WORLD TRADE ORGANIZATION


WT/DS264/RW

3 April 2006


(06-1385)

image

Original: English


UNITED STATES – FINAL DUMPING DETERMINATION ON SOFTWOOD LUMBER FROM CANADA -

RECOURSE TO ARTICLE 21.5 OF THE DSU BY CANADA


Report of the Panel

WT/DS264/RW

Page i


TABLE OF CONTENTS

Page


  1. INTRODUCTION 1

  2. FACTUAL ASPECTS 1

  3. PARTIES' REQUESTS FOR FINDINGS AND RECOMMENDATIONS 4

  4. INTERIM REVIEW 4

    1. REVIEW REQUESTED BY CANADA 4

    2. REVIEW REQUESTED BY THE UNITED STATES 5

  5. FINDINGS 5

    1. INTRODUCTION AND GENERAL ISSUES 5

    2. CANADA'S ARTICLE 2.4.2 CLAIM 7

      1. Main arguments of the parties 8

      2. Evaluation by the Panel 9

        1. The Issue 9

        2. Text 9

        3. Scope of Appellate Body Findings in US – Softwood Lumber V 10

        4. "Product As A Whole" Under The T-T Methodology 12

        5. "Margins of Dumping" Under The T-T Methodology 13

        6. Margins of Dumping Without Averaging 14

        7. Summary 15

        8. Broader Contextual Considerations 15

        (i) Article 2.4.2, second sentence 16

        1. Prospective normal value duty assessment 24

        2. Article 2.2 26

        3. Past GATT Discussion 27

        (i) Conclusion 28

    3. CANADA'S ARTICLE 2.4 CLAIM 29

      1. Main Arguments of the Parties 29

      2. Evaluation by the Panel 30

  6. CONCLUSIONS AND RECOMMENDATION 32

WT/DS264/RW

Page ii


LIST OF ANNEXES

ANNEX A


FIRST SUBMISSION BY THE PARTIES


Contents

Page

A-1 First Written Submission of Canada

A-2

A-2 Executive Summary of First Written Submission of the United States

A-10


ANNEX B


THIRD PARTIES' SUBMISSIONS


Contents

Page

B-1 Executive Summary Third Party Submission of China

B-2

B-2 Third Party Written Submission by the European Community Executive Summary

B-6

B-3 Executive Summary of the Third Party Submission of Japan

B-13

B-4 Executive Summary of New Zealand's third Party Submission

B-18


ANNEX C


SECOND SUBMISSION BY THE PARTIES


Contents

Page

C-1 Executive Summary of the Second Written Submission of Canada

C-2

C-2 Executive Summary of Rebuttal Submission of the United States

C-8


ANNEX D


ORAL STATEMENTS, 1ST/2ND MEETINGS


Contents

Page

D-1 Oral Statement of Canada

D-2

D-2 Opening Statement of the United States at the Substantive Meeting of the Panel

D-8

D-3 Oral Statement of the People's Republic of China

D-16

D-4 Third Party Oral Submission by the European Communities

D-18

D-5 Oral Statement by India

D-26

D-6 Oral Statement of Japan at the Third Party Session

D-28

D-7 New Zealand's Oral Statement

D-36

D-8 Oral Statement of Thailand

D-39

D-9 Closing Statement of Canada

D-42

D-10 Closing Statement of the United States at the Substantive Meeting

D-45

WT/DS264/RW

Page iii


ANNEX E


QUESTIONS AND ANSWERS


Contents

Page

E-1

Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel

E-2

E-2

Answers of the United States to the Panel's Questions of 18 November

E-19

E-3

Answers to Questions for parties and Third Parties of the People's Republic of China

E-41

E-4

European Communities Responses to the Panel's Questions

E-43

E-5

Replies of Japan to the Questions from the Panel

E-63

E-6

New Zealand's Responses to Questions by the Panel to the Third Parties

E-78

E-7

Thailand's Responses to the Panel's Questions for Parties and Third Parties

E-82

E-8

Canada's Comments on the United States' Answers to the Panel's Questions

E-90

E-9

Comments of the United States on Canada's and the third Parties Responses to the Panel's Questions

E-95


ANNEX F


REQUEST FOR THE ESTABLISHMENT OF A PANEL – DOCUMENT WT/DS264/16


Contents

Page

Request for the Establishment of a Panel – Document WT/DS264/16

F-2

WT/DS264/RW

Page 1


  1. INTRODUCTION


      1. On 19 May 2005, Canada requested the establishment of a panel pursuant to Article 21.5 of the Understanding on Rules and Procedures Governing the Settlement of Disputes (hereinafter "DSU") concerning the United States' alleged failure to implement the recommendations and rulings of the Dispute Settlement Body (hereinafter "DSB") in the dispute "United States – Final Dumping Determination on Softwood Lumber from Canada".1


      2. At its meeting on 1 June 2005, the DSB referred this dispute to the original panel, in accordance with Article 21.5 of the DSU, to examine the matter referred to the DSB by Canada in document WT/DS264/16. At that meeting, the parties to the dispute also agreed that the Panel should have standard terms of reference. The terms of reference are, therefore, the following:


        "To examine, in the light of the relevant provisions of the covered agreements cited by Canada in document WT/DS264/16, the matter referred to the DSB by Canada in that document and to make such findings as will assist the DSB in making the recommendations or in giving the rulings provided for in those agreements".2


      3. On 3 June 2005, the Panel was composed as follows: Chairman: Mr. Harsha V. Singh

        Members: Mr. Gerhard Hannes Welge

        Mr. Adrián Makuc.3


      4. China, the European Communities, India, Japan, New Zealand and Thailand reserved their third-party rights.


      5. On 3 August 2005, following his appointment as a Deputy Director-General of the WTO Secretariat, Mr. Harsha Singh resigned from his position as Chairman of the Panel. Subsequently, on

        18 August 2005 Canada requested the Director-General to appoint a replacement Chairperson, pursuant to paragraph 7 of Article 8 of the DSU.


      6. On 26 August 2005, the Director-General appointed a new Chairman of the Panel. Accordingly, the composition of the Panel is as follows:


        Chairman: Dr. Toufiq Ali


        Members: Mr. Gerhard Hannes Welge

        Mr. Adrián Makuc4


      7. The Panel met with the parties on 15-17 November 2005. It met with the third parties on 16 November 2005. The Panel issued its interim report to the parties on 14 February 2006.


  2. FACTUAL ASPECTS


      1. This dispute concerns the implementation by the United States of part of the rulings and recommendations of the DSB in respect of United States – Final Dumping Determination on Softwood Lumber from Canada (DS264). The relevant part of the rulings and recommendations of



        image

        1 WT/DS264/16, 20 May 2005.

        2 WT/DS264/20/Rev.2, 17 June 2005.

        3 WT/DS264/20/Rev.2, 17 June 2005.

        4 WT/DS264/23, 29 August 2005.

        WT/DS264/RW

        Page 2


        the DSB concerned the finding that the use of "zeroing" by the US Department of Commerce (DOC) in the underlying investigation was inconsistent with Article 2.4.2 of the AD Agreement in the context of a comparison of "a weighted average normal value with a weighted average of all comparable export transactions".


      2. In the original anti-dumping investigation underlying this dispute, DOC divided the product under investigation into groups of identical, or broadly similar, product types. After making certain adjustments within each product type, DOC calculated a weighted average normal value and export price for each product type, and then compared the weighted averages for each product type. This process resulted in multiple results, one for each product type. In some instances this comparison showed that the weighted average export price for a specific product type was less than the weighted average normal value, i.e., an amount of dumping, while in other instances, the comparison showed that the weighted average export price was more than the weighted average normal value, i.e., no dumping. These results were then aggregated to produce one single margin of dumping for the product under investigation for each investigated exporter. In the aggregation process, a value of "zero" was attributed as the amount of dumping for those product type comparisons where the weighted average export price was more than the weighted average normal value. DOC then aggregated the positive amounts of dumping from the individual product type comparisons, that is, those instances where the weighted average export price was less than the weighted normal value, and divided the result by the total value of exports, to arrive at a weighted average margin of dumping. This process of attributing a "zero" value as the amount of dumping for individual product type comparisons where the weighted average export price is more than the weighted average normal value for the same product type is the process of "zeroing" which was at issue before the original panel and Appellate Body in this dispute.


      3. The issue before the original panel was whether zeroing is allowed when an investigating authority is calculating an overall margin of dumping under the weighted average-to-weighted average methodology set forth in Article 2.4.2. The original panel observed that "It is clear that Article 2.4.2 requires that all comparable export transactions have to be taken into account when the weighted average normal value is compared to the weighted average of prices of all comparable export transactions. ... Through the use of zeroing, it is clear to us that the entirety of the prices of some export transactions, i.e., those export transactions where the weighted-average-export-price is greater than the weighted-average-normal-value, in the second stage of the process, are not taken into account."5 Consequently, the original panel concluded (subject to a dissenting opinion6) that:


        "...the United States has violated Article 2.4.2 of the AD Agreement by not taking into account all comparable export transactions when DOC calculated the overall margin of dumping as Article 2.4.2 requires that the existence of margins of dumping has to be established for softwood lumber on the basis of a comparison of the weighted- average-normal-value with the weighted average of prices of all comparable export transactions, that is, for all transactions involving all types of the product under investigation".7


      4. The original panel concluded that the United States "acted inconsistently with Article 2.4.2 of the AD Agreement in determining the existence of margins of dumping on the basis of a methodology incorporating the practice of 'zeroing'."8



        image

        5 Panel Report, United States – Final Dumping Determination on Softwood Lumber from Canada ("US

        – Softwood Lumber V"), WT/DS264/R, adopted 31 August 2004, as modified by Appellate Body Report, WT/DS264/AB/R, paras. 7.215-216.

        6 Panel Report, US – Softwood Lumber V, supra note 5, paras. 9.1-9-24.

        7 Panel Report, US – Softwood Lumber V, supra note 5, para. 7.224.

        8 Panel Report, US – Softwood Lumber V, supra note 5, para. 8.1(a).

        WT/DS264/RW

        Page 3


      5. The United States appealed that conclusion by the original panel. The Appellate Body upheld the original panel's finding that the United States acted inconsistently with Article 2.4.2 of the AD Agreement in determining the existence of margins of dumping on the basis of a methodology incorporating the practice of zeroing.9


      6. In reaching its conclusions, the Appellate Body emphasized that it was addressing the issue specifically in the context of the weighted average-to-weighted average comparison methodology. The Appellate Body concluded that dumping, and "margins of dumping", can be found only for the product under investigation as a whole, and cannot be found to exist for a product type, model, or category of that product, and further stated that the weighted average normal value/weighted average export price comparisons for the product types in the DOC analysis did not result in the calculation of "margins of dumping", but only in intermediate calculations on the way to the calculation of a margin of dumping for the product (as defined by DOC), lumber, as a whole. In looking at the text of Article 2.4.2, the Appellate Body relied on the presence of the phrase "all comparable export transactions", and stated, similarly to the original panel: "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." (footnote omitted) 10 The Appellate Body specifically stated that the issue of whether zeroing is permitted under the transaction- to-transaction methodology or the average-to-transaction methodology was not before it, and it refused to consider the United States' arguments that this issue should be considered as context in assessing the permissibility of zeroing under the average-to-average methodology.


      7. The Appellate Body's report was adopted 31 August 2004.


      8. Under US law (commonly referred to as "Section 129"), if a WTO Panel or Appellate Body report finds that a determination by the DOC is not consistent with US obligations, then, following consultations with and upon receipt of a written request from the USTR, the DOC "shall ... issue a determination in connection with the particular proceeding that would render the [DOC's] action ... not inconsistent with the findings of the panel or the Appellate Body".11 In this dispute, the USTR made such a request to the DOC on 5 November 2004. The DOC issued its "Section 129" determination within the statutory deadline set out in US law, on 2 May 2005. In that determination, the DOC calculated new rates for the exporters subject to the anti-dumping duty order, based on a comparison of normal value and export prices on a transaction-to-transaction basis. Specifically, the DOC matched individual sales of Canadian lumber in the United States (export transactions) with individual sales of Canadian lumber in Canada (normal value transactions), using criteria developed in the original investigation for matching comparable transactions, and then compared the price of each export transaction for which it had data with the price of a comparable normal value transaction. The resulting comparisons yielded a positive amount where normal value exceeded export price, or a negative amount where normal value was less than export price. DOC then calculated a final margin of dumping for individually examined exporters by adding together the positive amounts, and dividing by the value of all export transactions for that exporter individually examined. DOC did not take into account the negative amounts.


      9. Canada claims that the methodology adopted by the DOC in the Section 129 determination is inconsistent with Articles 2.4.2 and 2.4 of the AD Agreement. The United States denies Canada's claims.


    image

    9 Appellate Body Report, United States – Final Dumping Determination on Softwood Lumber from Canada ("US – Softwood Lumber V "), WT/DS264/AB/R, adopted 31 August 2004, para. 117.

    10 Appellate Body Report, US – Softwood Lumber V , supra note 9, para. 101.

    11 19 U.S.C. §3538(b)(2).

