英國牛津大學國際法牛津評注由英國dissertation網提供資料，本文涉及有關世貿組織關于國際法反傾銷協定研究（The WTO Anti-Dumping Agreement），對于研究西方國際法的英國留學生有一定指導性作用，另本站提供英國留學生法律專業dissertation代寫作服務。OXFORD COMMENTARIES ON INTERNATIONAL LAW
General Editors: Professor Philip Alston, Professor of International Law
at New York University, and Professor Vaughan Lowe, Chichele Professor
of Public International Law in the University of Oxford and Fellow of
All Souls College, Oxford.
Sub-Series:
OXFORD COMMENTARIES ON
THE GATT/WTO AGREEMENTS
General Editor: Professor Robert Howse, Professor of International Law at the University of Michigan.
The WTO Anti-Dumping Agreement

The WTO Anti-Dumping Agreement
A Commentary EDWIN VERMULST
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
© E. Vermulst, 2005
The moral rights of the author have been asserted
Database right Oxford University Press (maker)
Crown copyright material is reproduced under Class Licence
Number C01P0000148 with the permission of OPSI
and the Queen’s Printer for Scotland
First published 2005
All rights reserved. No part of this publication may be reproduced,tored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,or as expressly permitted by law, or under terms agreed with the appropriatereprographics rights organization. Enquiries concerning reproductionoutside the scope of the above should be sent to the Rights Department,Oxford University Press, at the address above
You must not circulate this book in any other binding or coverand you must impose the same condition on any acquirer
British Library Cataloguing in Publication DataData available#p#分頁標題#e#
Library of Congress Cataloging in Publication Data
Vermulst, Edwin A., 1958–
The WTO anti-dumping agreement : a commentary / Edwin Vermulst
p. cm. – (Oxford commentaries on international law)
Includes index.
ISBN 0–19–927707–9 (hardback : alk. paper) 1. Dumping (International
trade) – Law and legislation. 2. Antidumping duties – Law and legislation.
I. Title. II. Series.
K4635.V47 2006
343.087–dc22 2005025476
Typeset by Newgen Imaging Systems (P) Ltd., Chennai, India
Printed in Great Britain
on acid-free paper by
Biddles Ltd., King’s Lynn
ISBN 0–19–927707–9 978–0–19–927707–0
1 3 5 7 9 10 8 6 4 2
For Jiae, Juni and Alix

Foreword
This book provides an analytical overview of the WTO Anti-Dumping
Agreement, as interpreted by WTO Panels and the Appellate Body up to May2005. The Anti-Dumping Agreement is often perceived as the most technical –and, some would say, most controversial – WTO agreement. While the basicconcepts of dumping and resulting injury may appear deceptively simple,complex calculation methods related to the calculation of dumping and injurymargins as well as the very detailed procedural requirements that authoritiesneed to comply with before taking anti-dumping action make the Anti-
Dumping Agreement difficult to apply – and understand – in practice.This book of necessity goes into great detail about the intricacies of antidumpinglaw and practice, but at the same time it attempts to explain the variousconcepts in a relatively non-technical manner by means of simplifiedexamples that are easy to grasp for experts and non-experts alike.
The book also pays extensive attention to interpretations of the various provisionsof the Anti-Dumping Agreement by WTO panels and the AppellateBody. There have been more WTO dispute settlement cases under the Anti-Dumping Agreement than under any other WTO Agreement, reflecting boththe increased recourse to anti-dumping measures by WTO members all overthe world and the complexity of abiding by the WTO rules. Panels and the
Appellate Body, called upon to review administrative determinations and – onseveral occasions – aspects of national legislation, have done a remarkable jobin interpreting the WTO rules. For WTO members wishing to impose antidumping
measures in a WTO-consistent manner, knowledge not only of theAgreement itself, but also of such interpretations, is therefore a must.
The Anti-Dumping Agreement is on the agenda of the Doha Round. At thetime of writing, however, it is too soon to predict how these negotiations willturn out. Nevertheless, it seems unlikely that the basic concepts will significantlychange.
The author would like to thank Jochen Beck, James Durling, FolkertGraafsma, Gary Horlick, John Louth, Francesca Sgritta and Daniel Tarullo forvarious forms of input.#p#分頁標題#e#
Edwin Vermulst
Brussels, June 2005

Contents
WTO ADA Panel and Appellate Body Reports xv
GATT AD Panel Reports xxiv
List of Abbreviations xxv
INTRODUCTION 1
CHAPTER 1: DUMPING 7
Introduction 8
1.1 The Like Product 9
1.1.1 Sub-dividing the Like Product (Product
Control Numbers) 12
1.2 Export Price 13
1.2.1 Traders 14
1.2.2 Constructed Export Price 15
1.3 Normal Value 19
1.3.1 Normal Value: Based on Domestic Sales 19
1.3.1.1 Destined for Consumption 19
1.3.1.2 Ordinary Course of Trade 20
Exclusion of sales below cost 20
Domestic sales to related parties 23
1.3.1.3 Five per cent rule 30
1.3.2 Normal Value: Based on Constructed Normal Value 32
1.3.2.1 Cost Calculations and Allocations 34
1.3.2.2 SGA and Profits 37
General and administrative costs 39
Same general category of products 39
Ordinary course of trade 40
Other exporters or producers/weighted average 41
Amounts for profits 42
1.3.3 Normal Value: Country of Origin of Export? 43
1.3.4 Normal Value: Non-market Economies 44
1.4 Fair Comparison 45
1.4.1 Adjustments 46
1.4.1.1 Conditions and Terms of Sale 47
1.4.1.2 Physical Differences 47
1.4.1.3 Any other Differences 48
1.4.1.4 Constructed Export Price 49
1.4.1.5 Burden of Proof 49
1.4.2 Currency Conversions 49
1.4.2.1 Local L/C Sales 50
1.4.2.2 Date of Sale 50
1.4.3 Comparison Methods: Zeroing 51
1.4.3.1 Inter-model Zeroing 57
1.4.3.2 Multiple Averaging Periods 61
1.4.3.3 Sampling 62
CHAPTER 2: INJURY 63
Introduction 63
2.1 The Like Product 64
2.1.1 Narrowest Group or Range of Products 64
2.2 The Domestic Industry 65
2.2.1 Sampling 68
2.2.2 Companies outside the Domestic Industry 69
2.2.3 Captive Production 69
2.2.4 Related Parties 72
2.2.5 Regional Industry 73
2.3 Material Injury 73
2.3.1 The Record 77
2.3.2 Dumped Imports 79
2.3.3 Injury Investigation Period 82
2.3.4 Cumulation 84
2.3.5 The Article 3.4 Injury Factors 86
2.4 Causation 90
2.4.1 Known Factors 91
2.4.2 Other Factors 92
2.4.3 Non-attribution 93
2.5 Threat of Material Injury 94
x Contents
CHAPTER 3: PROCEDURES 100
Introduction 102
3.1 Initiation 102
3.1.1 The Application 103
3.1.2 Pre-initiation Examination of the Evidence 106
3.1.3 Pre-initiation Standing Determination 109
3.1.4 Pre-initiation Notification/Publicizing of the Application 113
3.1.4.1 Pre-initiation Notification 114
3.1.4.2 Pre-initiation Publicizing of the Application 115
3.2 The Investigation 116
3.2.1 Simultaneous Consideration of Dumping and Injury 116
3.2.2 De Minimis Dumping; Negligible Injury 117
3.2.2.1 De Minimis Dumping 119#p#分頁標題#e#
3.2.2.2 Negligible Injury 121
3.2.3 Customs Clearance 123
3.2.4 Time Limit Investigations 123
3.2.5 Investigation Techniques and Due Process Rights 124
3.2.5.1 Ample Opportunity to Present Evidence 124
3.2.5.2 Questionnaires 126
3.2.5.3 Access to Information 129
3.2.5.4 Access to the Application 133
3.2.5.5 Right of Defence; Hearings 134
3.2.5.6 Confidentiality of Information 136
3.2.5.7 Verifications 140
Accuracy check 140
Exporters’ verifications 142
3.2.5.8 Facts Available 146
3.2.5.9 Disclosure 158
3.2.5.10 Individual Dumping Margins/Sampling 161
3.2.5.11 Interested Parties 165
3.2.5.12 Small Companies 165
3.2.5.13 Final Provision 166
3.3 Measures 166
3.3.1 Provisional Measures 166
3.3.2 Undertakings 168
3.3.3 Anti-dumping Duties 170
3.3.3.1 Lesser Duty Rule 171
Contents xi
3.3.3.2 Non-discrimination 173
3.3.3.3 Collection Systems 174
Refunds in constructed export price situations 176
3.3.3.4 Sampling 178
3.4 Retroactivity 182
3.4.1 Provisional Measures 183
3.4.2 Critical Circumstances 184
3.5 Reviews 190
3.5.1 Interim Review 191
3.5.2 Sunset Review 194
3.5.2.1 The Likelihood Determination 195
3.5.2.2 Dumping Margin Calculation 199
3.5.2.3 Order-wide Determination 201
3.5.2.4 Non-cooperation 202
3.5.2.5 De Minimis Dumping and Cumulation 203
3.5.3 Common Provisions 204
3.5.4 Newcomer Review 205
3.6 Public Notice Requirements 206
3.6.1 Initiation 207
3.6.2 Preliminary and Final Determinations 210
3.6.3 Reviews and Retroactivity Decisions 213
3.7 Judicial Review 213
3.8 Third Country Dumping 214
3.9 Developing Countries 215
3.10 Committee on Anti-Dumping Practices 217
3.11 Dispute Settlement 218
3.11.1 The Matter 220
3.11.2 Challenging Legislation 221
3.11.3 Specificity of Claims 222
3.11.4 Nullification or Impairment 223
3.11.5 Facts made Available 223
3.11.6 Standard of Review 224
3.12 Final Provisions 232
xii Contents
CHAPTER 4: RESIDUAL ADA DISPUTE
SETTLEMENT ISSUES 238
Introduction 238
4.1 Exhaustion of Local/Administrative Remedies 239
4.2 Consultations 240
4.3 Request for Establishment of a Panel 241
4.3.1 Multiple Measures 242
4.3.2 Co-complainants and Parallel Panels 242
4.3.3 Third Parties 242
4.3.4 Matter, Claims and Arguments 243
4.4 Establishment and Composition of Panels 245
4.5 Function of Panels and Terms of Reference 246
4.6 Submissions to and Meetings with the Panel 247
4.7 Confidentiality 249
4.8 Burden of Proof 251
4.9 The Record 253
4.10 Right to Seek Information/Adverse Inferences 255
4.11 Amicus Curiae Briefs 257
4.12 Judicial Economy 258
4.13 Interim Review 259#p#分頁標題#e#
4.14 Recommendations and Suggestions 260
4.15 Adoption and Implementation 261
ANNEXES 265
Annex 1 Agreement on Implementation of Article VI of the General
Agreement on Tariffs and Trade 1994 267
Annex 2 Article VI GATT 1994 292
Contents xiii
Annex 3 Recommendations and other documents from the
Committee on Anti-Dumping Practices 295
Annex 4 WTO ADA Panel and Appellate Body Reports
In reverse chronological order 322
Annex 5 Statistics 326
Index 327
About the Author 333
xiv Contents
WTO ADA Panel and Appellate Body Reports
WTO ADA Panel and Appellate Body Reports (Alphabetical)
Short Title Full Case Title and Citation Page Reference
Argentina–Ceramic Argentina–Definitive anti-dumping 47–8, 58, 125,
floor tiles ( from Italy) measures on imports of ceramic floor 137–9, 151, 152–3,
tiles from Italy, WT/DS/189/R of 156–7, 159–60, 161,
28 September 2001 163–4, 227–8, 256
Argentina–Poultry Argentina–Definitive anti-dumping 10, 47–8, 62, 67, 75,
( from Brazil) duties on poultry from Brazil, 82, 83, 87, 105–6,
WT/DS241/R of 22 April 2003 107, 108, 109,
116–8, 127, 130–1,
134, 156, 161,
164–6, 171, 174,
208, 211, 227–8,
232, 240, 243, 244,
255, 260, 261
EC–Bed Linen EC–Anti-dumping duties on imports 13, 34, 38, 40, 41,
( from India) of cotton-type bed linen from India, 42, 43, 69, 79, 86,
WT/DS141/R of 30 October 2000 88, 89, 113, 200,
211, 216–7, 223,
231, 241, 242, 246,
257, 259, 263
EC–Bed Linen EC–Anti-dumping duties on imports 41, 53, 56, 58, 59
( from India) AB of cotton-type bed linen from India, 231
WT/DS141/AB/R of 1 March 2001
EC–Bed linen EC–Anti-dumping duties on imports 79, 181–2, 199
( from India) 21.5 of cotton-type bed linen from India,
WT/DS141/RW of 29 November 2002
EC–Bed linen EC–Anti-dumping duties on imports 75, 81, 82, 228, 231
(from India) 21.5 AB of cotton-type bed linen from India,
WT/DS141/AB/RW of 8 April 2003
EC–Malleable cast EC–Anti-dumping duties on malleable 13, 124, 211, 244
iron tube or pipe cast iron tube or pipe fittings from Brazil,
fittings ( from Brazil) WT/DS219/R of 7 March 2003
Short Title Full Case Title and Citation Page Reference
EC–Malleable cast iron EC–Anti-dumping duties on malleable 38, 60, 75, 85, 87,
tube or pipe fittings cast iron tube or pipe fittings from Brazil, 90, 91, 92, 131–3,
( from Brazil) AB WT/DS219/AB/R of 22 July 2003 135–6, 228
Egypt–Steel rebar Egypt–Anti-dumping measures on steel 36, 87, 89, 90,
( from Turkey) rebar from Turkey, WT/DS211/R of 125, 141, 143,
8 August 2002 151, 224, 231,#p#分頁標題#e#
244, 245, 254
Guatemala–Cement Guatemala–Anti-dumping investigation 103, 106, 107,
( from Mexico) I regarding Portland cement from Mexico, 114, 117, 118,
WT/DS60/R of 19 June 1998 223, 260
Guatemala–Cement Guatemala–Anti-dumping investigation 219–21, 238–9
( from Mexico) AB regarding Portland cement from Mexico,
WT/DS60/AB/R of 2 November 1998
Guatemala–Cement Guatemala–Definitive anti-dumping 10, 21, 83, 87,
( from Mexico) II measures on grey Portland cement from 88, 104, 107,
Mexico, WT/DS156/R of 108, 111, 114,
24 October 2000 116, 118, 125,
129–31, 133–5,
137, 139–40, 142,
143–4, 145–8,
151–2, 160, 209,
210, 211, 212,
246, 254, 255,
259, 260
Mexico–HFCS Mexico–Anti-dumping investigation of 10, 65, 77, 87, 88,
( from the high fructose corn syrup (HFCS) from 97, 98, 99, 103,
United States) the United States, WT/DS132/R of 105, 106–7, 117,
28 January 2000 168, 182, 183–4,
185, 208–9, 210,
211, 212, 222,
223, 241, 244,
253, 263
Mexico–HFCS Mexico–Anti-dumping investigation of 68, 211, 224, 225
( from the United high fructose corn syrup from the
States) 21.5 United States, WT/DS132/RW
of 22 June 2001
Mexico–HFCS Mexico–Anti-dumping investigation of 68, 95, 96, 97, 99
( from the United high fructose corn syrup from the 227
States) 21.5 AB United States, WT/DS132/AB/RW of
22 October 2001
xvi WTO ADA Panel and Appellate Body Reports
Short Title Full Case Title and Citation Page Reference
Mexico–Rice Mexico–Definitive anti-dumping 75, 76, 77, 82,
( from the measures on beef and rice from the 83, 84, 120–1,
United States) United States, Complaint with respect 124–5, 127–8,
to rice, WT/DS/295/R of 6 June 2005 151, 158, 161–2,
194, 205, 206,
208, 211, 234,
241, 258
Thailand–H-Beams Thailand–Anti-dumping duties on angles, 8, 34, 37, 38, 40,
( from Poland) shapes and sections of iron or non-alloy 42, 77, 78, 88,
steel and H-beams from Poland, 89, 92, 102, 105,
WT/DS122/R of 28 September 2000 114, 115, 124,
135, 224, 225–6,
244, 245, 251,
253, 257–8
Thailand–H-Beams Thailand–Anti-dumping duties on angles, 74, 75, 87,
( from Poland) AB shapes and sections of iron or non-alloy 206–7, 224,
steel and H-beams from Poland, 226, 240, 254
WT/DS122/AB/R of 12 March 2001
United States–1916 United States–Anti-Dumping Act of 1916, 67–8, 103, 106,
Act ( Japan) WT/DS162/R of 29 May 2000 112, 235–6,
242, 261–2
United States–1916 United States–Anti-Dumping Act of 1916, 3, 67–8, 102,
Act (EC) WT/DS136/R of 31 March 2000 114, 232–3, 242#p#分頁標題#e#
United States–1916 United States–Anti-Dumping Act of 1916, 221–2
Act AB WT/DS136/AB/R and WT/DS162/R of
28 August 2000
United States–1916 United States–Anti-Dumping Act of 1916, 323
Act AD 21.3 (c) WT/DS136/11 and WT/DS162/14 of
28 February 2001
United States– United States–Continued Dumping and 113, 242, 261,
CDSOA (Byrd Subsidy Offset Act of 2000, WT/DS217, 263
amendment) 234/R of 16 September 2002
United States– United States–Continued Dumping and 232, 233–4, 254
CDSOA (Byrd Subsidy Offset Act of 2000, WT/DS217,
amendment) AB 234/AB/R of 16 January 2003
United States– United States–Continued Dumping and 322
CDSOA (Byrd Subsidy Offset Act of 2000, WT/DS217/14,
amendment) WT/DS234/22 of 13 June 2003
AB 21.3(c)
WTO ADA Panel and Appellate Body Reports xvii
Short Title Full Case Title and Citation Page Reference
United States– United States–Sunset review of 85, 120, 194,
Corrosion-resistant anti-dumping duties on corrosion-resistant 195, 201, 203
carbon steel flat carbon steel flat products from Japan,
products ( from Japan) WT/DS244/R of 14 August 2003
United States– United States–Sunset review of 174, 195–7,
Corrosion-resistant anti-dumping duties on corrosion-resistant 200–2, 219, 222,
carbon steel flat carbon steel flat products from Japan, 236, 237
products WT/DS244/AB/R of 15 December 2003
( from Japan) AB
United States– United States–Anti-dumping duty on 35, 120, 141–2,
DRAMS 21.5 dynamic random access memory 176, 190–1,
( from Korea) semi-conductors (DRAMS) of one 192–3, 199,
megabit or above from Korea, 235–6, 246
WT/DS99/RW of 29 January 1999
United States–Hot United States–Anti-dumping measures 20, 23, 24, 25,
Rolled Steel on certain hot-rolled steel products 26–7, 87, 88, 90,
( from Japan) from Japan, WT/DS184/R of 93, 146, 148–9,
28 February 2001 180, 185,
186–90, 220,
224, 227, 254
United States–Hot United States–Anti-dumping measures 27, 28, 46, 49,
Rolled Steel on certain hot-rolled steel products 70, 71, 93, 94,
( from Japan) AB from Japan, WT/DS184/AB//R of 128, 149–50,
24 July 2001 153, 157–8, 166,
179–81, 225,
227, 229, 230,
231, 237, 259
United States–Hot United States–Anti-dumping 323
Rolled Steel ( from measures on certain hot-rolled steel
Japan) AB 21.3(c) products from Japan,WT/DS184/13 of
19 February 2002
United States–Oil United States–Sunset reviews of 136, 197, 201,
country tubular anti-dumping measures on oil country 204–5, 236–7,
goods ( from tubular goods from Argentina, 244
Argentina) WT/DS268/R of 16 July 2004#p#分頁標題#e#
United States–Oil United States–Sunset reviews of 85, 124–6,
country tubular anti-dumping measures on oil country 198–9, 202–3,
goods ( from tubular goods from Argentina, 204
Argentina) AB WT/DS268/AB/R of 29 November 2004
xviii WTO ADA Panel and Appellate Body Reports
Short Title Full Case Title and Citation Page Reference
United States– United States–Section 129(c)(1) of the 222
Section 129(c)(1) Uruguay Round Agreements Act,
WT/DS221/R of 15 July 2002
United States– United States–Investigation of the 95, 97, 98, 99,
Softwood lumber International Trade Commission, 257, 261
( from Canada) I WT/DS277/R of 22 March 2004
United States– United States–Final dumping 9, 10, 13, 14, 32,
Softwood lumber determination on softwood lumber from 35, 36, 39, 48,
( from Canada) II Canada, WT/DS264/R of 13 April 2004 49, 58, 59, 60,
62, 104, 109,
118–9
United States– United States–Final dumping 254
Softwood lumber determination on softwood lumber from
( from Canada) I AB Canada, WT/DS264/AB/R of
11 August 2004
United States– United States–Anti-dumping measures 15, 17, 18, 47,
Stainless steel plate in on stainless steel plate in coils and 50, 53, 61, 242,
coils and stainless stainless steel sheet and strip from Korea, 261
steel sheet and strip WT/DS179/R of 22 December 2000
( from Korea)
United States–Steel United States–Anti-dumping and 58, 148, 150–1,
plate ( from India) countervailing measures on steel 153–6, 224,
plate from India, WT/DS206/R 237, 254–5
of 28 June 2002
WTO ADA Panel Reports (chronological from
