in some cases (as in the EU–South Africa RTA) and shallower in others (as in the Chile–Korea FTA) than conveyed by the simple counts of tariff lines. To be sure, the result does not 120
100
% Duty Free
80
60
40
20
Tariff Lines Trade-Weighted Tariff Lines Imports
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Year
Figure 3.10a Evolution of duty-free access in Canadian market for Costa Rican goods: a comparison using various measures
132
estevadeordal, shearer, suominen 120
100
% Duty Free
80
60
40
20
Tariff Lines Trade-Weighted Tariff Lines Imports
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Year
Figure 3.10b Evolution of duty-free access in Costa Rican market for Canadian goods: a comparison using various measures 120
100
% Duty Free
80
60
40
20
Tariff Lines Trade-Weighted Tariff Lines Imports
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Year
Figure 3.10c Evolution of duty-free access in Chilean market for Korean goods: a comparison using various measures
market access provisions in rtas
133
120
100
% Duty Free
80
60
40
20
Tariff Lines Trade-Weighted Tariff Lines Imports
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Year
Figure 3.10d Evolution of duty-free access in Korean market for Chilean goods: a comparison using various measures 100 90 80
% Duty Free
70 60 50 40 30 20 Tariff Lines Trade-Weighted Tariff Lines
10
Imports 0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Year
Figure 3.10e Evolution of duty-free access in EU market for South African goods: a comparison using various measures
134
estevadeordal, shearer, suominen 100
90
80
70
% Duty Free
60
50
40
30
20 Tariff Lines Trade-Weighted Tariff Lines
10
Imports
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Year
Figure 3.10f Evolution of duty-free access in South African market for EU goods: a comparison using various measures 120
100
% Duty Free
80
60
40
20 Tariff Lines Trade-Weighted Tariff Lines Imports 0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Year
Figure 3.10g Evolution of duty-free access in Moroccan market for US goods: a comparison using various measures
market access provisions in rtas
135
120
100
% Duty F ree
80
60
40
20
Tariff Lines Trade-Weighted Tariff Lines Imports
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Y ear
Figure 3.10h Evolution of duty-free access in US market for Morrocan goods: a comparison using various measures
necessarily hold throughout the entire analyzed time period. In any case, measuring liberalization as a share of tariff lines often either under- or over-estimates (1) the importance of the bilateral trade in some tariff lines; and (2) the extent of trade that actually enters free of duty. RTAs where liberalization appears deeper when trade data is entered are not necessarily highly liberalizing. Rather, the results may simply indicate that much of the trade between the partners is free of duty even prior to the RTA’s entry into effect, which in essence would confirm the endogeneity of trade to the level of protection. In contrast, in some cases – Chile–Korea FTA and US schedule in the US–Morocco FTA – the share of liberalized imports trails that of liberalized lines.26 The bulk of actual trade in these schedules is subject to tariffs, which can be taken to indicate that, while tariffs persist on much of trade, they are not prohibitive of trade. Further iterations of this paper can explore these hypotheses. Overall, the results here echo the above findings on the differences between Northern and Southern parties’ liberalization paths. However, 26
The Korea–Chile FTA is analyzed in detail in WTO (2005).
136
estevadeordal, shearer, suominen
there are some notable exceptions. For instance, Morocco markedly attains the 90 per cent mark under each of the three liberalization measures by year 10 in its FTA with the EU, and Costa Rica does so in the Costa Rica–Canada FTA when liberalization is measured as a share of duty-free imports. Meanwhile, the EU does not attain the 90 per cent mark even by year 20 in the EU–South Africa FTA except under the trade-weighted tariff lines measure, which evinces the persistence of barriers in sensitive agricultural products in the EU’s schedule. Again, none of the diagrams captures the extent of potential trade between partners. Sectors with backloaded liberalization also tend to feature the highest tariffs, which means that they could also generate the greatest payoffs from liberalization. Indeed, measures based on potential trade, if possible, could well yield very different results. To be sure, a thorough analysis of potential trade would need to take into account such factors as total trade with the world and/or domestic production or consumption data. In a sense the question of potential trade is only of academic interest because WTO members are unlikely to agree on estimating the amount of trade that would occur in the absence of restrictions. Nevertheless, the issue is potentially of considerable practical importance in cases where prohibitive or near-prohibitive tariffs or quantitative restrictions seriously inhibit trade in sensitive products. In such cases, the party that aims to insulate its sensitive products might be able to do so while claiming that only a minimal percentage of the existing trade is affected by the exclusion – yet the potential trade that is being excluded could be very substantial.
2.2
RTA coverage: aggregate provisions
This section strives to complement the tariff-line data by exploring the coverage of the various market access commitments contained in the main texts of fifty RTAs. The mapping is based on a three-level matrix. The first level covers five main issue areas: NTMs, other measures, special regimes, RoO, and customs procedures. The second level is composed of disaggregated sets of issues falling under the main issue areas; the third level disaggregates the matrix further, carrying nineteen provisions for NTMs, one provision for other measures, fifteen provisions for special regimes, twenty-six for RoO, and twenty-two provisions for customs procedures. The three levels are displayed in the Annex to this chapter, along with definitions for the second-level provisions.
market access provisions in rtas
137
While the structure of the matrix draws on various different RTAs and RTA types, perhaps the greatest weight in the design of the matrix is given to RTAs based on the structure of the North American Free Trade Agreement (NAFTA). The justification for doing so is that NAFTA is among the most detailed and comprehensive of trade agreements, and also increasingly important as a blueprint for many RTAs in the Americas and around the world. As such, the methodology strives to ensure that no major provisions are being left out – which could be the case if the matrix were based on ‘thinner’ agreements. However, the matrix is general enough to accommodate differently structured agreements.
2.2.1 Methodology The methodology of the mapping as conducted at the second level of disaggregation aims to capture the coverage of market-access provisions in the various RTAs. A provision in an agreement is simply assigned 1 when the RTA covers it, and 0 when it does not. The underlying criteria for the choice of the various provisions are two-fold: (1) to cover all provisions for which there are counterparts in the multilateral trade regimes (GATT, WTO Agreements, and the World Customs Organization (WCO) rules per the 1999 Revised International Convention on the Simplification and Harmonisation of Customs Procedures (Kyoto Convention)); and (2) to capture also provisions that are not included in the multilateral rules but that figure prominently in RTAs. The coding is based on a liberal interpretation, so that 1 is assigned even when an agreement makes a relatively marginal reference to the provision; a stricter interpretation would result in a large number of zeros. For example, while some RTAs in the area of customs procedures have very detailed disciplines on the release of goods, including a specific timeline for the release, a clause on rapid release, and provisions on control and release systems, others include only a general statement that goods should be released expeditiously. In both cases, the coding methodology here would consider the provision covered. A provision is also considered covered if it is outlined in another, more general chapter in the RTA rather than in the chapters pertaining to an issue area examined here. For instance, while some agreements deal with the provisions of transparency and due process in the chapter on customs procedures and trade facilitation, others relegate them to a separate, general chapter on transparency that applies to the entire agreement. Again, the methodology here considers the provision covered in both cases.
138
estevadeordal, shearer, suominen
The third level of analysis goes beyond the simple 0–1 coding to qualify the coding conducted at the second level. For instance, in the case of cumulation of rules of origin, whereas the second-level coding assigns 1 for RTAs employing any type of cumulation – bilateral, diagonal or full 