    WT/DS264/RW

    Page 4


  3. PARTIES' REQUESTS FOR FINDINGS AND RECOMMENDATIONS


      1. Canada asks the Panel to find that the United States failed to implement the DSB’s recommendations and rulings and rule that it has acted inconsistently with Articles 2.4 and 2.4.2 of the Anti-Dumping Agreement by continuing to determine dumping on the basis of a methodology that incorporates the practice of zeroing.


      2. Canada also requests that the Panel recommend pursuant to Article 19.1 of the DSU that the United States bring its measures into conformity with its obligations under Articles 2.4 and 2.4.2 of the Anti-Dumping Agreement by recalculating dumping margins for all investigated exporters and the “all others rate” on the basis of a methodology that does not incorporate the practice of zeroing and that it return all anti-dumping cash deposits collected as a result of its failure to eliminate the practice of zeroing.


      3. The United States requests that the Panel reject Canada’s claims in their entirety and find that the United States properly implemented the recommendations and rulings of the DSB in this dispute.


      4. With respect to Canada's request that the Panel recommend that the United States bring its measures into conformity in a particular manner, the United States notes its understanding that Canada is asking for a “suggestion” under Article 19.1 of the DSU. The United States asserts that there should be no need for such a suggestion, as the United States has come into compliance with its WTO obligations. However, in the event the Panel were to accept Canada’s arguments, the United States requests that the Panel decline Canada’s request as inappropriate, asserting that it goes beyond anything relevant to implementing a recommendation and seeks to impose an obligation – to return anti-dumping cash deposits – nowhere called for under the WTO agreements.


      5. The arguments of the parties and third parties are set out in their written submissions and oral statements to the Panel. Executive summaries of those submissions and statements are appended to this report.


  4. INTERIM REVIEW


      1. On 31 January 2006, we submitted our interim report to the parties. On 20 February 2006, Canada and the United States submitted written requests for review of precise aspects of the interim report. On 27 February 2006, Canada and the United States submitted written comments on each other's request for interim review.


        1. REVIEW REQUESTED BY CANADA


      2. Canada requested review of paragraphs 5.11, 5.14, 5.19, 5.37, 5.39 (footnote 54), 5.70 (footnote 82), and 6.80 of the interim report. Canada asked the Panel to correct alleged mischaracterizations of Canada's position in these paragraphs.


      3. In the absence of any objections from the United States, we made the changes requested by Canada to paragraphs 5.11, 5.19, 5.37, 5.39 (footnote 54), and 6.80 of the interim report (mis- numbered in the interim report, paragraph 6.2 of the final report).


      4. In respect of paragraph 5.14, in the absence of any objections from the United States, we made changes to the text to reflect Canada's argument, based on the comments made by Canada.


      5. In respect of paragraph 5.70, footnote 82, Canada asked the Panel to insert a reference to a letter it had submitted to the Appellate Body on 28 September 2004 regarding Canada's position in the US – Softwood Lumber V Appellate Body proceedings. Canada asked the Panel to insert specific language regarding this matter. The United States objected to the language proposed by Canada,

        WT/DS264/RW

        Page 5


        alleging that it referred in an unbalanced way to Canada's letter. The United States proposed alternative language to be used in the event that the Panel decided to include a reference to Canada's letter in footnote 82.


      6. In response to Canada's request, we included a reference to Canada's 28 September 2004 in footnote 82 (footnote 84 in the final report). In doing so, we relied in part on the language proposed by the United States.


        1. REVIEW REQUESTED BY THE UNITED STATES


      7. The United States requested review of paragraphs 5.28, 5.31 and 5.73 of the Panel's interim report. The United States also referred the Panel to a number of typographical errors in the interim report.


      8. In the absence of any objections from Canada, we made the changes requested by the United States to paragraphs 5.31 and 5.73 of the interim report.


      9. In respect of paragraph 5.28, the United States requested the deletion of the phrase "even though it does not reflect the full results of all comparisons" from the final sentence of that paragraph. The United States asserted that, although the phrase is not inconsistent with the Panel's findings, it might be mis-read as being inconsistent. In its comments on the US request for interim review, Canada asserted that the phrase accurately reflects the Panel's finding that an investigating authority may treat non-dumped or negative transaction-specific comparison results as a zero value (i.e., less than they actually were) in calculating a margin of dumping.


      10. We see no need to make the change requested by the United States. In the context of paragraph 5.28, the meaning of the phrase "even though it does not reflect the full results of all comparisons" should be clear. However, we have amended the text of paragraph 5.28 in order to ensure that it clearly reflects our view that, when establishing the amount of dumping for the purpose of calculating a margin of dumping under the T-T comparison methodology, an investigating authority need not include in its calculations the results of comparisons where export price exceeds normal value.


      11. We also corrected the typographical errors identified by the United States, as well as others we identified.


  5. FINDINGS


    1. INTRODUCTION AND GENERAL ISSUES


        1. The claims put forward by Canada in this case challenge the DOC's Section 129 determination, and specifically one aspect of its methodology in calculating dumping margins for individual exporters examined. Canada has not challenged any of the procedural aspects of the Section 129 process, including the DOC's methodology in matching transactions for purposes of the transaction-to-transaction comparisons in the dumping margin calculation.


        2. The role of a Panel in an Article 21.5 proceeding is to evaluate the challenged measure to determine its consistency with the defending Member's obligations under the relevant WTO Agreements. Thus, the Panel is not limited by the original analysis and decision – rather, it is to consider, with a fresh eye, the new determination before it, and evaluate it in light of the claims and arguments of the parties in the Article 21.5 proceeding.


        3. In this case, there is no dispute as to the measure at issue, or as to the propriety of the claims raised by Canada. The principal task for us is to assess, applying the familiar concepts regarding

          WT/DS264/RW

          Page 6


          standard of review and burden of proof, whether the methodology used by DOC in its dumping margin calculation in the Section 129 determination is consistent with the asserted obligations in Articles 2.4 and 2.4.2 of the AD Agreement.


        4. The concepts of standard of review and burden of proof applicable in this dispute are the same as those applied in the original Panel's report. As the Panel noted in that report, Article 11 of the DSU sets forth the appropriate standard of review, in general, for panels for all covered agreements. Article 11 imposes upon panels a comprehensive obligation to make an "objective assessment of the matter", an obligation which embraces all aspects of a panel's examination of the "matter", both factual and legal.12


        5. Furthermore, Article 17.6(i) of the AD Agreement sets forth the special standard of review applicable to anti-dumping disputes. It provides with regard to factual issues that:


          "in its assessment of the facts of the matter, the panel shall determine whether the authorities' establishment of the facts was proper and whether their evaluation of those facts was unbiased and objective. If the establishment of the facts was proper and the evaluation was unbiased and objective, even though the panel might have reached a different conclusion, the evaluation shall not be overturned". (emphasis added)


          In this case, there is no dispute between the parties as to the establishment of the facts relevant to the claims before us.


        6. Article 3.2 of the DSU notes that the dispute settlement system serves, inter alia, to "clarify the existing provisions of [covered] agreements in accordance with customary rules of interpretation of public international law". Furthermore, pursuant to Article 3.2 of the DSU, it is clear that a panel's decision "must not add to or diminish rights and obligations provided in the WTO Agreement".13 With respect to questions of the interpretation of the AD Agreement, Article 17.6(ii) provides:


          "the panel shall interpret the relevant provisions of the Agreement in accordance with customary rules of interpretation of public international law. Where the panel finds that a relevant provision of the Agreement admits of more than one permissible interpretation, the panel shall find the authorities' measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations". (emphasis added)


          Thus, since Article 17.6(ii) requires us to apply the customary rules of interpretation of treaties in interpreting the AD Agreement, our task is in this respect no different from the task of all panels. Article 31.1 of the Vienna Convention,14 which is generally accepted as reflecting such customary rules, provides that:


          "[a] treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose".


          It is thus clear that our interpretation of the relevant provisions in this dispute must be based, first and foremost, on the text of the treaty, while context and object and purpose may also play a role.15 It is


          image

          12 Panel Report, US – Softwood Lumber V, supra note 5, para. 7.6.

          13 Appellate Body Report, India – Patent Protection for Pharmaceutical and Agricultural Chemical Products ("India – Patents (US)"), WT/DS50/AB/R, adopted 16 January 1998, DSR 1998:I, 9, para. 46.

          14 Done at Vienna, 23 May 1969, 1155 U.N.T.S. 331; 8 International Legal Materials 679.

          15 Appellate Body Report, Japan – Taxes on Alcoholic Beverages ("Japan – Alcoholic Beverages II "), WT/DS8/AB/R, WT/DS10/AB/R, WT/DS11/AB/R, adopted 1 November 1996, DSR 1996:I, 97, p. 11. The

          WT/DS264/RW

          Page 7


          also well-established that these principles of interpretation "neither require nor condone the imputation into a treaty of words that are not there or the importation into a treaty of concepts that were not intended".16


        7. What Article 17.6(ii) of the AD Agreement adds is an explicit recognition that the provisions of the AD Agreement may admit of more than one permissible interpretation, and an instruction that, if the process of treaty interpretation leads us to the conclusion that the interpretation of a provision in question put forward by the defending party is permissible, we shall find the measure in conformity with the AD Agreement if it is based on that permissible interpretation.


        8. Finally, we recall that, in WTO dispute settlement proceedings, the burden of proof rests with the party that asserts the affirmative of a particular claim or defence.17 Canada as the complaining party must therefore make a prima facie case of violation of the relevant provisions of the AD Agreement, which the United States as respondent must refute. We also note, however, that it is generally for each party asserting a fact, whether complainant or respondent, to provide proof thereof.18 We also recall that a prima facie case is one which, in the absence of effective refutation by the other party, requires a panel, as a matter of law, to rule in favour of the party presenting the prima facie case. The role of the Panel is not to make the case for either party, but it may pose questions to the parties "in order to clarify and distil the legal arguments".19


    2. CANADA'S ARTICLE 2.4.2 CLAIM


        1. Canada claims that the United States violated Article 2.4.2 of the AD Agreement because DOC engaged in zeroing when applying the transaction-to-transaction comparison methodology in the Section 129 Determination.


        2. Article 2.4.2 of the AD Agreement provides:


          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.


          image

          Appellate Body has recently emphasized the importance of the "ordinary meaning" of the terms used in the treaty text in, for example, Appellate Body Report, United States – Continued Dumping and Subsidy Offset Act of 2000 ("US – Offset Act (Byrd Amendment )"), WT/DS217/AB/R, WT/DS234/AB/R, adopted 27 January 2003, DSR 2003:I, 375.

          16 Appellate Body Report, India – Patents (US), supra note 13, para. 45.

          17 Appellate Body Report, United States – Measure Affecting Imports of Woven Wool Shirts and

          Blouses from India ("US – Wool Shirts and Blouses"), WT/DS33/AB/R and Corr.1, adopted 23 May 1997, DSR 1997:I, 323, p. 14, et seq.

          18 See Appellate Body Report, US – Wool Shirts and Blouses, supra note 17.

          19 Appellate Body Report, Thailand – Anti-Dumping Duties on Angles, Shapes and Sections of Iron or Non-Alloy Steel and H-Beams from Poland ("Thailand – H-Beams"), WT/DS122/AB/R, adopted 5 April 2001,

          DSR 2001:VII, 2701, para. 136.

          WT/DS264/RW

          Page 8


          1. Main arguments of the parties


        3. Canada submits that the Appellate Body has already found that zeroing is prohibited in the context of the weighted average-to-weighted average ("W-W") comparison methodology. Canada asserts that the Appellate Body's reasoning should also apply in respect of zeroing in the context of the transaction-to-transaction ("T-T") comparison methodology at issue in these proceedings.


        4. Canada asserts that the Appellate Body ruled, in the appeal of the original Panel's decision in this case, that, when establishing "margins of dumping" "on the basis of a comparison of a weighted average normal value with a weighted average of prices of all comparable export transactions", margins of dumping can be found only for the product under investigation as a whole, and cannot be found to exist for a product type, model, or category of that product. In this regard, Canada refers to the Appellate Body's finding that:


          … the results of the multiple comparisons at the sub-group level are … not “margins of dumping” within the meaning of Article 2.4.2. Rather, those results reflect only intermediate calculations made by an investigating authority in the context of establishing margins of dumping for the product under investigation. Thus, it is only on the basis of aggregating all these “intermediate values” that an investigating authority can establish margins of dumping for the product under investigation as a whole.20


        5. Canada also relies on the Appellate Body's statement that it:


          ... fail[ed] to see how an investigating authority could properly establish margins of dumping for the product under investigation as a whole without aggregating all of the “results” of the multiple comparisons for all product types. There is no textual basis under Article 2.4.2 that would justify taking into account the “results” of only some multiple comparisons in the process of calculating margins of dumping, while disregarding other “results”. If an investigating authority has chosen to undertake multiple comparisons, the investigating authority necessarily has to take into account the results of all those comparisons in order to establish margins of dumping for the product as a whole under Article 2.4.2.21


        6. Canada notes that this approach led the Appellate Body to find that zeroing is prohibited in the context of the W-W comparison methodology because the margin of dumping was not established for the product as a whole, in the sense that zeroing meant that non-dumped comparisons were treated as if the result were zero rather than the actual negative result, such that not all the results for all the multiple comparisons were taken into account. Canada submits that the same approach should apply in the context of the T-T comparison methodology. In particular, Canada asserts that, as with the W- W methodology, the T-T methodology involves multiple comparisons, in the sense that every transaction-specific comparison made by DOC represents an intermediate calculation in the context of establishing a margin of dumping for the product under investigation. According to Canada, therefore, DOC must aggregate all (i.e., without zeroing any) of these transaction-specific “intermediate values” in order to arrive at a single margin of dumping for the product as a whole. In other words, Canada claims that zeroing is prohibited under the T-T comparison methodology for the same reasons that it is prohibited under the W-W comparison methodology.22



          image

          20 Appellate Body Report, US – Softwood Lumber V , para. 97 (emphasis in original).

          21 Appellate Body Report, US – Softwood Lumber V , para. 98 (emphasis in original).

          22 Canada's arguments are broadly supported by China, the European Communities, India, Japan and Thailand, as third parties.