date of circulation)
Short Title Full Case Title and Citation Page Reference
Guatemala–Cement Guatemala–Anti-dumping investigation 103, 106, 107,
( from Mexico) I regarding Portland cement from Mexico, 114, 117, 118,
WT/DS60/R of 19 June 1998 223, 260
United States– United States–Anti-dumping duty on 35, 120, 141–2,
DRAMS 21.5 dynamic random access memory 176, 190–1,
( from Korea) semi-conductors (DRAMS) of one 192–3, 199,
megabit or above from Korea, 235–6, 246
WT/DS99/RW of 29 January 1999
WTO ADA Panel and Appellate Body Reports xix
Short Title Full Case Title and Citation Page Reference
Mexico–HFCS Mexico–Anti-dumping investigation 10, 65, 77, 87, 88,
( from the of high fructose corn syrup (HFCS) 97, 98, 99, 103,
United States) from the United States, WT/DS132/R 105, 106–7, 117,
of 28 January 2000 168, 182, 183–4,
185, 208–9, 210,
211, 212, 222, 223,
241, 244, 253, 263
Thailand–H-Beams Thailand–Anti-dumping duties on 8, 34, 37, 38, 40,
( from Poland) angles, shapes and sections of iron or 42, 77, 78, 88,#p#分頁標題#e#
non-alloy steel and H-beams from 89, 92, 102, 105,
Poland, WT/DS122/R of 114, 115, 124,
28 September 2000 135, 224, 225–6,
244, 245, 251, 253,
257–8
Guatemala–Cement Guatemala–Definitive anti-dumping 10, 21, 83, 87,
( from Mexico) II measures on grey Portland cement 88, 104, 107,
from Mexico, WT/DS156/R of 108, 111, 114,
24 October 2000 116, 118, 125,
129–31, 133–5,
137, 139–40, 142,
143–4, 145–8,
151–2, 160, 209,
210, 211, 212, 246,
254, 255, 259, 260
EC–Bed Linen EC–Anti-dumping duties on imports 13, 34, 38, 40, 41,
( from India) of cotton-type bed linen from India, 42, 43, 69, 79, 86,
WT/DS141/R of 30 October 2000 88, 89, 113, 200,
211, 216–7, 223,
231, 241, 242, 246,
257, 259, 263
United States– United States–Anti-dumping measures 15, 17, 18, 47, 50,
Stainless steel plate in on stainless steel plate in coils and 53, 61, 242, 261
coils and stainless stainless steel sheet and strip from Korea,
steel sheet and strip WT/DS179/R of 22 December 2000
( from Korea)
United States–Hot United States–Anti-dumping measures 20, 23, 24, 25,
Rolled Steel on certain hot-rolled steel products 26–7, 87, 88, 90,
( from Japan) from Japan, WT/DS184/R of 93, 146, 148–9,
28 February 2001 180, 185, 186–90,
220, 224, 227, 254
xx WTO ADA Panel and Appellate Body Reports
Short Title Full Case Title and Citation Page Reference
Mexico–HFCS Mexico–Anti-dumping investigation 68, 211, 224, 225
( from the United of high fructose corn syrup from the
States) 21.5 United States, WT/DS132/RW
of 22 June 2001
Argentina–Ceramic Argentina–Definitive anti-dumping 47–8, 58, 125,
floor tiles ( from Italy) measures on imports of ceramic floor 137–9, 151, 152–3,
tiles from Italy, WT/DS/189/R of 156–7, 159–60, 161,
28 September 2001 163–4, 227–8, 256
United States–Steel United States–Anti-dumping and 58, 148, 150–1,
plate ( from India) countervailing measures on steel 153–6, 224,
plate from India, WT/DS206/R 237, 254–5
of 28 June 2002
United States– United States–Section 129(c)(1) of the 222
Section 129(c)(1) Uruguay Round Agreements Act,
WT/DS221/R of 15 July 2002
Egypt–Steel rebar Egypt–Anti-dumping measures on steel 36, 87, 89, 90, 125,
( from Turkey) rebar from Turkey, WT/DS211/R of 141, 143, 151, 224,
8 August 2002 231, 244, 245, 254
United States– United States–Continued Dumping 113, 242, 261, 263
CDSOA (Byrd and Subsidy Offset Act of 2000,
amendment) WT/DS217, 234/R of
16 September 2002
EC–Bed linen EC–Anti-dumping duties on imports 79, 181–2, 199#p#分頁標題#e#
( from India) 21.5 of cotton-type bed linen from India,
WT/DS141/RW of 29 November 2002
EC–Malleable cast EC–Anti-dumping duties on malleable 13, 124, 211, 244
iron tube or pipe cast iron tube or pipe fittings from
fittings Brazil, WT/DS219/AB/R
( from Brazil) of 22 July 2003
Argentina–Poultry Argentina–Definitive anti-dumping 10, 47–8, 62, 67, 75,
( from Brazil) duties on poultry from Brazil, 82, 83, 87, 105–6,
WT/DS241/R of 22 April 2003 107, 108, 109,
116–8, 127, 130–1,
134, 156, 161,
164–6, 171, 174,
208, 211, 227–8,
232, 240, 243, 244,
255, 260, 261
WTO ADA Panel and Appellate Body Reports xxi
Short Title Full Case Title and Citation Page Reference
United States– United States–Sunset review of 85, 120, 194,
Corrosion-resistant anti-dumping duties on corrosion- 195, 201, 203
carbon steel flat resistant carbon steel flat products
products ( from Japan) from Japan, WT/DS244/R of
14 August 2003
United States– United States–Investigation of the 95, 97, 98, 99,
Softwood lumber International Trade Commission, 257, 261
( from Canada) I WT/DS277/R of 22 March 2004
United States– United States–Final dumping 9, 10, 13, 14, 32,
Softwood lumber determination on softwood lumber 35, 36, 39, 48,
( from Canada) II from Canada, WT/DS264/R of 49, 58, 59, 60, 62,
13 April 2004 104, 109, 118–9
United States–Oil United States–Sunset reviews of 136, 197, 201,
country tubular anti-dumping measures on oil country 204–5, 236–7,
goods ( from tubular goods from Argentina, 244
Argentina) WT/DS268/R of 16 July 2004
Mexico–Rice Mexico–Definitive anti-dumping 75, 76, 77, 82, 83,
( from the measures on beef and rice from the 84, 120–1, 124–5,
United States) United States, Complaint with respect 127–8, 151, 158,
to rice, WT/DS/295/R of 161–2, 194, 205,
6 June 2005 206, 208, 211, 234,
241, 258
WTO ADA Appellate Body Reports (chronological from
date of circulation)
Short Title Full Case Title and Citation Page Reference
Guatemala–Cement Guatemala–Anti-dumping 219–21, 238–9
( from Mexico) AB investigation regarding Portland
cement from Mexico, WT/DS60/AB/R
of 2 November 1998
United States–1916 United States–Anti-Dumping Act of 221–2
Act AB 1916, WT/DS136/AB/R and
WT/DS162/R of 28 August 2000
United States–1916 United States–Anti-Dumping Act of 323
Act AD 21.3(c) 1916, WT/DS136/11 and
WT/DS162/14 of 28 February 2001
xxii WTO ADA Panel and Appellate Body Reports
WTO ADA Panel and Appellate Body Reports xxiii
Short Title Full Case Title and Citation Page Reference#p#分頁標題#e#
Thailand–H-Beams Thailand–Anti-dumping duties on 74, 75, 87,
( from Poland) AB angles, shapes and sections of iron or 206–7, 224,
non-alloy steel and H-beams from 226, 240, 254
Poland, WT/DS122/AB/R of
12 March 2001
United States–Hot United States–Anti-dumping measures 27, 28, 46, 49, 70,
Rolled Steel on certain hot-rolled steel products 71, 93, 94, 128,
( from Japan) AB from Japan, WT/DS184/AB//R of 149–50, 153,
24 July 2001 157–8, 166, 179–81,
225, 227, 229, 230,
231, 237, 259
Mexico–HFCS Mexico–Anti-dumping investigation 68, 95, 96, 97, 99
( from the United of high fructose corn syrup from the 227
States) 21.5 AB United States, WT/DS132/AB/RW
of 22 October 2001
United States–Hot United States–Anti-dumping measures 323
Rolled Steel on certain hot-rolled steel products
( from Japan) from Japan, WT/DS184/13 of
AB 21.3(c) 19 February 2002
United States– United States–Continued Dumping 232, 233–4, 254
CDSOA (Byrd and Subsidy Offset Act of 2000,
amendment) AB WT/DS217, 234/AB/R of
16 January 2003
EC–Bed linen EC–Anti-dumping duties on imports 75, 81, 82, 228, 231
( from India) of cotton-type bed linen from India,
21.5 AB WT/DS141/AB/RW of 8 April 2003
United States– United States–Continued Dumping 322
CDSOA (Byrd and Subsidy Offset Act of 2000,
amendment) WT/DS217/14, WT/DS234/22
AB 21.3(c) of 13 June 2003
EC–Malleable cast EC–Anti-dumping duties on malleable 38, 60, 75, 85, 87,
iron tube or pipe cast iron tube or pipe fittings from 90, 91, 92, 131–3,
fittings ( from Brazil, WT/DS219/AB/R of 135–6, 228
Brazil) AB 22 July 2003
United States– United States–Sunset review of 174, 195–7, 200–2,
Corrosion-resistant anti-dumping duties on corrosion- 219, 222, 236, 237
carbon steel flat resistant carbon steel flat products
products ( from from Japan, WT/DS244/AB/R of
Japan) AB 15 December 2003
Short Title Full Case Title and Citation Page Reference
United States– United States–Final dumping 254
Softwood lumber determination on softwood lumber
( from Canada) AB from Canada, WT/DS264/AB/R
of 11 August 2004
United States-Oil United States–Sunset reviews of 85, 124–6, 198–9
country tubular anti-dumping measures on oil country 202–3, 204
goods ( from tubular goods from Argentina,
Argentina) AB WT/DS268/AB/R of
29 November 2004
GATT AD Panel Reports (in alphabetical order)
Short Title Full Case Title and Citation Page Reference
EC-Audio tapes EC–Anti-dumping proceeding 28, 55, 113, 252
in cassettes against audio tapes and audio tapes
in cassettes from Japan, unadopted,#p#分頁標題#e#
ADP/136/Corr. 1, 8 June 1995
EC-Cotton yarn EC–Imposition of anti-dumping 8, 55, 56, 215, 216
duties on imports of cotton yarns
from Brazil, adopted 30 October 1995,
ADP/137, 4 July 1995
United States- United States–Fresh and chilled 94, 113, 239, 252
Fresh and chilled Atlantic salmon from Norway, adopted
Atlantic salmon 27 April 2004, ADP/87,
30 November 1992
United States-Grey United States–Anti-dumping duties on 110, 239, 260
Portland cement and grey Portland cement and cement
cement clinker clinker from Mexico, unadopted,
ADP/82, 7 September 1992
United States- United States–Imposition of anti- 109, 260
Seamless steel dumping duties on imports of seamless
hollow products steel hollow products from Sweden,
unadopted ADP/47, 20 August 1990
United States- United States–Anti-dumping duties 240
Stainless steel plate on imports of stainless steel plate
from Sweden, unadopted, ADP/47,
20 August 1990
xxiv WTO ADA Panel and Appellate Body Reports
List of Abbreviations
AB Appellate Body
AD anti-dumping
ADA Anti-Dumping Agreement (Agreement on Implementation of
Article VI of GATT 1994)
ADD anti-dumping duties
ADP Committee on Anti-Dumping Practices
ASCM Agreement on Subsidies and Countervailing Measures
ASG Agreement on Safeguards
BIA best information available
CCDP car compact disc player
CDSOA Continued Dumping and Subsidy Offset Act (‘Byrd amendment’)
CIF cost, insurance and freight
CVD countervailing duty
DDA Doha Development Agenda
DRAMS dynamic random access memory semiconductor
DSB Dispute Settlement Body
DSU Dispute Settlement Understanding
G&A general and administrative (costs)
GAAP generally accepted accounting principles
GATS General Agreement on Trade in Services
GATT General Agreement on Tariffs and Trade
HS Harmonized System
IIP injury investigation period
IP investigation period
ITC International Trade Commission
L/C letter of credit
MEG monoethylene glycol
NOI notice of initiation
NV normal value
PCN product control number
PET polyethylene terephthalate
POI period of investigation
PSF polyester staple fibres
PTA purified terephthalic acid
PTY polyester textured yarn
REP request for establishment of a panel
SGA selling, general and administrative expense
SPB Sunset Policy Bulletin
USDOC US Department of Commerce
USITC US International Trade Commission
VCLT Vienna Convention on the Law of Treaties
WTO World Trade Organization
xxvi List of Abbreviations
INTRODUCTION
This book will analyse the 1994 WTO Agreement on Implementation of
Article VI or the Anti-Dumping Agreement (ADA), as interpreted by Panels#p#分頁標題#e#
and the Appellate Body.¹ It is divided into three main parts: the dumping
determination (Chapter 1), the injury determination (Chapter 2) and
procedural rules (Chapter 3). A shorter fourth part will analyse non-ADA
specific WTO jurisprudence that nevertheless has direct consequences for
ADA disputes (Chapter 4).²
The ADA is often considered to be the most complex and technical
agreement of the WTO.³ In this book I will attempt to explain the provisions
of the ADA in a relatively non-technical manner. Experts in the field are
referred to more technical treatises, such as the Handbook on Anti-Dumping
Investigations, co-authored by three experts in the WTO’s Rules Division.?
I will also pay extensive attention to important interpretations of the ADA
provisions by the WTO dispute settlement bodies.?
Dumping has traditionally been defined as the type of price discrimination
between national markets,? in which a producer sells at a lower price abroad than
in his home market ( price dumping). It is often considered unfair ? that a producer
¹ See also the recently published WTO Appellate Body Repertory of Reports and Awards
1995–2004, pp 42–101 (2005).
² Chapter 4 is based on Chapter 1 of Vermulst and Graafsma, WTO Disputes with respect to Anti-
Dumping, Subsidies and Safeguards (2002).
³ Durling and Nicely, Understanding the WTO Anti-Dumping Agreement: Negotiating History and
Subsequent Interpretation (2002) 2, note that the ADA may be the most important substantive agreement
from the Uruguay Round negotiations.
? Czako, Human and Miranda, A Handbook on Anti-Dumping Investigations (Cambridge
University Press, 2003).
? Interested readers are also referred to the excellent and always up-to-date Dispute Settlement
Commentaries, reported on-line by WorldTradeLaw.net, available at http://www.worldtradelaw.net/
dsc/dscpage.htm). It is to be noted that where I quote excerpts from panel or AB reports, I have used the
term ADA in the quotes for the sake of consistency. I have further deleted footnotes in quotations.
? Viner, Dumping, A Problem in International Trade (1966 edn) 3; Dale, Anti-Dumping Law in a
Liberal Trade Order (1980) 1.
? As originally envisaged by economists such as Viner, note 6 above, anti-dumping action would be
justified against predatory dumping only. However, predatory dumping has never been proven to exist.
Anti-dumping laws have also seldom contained a predatory intent requirement, presumably because
of recognition of the difficulty of proving such intent.
2 Introduction
? For a more sceptical view, see, for example, Kerr and Loppacher, ‘Anti-Dumping in the Doha
Negotiations, Fairy Tales at the World Trade Organization’ (2004) 38 Journal of World Trade 211,#p#分頁標題#e#
212–214.
? See for more detail Vermulst, (1986) 7 New York Law School Journal of International and
Comparative Law 301–421. ¹? Art VI:1 GATT 1994.
¹¹ Compare GATT Analytical Index, 223 (vol 1 1995):
‘[i]n discussions at the Review Session in 1954–55, in connection with the rejection of a proposal to
add a clause specifically obligating contracting parties to prevent dumping by their commercial
enterprises, it was agreed to add the following statement to the Working Party’s Report:
“In connexion with the effect of Article VI on the practice of dumping itself, they agreed that it follows
from paragraph 1 of Article VI, that contracting parties should, within the framework of their
who benefits from protection in his home market and therefore can charge highprices there, subsequently uses the artificially high profits generated on theprotected home market to subsidize low-priced export sales. He is perceived tothereby unfairly compete with domestic producers in the – open – importingcountry market which cannot afford such low prices.? Over time, a second formof dumping, so-called cost dumping, has been treated as objectionable.The focus on the effects of dumping on the domestic industry in the importingcountry led to the early realization that protective measures against dumpingshould only be permissible if such dumping causes injury in the importingcountry. This conclusion was drawn almost from the date of enactment of thefirst national anti-dumping laws at the beginning of the twentieth century.?Therefore, the national anti-dumping laws required two tests before theimporting country could impose so-called anti-dumping duties: there had tobe dumping and such dumping had to cause injury in the importing country.However, the methods used to assess dumping and injury as well as investigatoryprocedures differed substantially among the countries that used such laws. Thisdivergence, and the growing realization that anti-dumping laws, as applied, had thepotential to become significant barriers to international trade, created international
interest in dumping as early as the 1920s, a mere 20 years after the passage of thefirst anti-dumping laws. It was not until the end of the Second World War, however,
that binding international rules were developed. Such rules were implemented aspart of the 1947 General Agreement on Tariffs and Trade (GATT).
The GATT rules do not prohibit injurious dumping, although theycondemn it.¹? However, they do allow the country in which a product is
dumped to take protective action if it can establish that the product is dumpedand thereby causes injury to the domestic industry of that product. To a certainextent this importing country emphasis follows logically from the definition of
dumping as price discrimination practised by private individuals. TheWTO addresses governmental behaviour and therefore could not possibly
prohibit dumping by private enterprises.¹¹ Moreover, importing countriesmay welcome dumped imports because their user industries and consumers#p#分頁標題#e#
will benefit from the low prices.GATT 1947 applies only to goods. This implies that dumping of services is
not covered.¹² It has furthermore long been accepted by the GATT signatoriesthat neither Article VI nor the Agreement cover exchange dumping, socialdumping, environmental dumping or freight http://www.ttc-switch.com/Thesis_Writing/Law/dumping.¹³ On the other hand,the reasons why companies dump are considered irrelevant as long as thetechnical definitions are met.¹?Dumping therefore potentially equally¹? coverspredatory dumping,¹? cyclical dumping,¹? market expansion dumping,¹? statetradingdumping¹? and strategic dumping.²?Since 1947, anti-dumping has received elaborate attention in the
GATT/WTO on several occasions. Following a 1958 GATT Secretariat studyof national anti-dumping laws, a Group of Experts was established that in
1960 agreed on certain common interpretations of ambiguous terms of Article
VI. Moreover, in an ambitious move, an Anti-Dumping Code was negotiatedduring the 1967 Kennedy Round and signed by 17 parties. The Code was
revised during the Tokyo Round. The Tokyo Round Code had 25 signatories,
counting the EC as one.²¹ Although the 1979 Code was not explicitly mentionedin the Ministerial Declaration starting the Uruguay Round,²² fairly early in thenegotiations a number of GATT Contracting Parties, notably the United States,
Introduction 3legislation, refrain from encouraging dumping, as defined in that paragraph, by private commercial
enterprises.” ’
See also, Jackson, World Trade and the Law of GATT (1969) 401–424.