          WT/DS264/RW

          Page 9


        7. According to the United States, there is no textual basis for any obligation to offset the results of comparisons in which export prices are less than normal value with the results of comparisons in which export prices are greater than normal value when aggregating the results of multiple transaction-to-transaction comparisons. With regard to the Appellate Body Report, the United States asserts that the Appellate Body's condemnation of zeroing under the W-W comparison methodology was based on an integrated interpretation of the phrases "margins of dumping" and "all comparable export transactions". The United States submits that the latter phrase is absent in that part of the first sentence of Article 2.4.2 providing for the T-T comparison methodology. In addition, the United States asserts that, rather than being treated as "intermediate values", the results of T – T comparisons between normal value and export price may themselves be treated as "margins of dumping". The United States asserts that the term "margins of dumping" should be interpreted in the context of the particular comparison methodology being applied by the investigating authority. The United States argues that although "margins of dumping" must be determined for all comparable export transactions, and therefore the product as a whole, in the context of the W – W methodology, transaction-specific "margins of dumping" may be determined in the context of the T – T methodology. According to the US, the prohibition of zeroing under the W-W comparison methodology therefore does not apply in the context of the T-T comparison methodology.23


          1. Evaluation by the Panel


            1. The Issue


        8. In its Section 129 Determination, the DOC calculated a single margin of dumping for softwood lumber for each respondent foreign producer or exporter. It calculated that margin of dumping by determining the total amount of dumping on the basis of individual comparisons of export price and normal value for each export transaction, and then expressing that total amount as a proportion of the total value of all export sales, including those sales for which export price exceeded normal value. In order to establish the amount of dumping, the DOC summed up the amounts by which, on individual transactions, export price was less than the normal value. The DOC did not include in that summing up the amounts by which, in individual transactions, export price exceeded the normal value. In other words, the DOC did not offset the amounts attributable to non-dumped transactions against the amounts attributable to dumped transactions. The issue presented by Canada's Article 2.4.2 claim is whether it was permissible for the DOC to not make such offsets when calculating the margin of dumping for each producer/exporter.24 In other words, we must decide whether the DOC was permitted to sum only the amounts derived from T-T comparisons showing dumping, or whether it was required to also include the amounts derived from comparisons involving non-dumped transactions in that aggregation.


          1. Text


        9. The starting point for our analysis of this issue is the relevant text of the first sentence of Article 2.4.2 of the AD Agreement, which provides, inter alia, that "[s]ubject to the provisions governing fair comparison in paragraph 4, the existence of margins of dumping" may be established


          image

          23 In effect, the US argues that the investigating authority may therefore concentrate on the positive (transaction-specific) margins of dumping, without needing to make an offsets for non-dumped transactions. New Zealand, as third party, agrees that zeroing is permitted in the context of the T-T comparison methodology, albeit for different reasons than those presented by the United States.

          24 Canada asserts that the onus is on the United States to demonstrate that Article 2.4.2 allows an

          investigating authority to disregard the results of (non-dumped) intermediate values. See, e.g., Canada's Comments on the United States' Answers to the Panel's Questions at page 4. However, as complaining Member, the onus is on Canada to establish a prima facie case of violation, which entails demonstrating that Article 2.4.2 requires an investigating authority using the T-T comparison methodology to treat each transaction-specific comparison as an intermediate value, and to sum up all such intermediate values in calculating a single margin of dumping.

          WT/DS264/RW

          Page 10


          "by a comparison of normal value and export prices on a transaction-to-transaction basis". This is all that the relevant text of the first sentence of Article 2.4.2 says. It contains no definition of the term "margins of dumping", nor any additional instruction, or even guidance, on practical issues to be addressed by investigating authorities using the transaction-to-transaction comparison methodology. For example, will the margin of dumping be expressed as an absolute amount, or will it be expressed as a proportion of the value of export sales? Which export transactions will be compared with which transactions in the exporting Member? Will the results of the transaction-to-transaction comparisons be aggregated and, if so, how? We stress that such issues are not explicitly addressed by the relevant text of the first sentence of Article 2.4.2.


        10. Thus, if an investigating authority chooses to express the margin of dumping in proportion to the value of export sales, and to aggregate the results of the relevant transaction-to-transaction comparisons, the text contains no explicit instruction that such aggregation should include offsets for non-dumped amounts. In particular, the relevant part of the first sentence of Article 2.4.2 does not explicitly prohibit zeroing. Nor does Canada argue that it does. Instead, Canada argues that the prohibition of zeroing stems from the proposition that the phrase "margins of dumping" (in the relevant part of the first sentence of Article 2.4.2) refers to the margin of dumping "for the product as a whole", fully reflecting the results of all transaction-to-transaction comparisons. Canada relies in this regard on the findings of the Appellate Body in US - Softwood Lumber V.


          1. Scope of Appellate Body Findings in US – Softwood Lumber V


        11. US - Softwood Lumber V concerned the DOC's treatment of the results of model-to-model, or sub-group, comparisons when applying the weighted-average-to-weighted-average comparison methodology, which is also provided for in the first sentence of Article 2.4.2. The Appellate Body found that "'margins of dumping' can be found only for the product under investigation as a whole, and cannot be found to exist for a product type, model, or category of that product".25 In the present case, Canada asserts that the Appellate Body's reasoning should have been applied by the DOC to establish a single margin of dumping (per exporter/producer) for "the product as a whole" by aggregating all the results of the transaction-to-transaction comparisons (concerning that exporter/producer), and specifically by taking into account the full amount by which, in some of those comparisons, export price exceeded normal value against the amounts by which, in other comparisons, export price was less than normal value, i.e., showed dumping. In other words, Canada asserts that the DOC was required to fully reflect the value of all import transactions included in the transaction-to-transaction comparisons, rather than focusing on only those import transactions where export price was less than normal value.


        12. We are not persuaded by Canada's argument, however. The Appellate Body's ratio decidendi were necessarily limited to the legal issues before it, and those issues concerned the application of the W-W comparison methodology.26 The Appellate Body did not make findings regarding the T-T comparison methodology which is at issue in these proceedings, and its decision does not contain any legal analysis of the permissibility of zeroing under that methodology. Indeed, the Appellate Body explicitly declined to apply its findings regarding zeroing under the W-W comparison methodology to zeroing under the T-T comparison methodology. In particular, the Appellate Body "fail[ed] to see


          image

          25 Appellate Body Report, US – Softwood Lumber V, supra note 9, para. 96.

          26 Canada also relies on the Appellate Body's finding in EC - Bed Linen that "[w]hatever the method used to calculate the margins of dumping, ... these margins must be, and can only be, established for the product

          under investigation as a whole" (Appellate Body Report, European Communities – Anti-Dumping Duties on Imports of Cotton-Type Bed Linen from India ("EC – Bed Linen"), WT/DS141/AB/R, adopted 12 March 2001, DSR 2001:V, 2049, para. 53) (see para. 12 of Canada's second written submission). Since the Appellate Body in US - Softwood Lumber V was simply reiterating its earlier finding in EC – Bed Linen to support its position in US - Softwood Lumber V, we see no need to address the findings of the Appellate Body in EC – Bed Linen in detail. Instead, we shall focus - as the parties have done – on the findings of the Appellate Body in US - Softwood Lumber V.

          WT/DS264/RW

          Page 11


          how [it] could find that the transaction-to-transaction and average-to-transaction methodologies could provide contextual support for the United States' interpretation of Article 2.4.2 without examining first whether zeroing is permitted under those methodologies".27 If the Appellate Body's interpretation of the term "margins of dumping" (in isolation from the phrase "all comparable export transactions) necessarily leads to the prohibition of zeroing under the T-T comparison methodology, as alleged by Canada,28 there would have been no need for the Appellate Body to first examine "whether zeroing is permitted" under the T-T methodology before addressing the contextual argument advanced by the United States. The fact that the Appellate Body declined to extend its interpretation of "margins of dumping" from the context of the W-W methodology to the T-T methodology suggests strongly that that interpretation was indeed limited to the situation where the term "margins of dumping" is used in conjunction with the phrase "all comparable export transactions".


        13. Furthermore, the first clause of the first sentence of Article 2.4.2 describes the W-W methodology as the establishment of margins of dumping "on the basis of a comparison of a weighted average normal value with a weighted average of prices of all comparable export transactions". The phrase "all comparable export transactions" is not included in the description of the T-T comparison methodology, which is set out in the second clause of the first sentence Article 2.4.2, and is the methodology at issue in these Article 21.5 proceedings. According to the Appellate Body, the parties in US - Softwood Lumber V "disagree[d] as to the proper interpretation of the terms 'all comparable export transactions' and 'margins of dumping' ...".29 Although Canada asserts that "[t]he phrase 'all comparable export transactions' was not central to the Appellate Body's findings that intermediate comparisons must be aggregated to arrive at margins of dumping",30 the Appellate Body explicitly "emphasize[d]" that because "both of these terms occur in the same sentence and relate to establishing the existence of margins of dumping under Article 2.4.2", "they should be interpreted in an integrated manner".31 Accordingly, we do not consider it appropriate to focus on the Appellate Body's interpretation of the term "margins of dumping", in isolation from the phrase "all comparable export transactions".32


          image

          27 Appellate Body Report, US – Softwood Lumber V, supra note 9, para. 105 (footnote omitted, emphasis supplied).

          28 Canada also asserts that the term "margins of dumping" appears, without modification, in a single sentence that applies to both the W-W and T-T comparison methodologies. According to Canada, therefore, it

          is not possible, as a matter of grammatical construction, for "margins of dumping" to have one meaning for one methodology and another meaning for the other. We are not persuaded by this argument, however, because the two comparison methodologies in the first sentence are separated and distinguished by the disjunctive "or". Since "margins of dumping" may therefore be established in different ways, using different methodologies, it is entirely possible that the nature of the resultant "margins of dumping" may also differ, in order to reflect the nature of the comparison methodology at issue.

          29 Appellate Body Report, US – Softwood Lumber V, supra note 9, para. 82 (emphasis supplied).

          30 Canada's second written submission, para. 16.

          31 The importance attached by the Appellate Body to the phrase "all comparable export transactions"

          may also be implied from the fact that it upheld, without modification, the finding of the original panel that had placed great emphasis upon this phrase. Indeed, the Panel had found that the United States violated Article 2.4.2

          "by not taking into account all comparable export transactions when DOC calculated the overall margin of dumping as Article 2.4.2 requires that the existence of margins of dumping has to be established for softwood lumber on the basis of a comparison of the weighted- average-normal-value with the weighted average of prices of all comparable export transactions, that is, for all transactions involving all types of the product under investigation."

          Panel Report, US – Softwood Lumber V, supra note 5, para. 7.224. The Panel's views did not address the term "margins of dumping" or the concept of "product as a whole" as discussed by the Appellate Body.

          32 We note that Canada also relies on the Appellate Body's statements regarding Article VI.1 of the

          GATT 1994 and Article 2.1 of the AD Agreement (see, for example, para. 9 of Canada's second written submission). Since these statements were made in the context of the Appellate Body's interpretation of the phrase "margins of dumping" (see, Appellate Body Report, US – Softwood Lumber V, supra note 9, paras. 92-

          WT/DS264/RW

          Page 12


          1. "Product As A Whole" Under The T-T Methodology


        14. Besides, in a case where the DOC had applied the W-W methodology, making a comparison for each of several sub-groups of the product, and then aggregating them to determine the margin of dumping, it was entirely logical for the Appellate Body to have concluded that the margin of dumping for the "product as a whole" must fully reflect those instances where, for a particular sub-group,33 the weighted average export price was greater than the weighted average normal value. We do not consider that this necessarily requires that in a calculation based on the T-T methodology, the same logic applies and the same result must obtain. In particular, although there is little doubt that a margin of dumping is established for each exporter/producer with respect to the product under investigation, further examination indicates that "product" need not necessarily be interpreted as "product as a whole", in the sense that Canada posits, that is, the summed results, fully reflecting negative and positive results, of all comparisons concerning the product under investigation. There are also good reasons why "margins of dumping" need not necessarily relate to "the product as a whole" in all circumstances in the AD Agreement.