¹² ibid 404–405. It may be noted that the WTO General Agreement on Trade in Services (WTO
GATS) does not contain provisions on dumping of services.
¹³ See EPCT/C.II/48, at 1, as quoted in GATT Analytical Index, at 222 (vol I (1995). Compare
Stewart, ‘Antidumping’ in The GATT Uruguay Round: A Negotiating History (1986–1992), Vol II,
1383–1711, pp 1406–1407 (1993). ¹? Compare United States–1916 Act (EC), AB, para 107.
¹? The distinctions are made by Willig, Competition Policy and Anti-Dumping, Ch 1: ‘The
Economic Effects of Anti-Dumping Policy’ (OECD 1995).
¹? Dumping in order to drive competitors out of business and establish a monopoly.
¹? Selling at low prices because of over-capacity due to a downturn in demand.
¹? Selling at a lower price for export than domestically in order to gain market share.
¹? Selling at low prices in order to earn hard currency.
²? Dumping by benefiting from an overall strategy which includes both low export pricing and
maintaining a closed home market in order to reap monopoly or oligopoly profits.
²¹ See GATT, B.I.S.D., 36th Supp., at 435 (1990). The Parties were: Australia, Austria, Brazil,#p#分頁標題#e#
Canada, Czechoslovakia, Egypt, the EC, Finland, Hong Kong, Hungary, India, Japan, Korea, Mexico,
New Zealand, Norway, Pakistan, Poland, Romania, Singapore, Spain, Sweden, Switzerland, the USA
and Yugoslavia.
²² GATT/1396, Ministerial Declaration on the Uruguay Round (1986). For a brief background of
the preparatory activities that led to the adoption of the Declaration, see GATT Activities 1985, 6–10
(1986).
Japan, Korea, Hong Kong and the EC proposed changes – sometimes radicalones – to the 1979 Code. These led to the new ADA.²³
As the ADA forms part of the single package, all WTO members areautomatically members of and bound by the ADA. This occasionally leads tothe misunderstanding that the WTO rules oblige WTO members to adoptanti-dumping legislation. This is not correct. However, WTO members thatdo adopt and utilize anti-dumping legislation, must do so in accordance with
the provisions of the ADA:
Article 1 ADA: Principles
An anti-dumping measure shall be applied only under the circumstances provided forin Article VI of GATT 1994 and pursuant to investigations initiated²? and conductedin accordance with the provisions of this Agreement. The following provisions governthe application of Article VI of GATT 1994 in so far as action is taken underanti-dumping legislation or regulations.
It is significant that international regulation has always focused more on thetrade-distorting effects of anti-dumping action than on the effects of the dumpingitself; most GATT refinements of the anti-dumping regime over the past 60years have limited the ability of the importing country to take protective action.Until 1990, there were four GATT members that actively used their antidumpinglaws as a means of protecting domestic industries against injuriously
dumped imports: Australia, Canada, the European Communities and theUnited States.²? These four developed countries mostly applied measures
against developing countries.
In the course of the last 15 years, however, more and more countries haveadopted anti-dumping laws and started using them.²? According to WTOdata, thus far 104 WTO members, counting the EC as one, have submittednotifications of anti-dumping legislation and/or regulations to the WTO
4 Introduction
²³ For an overview of the negotiating history, see Koulen, ‘The New Anti-Dumping Code
Through Its Negotiating History’ in Bourgeois, Berrod and Gippini Fournier (eds), The Uruguay
Round Results (1995) 151–233; Stewart, ‘Antidumping’ in The GATT Uruguay Round: A NegotiatingHistory (1986–1992), Vol II, 1383–1711 (1993). For a shorter summary, see Horlick and Shea, ‘TheWorld Trade Organization Anti-Dumping Agreement’ (1995) 29 Journal of World Trade 5–31.
Durling and Nicely, Understanding the WTO Anti-Dumping Agreement: Negotiating History andSubsequent Interpretation (2002) contains all of the negotiating drafts.#p#分頁標題#e#
²? Fn 1 in ADA: The term ‘initiated’ as used in this Agreement means the procedural action bywhich a Member formally commences an investigation as provided in Article 5.
²? Compare Stewart, ‘Antidumping’ in The GATT Uruguay Round: A Negotiating History(1986–1992), Vol II, 1383–1711, p 1395 (1993) who reports that from 1980 to 1989 these fourmembersaccounted for 95 per cent of the 1,456 anti-dumping cases reported to the GATT during the 1980–1989period. See also Jackson and Vermulst, Anti-Dumping Law and Practice: A Comparative Study (1989).
²? Compare Kerr and Loppacher, ‘Anti-Dumping in the Doha Negotiations, Fairy Tales at theWorld Trade Organization’ (2004) 38 Journal of World Trade 211, 211.Secretariat.²? During the period from 1 January 1995 to 31 December 2004,38 WTO members applied a total of 1,656 anti-dumping measures.²?
As will be clear from the table, the 10 major users now are India, the United States,the EC, Argentina, South Africa, Canada, Turkey, Mexico, Brazil and Australia.
Details on the 40 major victims of these measures are provided in the table
below:
Anti-dumping measures by victim from 01/01/95 to 31/12/04:
Argentina (9) Australia (8) Belarus (10) Belgium (13)
Brazil (57) Bulgaria (10) Canada (13) Chile (14)
China (297) Czech Republic (14) EC (38) France (24)
Germany (35) Hong Kong (12) Hungary (7) India (60)
Indonesia (54) Italy (25) Japan (82) Kazakstan (18)
Korea (118) Latvia (7) Malaysia (34) Mexico (21)
Netherlands (13) Poland (18) Romania (24) Russia (76)
Singapore (21) Slovak Republic (7) South Africa (33) Spain (21)
Sweden (6) Taiwan (89) Thailand (63) Turkey (22)
Ukraine (47) UK (20) United States (83) Venezuela (12)
Thus, the 10 major victims have become China, the EC-25 and its Member
States, Korea, Taiwan, the United States, Japan, Russia, Thailand, India and
Brazil. Anti-dumping disputes therefore are no longer mainly conflicts
Anti-dumping measures by WTO member from 01/01/95 to 31/12/04:
Argentina (139) Australia (54) Brazil (62) Canada (80)
Chile (6) China (52) Colombia (11) Costa Rica (1)
Czech Republic (1) EC (193) Egypt (30) Guatemala (1)
India (302) Indonesia (23) Israel (15) Jamaica (4)
Japan (3) Korea (43) Latvia (1) Lithuania (7)
Malaysia (18) Mexico (69) New Zealand (14) Nicaragua (1)
Pakistan (4) Paraguay (1) Peru (34) Philippines (9)
Poland (9) Singapore (2) South Africa (113) Taiwan (2)
Thailand (23) Trinidad & Turkey (77) United States (219)
Tobago (7)
Uruguay (1) Venezuela (25)
Introduction 5
²? WTO Annual Report, pp 45–47 (WTO Geneva 2004).
²? WTO website, consulted on 4 June 2005. See Annex 5 for more details.
between developed country WTO members as users and developing country
WTO members as victims. Many important WTO members are both!#p#分頁標題#e#
In the framework of the Doha Development Agenda, WTO members are once
again negotiating about possible revisions of the ADA. At the time of writing
( June 2005), it is too soon to predict how these negotiations will turn out.
However, it seems unlikely that the basic concepts will significantly change.
Indeed, the 4th Ministerial WTO Conference in Doha decided in relevant part
that:
DDA negotiating mandate
28. In the light of experience and of the increasing application of these instruments byMembers, we agree to negotiations aimed at clarifying and improving disciplinesunder the Agreement[s] on Implementation of Article VI of the GATT 1994 . . . whilepreserving the basic concepts, principles and effectiveness of these Agreements andtheir instruments and objectives, and taking into account the needs of developing andleast-developed participants.
Paragraph 28 of this Ministerial Declaration²? is clearly the result of a carefullydrafted compromise between two factions. On the one hand, negotiations areto be ‘aimed at clarifying and improving disciplines’, reflecting the desire ofvictim (including some developing) countries and, on the other hand, suchnegotiations will have to ‘preserv[e] the basic concepts, principles and effectivenessof these Agreements’ reflecting the concerns of other WTO members³? that theinstrument might be tinkered with too much.
The widespread and ever-expanding use of the anti-dumping instrumentmakes it imperative that the ADA is transparent, fair and predictable.³¹ Aforeign supplier should be able to make in advance his own calculationsregarding normal value and export price so that he can determine whether heis dumping and take steps to avoid it, if he wants to. The artificiality of dumpingcalculations would appear to mandate such an interpretation.
6 Introduction
²? WTO, Ministerial Declaration adopted on 14 November, WT/MIN(01)/DEC/1
(20 November 2001).
³? Compare Tarullo, ‘Paved with Good Intentions: The Dynamic Effects of WTO Review of Anti-
Dumping Action’ (2003) 2 World Trade Review 373, 388.
³¹ Compare Horlick and Vermulst, ‘The 10 Major Problems with the Anti-Dumping Instrument:
An Attempt at Synthesis’ (2005) 39 Journal of World Trade 67.
Chapter 1
DUMPING
Introduction 8
1.1 The Like Product 9
1.1.1 Sub-dividing the Like Product (Product
Control Numbers) 12
1.2 Export Price 13
1.2.1 Traders 14
1.2.2 Constructed Export Price 15
1.3 Normal Value 19
1.3.1 Normal Value: Based on Domestic Sales 19
1.3.1.1 Destined for Consumption 19
1.3.1.2 Ordinary Course of Trade 20
Exclusion of sales below cost 20
Domestic sales to related parties 23
1.3.1.3 Five per cent rule 30
1.3.2 Normal value: Based on Constructed Normal Value 32
1.3.2.1 Cost Calculations and Allocations 34#p#分頁標題#e#
1.3.2.2 SGA and Profits 37
General and administrative costs 39
Same general category of products 39
Ordinary course of trade 40
Other exporters or producers/weighted average 41
Amounts for profits 42
1.3.3 Normal Value: Country of Origin of Export? 43
1.3.4 Normal Value: Non-market Economies 44
1.4 Fair Comparison 45
1.4.1 Adjustments 46
1.4.1.1 Conditions and Terms of Sale 47
1.4.1.2 Physical Differences 47
1.4.1.3 Any other Differences 48
1.4.1.4 Constructed Export Price 49
1.4.1.5 Burden of Proof 49
1.4.2 Currency Conversions 49
1.4.2.1 Local L/C Sales 50
1.4.2.2 Date of Sale 50
8 WTO Anti-Dumping Agreement
¹ Thailand–H-Beams from Poland, Panel, para 7.35.
² EC–Cotton yarns from Brazil, paras 449–456, ADP/137, 4 July 1995.
1.4.3 Comparison Methods: Zeroing 51
1.4.3.1 Inter-model Zeroing 57
1.4.3.2 Multiple Averaging Periods 61
1.4.3.3 Sampling 62
Introduction
Article 2 ADA covers the determination of dumping. The fact that only
one – albeit long – article of the ADA is devoted to this complex issue shows
that the ADA continues to function as a framework agreement, leaving substantial
discretion to the WTO members. Nevertheless, the ADA provisions on
dumping are much more detailed than previous ones and have been further
clarified by WTO and Appellate Body reports.
The Thailand–H-Beams case established early on that Article 2 contains
multiple obligations with respect to the determination of dumping.¹ As a result,
a WTO member wishing to challenge a dumping determination made by
another member must take great care that it describe in sufficient detail in its
request for establishment of a panel exactly which provisions it considers to
have been violated. Failure to do so may lead the panel/Appellate Body to reject
dumping claims on the ground that they were too vague. This confirms the
practice of earlier GATT panels, such as EC–Cotton yarns:
[F]or a claim to be before the Panel it would have to be specified in the document
requesting establishment of a panel . . . The panel found that Brazil’s claim that the EC
had made an ‘incorrect determination that certain domestic sales were not in the ordinary
course of trade’ was not expressly referred to anywhere in document ADP/121,
and accordingly dismissed Brazil’s argument on this point.²
Article 2.1 sets out that for the purpose of the ADA, a product is to be considered
as being dumped, i.e. introduced into the commerce of another country
at less than its normal value, if the export price of the product exported from
one country to another is less than the comparable price, in the ordinary course
of trade, for the like product when destined for consumption in the exporting#p#分頁標題#e#
country:
Article 2.1 ADA
For the purpose of this Agreement, a product is to be considered as being dumped,
i.e. introduced into the commerce of another country at less than its normal value, if
the export price of the product exported from one country to another is less than the
comparable price, in the ordinary course of trade, for the like product when destined
for consumption in the exporting country.
This definition of dumping introduces four important concepts, in order of
appearance: normal value, export price, ordinary course of trade and like
product. These concepts are discussed in more detail in the following
paragraphs.
The comparison between prices or costs in different markets is normally
performed during a period of time set by the authorities at the outset of the
investigation. This is typically called the investigation period (IP) or the period
of investigation (POI). It is noteworthy that Article 2 does not contain any
rules on the selection of this period, which may sometimes have ramifications
for the outcome of the proceeding.
However, a recommendation adopted by the Committee on Anti-Dumping
Practices in 2000 provides that the investigation period for dumping purposes
should normally be 12 months, and in any case no less than 6 months, ending
as close to the date of initiation of the investigation as is practicable.³
1.1 The Like Product
As referred to in Article 2.1 the ‘like product’ is the product sold on the domestic
market that is like the exported – dumped – product.?
Article 2.6 ADA stipulates more in general that throughout the ADA the like
product is a product which is identical, ie alike in all respects to the product
under consideration, or in the absence of such a product, another product
which, although not alike in all respects, has characteristics closely resembling
those of the product under consideration. Thus Article 2.6 both defines the
like product and clarifies that the definition applies throughout the ADA:
Article 2.6 ADA
Throughout this Agreement the term ‘like product’ (‘produit similaire’) shall be interpreted
to mean a product which is identical, i.e. alike in all respects to the product
under consideration, or in the absence of such a product, another product which,
although not alike in all respects, has characteristics closely resembling those of the
product under consideration.
Dumping 9
³ Committee on Anti-Dumping Practices, Recommendation concerning the periods of data collection
for anti-dumping investigations, G/ADP/6 (16 May 2000), reproduced in Annex 3.
? Compare United States–Softwood lumber from Canada II, Panel, para 7.152.
The panel in United States–Softwood lumber II noted that Article 2.6 is a
definitional article, and then appeared to draw the conclusion that it does not#p#分頁標題#e#
contain in itself obligations on Members, or in any event that it could not be
the basis for an independent violation.? In a similar vein, the panel in
Guatemala–Cement II found that Guatemala had violated Articles 3.1, 3.2 and
3.5 by wrongly characterizing some imports by MATINSA as not of the like
product and failing to take into account these imports in its determination of
injury and causality.?
These approaches may be contrasted with Mexico–HFCS, where the panel
noted ? that the United States had not claimed that imported HFCS was not
‘like’ domestically produced sugar and therefore would not rule on this (in
application of the WTO principle of non ultra petita).
In my view, there is no reason why Article 2.6 cannot form the basis for an
independent claim, whether it is a definitional article or not.? If an authority
treats imported coffee as a like product to domestically-produced tea, he violates
the like product definition, in addition to any other ADA articles that may
have been violated.?
For dumping purposes, the like product is to be compared with the allegedly
dumped product, which is referred to in Article 2.6 as the product under
consideration. In the context of the injury determination, on the other hand,
the term refers to the product produced by the domestic industry allegedly
being injured by the dumped product.¹? As a result, the ‘like product’ has a
different meaning, depending whether it is used for the dumping or for the
injury determination.
In both cases, the two products should ideally be identical. Only in the
absence of identical products may products with closely resembling characteristics
be compared with each other.
The ‘like product’ concept in the ADA and, indeed, the ASCM must be
distinguished from the concept of ‘like or directly competitive product’ used in the
10 WTO Anti-Dumping Agreement
? United States–Softwood lumber from Canada II, Panel, para 7.146.
? Guatemala–Cement from Mexico II, Panel, para 8.272.
? Mexico–HFCS from the United States, Panel, para 7.59, fn 555.
? Compare Argentina–Poultry from Brazil, Panel, paras 7.331; 7.338, with respect to the Article
4.1 definition of the domestic industry, discussed in section 2.2, below. But see
WorldTradeLaw.net, Dispute Settlement Commentary on Argentina–Poultry from Brazil, p 24,
checked on 23 April 2005.
? Durling and Nicely, Understanding the WTO Anti-Dumping Agreement: Negotiating History and
Subsequent Interpretation (2002) 113, implicitly take the same approach.
¹? Compare United States–Softwood lumber from Canada II, Panel, para 7.152.
ASG which is obviously broader.¹¹ In order for products to be considered like,
they must have identical or closely resembling (physical) characteristics, while the#p#分頁標題#e#
focus for directly competitive products is the extent to which they compete in
the marketplace. As the objective of the ADA is to protect the domestic market
against unfairly traded imports while the objective of the ASG is to protect the
market against fairly traded imports, it is surprising that the product definition in
the ASG is in fact broader than the one in the ADA and the ASCM.
When an anti-dumping investigation is initiated, the product under
consideration is often defined both by descriptive language and by reference to
Harmonized System (HS) numbers, or national versions thereof. If, for example,
the product under consideration is certain types of PET film, it could be defined
as non-self-adhesive film of polyethylene terephthalate (PET film), excluding film
for the manufacture of flexible magnetic discs and classifiable within HS numbers
392062. This definition will then be provided to all interested parties both in the
notice of initiation of the investigation and in the questionnaires that are sent out.
The accuracy of the initial definition is important because it sets the parameters
of the investigation and, at the end of the day, arguably limits the scope of the
application of anti-dumping measures. If in the example above, the administering
authorities exclude PET film for the manufacture of flexible magnetic discs at the
outset of the investigation, they cannot impose anti-dumping measures on this
type of film when the time comes to impose such measures, among others, because
they would not have investigated whether such PET film in fact has been
injuriously dumped.
Dumping 11
¹¹ Argentina–Footwear (EC), AB, para 94.
Dumping:
• like product  product sold in exporting country
to be compared with:
• product under consideration  allegedly dumped product
Injury:
• like product  product sold by domestic industry
to be compared with:
• product under consideration  allegedly dumped product
1.1.1 Sub-dividing the Like Product (Product Control Numbers)
For the purpose of dumping (and injury) margin calculations, it is customary
in most countries to sub-divide the like product in various types or models,
often on the basis of product control numbers (PCNs).¹² Such sub-divisions
are normally prepared by the administering authorities before the case is initiated
and then communicated to all interested parties in the various types of
questionnaires.¹³
If, for example, the like product is polyester textured yarn (PTY), different
types of PTY could be distinguished on the basis of quality, denier, filament
count, flame retardance, colouring and number of twists. If the like product is
steel fasteners, it could be sub-divided on the basis of type (eg nuts, bolts and#p#分頁標題#e#
screws), raw material, Department Idenfification Number (DIN), diameter
(10th of mm) and length (10th of mm):
12 WTO Anti-Dumping Agreement
¹² Compare Czako, Human and Miranda, A Handbook on Anti-Dumping Investigations (2003) 99.
¹³ See section 3.2.5.2, below.
Field description Explanation
Type of fastener Indicate the type of the fastener
Wood screws 1
Self tapping screws 2
Other screws without head 3
Slot and cross-recessed screws 4
Hexagon socket head screws 5
Hexagon bolts 6
Nuts 7
Raw material Indicate the raw material used for producing the
product concerned:
Austenitic steel A1
Austenitic steel A2
Austenitic steel A4
Chromium ferritic steel C2
Chromium ferritic steel C4
Chromium martensitic steel C1
Chromium martensitic steel C3
(Continued)
If the like product is fish, one might distinguish different types on the basis of
quality, fresh/chilled or frozen, presentation (whole, gutted, with/without
head, off-cuts and fillets) and weight of presentation.
Such sub-divisions are arguably necessary in order to make accurate typeby-
type comparisons before calculating producer-specific margins. Thus, for
example, one cannot compare for dumping purposes carbon steel nuts sold by
Chinese producers in China with their exports of stainless steel bolts to Canada.
Similarly, for injury purposes, one should not compare frozen whole salmon
exported from Chile to the EC with fresh salmon fillets produced in Scotland.
The ADA is silent on the legality of such sub-divisions. This is perhaps not
surprising as the need to make them seems so obvious. However, as we will see
below, administering authorities will apply many rules¹? also on a type-by-type,
model-by-model basis, although this does not necessarily appear foreseen by
the ADA. This may make the calculation of particularly dumping margins a
very artificial and speculative exercise.
1.2 Export Price
The key product in an anti-dumping case is the allegedly dumped product, often
referred to as the product under consideration. Not only is this the product
which is allegedly dumped, but it is also the product which is allegedly causing
injury to the domestic industry producing the like product. This product
Dumping 13
¹? For example the 5% home market viability test, the ordinary course of trade test, the 20% sales
below cost test and the concept of inter-model zeroing. As we will see in section 1.4.3.1, below, the AB
ruled that inter-model zeroing is not allowed under the ADA in EC–Bed linen from India, EC–Malleable
cast iron tube and pipe fittings from Brazil and United States–Softwood lumber from Canada II.
Field description Explanation
Other combination D1 (please specify the#p#分頁標題#e#
components in a separate
note)
DIN number Indicate the corresponding DIN number
Diameter Indicate the diameter of the product concerned, in
10th of millimetres, e.g. 045, or 120, etc.
Length Indicate the length of the screw/bolt or the height of the
nut in 10th of millimetres
under consideration is, in other words, the product exported by the producers
or exporters in the exporting country to importers in the importing country. As
a result, it is logical to start the discussion with the analysis of the price of this
product,¹? ie the export price, as we will do here, although this is not the order
followed by the ADA, which starts with the analysis of the price of the product
sold on the domestic market of the foreign producers, the so-called normal value.
Article 2.1 merely provides indirectly that the export price is the price of the
product exported from one country (the exporting country) to another (the
importing country). If, for example, producer x in country X sells T-shirts to
importer y in country Y, then the price charged by producer x to importer y is
the export price. At this stage, the terms and conditions agreed upon by the
buyer and the seller do not matter yet, although they will need to be taken into
account later on. If, for example, producer x sells his T-shirts to importer y
ex-warehouse, cash at delivery, for $1 per T-shirt, then this is the export price.
If producer x sells to importer y delivered, duty-paid, 90 days’ credit, for $2 per
T-shirt, then this is the export price. In other words, what matters at this stage
is the price agreed upon between the buyer and the seller.
As this is virtually always an international transaction, the price is normally
well documented in various transaction documents, such as the commercial
invoice, the letter of credit, the bill of lading, the packing list, duty drawback
claims, etc. Furthermore, the accuracy of these documents in theory¹? can be
verified both at the exporter and at the importer. Consequently, it generally is
not complicated to establish the export price.
1.2.1 Traders
It may happen that foreign producers sell the product under consideration to
other parties, typically traders, in the exporting country which then resell it to
the importing country. On the basis of Article 2.1 it seems clear that the export
price then is the price charged by the trader to the importer. Indeed, the transaction
between the producer and the trader is an internal sale in the exporting
country and in many cases the producer may not even know whether the trader
will export the product under consideration,¹? let alone to which destination.
14 WTO Anti-Dumping Agreement
¹? Compare United States–Softwood lumber from Canada II, Panel, paras 7.152–7.153.#p#分頁標題#e#
¹? In practice, most authorities verify the export price during the verification of the exporter.
¹? For the producer it may not always be evident how to report such sales to traders. If, for example,
he does not know whether the trader will resell the product on the domestic market or for export,
should he then report it as a domestic sale or as an export sale? If the producer knows that the trader
will export the product, but he does not know the destination, should he then report the transactions
as an export sale to the importing country or as third country exports? Obviously, the answer to such
questions is highly fact-specific.
Furthermore, it is the trader that sets the export price and therefore possibly
engages in dumping.
On the other hand, it may be administratively difficult for administering
authorities to make individual calculations for traders. Even if this can be done,
the imposition of anti-dumping duties on traders is of limited use because
traders can easily switch sources of supply. Presumably for these reasons, some
countries normally do not calculate separate margins for traders.
A different approach is often taken when the producer and the trader qualify
as related parties.¹? Administering authorities may then take the position
that they belong to a single economic entity as a result of which, on the one
hand, prices between the two may be unreliable and, on the other hand, actions
by one may be attributed to the other. The authorities may then require the
producer and the related trader to cooperate to ensure that all the producer’s
export sales to the importing country, whether direct or indirect through the
related trader, are reported in the questionnaire response of the producer. In the
example below, the relevant export price would then be the price of 105
charged by the related trader to his independent customers in the importing
country:
1.2.2 Constructed Export Price
There may be cases where the exporter and the importer do not establish an
export price, for example in the case of barter, or where the export price that
they agree upon is unreliable. The latter might be the case if the exporter and
the importer are related parties¹? (in which case the price agreed upon by them
might not be an arm’s length price) or if the administering authorities have reason
to believe, for example, that while the importer pays a high invoiced price, he
receives compensation from the exporter in some other manner. Articles 2.3
and 2.4 (second part) contain important rules for such situations.²?
Article 2.3 ADA authorizes the authorities to construct the export price on
the basis of the price at which the imported products are first resold to an independent
buyer, or if the products are not resold to an independent buyer, or not#p#分頁標題#e#
Producer → Related trader → Independent customer
100 105
Dumping 15
¹? See section 2.2.2, below. ¹? See section 2.2.2, below.
²? Compare United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea,
Panel, para 6.90.
resold in the condition as imported, on such reasonable basis as the authorities
may determine in cases where there is no export price or where it appears to the
authorities concerned that the export price is unreliable because of association
or a compensatory arrangement between the exporter and the importer or a
third party.
By far the most common of these situations is the one where a foreign
producer sells the product under consideration to a related party which then
resells to unrelated customers. In application of Article 2.3 ADA many administering
authorities will then almost automatically²¹ proceed to construct the
export price on the basis of the resale price to the first independent customer.
However, authorities are under no obligation to do so, as the use of the word
‘may’ makes clear.
Thus, in the example above, most administering authorities will use the price
of 140 charged by the related importer to the independent customer. The construction
of the export price on the basis of the price to the first independent
customer is not problematic in and of itself because it is a market price.
However, the construction of the export price may become problematic
because Article 2.4 ADA provides in relevant part that in such cases allowances
for costs, including duties and taxes, incurred between importation and resale,
and for profits accruing, should also be made.
Thus, Article 2.4 authorizes administering authorities to deduct all costs
incurred between importation and resale as well as profits accruing from the
price to the first independent customer.
Producer → Related importer → Independent customer
100 Duty: 14 140
Freight: 5
Storage: 2.8
SGA: 14 (10%)
Profit: 4.2 (3%)
Producer → Related importer → Independent customer
100 140
16 WTO Anti-Dumping Agreement
²¹ This automaticity may be questioned as the authorities must technically establish that the export
price is unreliable.
In the example above, the costs²² incurred between importation and resale are
35.8 while the 3 per cent profit realized by the related importer on his turnover
amounts to 4.2. If such costs and profit were to be deducted by the administering
authorities, the constructed export price would be 100. In other words,
it would be the same as the export price charged by the exporter to the related
importer and the construction of the export price therefore would be ‘neutral’.#p#分頁標題#e#
In practice, however, some administering authorities will not consider the
profit margin realized by the related importer on the ground that it is influenced
in part by the unreliable purchase price. They will therefore impute a
profit margin to the related importer, supposedly based on the profit margins
generated by independent importers cooperating in the same proceeding.
While it is not clear whether the imputation of such profit is allowed, the ADA
in any event provides no guidance on how to do this, as a result of which
different administering authorities use different methods.
In United States–Stainless steel plate in coils and stainless steel sheet and strip,
the panel ruled that administering authorities on the one hand are not required
to make these special allowances, but on the other hand in no event may make
allowances that are not provided for in Article 2.4 when they construct the
export price:²³
The term ‘should’ in its ordinary meaning generally is non-mandatory, i.e., its use in
Article 2.4 indicates that a Member is not required to make allowance for costs and
profits when constructing an export price. We believe that, because the failure to make
allowance for costs and profits could only result in a higher export price – and thus a
lower dumping margin – the ADA merely permits, but does not require, that such
allowances be made.
. . . Article 2.4 provides an authorization to make certain specific allowances.
Allowances not within the scope of that authorization cannot be made. If a Member
were free to make any additional allowances it desired, there would be no effective
disciplines on the methodology for construction of an export price and the provision
in question would be reduced to inutility.²?
Dumping 17
²² In the case of multi-product importers, SGA (selling, general and administrative expense) will
normally need to be allocated between the product under consideration and other products.
²³ Cunningham and Cribb, ‘Dispute Settlement Through the Lens of “Free Flow of Trade”: A
Review of WTO Dispute Settlement of US Anti-Dumping and Countervailing Duty Measures’
(2003) 6 Journal of International Economic Law 155, 157–158 criticize this aspect of the panel
report: ‘In plain-speak, the panel’s conclusion was “When it says you ‘should’ examine these things, it
means you ‘may’ examine these things and when we say you ‘may’ examine these things, you may
examine only these specifically enumerated things and no other things.” ’
²? United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea, Panel, paras
6.93–6.94.
18 WTO Anti-Dumping Agreement
Domestic market Export market#p#分頁標題#e#
Manufacturer
100 100 CIF
Unrelated distributor
Dealer Dealer
Related distributor
-indirect costs: 16
-direct costs: 7
-profit: 7
-indirect costs: 16
-direct costs: 7
-profit: 7
130 130
In the same case, the panel ruled that the concept of ‘costs incurred between
importation and resale’ cannot be stretched to include costs, in casu bad debt as
a result of bankruptcy, that not only were not incurred in an accounting sense
until after the date of resale but which were entirely unforeseen at that time.²?
²? United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea, Panel, para
6.100.
Article 2.4, fourth and fifth sentences, provides that in cases where price
comparability has been affected by the construction of the export price,
the authorities shall establish the normal value at a level of trade equivalent to
the level of trade of the constructed export price, or shall make due allowance
as warranted under Article 2.4.²?
It seems clear that after all costs incurred between importation and resale, as
well as profits accruing, have been deducted from the price to the first independent
customer, what is left is the price of the product under consideration
as if it had been sold to an unrelated importer. In the distribution chain, such
an unrelated importer effectively functions as a distributor and therefore this
constructed export price is the price to an unrelated distributor.
Thus, Article 2.4 would appear to require that where such costs and profits are
deducted, the remaining price must be compared with a normal value, similarly
established at the level of sale to a domestic distributor or, alternatively, that due
allowance for the level of trade difference be made on the normal value side.
Thus, in the example on page 18, the constructed export price of 100
(130  7  7  16  100) should be compared with the domestic price to
the unrelated distributor of 100.27
1.3 NormalValue
1.3.1 Normal Value: Based on Domestic Sales
Article 2.1 indirectly defines normal value as the comparable price in the
ordinary course of trade for the like product when destined for consumption in
the exporting country. Thus, the normal value is normally the price of the
like product in the exporting country, as long as the product is destined for
consumption there and the price is in the ordinary course of trade.
1.3.1.1 Destined for Consumption
The requirement that the like product be destined for consumption in the
exporting country thus far has not given rise to WTO disputes. However, this
does not mean that it may not be problematic in practice. Some countries, for
example, track imported raw materials all the way to exportation of the finished#p#分頁標題#e#
Dumping 19
²? This sentence was added during the Uruguay Round negotiations and has been called a ‘significant
change’, see Koulen, ‘The New Anti-Dumping Code Through its Negotiating History’, in
Bourgeois, Berrod and Gippini Fournier (eds), The Uruguay Round Results (1995) 151–232.
²? Note that at this stage we are only discussing the special level of trade adjustment of Art 2.4 ADA.
Direct selling costs included in both prices will still need to be taken out later on, see section 1.4.1, below.
product. If the like product is an intermediate product to be used in a final
product which will be exported, some authorities may then question whether
the intermediate product is destined for consumption in the country of export,
particularly in situations where the producer of the intermediate product
knows at the moment of sale to a domestic customer that the processed product
will eventually be exported. The following example may clarify this:
In this example the like product is polyester staple fibres (PSF). The producer
under investigation imports the raw materials PTA and MEG from which he
manufactures PSF. He then sells this PSF to domestic companies which
produce polyester yarns which they export. On the basis of the wording of
Article 2.1 ADA it seems clear that the like product under investigation, PSF,
is destined for consumption in the exporting country as it is consumed in the
product of the processed product polyester yarn which is equally manufactured
in the exporting country market. Nevertheless, some administering authorities
will exclude such domestic sales of the like product on the ground that they are
incorporated in a processed product which is exported.
1.3.1.2 Ordinary Course of Trade
The ADA does not define the concept ‘ordinary course of trade’.²? However,
Article 2.1 appears to establish that only domestic prices in the ordinary
course of trade are to be used as normal value. This implies that administering
authorities would normally need to check whether domestic sales are or are not
made in the ordinary course of trade. While there may be various reasons why
domestic sales are not made in the ordinary course of trade, two situations are
most common:
• Some or all domestic transactions are sold below cost; or
• Domestic sales are made to related parties.
Exclusion of sales below cost Many authorities consider that under certain
circumstances sales below cost are not in the ordinary course of trade. In order
to determine whether domestic sales are in the ordinary course of trade, the
authorities therefore will need not only information on domestic prices, but
also on the costs of the domestic transactions concerned in order to determine
PTA/MEG → Polyester staple fibres → Polyester yarns#p#分頁標題#e#
Imported Produced & sold domestically Produced/exported
20 WTO Anti-Dumping Agreement
²? Compare United States–Hot Rolled Steel from Japan, Panel, para 7.108.
whether domestic prices are above or below cost. While many authorities
routinely request cost information in the questionnaires and then check such cost
information during the investigation, others will only request and check cost
information if there are allegations by the domestic industry that the exporters
have been selling below cost. In this context, the panel in Guatemala–Cement II
held that nothing in the ADA prevents an investigating authority from requesting
cost information, even if the domestic industry does not allege sales below cost.²?
Thus, the authorities are free to proactively seek information in this regard.
It is important to note that the ADA does not require authorities to treat
domestic sales below cost as not in the ordinary course of trade.³? Article 2.2.1
ADA merely provides that such sales may be treated as not in the ordinary
course of trade and may be disregarded only if the authorities determine that
sales below cost are made:
• within an extended period of time (normally one year, but never less than
six months);
• in substantial quantities (meaning that either the weighted average selling
price of the transactions under consideration for the determination of the
normal value is below the weighted average per unit costs, or the volume
of sales below per unit costs represents not less than 20 per cent of the
volume sold in transactions under consideration for the determination of
the normal value);
• at prices which do not provide for the recovery of all costs within a
reasonable period of time.
Article 2.2.1 ADA
Sales of the like product in the domestic market of the exporting country or sales to a
third country at prices below per unit (fixed and variable) costs of production plus
administrative, selling and general costs may be treated as not being in the ordinary
course of trade by reason of price and may be disregarded in determining normal value
only if the authorities³¹ determine that such sales are made within an extended period
of time³² in substantial quantities³³ and are at prices which do not provide for the recovery
Dumping 21
²? Compare Guatemala–Cement from Mexico II, Panel, para 8.183.
³? Compare Czako, Human and Miranda, A Handbook on Anti-Dumping Investigations (2003) 152.
³¹ Fn 3 in ADA: When in this Agreement the term ‘authorities’ is used, it shall be interpreted as
meaning authorities at an appropriate senior level.
³² Fn 4 in ADA: The extended period of time should normally be one year but shall in no case be#p#分頁標題#e#
less than six months.
³³ Fn 5 in ADA: Sales below per unit costs are made in substantial quantities when the authorities
establish that the weighted average selling price of the transactions under consideration for the determination
of the normal value is below the weighted average per unit costs, or that the volume of sales
below per unit costs represents not less than 20 per cent of the volume sold in transactions under
consideration for the determination of the normal value.
of all costs within a reasonable period of time. If prices which are below per unit costs at
the time of sale are above weighted average per unit costs for the period of investigation,
such prices shall be considered to provide for recovery of costs within a reasonable
period of time.
It is important to bear in mind that many authorities apply these tests on a
PCN-specific basis. This is therefore an area where the PCN classification³?
may play an important role.
In practice, the second test is often the most important. In applying this test,
the authorities will normally first check whether the weighted average domestic
selling price of a PCN is lower than its weighted average per unit cost. If this is
the case, the domestic sales of that PCN will be ignored and recourse will be
had to constructed normal value. This alternative manner of establishing
normal value is discussed in more detail in section 1.3.2, below. Where the
weighted average domestic selling price of a PCN is higher than its weighted
average per unit cost, the authorities will then check whether the volume of
sales below per unit costs represents 20 per cent or more of the volume sold in
transactions under consideration for the determination of the normal value.
Where sales below cost represent less than 20 per cent of the total domestic sales of
the PCN, such sales below costs must be included in the calculation of normal value,
based on domestic prices. However, where sales below cost represent 20 per cent
or more of the total domestic sales of the PCN, such sales below cost may be
excluded and normal value may then be based on the remaining domestic sales of
the PCN above cost. As domestic sales below costs are obviously lower-priced than
sales at or above cost, such exclusion will increase the dumping margin, compared
with inclusion of the below cost sales, as the following example may show:
22 WTO Anti-Dumping Agreement
³? See section 1.1.1, above.
Suppose the PCN cost is 90 and four transactions of equal weight took place in
the domestic and the export market as follows:
Domestic Export
(1) 1 June 50 50
(2) 10 June 100 100
(3) 15 June 150 150
(4) 20 June 200 200
(Continued)
Despite the discretion that Article 2.2.1 grants investigating authorities, many
authorities will exclude sales below costs as a matter of course where such sales#p#分頁標題#e#
represent 20 per cent or more of total domestic sales.
Domestic sales to related parties Some authorities also consider that domestic
sales to related parties may not be in the ordinary course of trade. Typically this
occurs if a producer sells to a related domestic distributor which then resells to
an unrelated dealer or end-user.
As is the case on the export side, domestic sales to related parties in fact may or
may not be made on an arm’s length basis and arguably – rather than the qualification
of the purchaser (ie related or unrelated) – the reliability of the prices
charged ought to be the focus of the enquiry. In practice, however, some
authorities assume as a matter of course that sales between related parties are
unreliable and therefore will automatically treat them as not being in the ordinary
course of trade. Thus far, this extreme interpretation has not been challenged
in the WTO.
Ironically, however, a less radical interpretation by the American
Department of Commerce was successfully challenged by Japan in United
States–Hot rolled steel.³? Under the so-called arm’s length test, the
Commerce Department tested whether there were differences in domestic
Producer → Related distributor → Unrelated customer
100 130
Dumping 23
³? United States–Hot Rolled Steel from Japan, Panel, paras 7.108–7.112.
The weighted average domestic price after the exclusion of the domestic sale
below cost made on 1 June is: 150 [(100  150  200)  3  150]. Compared
with a weighted average export price of 125 [(50  100  150 
200)  4 125], this gives a dumping amount of 100 [(150  125)  4  100]
and an approximate dumping margin of 20% (100  500  100  20%).
Suppose that the authority would include the domestic sale below cost, the
weighted average domestic price would be 125 [(50  100  150  200)  4
 125], as a result of which no dumping would be found.
pricing to affiliated customers as compared with domestic pricing to
unaffiliated customers:
The USDOC 99.5% test
Under this test, the DOC checked whether the weighted average sales price of
domestic sales to an affiliated party was lower than the weighted average sales price
to all non-affiliated customers. If it was more than 0.5% lower, the domestic related
party sales concerned were automatically excluded from the calculation of
normal value. However, the DOC did not check whether affiliated party sales were
higher-priced.
The panel considered this test objectionable and therefore not a permissible
interpretation of the ADA because of its one-sided nature: the arm’s length test
only checked whether prices to affiliated customers were lower, on average,
than prices to unaffiliated customers and therefore would treat only the#p#分頁標題#e#
lower-priced sales to affiliates as not in the ordinary course of trade. Relatively
high-priced sales to affiliates, on the other hand, were not tested for and would
be treated as ordinary course of trade sales by the Commerce Department. The
panel considered this one-sided test result-oriented:
The result of application of the ‘arm’s length’ test . . . is the exclusion from the determination
of normal value of prices that are, on average, lower. As a result, the application
of the ‘arm’s length’ test cannot but skew the normal value upward, thereby making a
finding of dumping, or a higher margin of dumping, more likely.³?