        15. The Appellate Body drew its conclusion that dumping is to be found for the "product as a whole" from its consideration of Article VI of GATT 1994 and Article 2.1 of the AD Agreement, which both define the concept of "dumping", in the case of the latter, by its own terms for the entire AD Agreement.34 To extend the Appellate Body's reference to the concept of "product as a whole" in the sense that Canada proposes to the T-T methodology would entail accepting that it applies throughout Article VI of GATT 1994, and the AD Agreement, wherever the term "product" or "products" appears.35 A review of the use of these terms does not support the proposition that "product" must always mean the entire universe of exported product subject to an anti-dumping investigation. For instance, Article VI:2 states that a contracting party "may levy on any dumped product" an anti-dumping duty. Article VI:3 provides that "no countervailing duty shall be levied on any product". Article VI:6(a) provides that no contracting party shall levy any anti-dumping or countervailing duty on the importation of any product...". Similarly, Article VI:6(b) provides that a contracting party may be authorized "to levy an anti-dumping or countervailing duty on the importation of any product". Taken together, these provisions suggest that "to levy a duty on a product" has the same meaning as "to levy a duty on the importation of that product". Canada's position, if applied to these provisions, would mean that the phrase "importation of a product" cannot


          image

          93), these statements should likewise be read in light of the phrase "all comparable export transactions". Since Canada has not raised any claims based on these provisions, there is no need for us to examine their application in the context of the T-T comparison methodology, i.e., independent of the phrase "all comparable export transactions". That being said, we consider that there is nothing inherent in the word "product[]" (as used in Article VI:1 of the GATT 1994 and Article 2.1 of the AD Agreement) to suggest that this word should preclude the possibility of establishing margins of dumping on a transaction-specific basis, although this should not imply that a margin of dumping established with respect to a particular transaction is sufficient to impose an anti- dumping measure on all subsequent imports of the product. The notion that "a product is introduced into the commerce of another country" (Article 2.1) clearly can meaningfully apply to a particular export sale and does not require consideration of different export sales taken together. Indeed, even Canada acknowledges that a margin of dumping may be established for a single export transaction when an investigation involves only one transaction. See Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 27.

          33 We note that the Appellate Body itself described the "product as a whole" in US - Softwood Lumber

          V as "softwood lumber", rather than "all comparisons involving softwood lumber". Appellate Body Report, US – Softwood Lumber V, supra note 9, para. 99. In light of the Appellate Body's own description of "the product as a whole", we believe that the Appellate Body simply used the phrase "product as a whole" to emphasise the difference between establishing a margin of dumping for a single model of the product under investigation on the one hand, and establishing a margin of dumping for the product under investigation writ large, in all its types, models or categories.

          34 Article 2.1 begins "For the purpose of this Agreement,...".

          35 We note that the phrase "product as a whole" does not appear in either Article VI of GATT 1994, or the AD Agreement.

          WT/DS264/RW

          Page 13


          refer to a single import transaction. In many places where the words product and products are used in Article VI of the GATT 1994, an interpretation of these words as necessarily referring to the entire universe of investigated export transactions is not compelling.36


        16. The Appellate Body supported its reference to "product as a whole" by having regard to Articles 6.10 and 9.2 of the AD Agreement. For the reasons set forth below, we do not consider that these provisions provide any guidance as to whether or not offsets should be made outside the context of the W-W comparison methodology at issue in US – Softwood Lumber V.


        17. Article 6.10 of the AD Agreement requires that, as a rule, "an individual margin of dumping" shall be determined for each known exporter or producer of "the product under investigation". That is to say, known exporters or producers must be treated individually for purposes of determining dumping. In addition, this provision can also reasonably be understood to imply that for each of these exporters or producers a single margin of dumping must be calculated. Thus, Article 6.10 arguably entails a need to aggregate the results of the comparisons made in respect of different transactions in order to establish "an individual margin of dumping" for a particular exporter or producer. Assuming such an obligation exists, this does not answer the question of what methodology can be used to calculate that margin. The AD Agreement contains no express provision on the precise methodology to be used to calculate an overall margin. It certainly does not preclude the view that, in the methodology at issue in this dispute, the DOC did, in fact, calculate a single overall margin for each investigated producer/exporter of softwood lumber, i.e. a margin derived by aggregating the results of multiple transaction-specific comparisons of export price and normal value. Except in the case of Article 5.8, the AD Agreement does not expressly indicate whether such a margin must be expressed as an absolute amount or as a percentage. Where a margin is expressed as a percentage, which implies that the magnitude of one variable is defined as a proportion of another variable, the AD Agreement does not contain any rules regarding the methodology for computing these variables, i.e,. the denominator and numerator. Absent such rules, the mere fact that Article 6.10 uses the term "product under investigation" is insufficient to conclude that this provision dictates the use of a particular methodology for calculating an overall margin of dumping whereby the numerator of that margin must include the sum total of all (positive and negative) differences between export prices and the normal value.


        18. Similarly, the fact that Article 9.2 of the AD Agreement provides for the imposition of anti- dumping duty on "product" is of limited relevance to the question of the precise methodology for calculating margins of dumping. It is obvious that there must be identity between the product subject to the anti-dumping duty and the product in respect of which determinations of dumping and injury are made.37 It also stands to reason that, because the duty is applied to all subsequent imports of the product, the determination of dumping on the basis of a T-T comparison must be based on an analysis that takes into account all transactions under consideration. It does not necessarily follow, however, that transactions in which export prices are above the normal value must be treated in the same manner as transactions in which export prices are below the normal value.


          1. "Margins of Dumping" Under The T-T Methodology


        19. We are also not persuaded that, outside the context of the W-W comparison methodology, "margins of dumping" must necessarily be established for "the product as a whole", on the basis of the


          image

          36 More generally, an analysis of the use of the words product and products throughout the GATT 1994, indicates that there is no basis to equate product with "product as a whole" in the sense in which Canada uses that term in this proceeding. Thus, for example, when Article VII:3 of the GATT refers to "the value for customs purposes of any imported product", this can only be interpreted to refer to the value of a product in a particular import transaction.

          37 Of course, the "product" on which the duty is actually imposed after the conclusion of an investigation is not the universe of exports that was subject to the investigation, as these exports have already

          been imported without duty.

          WT/DS264/RW

          Page 14


          full results of all comparisons. While "margins of dumping" is not defined by the AD Agreement, Article VI:2 of the GATT 1994 provides that, for the purposes of Article VI, "the margin of dumping is the price difference determined in accordance with the provisions of paragraph 1" of Article VI. Paragraph 1 of Article VI defines dumping as a practice "by which products of one country are introduced into the commerce of another country at less than the normal value of the products" (emphasis supplied). Leaving aside the use of constructed normal value and third country reference markets, addressed in Article VI:1(b)(i) and (ii), Article VI:1 provides that "a product is to be considered as being introduced into the commerce of an importing country at less than its normal value, if the price of the product exported from one country to another (a) is less than the comparable price, in the ordinary course of trade, for the like product in the exporting country" (emphasis supplied). In other words, there is dumping when the export "price" is less than the normal value. Given this definition of dumping, and the express linkage between this definition and the phrase "price difference", it would be permissible38 for a Member to interpret the "price difference" referred to in Article VI:2 as the amount by which the export price is less than normal value, and to refer to that "price difference" as the "margin of dumping".


        20. In the absence of any definition of the phrase "margins of dumping" in Article 2.4.2, and in the absence of any obligation under the T-T methodology to ensure that "all comparable export transactions" are represented in a weighted average export price, we see no reason why a Member may not, when applying the transaction-to-transaction comparison methodology, establish the "margin of dumping" on the basis of the total amount by which transaction-specific export prices are less than the transaction-specific normal values. In such cases, the margin of dumping clearly would reflect the price difference for dumped, rather than non-dumped, exports of the product by a particular exporter.39 In our view, this would be a permissible interpretation of the relevant part of the first sentence of Article 2.4.2, even though it does not reflect the full results of all comparisons. In other words, when establishing the amount of dumping for the purpose of calculating a margin of dumping under the T-T comparison methodology, an investigating authority need not include in its calculations the results of comparisons where export price exceeds normal value.


          1. Margins of Dumping Without Averaging


        21. In any event, even if we were to accept the argument that "margins of dumping" must be established "for the product as a whole", this concept in and of itself does not explain whether offsets should be made in the context of the T-T comparison methodology. It seems clear to us that the concept of a "weighted average export price", and particularly combined with the phrase "all comparable export transactions" is reasonably understood to require taking into account the entire universe of export prices for all transactions under consideration, regardless of whether they are above or below normal value. Thus, that the Appellate Body, drawing on the concept of "product as a whole", concluded that the margin of dumping should reflect all comparisons involving the product, or all import transactions involving the product, seems entirely consistent with the obligation to establish margins of dumping under the W-W comparison methodology, which is an averaging methodology that requires consideration of "all comparable export transactions". But, since this latter phrase does not appear in the text with reference to the T-T comparison methodology, we see no basis to conclude that the phrase "margins of dumping for the product as a whole" must be understood in the same way when applying the T-T comparison methodology. Moreover, although the T-T methodology might involve aggregation or summing up of results of comparisons of transaction- specific prices, this should not be confused with averaging. There is no requirement that aggregation


          image

          38 We recall in this regard, the applicable standard of review set out Article 17.6(ii) of the AD Agreement, as discussed in paras. 5.5 - 5.7 above.

          39 In the Section 129 Determination, the DOC expressed the amount of dumping as a proportion of the

          total value of export sales. The denominator (i.e., the total value of export sales) included the value of both dumped and non-dumped export transactions. Only the numerator included only dumped export transactions. Canada has not challenged the DOC's calculation of the denominator.

          WT/DS264/RW

          Page 15


          under the T-T methodology should result in, or reflect, averages. Thus, we see no reason to conclude that a concept of a "margin of dumping for the product as a whole" must be understood, as Canada has argued, to be a margin of dumping that incorporates (through aggregation) price differences on individual transactions where export price is both greater than and less than normal value. Rather than investigating average pricing behaviour over a given period, the T-T methodology allows authorities to investigate transaction-specific instances of dumping, where export price is less than normal value, and calculate an overall margin of dumping which reflects the total amount of dumping on the imports subject to the investigation.40 As a result, the US interpretation of the first sentence of Article 2.4.2, in the context of the T-T comparison methodology, as not precluding zeroing would seem at a minimum to be permissible.41


          1. Summary


        22. In summary, we recall that Canada has not argued that the relevant text of Article 2.4.2 explicitly prohibits zeroing during the aggregation of transaction-to-transaction comparisons. Canada's Article 2.4.2 claim is instead dependent on the Appellate Body's finding in US - Softwood Lumber V that margins of dumping should be established for "the product as a whole". However, we have demonstrated that the Appellate Body's findings regarding the need to establish margins of dumping for "the product as a whole" should not necessarily be applied in the same manner outside the W-W comparison methodology. Those findings may in any event not be isolated from its consideration of the phrase "all comparable export transactions", which phrase does not appear in the relevant text concerning the transaction-to-transaction comparison methodology. This difference in language reflects a fundamental distinction between the nature of the W-W and T-T comparison methodologies. Although both methodologies might involve aggregation, the W-W methodology is based on an analysis of average price behaviour, while the T-T methodology allows an investigating authority to identify transaction-specific instances of dumping. In these circumstances, we conclude that there is no basis to uphold Canada's claim that Article 2.4.2 required the DOC to establish margins of dumping by aggregating the results of all transaction-to-transaction comparisons, offsetting non-dumped comparisons against dumped comparisons.


          1. Broader Contextual Considerations


        23. The above analysis is confirmed by a number of broader contextual considerations which highlight a number of difficulties that would result from simply extending the findings of the Appellate Body in US - Softwood Lumber V, as Canada would have us do. One of these broader contextual considerations concerns the impact of Canada's "margins of dumping for the product as a whole" argument on the targeted dumping comparison methodology set forth in the second sentence of Article 2.4.2. Further contextual considerations concern the consequences of applying Canada's "margins of dumping for the product as a whole" argument to other provisions of the AD Agreement. In its final submission to the Panel, Canada sought to discourage the Panel from taking the latter


          image

          40 We note that there is a fundamental underlying question whether the concept of dumping may relate to individual export transactions or whether it relates exclusively to the average pricing behaviour of an exporter over time. Although Canada has not expressly argued that dumping in the AD Agreement and in the GATT 1994 should be conceptualized in terms of average pricing behaviour, the argument that positive differences between individual export prices and normal values in a T-T comparison must be offset by negative differences between individual export prices and normal values, leads to a result tantamount to a comparison on an average-to-average basis. However, there is no indication in the AD Agreement or GATT 1994 to suggest any agreement or common understanding among Members on this underlying question, and thus no principled basis on which to conclude that one or the other view should influence the interpretation of the text of the AD Agreement. Consequently, there would seem no basis to prohibit a calculation methodology which generates an overall margin of dumping which reflects precisely the amount of anti-dumping duty that could have been collected on the import transactions from which it was calculated.

          41 We recall that, pursuant to Article 17.6(ii) of the AD Agreement, we shall find in favour of the defending party if its measure is based on a permissible interpretation of the AD Agreement.