On appeal, the Appellate Body (AB) upheld the conclusion of the panel, albeit
for partially different reasons.
In our view, the duties of investigating authorities, under Article 2.1 of the ADA,
are precisely the same, whether the sales price is higher or lower than the ‘ordinary
course’ price, and irrespective of the reason why the transaction is not ‘in the ordinary
course of trade’. Investigating authorities must exclude, from the calculation of normal
value, all sales which are not made ‘in the ordinary course of trade’. To include such
sales in the calculation, whether the price is high or low, would distort what is defined
as ‘normal value’.
. . . Although . . . the ADA affords WTO Members discretion to determine how to
ensure that normal value is not distorted through the inclusion of sales that are not ‘in
the ordinary course of trade’, that discretion is not without limits. In particular, the
discretion must be exercised in an even-handed way that is fair to all parties affected by
an anti-dumping investigation. If a Member elects to adopt general rules to prevent
distortion of normal value through sales between affiliates, those rules must reflect,
24 WTO Anti-Dumping Agreement
³? United States–Hot Rolled Steel from Japan, Panel, para 7.112.
even-handedly, the fact that both high and low-priced sales between affiliates might
not be ‘in the ordinary course of trade’.³?
Thus, both the panel and the AB rejected the 99.5 per cent test because it was
one-sided and not even-handed.³?
Assuming that authorities do apply an appropriate test for determining
whether sales to related parties in the domestic market are in the ordinary
course of trade and find, on the basis of the application of the test, that this is
not the case, the consequence of this finding will be that they may then
conclude that the domestic sales to the related parties are not in the ordinary
course of trade. However, on what basis should they then determine normal
value? The obvious answer might appear to be: on the basis of domestic sales to#p#分頁標題#e#
unrelated customers.³? However, such sales may not exist, may be unrepresentative??
or may lead to exclusion of a significant portion of domestic sales.
Possibly for such reasons, some authorities will not in fact exclude sales
transactions to related parties, but will rather ignore the prices charged between
the related parties and instead move down the distribution chain and base
normal value on the resale price by the related party to independent customers.
Thus, in the example above, the intra-group price of 100 will be ignored and
instead normal value will be based on the price of 130 charged by the related
distributor to its independent customer. In most cases, this price will be higher
because it is a level down the distribution chain. The independent customer
will typically be a dealer or an end-user who pays a higher price than the
distributor because the distributor should normally cover its costs as well as a
reasonable profit from the difference between its purchase price and its resale
price.
In fact, there is conceptually a certain symmetry with sales to related parties
on the export side, where Article 2.3 jo. 2.4 ADA provide special rules for
Producer → Related distributor → Independent customer
100 130
Dumping 25
³? United States–Hot Rolled Steel from Japan, AB, paras 141–148. WorldTradeLaw.net Dispute
Settlement Commentary on US–Hot rolled steel, p 19, checked 23 April 2005, considers that ‘the fact
that the Appellate Body did not rely on specific language in Article 2.1 as support for these statements
suggests that this principle [of fundamental fairness] applies throughout’ the ADA.
³? Durling, ‘Deference, But Only When Due: WTO Review of Anti-Dumping Measures’ (2003)
6 Journal of International Economic Law 125, 135.
³? Czako, Human and Miranda, A Handbook on Anti-Dumping Investigations (2003) 150.
?? See section 1.3.1.3, below.
handling this situation through the constructed export price mechanism.?¹On
the normal value side, however, a similar provision is lacking.
In United States–Hot rolled steel, the panel had determined that the ADA
does not allow for the possibility of using such ‘replacement sales’ because the
26 WTO Anti-Dumping Agreement
?¹ See section 1.2.2, above.
100
Domestic market Export market
Manufacturer
100 CIF
Unrelated distributor Related distributor
-indirect costs: 16
-direct costs: 7
-profit: 7
-indirect costs: 16
-direct costs: 7
-profit: 7
130 130
Dealer Dealer
ADA is producer/exporter based and because a provision similar to Article 2.3
does not exist for the normal value side.?² This reasoning was arguably too
simplistic because it treated the producer and its related distributors as#p#分頁標題#e#
different entities although – not contestedly – they belong to the same corporate
group. To treat group members as a single economic entity is in fact quite normal,
not only for dumping, but also for other purposes including accounting,?³
taxation, corporate law, etc.
The AB did indeed overrule the panel by holding that the use of downstream
domestic sales by the USDOC constituted a permissible interpretation of
Article 2.1 ADA. However, it followed different reasoning:
The text of Article 2.1 is . . . silent as to who the parties to relevant sales transactions
should be. Thus, Article 2.1 does not expressly mandate that the sale be made by the
exporter for whom a margin of dumping is being calculated. Nor does Article 2.1
expressly preclude that relevant sales transactions might be made downstream,
between affiliates of the exporter and independent buyers. In our view, provided that
all of the explicit conditions in Article 2.1 of the ADA are satisfied, the identity of the
seller of the ‘like product’ is not a ground for precluding the use of a downstream sales
transaction when calculating normal value. In short, we see no reason to read into
Article 2.1 an additional condition that is not expressed.??
Like the panel, the AB therefore treated the producer and the related party as
different entities, but considered that as Article 2.1 is silent as to who should
sell the product on the domestic market, it was not necessary that the domestic
seller be the same as the producer/exporter. It therefore proceeded from the
same assumption (different parties) as the panel, but drew the opposite
conclusion from the ‘silence’ of Article 2.1 as to the identity of the seller.
While one may have doubts about the correctness of this two-step approach
(indeed, the single economic entity approach avoids this quagmire), the AB’s
subsequent reasoning makes it clear that the AB understood the underlying
problems inherent in the use of downstream sales very well:
. . . The use of downstream sales prices to calculate normal value may affect the comparability
of normal value and export price because, for instance, the downstream sales
may have been made at a different level of trade from the export sales. Other factors
may also affect the comparability of prices, such as . . . the costs and profits of the
reseller. Thus, we believe that when investigating authorities decide to use downstream
sales to independent buyers to calculate normal value, they come under a particular
Dumping 27
?² United States–Hot Rolled Steel from Japan, Panel, paras 7.114–7.118.
?³ Thus, for example, results of different group members will normally be consolidated for
accounting purposes. ?? United States–Hot Rolled Steel from Japan, AB, para 166.#p#分頁標題#e#
duty to ensure the fairness of the comparison because it is more than likely that
downstream sales will contain additional price components which could distort the
comparison.??
. . . By making the allowances required under Article 2.4 of the ADA, the
investigating authorities should, in effect, arrive at a price which corresponds to the
‘ex-factory’ price of the ‘like product’ for the specific exporter concerned, as required by
that provision.??
Thus without explicitly saying so, the AB essentially instructs investigating
authorities that where they use domestic downstream sales, they must deduct
the same price components, notably the reseller’s costs and profit, that are
deducted in a constructed export price situation, in order to arrive at an ex
factory domestic price. The AB thereby took the same line that a previous
GATT panel had taken in the – unadopted – panel report EC–Audio tapes in
cassettes.?? While the Article 2.4 allowances will be discussed in more detail
below,?? at this juncture the application of the AB’s reasoning to the facts in the
flowchart opposite may indicate its importance.
On the export side, the starting point on the basis of Article 2.3 will be the
price charged by the related distributor to the unrelated dealer of 130. From
this price, the investigating authority ex Article 2.4 will then deduct both direct
(7) and indirect (16) costs as well a reasonable profit (eg 7) of the reseller in
order to arrive at a CIF price of 100. From this price, similarly ex Article 2.4,
ocean freight (eg 4), insurance (eg 0.2) and inland freight (eg 0.8) will then be
deducted to arrive at an ex factory export price of 95 (130  16  7  7 
4  0.2 0.8  95).
The AB has now clarified that on the domestic side investigating authorities
may also use the price charged by the related distributor to the unrelated dealer
of 130, but they should then ex Article 2.4 effectively deduct the same price
components that they deduct on the export side, ie in the example above, both
direct (7) and indirect (16) costs as well as a reasonable profit (eg 7) of the
reseller and the inland freight paid by the producer (eg 0.8) in order to arrive at
an ex factory domestic price of 99.2 (130  16  7  7  0.8  99.2). This
would lead to a dumping margin, expressed as a percentage of the CIF price, of:
99.2  95  4.2  100  4.2%, a result that most of us would consider fair
on the basis of the facts above.
28 WTO Anti-Dumping Agreement
?? United States–Hot Rolled Steel from Japan, AB, para 168
?? United States–Hot Rolled Steel from Japan, AB, para 170.
?? EC–Audio tapes in cassettes, GATT panel report, paras 374 and 377.
?? See section 1.4.1, below.
Prior to the AB’s clarification, some investigating authorities would, on the#p#分頁標題#e#
domestic side, indeed have used the price charged by the related distributor to
the unrelated dealer of 130, but they would then have deducted only direct
selling expenses incurred by either the related distributor (7) or the producer
(0.8) to arrive at a domestic price of 122.2 (130  7  0.8  122.2). In the
example above, this would have resulted in a dumping margin, expressed as a
Dumping 29
100 delivered
Domestic market Export market
Manufacturer
100 CIF
Related distributor Related distributor
130 130
Dealer Dealer
-indirect costs: 16
-direct costs: 7
-profit: 7
-indirect costs: 16
-direct costs: 7
-profit: 7
percentage of the CIF export price, of: 122.2  95  27.2  100  27.2%.
It may therefore be clear that the ruling of the AB is of significant practical
importance.
1.3.1.3 Five per cent rule
Normal value obviously cannot be based on domestic sales where there are no
domestic sales. Thus, if a producer sells only for export, his normal value will
have to be determined on another basis. Similarly, if a producer sells frozen
shrimp for exports (for example because of geographical distance between the
markets), while he only sells fresh shrimp domestically, the two different types
cannot be compared with each other and again export prices of frozen shrimp
will have to be compared with a normal value for frozen shrimp, established on
a different basis. These alternative bases are the subject of the next section.
However, it is also possible that a producer does sell the like product on both
markets, but that the quantity sold on the domestic market is relatively small.
This happens often in countries with small home markets such as Singapore or
Hong Kong.
Even in larger markets, however, it may happen frequently that while the
like product is sold in sufficient quantities on both markets, certain types are
sold more on the export market, while other types are sold more domestically,
for example, because of differences in consumer tastes, technical specifications,
etc. It may then occur that the exported types are sold only in small quantities
on the domestic market.
Footnote 2 ADA
Sales of the like product destined for consumption in the domestic market of the
exporting country shall normally be considered a sufficient quantity for the determination
of the normal value if such sales constitute 5 per cent or more of the sales of the
product under consideration to the importing Member, provided that a lower ratio
should be acceptable where the evidence demonstrates that domestic sales at such
lower ratio are nonetheless of sufficient magnitude to provide for a proper comparison.
Footnote 2 ADA provides as a general rule that domestic sales of the like
product are sufficient to base normal value on if they account for 5 per cent or#p#分頁標題#e#
more of the sales of the product under consideration to the importing country
market. This is often called the five per cent or home market viability test. It is
important to note that footnote 2 clearly contemplates that this test be applied
on a global basis by comparing sales of the like product in the domestic market
with sales of the product under consideration in the importing market. Despite
this clear admonition, many investigating authorities apply the five per cent
30 WTO Anti-Dumping Agreement
test not only on a global basis, but also on a type-by-type, model-by-model
basis.?? This is therefore one of the areas where PCNs?? have gained great
importance.
Suppose, for example, that the product under consideration is farmed
salmon, whether fresh or frozen, and that quantities sold in the two relevant
markets are as follows:
In the example above, the global five per cent test is met because sales of the like
product on the domestic market account for 6 per cent of export sales of the
product under consideration (12,000  200,000  100  6%). However,
when applied to each type of salmon (fresh and frozen), the five per cent test is
met only for the relatively small quantity of exported fresh salmon
(9,000  1,000  100  900%), but not for the far larger quantity of
exported frozen salmon (3,000  199,000  100  1.51%). Many investigating
authorities will therefore consider that domestic sales of frozen salmon
are not representative and refuse to base normal value on these domestic sales.
In this context, it is to be noted that administering authorities will typically
apply both tests, with the global test operating as an overall benchmark test and
the PCN-based test coming into play only in cases where the global test has been
met. Needless to say, the more detailed the PCNs, the higher the likelihood that
one or more exported PCNs are not sold in sufficient quantities on the domestic
market, as a result of which recourse must be had to one of the alternative bases
for calculating normal value for such PCNs. Much depends, therefore, on the
level of classification details requested by the investigating authorities.
An example may clarify this. If the product under investigation is colour televisions
from Korea, the investigating authorities might determine that colour
televisions with different broadcasting standards constitute different types and
require a different PCN for each. Such determination would almost certainly
result in a finding that colour televisions with the NTSC broadcasting system
Type Export market Domestic market
Frozen 199,000 3,000
Fresh 1,000 9,000
Total 200,000 12,000
Dumping 31
?? Compare Didier, ‘The WTO Anti-Dumping Code and EC practice’ (2001) 35 Journal of
World Trade 33, 36. ?? See section 1.1.1, above.#p#分頁標題#e#
sold in Korea cannot be compared with colour televisions exported to the EC
with the PAL/SECAM broadcasting system, as a result of which none of the
colour televisions sold in Korea would be viable.
However, an investigating authority could just as – if not more – plausibly
determine that the differences in broadcasting standards constitute a minor
physical difference, for which an allowance can be made on the basis of Article
2.4 ADA. Such a determination, in turn, might well lead to a finding that most
exported television types have a viable domestic equivalent.
As the practice of sub-dividing the like product into PCNs is not discussed
in the ADA, this is an area where investigating authorities have almost unfettered
discretion. One could argue that the fact that the ADA is silent on the
practice, coupled to repeated ADA references to the concepts of ‘like product’
and ‘product under consideration’, evidences that the ADA in fact does not
allow such sub-divisions.
Such a broad claim, however, seems too blunt. A certain level of precision is
required in order to effect a fair comparison between normal value and export
price and the intermediate steps of sub-dividing the like product into PCNs
and calculating margins per PCN before calculating a weighted average margin
per producer are often a justifiable means to attain that objective.?¹ Thus, rather
than attacking the practice as such, it may be more appropriate to challenge
case-specific, result-oriented PCNs devised by shrewd investigating authorities
with a view to inflating dumping margins on the ground that unbiased and
objective authorities would not have established such PCNs.
1.3.2 Normal Value: Based on Constructed Normal Value
Article 2.2 ADA provides that normal value may be based on third country
exports or constructed normal value in three situations:
• no domestic sales of the like product in the ordinary course of trade;
• insufficient (less than 5 per cent) domestic sales of the like product; or
• particular market situation in the domestic market.
Article 2.2 ADA
When there are no sales of the like product in the ordinary course of trade in the domestic
market of the exporting country or when, because of the particular market situation or
the low volume of the sales in the domestic market of the exporting country,?² such
32 WTO Anti-Dumping Agreement
?¹ Compare the discussion by the AB in the context of inter-model zeroing in United States–Softwood
lumber from Canada II, discussed below in section 1.4.3.1.
?² Fn 2 in ADA: Sales of the like product destined for consumption in the domestic market of the
exporting country shall normally be considered a sufficient quantity for the determination of the
sales do not permit a proper comparison, the margin of dumping shall be determined#p#分頁標題#e#
by comparison with a comparable price of the like product when exported to an
appropriate third country, provided that this price is representative, or with the cost of
production in the country of origin plus a reasonable amount for administrative,
selling and general costs and for profits.
While the third situation does not occur very often, it happens frequently that
investigating authorities determine that one of the first two situations exists.
As we have seen above, such findings are facilitated by the practice of many
investigating authorities to apply the tests both on a global basis and on a PCNspecific
basis.
With respect to the third country exports alternative, Article 2.2 merely
provides that export sales to an appropriate third country may be used,
provided that the price is representative. Use of third country export sales is
relatively rare, with only the United States using it on a regular basis. Many
other investigating authorities prefer to use the second alternative, constructed
normal value, without even checking third country exports’ data.?³
At least some investigating authorities have expressed a reluctance to use
third country exports as the basis for normal value on the ground that, if
exports are allegedly dumped into their market, it cannot be precluded that
such exports are dumped into third country markets also. As the very definition
of dumping depends in part on the price of the third country exports, one may
wonder whether such simplistic reasoning passes muster. However, as the ADA
leaves investigating authorities a full choice between the two methods, without
even a motivation requirement, use of one rather than the other method seems
difficult to challenge, unless one could make a case that the methodology used
does not lead to a fair comparison.
Article 2.2 defines constructed normal value as the cost of production?? in
the country of origin plus a reasonable amount for administrative, selling and
general costs and for profits. Cost of production in this sense is often called cost
of manufacture and typically includes all the costs incurred in the factory.
Administrative, selling and general costs are often called SGA and include both
direct and indirect selling costs as well as administrative and other general costs
Dumping 33
normal value if such sales constitute 5 per cent or more of the sales of the product under consideration
to the importing member, provided that a lower ratio should be acceptable where the evidence demonstrates
that domestic sales at such lower ratio are nonetheless of sufficient magnitude to provide for a
proper comparison.
?³ Compare Durling and Nicely, Understanding the WTO Anti-Dumping Agreement: Negotiating
History and Subsequent Interpretation (2002) 35.
?? Art 2.2.1 clarifies that costs include both fixed and variable costs.#p#分頁標題#e#
incurred after the product leaves the factory. If, for example, a company has a
factory and a headquarters, the SGA costs will normally be the costs incurred
by the headquarters.
It is important to bear in mind that the purpose of the constructed normal
value is to construct a price of the exported product, as if it would have been
sold on the domestic market. This has two consequences. First, as far as cost of
production is concerned, one should calculate the cost of production of the
exported product. If, for example, a car compact disc player (CCDP) is sold for
export with an anti-theft device, while the domestically sold CCDP is sold
without such a device,?? the cost of production of the CCDP should include
the cost of the anti-theft device. Similarly, if, for example, the country of export
uses a duty drawback system, the cost of production of the exported product
would normally be reported exclusive of import duties because the exported
product does not include import duties.
Second, one should add to this cost of production of the exported product
the SGA and profit on the domestic market. With regard to the reasonable
profit requirement, explored in more detail below, it may be noted at this stage
that WTO panels have repeatedly held that Article 2.2 does not provide an
overarching reasonableness requirement and that where one of the methods
provided in Article 2.2.2 is used in a legitimate manner, the result will qualitate
qua be reasonable for purposes of Article 2.2.?? Thus, in EC–Bed linen, the use
of a profit margin of 18.6 per cent was not considered illegal as such.?? In
Thailand–H-Beams, a profit margin of 36.3 per cent was used.??
1.3.2.1 Cost Calculations and Allocations
With regard to the calculation of constructed normal values, Article 2.2.1.1
sets out two important general principles, applicable to both cost of manufacture
and SGA:
• Costs must normally be calculated on the basis of records kept by the
exporter or producer under investigation, provided that such records:
• are in accordance with the generally accepted accounting principles
(GAAP) of the exporting country; and
• reasonably reflect the costs associated with the production and sale of
the product under consideration;
34 WTO Anti-Dumping Agreement
?? It is assumed that the domestically sold and the exported CCDP fall under the same PCN. If a
different PCN has been assigned to each type, the problem does not arise.
?? Compare Thailand–H-Beams from Poland, Panel, para 7.121; EC–Bed linen from India, Panel,
para 6.96. ?? EC–Bed linen from India, Panel, para 6.100.
?? Thailand–H-Beams from Poland, Panel, paras 7.109 and 7.128.
• Authorities must consider all available evidence on the proper allocation of#p#分頁標題#e#
costs, including that which is made available by the exporter or producer
in the course of the investigation, provided that such allocations have been
historically utilized by the exporter or producer, in particular in relation to
establishing appropriate amortization and depreciation periods and
allowances for capital expenditures and other development costs.
Compared to a price-based normal value, constructed normal value calculations
are often controversial because the ADA offers investigating authorities
substantial leeway to calculate and allocate costs and profits and it happens
frequently that the authorities do not agree with the calculations made by the
exporters concerned. Not surprisingly, therefore, these issues have come up in
a number of dispute settlement cases.
Article 2.2.1.1 ADA
For the purpose of paragraph 2, costs shall normally be calculated on the basis of
records kept by the exporter or producer under investigation, provided, that such
records are in accordance with the generally accepted accounting principles of the
exporting country and reasonably reflect the costs associated with the production and
sale of the product under consideration. Authorities shall consider all available
evidence on the proper allocation of costs, including that which is made available by
the exporter or producer in the course of the investigation provided that such
allocations have been historically utilized by the exporter or producer, in particular in
relation to establishing appropriate amortization and depreciation periods and
allowances for capital expenditures and other development costs. Unless already
reflected in the cost allocations under this sub-paragraph, costs shall be adjusted
appropriately for those non-recurring items of cost which benefit future and/or current
production, or for circumstances in which costs during the period of investigation are
affected by start-up operations.??
In United-States–DRAMS, the panel ruled that the rules concerning GAAP
and reflection of costs of production and sale only apply to records kept by the
exporter or producer under investigation. The USDOC rejection of projections
for future costs, prepared by an outside consultant on behalf of one of the
Korean producers, therefore did not violate Article 2.2.1.1.??
United States–Softwood lumber II raised an interesting cost allocation issue. The
panel had determined that Article 2.2.1.1 simply provided that investigating
authorities must consider all available evidence on the proper allocation of
Dumping 35
?? Fn 6 in ADA: The adjustment made for start-up operations shall reflect the costs at the end of
the start-up period or, if that period extends beyond the period of investigation, the most recent costs
which can reasonably be taken into account by the authorities during the investigation.#p#分頁標題#e#
?? United States–DRAMS from Korea, Panel, para 6.66.
costs, including that made available by respondents, in so far as such allocations
have been historically utilized by the exporter or producer?¹ and that it therefore
did not require the authorities to compare various possible allocation
methods. On appeal, the AB considered this too categorical. The AB held that
the term consider means that an investigating authority must ‘reflect on’ and
‘weigh the merits of ’ all available evidence on the proper allocation of costs,?²
which may have to include a comparison between allocation methods:
. . . the parameters of the obligation to ‘consider all available evidence’ will vary case-bycase.
It may well be that, in the light of the facts of a particular case, the requirement to
‘consider all available evidence’ may be satisfied by the investigating authority without
comparing allocation methodologies or aspects thereof. However, in other instances –
such as where there is compelling evidence available to the investigating authority that
more than one allocation methodology potentially may be appropriate to ensure that
there is a proper allocation of costs – the investigating authority may be required to
‘reflect on’ and ‘weigh the merits of ’ evidence that relates to such alternative allocation
methodologies in order to satisfy the requirement to ‘consider all available evidence’.
Thus, although . . . Article 2.2.1.1 does not, as a general rule, require investigating
authorities to compare allocation methodologies to assess their respective advantages
and disadvantages in each and every case, there may be particular instances in which
the investigating authority may be required to compare them . . . ?³
In Egypt–Steel rebar, Turkey had claimed that the failure of the Egyptian
authorities to offset short-term interest income against interest expenses
violated Article 2.2.1.1.?? The panel determined that it must consider the
evidence on the record in order to decide whether there was evidence that
short-term interest income was reasonably related to the cost of producing and
selling rebar, as a result of which it should have been taken into account by the
Egyptian authorities in the cost of production calculation.?? The panel noted
that the Turkish exporters had offset the interest income in their questionnaire
responses, but that Turkey had not identified, nor had the panel found, any
evidence that would demonstrate a relationship between the short-term
interest income and the cost of producing rebar.?? As a result, Turkey had not
established a prima facie case.
36 WTO Anti-Dumping Agreement
?¹ United States–Softwood lumber from Canada II, Panel, para 7.237.#p#分頁標題#e#
?² United States–Softwood lumber from Canada II, AB, para 133.
?³ United States–Softwood lumber from Canada II, AB, paras 137–139.
?? Egypt–Rebar from Turkey, Panel, para 7.389.
?? Egypt–Rebar from Turkey, Panel, para 7.393.
?? Egypt–Rebar from Turkey, Panel, paras 7.422, 7.426.
1.3.2.2 SGA and Profits
One of the most complex – and arbitrary – aspects of constructed normal value
calculations concerns the selection of appropriate SGA and profits. It is recalled
that the objective of constructing a normal value is to calculate a price of the
exported product, as if it has been sold on the domestic market.??
As the product has been exported, a cost of manufacture of the exported
product will therefore exist, although the allocation of this cost, for example, to
specific PCNs, may create controversy. On the other hand, it will happen often
that the producer under investigation does not sell the like product or specific
PCNs on the domestic market or that he does not sell in the ordinary course of
trade. Indeed, these will tend to be the very reasons why normal value had to be
constructed in the first place. The question then becomes which domestic SGA
and profits to use.
Article 2.2.2 technically provides four possibilities:
• [Actual data pertaining to production and sales in the ordinary course of
trade of the like product by the exporter or producer under investigation;
or, when this is not possible:]
• the actual amounts incurred and realized by the exporter or producer in
question in respect of production and sales in the domestic market of the
country of origin of the same general category of products;
• the weighted average of the actual amounts incurred and realized by other
exporters or producers subject to investigation in respect of production and
sales of the like product in the domestic market of the country of origin;
• any other reasonable method, provided that the amount for profit so
established shall not exceed the profit normally realized by other exporters
or producers on sales of products of the same general category in the
domestic market of the country of origin.
Article 2.2.2 ADA
For the purpose of paragraph 2, the amounts for administrative, selling and general
costs and for profits shall be based on actual data pertaining to production and sales in
the ordinary course of trade of the like product by the exporter or producer under
investigation. When such amounts cannot be determined on this basis, the amounts
may be determined on the basis of:
(i) the actual amounts incurred and realized by the exporter or producer in question
in respect of production and sales in the domestic market of the country of origin
of the same general category of products;#p#分頁標題#e#
Dumping 37
?? Thailand–H-Beams from Poland, Panel, para 7.112.
(ii) the weighted average of the actual amounts incurred and realized by other
exporters or producers subject to investigation in respect of production and sales
of the like product in the domestic market of the country of origin;
(iii) any other reasonable method, provided that the amount for profit so established
shall not exceed the profit normally realized by other exporters or producers on
sales of products of the same general category in the domestic market of the
country of origin.
Under the first method, the investigating authority sticks with the like product
data of the producer/exporter concerned. This method must therefore be used
where a producer has viable domestic sales in the ordinary course of trade of at
least one exported PCN. The SGA and particularly the profit will then be based
on the SGA incurred and the profit realized on the viable domestic sales in the
ordinary course of trade of this PCN. It does not require much imagination to
understand that the profit thus calculated may be very arbitrary.
In EC–Malleable cast iron tube or pipe fittings, the AB upheld the panel
finding that SGA/profit calculations made per the Article 2.2.2 chapeau may
include SGA/profits on low volume domestic sales.?? The AB considered
relevant in this regard that Article 2.2 separately identified low volume sales
and sales not in the ordinary course of trade while the Article 2.2.2 chapeau
explicitly excluded only the latter.??
Where the first method is impossible to use, typically because a producer
does not have domestic sales in the ordinary course of trade of any exported
PCNs, Article 2.2.2 ADA offers three alternative allocation methods. It is
important to note that Article 2.2.2 does not establish a hierarchy between the
three second-best methods in sub-paragraphs (i) to (iii), thereby leaving full
discretion to the investigating authorities as to which method to use.??
The method in sub-paragraph (i) sticks with the producer under investigation,
but authorizes use of his SGA incurred and profits realized with respect to
production and sales in the domestic market of the same general category of
products. The method in sub-paragraph (ii) departs from the data pertaining to
the producer under investigation, but sticks with the like product in authorizing
the use of SGA incurred and profits realized on the like product in the domestic
market by other producers or exporters. It is noted here that use of this method
raises a major due process issue because such data are typically confidential and
38 WTO Anti-Dumping Agreement
?? EC–Malleable cast iron tube or pipe fittings from Brazil, AB, para 101.
?? EC–Malleable cast iron tube or pipe fittings from Brazil, AB, para 98.#p#分頁標題#e#
?? Compare Thailand–H-Beams from Poland, Panel, para 7.123; EC–Bed linen from India, Panel,
para 6.62; Koulen, ‘The New Anti-Dumping Code Through its Negotiating History’ in Bourgeois,
Berrod and Gippini Fournier (eds), The Uruguay Round Results (1995) 151, 208.
consequently cannot be checked by the producer for which the method is used.
Last, the sub-paragraph (iii) method authorizes any other reasonable method,
provided that the amount for profit so established shall not exceed the profit
normally realized by other exporters or producers on sales of products of the
same general category in the domestic market of the country of origin.
General and administrative costs General and administrative (G&A) costs
tend to be costs that are incurred for the company as a whole. As a result, such
costs are often allocated company-wide on the basis of a company’s turnover.
However, it may happen that companies take the position that certain general
and/or administrative costs do not relate to domestic sales of the product
concerned, as a result of which they should be excluded from the calculation.
Investigating authorities may not always agree with such exclusion.
This happened, for example, in United States–Softwood lumber II. The panel
started out by noting that general costs are costs affecting all or nearly all
products manufactured by a company while administrative costs are costs
concerning the management of the company’s affairs and considered that both
types of costs can only have a bearing on all the products manufactured by a
company, although in varying degrees. Thus, the panel found that by their
nature, G&A costs are costs that will normally affect all products produced or
sold by a company.?¹ Unless a producer/exporter can therefore demonstrate
that the product under investigation does not benefit from a particular G&A
cost item, an investigating authority is not precluded from attributing at least
a portion of that cost to the product under investigation.?²
On this basis, the panel reached the conclusion that an unbiased and
objective investigating authority could treat legal settlement costs related to
hardboard siding that was produced and sold between 1 and 18 years before the
period of investigation and not related to the production process as such of
hardboard siding, as a general expense.?³
Same general category of products Sub-paragraphs 2.2.2(i) and 2.2.2(iii)
introduce the concept of the same general category of products, but do not define
it. In many cases, this will not necessarily present a problem because the
answer can often be found in the accounting records of the producer. If, for
example, a company produces audio and video apparatus in two separate
divisions and maintains divisional accounting records, the same general#p#分頁標題#e#
category of products in an investigation involving DVD players will be the
Dumping 39
?¹ United States–Softwood lumber from Canada II, Panel, para 7.263.
?² United States–Softwood lumber from Canada II, Panel, para 7.267.
?³ United States–Softwood lumber from Canada II, Panel, para 7.267.
video apparatus division and not the audio apparatus division nor the
company-wide range of products.
In Thailand–H-Beams, the panel considered that the intention of the
sub-paragraphs is to obtain results that approximate as closely as possible the
price of the like product in the ordinary course of trade in the domestic market
of the exporting country.?? It therefore rejected Polish arguments that a
broader rather than a narrower definition is required ?? and hinted that in fact
the opposite approach might be more appropriate:
. . . the use under subparagraph (i) of a narrower rather than a broader ‘same general
category of products’ certainly is permitted. Indeed, the narrower the category,
the fewer products other than the like product will be included in the category, and this
would seem to be fully consistent with the goal of obtaining results that approximate
as closely as possible the price of the like product in the ordinary course of trade in the
domestic market of the exporting country.??
Given that nothing in the text of the ADA or anywhere else mandates the use of HS
categories in the context of Article 2.2.2 (i), we do not find that Thailand was
‘obligated’ to use the HS category proposed by Poland.??
Ordinary course of trade It is important to note that methods (i)–(iii) do not
contain the ordinary course of trade limitation, contained in the primary method.
In EC–Bed linen from India, the panel had ruled that while an authority is not
obligated to exclude sales not in the ordinary course of trade for purposes of
determining the profit rate under the subparagraphs of Article 2.2.2, neither is
such exclusion prohibited. The panel based this finding on what it saw as a general
principle in Article 2 that authorities may base their calculations of normal
value only on sales made in the ordinary course of trade.?? The panel therefore
considered the exclusion by the EC of sales not in the ordinary course of trade
from the determination of the profit amount to be used in the calculation of a
constructed normal value to be a permissible interpretation of the ADA.??
On appeal by India, however, the AB overruled the panel on the basis of a
textual interpretation of Article 2.2.2(ii). The AB first noted that Article
2.2.2(ii) refers to the weighted average of the actual amounts incurred and
realized by other exporters or producers and did not make any exceptions or
qualifications. Thus, there was no basis in Article 2.2.2(ii) for excluding some#p#分頁標題#e#
40 WTO Anti-Dumping Agreement
?? Thailand–H-Beams from Poland, Panel, para 7.112.
?? Thailand–H-Beams from Poland, Panel, para 7.111.
?? Thailand–H-Beams from Poland, Panel, para 7.113.
?? Thailand–H-Beams from Poland, Panel, para 7.116.
?? EC–Bed linen from India, Panel, para 6.85.
?? EC–Bed linen from India, Panel, para 6.87.
amounts that were actually incurred or realized from the actual amounts
incurred or realized and, as a result, an investigating authority is not allowed to
exclude sales not made in the ordinary course of trade from the calculation of
the weighted average under Article 2.2.2(ii).
The AB found support for this interpretation in the context of Article
2.2.2(ii), in particular, in the first sentence of the chapeau of Article 2.2.2. In
contrast to Article 2.2.2(ii), the first sentence of the chapeau of Article 2.2.2
refers to actual data pertaining to production and sales in the ordinary course
of trade. Thus, it was clear that sales not in the ordinary course of trade are to
be excluded when calculating amounts for SGA and profits using the method
set out in the chapeau of Article 2.2.2.
The exclusion in the chapeau led the AB to believe that, where there is no
such explicit exclusion elsewhere in the same Article of the ADA, no exclusion
should be implied. Reading into the text of Article 2.2.2(ii) a requirement provided
for in the chapeau of Article 2.2.2 was not justified either by the text or
by the context of Article 2.2.2(ii). The AB therefore reversed the panel finding
that, in calculating the amount for profits under Article 2.2.2(ii) of the ADA,
an investigating authority may exclude sales by other exporters or producers
that are not made in the ordinary course of trade.??
Other exporters or producers/weighted average Sub-paragraphs 2.2.2(ii) and
2.2.2(iii) authorize investigating authorities to use data of other exporters or producers
for, respectively, the like product or the same general category of products.
The question may therefore arise whether investigating authorities can
resort to these methods where there is only one producer or exporter with such
domestic sales, particularly in the case of option 2.2.2(ii) which not only uses
the plural form, but also refers to weighted average amounts.
This situation arose in EC–Bed linen where only the Indian producer
Bombay Dyeing had viable sales of the like product in India in the ordinary
course of trade. The panel considered that the phrase ‘other exporters or producers’
as a general matter admits an understanding where the plural form
includes the singular case and that in both common speech and legal texts, the
ordinary meaning of the plural form may include the singular case. Thus, the#p#分頁標題#e#
plural form did not necessarily preclude resort to option (ii) where there is only
one other producer or exporter of the like product.?¹ The panel also found that
the phrases ‘weighted average’ and ‘other producers and exporters’ did not
constitute two separate requirements, but rather that the concept of weighted
averaging came into play only when there is information from more than one
Dumping 41
?? EC–Bed linen from India, AB, paras 80–84. ?¹ EC–Bed linen from India, Panel, para 6.70.
other producer or exporter available and that this was the reason why it had
been put there.
On appeal, the AB rejected both points and considered the term ‘weighted
average’ the key:
. . . First of all, and obviously, an ‘average’ of amounts for SG&A and profits cannot be
calculated on the basis of data on SG&A and profits relating to only one exporter or
producer. Moreover, the textual directive to ‘weight’ the average further supports this
view because the ‘average’ which results from combining the data from different
exporters or producers must reflect the relative importance of these different exporters
or producers in the overall mean. In short, it is simply not possible to calculate the
‘weighted average’ relating to only one exporter or producer . . .
The requirement to calculate a ‘weighted average’ in Article 2.2.2(ii) is . . . the key to
interpreting that provision. It is indispensable to the calculation method set forth in
this provision, and, thus, it is indispensable to the entire provision – which deals only
with the mechanics of that calculation. We disagree with the Panel that ‘the concept of
weighted averaging is relevant only when there is information from more than one
other producer or exporter available to be considered’. We see no justification, textual
or otherwise, for concluding that amounts for SG&A and profits are to be determined
on the basis of the weighted average some of the time but not all of the time . . .
. . . the use of the phrase ‘weighted average’, combined with the use of the words
‘amounts’ and ‘exporters or producers’ in the plural in the text of Article 2.2.2(ii),
clearly anticipates the use of data from more than one exporter or producer. We conclude
that the method for calculating amounts for SG&A and profits set out in this
provision can only be used if data relating to more than one other exporter or producer
is available.?²
Amounts for profits The application of various calculation methods on a
type-by-type basis may lead to findings of high profit margins. Thus, as we have
seen above,?³ in Thailand–H-Beams, the Thai authorities used a profit margin#p#分頁標題#e#
of 36.3 per cent and in EC–Bed linen, the EC authorities a profit margin of 18.6
per cent. Use of such high profit margins will often lead to findings of dumping.
Not surprisingly, therefore, in both cases the argument was made that such high
profit margins were not reasonable. However, both panels rejected this argument
by juxtaposing the texts and the methods of sub-paragraphs 2.2.2(i) and 2.2.2(ii)
on the one hand and sub-paragraph 2.2.2(iii) on the other hand.??
Whatever the argument about the ‘reasonability’ of a particular result – a 50% profit
rate, for instance – if it is based on actual data and properly calculated, then that is the
42 WTO Anti-Dumping Agreement
?² EC–Bed linen from India, AB, paras 72–77. ?³ See section 1.3.2.2, above.
?? Thailand–H-Beams from Poland, Panel, paras 7.124–7.125; EC–Bed linen from India, Panel,
paras 6.96–6.98.
reality. An important object and purpose of Article 2.2.2 is to base the calculation of
the profit amount on actual data. Similarly, while the methods set out in paragraphs (i)
and (ii) are derivatives of the chapeau methodology, where actual data are used as
required and the calculation is correct, the results obtained themselves reflect objective
reality. Thus, the use of actual data itself ensures that subjective judgments about the
reasonability of the results do not affect the calculation of constructed normal value.
No purpose would be served by testing the results obtained under the chapeau and subparagraphs
(i) and (ii) against some arbitrary or subjective standard of reasonability.??
1.3.3 Normal Value: Country of Origin of Export?
It may happen that the country of origin and the country of export of an
exported product are not the same. The question then arises whether normal
value should be based on the domestic sales in the country of origin or in the
country of export. Article 2.5 ADA sets out the principle that in such situations
the export price shall normally be compared with the comparable price in the
country of export.
2.5 In the case where products are not imported directly from the country of origin
but are exported to the importing Member from an intermediate country, the price at
which the products are sold from the country of export to the importing Member shall
normally be compared with the comparable price in the country of export. However,
comparison may be made with the price in the country of origin, if, for example, the
products are merely trans-shipped through the country of export, or such products
are not produced in the country of export, or there is no comparable price for them in
the country of export.
This preference for country of export over country of origin is
surprising because reliance on the price in the country of origin at first sight#p#分頁標題#e#
seems more appropriate in that this is the country where the last substantial
process or transformation of the exported product took place. The country
of export (in cases where it differs from the country of origin), on the other
hand, will tend to depend mostly on logistical reasons. Typical examples
of this might be shipments from Hong Kong of Chinese origin merchandise
and shipments from Singapore of Malaysian or Indonesian origin
merchandise.
However, Article 2.5 provides that the export price may be compared with
the price in the country of origin if, for example, the exported products are
merely trans-shipped through or not produced in the country of export or if
there is no comparable price for them in the country of export. Thus, these
Dumping 43
?? EC–Bed Linen, Panel, para 6.99.
exceptions will take care of the most problematic situations. Furthermore, the
use of the words ‘normally’ in the main rule and ‘for example’ in the exceptions
indicates that even in other situations, the authorities may resort to prices in
the country of origin. Suppose, for example, that some finishing operations
take place in the country of export which are insufficient to confer origin and
that the product concerned is produced and sold by others in the country of
export. None of the three exceptions is then applicable and normal value could
therefore be established on the basis of the prices in the country of export, even
though this might not be the most appropriate basis. Arguably, however, the
authorities even then could still decide to use prices in the country of origin
because the exceptions given are merely illustrative.