          WT/DS264/RW

          Page 16


          considerations into account. In particular, Canada stated that it "has not argued that the term 'margins of dumping' must have the same meaning throughout the Agreement".42 Instead, Canada asserts that "margins of dumping" simply has the same meaning throughout Article 2.4.2. In its oral statement at the substantive meeting, however, Canada explicitly stated that "the Appellate Body interpreted 'margins of dumping' in the context of Article VI of the GATT and Article 2.1 of the Anti-Dumping Agreement – provisions that apply to the entire Anti-Dumping Agreement".43 Despite its protestations in its final submission, therefore, Canada has clearly argued that the phrase "margins of dumping" should have the same meaning throughout the AD Agreement, and that that meaning must necessarily be that set forth in the findings of the Appellate Body in US - Softwood Lumber V.


        24. Since Canada's Article 2.4.2 claim is essentially based on the extension of the Appellate Body's US - Softwood Lumber V findings beyond the scope of that case, contextual considerations demonstrating the difficulties of applying those Appellate Body findings universally, throughout the AD Agreement, are highly relevant to the case before us. We begin by discussing the difficulty of applying the Appellate Body's findings in the context of the second sentence of Article 2.4.2. We then examine the implications of extending those findings to prospective normal value duty assessment systems. Thereafter, we examine the consequences of finding that margins of dumping must be established for the product as a whole in the context of Article 2.2 of the AD Agreement. We conclude our analysis by identifying the problem of reconciling the "margins of dumping for the product as a whole" approach with prior GATT discussion of the use of transaction-specific margins of dumping.


          1. Article 2.4.2, second sentence


            The US mathematical equivalence argument


        25. The second sentence of Article 2.4.2 provides for a departure from the two comparison methodologies provided for in the first sentence of that provision "if the authorities find a pattern of export prices which differ significantly among different purchasers, regions or time periods". In instances of such targeted dumping, "[a] normal value established on a weighted average basis may be compared to prices of individual export transactions" (the "W-T comparison methodology"). In such cases, the margins of dumping are established using the W-T comparison methodology. If Canada is correct in arguing that the Appellate Body "did not limit its analysis of 'dumping' and 'margins of dumping' to the meaning of these terms under the weighted-average-to-weighted-average methodology",44 there is seemingly nothing to prevent the Appellate Body's interpretation applying not only to the T-T methodology set forth in the second part of the first sentence of Article 2.4.2, but also to the W-T methodology provided for in the second sentence thereof.45 This would effectively prohibit zeroing under the W-T comparison methodology (just as Canada alleges that it prohibits zeroing under the T-T comparison methodology). The United States asserts that such prohibition of zeroing in the context of the W-T methodology would mean that the margin of dumping using the W- T methodology would be mathematically equivalent to the margin of dumping established using the W-W methodology, thereby depriving the second sentence of Article 2.4.2 of effect. It is by now well established that panels may not interpret provisions in a way that would reduce those provisions to redundancy or inutility.46 If the United States is correct, therefore, we are precluded from endorsing a prohibition of zeroing on the basis of the Appellate Body's interpretation of the term "margins of dumping" in US - Softwood Lumber V., as argued by Canada. For this reason, we explored this issue



          image

          42 See Canada's Comments on the United States' Answers t the Panel's Questions, page 1.

          43 Oral Statement of Canada, para. 13.

          44 Ibid. para. 9.

          45 Canada has not provided any basis for restricting the application of the Appellate Body's interpretation to only the first sentence of Article 2.4.2.

          46 See, for example, Appellate Body Report, United States – Standards for Reformulated and

          Conventional Gasoline ("US – Gasoline"), WT/DS2/AB/R, adopted 20 May 1996, DSR 1996:I, 3, para. 21.

          WT/DS264/RW

          Page 17


          in detail with the parties and third parties, both at the substantive meeting, and through subsequent written questions. We did not explore this issue in order to make findings under the second sentence of Article 2.4.2, which is, of course, not before us in this dispute. Instead, we explored this issue to test the validity of the US mathematical equivalence argument. Our analysis of the various arguments made in respect of this issue led us to accept the US mathematical equivalence argument.


          Separate margins of dumping for patterns of export prices


        26. In response to the US argument, Canada and certain third parties sought to demonstrate that the W-T comparison methodology may be applied so as to give a result that is mathematically different from the W-W methodology, even without zeroing in the manner practised by the United States. In other words, they sought to demonstrate that the W-T methodology would not be nullified in the manner alleged by the United States. As described below, they do so by effectively arguing that the term "margins of dumping" has a different meaning under the W-T comparison methodology than under the W-W and T-T comparison methodologies, and that this difference in meaning is linked to the targeted nature of the dumping examined in the second sentence of Article 2.4.2.47


        27. We presented Canada with a written question asking how the targeted dumping mechanism might operate to produce a mathematically distinct result from that obtained in a W-W comparison, without zeroing, in a hypothetical case regarding regional targeting, where imports into one region are dumped, and imports into other regions are not dumped. In reply, Canada stated that, having proven the regional pattern of dumping:


          an investigating authority would ... have a choice. It could either continue its anti- dumping investigation into export transactions to other parts of the country or it could terminate the investigation outside the targeted region. As the example in the question provides that transactions to other regions were not dumped, Canada assumes that the investigating authority would terminate its investigation into the non-targeted regions and continue the investigation with respect to transactions to the targeted region only.


          The investigating authority could then calculate the amount of dumping to the targeted region and place this value in the numerator. Canada is of the view that zeroing would not be permitted in the aggregation of any intermediate comparisons, just as it is not permitted in the two other methodologies found in the first sentence of Article 2.4.2. Similarly, the denominator would consist of the total value of dumped and non-dumped export transactions to the targeted region. The necessary injury analysis would also consider the same series or group of export transactions.


          As this example resulted in the termination of the investigation outside the targeted region, anti-dumping duties would only be applied to imports into the targeted region. Were an investigating authority to use the targeted dumping methodology in this example, it necessarily would produce a different margin of dumping than the weighted-average-to-weighted average methodology which would be applied to all imports into a country because the latter methodology would necessarily examine a different data set (i.e., all export transactions, rather than the subset of export transactions that make up transaction involving the targeted region).



          image

          47 We note that this was not the position taken by Thailand. Thailand explicitly stated that the results of the W-T comparisons within the targeted pattern were merely intermediate comparisons that needed to be aggregated. See Thailand's Responses to the Panel's Questions for Parties and Third Parties, Response to Question 45, para. 12.

          WT/DS264/RW

          Page 18


        28. In reply to a further question, Canada described how the second sentence of Article 2.4.2 might be applied in respect of purchaser-specific targeting. Canada asserted that, having identified such a pattern, an investigating authority could conduct a W-T calculation and apply that margin of dumping to the transactions involving that purchaser. Canada acknowledged that


          [t]his procedure would result in two margins of dumping being applied to an exporter or producer who has some sales to the purchaser to which the targeted dumping methodology was applied and other sales to other purchasers that were not subject to the targeted dumping methodology.48


        29. Regarding the application of the second sentence of Article 2.4.2 in respect of time-specific targeting, Canada asserts that "the investigating authority would ... calculate a margin for the transactions that fall within the pattern", and "apply that margin of dumping to those transactions".49 Canada submits that "only one margin of dumping would be applied at any particular point in time. If an import occurred within the period of time of the year in which targeted dumping was found, then the targeted dumping rate would be applied. If the import occurred at a point in time that was outside the targeted dumping time period, then the rate calculated for the balance of the year would be applied."50


        30. We note as an initial matter that each of Canada's proposed methods of applying the targeted dumping provision changes the parameters of the analysis from those that would apply in a W-W analysis. Thus, Canada's arguments do not address the question of how a targeted dumping analysis based on W-T comparison without zeroing could yield a result different from a W-W comparison, in a situation holding everything except the comparison methodology equal. Nor does Canada address how any anti-dumping duty calculated under one of its proposed methodologies could be applied consistently with the AD Agreement. For instance, Article 9.2 provides that "[w]hen an anti-dumping duty is imposed in respect of any product, such duty shall be collected in the appropriate amounts in each case, on a non-discriminatory basis on imports of such product form all sources found to be dumped and causing injury". It is not clear that collection of duty only on imports into certain regions, to certain purchasers, or during certain time-periods, even if otherwise possible51, would be consistent with this requirement. We note also in this regard Article 6.10, which provides that "[t]he authorities shall, as a rule determine an individual margin of dumping for each known exporter or producer concerned of the product under investigation" (emphasis added). It is not clear that calculation of two different duty rates, for imports forming the targeted pattern, and other imports, would be consistent with this provision. On the other hand, the alternative, of imposing the duty calculated for the imports forming the targeted pattern on all imports from the producer/exporter in question without limitation would, presumably, be an even worse outcome than the imposition of a duty calculated with zeroing, and has not been suggested by any party or third party as a possibility.


          Separate universes of export transactions treated as the "product as a whole"


        31. Canada addresses the US mathematical equivalence argument by showing that the results of the W-T methodology will differ from those of the W-W methodology when one margin of dumping is calculated (using the W-T methodology, without zeroing) for the targeted sub-category of product, and another is calculated for the non-targeted sub-category.52 At first glance, the calculation of multiple margins of dumping for different sub-categories of product would appear to be at odds with


          image

          48 Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 6.

          49 Ibid., para. 30.

          50 Ibid., para. 31.

          51 See discussion below at para. 5.44 regarding imposition of duties in regional industry cases, for which special provision is made in Article 4.2 of the AD Agreement.

          52 In each case, the amount of dumping attributable to the imports forming the targeted pattern is not

          offset by non-dumping outside of that pattern.

          WT/DS264/RW

          Page 19


          Canada's argument that a single53 margin of dumping should be established for the product as a whole, regardless of which Article 2.4.2 comparison methodology is being used.54 Canada seeks to reconcile this apparent inconsistency by defining the "product as a whole" as "the universe of transactions that would be aggregated to arrive at a margin of dumping". In other words, Canada treats the imports forming the targeted pattern of dumping as a separate universe of transactions, calculates a single margin of dumping for that universe, and refers to that margin of dumping as being the margin for the product as a whole. Canada claims that this is the sense in which the Appellate Body used the phrase "product as a whole". Specifically, Canada asserts that the Appellate Body "used the phrase 'product as a whole' to refer to the universe of transactions that would be aggregated to arrive at a margin of dumping".55 However, Canada points to nothing in the Appellate Body Report to support its argument that the Appellate Body used the phrase "product as a whole" to mean a separate universe, or sub-category, of export transactions, either in general, or in the context of the second sentence of Article 2.4.2.56 Nor have we been able to find anything to this effect in the Appellate Body Report. Rather, we note that the Appellate Body frequently refers instead to "the product under investigation as a whole", as opposed to "a type, model, or category of that product"57. By the phrase "the product under investigation as a whole", we understand the Appellate Body to mean the product as defined upon initiation of the investigation. This understanding is shared by Canada, which itself points out in reply to Question 11 that "[t]he Appellate Body properly concluded that 'product under investigation' referred to the 'product' defined at the outset of an investigation".58 We therefore see no basis for concluding that the Appellate Body, as alleged by Canada, used the phrase "product as a whole" to refer to a sub-category, or universe, of transactions that could be carved out from the totality of the transactions initially under investigation. In addition, Canada's argument would mean that an investigating authority might calculate different "margins of dumping" in one investigation for one exporter for different sub-categories of products based on purchasers, regions, or time periods. This is at odds with Canada's argument that, throughout Article 2.4.2, "margins of dumping" always means a single margin of dumping (per exporter/producer) for the product as a whole, that is, the entire universe of exports subject to the investigation.


        32. Although Canada has sought to rely on the findings of the Appellate Body in US - Softwood Lumber V, it does not point to anything in the text of the second sentence of Article 2.4.2 that would justify the calculation of separate margins of dumping for separate universes of transactions. Nor have we been able to identify any language that might support such an approach.59 Rather, because


          image

          53 We note in this regard that Canada asserted that "an investigating authority [] must aggregate all the results of transaction-specific comparisons to arrive at a single margin of dumping for the product as a whole". Canada's second written submission, para. 13 (bold emphasis supplied).

          54 We recall Canada's statement that "[t]he consistent meaning of the phrase 'margins of dumping' within Article 2.4.2 warrants emphasis". Canada's Comments on the United States' Answers to the Panel's

          Questions, page 2.

          55 Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 6, para. 28 and Question 11, paras. 46-48.

          56 Indeed, as noted above, the Appellate Body declined to consider the second sentence of Article 2.4.2 as context in its analysis, and thus did not address it at all.

          57 See, for example, Appellate Body Report, US – Softwood Lumber V,, supra note 9, at paras, 93, 96

          and 103..

          58 Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 11, para. 46, emphasis supplied. Likewise, the EC asserts that the Appellate Body used

          the words "as a whole" "to emphasise that 'the product' is 'the product' as defined by the investigating authority at the outset of the original proceeding, and not some type, model or category of that product". European Communities Responses to the Panel's Questions, Response to Question 32 (emphasis supplied).

          59 In our view, the only text in the second sentence of Article 2.4.2 that might possibly be of relevance in this regard is the phrase "individual export transactions". However, there is nothing to suggest that the group

          of "individual export transactions" to be compared with normal value (established on a weighted average basis) may be confined to the "pattern of export prices" identified by the investigating authority. Indeed, had the concepts of "individual export transactions" and "pattern of export prices" been identical, we would have expected them to be described within a single sentence in identical terms.