It is noted that both the main rule and the exceptions only refer to the ‘price’
concept. Presumably, however, Article 2.5 also authorizes resort to constructed
normal value in cases where there are no sales in the ordinary course of trade.
Indeed, an unduly literal reading of the word ‘price’ might lead to situations
where a normal value could not be established, for example in cases where the
exported product is sold neither in the country of origin nor in the country of
export.
1.3.4 Normal Value: Non-market Economies
Article 2.7 ADA provides that Article 2 is without prejudice to the second
Supplementary Provision to paragraph 1 of Article VI in Annex I to GATT
1994. The second Supplementary Provision to Article VI:1 states that it is
recognized that, in the case of imports from a country which has a complete or
substantially complete monopoly of its trade and where all domestic prices
are fixed by the State, special difficulties may exist in determining price
comparability for the purposes of paragraph 1, and in such cases importing
contracting parties may find it necessary to take into account the possibility#p#分頁標題#e#
that a strict comparison with domestic prices in such a country may not always
be appropriate.
The second Supplementary Provision traditionally has been used by various
authorities to ignore prices and costs in ‘non-market economies’ on the
ground that they are unreliable because they are not set by market forces but
by the State. Rather, therefore, these authorities would then use prices or
costs in a market economy country as the basis for normal value in cases
involving non-market economies (the so-called surrogate or analogue
country concept).?? A further consequence would be that producers in a
44 WTO Anti-Dumping Agreement
?? Compare Czako, Human and Miranda, A Handbook on Anti-Dumping Investigations (2003) 35.
non-market economy might all be subjected to the same duty rate to avoid
circumvention.
This blunt approach has been criticized on two basic grounds. First, it has
been argued that there are very few countries which have a complete or
substantially complete monopoly of their trade and where all domestic prices
are fixed by the State; thus, the second Supplementary Provision should not be
used against countries such as China, Vietnam, Ukraine and other economies
in transition which have a vibrant and ever-increasing private sector. Second,
the use of the surrogate country concept as such has often been criticized
because it tends to lead to high dumping margins, among others, because the
surrogate countries chosen may be at a different stage of economic development
and because the only reason that producers in the surrogate countries have
typically cooperated is to ensure that their non-market economy competitors
receive a high anti-dumping duty.
In recognition of the fast changes occurring in many non-market economies,
many WTO members have loosened their treatment of non-market economy
producers in recent years. Countries that were long treated as non-market
economies, such as Russia, are now treated as market economies by many
WTO members for anti-dumping purposes. In addition, many WTO
members acknowledge the existence of ‘bubbles of capitalism’ in non-market
economies which may lead them to grant non-market economy producers
individual duty rates or even to accept their domestic prices and/or costs for the
determination of normal value.
1.4 Fair Comparison
Once export price and normal value have been determined on the basis of the
rules described in the preceding sections, they must be compared with each other.
It is recalled here that normally, as an intermediate step, these comparisons are
made on a PCN-per-PCN basis.
Article 2.4, first sentence, sets out the overarching principle or ‘generic
rule’?? that a fair comparison shall be made between the export price and the
normal value. It has been observed that the pre-eminence of the explicit fair#p#分頁標題#e#
comparison requirement in the ADA was one of the major accomplishments of
Dumping 45
?? Koulen, ‘The New Anti-Dumping Code Through its Negotiating History’ in Bourgeois, Berrod
and Gippini Fournier (eds), The Uruguay Round Results (1995) 151, 210.
the negotiations.?? The AB has held that this sentence constitutes an obligation
for the administering authorities:??
2.4 A fair comparison shall be made between the export price and the normal value.
This comparison shall be made at the same level of trade, normally at the ex-factory
level, and in respect of sales made at as nearly as possible the same time. Due allowance
shall be made in each case, on its merits, for differences which affect price comparability,
including differences in conditions and terms of sale, taxation, levels of trade, quantities,
physical characteristics, and any other differences which are also demonstrated to
affect price comparability.?? In the cases referred to in paragraph 3, allowances for
costs, including duties and taxes, incurred between importation and resale, and for
profits accruing, should also be made. If in these cases price comparability has been
affected, the authorities shall establish the normal value at a level of trade equivalent to
the level of trade of the constructed export price, or shall make due allowance as
warranted under this paragraph. The authorities shall indicate to the parties in
question what information is necessary to ensure a fair comparison and shall not
impose an unreasonable burden of proof on those parties.
1.4.1 Adjustments
Article 2.4, second sentence, concretizes the general fair comparison obligation
by providing that the comparison must be made at the same level of trade,
normally at the ex-factory level, and in respect of sales made at, as nearly as
possible, the same time.
The export price and the normal value are essentially the gross prices that are
found in the market, as evidenced by invoices and other relevant sales
documents. However, these prices may be prices to distributors, wholesalers or
end-users. In addition, various cost items may be included in the price. Thus,
to give a simple example, a delivered price will normally be higher than an
ex-warehouse price because the freight costs from the factory to the customer
are borne by the producer and therefore included in the price. Such costs
should therefore be deducted. In practice, most administering authorities will
deduct all expenses included in the price from the moment the product leaves
the factory in order to calculate the ex-factory price. Some authorities will also
deduct packing costs. This process is sometimes called netting back.
46 WTO Anti-Dumping Agreement
?? Holmer, Horlick and Stewart, ‘Enacted and Rejected Amendments to the Anti-Dumping Law:#p#分頁標題#e#
In Implementation or Contravention of the Anti-Dumping Agreement’ (1995) 29 The International
Lawyer 483–511, at 497 (comment by Horlick).
?? United States–Hot Rolled Steel from Japan, AB, para 180.
?? Fn 7 in ADA: It is understood that some of the above factors may overlap, and authorities shall
ensure that they do not duplicate adjustments that have been already made under this provision.
On the other hand, the requirement that the comparison must be made in
respect of sales made at, as nearly as possible, the same time is often ignored as
it is the practice of most authorities to calculate weighted averages during the
investigation period, particularly where the volume of transactions is high.
Thus, only in cases where the volume of export transactions is low, will administering
authorities try to locate domestic sales made on or around the same
date and use those as the basis or normal value, rather than an average of all
domestic sales during the investigation period. In relatively small importing
markets, such as New Zealand, such method may be more administratively
convenient.
Article 2.4, third sentence, then requires due allowance to be made for all
differences which affect price comparability. These include differences in
conditions and terms of sale, taxation, levels of trade, quantities and physical
characteristics.
1.4.1.1 Conditions and Terms of Sale
In United States–Stainless steel plate in coils and stainless steel sheet and strip, the
panel held that the bad debt expense resulting from the failure of a customer to
pay because of bankruptcy is not a condition or term of sale within the meaning
of Article 2.4 for which adjustment may be made. The panel considered
that the term ‘conditions and terms of sale’ refers to differences in the bundle
of contractual rights and obligations arising from a sales agreement. Failure of
a customer to pay constituted a breach of such conditions, rather than a
condition as such.?¹
In Argentina–Poultry, Argentina claimed to have rejected an adjustment for
freight costs claimed by one exporter on the ground that this exporter had
reported its freight costs on an annualized basis, without supporting evidence.
The panel considered, inter alia, that the questionnaire did not exclude reporting
freight costs on such basis and that this was therefore insufficient reason to
reject the adjustment.?²
1.4.1.2 Physical Differences
In Argentina–Ceramic floor tiles, the question arose to what extent the authorities
should proactively make an allowance for differences in physical characteristics.
The panel in that case adopted a fairly high standard by considering that at
a minimum the authorities have to evaluate identified differences in physical#p#分頁標題#e#
characteristics to see whether an adjustment is required to maintain price
Dumping 47
?¹ United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea, Panel,
para 6.75. ?² Argentina–Poultry from Brazil, Panel, para 7.236.
comparability and to ensure a fair comparison, and to adjust where necessary.
The case record indicated that ceramic tiles can be distinguished on the basis of
a number of characteristics, such as dimensions (length and width), colour,
degree of processing (polished/unpolished), and quality, and that the price of
the products differs as a function of these differences in physical characteristics.
The authority had made an allowance for three types of physical differences,
either directly (differences in size) or indirectly (by collecting data only for firstquality,
unpolished tiles), but not for other important differences. The panel
decided that an objective and unbiased evaluation of the facts of the case would
have required the authority to make additional adjustments and, as a result, the
authority acted inconsistently with Article 2.4 by failing to make adjustments
for physical differences affecting price comparability.?³
In Argentina–Poultry, the panel considered that a 9.09 per cent upward
adjustment to the normal value to take into account physical differences
between poultry sold in Brazil and poultry exported to Argentina was not justified
by the information on the record.??
In United States–Softwood lumber II, the panel, faced with a claim for an
allowance based on differences in dimension, seemed to shift the burden of
proof from the authority to the party claiming the allowance. The panel agreed
with previous panels that the authority must at least evaluate identified
differences. However, it considered that Article 2.4 does not require that an
adjustment be made automatically in all cases where a difference is found to
exist, but only where – based on the merits of the case – that difference is
demonstrated to affect price comparability. In the view of this panel, an
interpretation that an adjustment would have to be made automatically where
a difference in physical characteristics is found to exist would render the term
‘which affect price comparability’ nugatory.??
1.4.1.3 Any other Differences
Under Article 2.4, third sentence, due allowance must also be made for any
other differences which are demonstrated to affect price comparability. The
Argentina–Ceramic floor tiles panel ruled that this is an open-ended provision
which places an obligation upon the administering authorities:??
. . . Article 2.4 expressly requires that ‘allowances’ be made for ‘any other differences
which are also demonstrated to affect price comparability’. There are, therefore, no#p#分頁標題#e#
48 WTO Anti-Dumping Agreement
?³ Argentina–Ceramic floor tiles from Italy, Panel, paras 6.113–6.117.
?? Argentina–Poultry from Brazil, Panel, para 7.260.
?? United States–Softwood lumber from Canada II, Panel, para 7.165.
?? Argentina–Ceramic floor tiles from Italy, Panel, para 6.113.
differences ‘affect[ing] price comparability’ which are precluded, as such, from being
the object of an ‘allowance’.
We would also emphasize that, under Article 2.4, the obligation to ensure a ‘fair
comparison’ lies on the investigating authorities, and not the exporters. It is those
authorities which, as part of their investigation, are charged with comparing normal
value and export price and determining whether there is dumping of imports . . . ??
1.4.1.4 Constructed Export Price
This has been discussed in section 1.2.2 above.
1.4.1.5 Burden of Proof
As regards the requirement in the last sentence of Article 2.4 that the authorities
shall not impose an unreasonable burden of proof on interested parties, the
United States–Softwood lumber II panel ruled that this requirement does not
remove the burden from interested parties to constructively substantiate their
claimed adjustments:
. . . the requirement in the last sentence of Article 2.4 that the authorities ‘shall not impose
an unreasonable burden of proof’ on interested parties does not remove the burden from
interested parties to substantiate their assertions concerning claimed adjustments.??
The panel further considered that the provision does not impose on investigating
authorities any particular method for examining whether any given difference
affects price comparability.??
1.4.2 Currency Conversions
Export prices are often expressed in currencies such as US dollars or euros while
normal value is typically expressed in the local currency of the exporters. Thus,
one of the two has to be converted in order to compare the export price with
the normal value. Many authorities will convert the currency of export into the
local currency of the exporting country.
Article 2.4.1 of the ADA contains rules on currency conversions,¹?? but
does not address this specific issue, thereby leaving discretion to the authorities.
Some authorities will provide a list of average exchange rates during the
investigation period in (an annex to) the questionnaire which interested parties
are then requested to use. While this may facilitate the workload, Article 2.4.1
Dumping 49
?? United States–Hot Rolled Steel from Japan, AB, paras 177–180.
?? United States–Softwood lumber from Canada II, Panel, para 7.167.
?? United States–Softwood lumber from Canada II, Panel, para 7.167.#p#分頁標題#e#
¹?? See Czako, Human and Miranda, A Handbook on Anti-Dumping Investigations (2003) 146–148
for more detail.
does require currency conversions to be made on the basis of the exchange rate
prevailing on the date of sale. Interested parties therefore have the right to use
such exchange rates rather than the average exchange rate provided in the
questionnaire, if they so prefer:
2.4.1 When the comparison under paragraph 4 requires a conversion of currencies,
such conversion should be made using the rate of exchange on the date of sale,¹?¹ provided
that when a sale of foreign currency on forward markets is directly linked to the
export sale involved, the rate of exchange in the forward sale shall be used. Fluctuations
in exchange rates shall be ignored and in an investigation the authorities shall allow
exporters at least 60 days to have adjusted their export prices to reflect sustained movements
in exchange rates during the period of investigation.
Article 2.4.1 provides rules for two special situations:
• First, if a sale in a foreign currency on forward markets (hedging) is shown
to be directly rated to the export sale involved, then the exchange rate in
the forward sale must be used; and
• Second, exchange rates fluctuations are to be ignored and exporters must
be given 60 days in which to adjust their export prices to reflect sustained
currency movements.¹?²
1.4.2.1 Local L/C Sales
In anti-dumping cases involving Korea, authorities may encounter the
phenomenon of so-called local L/C sales. These are domestic sales of the like
product, made by letter of credit and expressed in US dollars. The like product
thus sold will be processed into another product which will eventually be
exported. The reason why this special sales channel exists is related to the
Korean duty drawback system. In United States–Stainless steel plate in coils and
stainless steel sheet and strip, the USDOC had converted the local L/C sales first
into Korean Won and then back again into US dollars. The panel ruled that the
United States thereby had acted inconsistently with Article 2.4.1 by performing
a currency conversion that was not required.¹?³
1.4.2.2 Date of Sale
Footnote 8 to Article 2.4.1 ADA provides that the date of sale would normally
be the date of contract, purchase order, order confirmation, or invoice,
50 WTO Anti-Dumping Agreement
¹?¹ Fn 8 in ADA: Normally, the date of sale would be the date of contract, purchase order, order
confirmation, or invoice, whichever establishes the material terms of sale.
¹?² Horlick and Shea, ‘The World Trade Organization Anti-Dumping Agreement’ (1995) 29
Journal of World Trade, 5, 25.
¹?³ United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea, Panel, para 7.1.#p#分頁標題#e#
whichever establishes the material terms of sale. While technically relevant
only for determining the date of sales for currency conversion purposes, the
footnote effectively provides a definition of what constitutes a sale.
This definition is sometimes also used by authorities in order to determine
whether specific sales transactions fall inside or outside the investigation
period. Suppose, for example, that the investigation period is the calendar year
2005 and that an export sale is invoiced 23 December 2005. Although this sale
would probably arrive in the importing country market only in the course of
2006, arguably it should nevertheless be reported in the questionnaire response
and used in the calculation because it was invoiced and thus sold during the
investigation period.
1.4.3 Comparison Methods: Zeroing
Once all costs included in the price and incurred from the moment the product
left the factory are deducted from both export price and normal value and
various other adjustments, such as level of trade and physical differences, have
been made, two sets of ex-factory prices remain: a set of export prices and a set
of domestic prices. At first sight, these sets of prices can be compared in two
obvious ways. The following simplified example may clarify this:
Dumping 51
Weighted average to weighted average and transaction to transaction
comparisons
Suppose four transactions of equal weight are sold in the domestic and the
export market as follows:
Domestic Export
(1) 1 June 50 50
(2) 10 June 100 100
(3) 15 June 150 150
(4) 20 June 200 200
Under the first method, one can calculate a weighted average domestic
price and a weighted average export price and then compare these weighted
averages with each other. This is called the weighted average to weighted
average method. In the example above, both the weighted average
domestic price (normal value) and the weighted average export price are
(Continued)
Conceptually, however, a third – less obvious – comparison is possible: one in
which one calculates a weighted average domestic price and then compares this
price with prices of export transactions on a transaction by transaction basis.
The following example shows the result of this third calculation method:
52 WTO Anti-Dumping Agreement
125 [(50  100  150  200)/4  125]. As a result, again no dumping
would be found.
Under the second method, one can compare domestic prices and export
prices sold on or around the same date. This is called the transaction-totransaction
method. Use of this method will lead to a finding of no dumping in
the example because the domestic prices are identical to the export prices on
the four dates that the sales take place.
Comparison of weighted average normal value with export prices on a#p#分頁標題#e#
transaction-by-transaction basis
Suppose four transactions of equal weight are sold in the domestic and the
export market as follows:
Domestic Export
(1) 1 June 50 50
(2) 10 June 100 100
(3) 15 June 150 150
(4) 20 June 200 200
WANV 125
When the weighted average normal value of 125 is compared with the export
prices on a transaction-by-transaction basis, the result becomes as follows:
Domestic Export Dumping amount
(1) 1 June 125 50  75
(2) 10 June 125 100  25
(3) 15 June 125 150  25
(4) 20 June 125 200  75
(Continued)
Under the third method, we now have found two transactions which are
dumped and two transactions which are not, or negatively, dumped. The total
dumping amount is 100 and the total non-dumped amount is also 100.
Prior to the entry into force of the Uruguay Round Anti-Dumping
Agreement, some authorities would routinely use this third method to compare
export prices with normal value and calculate dumping margins. They would
then take the position that the third and the fourth transaction are not dumped
and therefore attribute a zero dumping amount to these third and fourth transactions.
This became known as the practice of zeroing.¹?? The result of this was
that non-dumped prices could not be used to offset dumped prices.
Dumping 53
¹?? For a well-written overview of GATT and the first WTO panel reports on zeroing (EC–Bed
linen from India and United States–Stainless steel plate in coils and stainless steel sheet and strip from
Korea), see Kim, ‘Fair Price Comparison in the WTO Anti-Dumping Agreement: Recent WTO Panel
Decisions Against the “Zeroing” Method’ (2002) 36 Journal of World Trade 39–56.
Thus, as a result of this comparison, the dumping amount on the first
transaction is 75 and the dumping amount on the second transaction is 25.
On the other hand, the dumping amount on the third transaction is 25
(because the export price of 150 is 25 higher than the weighted average
normal value of 125) and the dumping amount on the fourth transaction is
75 (because the export price of 200 is 75 higher than the weighted average
normal value of 125). In the case of the third and the fourth transaction,
there is therefore a negative dumping amount.
Comparison of weighted average normal value with export prices on a
transaction-by-transaction basis: the magic of zeroing
Domestic Export Dumping amount
(1) 1 June 125 50 75
(2) 10 June 125 100 25
(3) 15 June 125 150  25  0
(4) 20 June 125 200  75  0
The total dumping amount in the example above is 100 (25 on the first
transaction and 75 on the second transaction). Assuming that the total CIF
price of the four export transactions is 525, this will give a dumping margin of#p#分頁標題#e#
100 / 525  100  19.05%.
Thus, in the example above, the first and the second obvious comparison
methods will lead to a finding of no dumping, while the use of the third – less
obvious – method, coupled to the practice of zeroing, will result in a dumping
margin of over 19 per cent. The logic often presented for using such an
asymmetrical comparison method is that it allows the authorities to focus on
targeted dumping.¹?? The analogy often invoked is that of getting a ticket for
speeding: if you drive too fast and are ticketed for speeding, you cannot claim
credit for the times that you drove within the speed limit.
It does not require much imagination to understand that use of this third
method will make it easier to find dumping in most cases, particularly when
compared with the first, weighted average to weighted average, method. In
fact, if just one export transaction is lower-priced than the weighted average
normal value, a finding of dumping will result from the use of the third
method, even if all other export transactions are higher-priced than the
weighted average normal value (if all export transactions are dumped, on the
other hand, the first and the third method will yield the same result).
Admittedly, the above example is of a symmetrical perfection that will seldom
be found in practice. In less symmetrically perfect ‘real life’ factual situations,
the result of the third method will not always be worse than the use of the
transaction-to-transaction method because the latter can lead to extremely
arbitrary results. However, it will virtually always be worse than the use of the
weighted average to weighted average method and, at best, will lead to an equal
result in the rare cases where all export transactions are dumped. This is of
course because the use of a weighted average to weighted average method
automatically offsets positive and negative dumping.
A first GATT challenge to the practice of zeroing, brought in 1992 by
Norway in United States–Fresh and chilled Atlantic salmon,¹?? failed because the
panel found that the weighted average normal value (NV) to transaction-bytransaction
export price (EP) comparison was not inherently biased. In
support, the panel noted that where all export transactions were priced below
the average normal value, the bias would not occur.¹?? While Norway had provided
an example showing the bias, the panel found that there was no evidence
that the example reflected the factual situation in the Salmon case.¹??
54 WTO Anti-Dumping Agreement
¹?? Compare United States–Fresh and chilled Atlantic salmon from Norway, para 476, ADP/87,
30 November 1992.
¹?? United States–Fresh and chilled Atlantic salmon from Norway, ADP/87, 30 November 1992.#p#分頁標題#e#
¹?? Compare United States–Fresh and chilled Atlantic salmon from Norway, para 482, ADP/87,
30 November 1992.
¹?? Compare United States–Fresh and chilled Atlantic salmon from Norway, para 484, ADP/87,
30 November 1992.
Similarly, a second challenge brought by Japan against the EC in 1994,
EC–Audio tapes in cassettes,¹?? failed because Japan had argued that the
weighted average NV to transaction-by-transaction EP comparison would
always inflate dumping margins.¹¹? The GATT panel decided to take the
transaction-to-transaction method as a benchmark and correctly concluded
that the claim was mathematically incorrect because either method might lead
to worse results than the other method, depending on the facts of the case:¹¹¹
. . . the average to average benchmark proposed by Japan also failed in some instances
accurately to reflect the results that would be obtained if the existence and extent of
dumping were determined on a transaction-to-transaction basis. In fact, the Panel was
aware of no averaging methodology that would not in some cases produce results that
differed from those obtained through the determination of the extent of dumping on
a transaction-to-transaction basis.
In light of this fact, and taking into account that Japan did not contend that the use
of averaging was inconsistent with the Agreement per se, the Panel could not conclude
that the EC’s methodology as applied in this case was unfair on the grounds of
arbitrariness.¹¹²
While Japan had submitted dumping margin calculations pertaining to the
largest exporter, which showed that the use of a weighted average NV to a
weighted average EP comparison would have led to a better result than the
method used by the EC, the panel again considered that the transaction to
transaction method was the appropriate benchmark, as Article 2 of the Tokyo
Round Code did not require the averaging method.¹¹³Observers have pointed
out that the panel’s insistence on using the transaction-to-transaction method
as the benchmark resulted in the panel effectively imposing an impossible burden
of proof on the applicant as it will be impossible for any applicant to predict
how the authorities would apply such method in a particular case.¹¹?
In a third GATT case, EC–Cotton yarn,¹¹? the panel rejected Brazil’s claim that
the EC should have made an adjustment¹¹? for distortions caused by zeroing in
hyper-inflationary economies. The panel considered that zeroing took place
Dumping 55
¹?? EC–Audio tapes in cassettes from Japan, ADP/136, 28 April 1995.
¹¹? EC–Audio tapes in cassettes from Japan, paras 125–126, ADP/136, 28 April 1995.#p#分頁標題#e#
¹¹¹ EC–Audio tapes in cassettes from Japan, para 354, ADP/136, 28 April 1995.
¹¹² EC–Audio tapes in cassettes from Japan, para 357, ADP/136, 28 April 1995.
¹¹³ EC–Audio tapes in cassettes from Japan, para 358, ADP/136, 28 April 1995.
¹¹? Horlick and Clarke, ‘Standards for Panels Reviewing Anti-Dumping Determinations under
the GATT and WTO’ in Petersman (ed), International Trade Law and the GATT/WTO Dispute
Settlement System (1997) 313, 321–322.
¹¹? EC–Imposition of anti-dumping duties on imports of cotton yarn from Brazil, ADP/137,
485–486, 4 July 1995. ¹¹? See section 1.4.1, above.
only after adjustments had already been made.¹¹?However, the battle between
users of the anti-dumping instrument and its victims did not end there. In the
then-ongoing Uruguay Round negotiations, a push to limit the use of the third
method resulted in the following compromise Article 2.4.2:
2.4.2 Subject to the provisions governing fair comparison in paragraph 4, the existence
of margins of dumping during the investigation phase shall normally be established on
the basis of a comparison of a weighted average normal value with a weighted average
of prices of all comparable export transactions or by a comparison of normal value and
export prices on a transaction-to-transaction basis. A normal value established on a
weighted average basis may be compared to prices of individual export transactions if
the authorities find a pattern of export prices which differ significantly among different
purchasers, regions or time periods, and if an explanation is provided as to why
such differences cannot be taken into account appropriately by the use of a weighted
average-to-weighted average or transaction-to-transaction comparison.
Thus, the weighted average to weighted average or transaction-to-transaction
methods became the preferred methods. The third method may be resorted to
only if the authority finds a pattern of export prices, which differs significantly
among different purchasers, regions or time periods and if the authority
provides an explanation as to why such differences cannot be taken into
account appropriately by the use of one of the preferred methods. This
compromise solution therefore allows the authorities to counter three forms of
hidden¹¹? or targeted dumping (purchaser; region; time period).¹¹?
It is emphasized that Article 2.4.2 contains the phrase ‘during the investigation
phase’. It is therefore possible that the obligation does not apply in the
context of reviews, including the duty assessment reviews that the USDOC
conducts annually. This issue is currently pending in the case United States–Laws,
regulations and methodology for calculating dumping margins (“zeroing”),#p#分頁標題#e#
brought by the EC.¹²?
56 WTO Anti-Dumping Agreement
¹¹? EC–Imposition of anti-dumping duties on imports of cotton yarn from Brazil, para 501, ADP/137,
4 July 1995.
¹¹? Horlick and Shea, ‘The World Trade Organization Anti-Dumping Agreement’ (1995) 29
Journal of World Trade, 5.
¹¹? Compare Croome, Reshaping the World Trading System (1995) 305:
Curbs were imposed on comparing isolated export selling prices with average prices in the home market,
unless evidence could be produced that the sellers were targeting particular regions, purchasers or
periods for dumping.
See also EC–Bed linen from India, AB, para 62.
¹²? United States–Laws, regulations and methodology for calculating dumping margins (“zeroing”),
WT/DS294/8, 2 November 2004, Constitution of the Panel Established at the Request of the
European Communities.
1.4.3.1 Inter-model Zeroing
Virtually as soon as the Uruguay Round Anti-Dumping Agreement entered
into force on 1 January 2005, some authorities started using a new type of
zeroing, the so-called inter-model zeroing (as opposed to the intra-model zeroing
that was used previously and which is arguably still allowed under the
exceptional third method of Article 2.4.2).
It is recalled that when comparing export price and normal value, such
comparisons are typically first made on a PCN-by-PCN basis, before the
results of these PCN-based calculations are weighted to come up with a dumping
margin for each cooperating exporter. Thus, in this intermediate step, a
positive or negative dumping amount will have been calculated for each PCN.
Using the inter-model zeroing concept, some authorities will then zero
negatively dumped PCNs, thereby not allowing such non-dumped PCNs to
offset positively dumped PCNs.
Dumping 57
¹²¹ Compare Durling, ‘Deference, But Only When Due: WTO Review of Anti-Dumping
Measures’ (2003) 6 Journal of International Economic Law 125, 136.
¹²² See for a good discussion of the panel report, Kim, ‘Fair Price Comparison in the WTO Anti-
Dumping Agreement: Recent WTO Panel Decisions Against the “Zeroing” Method’ (2002) 36
New kid on the block: inter-model zeroing
Domestic Export Dumping amount
(1) Model A 1,250 500 750
(2) Model B 1,250 1,000 250
(3) Model C 1,250 1,500 250  0
(4) Model D 1,250 2,000 750  0
The total dumping amount in the example above is 1,000 (250 on model A
and 750 on model B). The negatively dumped models C and D are zeroed.
Assuming that the total CIF price of the four models is 5,250, this will give a
dumping margin of 1,000 / 5,250  100  19.05%.
As a result of inter-model zeroing, dumping will be found as soon as one model#p#分頁標題#e#
is dumped even if all other models are not dumped. Again, therefore, intermodel
zeroing facilitates dumping findings.¹²¹
In 1999, India challenged the EC on inter-model zeroing in EC–Bed linen.
The panel agreed with India that inter-model zeroing was not allowed under
Article 2.4.2.¹²² The panel started out by noting that in light of Article 2.1 of
the ADA, the ‘margins of dumping’ established under Article 2.4.2 must relate
to the product at issue. Thus, a margin of dumping can only be established for
the product at issue, and not for individual transactions concerning that
product, or discrete models of that product.¹²³
The panel attached importance to the fact that Article 2.4.2 specifies that
the weighted average normal value shall be compared with a weighted average
of prices of all comparable export transactions. By counting as zero the results
of comparisons showing a negative margin, the EC, in effect, changed the
prices of the export transactions in those comparisons. The panel considered it
impermissible to zero such negative margins in establishing the existence of
dumping for the product under investigation, since this had the effect of
changing the results of an otherwise proper comparison.¹²?
The panel took pains to point out that it did not mean to suggest that Article
2.4.2 prohibits an investigating authority from undertaking multiple comparisons
as such, ie the intermediate step.¹²? (This view was echoed by subsequent
panels in United States–Stainless Steel, Argentina–Ceramic Tiles and United
States-Softwood lumber II.)
On appeal, the AB agreed with the findings of the panel:
By ‘zeroing’ the ‘negative dumping margins’, the European Communities . . . did
not take fully into account the entirety of the prices of some export transactions,
namely, those export transactions involving models of cotton-type bed linen
where ‘negative dumping margins’ were found. Instead, the European Communities
treated those export prices as if they were less than what they were. This, in turn,
inflated the result from the calculation of the margin of dumping . . . Furthermore,
we are also of the view that a comparison between export price and normal value
that does not take fully into account the prices of all comparable export transactions
– such as the practice of ‘zeroing’ at issue in this dispute – is not a ‘fair comparison’
between export price and normal value, as required by Article 2.4 and by Article
2.4.2.¹²?
The AB also rejected the argument that inter-model zeroing should be
allowed to offset targeted model dumping. In the view of the AB, Article 2.4.2,
second sentence, ie the exceptional method, only allowed members to#p#分頁標題#e#
58 WTO Anti-Dumping Agreement
Journal of World Trade 39–56. For criticism, see Greenwald, ‘WTO Dispute Settlement: An Exercise
in Trade Law Legislation’ (2003) 6 Journal of International Economic Law 113, 118–120. But see
Durling, ‘Deference, But Only When Due: WTO Review of Anti-Dumping Measures’ (2003) 6
Journal of International Economic Law 125, 136.
¹²³ EC–Bed linen from India, Panel, para 6.114.
¹²? EC–Bed linen from India, Panel, para 6.115.
¹²? EC–Bed linen from India, Panel, para 6.117.
¹²? EC–Bed linen from India, AB, para 55.
address three kinds of targeted dumping: dumping that targets certain
purchasers, certain regions or certain periods. Neither Article 2.4.2, second
sentence, nor any other provision of the ADA referred to dumping targeted to
certain models or types of the same product under investigation: had the drafters
of the ADA intended to authorize members to respond to such kind of dumping,
they would have done so explicitly in Article 2.4.2, second sentence.¹²?
Last, the AB summarily rejected the argument of the EC that the panel had
violated Article 17.6(ii) ADA by not considering inter-model zeroing a
permissible interpretation:¹²?
It appears clear to us from the emphatic and unqualified nature of this finding of
inconsistency that the Panel did not view the interpretation given by the European
Communities of Article 2.4.2 of the ADA as a ‘permissible interpretation’ within the
meaning of Article 17.6(ii) of the ADA. Thus, the Panel was not faced with a choice
among multiple ‘permissible’ interpretations which would have required it, under
Article 17.6(ii), to give deference to the interpretation relied upon by the European
Communities. Rather, the Panel was faced with a situation in which the interpretation
relied upon by the European Communities was, to borrow a word from the European
Communities, ‘impermissible’. We do not share the view of the European
Communities that the Panel failed to apply the standard of review set out in Article
17.6(ii) of the ADA.¹²?
In rejecting the practice of inter-model zeroing, the AB had put much emphasis
on the fact that dumping margins are established for the product under
investigation. This had led some to conclude that the AB had prohibited the
concept of multiple averaging as such. In United States–Softwood lumber II, the
AB corrected this misperception:
We agree with the participants in this dispute that multiple averaging is permitted
under Article 2.4.2 to establish the existence of margins of dumping for the product
under investigation. We disagree with those who suggest that the Appellate Body
Report in EC – Bed Linen is premised on an assumption that multiple averaging is prohibited.#p#分頁標題#e#
The issue of multiple averaging was not before the Appellate Body in EC – Bed
Linen and the reasoning of the Appellate Body in that case should therefore not be
read as prohibiting that practice. This is not to say that EC –Bed Linen is not relevant
in this appeal. Indeed, there are a number of relevant findings to which we refer to
Dumping 59
¹²? EC-Bed linen from India, AB, para 62. ¹²? EC–Bed linen from India, AB, para 65.
¹²? EC–Bed linen from India, AB, para 65. For criticism of the AB report on this point, see Tarullo,
‘Paved with Good Intentions: The Dynamic Effects of WTO Review of Anti-Dumping Action’
(2003) 2 World Trade Review 373, 377–378; Tarullo, ‘The Hidden Costs of International Dispute
Settlement: WTO Review of Domestic Anti-Dumping Decisions’ (2002) Law and Policy in
International Business 109, 132–136.
below. However, the Appellate Body did not rule on multiple averaging in that case and
therefore it is incorrect to argue, as the United States does, that ‘[t]he agreement of both
parties to this dispute and a unanimous Panel that Article 2.4.2 permits multiple
comparisons is a fundamental departure from the premise’ of the Appellate Body
Report in EC – Bed Linen.¹³?
In EC–Malleable cast iron tube or pipe fittings, the panel again rejected inter-model
zeroing and considered irrelevant the EC assertion that the zeroing had had
limited impact on the dumping margin (2.73 per cent).¹³¹
In United States–Softwood lumber II, Canada attacked inter-model zeroing
applied by the USDOC. While two of the three panellists firmly rejected the
practice, one panel member strongly disagreed in a dissenting opinion. On
appeal, however, the AB summarily confirmed its findings in EC–Bed linen and
paid no attention to the dissenting opinion:
Zeroing means, in effect, that at least in the case of some export transactions, the export
prices are treated as if they were less than what they actually are. Zeroing, therefore,
does not take into account the entirety of the prices of some export transactions,
namely, the prices of export transactions in those sub-groups in which the weighted
average normal value is less than the weighted average export price. Zeroing thus
inflates the margin of dumping for the product as a whole.¹³²
As the treatment of zeroing in WTO dispute settlement has created some
controversy, ¹³³ it is important to emphasize that thus far the AB has only ruled
on the illegality of inter-model zeroing under the weighted average method in
original investigations.¹³? Although the AB used relatively broad language to
condemn the zeroing practice in these three cases, it remains to be seen which#p#分頁標題#e#
position the AB will take with respect to:
• inter-model zeroing under the transaction-to-transaction method;
• (intra-model) zeroing under the exceptional weighted average normal
value to transaction-by-transaction export price method;¹³?
60 WTO Anti-Dumping Agreement
¹³? United States–Softwood lumber from Canada II, AB, para 81.
¹³¹ EC–Malleable cast iron tube or pipe fittings from Brazil, Panel, paras 7.216–7.219.
¹³² United States–Softwood lumber from Canada II, AB, para 98.
¹³³ Compare WorldTradeLaw.net, Dispute Settlement Commentary on United States–Final lumber
AD determination, p 13, checked on 13 April 2005.
¹³? But see Kim, ‘Fair Price Comparison in the WTO Anti-Dumping Agreement: Recent WTO
Panel Decisions Against the “Zeroing” Method’ (2002) 36 Journal of World Trade 39, 56:
If the reasoning applied by the Bed linen panel is broadly upheld, then the practice of forming subgroupsof transactions and assigning zero values to the margins for the sub-groups for which dumpingmargins were negative would also be found inconsistent with the WTO Anti-Dumping Agreement.
¹³? If the AB were to reject zeroing under this third, exceptional method, the outcome of thethird method would be identical to that of the weighted average to weighted average method.
• (inter-model) zeroing in interim and sunset reviews;¹³? and
• (inter-model) zeroing US duty assessment reviews.
1.4.3.2 Multiple Averaging Periods
In United States–Stainless steel plate in coils and stainless steel sheet and strip, theUSDOC had cut the investigation period which fell in the middle of the Asianfinancial crisis into two parts: a pre- and a post-Korean Won devaluation phase,
based on a finding that the normal value in the latter phase, expressed indollars, differed significantly from the normal value in the earlier phase of theinvestigation period and initially calculated separate dumping margins for thetwo periods.¹³? The USDOC had then calculated one average dumpingmargin for the whole period, but in that process it had applied zeroing for thesub-periods where the average export price was higher than the average normalvalue. Korea had not attacked the zeroing¹³? as such,¹³? but rather hadcontested the appropriateness of sub-dividing the investigation period.
The panel considered that this sub-division was inconsistent with therequirement of Article 2.4.2 to compare a weighted average normal value witha weighted average of all comparable export transactions.¹?? In light of the
requirement that a comparison be made between sales made at as nearly as possiblethe same time, the panel considered that as a general matter the periods onthe basis of which the weighted average normal value and the weighted averageexport price are calculated must be the same.¹?¹#p#分頁標題#e#
The panel did not preclude that there might be factual circumstances wherethe use of multiple averaging periods could be appropriate in order to insure thatcomparability is not affected by differences in the timing of sales within the
averaging periods in the home and export markets. This might be the case, for
Dumping 61
The exceptional method then would have no meaning. Compare Durling and Nicely, Understandingthe WTO Anti-Dumping Agreement: Negotiating History and Subsequent Interpretation (2002) 109:
. . . it was resolved [during the Uruguay Round negotiations] that the ability to make weightedaverage-to-transaction comparisons when export prices vary significantly would permit authorities toaddress the issue of targeted dumping – the practice meant to be avoided by zeroing.
¹³? Compare Durling and Nicely, Understanding the WTO Anti-Dumping Agreement: Negotiating
History and Subsequent Interpretation (2002) 109.
¹³? United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea, Panel,
para 6.124. ¹³? See section 1.4.3.1, above.
¹³? Compare Kim, ‘Fair Price Comparison in the WTO Anti-Dumping Agreement: Recent WTO
Panel Decisions Against the “Zeroing” Method’ (2002) 36 Journal of World Trade 39, 52–53.
¹?? United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea, Panel, para 7.3.
¹?¹ United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea, Panel,
para 6.121.
example, where changes in normal value or export price during the course ofthe investigation period are combined with differences in the relative weightsby volume within the investigation period of sales in the home market as comparedto the export market. The use of weighted averages for the entire investigation
period might then indicate the existence of a margin of dumping thatdid not reflect the situation at any given moment within the investigation
period.¹² However, such considerations had not been an element in thedecision of the USDOC.¹³
As we have seen above, the AB also ruled in the later United States–Softwoodlumber II case that there is nothing wrong with multiple averaging as such,albeit in the context of inter-model zeroing.¹
1.4.3.3 Sampling
英國牛津大學國際法牛津評注由英國dissertation網提供資料，本文涉及有關世貿組織關于國際法反傾銷協定研究（The WTO Anti-Dumping Agreement），對于研究西方國際法的英國留學生有一定指導性作用，另本站提供英國留學生法律專業dissertation代寫#p#分頁標題#e#作服務。In Argentina–Poultry, Argentina had compared the weighted average exportprice with a weighted average allegedly statistical sample of domestic salestransactions, even though the Brazilian exporters had reported all domesticsales transactions.¹ The panel considered Article 2.2.1 which contains theconditions under which domestic sales may be excluded (may be disregardedin determining normal value only if . . .) as relevant context and concluded that
Argentina had violated Article 2.4.2 by failing to compare the weighted averageexport price with a proper weighted average normal value.¹
62 WTO Anti-Dumping Agreement
¹ United States–Stainless steel plate in coils and stainless steel sheet and strip from Korea, Panel, para
6.123.
¹³ For criticism, see Tarullo, ‘The Hidden Costs of International Dispute Settlement: WTO
Review of Domestic Anti-Dumping Decisions’ (2002) 34 Law and Policy in International Business
109, 142–143. ¹ United States–Softwood lumber from Canada II, AB, para 81.
Argentina–Poultry from Brazil, Panel, para 7.268.
 Argentina–Poultry from Brazil, Panel, paras 7.272; 7.276.