          WT/DS264/RW

          Page 20


          the W-T methodology is presented as a limited alternative to the W-W and T-T methodologies, the text of Article 2.4.2 would appear to indicate that an investigating authority should use only the W-T methodology if the factual circumstances are such that it is entitled to not use either the W-W methodology or the T-T methodology. In other words, if targeted dumping is identified, and the investigating authority invokes the second sentence of Article 2.4.2, the whole investigation would be conducted on the basis of the W-T methodology.60 Just as there is no basis in Article 2.4.2 for combining the W-W and T-T methodologies, so too there would seem to be no basis for combining one, or both, of these methodologies with the W-T methodology under the second sentence of Article 2.4.2.


        33. In addressing the US mathematical equivalence argument, therefore, Canada's position exposes a fundamental inconsistency between its argument that the term "margins of dumping" has the same meaning throughout Article 2.4.2 on the one hand, and its assertion that the Appellate Body has found that "margins of dumping" may only ever be calculated for a product as a whole on the other. This inconsistency means either that the term "margins of dumping" does not have the same meaning throughout Article 2.4.2, or that "margins of dumping" need not necessarily (outside of the W-W comparison methodology) be established for the product as a whole by aggregating the results of all transaction-specific comparisons. Either way, this inconsistency fundamentally undermines Canada's claim that the term "margins of dumping" must be interpreted and applied in the same way under both the W-W and T-T comparison methodologies.


          Targeted dumping as an exception to the need to establish margins of dumping for the "product as a whole"


        34. The EC, as third party, seeks to avoid this inconsistency by arguing that the second sentence of Article 2.4.2 is an exception.61 Thus, the EC asserts that a single "margin of dumping" must be established for the product as a whole throughout the AD Agreement, except in the case of targeted dumping.62 However, the ordinary meaning of the text of the second sentence of Article 2.4.2 provides no basis for such an approach. The second sentence simply states that the two primary comparison methodologies provided for in the first sentence need not be applied when certain


          image

          60 In its comments on the US replies to our questions, Canada further asserts that the mathematical equivalence argument would be undermined when the W-T methodology is used to calculate a margin of dumping for the targeted export transactions, and the T-T methodology is used to calculate a margin of dumping for the non-targeted export transactions. Canada's Comments on the United States' Answers to the Panel's Questions, page 5. In our view, such a combination of the W-T and T-T methodologies is not envisaged by Article 2.4.2. In any event, we are examining ways in which the W-T methodology might continue to have meaning in and of itself. We do not consider it sufficient to reject the mathematical equivalence argument on the basis that the second sentence of Article 2.4.2 continues to have some limited applicability provided the W-T methodology is combined with the T-T methodology.

          61 We note that the European Communities has brought its own case against US zeroing measures (see WT/DS294, currently on appeal). We have only considered the arguments presented by the European Communities in its capacity as third party in these proceedings.

          62 In response to a Question from the Panel, the EC asserts that the term "margin of dumping"

          "has the same meaning throughout Article VI:2 of the GATT 1994 and the ADA, subject to the targeted dumping provisions. [The EC] agree[s] with Canada that if there is targeted dumping by purchaser, region or time, an investigating authority is entitled – for example - to calculate the dumped amount relating to a purchaser, region or time so identified. The second sentence of Article 2.4.2 is an exception to the first sentence of Article 2.4.2; thus, similarly, the targeted dumped amount may, subject to the conditions provided for in the second sentence of Article 2.4.2, be expressed as a percentage of export price, or characterized as a “margin of dumping”. Rather than expressing this in terms of the second sentence of Article 2.4.2 constituting an exception to the general requirement to calculate a margin of dumping for the product as a whole, the EC would say that the targeted dumping provisions provide for a specific methodology to determine such margin in exceptional circumstances."

          European Communities Responses to the Panel's Questions, Response to Question 1.

          WT/DS264/RW

          Page 21


          conditions are met, i.e., when there is evidence of particular types of targeted dumping. In such circumstances, the second sentence provides for the use of a third comparison methodology. There is nothing in the second sentence to suggest that the scope of the investigation is different when the targeted dumping provision applies than when the W-W or T-T comparison methodology is used. Nor is there anything in the second sentence to suggest that an investigating authority may focus on (and calculate separate margins of dumping for) sub-categories of products that it would not otherwise be able to focus on in the context of a W-W or T-T comparison. In other words, the second sentence merely allows the investigating authority to compare the investigated data differently than in the first sentence (by using a W-T approach, rather than W-W or T-T); it does not allow the investigating authority to change that data, or ignore parts thereof.63 Thus, in examining the US mathematical equivalence argument, we are exploring ways in which the W-T targeted dumping methodology might be applied without zeroing so as to produce results that are different from the W-W methodology on a ceteris paribus basis.


        35. Indeed, the European Communities appears to implicitly acknowledge that Article 2.4.2 per se does not allow an investigating authority to focus on sub-sets of transactions because, in the context of regional targeted dumping, it relies on Article 4.2 of the AD Agreement to justify the imposition of the resultant anti-dumping measure on a regional basis. Thus, after stating that the investigating authority could focus its analysis on the set of transactions in the targeted region, the EC asserts that the resultant anti-dumping duty "could, for example, be imposed on the products consigned for final consumption to [that] region [], when the domestic industry has been interpreted as referring to the producer in [that] region [] consistent with Article 4.2".64 We recall that this argument has been advanced by the EC in response to the US argument that a broad prohibition of zeroing (based on the proposition that a single "margin of dumping" must be calculated for "the product as a whole") would nullify the comparison methodology set forth in the second sentence of Article 2.4.2. In this context, we are not persuaded by an argument that the W-T comparison methodology still has meaning, despite a broad prohibition of zeroing, provided the second sentence of Article 2.4.2 is applied in conjunction with Article 4.2. In testing the continued utility of the W-T comparison methodology, the Panel is interested in exploring ways in which the second sentence of Article 2.4.2 continues to have meaning in and of itself.65 The EC's argument says nothing regarding how the targeted dumping provision would apply in a case where the "region" did not satisfy the strict requirements of Article 4.1(ii) regarding regional markets, and thus Article 4.2 did not apply.


        36. Furthermore, unlike in Article 4.2, there is no provision in Article 2.4.2 for application of the anti-dumping duty resulting from a targeted dumping analysis to only imports into the region. Article 4.2 requires the imposition of anti-dumping duties only on imports into the defined geographical market, unless the Constitutional law of the importing Member precludes such selective


          image

          63 For this reason, we also reject Canada's argument that a regional targeted dumping analysis would produce a margin of dumping different from a W-W analysis "because the latter methodology would necessarily examine a different data set (i.e., all export transactions, rather than the subset of export transactions that make up transaction[s] involving the targeted region". Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 5, para. 20.

          64 European Communities Responses to the Panel's Questions, Response to Question 33.

          65 We also note the EC assertion. in response to Question 33 from the Panel, that if "the domestic industry has [not] been interpreted as referring to the producers in a certain area" (pursuant to Article 4.2), then

          the investigation would not focus on transactions in the targeted region. Instead, the investigation would cover both targeted and non-targeted regions together, and calculate a total dumped amount for both regions. In the numerical example provided by the EC in its response to our Question 33, non-dumped transactions are zeroed, i.e., there is no offset for non-dumped transactions. Although the EC purports to justify such zeroing on the basis of "the targeted dumping provisions" (See European Communities Responses to the Panel's Questions, Response to Question 33), we note that the comparison methodology employed is not targeted, because it is based on the weighted average export price for both targeted and non-targeted regions. In the context of such a non-targeted investigation, it makes no sense to speak of something being justified by the targeted pattern of export prices.

          WT/DS264/RW

          Page 22


          imposition. In such cases, Article 4.2 permits the imposition of the duties without limitation, provided certain conditions are met. In the context of Article 4.2, such non-selective duty imposition is generally not problematic because the dumped imports must be concentrated into an isolated geographic market as a prerequisite for regional analysis under Article 4.1(ii). In other words, there are generally few or no imports of the subject merchandise into other areas of the importing Member. This is not necessarily the case in the context of targeted dumping, however. The fact that Members found it necessary to include a specific provision dealing with this issue in the context of regional industries strongly suggests to us that selective imposition of anti-dumping duties in the case of targeted dumping would also have been specifically provided for.


          Targeted dumping as a difference affecting price comparability


        37. The EC also seeks to justify its focus on the set of transactions in the targeted region by reference to the allowances permitted under Article 2.4. In particular, the EC asserts that if there is a pattern of dumped prices in one region, but not in the other, then there is a difference between those regions that "affect[s] price comparability" in the meaning of Article 2.4.66 The EC states that "price comparability" is affected because the dumping into the targeted region would be masked if a single W-W comparison had to be made covering both the targeted and non-targeted regions. The EC asserts that, in such cases, the export price into the non-targeted region could be "adjusted" so that the margin of dumping for that region is zero, i.e., there would be no non-dumping to offset against the dumping in the targeted region.67


        38. The United States asserts that the EC essentially confuses two concepts that have nothing to do with one another: (1) differences in prices in the export market between regions, purchasers and time-periods; and (2) differences (which affect price comparability) in the importing country market between export price and normal value.68 The United States asserts that the EC has identified a difference that falls within the first category, whereas Article 2.4 only justifies allowances for difference falling under the second category.


        39. We note that the EC made the same argument to the US – Zeroing (EC) panel. That panel rejected the EC's argument in the following terms:


          This argument reflects a misinterpretation of the very concept of price comparability as used in Article 2.4 of the AD Agreement. Differences in price comparability in Article 2.4 for which an adjustment or allowance may have to be are differences between the product as sold in the export market and the product as sold in the domestic market with respect to factors such as level of trade, taxation, quantities, etc. The existence of differences in prices in the export market between regions, purchasers and time-periods is conceptually wholly irrelevant to, and outside the scope of, Article 2.4 because such differences have nothing to with whether or not export sales and domestic sales are comparable with regard to factors such as level of trade, taxation, quantities, etc. (...)69


        40. We agree with this reasoning. We therefore reject the EC argument that the Article 2.4 allowance mechanism provides a justification for focusing an investigation on the set of transactions


          image

          66 See, for example, European Communities Responses to the Panel's Questions, Response to Question 35(d).

          67 See, for example, Third Party Oral Submission by the European Communities, para. 34.

          68 See Answers to the United States to the Panel's Questions of 18 November, Reply to Question 15,

          para. 14.


          69 Panel Report, United States – Laws, Regulations and Methodology for Calculating Dumping

          Margins ("Zeroing") ("US – Zeroing (EC)"), WT/DS294/R, 31 October 2005 (appeal pending) at para. 7.279.

          WT/DS264/RW

          Page 23


          in the targeted region (or any other pattern of export prices), without making offsets for non-dumped transactions outside of that region (or outside of the relevant pattern).70


          Change in the normal value data set


        41. Finally, the EC, Japan and Thailand also challenge the US mathematical equivalence argument by suggesting that the result of a W-T comparison will differ from the result of a W-W methodology to the extent that different weighted average normal values are used, i.e., to the extent that the "W" in the W-W methodology is not the same as the "W" in the W-T methodology. Thus, the EC asserts that the second sentence of Article 2.4.2 permits the use of a normal value "calculated by reference to the sub-set of transactions within the pattern identified by the investigating authority".71 Japan states that "[s]o long as Members are not prohibited from using different bases or methods (including different time periods) to calculate the 'W' in the W-to-W and W-to- T comparisons, the outcomes of the comparisons will almost inevitably differ because the groups of transactions making up the weighted average normal value will differ".72 In respect of time-specific targeted dumping, Thailand asserts that "[i]f a W to T methodology is used for one period, and a W to W or T to T methodology for another, export sales prices in the different periods will be compared to different normal values ..."73


        42. In response, the United States submits that:


          nothing in the text of Article 2.4.2 supports such a leap from the treatment of prices in the export market to the treatment of prices in the normal value market. Moreover, there is no logic to the proposition that targeting in the export market according to purchaser, region, or time, which might justify special treatment of export prices, would also justify corresponding special treatment of normal value prices. That proposition assumes without any basis that the events that justify special treatment of prices in the export market also are occurring in the normal value market. Just because prices in the export market exhibit a targeted pattern with respect to purchaser, region or time does not mean that a corresponding targeted pattern is being exhibited in the normal value market. 74


        43. In our view, the second sentence of Article 2.4.2 provides for a specific comparison methodology to address situations where there is a "pattern of export prices" that result in targeted dumping that might not be identified through the W-W or T-T comparison methodologies. The relevant pattern, therefore, is the pattern of export sales into the importing Member. There is no reference in the second sentence of Article 2.4.2 to patterns of home market sales in the exporting Member. Once the relevant pattern has been identified, the symmetrical comparison methodologies provided for in the first sentence of Article 2.4.2 may be replaced by the asymmetrical comparison


          image

          70 Furthermore, we note the US argument that the possibility of restricting the universe of export transactions in the context of a targeted dumping analysis ignores Article 6.10 of the AD Agreement. This provision provides in relevant part that, "[i]n cases where the number of exporters, producers, importers or types of products involved is so large" as to make it impracticable to determine an individual margin of dumping for each know exporter or producer of the product under investigation, "the authorities may limit their examination

          ... to the largest percentage of the volume of the exports from the country in Question which can reasonably be investigated". In our view, the fact that this provision addresses explicitly the circumstances in which less than all export transactions may be examined in an investigation undermines the suggestion that the scope of an investigation may be limited – implicitly – through the second sentence of Article 2.4.2.

          71 See European Communities Responses to the Panel's Questions, Response to Question 9.

          72 See Replies of Japan to the Questions of the Panel, Response to Question 44.

          73 See Thailand's Responses to the Panels Questions for Parties and Third Parties, Response to Question 45, at para. 8.

          74 See Comments of the United States on Canada's and the third Parties Responses to the Panel's

          Questions, para 45 (footnote omitted).

          WT/DS264/RW

          Page 24


          methodology provided for in the second sentence. That asymmetrical comparison methodology employs a "normal value established on a weighted average basis". In order for the argument of the EC, Japan and Thailand to succeed, that normal value would have to differ (or at least have the potential to differ) from the "weighted average normal value" provided for in the first sentence. However, the abovementioned third parties have provided no textual analysis justifying any such conclusion. Given that the second sentence of Article 2.4.2 is designed to address a problem that might result from particular patterns of export prices, rather than home market prices, we ourselves see no basis in the text of that provision to conclude that the "normal value established on a weighted average basis" is different from the "weighted average normal value". Furthermore, we note that even Canada, the complaining Member, does not support the contrary argument of the EC, Japan and Thailand.75 As a result, we are not persuaded by these third parties' assertion that the US mathematical equivalence argument may be disposed of by varying the "W" in the W-T comparison methodology.


          Summary


        44. In response to a very simple argument of mathematical equivalence, Canada and certain third parties have provided convoluted explanations of how the W-T comparison methodology might be applied, without zeroing, so as to give results that are mathematically different from the results of the W-W comparison methodology. Their arguments are essentially based on the notion that targeted dumping allows investigating authorities to reach their determinations on the basis of the pricing behaviour in respect of universes of transactions that are narrower than the initial scope of the investigation, and narrower than those analysed under the W-W methodology. Such an approach is at odds with the very text of the second sentence of Article 2.4.2 and/or other provisions of the AD Agreement. In other words, Canada and the relevant third parties have failed to explain how the second sentence of Article 2.4.2 might be applied, without zeroing, in a WTO-consistent manner, so as to give results that are mathematically different from the results of the W-W comparison methodology. Such an approach also means that Canada and the relevant third parties have failed to address the US mathematical equivalence argument on a ceteris paribus basis. In these circumstances, and noting that the US mathematical equivalence argument has been confirmed by the panel in US – Zeroing (EC),76 we accept the US mathematical equivalence argument. We therefore agree with the United States that a general prohibition of zeroing based purely on the Appellate Body's interpretation of the phrase "margins of dumping" in US - Softwood Lumber V would deprive the second sentence of Article 2.4.2 of effect.


          1. Prospective normal value duty assessment77


        45. Under a prospective normal value duty assessment system, anti-dumping duties are assessed as individual import transactions occur, by comparing a transaction-specific export price against a


          image

          75 In response to Question 9 from the Panel, Canada asserted that the "weighted average normal value" (Article 2.4.2, first sentence) is the same as the "normal value established on a weighted average basis" (Article 2.4.2, second sentence). Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 9. Canada "understands these terms to both refer to an aggregate weighted-average normal value for the 'like product'." Furthermore, the only change in the data set referred to by Canada in its discussion of targeted dumping hypotheticals concerns the export transactions, not the home market transactions in the exporting Member. Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 5, para. 20).

          76 Panel Report, United States – Laws, Regulations and Methodology for Calculating Dumping Margins ("Zeroing") ("US – Zeroing (EC)"), WT/DS294/R, 31 October 2005 (appeal pending) at para. 7.266.

          77 We have focused our analysis on the type of prospective normal value assessment system applied,

          for example, by Canada. Similar issues will likely arise in respect of other prospective assessment systems applied by other Members. For example, the issue of offsets could arise in the context of export prices exceeding the minimum export price in the duty assessment system applied by Argentina, or export prices exceeding the floor price occasionally applied by the European Communities.

          WT/DS264/RW

          Page 25


          prospective normal value. Article 9.3 of the AD Agreement provides that the amount of anti-dumping duty shall not exceed the "margin of dumping" established under Article 2. Canada asserts that "[a] prospective normal value assessment system ... assesses anti-dumping duties as imports occur through a comparison between the export price and the prospective normal value. An investigating authority assesses anti-dumping duties when the export price is lower than the weighted-average normal value, but applies no anti-dumping duties to non-dumped transactions when the opposite is true."78 We therefore understand Canada to accept that, in the context of a prospective normal value duty assessment system, the "margin of dumping" referred to in Article 9.3 is the transaction-specific margin of dumping established in respect of the specific import transaction being assessed. This approach is confirmed by Article 9.2 of the AD Agreement, which provides that anti-dumping duties are levied in respect of "imports of [the relevant] product". In the context of such transaction-specific duty assessment, it makes no sense to talk of a margin of dumping being established for the product as a whole, by aggregating the results of all comparisons, since there is only one comparison at issue.


        46. If other comparisons were somehow relevant in the context of a prospective normal value duty assessment system, the application of the Appellate Body's interpretation of the phrase "margins of dumping" in the sense argued by Canada would require offsets for non-dumped transactions, in light of the obligation to take all (including non-dumped) comparisons into account in determining the margin of dumping for the product as a whole. In other words, an importer being assessed an anti- dumping duty for a dumped transaction, i.e., one in which export price was less than the prospective normal value, would receive an offset for non-dumped transactions, i.e., those in which export price was more than the prospective normal value, even if those transactions are made by other importers.79 This is illogical, as it would provide importers clearing dumped transactions with a double competitive advantage vis-à-vis other importers: first, they would benefit from the lower price inherent in a dumped transaction; second, they would benefit from offsets, or credits, "financed" by the higher prices paid by other importers clearing non-dumped, or even less-dumped, transactions.


        47. When asked a question regarding this issue, Canada "observe[d]that a prospective normal value system does not employ the practice of zeroing. ... An investigating authority assesses anti- dumping duties when the export price is lower than the weighted-average normal value, but applies no anti-dumping duties to non-dumped transactions when the opposite is true. It is not the same as the practice of zeroing, i.e., the changing of the results of intermediate values prior to their aggregation into a margin of dumping" (see Canada's reply to Question 4 from the Panel). Canada's answer ignores the anomaly that we have identified, since it is premised on the fact that investigating authorities apply prospective normal value assessment systems on a transaction-specific basis, without any obligation to offset for non-dumped amounts. Canada's answer does not explain how investigating authorities may assess on a transaction-specific basis notwithstanding the alleged obligation to calculate margins of dumping for the product as a whole. Nor does Canada explain why, despite the establishment of margins of dumping for the product as a whole, there is no need to offset non-dumped amounts against dumped amounts at the time of assessment.


        48. Furthermore, the first sentence of Article 9.3.2 of the AD Agreement provides:


          When the amount of the anti-dumping duty is assessed on a prospective basis, provision shall be made for a prompt refund, upon, request, of any duty paid in excess of the margin of dumping.



          image

          78 See Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 4, para. 13.

          79 This is because margins of dumping are established per exporter/producer, not per importer. Thus,

          "the single margin of dumping for the product as a whole" referred to by Canada is actually a margin specific to a given exporter/producer, on the basis of all import transactions by all importers sourcing the product from that exporter/producer.

          WT/DS264/RW

          Page 26


        49. If the Appellate Body's interpretation of "margins of dumping" were to apply throughout the AD Agreement in the sense argued by Canada, this provision, which expressly applies in respect of prospective duty assessment systems, would mean that a refund becomes payable if an anti-dumping duty is paid in excess of the single margin of dumping for the product as a whole, calculated by aggregating the results of all intermediate comparisons, without zeroing. Again, this makes no sense in the context of a prospective normal value duty assessment system, because (as even Canada acknowledges) the "margin of dumping" at issue is a transaction-specific price difference calculated for a specific import transaction. And if other comparisons for the product as a whole were somehow relevant, offsets would have to be provided for non-dumped transactions, with the result that one importer could request a refund on the basis of a margin of dumping calculated by reference to non- dumped transactions made by other importers. We are unable to accept that the Appellate Body could have intended such absurd results to follow from its interpretation of the phrase "margins of dumping" in US - Softwood Lumber V.


          1. Article 2.2


        50. Article 2.2 provides:


          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, such sales do not permit a proper comparison, the margin of dumping shall be determined 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. (footnote omitted)


        51. The Panel asked the parties the following question regarding this provision:


          Would the parties, and third parties, please describe how their investigating authorities would apply the provisions of Article 2.2 in a case involving multiple allegedly dumped models of a product under consideration, where there are no sales in the home market of some of those models? Specifically, would the parties consider it obligatory, under that provision, to determine normal value on one single basis for all models, or would the parties consider that Article 2.2 permits the determination of normal value on, for instance, the basis of home market sales for some models, and constructed normal value for others?


        52. Canada replied in relevant part that:


          most investigating authorities, including CBSA, examine whether there are sufficient “like product” sales in the ordinary course of trade for a particular sub-group or model when these authorities conduct a weighted-average-to-weighted-average calculation. Canada does not believe that this practice is problematic – investigating authorities are simply conducting a more detailed analysis to ensure that the “margin of dumping” is calculated in a more accurate manner. If investigating authorities failed to conduct such an analysis it could lead to less accurate sub-group calculations with the weighted-average export price being compared to a weighted-average normal value that is comprised of a handful of normal value transactions. The Appellate Body found that the general language of Article 2.4.2 permitted the use of “multiple averaging” or “model matching” provided that the intermediate values are properly aggregated. Article 2.2 also uses general language that should be interpreted in a permissive manner. Article 2.2 provides a conceptual description of the practice of using constructed normal values or third country sales. It should not be interpreted

          WT/DS264/RW

          Page 27


          to prohibit a more detailed form of this analysis that increases the accuracy of the calculation methodology.


        53. This provision governs, inter alia, the use by investigating authorities of a constructed normal value. If the reference in Article 2.2 to "margin of dumping" were understood to mean a single margin of dumping for the product as a whole, this would suggest to the Panel that a constructed normal value would have to be used to establish a single margin of dumping for the product as a whole "[w]hen 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, such sales do not permit a proper comparison". In other words, once the conditions for use of a constructed normal value were triggered, a constructed normal value would necessarily be used in all aspects of the determination of the margin of dumping for the product as a whole.


        54. In our view, Canada's suggestion that Article 2.2 may be applied on a model basis to increase the accuracy of the calculation methodology is not consistent with its view that investigating authorities must calculate a single margin of dumping for the product as a whole. For example, if there were ten models of the like product, and the Article 2.2 trigger conditions applied with respect to only one model, Canada asserts that an investigating authority may use a constructed normal value for only that one model (by making a "sub-group calculation"). However, Article 2.2 stipulates that the "margin of dumping" shall be established using a constructed normal value whenever the trigger conditions are fulfilled. If "margin of dumping" in Article 2.2 means a single margin of dumping for the product as a whole (and not for a specific model), this must mean that the margin of dumping for the product as a whole must be calculated using a constructed normal value for all models, even if the trigger conditions only apply in respect of one model. In other words, there would never be a "sub- group calculation" of the sort envisaged by Canada. We can see nothing in the text of Article 2.2 that would require this result, and, as Canada indicates, investigating authorities do not understand it to require this result. Indeed, the use of different methods for establishing normal value for different models or sub-groups of product is an integral part of model averaging, which itself is permitted under Article 2.4.2 of the AD Agreement.80 Moreover, such mandatory use of constructed normal value in respect of all models would run counter to the principle that constructed normal value is an alternative to be used only in the limited circumstances provided for in Article 2.2. It would also increase the burden on respondents, who would be required to produce cost data for all models, rather than just one or a few for which constructed normal value is necessary. We are not convinced that the Appellate Body could have intended its US - Softwood Lumber V findings to be applied in this manner. We also note that this approach is at odds with Canada's description of its own application of Article 2.2. These considerations serve to confirm our doubts regarding the broader application of the Appellate Body's US - Softwood Lumber V interpretation of the term "margins of dumping" sought by Canada.


          1. Past GATT Discussion


        55. On 27 May 1960, GATT Contracting Parties adopted the Second Report of the Group of Experts on Anti-Dumping and Countervailing Duties. That Report discussed what was referred to as the "pre-selection system" in the following terms:


          The pre-selection system


            1. In considering the pre-selection system in comparison with other systems the Group of Experts thought it desirable to re-affirm the following principles:


              image

              80 Appellate Body Report, US - Softwood Lumber V, supra note 9, at para. 80.

              WT/DS264/RW

              Page 28


              1. Anti-dumping duties should never be used for the purpose of ensuring normal protection for a domestic industry; such protection was the task of the tariff.


              2. The imposition of anti-dumping duties was justified only:


                1. where a product was in fact found to be dumped, and,

                2. where the dumping caused or threatened material injury to a domestic industry

              - the judgment of which rested with the governmental authorities of the importing country.


            2. The Group considered that the ideal method of fulfilling these principles was to make a determination in respect of both dumping and material injury in respect of each single importation of the product concerned. This, however, was clearly impracticable, particularly as regards injury.


            3. Failing such a method, the pre-selection system seemed to be the most satisfactory, since under such a system anti-dumping duties were applied only after a specific complaint had been investigated and a finding of dumping and material injury made. Provided the pro- selection system was administered at a high level, it could substantially reduce the number of cases in which anti-dumping duties were actually applied. An additional advantage of the system was that it involved a certain amount of publicity which in itself might serve as a deterrent to dumping.


            4. The Group was, in general, of the opinion that anti-dumping measures adopted after the pre-selection procedure had been followed should be directed only against such firms as had been found responsible for the dumping, or at most against those countries from which the dumped imports came."81


        56. In referring to a "determination ... of ... dumping ... in respect of each single importation of the product concerned", the Group of Experts clearly envisaged the calculation of transaction-specific margins of dumping. This would suggest that the Group of Experts did not consider that there was anything in the definition of dumping set forth in Article VI of the GATT that would preclude the calculation of such transaction-specific margins. This in turn would suggest that the GATT Contracting Parties would have disagreed with Canada's reliance on the same provision of the GATT 1994 to support its argument that "margins of dumping" must always be calculated "for the product as a whole" by aggregating all transaction-specific comparisons.


          (i) Conclusion


        57. To conclude, neither the ordinary meaning of the first sentence of Article 2.4.2 as a whole, nor the ordinary meaning of the phrase "margins of dumping" in particular, require that all transaction-specific comparisons under the T-T comparison methodology must be treated as "intermediate values" and aggregated, without zeroing, in order to arrive at a single margin of dumping for the product as a whole. Nor is such an approach mandated – in the context of the T-T comparison methodology – by the Appellate Body's interpretation of the phrase "margins of dumping" in US - Softwood Lumber V. Indeed, broader contextual considerations demonstrate that the application of the Appellate Body's interpretation beyond the confines of the W-W comparison methodology would lead to absurd results that could never have been intended by the Appellate Body, let alone the drafters of the AD Agreement.


        58. Conscious that Article 3.2 of the DSU precludes the DSB from "add[ing] to ... the ... obligations" of the United States, and taking into account the standard of review provided for in


          image

          81 BISD 9S/194, para. 7 (emphasis supplied)

          WT/DS264/RW

          Page 29


          Article 17.6(ii) of the AD Agreement, we reject Canada's interpretation of the phrase "margins of dumping" in the context of the transaction-to-transaction comparison methodology provided for in the first sentence of Article 2.4.2, and find that the interpretation put forward by the United States is permissible. We therefore find that the DOC was entitled not to offset the non-dumped transactions against the dumped transactions when calculating the margin of dumping for each respondent foreign producer or exporter. Accordingly, we reject Canada's claim that the DOC's use of zeroing in the T-T comparison methodology at issue is inconsistent with Article 2.4.2 of the AD Agreement.


    3. CANADA'S ARTICLE 2.4 CLAIM


      1. Canada claims that DOC's zeroing in the Section 129 Determination is inconsistent with Article 2.4. Canada's Article 2.4 claim is based on the first sentence of that provision, which provides:


        A fair comparison shall be made between the export price and the normal value.


        1. Main Arguments of the Parties


      2. Canada asserts that DOC manipulated the comparisons where the export price was higher than the home market price by disregarding the difference between these prices and replacing it with a zero value. According to Canada, this manipulation of transaction-to-transaction comparisons cannot be considered a “fair comparison” between the export price and the normal value as it inflated the margins of dumping. In support, Canada relies on the Appellate Body's statement in EC – Bed Linen that by not taking into account all comparisons, the practice of “zeroing” does not provide a fair comparison between the export price and the normal value and is, therefore, inconsistent with Article 2.4 of the Anti-Dumping Agreement. Canada also argues that, according to the Appellate Body in U.S. – Corrosion-Resistant Steel Sunset Review, zeroing introduces an “inherent bias” that “may distort not only the magnitude of a dumping margin, but also a finding of the very existence of dumping.”82 Canada submits that, because of such "inherent bias", the practice of zeroing in the context of the transaction-to-transaction methodology is by definition inconsistent with Article 2.4 of the Anti-Dumping Agreement.83


      3. The US submits that the case law relied on by Canada is without relevance. The US asserts that, because Article 2.4 was not at issue in the EC – Bed Linen dispute, any reference by the Appellate Body in that case to Article 2.4 was obiter dictum. The US asserts that the Appellate Body's reference to Article 2.4 in U.S. – Corrosion-Resistant Steel Sunset Review is not relevant because it was made in the context of an average-to-transaction comparison methodology.


      4. The United States further submits that Canada's interpretation of Article 2.4 would mean that that provision pertains to steps an investigating authority takes after making a comparison between export price and normal value, whereas Article 2.4 addresses only adjustments that must be made before a comparison is performed. The United States also submits that Canada’s suggestion that the “fair comparison” requirement in Article 2.4 contains a general obligation to offset dumping margins also cannot be reconciled with Article 2.4.2. The United States asserts that, if Canada were correct in arguing that there is a general obligation to offset, then the fair comparison obligation would also require the investigating authority to provide for an offset for transactions that exceed normal value



        image

        82 Appellate Body Report, United States – Sunset Review of Anti-Dumping Duties on Corrosion- Resistant Carbon Steel Flat Products from Japan ("US – Corrosion-Resistant Steel Sunset Review"), WT/DS244/AB/R, adopted 9 January 2004, at para. 135.

        83 Canada's arguments are broadly supported by China, the European Communities, India, Japan and Thailand.

        WT/DS264/RW

        Page 30


        even when using the targeted dumping methodology.84 The United States argues that if offsetting were required, the overall dumping margin calculated for an exporter must, mathematically, be the same under both a symmetrical comparison of weighted averages of normal values and export prices and an asymmetrical comparison of weighted average normal values and individual export prices. According to the United States, the “general obligation” that Canada posits therefore cannot exist, because if it existed it would nullify any distinction between the average-to-average and the average- to-transaction methodologies in Article 2.4.2. The United States notes that this approach was adopted by the panel in US – Zeroing (EC).85


      5. The United States further notes that the panel in US – Zeroing (EC) found that any analysis of the "fairness" of a methodology should be discerned from a "standard of appropriateness or rightness within the four corners of the Antidumping Agreement which would provide a basis for reliably judging that there has been an unfair departure from that standard."86 According to the United States, that panel found that the fact that one assessment methodology may result in a higher margin than another may only be deemed "unfair" if the other methodology could be determined to be the only "correct" methodology pursuant to the text of the AD Agreement.87 88


        1. Evaluation by the Panel


      6. Canada claims that zeroing is "by definition" inconsistent with Article 2.4 because it fails to take into account all comparisons, and because it introduces an "inherent bias" that inflates the margin of dumping.


      7. Canada's arguments are primarily based on statements made by the Appellate Body in EC – Bed Linen and US – Corrosion-Resistant Steel Sunset Review. We are not persuaded of the relevance of these statements by the Appellate Body to the case in hand. First, we note that, in the EC – Bed Linen case, none of the legal issues before the Appellate Body concerned Article 2.4 of the AD Agreement. Anything the Appellate Body may have said regarding Article 2.4 was therefore obiter dictum. Second, we note that, as relevant here, the US – Corrosion-Resistant Steel Sunset Review case concerned the application of the W-T comparison methodology in the context of an assessment review, as opposed to an original investigation, whereas the present Article 21.5 proceedings concern DOC's use of the T-T comparison methodology in an original investigation.


      8. Turning to the substance of Canada's claim, we believe that a claim based on a highly general and subjective test such as "fair comparison" should be approached with caution by treaty interpreters. For this reason, any conception of "fairness" should be solidly rooted in the context provided by the


        image

        84 The United States asserts that, in the underlying proceeding before the Appellate Body, Canada conceded that “zeroing is permitted under the third methodology [i.e., the targeted dumping methodology].” See Appellate Body Report, US – Softwood Lumber V, supra note 9, at para. 105 n.164. We note that, in response to the US assertion, Canada filed a letter with this Panel that it had provided to the Appellate Body on 28 September 2004. In that letter, Canada stated that it "did not take the position [during the appeal] that zeroing is permitted under the third methodology." Canada's letter recognized that the Appellate Body report had already been adopted, and for that reason, Canada's letter did not ask the Appellate Body to take any specific action in this regard. According to the United States, Canada did not offer then and does not offer now any textual basis for a distinction between the fair comparison requirement as applied to the targeted dumping methodology and the fair comparison requirement as applied to the other two methodologies provided for in Article 2.4.2. The United States asserts that, in fact, Canada expressly characterizes the obligation it asserts as a “general obligation.” Canada's first written, para. 28.

        85 Panel Report, US – Zeroing (EC), supra note 76, at para. 7.266.

        86 Ibid, para 7.260.

        87 Ibid.

        88 New Zealand broadly supports the US interpretation of Article 2.4, claiming that the calculation of margins of dumping using the T-T comparison methodology is inherently a "fair comparison" in terms of

        Article 2.4 of the AD Agreement..

        WT/DS264/RW

        Page 31


        AD Agreement, and perhaps the WTO Agreement more generally. As such, there must be a discernable standard within the AD Agreement, and perhaps the WTO Agreement, by which to assess whether or not a comparison has been "fair" or "unfair". Thus, the fact that comparison methodology A produces a higher margin of dumping than comparison methodology B would only make comparison methodology A unfair if comparison methodology B were the applicable standard.89 If, however, the AD Agreement were to permit either comparison methodology A or B, this would not be the case.


      9. Since we have already concluded that T-T with zeroing (resulting in higher margins) is not inconsistent with Article 2.4.2, one can not conclude that failure to use a comparison methodology that would have resulted in lower margins (i.e., T-T without zeroing) is "unfair". For the same reason, one can not conclude that failure to use a comparison methodology that would have reflected all comparisons, by offsetting non-dumped amounts against dumped amounts (i.e., T-T without zeroing), is "unfair". The principle of effective treaty interpretation implies that the "fair comparison" obligation in article 2.4 must not be interpreted in a manner so as to trump the more specific provisions of Article 2.4.2.90


      10. The principle of effective treaty interpretation is also relevant to Canada's argument that zeroing is "by definition" inconsistent with Article 2.4. Since Canada's argument (that zeroing when comparing normal value and export price is unfair) is unqualified, it must apply to zeroing under any of the three comparison methodologies set forth in Article 2.4.2. We have already established, however, that a prohibition of zeroing under the targeted dumping comparison methodology would render the second sentence of Article 2.4.2 meaningless, in the sense that it would result in a margin of dumping mathematically equivalent to that established under the W-W comparison methodology. Since zeroing therefore can not be prohibited as "by definition" unfair in the context of Article 2.4.2, Article 2.4 can not provide for the unqualified, "by definition" prohibition suggested by Canada.


      11. In addition, we recall our discussion of certain contextual considerations when examining Canada's Article 2.4.2 claim.91 In our view, these contextual considerations suggest that there is no general obligation under the AD Agreement to offset non-dumped amounts against dumped amounts. Indeed, any such general obligation would have profound implications for prospective normal value duty assessment procedures of the sort applied by Canada. In light of these contextual considerations, we are unable to accept that mandatory offsets establish a standard by which to assess fairness for the purpose of Article 2.4.


      12. For the above reasons, we reject Canada's claim that the United States has violated the fair comparison obligation provided for in the first sentence of Article 2.4 of the AD Agreement.


    image

    89 This approach was set forth in the dissenting opinion in the original panel report, see, Panel Report, US - Softwood Lumber V, supra note 5, at para. 9.16-9.22, and was also followed by the panel in Panel Report, US – Zeroing (EC), supra note 76, at para. 7.260.

    90 In its reply to Question 22 from the Panel, Canada appears to endorse such an approach to the issue.

    Responses of Canada to Questions to the Parties following the Substantive Meeting of the Panel, Response to Question 22. In particular, having cited the statement of the panel in US – Zeroing (EC) regarding the "discernable standard" (cited above), Canada asserts that "[i]t follows that the concept of 'fairness' relates to the substantive rules concerning the calculation of margins of dumping under the methodologies found in Article 2.4.2 – rules which prohibit zeroing when aggregating intermediate comparisons." In other words, Canada agrees that the fairness obligation in Article 2.4 may be interpreted in light of the substantive rules of Article 2.4.2. Canada errs in its conclusion, however, because, as we concluded above, Article 2.4.2 does not prohibit zeroing in the context of the T-T comparison methodology.

    91 See paras. 5.31 - 5.64 above.

    WT/DS264/RW

    Page 32


  6. CONCLUSIONS AND RECOMMENDATION


    1. Based on the foregoing, we conclude that the determination of the DOC in the section 129 proceeding investigation is not inconsistent with the asserted provisions of Articles 2.4 and 2.4.2 of the AD Agreement.


    2. We therefore consider that the United States has implemented the recommendations and rulings of the DSB in US – Softwood Lumber V, to bring its measure into conformity with its obligations under the AD Agreement.


    3. Having found that the United States did not act inconsistently with its obligations under the asserted WTO Agreements, we consider that no recommendation under Article 19.1 of the DSU is necessary, and we make none.


image